Drug Therapy

In Reply: We reported on publications from 2016–2017 and, unfortunately, at the time we were writing our paper, the European Society of Cardiology (ESC) update on dual antiplatelet therapy¹ had not yet been published. We presented the recommendations from the American College of Cardiology (ACC) and American Heart Association (AHA),² which differ from the recently published ESC guidelines. The ESC suggests that the minimum waiting period after drug-eluting stent placement before noncardiac surgery should be 1 month rather than 3 months but acknowledges that in the setting of complex stenting or recent acute coronary syndrome, 6 months is preferred. The recommendation in this latter scenario is a class IIb C recommendation—essentially expert consensus opinion.

Further, in the study by Egholm et al,³ the event rates in patients undergoing noncardiac surgery in the 1- to 2-month period were numerically higher than in the control group, and no adjusted odds ratios were given. The numbers of events were very low, and a change of only 1 or 2 events in the other direction in the groups would likely make it statistically significant.

All of these recommendations are based on observational studies and registry data, as there are no randomized controlled trials to address this issue. There are many complexities to be accounted for including the type of stent, timing, circumstances surrounding stenting, anatomy, number of stents, patient comorbidities (particularly age, diabetes mellitus, cardiac disease), type of surgery and anesthesia, and perioperative management of antiplatelet therapy. While we acknowledge the ESC recommendation, we would urge caution in the recommendation to wait only 1 month, and in the United States most would prefer to wait 3 months if possible.

In Reply: We reported on publications from 2016–2017 and, unfortunately, at the time we were writing our paper, the European Society of Cardiology (ESC) update on dual antiplatelet therapy¹ had not yet been published. We presented the recommendations from the American College of Cardiology (ACC) and American Heart Association (AHA),² which differ from the recently published ESC guidelines. The ESC suggests that the minimum waiting period after drug-eluting stent placement before noncardiac surgery should be 1 month rather than 3 months but acknowledges that in the setting of complex stenting or recent acute coronary syndrome, 6 months is preferred. The recommendation in this latter scenario is a class IIb C recommendation—essentially expert consensus opinion.

Further, in the study by Egholm et al,³ the event rates in patients undergoing noncardiac surgery in the 1- to 2-month period were numerically higher than in the control group, and no adjusted odds ratios were given. The numbers of events were very low, and a change of only 1 or 2 events in the other direction in the groups would likely make it statistically significant.

All of these recommendations are based on observational studies and registry data, as there are no randomized controlled trials to address this issue. There are many complexities to be accounted for including the type of stent, timing, circumstances surrounding stenting, anatomy, number of stents, patient comorbidities (particularly age, diabetes mellitus, cardiac disease), type of surgery and anesthesia, and perioperative management of antiplatelet therapy. While we acknowledge the ESC recommendation, we would urge caution in the recommendation to wait only 1 month, and in the United States most would prefer to wait 3 months if possible.

In Reply: We reported on publications from 2016–2017 and, unfortunately, at the time we were writing our paper, the European Society of Cardiology (ESC) update on dual antiplatelet therapy¹ had not yet been published. We presented the recommendations from the American College of Cardiology (ACC) and American Heart Association (AHA),² which differ from the recently published ESC guidelines. The ESC suggests that the minimum waiting period after drug-eluting stent placement before noncardiac surgery should be 1 month rather than 3 months but acknowledges that in the setting of complex stenting or recent acute coronary syndrome, 6 months is preferred. The recommendation in this latter scenario is a class IIb C recommendation—essentially expert consensus opinion.

Further, in the study by Egholm et al,³ the event rates in patients undergoing noncardiac surgery in the 1- to 2-month period were numerically higher than in the control group, and no adjusted odds ratios were given. The numbers of events were very low, and a change of only 1 or 2 events in the other direction in the groups would likely make it statistically significant.

All of these recommendations are based on observational studies and registry data, as there are no randomized controlled trials to address this issue. There are many complexities to be accounted for including the type of stent, timing, circumstances surrounding stenting, anatomy, number of stents, patient comorbidities (particularly age, diabetes mellitus, cardiac disease), type of surgery and anesthesia, and perioperative management of antiplatelet therapy. While we acknowledge the ESC recommendation, we would urge caution in the recommendation to wait only 1 month, and in the United States most would prefer to wait 3 months if possible.

Study Overview

Objective. To determine if romosuzumab, an antisclerostin antibody, is superior to alendronate in reducing the incidence of fracture in postmenopausal women with osteoporosis at high-risk for fracture.

Design. Multicenter, international, double-blind, randomized clinical trial.

Setting and participants. 4093 postmenopausal women with osteoporosis and a previous fragility fracture were enrolled from over 40 countries worldwide. Patients were eligible for the study if they were 55 to 90 years old and were deemed at high risk for future fracture based on bone mineral density (BMD) T score at the total hip or femoral neck and fracture history. This included T score ≤ –2.5 and ≥ 1 moderate or severe vertebral fractures or ≥ 2 mild vertebral fractures; T score ≤ –2.0 and either ≥ 2 moderate or severe vertebral fractures or proximal femur fracture within 3 to 24 months before randomization. Subjects with a history of prior use of medications that affect bone metabolism were excluded, as were those with other metabolic bone disease, vitamin D deficiency, uncontrolled metabolic disease, malabsorption syndromes, history of transplant, severe renal insufficiency, malignancy or severe illness.

Intervention. Patients were randomized to either subcutaneous romosuzumab 210 mg monthly or oral alendronate 70 mg weekly for 12 months. Following the 12-month double-blind period, all patients received open-label weekly alendronate until the end of the trial, with maintenance of blinding to the initial treatment assignment. Primary analysis occurred when all subjects had completed the 24-month visit and clinical fractures had been confirmed in at least 330 patients. All patients received daily calcium and vitamin D. Lateral radiographs of the thoracic and lumbar spine were obtained at screening and months 12 and 24. The BMD at the lumbar spine and proximal femur was evaluated by dual-energy x-ray absorptiometry at baseline and every 12 months thereafter. Serum concentrations of bone-turnover markers were measured in a subgroup of patients.

Main outcome measures. The primary outcomes were the incidence of new vertebral fracture and the incidence of clinical fracture at 24 months. Clinical fractures included symptomatic vertebral fracture and nonvertebral fractures. The secondary outcomes were the BMD at the lumbar spine, total hip, and femoral neck at 12 and 24 months, the incidence of nonvertebral fracture, and fracture category. Safety outcomes included the incidence of adjudicated clinical events, including serious cardiovascular adverse events, osteonecrosis of the jaw, and atypical femoral fracture. Serious cardiovascular events were defined as cardiac ischemic event, cerebrovascular event, heart failure, death, non-coronary revascularization and peripheral vascular ischemic event not requiring revascularization.

Analysis. An intention to treat approach was used for data analysis. For the incidence of fractures, the treatment groups were compared using a Cox proportional-hazards model and the Mantel-Haenszel method with adjustment for age (< 75 vs ≥ 75 years), the presence or absence of severe vertebral fracture at baseline, and baseline BMD T score at the total hip. Between-group comparisons of the percentage change in BMD from baseline were analyzed by means of a repeated-measures model with adjustment for treatment, age category, baseline severe vertebral fracture, visit, treatment-by-visit interaction, and baseline BMD. Percentage changes from baseline in bone turnover were assessed using a Wilcoxon rank-sum test. The safety analysis included cumulated incidence rates of adverse outcomes. Odds ratios and confidence intervals were estimated for serious cardiovascular adverse events with the use of a logistic regression model.

Main results. 2046 participants were randomized to the romosozumab group and 2047 to the alendronate group. A total of 3654 participants from both groups (89.3%) completed 12 months of the trial, and 3150 (77.0%) completed the primary analysis period. The treatment groups were similar in baseline age, ethnicity, and fracture history. The majority of patients in both groups were non-Hispanic (> 60%) and ≥ 75 years old (> 50%). The mean age of the patients was 74.3 years. Baseline mean bone mineral density T scores were –2.96 at the lumbar spine, –2.8 at the total hip, and –2.9 at the femoral neck.

After 24 months of treatment, 6.2% of patients in the romosozumab-alendronate group had a new vertebral fracture as compared to 11.9% in the alendronate-alendronate group. This represents a 48% lower risk (risk ratio 0.52, 95% confidence interval [CI] 0.4–0.66; P < 0.001) of new vertebral fractures with romosozumab. At the time of the primary analysis, romosozumab followed by alendronate resulted in a 27% lower risk of clinical fracture than alendronate alone (hazard ratio 0.73, 95% CI 0.61–0.88; P < 0.001). 8.7% of the romosozumab-alendronate group had a nonvertebral fracture versus 10.6% in the alendronate-alendronate group, representing a 19% lower risk with romosozumab (hazard ratio 0.81, 95% CI 0.66–0.99; P = 0.04). Hip fractures occurred in 2.0% of the romosozumab-alendronate group as compared with 3.2% in the alendronate-alendronate group, representing a 38% lower risk with romosozumab (hazard ratio 0.62, 95% CI 0.42–0.92; P = 0.02).

Patients in the romosozumab-alendronate group had greater gains in BMD from baseline at the lumbar spine (14.9% vs 8.5%) and total hip (7% vs 3.6%) compared to the alendronate-alendronate group. (P < 0.001 for all comparisons). At 12 months, romosozumab treatment resulted in decreased levels of bone resorption marker β-CTX and increased levels of bone formation marker P1NP. β-CTX and P1NP decreased and remained below baseline levels after transitioning to alendronate. In the alendronate-alendronate group, P1NP and β-CTX decreased within 1 month and remained below baseline levels at 36 months.

Overall, the adverse events and serious event rates were similar between the 2 treatment groups during the double-blind period with 2 exceptions. In the first 12 months, injection-site reactions were reported in 4.4% of patients receiving romosozumab compared to 2.6% in those receiving alendronate. Patients in the romosozumab group had an increased incidence of adjudicated serious cardiovascular outcomes during the double-blind period, 2.5% (50 of 2040 patients) compared to 1.9% (38 of 2014 patients) in the alendronate group. During the open-label period, osteonecrosis of the jaw occurred in one patient in each group. Two atypical femoral fractures occurred in the romosozumab-alendronate group, compared to 4 in the alendronate-alendronate group. During the first 18 months of the study, binding anti-romosozumab antibodies were observed in 15.3% of the romosozumab group, with neutralizing antibodies in 0.6%.

Conclusion. In postmenopausal woman with osteoporosis and high fracture risk, 12 months of romosozumab treatment followed by alendronate resulted in significantly lower risk of fracture than use of alendronate alone.

Commentary

Osteoporosis-related fragility fractures carry a substantial risk of morbidity and mortality [1]. The goal of osteoporosis treatment is to ameliorate this risk. The current FDA-approved medications for osteoporosis can be divided into anabolic (teriparatide, abaloparatide) and anti-resorptive (bisphosphonate, denosumab, selective estrogen receptor modulators) categories. Sclerostin is a glycoprotein produced by osteocytes that inhibits the Wnt signaling pathway, thereby impeding osteoblast proliferation and activity. Romosozumab is a monoclonal antisclerostin antibody that results in both increased bone formation and decreased bone resorption [1]. By apparently uncoupling bone formation and resorption to increase bone mass, this medication holds promise to become the ideal osteoporosis drug.

Initial studies have shown that 12 months of romosozumab treatment significantly increased BMD at the lumbar spine (+11.3%), as compared to placebo (–0.1%), alendronate (+4.1%), and teriparatide (+7.1%) [2]. The Fracture Study in Postmenopausal Women with Osteoporosis (FRAME) was a large (7180 patients) randomized controlled trial that demonstrated that 12 months of romosozumab resulted in a 73% lower risk of vertebral fracture and 36% lower risk of clinical fracture compared to placebo [3]. However, there was no significant reduction in non-vertebral facture [3]. This may be due to the fact that FRAME excluded women at the highest risk for fracture. That is, exclusion criteria included history of hip fracture, any severe vertebral facture, or more than 2 moderate vertebral fractures. The current phase 3 ARCH trial (Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk) attempts to clarify the potential benefit of romosozumab treatment in this very high-risk patient population, compared to a common first-line osteoporosis treatment, alendronate.

Indeed, ARCH demonstrates that sequential therapy with romosozumab followed by alendronate is superior to alendronate alone in improving BMD at all sites and preventing new vertebral, clinical, and non-vertebral fractures in postmenopausal women with osteoporosis and a history of fragility fracture. While ARCH was not designed as a cardiovascular outcomes trial, the higher rate of serious cardiovascular adverse events in the romosozumab group raises concern that romosozumab may have a negative effect on vascular tissue. Sclerostin is expressed in vascular smooth muscle [4] and upregulated at sites of vascular calcification [5]. It is possible that inhibiting sclerostin activity could alter vascular remodeling or increase vascular calcification. However, it is interesting that in the larger FRAME trial, no increase in adverse cardiovascular events was seen in the romosozumab group compared to placebo. This may be due to the fact that the average age of patients in FRAME was lower than ARCH. However, it also raises the hypothesis that alendronate itself may be protective in terms of cardiovascular risk. It has been postulated that bisphosphonates may have cardiovascular protective effects, given animal studies have demonstrated that alendronate downregulates monocyte chemoattractant protein 1 and macrophage inflammatory protein 1 [6]. However no cardioprotective benefit was seen in meta-analysis [7].

ARCH has several strengths, including its design as an international, double-blind, and randomized clinical trial. The primary outcome of cumulative fracture incidence is a hard endpoint and is clinically relevant. The intervention is simple and the results are clearly defined. The statistical assessment yields significant results. However, there are some limitations to the study. The lead author has received research support from Amgen and UCB Pharma, the makers of romosuzumab. Amgen and UCB Pharma designed the trial, and Amgen was responsible for trial oversight and data analyses per a pre-specified statistical analysis plan. An external independent data monitoring committee monitored unblinded safety data. Because there was no placebo-controlled arm, it is difficult to determine whether the unexpected cardiovascular signal was due to romosuzumab itself or a protective effect of alendronate. In addition, the majority of study participants were non-Hispanic from Central or Eastern Europe and Latin America, with only ~2% of patients from North America. As a result, ARCH findings may not be generalizable to other regional or ethnic populations. Furthermore, the majority of the patients were ≥ 75 years of age and were at very high fracture risk. It is unclear if younger patients or those with lower risk of fracture would see the same fracture prevention and BMD gain. In addition, because of the relatively short length of the trial, the durability of the metabolic bone benefit and cardiovascular risk is unknown. While the authors reported the increased anti-romosozumab antibodies in the romosozumab group had no detectable effect on efficacy or safety, given the short duration of the trial, this has not been proven.

Applications for Clinical Practice

The dual anti-resorptive and anabolic effect of romosozumab makes it an attractive and promising new osteoporosis therapy. ARCH suggests that sequential therapy with romosuzumab and alendronate is superior in terms of fracture prevention to alendronate alone in elderly postmenopausal women with osteoporosis and a history of fragility fractures, although longer term studies are needed to define the durability of this effect. While the absolute number of serious adjudicated cardiovascular events was low, the increased incidence in the romosuzumab group will likely prevent the FDA from approving this medication for widespread use at this time. Additional studies are needed to clarify the cause and magnitude of this cardiovascular risk and to determine whether prevention of fracture-associated morbidity and mortality is enough to mitigate it.

—Simona Frunza-Stefan, MD, and Hillary B. Whitlach, MD, University of Maryland School of Medicine, Baltimore, MD

Study Overview

Objective. To determine if romosuzumab, an antisclerostin antibody, is superior to alendronate in reducing the incidence of fracture in postmenopausal women with osteoporosis at high-risk for fracture.

Design. Multicenter, international, double-blind, randomized clinical trial.

Setting and participants. 4093 postmenopausal women with osteoporosis and a previous fragility fracture were enrolled from over 40 countries worldwide. Patients were eligible for the study if they were 55 to 90 years old and were deemed at high risk for future fracture based on bone mineral density (BMD) T score at the total hip or femoral neck and fracture history. This included T score ≤ –2.5 and ≥ 1 moderate or severe vertebral fractures or ≥ 2 mild vertebral fractures; T score ≤ –2.0 and either ≥ 2 moderate or severe vertebral fractures or proximal femur fracture within 3 to 24 months before randomization. Subjects with a history of prior use of medications that affect bone metabolism were excluded, as were those with other metabolic bone disease, vitamin D deficiency, uncontrolled metabolic disease, malabsorption syndromes, history of transplant, severe renal insufficiency, malignancy or severe illness.

Intervention. Patients were randomized to either subcutaneous romosuzumab 210 mg monthly or oral alendronate 70 mg weekly for 12 months. Following the 12-month double-blind period, all patients received open-label weekly alendronate until the end of the trial, with maintenance of blinding to the initial treatment assignment. Primary analysis occurred when all subjects had completed the 24-month visit and clinical fractures had been confirmed in at least 330 patients. All patients received daily calcium and vitamin D. Lateral radiographs of the thoracic and lumbar spine were obtained at screening and months 12 and 24. The BMD at the lumbar spine and proximal femur was evaluated by dual-energy x-ray absorptiometry at baseline and every 12 months thereafter. Serum concentrations of bone-turnover markers were measured in a subgroup of patients.

Main outcome measures. The primary outcomes were the incidence of new vertebral fracture and the incidence of clinical fracture at 24 months. Clinical fractures included symptomatic vertebral fracture and nonvertebral fractures. The secondary outcomes were the BMD at the lumbar spine, total hip, and femoral neck at 12 and 24 months, the incidence of nonvertebral fracture, and fracture category. Safety outcomes included the incidence of adjudicated clinical events, including serious cardiovascular adverse events, osteonecrosis of the jaw, and atypical femoral fracture. Serious cardiovascular events were defined as cardiac ischemic event, cerebrovascular event, heart failure, death, non-coronary revascularization and peripheral vascular ischemic event not requiring revascularization.

Analysis. An intention to treat approach was used for data analysis. For the incidence of fractures, the treatment groups were compared using a Cox proportional-hazards model and the Mantel-Haenszel method with adjustment for age (< 75 vs ≥ 75 years), the presence or absence of severe vertebral fracture at baseline, and baseline BMD T score at the total hip. Between-group comparisons of the percentage change in BMD from baseline were analyzed by means of a repeated-measures model with adjustment for treatment, age category, baseline severe vertebral fracture, visit, treatment-by-visit interaction, and baseline BMD. Percentage changes from baseline in bone turnover were assessed using a Wilcoxon rank-sum test. The safety analysis included cumulated incidence rates of adverse outcomes. Odds ratios and confidence intervals were estimated for serious cardiovascular adverse events with the use of a logistic regression model.

Main results. 2046 participants were randomized to the romosozumab group and 2047 to the alendronate group. A total of 3654 participants from both groups (89.3%) completed 12 months of the trial, and 3150 (77.0%) completed the primary analysis period. The treatment groups were similar in baseline age, ethnicity, and fracture history. The majority of patients in both groups were non-Hispanic (> 60%) and ≥ 75 years old (> 50%). The mean age of the patients was 74.3 years. Baseline mean bone mineral density T scores were –2.96 at the lumbar spine, –2.8 at the total hip, and –2.9 at the femoral neck.

After 24 months of treatment, 6.2% of patients in the romosozumab-alendronate group had a new vertebral fracture as compared to 11.9% in the alendronate-alendronate group. This represents a 48% lower risk (risk ratio 0.52, 95% confidence interval [CI] 0.4–0.66; P < 0.001) of new vertebral fractures with romosozumab. At the time of the primary analysis, romosozumab followed by alendronate resulted in a 27% lower risk of clinical fracture than alendronate alone (hazard ratio 0.73, 95% CI 0.61–0.88; P < 0.001). 8.7% of the romosozumab-alendronate group had a nonvertebral fracture versus 10.6% in the alendronate-alendronate group, representing a 19% lower risk with romosozumab (hazard ratio 0.81, 95% CI 0.66–0.99; P = 0.04). Hip fractures occurred in 2.0% of the romosozumab-alendronate group as compared with 3.2% in the alendronate-alendronate group, representing a 38% lower risk with romosozumab (hazard ratio 0.62, 95% CI 0.42–0.92; P = 0.02).

Patients in the romosozumab-alendronate group had greater gains in BMD from baseline at the lumbar spine (14.9% vs 8.5%) and total hip (7% vs 3.6%) compared to the alendronate-alendronate group. (P < 0.001 for all comparisons). At 12 months, romosozumab treatment resulted in decreased levels of bone resorption marker β-CTX and increased levels of bone formation marker P1NP. β-CTX and P1NP decreased and remained below baseline levels after transitioning to alendronate. In the alendronate-alendronate group, P1NP and β-CTX decreased within 1 month and remained below baseline levels at 36 months.

Overall, the adverse events and serious event rates were similar between the 2 treatment groups during the double-blind period with 2 exceptions. In the first 12 months, injection-site reactions were reported in 4.4% of patients receiving romosozumab compared to 2.6% in those receiving alendronate. Patients in the romosozumab group had an increased incidence of adjudicated serious cardiovascular outcomes during the double-blind period, 2.5% (50 of 2040 patients) compared to 1.9% (38 of 2014 patients) in the alendronate group. During the open-label period, osteonecrosis of the jaw occurred in one patient in each group. Two atypical femoral fractures occurred in the romosozumab-alendronate group, compared to 4 in the alendronate-alendronate group. During the first 18 months of the study, binding anti-romosozumab antibodies were observed in 15.3% of the romosozumab group, with neutralizing antibodies in 0.6%.

Conclusion. In postmenopausal woman with osteoporosis and high fracture risk, 12 months of romosozumab treatment followed by alendronate resulted in significantly lower risk of fracture than use of alendronate alone.

Commentary

Osteoporosis-related fragility fractures carry a substantial risk of morbidity and mortality [1]. The goal of osteoporosis treatment is to ameliorate this risk. The current FDA-approved medications for osteoporosis can be divided into anabolic (teriparatide, abaloparatide) and anti-resorptive (bisphosphonate, denosumab, selective estrogen receptor modulators) categories. Sclerostin is a glycoprotein produced by osteocytes that inhibits the Wnt signaling pathway, thereby impeding osteoblast proliferation and activity. Romosozumab is a monoclonal antisclerostin antibody that results in both increased bone formation and decreased bone resorption [1]. By apparently uncoupling bone formation and resorption to increase bone mass, this medication holds promise to become the ideal osteoporosis drug.

Initial studies have shown that 12 months of romosozumab treatment significantly increased BMD at the lumbar spine (+11.3%), as compared to placebo (–0.1%), alendronate (+4.1%), and teriparatide (+7.1%) [2]. The Fracture Study in Postmenopausal Women with Osteoporosis (FRAME) was a large (7180 patients) randomized controlled trial that demonstrated that 12 months of romosozumab resulted in a 73% lower risk of vertebral fracture and 36% lower risk of clinical fracture compared to placebo [3]. However, there was no significant reduction in non-vertebral facture [3]. This may be due to the fact that FRAME excluded women at the highest risk for fracture. That is, exclusion criteria included history of hip fracture, any severe vertebral facture, or more than 2 moderate vertebral fractures. The current phase 3 ARCH trial (Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk) attempts to clarify the potential benefit of romosozumab treatment in this very high-risk patient population, compared to a common first-line osteoporosis treatment, alendronate.

Indeed, ARCH demonstrates that sequential therapy with romosozumab followed by alendronate is superior to alendronate alone in improving BMD at all sites and preventing new vertebral, clinical, and non-vertebral fractures in postmenopausal women with osteoporosis and a history of fragility fracture. While ARCH was not designed as a cardiovascular outcomes trial, the higher rate of serious cardiovascular adverse events in the romosozumab group raises concern that romosozumab may have a negative effect on vascular tissue. Sclerostin is expressed in vascular smooth muscle [4] and upregulated at sites of vascular calcification [5]. It is possible that inhibiting sclerostin activity could alter vascular remodeling or increase vascular calcification. However, it is interesting that in the larger FRAME trial, no increase in adverse cardiovascular events was seen in the romosozumab group compared to placebo. This may be due to the fact that the average age of patients in FRAME was lower than ARCH. However, it also raises the hypothesis that alendronate itself may be protective in terms of cardiovascular risk. It has been postulated that bisphosphonates may have cardiovascular protective effects, given animal studies have demonstrated that alendronate downregulates monocyte chemoattractant protein 1 and macrophage inflammatory protein 1 [6]. However no cardioprotective benefit was seen in meta-analysis [7].

ARCH has several strengths, including its design as an international, double-blind, and randomized clinical trial. The primary outcome of cumulative fracture incidence is a hard endpoint and is clinically relevant. The intervention is simple and the results are clearly defined. The statistical assessment yields significant results. However, there are some limitations to the study. The lead author has received research support from Amgen and UCB Pharma, the makers of romosuzumab. Amgen and UCB Pharma designed the trial, and Amgen was responsible for trial oversight and data analyses per a pre-specified statistical analysis plan. An external independent data monitoring committee monitored unblinded safety data. Because there was no placebo-controlled arm, it is difficult to determine whether the unexpected cardiovascular signal was due to romosuzumab itself or a protective effect of alendronate. In addition, the majority of study participants were non-Hispanic from Central or Eastern Europe and Latin America, with only ~2% of patients from North America. As a result, ARCH findings may not be generalizable to other regional or ethnic populations. Furthermore, the majority of the patients were ≥ 75 years of age and were at very high fracture risk. It is unclear if younger patients or those with lower risk of fracture would see the same fracture prevention and BMD gain. In addition, because of the relatively short length of the trial, the durability of the metabolic bone benefit and cardiovascular risk is unknown. While the authors reported the increased anti-romosozumab antibodies in the romosozumab group had no detectable effect on efficacy or safety, given the short duration of the trial, this has not been proven.

Applications for Clinical Practice

The dual anti-resorptive and anabolic effect of romosozumab makes it an attractive and promising new osteoporosis therapy. ARCH suggests that sequential therapy with romosuzumab and alendronate is superior in terms of fracture prevention to alendronate alone in elderly postmenopausal women with osteoporosis and a history of fragility fractures, although longer term studies are needed to define the durability of this effect. While the absolute number of serious adjudicated cardiovascular events was low, the increased incidence in the romosuzumab group will likely prevent the FDA from approving this medication for widespread use at this time. Additional studies are needed to clarify the cause and magnitude of this cardiovascular risk and to determine whether prevention of fracture-associated morbidity and mortality is enough to mitigate it.

—Simona Frunza-Stefan, MD, and Hillary B. Whitlach, MD, University of Maryland School of Medicine, Baltimore, MD

Study Overview

Objective. To determine if romosuzumab, an antisclerostin antibody, is superior to alendronate in reducing the incidence of fracture in postmenopausal women with osteoporosis at high-risk for fracture.

Design. Multicenter, international, double-blind, randomized clinical trial.

Setting and participants. 4093 postmenopausal women with osteoporosis and a previous fragility fracture were enrolled from over 40 countries worldwide. Patients were eligible for the study if they were 55 to 90 years old and were deemed at high risk for future fracture based on bone mineral density (BMD) T score at the total hip or femoral neck and fracture history. This included T score ≤ –2.5 and ≥ 1 moderate or severe vertebral fractures or ≥ 2 mild vertebral fractures; T score ≤ –2.0 and either ≥ 2 moderate or severe vertebral fractures or proximal femur fracture within 3 to 24 months before randomization. Subjects with a history of prior use of medications that affect bone metabolism were excluded, as were those with other metabolic bone disease, vitamin D deficiency, uncontrolled metabolic disease, malabsorption syndromes, history of transplant, severe renal insufficiency, malignancy or severe illness.

Intervention. Patients were randomized to either subcutaneous romosuzumab 210 mg monthly or oral alendronate 70 mg weekly for 12 months. Following the 12-month double-blind period, all patients received open-label weekly alendronate until the end of the trial, with maintenance of blinding to the initial treatment assignment. Primary analysis occurred when all subjects had completed the 24-month visit and clinical fractures had been confirmed in at least 330 patients. All patients received daily calcium and vitamin D. Lateral radiographs of the thoracic and lumbar spine were obtained at screening and months 12 and 24. The BMD at the lumbar spine and proximal femur was evaluated by dual-energy x-ray absorptiometry at baseline and every 12 months thereafter. Serum concentrations of bone-turnover markers were measured in a subgroup of patients.

Main outcome measures. The primary outcomes were the incidence of new vertebral fracture and the incidence of clinical fracture at 24 months. Clinical fractures included symptomatic vertebral fracture and nonvertebral fractures. The secondary outcomes were the BMD at the lumbar spine, total hip, and femoral neck at 12 and 24 months, the incidence of nonvertebral fracture, and fracture category. Safety outcomes included the incidence of adjudicated clinical events, including serious cardiovascular adverse events, osteonecrosis of the jaw, and atypical femoral fracture. Serious cardiovascular events were defined as cardiac ischemic event, cerebrovascular event, heart failure, death, non-coronary revascularization and peripheral vascular ischemic event not requiring revascularization.

Analysis. An intention to treat approach was used for data analysis. For the incidence of fractures, the treatment groups were compared using a Cox proportional-hazards model and the Mantel-Haenszel method with adjustment for age (< 75 vs ≥ 75 years), the presence or absence of severe vertebral fracture at baseline, and baseline BMD T score at the total hip. Between-group comparisons of the percentage change in BMD from baseline were analyzed by means of a repeated-measures model with adjustment for treatment, age category, baseline severe vertebral fracture, visit, treatment-by-visit interaction, and baseline BMD. Percentage changes from baseline in bone turnover were assessed using a Wilcoxon rank-sum test. The safety analysis included cumulated incidence rates of adverse outcomes. Odds ratios and confidence intervals were estimated for serious cardiovascular adverse events with the use of a logistic regression model.

Main results. 2046 participants were randomized to the romosozumab group and 2047 to the alendronate group. A total of 3654 participants from both groups (89.3%) completed 12 months of the trial, and 3150 (77.0%) completed the primary analysis period. The treatment groups were similar in baseline age, ethnicity, and fracture history. The majority of patients in both groups were non-Hispanic (> 60%) and ≥ 75 years old (> 50%). The mean age of the patients was 74.3 years. Baseline mean bone mineral density T scores were –2.96 at the lumbar spine, –2.8 at the total hip, and –2.9 at the femoral neck.

After 24 months of treatment, 6.2% of patients in the romosozumab-alendronate group had a new vertebral fracture as compared to 11.9% in the alendronate-alendronate group. This represents a 48% lower risk (risk ratio 0.52, 95% confidence interval [CI] 0.4–0.66; P < 0.001) of new vertebral fractures with romosozumab. At the time of the primary analysis, romosozumab followed by alendronate resulted in a 27% lower risk of clinical fracture than alendronate alone (hazard ratio 0.73, 95% CI 0.61–0.88; P < 0.001). 8.7% of the romosozumab-alendronate group had a nonvertebral fracture versus 10.6% in the alendronate-alendronate group, representing a 19% lower risk with romosozumab (hazard ratio 0.81, 95% CI 0.66–0.99; P = 0.04). Hip fractures occurred in 2.0% of the romosozumab-alendronate group as compared with 3.2% in the alendronate-alendronate group, representing a 38% lower risk with romosozumab (hazard ratio 0.62, 95% CI 0.42–0.92; P = 0.02).

Patients in the romosozumab-alendronate group had greater gains in BMD from baseline at the lumbar spine (14.9% vs 8.5%) and total hip (7% vs 3.6%) compared to the alendronate-alendronate group. (P < 0.001 for all comparisons). At 12 months, romosozumab treatment resulted in decreased levels of bone resorption marker β-CTX and increased levels of bone formation marker P1NP. β-CTX and P1NP decreased and remained below baseline levels after transitioning to alendronate. In the alendronate-alendronate group, P1NP and β-CTX decreased within 1 month and remained below baseline levels at 36 months.

Overall, the adverse events and serious event rates were similar between the 2 treatment groups during the double-blind period with 2 exceptions. In the first 12 months, injection-site reactions were reported in 4.4% of patients receiving romosozumab compared to 2.6% in those receiving alendronate. Patients in the romosozumab group had an increased incidence of adjudicated serious cardiovascular outcomes during the double-blind period, 2.5% (50 of 2040 patients) compared to 1.9% (38 of 2014 patients) in the alendronate group. During the open-label period, osteonecrosis of the jaw occurred in one patient in each group. Two atypical femoral fractures occurred in the romosozumab-alendronate group, compared to 4 in the alendronate-alendronate group. During the first 18 months of the study, binding anti-romosozumab antibodies were observed in 15.3% of the romosozumab group, with neutralizing antibodies in 0.6%.

Conclusion. In postmenopausal woman with osteoporosis and high fracture risk, 12 months of romosozumab treatment followed by alendronate resulted in significantly lower risk of fracture than use of alendronate alone.

Commentary

Osteoporosis-related fragility fractures carry a substantial risk of morbidity and mortality [1]. The goal of osteoporosis treatment is to ameliorate this risk. The current FDA-approved medications for osteoporosis can be divided into anabolic (teriparatide, abaloparatide) and anti-resorptive (bisphosphonate, denosumab, selective estrogen receptor modulators) categories. Sclerostin is a glycoprotein produced by osteocytes that inhibits the Wnt signaling pathway, thereby impeding osteoblast proliferation and activity. Romosozumab is a monoclonal antisclerostin antibody that results in both increased bone formation and decreased bone resorption [1]. By apparently uncoupling bone formation and resorption to increase bone mass, this medication holds promise to become the ideal osteoporosis drug.

Initial studies have shown that 12 months of romosozumab treatment significantly increased BMD at the lumbar spine (+11.3%), as compared to placebo (–0.1%), alendronate (+4.1%), and teriparatide (+7.1%) [2]. The Fracture Study in Postmenopausal Women with Osteoporosis (FRAME) was a large (7180 patients) randomized controlled trial that demonstrated that 12 months of romosozumab resulted in a 73% lower risk of vertebral fracture and 36% lower risk of clinical fracture compared to placebo [3]. However, there was no significant reduction in non-vertebral facture [3]. This may be due to the fact that FRAME excluded women at the highest risk for fracture. That is, exclusion criteria included history of hip fracture, any severe vertebral facture, or more than 2 moderate vertebral fractures. The current phase 3 ARCH trial (Active-Controlled Fracture Study in Postmenopausal Women with Osteoporosis at High Risk) attempts to clarify the potential benefit of romosozumab treatment in this very high-risk patient population, compared to a common first-line osteoporosis treatment, alendronate.

Indeed, ARCH demonstrates that sequential therapy with romosozumab followed by alendronate is superior to alendronate alone in improving BMD at all sites and preventing new vertebral, clinical, and non-vertebral fractures in postmenopausal women with osteoporosis and a history of fragility fracture. While ARCH was not designed as a cardiovascular outcomes trial, the higher rate of serious cardiovascular adverse events in the romosozumab group raises concern that romosozumab may have a negative effect on vascular tissue. Sclerostin is expressed in vascular smooth muscle [4] and upregulated at sites of vascular calcification [5]. It is possible that inhibiting sclerostin activity could alter vascular remodeling or increase vascular calcification. However, it is interesting that in the larger FRAME trial, no increase in adverse cardiovascular events was seen in the romosozumab group compared to placebo. This may be due to the fact that the average age of patients in FRAME was lower than ARCH. However, it also raises the hypothesis that alendronate itself may be protective in terms of cardiovascular risk. It has been postulated that bisphosphonates may have cardiovascular protective effects, given animal studies have demonstrated that alendronate downregulates monocyte chemoattractant protein 1 and macrophage inflammatory protein 1 [6]. However no cardioprotective benefit was seen in meta-analysis [7].

ARCH has several strengths, including its design as an international, double-blind, and randomized clinical trial. The primary outcome of cumulative fracture incidence is a hard endpoint and is clinically relevant. The intervention is simple and the results are clearly defined. The statistical assessment yields significant results. However, there are some limitations to the study. The lead author has received research support from Amgen and UCB Pharma, the makers of romosuzumab. Amgen and UCB Pharma designed the trial, and Amgen was responsible for trial oversight and data analyses per a pre-specified statistical analysis plan. An external independent data monitoring committee monitored unblinded safety data. Because there was no placebo-controlled arm, it is difficult to determine whether the unexpected cardiovascular signal was due to romosuzumab itself or a protective effect of alendronate. In addition, the majority of study participants were non-Hispanic from Central or Eastern Europe and Latin America, with only ~2% of patients from North America. As a result, ARCH findings may not be generalizable to other regional or ethnic populations. Furthermore, the majority of the patients were ≥ 75 years of age and were at very high fracture risk. It is unclear if younger patients or those with lower risk of fracture would see the same fracture prevention and BMD gain. In addition, because of the relatively short length of the trial, the durability of the metabolic bone benefit and cardiovascular risk is unknown. While the authors reported the increased anti-romosozumab antibodies in the romosozumab group had no detectable effect on efficacy or safety, given the short duration of the trial, this has not been proven.

Applications for Clinical Practice

The dual anti-resorptive and anabolic effect of romosozumab makes it an attractive and promising new osteoporosis therapy. ARCH suggests that sequential therapy with romosuzumab and alendronate is superior in terms of fracture prevention to alendronate alone in elderly postmenopausal women with osteoporosis and a history of fragility fractures, although longer term studies are needed to define the durability of this effect. While the absolute number of serious adjudicated cardiovascular events was low, the increased incidence in the romosuzumab group will likely prevent the FDA from approving this medication for widespread use at this time. Additional studies are needed to clarify the cause and magnitude of this cardiovascular risk and to determine whether prevention of fracture-associated morbidity and mortality is enough to mitigate it.

—Simona Frunza-Stefan, MD, and Hillary B. Whitlach, MD, University of Maryland School of Medicine, Baltimore, MD

Alzheimer disease is the most common form of dementia. In 2016, an estimated 5.2 million Americans age 65 and older had Alzheimer disease. The prevalence is projected to increase to 13.8 million by 2050, including 7 million people age 85 and older.¹

Although no cure for dementia exists, several cognition-enhancing drugs have been approved by the US Food and Drug Administration (FDA) to treat the symptoms of Alzheimer dementia. The purpose of these drugs is to stabilize cognitive and functional status, with a secondary benefit of potentially reducing behavioral problems associated with dementia.

CURRENTLY APPROVED DRUGS

Two classes of drugs are approved to treat Alz­heimer disease: cholinesterase inhibitors and an N-methyl-d-aspartate (NMDA) receptor antagonist (Table 1).

Cholinesterase inhibitors

The cholinesterase inhibitors act by reversibly binding and inactivating acetylcholinesterase, consequently increasing the time the neurotransmitter acetylcholine remains in the synaptic cleft. The 3 FDA-approved cholinesterase inhibitors are donepezil, galantamine, and rivastigmine. Tacrine, the first approved cholinesterase inhibitor, was removed from the US market after reports of severe hepatic toxicity.²

The clinical efficacy of cholinesterase inhibitors in improving cognitive function has been shown in several randomized controlled trials.^3–10 However, benefits were generally modest, and some trials used questionable methodology, leading experts to challenge the overall efficacy of these agents.

All 3 drugs are approved for mild to moderate Alzheimer disease (stages 4–6 on the Global Deterioration Scale; Table 2)^11,12; only donepezil is approved for severe Alzheimer disease. Rivastigmine has an added indication for treating mild to moderate dementia associated with Parkinson disease. Cholinesterase inhibitors are often used off-label to treat other forms of dementia such as vascular dementia, mixed dementia, and dementia with Lewy bodies.¹³

NMDA receptor antagonist

Memantine, currently the only FDA-approved NMDA receptor antagonist, acts by reducing neuronal calcium ion influx and its associated excitation and toxicity. Memantine is approved for moderate to severe Alzheimer disease.

Combination therapy

Often, these 2 classes of medications are prescribed in combination. In a randomized controlled trial that added memantine to stable doses of donepezil, patients had significantly better clinical response on combination therapy than on cholinesterase inhibitor monotherapy.¹⁴

In December 2014, the FDA approved a capsule formulation combining donepezil and memantine to treat symptoms of Alzheimer dementia. However, no novel pharmacologic treatment for Alzheimer disease has been approved since 2003. Furthermore, recently Pfizer announced a plan to eliminate 300 research positions aimed at finding new drugs to treat Alzheimer disease and Parkinson disease.¹⁵

CONSIDERATIONS WHEN STARTING COGNITIVE ENHANCERS

Cholinesterase inhibitors

Adverse effects of cholinesterase inhibitors are generally mild and well tolerated and subside within 1 to 2 weeks. Gastrointestinal effects are common, primarily diarrhea, nausea, and vomiting. They are transient but can occur in about 20% of patients (Table 3).

Other potential adverse effects include bradycardia, syncope, rhabdomyolysis, neuroleptic malignant syndrome, and esophageal rupture. Often, the side-effect profile helps determine which patients are appropriate candidates for these medications.

As expected, higher doses of donepezil (23 mg vs 5–10 mg) are associated with higher rates of nausea, diarrhea, and vomiting.

Dosing. The cholinesterase inhibitors should be slowly titrated to minimize side effects. Starting at the lowest dose and maintaining it for 4 weeks allows sufficient time for transient side effects to abate. Some patients may require a longer titration period.

As the dose is escalated, the probability of side effects may increase. If they do not subside, dose reduction with maintenance at the next lower dose is appropriate.

Gastrointestinal effects. Given the adverse gastrointestinal effects associated with this class of medications, patients experiencing significant anorexia and weight loss should generally avoid cholinesterase inhibitors. However, the rivastigmine patch, a transdermal formulation, is an alternative for patients who experience gastrointestinal side effects.

Bradycardia risk. Patients with significant bradycardia or who are taking medications that lower the heart rate may experience a worsening of their bradycardia or associated symptoms if they take a cholinesterase inhibitor. Syncope from bradycardia is a significant concern, especially in patients already at risk of falls or fracture due to osteoporosis.

NMDA receptor antagonist

The side-effect profile of memantine is generally more favorable than that of cholinesterase inhibitors. In clinical trials, it has been better tolerated with fewer adverse effects than placebo, with the exception of an increased incidence of dizziness, confusion, and delusions.^16,17

Caution is required when treating patients with renal impairment. In patients with a creatinine clearance of 5 to 29 mL/min, the recommended maximum total daily dose is 10 mg (twice-daily formulation) or 14 mg (once-daily formulation).

Off-label use to treat behavioral problems

These medications have been used off-label to treat behavioral problems associated with dementia. A systematic review and meta-analysis showed cholinesterase inhibitor therapy had a statistically significant effect in reducing the severity of behavioral problems.¹⁸ Unfortunately, the number of dropouts increased in the active-treatment groups.

Patients with behavioral problems associated with dementia with Lewy bodies may experience a greater response to cholinesterase inhibitors than those with Alzheimer disease.¹⁹ Published post hoc analyses suggest that patients with moderate to severe Alzheimer disease receiving memantine therapy have less severe agitation, aggression, irritability, and other behavioral disturbances compared with those on placebo.^20,21 However, systematic reviews have not found that memantine has a clinically significant effect on neuropsychiatric symptoms of dementia.^18,22,23

Combination therapy

In early randomized controlled trials, adding memantine to a cholinesterase inhibitor provided additional cognitive benefit in patients with Alzheimer disease.^15,24 However, a more recent randomized controlled trial did not show significant benefits for combined memantine and donepezil vs donepezil alone in moderate to severe dementia.²⁵

In patients who had mild to moderate Alzheimer disease at 14 Veterans Affairs medical centers who were already on cholinesterase inhibitor treatment, adding memantine did not show benefit. However, the group receiving alpha-tocopherol (vitamin E) showed slower functional decline than those on placebo.²⁶ Cognition and function are not expected to improve with memantine.

CONSIDERATIONS WHEN STOPPING COGNITIVE ENHANCERS

The cholinesterase inhibitors are usually prescribed early in the course of dementia, and some patients take these drugs for years, although no studies have investigated benefit or risk beyond 1 year. It is generally recommended that cholinesterase inhibitor therapy be assessed periodically, eg, every 3 to 6 months, for perceived cognitive benefits and adverse gastrointestinal effects.

These medications should be stopped if the desired effects—stabilizing cognitive and functional status—are not perceived within a reasonable time, such as 12 weeks. In some cases, stopping cholinesterase inhibitor therapy may cause negative effects on cognition and neuropsychiatric symptoms.²⁷

Deciding whether benefit has occurred during a trial of cholinesterase inhibitors often requires input and observations from the family and caregivers. Soliciting this information is key for practitioners to determine the correct treatment approach for each patient.

Although some patients with moderately severe disease experience clinical benefits from cholinesterase inhibitor therapy, it is reasonable to consider discontinuing therapy when a patient has progressed to advanced dementia with loss of functional independence, thus making the use of the therapy—ie, to preserve functional status—less relevant. Results from a randomized discontinuation trial of cholinesterase inhibitors in institutionalized patients with moderate to severe dementia suggest that discontinuation is safe and well tolerated in most of these patients.²⁸

Abruptly stopping high-dose cholinesterase inhibitors is not recommended. Most clinical trials tapered these medications over 2 to 4 weeks. Patients taking the maximum dose of a cholinesterase inhibitor should have the dose reduced to the next lowest dose for 2 weeks before the dose is reduced further or stopped completely.

Behavioral and psychiatric problems often accompany dementia; however, no drugs are approved to treat these symptoms in patients with Alzheimer disease. Nonpharmacologic interventions are recommended as the initial treatment.²⁹ Some practitioners prescribe psychotropic drugs off-label for Alzheimer disease, but most clinical trials have not found these therapies to be very effective for psychiatric symptoms associated with Alzheimer disease.^30,31

Recently, a randomized controlled trial of dextromethorphan-quinidine showed mild reduction in agitation in patients with Alzheimer disease, but there were significant increases in falls, dizziness, and diarrhea.³²

Patients prescribed medications for behavioral and psychological symptoms of dementia should be assessed every 3 to 6 months to determine if the medications have been effective in reducing the symptoms they were meant to reduce. If there has been no clear reduction in the target behaviors, a trial off the drug should be initiated, with careful monitoring to see if the target behavior changes. Dementia-related behaviors may worsen off the medication, but a lower dose may be found to be as effective as a higher dose. As dementia advances, behaviors initially encountered during one stage may diminish or abate.

In a long-term care setting, a gradual dose-reduction trial of psychotropic medications should be conducted every year to determine if the medications are still necessary.³³ This should be considered during routine management and follow-up of patients with dementia-associated behavioral problems.

REASONABLE TO TRY

Cognitive enhancers have been around for more than 10 years and are reasonable to try in patients with Alzheimer disease. All the available drugs are FDA-approved for reducing dementia symptoms associated with mild to moderate Alzheimer disease; donepezil and memantine are also approved for severe Alz­heimer disease, either in combination or as monotherapy.

When selecting a cognitive enhancer, practitioners need to consider the potential for adverse effects. And if a cholinesterase inhibitor is prescribed, it is important to periodically assess for perceived cognitive benefits and adverse gastrointestinal effects. The NMDA receptor antagonist has a more favorable side effect profile. Combining the drugs is also an option.

Similarly, patients prescribed psychotropic medications for behavioral problems related to dementia should be reassessed to determine if the dose could be reduced or eliminated, particularly if targeted behaviors have not responded to the treatment or the dementia has advanced.

For patients on cognitive enhancers, discontinuation should be considered when the dementia advances to the point where the patient is totally dependent for all basic activities of daily living, and the initial intended purpose of these medications—preservation of cognitive and functional status—is no longer achievable.

Alzheimer disease is the most common form of dementia. In 2016, an estimated 5.2 million Americans age 65 and older had Alzheimer disease. The prevalence is projected to increase to 13.8 million by 2050, including 7 million people age 85 and older.¹

Although no cure for dementia exists, several cognition-enhancing drugs have been approved by the US Food and Drug Administration (FDA) to treat the symptoms of Alzheimer dementia. The purpose of these drugs is to stabilize cognitive and functional status, with a secondary benefit of potentially reducing behavioral problems associated with dementia.

CURRENTLY APPROVED DRUGS

Two classes of drugs are approved to treat Alz­heimer disease: cholinesterase inhibitors and an N-methyl-d-aspartate (NMDA) receptor antagonist (Table 1).

Cholinesterase inhibitors

The cholinesterase inhibitors act by reversibly binding and inactivating acetylcholinesterase, consequently increasing the time the neurotransmitter acetylcholine remains in the synaptic cleft. The 3 FDA-approved cholinesterase inhibitors are donepezil, galantamine, and rivastigmine. Tacrine, the first approved cholinesterase inhibitor, was removed from the US market after reports of severe hepatic toxicity.²

The clinical efficacy of cholinesterase inhibitors in improving cognitive function has been shown in several randomized controlled trials.^3–10 However, benefits were generally modest, and some trials used questionable methodology, leading experts to challenge the overall efficacy of these agents.

All 3 drugs are approved for mild to moderate Alzheimer disease (stages 4–6 on the Global Deterioration Scale; Table 2)^11,12; only donepezil is approved for severe Alzheimer disease. Rivastigmine has an added indication for treating mild to moderate dementia associated with Parkinson disease. Cholinesterase inhibitors are often used off-label to treat other forms of dementia such as vascular dementia, mixed dementia, and dementia with Lewy bodies.¹³

NMDA receptor antagonist

Memantine, currently the only FDA-approved NMDA receptor antagonist, acts by reducing neuronal calcium ion influx and its associated excitation and toxicity. Memantine is approved for moderate to severe Alzheimer disease.

Combination therapy

Often, these 2 classes of medications are prescribed in combination. In a randomized controlled trial that added memantine to stable doses of donepezil, patients had significantly better clinical response on combination therapy than on cholinesterase inhibitor monotherapy.¹⁴

In December 2014, the FDA approved a capsule formulation combining donepezil and memantine to treat symptoms of Alzheimer dementia. However, no novel pharmacologic treatment for Alzheimer disease has been approved since 2003. Furthermore, recently Pfizer announced a plan to eliminate 300 research positions aimed at finding new drugs to treat Alzheimer disease and Parkinson disease.¹⁵

CONSIDERATIONS WHEN STARTING COGNITIVE ENHANCERS

Cholinesterase inhibitors

Adverse effects of cholinesterase inhibitors are generally mild and well tolerated and subside within 1 to 2 weeks. Gastrointestinal effects are common, primarily diarrhea, nausea, and vomiting. They are transient but can occur in about 20% of patients (Table 3).

Other potential adverse effects include bradycardia, syncope, rhabdomyolysis, neuroleptic malignant syndrome, and esophageal rupture. Often, the side-effect profile helps determine which patients are appropriate candidates for these medications.

As expected, higher doses of donepezil (23 mg vs 5–10 mg) are associated with higher rates of nausea, diarrhea, and vomiting.

Dosing. The cholinesterase inhibitors should be slowly titrated to minimize side effects. Starting at the lowest dose and maintaining it for 4 weeks allows sufficient time for transient side effects to abate. Some patients may require a longer titration period.

As the dose is escalated, the probability of side effects may increase. If they do not subside, dose reduction with maintenance at the next lower dose is appropriate.

Gastrointestinal effects. Given the adverse gastrointestinal effects associated with this class of medications, patients experiencing significant anorexia and weight loss should generally avoid cholinesterase inhibitors. However, the rivastigmine patch, a transdermal formulation, is an alternative for patients who experience gastrointestinal side effects.

Bradycardia risk. Patients with significant bradycardia or who are taking medications that lower the heart rate may experience a worsening of their bradycardia or associated symptoms if they take a cholinesterase inhibitor. Syncope from bradycardia is a significant concern, especially in patients already at risk of falls or fracture due to osteoporosis.

NMDA receptor antagonist

The side-effect profile of memantine is generally more favorable than that of cholinesterase inhibitors. In clinical trials, it has been better tolerated with fewer adverse effects than placebo, with the exception of an increased incidence of dizziness, confusion, and delusions.^16,17

Caution is required when treating patients with renal impairment. In patients with a creatinine clearance of 5 to 29 mL/min, the recommended maximum total daily dose is 10 mg (twice-daily formulation) or 14 mg (once-daily formulation).

Off-label use to treat behavioral problems

These medications have been used off-label to treat behavioral problems associated with dementia. A systematic review and meta-analysis showed cholinesterase inhibitor therapy had a statistically significant effect in reducing the severity of behavioral problems.¹⁸ Unfortunately, the number of dropouts increased in the active-treatment groups.

Patients with behavioral problems associated with dementia with Lewy bodies may experience a greater response to cholinesterase inhibitors than those with Alzheimer disease.¹⁹ Published post hoc analyses suggest that patients with moderate to severe Alzheimer disease receiving memantine therapy have less severe agitation, aggression, irritability, and other behavioral disturbances compared with those on placebo.^20,21 However, systematic reviews have not found that memantine has a clinically significant effect on neuropsychiatric symptoms of dementia.^18,22,23

Combination therapy

In early randomized controlled trials, adding memantine to a cholinesterase inhibitor provided additional cognitive benefit in patients with Alzheimer disease.^15,24 However, a more recent randomized controlled trial did not show significant benefits for combined memantine and donepezil vs donepezil alone in moderate to severe dementia.²⁵

In patients who had mild to moderate Alzheimer disease at 14 Veterans Affairs medical centers who were already on cholinesterase inhibitor treatment, adding memantine did not show benefit. However, the group receiving alpha-tocopherol (vitamin E) showed slower functional decline than those on placebo.²⁶ Cognition and function are not expected to improve with memantine.

CONSIDERATIONS WHEN STOPPING COGNITIVE ENHANCERS

The cholinesterase inhibitors are usually prescribed early in the course of dementia, and some patients take these drugs for years, although no studies have investigated benefit or risk beyond 1 year. It is generally recommended that cholinesterase inhibitor therapy be assessed periodically, eg, every 3 to 6 months, for perceived cognitive benefits and adverse gastrointestinal effects.

These medications should be stopped if the desired effects—stabilizing cognitive and functional status—are not perceived within a reasonable time, such as 12 weeks. In some cases, stopping cholinesterase inhibitor therapy may cause negative effects on cognition and neuropsychiatric symptoms.²⁷

Deciding whether benefit has occurred during a trial of cholinesterase inhibitors often requires input and observations from the family and caregivers. Soliciting this information is key for practitioners to determine the correct treatment approach for each patient.

Although some patients with moderately severe disease experience clinical benefits from cholinesterase inhibitor therapy, it is reasonable to consider discontinuing therapy when a patient has progressed to advanced dementia with loss of functional independence, thus making the use of the therapy—ie, to preserve functional status—less relevant. Results from a randomized discontinuation trial of cholinesterase inhibitors in institutionalized patients with moderate to severe dementia suggest that discontinuation is safe and well tolerated in most of these patients.²⁸

Abruptly stopping high-dose cholinesterase inhibitors is not recommended. Most clinical trials tapered these medications over 2 to 4 weeks. Patients taking the maximum dose of a cholinesterase inhibitor should have the dose reduced to the next lowest dose for 2 weeks before the dose is reduced further or stopped completely.

Behavioral and psychiatric problems often accompany dementia; however, no drugs are approved to treat these symptoms in patients with Alzheimer disease. Nonpharmacologic interventions are recommended as the initial treatment.²⁹ Some practitioners prescribe psychotropic drugs off-label for Alzheimer disease, but most clinical trials have not found these therapies to be very effective for psychiatric symptoms associated with Alzheimer disease.^30,31

Recently, a randomized controlled trial of dextromethorphan-quinidine showed mild reduction in agitation in patients with Alzheimer disease, but there were significant increases in falls, dizziness, and diarrhea.³²

Patients prescribed medications for behavioral and psychological symptoms of dementia should be assessed every 3 to 6 months to determine if the medications have been effective in reducing the symptoms they were meant to reduce. If there has been no clear reduction in the target behaviors, a trial off the drug should be initiated, with careful monitoring to see if the target behavior changes. Dementia-related behaviors may worsen off the medication, but a lower dose may be found to be as effective as a higher dose. As dementia advances, behaviors initially encountered during one stage may diminish or abate.

In a long-term care setting, a gradual dose-reduction trial of psychotropic medications should be conducted every year to determine if the medications are still necessary.³³ This should be considered during routine management and follow-up of patients with dementia-associated behavioral problems.

REASONABLE TO TRY

Cognitive enhancers have been around for more than 10 years and are reasonable to try in patients with Alzheimer disease. All the available drugs are FDA-approved for reducing dementia symptoms associated with mild to moderate Alzheimer disease; donepezil and memantine are also approved for severe Alz­heimer disease, either in combination or as monotherapy.

When selecting a cognitive enhancer, practitioners need to consider the potential for adverse effects. And if a cholinesterase inhibitor is prescribed, it is important to periodically assess for perceived cognitive benefits and adverse gastrointestinal effects. The NMDA receptor antagonist has a more favorable side effect profile. Combining the drugs is also an option.

Similarly, patients prescribed psychotropic medications for behavioral problems related to dementia should be reassessed to determine if the dose could be reduced or eliminated, particularly if targeted behaviors have not responded to the treatment or the dementia has advanced.

For patients on cognitive enhancers, discontinuation should be considered when the dementia advances to the point where the patient is totally dependent for all basic activities of daily living, and the initial intended purpose of these medications—preservation of cognitive and functional status—is no longer achievable.

Alzheimer disease is the most common form of dementia. In 2016, an estimated 5.2 million Americans age 65 and older had Alzheimer disease. The prevalence is projected to increase to 13.8 million by 2050, including 7 million people age 85 and older.¹

Although no cure for dementia exists, several cognition-enhancing drugs have been approved by the US Food and Drug Administration (FDA) to treat the symptoms of Alzheimer dementia. The purpose of these drugs is to stabilize cognitive and functional status, with a secondary benefit of potentially reducing behavioral problems associated with dementia.

CURRENTLY APPROVED DRUGS

Two classes of drugs are approved to treat Alz­heimer disease: cholinesterase inhibitors and an N-methyl-d-aspartate (NMDA) receptor antagonist (Table 1).

Cholinesterase inhibitors

The cholinesterase inhibitors act by reversibly binding and inactivating acetylcholinesterase, consequently increasing the time the neurotransmitter acetylcholine remains in the synaptic cleft. The 3 FDA-approved cholinesterase inhibitors are donepezil, galantamine, and rivastigmine. Tacrine, the first approved cholinesterase inhibitor, was removed from the US market after reports of severe hepatic toxicity.²

The clinical efficacy of cholinesterase inhibitors in improving cognitive function has been shown in several randomized controlled trials.^3–10 However, benefits were generally modest, and some trials used questionable methodology, leading experts to challenge the overall efficacy of these agents.

All 3 drugs are approved for mild to moderate Alzheimer disease (stages 4–6 on the Global Deterioration Scale; Table 2)^11,12; only donepezil is approved for severe Alzheimer disease. Rivastigmine has an added indication for treating mild to moderate dementia associated with Parkinson disease. Cholinesterase inhibitors are often used off-label to treat other forms of dementia such as vascular dementia, mixed dementia, and dementia with Lewy bodies.¹³

NMDA receptor antagonist

Memantine, currently the only FDA-approved NMDA receptor antagonist, acts by reducing neuronal calcium ion influx and its associated excitation and toxicity. Memantine is approved for moderate to severe Alzheimer disease.

Combination therapy

Often, these 2 classes of medications are prescribed in combination. In a randomized controlled trial that added memantine to stable doses of donepezil, patients had significantly better clinical response on combination therapy than on cholinesterase inhibitor monotherapy.¹⁴

In December 2014, the FDA approved a capsule formulation combining donepezil and memantine to treat symptoms of Alzheimer dementia. However, no novel pharmacologic treatment for Alzheimer disease has been approved since 2003. Furthermore, recently Pfizer announced a plan to eliminate 300 research positions aimed at finding new drugs to treat Alzheimer disease and Parkinson disease.¹⁵

CONSIDERATIONS WHEN STARTING COGNITIVE ENHANCERS

Cholinesterase inhibitors

Adverse effects of cholinesterase inhibitors are generally mild and well tolerated and subside within 1 to 2 weeks. Gastrointestinal effects are common, primarily diarrhea, nausea, and vomiting. They are transient but can occur in about 20% of patients (Table 3).

Other potential adverse effects include bradycardia, syncope, rhabdomyolysis, neuroleptic malignant syndrome, and esophageal rupture. Often, the side-effect profile helps determine which patients are appropriate candidates for these medications.

As expected, higher doses of donepezil (23 mg vs 5–10 mg) are associated with higher rates of nausea, diarrhea, and vomiting.

Dosing. The cholinesterase inhibitors should be slowly titrated to minimize side effects. Starting at the lowest dose and maintaining it for 4 weeks allows sufficient time for transient side effects to abate. Some patients may require a longer titration period.

As the dose is escalated, the probability of side effects may increase. If they do not subside, dose reduction with maintenance at the next lower dose is appropriate.

Gastrointestinal effects. Given the adverse gastrointestinal effects associated with this class of medications, patients experiencing significant anorexia and weight loss should generally avoid cholinesterase inhibitors. However, the rivastigmine patch, a transdermal formulation, is an alternative for patients who experience gastrointestinal side effects.

Bradycardia risk. Patients with significant bradycardia or who are taking medications that lower the heart rate may experience a worsening of their bradycardia or associated symptoms if they take a cholinesterase inhibitor. Syncope from bradycardia is a significant concern, especially in patients already at risk of falls or fracture due to osteoporosis.

NMDA receptor antagonist

The side-effect profile of memantine is generally more favorable than that of cholinesterase inhibitors. In clinical trials, it has been better tolerated with fewer adverse effects than placebo, with the exception of an increased incidence of dizziness, confusion, and delusions.^16,17

Caution is required when treating patients with renal impairment. In patients with a creatinine clearance of 5 to 29 mL/min, the recommended maximum total daily dose is 10 mg (twice-daily formulation) or 14 mg (once-daily formulation).

Off-label use to treat behavioral problems

These medications have been used off-label to treat behavioral problems associated with dementia. A systematic review and meta-analysis showed cholinesterase inhibitor therapy had a statistically significant effect in reducing the severity of behavioral problems.¹⁸ Unfortunately, the number of dropouts increased in the active-treatment groups.

Patients with behavioral problems associated with dementia with Lewy bodies may experience a greater response to cholinesterase inhibitors than those with Alzheimer disease.¹⁹ Published post hoc analyses suggest that patients with moderate to severe Alzheimer disease receiving memantine therapy have less severe agitation, aggression, irritability, and other behavioral disturbances compared with those on placebo.^20,21 However, systematic reviews have not found that memantine has a clinically significant effect on neuropsychiatric symptoms of dementia.^18,22,23

Combination therapy

In early randomized controlled trials, adding memantine to a cholinesterase inhibitor provided additional cognitive benefit in patients with Alzheimer disease.^15,24 However, a more recent randomized controlled trial did not show significant benefits for combined memantine and donepezil vs donepezil alone in moderate to severe dementia.²⁵

In patients who had mild to moderate Alzheimer disease at 14 Veterans Affairs medical centers who were already on cholinesterase inhibitor treatment, adding memantine did not show benefit. However, the group receiving alpha-tocopherol (vitamin E) showed slower functional decline than those on placebo.²⁶ Cognition and function are not expected to improve with memantine.

CONSIDERATIONS WHEN STOPPING COGNITIVE ENHANCERS

The cholinesterase inhibitors are usually prescribed early in the course of dementia, and some patients take these drugs for years, although no studies have investigated benefit or risk beyond 1 year. It is generally recommended that cholinesterase inhibitor therapy be assessed periodically, eg, every 3 to 6 months, for perceived cognitive benefits and adverse gastrointestinal effects.

These medications should be stopped if the desired effects—stabilizing cognitive and functional status—are not perceived within a reasonable time, such as 12 weeks. In some cases, stopping cholinesterase inhibitor therapy may cause negative effects on cognition and neuropsychiatric symptoms.²⁷

Deciding whether benefit has occurred during a trial of cholinesterase inhibitors often requires input and observations from the family and caregivers. Soliciting this information is key for practitioners to determine the correct treatment approach for each patient.

Although some patients with moderately severe disease experience clinical benefits from cholinesterase inhibitor therapy, it is reasonable to consider discontinuing therapy when a patient has progressed to advanced dementia with loss of functional independence, thus making the use of the therapy—ie, to preserve functional status—less relevant. Results from a randomized discontinuation trial of cholinesterase inhibitors in institutionalized patients with moderate to severe dementia suggest that discontinuation is safe and well tolerated in most of these patients.²⁸

Abruptly stopping high-dose cholinesterase inhibitors is not recommended. Most clinical trials tapered these medications over 2 to 4 weeks. Patients taking the maximum dose of a cholinesterase inhibitor should have the dose reduced to the next lowest dose for 2 weeks before the dose is reduced further or stopped completely.

Behavioral and psychiatric problems often accompany dementia; however, no drugs are approved to treat these symptoms in patients with Alzheimer disease. Nonpharmacologic interventions are recommended as the initial treatment.²⁹ Some practitioners prescribe psychotropic drugs off-label for Alzheimer disease, but most clinical trials have not found these therapies to be very effective for psychiatric symptoms associated with Alzheimer disease.^30,31

Recently, a randomized controlled trial of dextromethorphan-quinidine showed mild reduction in agitation in patients with Alzheimer disease, but there were significant increases in falls, dizziness, and diarrhea.³²

Patients prescribed medications for behavioral and psychological symptoms of dementia should be assessed every 3 to 6 months to determine if the medications have been effective in reducing the symptoms they were meant to reduce. If there has been no clear reduction in the target behaviors, a trial off the drug should be initiated, with careful monitoring to see if the target behavior changes. Dementia-related behaviors may worsen off the medication, but a lower dose may be found to be as effective as a higher dose. As dementia advances, behaviors initially encountered during one stage may diminish or abate.

In a long-term care setting, a gradual dose-reduction trial of psychotropic medications should be conducted every year to determine if the medications are still necessary.³³ This should be considered during routine management and follow-up of patients with dementia-associated behavioral problems.

REASONABLE TO TRY

Cognitive enhancers have been around for more than 10 years and are reasonable to try in patients with Alzheimer disease. All the available drugs are FDA-approved for reducing dementia symptoms associated with mild to moderate Alzheimer disease; donepezil and memantine are also approved for severe Alz­heimer disease, either in combination or as monotherapy.

When selecting a cognitive enhancer, practitioners need to consider the potential for adverse effects. And if a cholinesterase inhibitor is prescribed, it is important to periodically assess for perceived cognitive benefits and adverse gastrointestinal effects. The NMDA receptor antagonist has a more favorable side effect profile. Combining the drugs is also an option.

Similarly, patients prescribed psychotropic medications for behavioral problems related to dementia should be reassessed to determine if the dose could be reduced or eliminated, particularly if targeted behaviors have not responded to the treatment or the dementia has advanced.

For patients on cognitive enhancers, discontinuation should be considered when the dementia advances to the point where the patient is totally dependent for all basic activities of daily living, and the initial intended purpose of these medications—preservation of cognitive and functional status—is no longer achievable.

A 45-year-old man with known human immunodeficiency virus infection presented with a 5-day history of dyspnea. When his dyspnea had become symptomatic, he had restarted his home dapsone prophylaxis, but his dyspnea had progressively worsened, and his urine became dark.

Based on these test results, the patient’s dapsone was stopped and replaced with atovaquone. Intravenous infusion of methylene blue was started, with subsequent improvement of the hypoxia and cyanosis (Figure 1). His urine became green, but it returned to a normal color in a matter of hours. He was ultimately diagnosed with P jirovecii pneumonia and completed a course of atovaquone with total resolution of his symptoms.

THE MECHANISMS BEHIND METHEMOGLOBINEMIA

Heme iron is normally in the ferrous state (Fe²⁺), which allows for hemoglobin to carry oxygen and release it to tissues.¹ Exposure to an oxidative stress can lead to methemoglobinemia from an increase in abnormal hemoglobin that contains iron in a ferric state (Fe³⁺).^1,2

Methemoglobin reduces oxygen-carrying capacity in two ways: it is unable to carry oxygen, and its presence shifts the oxy­gen dissociation curve to the left, causing any remaining normal hemoglobin to be unable to release oxygen to the tissues.^1,2

Causes of acquired methemoglobinemia include topical anesthetics (eg, benzocaine, lidocaine) and antibiotics (eg, dapsone).^2,3 Signs and symptoms include cyanosis, headache, fatigue, dyspnea, lethargy, respiratory distress, and dark-colored urine.^1,2

MANAGEMENT

Treatment consists of intravenous methylene blue, which reduces the hemoglobin from a ferric state to a ferrous state.^1–4 Methylene blue is a water-soluble dye excreted primarily in the urine, and common side effects include dizziness, nausea, and green urine.^5–7 The blue pigments from methylene blue combine with urobilin (a yellow pigment in the urine), producing a green color.⁷ This is not pathological and requires no treatment, as the urine returns to normal color after the body fully excretes the dye.^5–7

If intravenous methylene blue fails to produce a response, other treatments to consider include hemodialysis, blood transfusion, exchange transfusion, and hyperbaric oxygen therapy.²

A 45-year-old man with known human immunodeficiency virus infection presented with a 5-day history of dyspnea. When his dyspnea had become symptomatic, he had restarted his home dapsone prophylaxis, but his dyspnea had progressively worsened, and his urine became dark.

Based on these test results, the patient’s dapsone was stopped and replaced with atovaquone. Intravenous infusion of methylene blue was started, with subsequent improvement of the hypoxia and cyanosis (Figure 1). His urine became green, but it returned to a normal color in a matter of hours. He was ultimately diagnosed with P jirovecii pneumonia and completed a course of atovaquone with total resolution of his symptoms.

THE MECHANISMS BEHIND METHEMOGLOBINEMIA

Heme iron is normally in the ferrous state (Fe²⁺), which allows for hemoglobin to carry oxygen and release it to tissues.¹ Exposure to an oxidative stress can lead to methemoglobinemia from an increase in abnormal hemoglobin that contains iron in a ferric state (Fe³⁺).^1,2

Methemoglobin reduces oxygen-carrying capacity in two ways: it is unable to carry oxygen, and its presence shifts the oxy­gen dissociation curve to the left, causing any remaining normal hemoglobin to be unable to release oxygen to the tissues.^1,2

Causes of acquired methemoglobinemia include topical anesthetics (eg, benzocaine, lidocaine) and antibiotics (eg, dapsone).^2,3 Signs and symptoms include cyanosis, headache, fatigue, dyspnea, lethargy, respiratory distress, and dark-colored urine.^1,2

MANAGEMENT

Treatment consists of intravenous methylene blue, which reduces the hemoglobin from a ferric state to a ferrous state.^1–4 Methylene blue is a water-soluble dye excreted primarily in the urine, and common side effects include dizziness, nausea, and green urine.^5–7 The blue pigments from methylene blue combine with urobilin (a yellow pigment in the urine), producing a green color.⁷ This is not pathological and requires no treatment, as the urine returns to normal color after the body fully excretes the dye.^5–7

If intravenous methylene blue fails to produce a response, other treatments to consider include hemodialysis, blood transfusion, exchange transfusion, and hyperbaric oxygen therapy.²

A 45-year-old man with known human immunodeficiency virus infection presented with a 5-day history of dyspnea. When his dyspnea had become symptomatic, he had restarted his home dapsone prophylaxis, but his dyspnea had progressively worsened, and his urine became dark.

Based on these test results, the patient’s dapsone was stopped and replaced with atovaquone. Intravenous infusion of methylene blue was started, with subsequent improvement of the hypoxia and cyanosis (Figure 1). His urine became green, but it returned to a normal color in a matter of hours. He was ultimately diagnosed with P jirovecii pneumonia and completed a course of atovaquone with total resolution of his symptoms.

THE MECHANISMS BEHIND METHEMOGLOBINEMIA

Heme iron is normally in the ferrous state (Fe²⁺), which allows for hemoglobin to carry oxygen and release it to tissues.¹ Exposure to an oxidative stress can lead to methemoglobinemia from an increase in abnormal hemoglobin that contains iron in a ferric state (Fe³⁺).^1,2

Methemoglobin reduces oxygen-carrying capacity in two ways: it is unable to carry oxygen, and its presence shifts the oxy­gen dissociation curve to the left, causing any remaining normal hemoglobin to be unable to release oxygen to the tissues.^1,2

Causes of acquired methemoglobinemia include topical anesthetics (eg, benzocaine, lidocaine) and antibiotics (eg, dapsone).^2,3 Signs and symptoms include cyanosis, headache, fatigue, dyspnea, lethargy, respiratory distress, and dark-colored urine.^1,2

MANAGEMENT

Treatment consists of intravenous methylene blue, which reduces the hemoglobin from a ferric state to a ferrous state.^1–4 Methylene blue is a water-soluble dye excreted primarily in the urine, and common side effects include dizziness, nausea, and green urine.^5–7 The blue pigments from methylene blue combine with urobilin (a yellow pigment in the urine), producing a green color.⁷ This is not pathological and requires no treatment, as the urine returns to normal color after the body fully excretes the dye.^5–7

If intravenous methylene blue fails to produce a response, other treatments to consider include hemodialysis, blood transfusion, exchange transfusion, and hyperbaric oxygen therapy.²

Sometimes the answer is straightforward—the pills were started to reduce knee pain, but the pain is still there. But sometimes if I suggest stopping a drug, I get surprising pushback from the patient: “But the pain may be worse without those pills.” And it can be hard to assess whether a medication has attained its therapeutic goal, such as when a drug is given to prevent or reduce the occurrence of intermittent events (eg, hydroxychloroquine to reduce flares in lupus, aspirin to prevent transient ischemic attacks). Unless a medication has been given sufficient time to have its effect and has clearly failed, we often have to trust its efficacy because those events might be more frequent if the drug were stopped.

In this issue of the Journal, Kim and Factora discuss a difficult scenario—discontinuing cognitive-enhancing therapy in patients with Alzheimer dementia. The stakes, hopes, and anxiety are high for the patient and caregivers. These drugs have only modest efficacy, and it is often difficult to know if they are working. To complicate matters, dementia is progressive, but the rate of progression differs among individuals, making it harder to be convinced that the drugs have lost their efficacy and that discontinuation is warranted in order to reduce the side effects, cost, and pill burden. Similar challenges arise for similar reasons when considering discontinuation of antipsychotics or other drugs given for behavioral reasons to elderly patients with dementia.¹

The issues surrounding downsizing medication lists—or deprescribing—extend far beyond patients with Alzheimer dementia. A disease may have progressed beyond the point where the drug can make a significant impact. Patients often develop tachyphylaxis to the newer protein drugs (biologics for rheumatoid arthritis, inflammatory bowel disease, psoriasis, or gout) due to the generation of neutralizing antidrug antibodies. At some point the patient’s life expectancy becomes a factor when considering medications directed at preventing long-term complications of a disease and the increased likelihood of significant adverse effects in patients as they age (eg, from aggressively prescribed antihypertensive and antidiabetic drugs). For drugs such as bisphosphonates, alkylating agents, and metoclopramide, complications of cumulative dosing over time should be considered a reason to discontinue therapy.

The take-home message from this article is to create opportunities to periodically revisit the rationale for all of a patient’s prescriptions, and to make sure patients are comfortable knowing why they should keep taking each of their medications. While revisiting a patient’s prescriptions may indeed reduce the medication burden, I believe it also enhances adherence to the remaining prescribed medications. The medication reconciliation process requires more than simply checking off the box in the EMR to indicate that the medications were “reviewed.”

Sometimes the answer is straightforward—the pills were started to reduce knee pain, but the pain is still there. But sometimes if I suggest stopping a drug, I get surprising pushback from the patient: “But the pain may be worse without those pills.” And it can be hard to assess whether a medication has attained its therapeutic goal, such as when a drug is given to prevent or reduce the occurrence of intermittent events (eg, hydroxychloroquine to reduce flares in lupus, aspirin to prevent transient ischemic attacks). Unless a medication has been given sufficient time to have its effect and has clearly failed, we often have to trust its efficacy because those events might be more frequent if the drug were stopped.

In this issue of the Journal, Kim and Factora discuss a difficult scenario—discontinuing cognitive-enhancing therapy in patients with Alzheimer dementia. The stakes, hopes, and anxiety are high for the patient and caregivers. These drugs have only modest efficacy, and it is often difficult to know if they are working. To complicate matters, dementia is progressive, but the rate of progression differs among individuals, making it harder to be convinced that the drugs have lost their efficacy and that discontinuation is warranted in order to reduce the side effects, cost, and pill burden. Similar challenges arise for similar reasons when considering discontinuation of antipsychotics or other drugs given for behavioral reasons to elderly patients with dementia.¹

The issues surrounding downsizing medication lists—or deprescribing—extend far beyond patients with Alzheimer dementia. A disease may have progressed beyond the point where the drug can make a significant impact. Patients often develop tachyphylaxis to the newer protein drugs (biologics for rheumatoid arthritis, inflammatory bowel disease, psoriasis, or gout) due to the generation of neutralizing antidrug antibodies. At some point the patient’s life expectancy becomes a factor when considering medications directed at preventing long-term complications of a disease and the increased likelihood of significant adverse effects in patients as they age (eg, from aggressively prescribed antihypertensive and antidiabetic drugs). For drugs such as bisphosphonates, alkylating agents, and metoclopramide, complications of cumulative dosing over time should be considered a reason to discontinue therapy.

The take-home message from this article is to create opportunities to periodically revisit the rationale for all of a patient’s prescriptions, and to make sure patients are comfortable knowing why they should keep taking each of their medications. While revisiting a patient’s prescriptions may indeed reduce the medication burden, I believe it also enhances adherence to the remaining prescribed medications. The medication reconciliation process requires more than simply checking off the box in the EMR to indicate that the medications were “reviewed.”

Sometimes the answer is straightforward—the pills were started to reduce knee pain, but the pain is still there. But sometimes if I suggest stopping a drug, I get surprising pushback from the patient: “But the pain may be worse without those pills.” And it can be hard to assess whether a medication has attained its therapeutic goal, such as when a drug is given to prevent or reduce the occurrence of intermittent events (eg, hydroxychloroquine to reduce flares in lupus, aspirin to prevent transient ischemic attacks). Unless a medication has been given sufficient time to have its effect and has clearly failed, we often have to trust its efficacy because those events might be more frequent if the drug were stopped.

In this issue of the Journal, Kim and Factora discuss a difficult scenario—discontinuing cognitive-enhancing therapy in patients with Alzheimer dementia. The stakes, hopes, and anxiety are high for the patient and caregivers. These drugs have only modest efficacy, and it is often difficult to know if they are working. To complicate matters, dementia is progressive, but the rate of progression differs among individuals, making it harder to be convinced that the drugs have lost their efficacy and that discontinuation is warranted in order to reduce the side effects, cost, and pill burden. Similar challenges arise for similar reasons when considering discontinuation of antipsychotics or other drugs given for behavioral reasons to elderly patients with dementia.¹

The issues surrounding downsizing medication lists—or deprescribing—extend far beyond patients with Alzheimer dementia. A disease may have progressed beyond the point where the drug can make a significant impact. Patients often develop tachyphylaxis to the newer protein drugs (biologics for rheumatoid arthritis, inflammatory bowel disease, psoriasis, or gout) due to the generation of neutralizing antidrug antibodies. At some point the patient’s life expectancy becomes a factor when considering medications directed at preventing long-term complications of a disease and the increased likelihood of significant adverse effects in patients as they age (eg, from aggressively prescribed antihypertensive and antidiabetic drugs). For drugs such as bisphosphonates, alkylating agents, and metoclopramide, complications of cumulative dosing over time should be considered a reason to discontinue therapy.

The take-home message from this article is to create opportunities to periodically revisit the rationale for all of a patient’s prescriptions, and to make sure patients are comfortable knowing why they should keep taking each of their medications. While revisiting a patient’s prescriptions may indeed reduce the medication burden, I believe it also enhances adherence to the remaining prescribed medications. The medication reconciliation process requires more than simply checking off the box in the EMR to indicate that the medications were “reviewed.”

Medications started for appropriate indications in middle age may need to be monitored more closely as the patient ages. Some drugs may become unnecessary or even dangerous as the patient ages, functional status and renal function decline, and goals of care change.

See related editorial

Older adults tend to have multiple illnesses and therefore take more drugs, and polypharmacy increases the risk of poor outcomes. The number of medications a person uses is a risk factor for adverse drug reactions, nonadherence, financial burden, drug-drug interactions, and worse outcomes.¹

The prevalence of polypharmacy increased from an estimated 8.2% to 15% from 1999 to 2011 based on the National Health and Nutrition Examination Survey.² Guideline-based therapy for specific diseases may lead to the addition of more medications to reach disease targets.³ Most older adults in the United States compound the risk of prescribed medications by also taking over-the-counter medications and dietary supplements.⁴

In addition, medications are often used in older adults based on studies of younger persons without significant comorbidities. Applying clinical guidelines based on these studies to older adults with comorbidity and functional impairment is challenging.⁵ Age-related pharmacokinetic and pharmacodynamic changes increase the risk of adverse drug reactions.⁶

In this article, we review commonly used medications that are potentially inappropriate based on clinical practice. We also review tools to evaluate appropriate drug therapy in older adults.

DRUGS THAT ARE COMMONLY USED, BUT POTENTIALLY INAPPROPRIATE

Statins

Statins are effective when used as secondary prevention in older adults,⁷ but their efficacy when used as primary prevention of atherosclerotic cardiovascular disease in people age 75 and older is questionable.⁸ Nevertheless, they are widely used for this purpose. For example, before the 2013 joint guidelines of the American College of Cardiology and the American Heart Association (ACC/AHA) were released, 22% of patients age 80 and older in the Geisinger health system were taking a statin for primary prevention.⁹

The 2013 ACC/AHA guidelines included a limited recommendation for statins for primary prevention of atherosclerotic cardiovascular disease in adults age 75 and older.¹⁰ The guideline noted, however, that few data were available to support this recommendation.¹⁰

In a systematic review of 18 randomized clinical trials of statins for primary prevention of atherosclerotic cardiovascular disease, the mean age was 57, yet conclusions were extrapolated to an older patient population.¹¹ The estimated 10-year risk of atherosclerotic cardiovascular disease based on pooled cohort risk equations of adults age 75 and older always exceeds the 7.5% treatment threshold recommended by the guidelines.⁸

Myopathy is a common adverse effect of statins. In addition, statins interact with other drugs that inhibit the cytochrome P450 3A4 isoenzyme, such as amlodipine, amiodarone, and diltiazem.^8,12 If statin therapy caused no functional limitation due to muscle pain or weakness, statins for primary prevention would be cost-effective, but even a small increase in adverse effects in an elderly patient can offset the cardiovascular benefit.¹³ A recent post hoc secondary analysis found no benefit of pravastatin for primary prevention in adults age 75 and older.¹⁴

Thus, statin treatment for primary prevention in older patients should be individualized, based on life expectancy, function, and cardiovascular risk. Statin therapy does not replace modification of other risk factors.

Anticholinergics

Drugs with anticholinergic properties are commonly prescribed in the elderly for conditions such as muscle spasm, overactive bladder, psychiatric disorders, insomnia, extrapyramidal symptoms, vertigo, pruritus, peptic ulcer disease, seasonal allergies, and even the common cold,¹⁵ and many of the drugs often prescribed have strong anticholinergic properties (Table 1). Taking multiple medications with anticholinergic properties results in a high “anticholinergic burden,” which is associated with falls, impulsive behavior, poor physical performance, loss of independence, dementia, delirium, and brain atrophy.^15–18

The 2014 American College of Physicians guideline on nonsurgical management of urinary incontinence in women recommends pharmacologic treatment for urgency and stress urinary incontinence after failure of nonpharmacologic therapy,¹⁹ and many drugs for these urinary symptoms have anticholinergic properties. If an anticholinergic is necessary, an agent that results in a lower anticholinergic burden should be considered in older patients.

A pharmacist-initiated medication review and intervention may be another way to adjust medications to reduce the patient’s anticholinergic burden.^20,21 The common use of anticholinergic drugs in older adults reminds us to monitor their use closely.²²

Benzodiazepines are among the most commonly prescribed psychotropics in developed countries and are prescribed mainly by primary care physicians rather than psychiatrists.²³

In 2008, 5.2% of US adults ages 18 to 80 used a benzodiazepine, and long-term use was more prevalent in older patients (ages 65–80).²³

Benzodiazepines are prescribed for anxiety,²⁴ insomnia,²⁵ and agitation. They can cause withdrawal²⁶ and have potential for abuse.²⁷ Benzodiazepines are associated with cognitive decline,²⁸ impaired driving,²⁹ falls,³⁰ and hip fractures³¹ in older adults.

In addition, use of nonbenzodiazepine hypnotics (eg, zolpidem) is on the rise,³² and these drugs are known to increase the risk of hip fracture in nursing home residents.³³

The American Geriatrics Society, through the American Board of Internal Medicine’s Choosing Wisely campaign, recommends avoiding benzodiazepines as a first-line treatment for insomnia, agitation, or delirium in older adults.³⁴ Yet prescribing practices with these drugs in primary care settings conflict with guidelines, partly due to lack of training in constructive strategies regarding appropriate use of benzodiazepines.³⁵ Educating patients on the risks and benefits of benzodiazepine treatment, especially long-term use, has been shown to reduce the rate of benzodiazepine-associated secondary events.³⁶

Antipsychotics

Off-label use of antipsychotics is common and is increasing in the United States. In 2008, off-label use of antipsychotic drugs accounted for an estimated $6 billion.³⁷ A common off-label use is to manage behavioral symptoms of dementia, despite a black-box warning about an increased risk of death in patients with dementia who are treated with antipsychotics.^38,39 The Choosing Wisely campaign recommends against prescribing antipsychotics as a first-line treatment of behavioral and psychological symptoms of dementia.³⁴

Antipsychotic drugs are associated with risk of acute kidney injury,⁴⁰ as well as increased risk of falls and fractures (eg, a 52% higher risk of a serious fall, and a 50% higher risk of a nonvertebral osteoporotic fracture).⁴¹

Patients with dementia often exhibit aggression, resistance to care, and other challenging or disruptive behaviors. In such instances, antipsychotic drugs are often prescribed, but they provide limited and inconsistent benefits, while causing oversedation and worsening of cognitive function and increasing the likelihood of falling, stroke, and death.^38,39,41

Because pharmacologic treatments for dementia are only modestly effective, have notable risks, and do not treat some of the behaviors that family members and caregivers find most distressing, nonpharmacologic measures are recommended as first-line treatment.⁴² These include caregiver education and support, training in problem-solving, and targeted therapy directed at the underlying causes of specific behaviors (eg, implementing nighttime routines to address sleep disturbances).⁴² Nonpharmacologic management of behavioral symptoms in dementia can significantly improve quality of life for patients and caregivers.⁴² Use of antipsychotic drugs in patients with dementia should be limited to cases in which nonpharmacologic measures have failed and patients pose an imminent threat to themselves or others.⁴³

Proton pump inhibitors

Proton pump inhibitors are among the most commonly prescribed medications in the United States, and their use has increased significantly over the decade. It has been estimated that between 25% and 70% of these prescriptions have no appropriate indication.⁴⁴

There is considerable excess use of acid suppressants in both inpatient and outpatient settings.^45,46 In one study, at discharge from an internal medicine service, almost half of patients were taking a proton pump inhibitor.⁴⁷

Evidence-based guidelines recommend these drugs to treat gastroesophageal reflux disease, nonerosive reflux disease, erosive esophagitis, dyspepsia, and peptic ulcer disease. However, long-term use (ie, beyond 8 weeks) is recommended only for patients with erosive esophagitis, Barrett esophagus, a pathologic hypersecretory condition, or a demonstrated need for maintenance treatment for reflux disease.⁴⁸

Although proton pump inhibitors are highly effective and have low toxicity, there are reports of an association with Clostridium difficile infection,⁴⁹ community-acquired pneumonia,⁵⁰ hip fracture,⁵¹ vitamin B₁₂ deficiency,⁵² atrophic gastritis,⁵³ kidney disease,⁵⁴ and dementia.⁵⁵

Nondrug therapies such as weight loss and elevation of the head of the bed may improve esophageal pH levels and reflux symptoms.⁵⁶

Deprescribing.org has practical advice for healthcare providers, patients, and caregivers on how to discontinue proton pump inhibitors, including videos, algorithms, and guidelines.

TOOLS TO EVALUATE APPROPRIATE DRUG THERAPY

Beers criteria

The Beers criteria (Table 2), developed in 1991 by a geriatrician as an approach to safer, more effective drug therapy in frail elderly nursing home patients,⁵⁷ were updated by the American Geriatrics Society in 2015 for use in any clinical setting.⁵⁸ (The criteria are also available as a smartphone application through the American Geriatrics Society at www.americangeriatrics.org.)

The Beers criteria offer evidence-based recommendations on drugs to avoid in the elderly, along with the rationale for use, the quality of evidence behind the recommendation, and the graded strength of the recommendation. The Beers criteria should be viewed through the lens of clinical judgment to offer safer nonpharmacologic and pharmacologic treatments.

The Joint Commission recommends medication reconciliation at every transition of care.⁵⁹ The Beers criteria are a good starting point for a comprehensive medication review.

STOPP/START criteria

Another tool to aid safe prescribing in older adults is the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions (STOPP), used in conjuction with the Screening Tool to Alert Doctors to Right Treatment (START). The STOPP/START criteria^60,61 are based on an up-to-date literature review and consensus (Table 3).

THE BOTTOM LINE

Physicians caring for older adults need to diligently weigh the benefits of drug therapy and consider the patient’s care goals, current level of functioning, life expectancy, values, and preferences. Statin therapy for primary prevention, anticholinergics, benzodiazepines, antipsychotics, and proton pump inhibitors are widely used without proper indications, pointing to the need for a periodic comprehensive review of medications to reevaluate the risks vs the benefits of the patient’s medications. The Beers criteria and the STOPP/ START criteria can be useful tools for this purpose.

Medications started for appropriate indications in middle age may need to be monitored more closely as the patient ages. Some drugs may become unnecessary or even dangerous as the patient ages, functional status and renal function decline, and goals of care change.

See related editorial

Older adults tend to have multiple illnesses and therefore take more drugs, and polypharmacy increases the risk of poor outcomes. The number of medications a person uses is a risk factor for adverse drug reactions, nonadherence, financial burden, drug-drug interactions, and worse outcomes.¹

The prevalence of polypharmacy increased from an estimated 8.2% to 15% from 1999 to 2011 based on the National Health and Nutrition Examination Survey.² Guideline-based therapy for specific diseases may lead to the addition of more medications to reach disease targets.³ Most older adults in the United States compound the risk of prescribed medications by also taking over-the-counter medications and dietary supplements.⁴

In addition, medications are often used in older adults based on studies of younger persons without significant comorbidities. Applying clinical guidelines based on these studies to older adults with comorbidity and functional impairment is challenging.⁵ Age-related pharmacokinetic and pharmacodynamic changes increase the risk of adverse drug reactions.⁶

In this article, we review commonly used medications that are potentially inappropriate based on clinical practice. We also review tools to evaluate appropriate drug therapy in older adults.

DRUGS THAT ARE COMMONLY USED, BUT POTENTIALLY INAPPROPRIATE

Statins

Statins are effective when used as secondary prevention in older adults,⁷ but their efficacy when used as primary prevention of atherosclerotic cardiovascular disease in people age 75 and older is questionable.⁸ Nevertheless, they are widely used for this purpose. For example, before the 2013 joint guidelines of the American College of Cardiology and the American Heart Association (ACC/AHA) were released, 22% of patients age 80 and older in the Geisinger health system were taking a statin for primary prevention.⁹

The 2013 ACC/AHA guidelines included a limited recommendation for statins for primary prevention of atherosclerotic cardiovascular disease in adults age 75 and older.¹⁰ The guideline noted, however, that few data were available to support this recommendation.¹⁰

In a systematic review of 18 randomized clinical trials of statins for primary prevention of atherosclerotic cardiovascular disease, the mean age was 57, yet conclusions were extrapolated to an older patient population.¹¹ The estimated 10-year risk of atherosclerotic cardiovascular disease based on pooled cohort risk equations of adults age 75 and older always exceeds the 7.5% treatment threshold recommended by the guidelines.⁸

Myopathy is a common adverse effect of statins. In addition, statins interact with other drugs that inhibit the cytochrome P450 3A4 isoenzyme, such as amlodipine, amiodarone, and diltiazem.^8,12 If statin therapy caused no functional limitation due to muscle pain or weakness, statins for primary prevention would be cost-effective, but even a small increase in adverse effects in an elderly patient can offset the cardiovascular benefit.¹³ A recent post hoc secondary analysis found no benefit of pravastatin for primary prevention in adults age 75 and older.¹⁴

Thus, statin treatment for primary prevention in older patients should be individualized, based on life expectancy, function, and cardiovascular risk. Statin therapy does not replace modification of other risk factors.

Anticholinergics

Drugs with anticholinergic properties are commonly prescribed in the elderly for conditions such as muscle spasm, overactive bladder, psychiatric disorders, insomnia, extrapyramidal symptoms, vertigo, pruritus, peptic ulcer disease, seasonal allergies, and even the common cold,¹⁵ and many of the drugs often prescribed have strong anticholinergic properties (Table 1). Taking multiple medications with anticholinergic properties results in a high “anticholinergic burden,” which is associated with falls, impulsive behavior, poor physical performance, loss of independence, dementia, delirium, and brain atrophy.^15–18

The 2014 American College of Physicians guideline on nonsurgical management of urinary incontinence in women recommends pharmacologic treatment for urgency and stress urinary incontinence after failure of nonpharmacologic therapy,¹⁹ and many drugs for these urinary symptoms have anticholinergic properties. If an anticholinergic is necessary, an agent that results in a lower anticholinergic burden should be considered in older patients.

A pharmacist-initiated medication review and intervention may be another way to adjust medications to reduce the patient’s anticholinergic burden.^20,21 The common use of anticholinergic drugs in older adults reminds us to monitor their use closely.²²

Benzodiazepines are among the most commonly prescribed psychotropics in developed countries and are prescribed mainly by primary care physicians rather than psychiatrists.²³

In 2008, 5.2% of US adults ages 18 to 80 used a benzodiazepine, and long-term use was more prevalent in older patients (ages 65–80).²³

Benzodiazepines are prescribed for anxiety,²⁴ insomnia,²⁵ and agitation. They can cause withdrawal²⁶ and have potential for abuse.²⁷ Benzodiazepines are associated with cognitive decline,²⁸ impaired driving,²⁹ falls,³⁰ and hip fractures³¹ in older adults.

In addition, use of nonbenzodiazepine hypnotics (eg, zolpidem) is on the rise,³² and these drugs are known to increase the risk of hip fracture in nursing home residents.³³

The American Geriatrics Society, through the American Board of Internal Medicine’s Choosing Wisely campaign, recommends avoiding benzodiazepines as a first-line treatment for insomnia, agitation, or delirium in older adults.³⁴ Yet prescribing practices with these drugs in primary care settings conflict with guidelines, partly due to lack of training in constructive strategies regarding appropriate use of benzodiazepines.³⁵ Educating patients on the risks and benefits of benzodiazepine treatment, especially long-term use, has been shown to reduce the rate of benzodiazepine-associated secondary events.³⁶

Antipsychotics

Off-label use of antipsychotics is common and is increasing in the United States. In 2008, off-label use of antipsychotic drugs accounted for an estimated $6 billion.³⁷ A common off-label use is to manage behavioral symptoms of dementia, despite a black-box warning about an increased risk of death in patients with dementia who are treated with antipsychotics.^38,39 The Choosing Wisely campaign recommends against prescribing antipsychotics as a first-line treatment of behavioral and psychological symptoms of dementia.³⁴

Antipsychotic drugs are associated with risk of acute kidney injury,⁴⁰ as well as increased risk of falls and fractures (eg, a 52% higher risk of a serious fall, and a 50% higher risk of a nonvertebral osteoporotic fracture).⁴¹

Patients with dementia often exhibit aggression, resistance to care, and other challenging or disruptive behaviors. In such instances, antipsychotic drugs are often prescribed, but they provide limited and inconsistent benefits, while causing oversedation and worsening of cognitive function and increasing the likelihood of falling, stroke, and death.^38,39,41

Because pharmacologic treatments for dementia are only modestly effective, have notable risks, and do not treat some of the behaviors that family members and caregivers find most distressing, nonpharmacologic measures are recommended as first-line treatment.⁴² These include caregiver education and support, training in problem-solving, and targeted therapy directed at the underlying causes of specific behaviors (eg, implementing nighttime routines to address sleep disturbances).⁴² Nonpharmacologic management of behavioral symptoms in dementia can significantly improve quality of life for patients and caregivers.⁴² Use of antipsychotic drugs in patients with dementia should be limited to cases in which nonpharmacologic measures have failed and patients pose an imminent threat to themselves or others.⁴³

Proton pump inhibitors

Proton pump inhibitors are among the most commonly prescribed medications in the United States, and their use has increased significantly over the decade. It has been estimated that between 25% and 70% of these prescriptions have no appropriate indication.⁴⁴

There is considerable excess use of acid suppressants in both inpatient and outpatient settings.^45,46 In one study, at discharge from an internal medicine service, almost half of patients were taking a proton pump inhibitor.⁴⁷

Evidence-based guidelines recommend these drugs to treat gastroesophageal reflux disease, nonerosive reflux disease, erosive esophagitis, dyspepsia, and peptic ulcer disease. However, long-term use (ie, beyond 8 weeks) is recommended only for patients with erosive esophagitis, Barrett esophagus, a pathologic hypersecretory condition, or a demonstrated need for maintenance treatment for reflux disease.⁴⁸

Although proton pump inhibitors are highly effective and have low toxicity, there are reports of an association with Clostridium difficile infection,⁴⁹ community-acquired pneumonia,⁵⁰ hip fracture,⁵¹ vitamin B₁₂ deficiency,⁵² atrophic gastritis,⁵³ kidney disease,⁵⁴ and dementia.⁵⁵

Nondrug therapies such as weight loss and elevation of the head of the bed may improve esophageal pH levels and reflux symptoms.⁵⁶

Deprescribing.org has practical advice for healthcare providers, patients, and caregivers on how to discontinue proton pump inhibitors, including videos, algorithms, and guidelines.

TOOLS TO EVALUATE APPROPRIATE DRUG THERAPY

Beers criteria

The Beers criteria (Table 2), developed in 1991 by a geriatrician as an approach to safer, more effective drug therapy in frail elderly nursing home patients,⁵⁷ were updated by the American Geriatrics Society in 2015 for use in any clinical setting.⁵⁸ (The criteria are also available as a smartphone application through the American Geriatrics Society at www.americangeriatrics.org.)

The Beers criteria offer evidence-based recommendations on drugs to avoid in the elderly, along with the rationale for use, the quality of evidence behind the recommendation, and the graded strength of the recommendation. The Beers criteria should be viewed through the lens of clinical judgment to offer safer nonpharmacologic and pharmacologic treatments.

The Joint Commission recommends medication reconciliation at every transition of care.⁵⁹ The Beers criteria are a good starting point for a comprehensive medication review.

STOPP/START criteria

Another tool to aid safe prescribing in older adults is the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions (STOPP), used in conjuction with the Screening Tool to Alert Doctors to Right Treatment (START). The STOPP/START criteria^60,61 are based on an up-to-date literature review and consensus (Table 3).

THE BOTTOM LINE

Physicians caring for older adults need to diligently weigh the benefits of drug therapy and consider the patient’s care goals, current level of functioning, life expectancy, values, and preferences. Statin therapy for primary prevention, anticholinergics, benzodiazepines, antipsychotics, and proton pump inhibitors are widely used without proper indications, pointing to the need for a periodic comprehensive review of medications to reevaluate the risks vs the benefits of the patient’s medications. The Beers criteria and the STOPP/ START criteria can be useful tools for this purpose.

Medications started for appropriate indications in middle age may need to be monitored more closely as the patient ages. Some drugs may become unnecessary or even dangerous as the patient ages, functional status and renal function decline, and goals of care change.

See related editorial

Older adults tend to have multiple illnesses and therefore take more drugs, and polypharmacy increases the risk of poor outcomes. The number of medications a person uses is a risk factor for adverse drug reactions, nonadherence, financial burden, drug-drug interactions, and worse outcomes.¹

The prevalence of polypharmacy increased from an estimated 8.2% to 15% from 1999 to 2011 based on the National Health and Nutrition Examination Survey.² Guideline-based therapy for specific diseases may lead to the addition of more medications to reach disease targets.³ Most older adults in the United States compound the risk of prescribed medications by also taking over-the-counter medications and dietary supplements.⁴

In addition, medications are often used in older adults based on studies of younger persons without significant comorbidities. Applying clinical guidelines based on these studies to older adults with comorbidity and functional impairment is challenging.⁵ Age-related pharmacokinetic and pharmacodynamic changes increase the risk of adverse drug reactions.⁶

In this article, we review commonly used medications that are potentially inappropriate based on clinical practice. We also review tools to evaluate appropriate drug therapy in older adults.

DRUGS THAT ARE COMMONLY USED, BUT POTENTIALLY INAPPROPRIATE

Statins

Statins are effective when used as secondary prevention in older adults,⁷ but their efficacy when used as primary prevention of atherosclerotic cardiovascular disease in people age 75 and older is questionable.⁸ Nevertheless, they are widely used for this purpose. For example, before the 2013 joint guidelines of the American College of Cardiology and the American Heart Association (ACC/AHA) were released, 22% of patients age 80 and older in the Geisinger health system were taking a statin for primary prevention.⁹

The 2013 ACC/AHA guidelines included a limited recommendation for statins for primary prevention of atherosclerotic cardiovascular disease in adults age 75 and older.¹⁰ The guideline noted, however, that few data were available to support this recommendation.¹⁰

In a systematic review of 18 randomized clinical trials of statins for primary prevention of atherosclerotic cardiovascular disease, the mean age was 57, yet conclusions were extrapolated to an older patient population.¹¹ The estimated 10-year risk of atherosclerotic cardiovascular disease based on pooled cohort risk equations of adults age 75 and older always exceeds the 7.5% treatment threshold recommended by the guidelines.⁸

Myopathy is a common adverse effect of statins. In addition, statins interact with other drugs that inhibit the cytochrome P450 3A4 isoenzyme, such as amlodipine, amiodarone, and diltiazem.^8,12 If statin therapy caused no functional limitation due to muscle pain or weakness, statins for primary prevention would be cost-effective, but even a small increase in adverse effects in an elderly patient can offset the cardiovascular benefit.¹³ A recent post hoc secondary analysis found no benefit of pravastatin for primary prevention in adults age 75 and older.¹⁴

Thus, statin treatment for primary prevention in older patients should be individualized, based on life expectancy, function, and cardiovascular risk. Statin therapy does not replace modification of other risk factors.

Anticholinergics

Drugs with anticholinergic properties are commonly prescribed in the elderly for conditions such as muscle spasm, overactive bladder, psychiatric disorders, insomnia, extrapyramidal symptoms, vertigo, pruritus, peptic ulcer disease, seasonal allergies, and even the common cold,¹⁵ and many of the drugs often prescribed have strong anticholinergic properties (Table 1). Taking multiple medications with anticholinergic properties results in a high “anticholinergic burden,” which is associated with falls, impulsive behavior, poor physical performance, loss of independence, dementia, delirium, and brain atrophy.^15–18

The 2014 American College of Physicians guideline on nonsurgical management of urinary incontinence in women recommends pharmacologic treatment for urgency and stress urinary incontinence after failure of nonpharmacologic therapy,¹⁹ and many drugs for these urinary symptoms have anticholinergic properties. If an anticholinergic is necessary, an agent that results in a lower anticholinergic burden should be considered in older patients.

A pharmacist-initiated medication review and intervention may be another way to adjust medications to reduce the patient’s anticholinergic burden.^20,21 The common use of anticholinergic drugs in older adults reminds us to monitor their use closely.²²

Benzodiazepines are among the most commonly prescribed psychotropics in developed countries and are prescribed mainly by primary care physicians rather than psychiatrists.²³

In 2008, 5.2% of US adults ages 18 to 80 used a benzodiazepine, and long-term use was more prevalent in older patients (ages 65–80).²³

Benzodiazepines are prescribed for anxiety,²⁴ insomnia,²⁵ and agitation. They can cause withdrawal²⁶ and have potential for abuse.²⁷ Benzodiazepines are associated with cognitive decline,²⁸ impaired driving,²⁹ falls,³⁰ and hip fractures³¹ in older adults.

In addition, use of nonbenzodiazepine hypnotics (eg, zolpidem) is on the rise,³² and these drugs are known to increase the risk of hip fracture in nursing home residents.³³

The American Geriatrics Society, through the American Board of Internal Medicine’s Choosing Wisely campaign, recommends avoiding benzodiazepines as a first-line treatment for insomnia, agitation, or delirium in older adults.³⁴ Yet prescribing practices with these drugs in primary care settings conflict with guidelines, partly due to lack of training in constructive strategies regarding appropriate use of benzodiazepines.³⁵ Educating patients on the risks and benefits of benzodiazepine treatment, especially long-term use, has been shown to reduce the rate of benzodiazepine-associated secondary events.³⁶

Antipsychotics

Off-label use of antipsychotics is common and is increasing in the United States. In 2008, off-label use of antipsychotic drugs accounted for an estimated $6 billion.³⁷ A common off-label use is to manage behavioral symptoms of dementia, despite a black-box warning about an increased risk of death in patients with dementia who are treated with antipsychotics.^38,39 The Choosing Wisely campaign recommends against prescribing antipsychotics as a first-line treatment of behavioral and psychological symptoms of dementia.³⁴

Antipsychotic drugs are associated with risk of acute kidney injury,⁴⁰ as well as increased risk of falls and fractures (eg, a 52% higher risk of a serious fall, and a 50% higher risk of a nonvertebral osteoporotic fracture).⁴¹

Patients with dementia often exhibit aggression, resistance to care, and other challenging or disruptive behaviors. In such instances, antipsychotic drugs are often prescribed, but they provide limited and inconsistent benefits, while causing oversedation and worsening of cognitive function and increasing the likelihood of falling, stroke, and death.^38,39,41

Because pharmacologic treatments for dementia are only modestly effective, have notable risks, and do not treat some of the behaviors that family members and caregivers find most distressing, nonpharmacologic measures are recommended as first-line treatment.⁴² These include caregiver education and support, training in problem-solving, and targeted therapy directed at the underlying causes of specific behaviors (eg, implementing nighttime routines to address sleep disturbances).⁴² Nonpharmacologic management of behavioral symptoms in dementia can significantly improve quality of life for patients and caregivers.⁴² Use of antipsychotic drugs in patients with dementia should be limited to cases in which nonpharmacologic measures have failed and patients pose an imminent threat to themselves or others.⁴³

Proton pump inhibitors

Proton pump inhibitors are among the most commonly prescribed medications in the United States, and their use has increased significantly over the decade. It has been estimated that between 25% and 70% of these prescriptions have no appropriate indication.⁴⁴

There is considerable excess use of acid suppressants in both inpatient and outpatient settings.^45,46 In one study, at discharge from an internal medicine service, almost half of patients were taking a proton pump inhibitor.⁴⁷

Evidence-based guidelines recommend these drugs to treat gastroesophageal reflux disease, nonerosive reflux disease, erosive esophagitis, dyspepsia, and peptic ulcer disease. However, long-term use (ie, beyond 8 weeks) is recommended only for patients with erosive esophagitis, Barrett esophagus, a pathologic hypersecretory condition, or a demonstrated need for maintenance treatment for reflux disease.⁴⁸

Although proton pump inhibitors are highly effective and have low toxicity, there are reports of an association with Clostridium difficile infection,⁴⁹ community-acquired pneumonia,⁵⁰ hip fracture,⁵¹ vitamin B₁₂ deficiency,⁵² atrophic gastritis,⁵³ kidney disease,⁵⁴ and dementia.⁵⁵

Nondrug therapies such as weight loss and elevation of the head of the bed may improve esophageal pH levels and reflux symptoms.⁵⁶

Deprescribing.org has practical advice for healthcare providers, patients, and caregivers on how to discontinue proton pump inhibitors, including videos, algorithms, and guidelines.

TOOLS TO EVALUATE APPROPRIATE DRUG THERAPY

Beers criteria

The Beers criteria (Table 2), developed in 1991 by a geriatrician as an approach to safer, more effective drug therapy in frail elderly nursing home patients,⁵⁷ were updated by the American Geriatrics Society in 2015 for use in any clinical setting.⁵⁸ (The criteria are also available as a smartphone application through the American Geriatrics Society at www.americangeriatrics.org.)

The Beers criteria offer evidence-based recommendations on drugs to avoid in the elderly, along with the rationale for use, the quality of evidence behind the recommendation, and the graded strength of the recommendation. The Beers criteria should be viewed through the lens of clinical judgment to offer safer nonpharmacologic and pharmacologic treatments.

The Joint Commission recommends medication reconciliation at every transition of care.⁵⁹ The Beers criteria are a good starting point for a comprehensive medication review.

STOPP/START criteria

Another tool to aid safe prescribing in older adults is the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions (STOPP), used in conjuction with the Screening Tool to Alert Doctors to Right Treatment (START). The STOPP/START criteria^60,61 are based on an up-to-date literature review and consensus (Table 3).

THE BOTTOM LINE

Physicians caring for older adults need to diligently weigh the benefits of drug therapy and consider the patient’s care goals, current level of functioning, life expectancy, values, and preferences. Statin therapy for primary prevention, anticholinergics, benzodiazepines, antipsychotics, and proton pump inhibitors are widely used without proper indications, pointing to the need for a periodic comprehensive review of medications to reevaluate the risks vs the benefits of the patient’s medications. The Beers criteria and the STOPP/ START criteria can be useful tools for this purpose.

According to a 2016 study, more than one-third of older adults in the United States take 5 or more medications.¹ This is a growing problem. Not only do older patients take more drugs than younger patients, they are also at higher risk of adverse drug events, drug-drug interactions, geriatric syndromes, and lower adherence.²

See related article

Many drugs that older patients are given are potentially inappropriate, ie, their risks outweigh the expected benefits, particularly when effective and safer alternative therapies exist. Although many clinicians are aware of the risks of polypharmacy, they may not be confident in discontinuing potentially inappropriate medications. The process of deliberately tapering, stopping, or reducing doses of medications with the goal of reducing harm and improving patient outcomes is known as deprescribing.³

In this issue, Kim et al⁴ review several medications that are overused or often used inappropriately in older adults: statins for primary prevention of atherosclerotic cardiovascular disease, anticholinergic drugs, benzodiazepines, antipsychotics, and proton pump inhibitors. They offer guidance about the situations in which these drugs may be inappropriate as well as alternative drug and nondrug treatments. Further, they suggest that, when prescribing or deprescribing drugs in older adults, clinicians consult tools such as the Beers criteria and the STOPP/START criteria (the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions, and the Screening Tool to Alert Doctors to Right Treatment).

The issues Kim et al review are highly relevant and may increase awareness of specific potentially inappropriate medications. They also remind us that nonpharmacologic treatments are first-line for many medical conditions. In an era of a pill for every ill and a quick-fix mentality among both patients and providers, lifestyle changes and other nonpharmacologic treatments may be overlooked. Similarly, the STOPP/START criteria, which are concrete, evidence-based recommendations that can be applied to patient care, are likely underused in clinical practice.

Although necessary and valuable, simply arming clinicians with knowledge is insufficient to tackle the problems of polypharmacy and inappropriate prescribing. As the authors note in their discussion of benzodiazepines, practice guidelines exist regarding prescribing these agents, and data from randomized trials support specific interventions to deprescribe them.⁵ Nevertheless, clinicians report feeling inadequately prepared to discontinue benzodiazepines, particularly when patients perceive benefit from them. As such, user-friendly tools and specific strategies for weighing risks vs benefits are critical for clinicians.

PUTTING KNOWLEDGE INTO PRACTICE

How do we translate knowledge into practice with regard to deprescribing potentially inappropriate medications in older patients—or prescribing drugs only if appropriate in the first place?

An opportunity arises when patients are in the hospital. Taking a medication history on admission and matching medications with indications are key starting points. Clinical pharmacists can help screen for side effects and potential interactions and can provide deprescribing recommendations. Meticulous discharge medication reconciliation, patient education, and communication of the updated medication list to the outpatient provider are central to ensuring that patients adhere to medication adjustments after they go home.

A MATTER OF BALANCE

Another factor to consider is the patient’s physiologic age compared with his or her chronologic age. If a patient has multiple comorbidities, frailty, limited life expectancy, or poor renal function, we may consider her older than her chronologic age. In this case, a drug’s risks may outweigh its benefit, which is something to be discussed. On the other hand, a high-functioning and relatively healthy elderly patient may be a candidate for medications known to reduce the risk of death or control a chronic disease better. Incorporating a patient’s goals of care and using shared decision-making are also likely to yield an optimal medication regimen.

Smartphone apps and resources embedded in electronic health records provide additional decision support. Used when prescribing or reconciling medications, these supplemental brains offer instant feedback and information on dose adjustments, drug interactions, clinical guidelines, and even potentially inappropriate medications. While the impacts of these electronic tools on prescribing patterns and outcomes in geriatric populations remain unclear, new ones are being developed and studied.⁶ This may be the most promising way to translate knowledge into practice, as it is more easily integrated with existing clinician workflows.

AN OPPORTUNITY TO IMPROVE

There is significant opportunity to reduce polypharmacy and optimize medication prescribing practices for older adults. Awareness of potentially inappropriate medications and clinical situations in which the use of certain classes of medications should be minimized is the first step in addressing this problem. Using tools such as the STOPP/START criteria, reviewing medications at critical transition points, prioritizing patient function and goals, and using electronic clinical decision support should aid prescribing decisions.

Whenever possible, collaborating with other care team members such as pharmacists may increase efficiency and effectiveness of medication management. Ultimately, inclusion of more older adults in clinical trials may provide data-driven guidance for weighing risks and benefits. Finally, further study of the effects of deprescribing on clinical outcomes may be the missing piece to help clinicians and patients find balance in prescription management.

According to a 2016 study, more than one-third of older adults in the United States take 5 or more medications.¹ This is a growing problem. Not only do older patients take more drugs than younger patients, they are also at higher risk of adverse drug events, drug-drug interactions, geriatric syndromes, and lower adherence.²

See related article

Many drugs that older patients are given are potentially inappropriate, ie, their risks outweigh the expected benefits, particularly when effective and safer alternative therapies exist. Although many clinicians are aware of the risks of polypharmacy, they may not be confident in discontinuing potentially inappropriate medications. The process of deliberately tapering, stopping, or reducing doses of medications with the goal of reducing harm and improving patient outcomes is known as deprescribing.³

In this issue, Kim et al⁴ review several medications that are overused or often used inappropriately in older adults: statins for primary prevention of atherosclerotic cardiovascular disease, anticholinergic drugs, benzodiazepines, antipsychotics, and proton pump inhibitors. They offer guidance about the situations in which these drugs may be inappropriate as well as alternative drug and nondrug treatments. Further, they suggest that, when prescribing or deprescribing drugs in older adults, clinicians consult tools such as the Beers criteria and the STOPP/START criteria (the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions, and the Screening Tool to Alert Doctors to Right Treatment).

The issues Kim et al review are highly relevant and may increase awareness of specific potentially inappropriate medications. They also remind us that nonpharmacologic treatments are first-line for many medical conditions. In an era of a pill for every ill and a quick-fix mentality among both patients and providers, lifestyle changes and other nonpharmacologic treatments may be overlooked. Similarly, the STOPP/START criteria, which are concrete, evidence-based recommendations that can be applied to patient care, are likely underused in clinical practice.

Although necessary and valuable, simply arming clinicians with knowledge is insufficient to tackle the problems of polypharmacy and inappropriate prescribing. As the authors note in their discussion of benzodiazepines, practice guidelines exist regarding prescribing these agents, and data from randomized trials support specific interventions to deprescribe them.⁵ Nevertheless, clinicians report feeling inadequately prepared to discontinue benzodiazepines, particularly when patients perceive benefit from them. As such, user-friendly tools and specific strategies for weighing risks vs benefits are critical for clinicians.

PUTTING KNOWLEDGE INTO PRACTICE

How do we translate knowledge into practice with regard to deprescribing potentially inappropriate medications in older patients—or prescribing drugs only if appropriate in the first place?

An opportunity arises when patients are in the hospital. Taking a medication history on admission and matching medications with indications are key starting points. Clinical pharmacists can help screen for side effects and potential interactions and can provide deprescribing recommendations. Meticulous discharge medication reconciliation, patient education, and communication of the updated medication list to the outpatient provider are central to ensuring that patients adhere to medication adjustments after they go home.

A MATTER OF BALANCE

Another factor to consider is the patient’s physiologic age compared with his or her chronologic age. If a patient has multiple comorbidities, frailty, limited life expectancy, or poor renal function, we may consider her older than her chronologic age. In this case, a drug’s risks may outweigh its benefit, which is something to be discussed. On the other hand, a high-functioning and relatively healthy elderly patient may be a candidate for medications known to reduce the risk of death or control a chronic disease better. Incorporating a patient’s goals of care and using shared decision-making are also likely to yield an optimal medication regimen.

Smartphone apps and resources embedded in electronic health records provide additional decision support. Used when prescribing or reconciling medications, these supplemental brains offer instant feedback and information on dose adjustments, drug interactions, clinical guidelines, and even potentially inappropriate medications. While the impacts of these electronic tools on prescribing patterns and outcomes in geriatric populations remain unclear, new ones are being developed and studied.⁶ This may be the most promising way to translate knowledge into practice, as it is more easily integrated with existing clinician workflows.

AN OPPORTUNITY TO IMPROVE

There is significant opportunity to reduce polypharmacy and optimize medication prescribing practices for older adults. Awareness of potentially inappropriate medications and clinical situations in which the use of certain classes of medications should be minimized is the first step in addressing this problem. Using tools such as the STOPP/START criteria, reviewing medications at critical transition points, prioritizing patient function and goals, and using electronic clinical decision support should aid prescribing decisions.

Whenever possible, collaborating with other care team members such as pharmacists may increase efficiency and effectiveness of medication management. Ultimately, inclusion of more older adults in clinical trials may provide data-driven guidance for weighing risks and benefits. Finally, further study of the effects of deprescribing on clinical outcomes may be the missing piece to help clinicians and patients find balance in prescription management.

According to a 2016 study, more than one-third of older adults in the United States take 5 or more medications.¹ This is a growing problem. Not only do older patients take more drugs than younger patients, they are also at higher risk of adverse drug events, drug-drug interactions, geriatric syndromes, and lower adherence.²

See related article

Many drugs that older patients are given are potentially inappropriate, ie, their risks outweigh the expected benefits, particularly when effective and safer alternative therapies exist. Although many clinicians are aware of the risks of polypharmacy, they may not be confident in discontinuing potentially inappropriate medications. The process of deliberately tapering, stopping, or reducing doses of medications with the goal of reducing harm and improving patient outcomes is known as deprescribing.³

In this issue, Kim et al⁴ review several medications that are overused or often used inappropriately in older adults: statins for primary prevention of atherosclerotic cardiovascular disease, anticholinergic drugs, benzodiazepines, antipsychotics, and proton pump inhibitors. They offer guidance about the situations in which these drugs may be inappropriate as well as alternative drug and nondrug treatments. Further, they suggest that, when prescribing or deprescribing drugs in older adults, clinicians consult tools such as the Beers criteria and the STOPP/START criteria (the Screening Tool of Older Persons’ Potentially Inappropriate Prescriptions, and the Screening Tool to Alert Doctors to Right Treatment).

The issues Kim et al review are highly relevant and may increase awareness of specific potentially inappropriate medications. They also remind us that nonpharmacologic treatments are first-line for many medical conditions. In an era of a pill for every ill and a quick-fix mentality among both patients and providers, lifestyle changes and other nonpharmacologic treatments may be overlooked. Similarly, the STOPP/START criteria, which are concrete, evidence-based recommendations that can be applied to patient care, are likely underused in clinical practice.

Although necessary and valuable, simply arming clinicians with knowledge is insufficient to tackle the problems of polypharmacy and inappropriate prescribing. As the authors note in their discussion of benzodiazepines, practice guidelines exist regarding prescribing these agents, and data from randomized trials support specific interventions to deprescribe them.⁵ Nevertheless, clinicians report feeling inadequately prepared to discontinue benzodiazepines, particularly when patients perceive benefit from them. As such, user-friendly tools and specific strategies for weighing risks vs benefits are critical for clinicians.

PUTTING KNOWLEDGE INTO PRACTICE

How do we translate knowledge into practice with regard to deprescribing potentially inappropriate medications in older patients—or prescribing drugs only if appropriate in the first place?

An opportunity arises when patients are in the hospital. Taking a medication history on admission and matching medications with indications are key starting points. Clinical pharmacists can help screen for side effects and potential interactions and can provide deprescribing recommendations. Meticulous discharge medication reconciliation, patient education, and communication of the updated medication list to the outpatient provider are central to ensuring that patients adhere to medication adjustments after they go home.

A MATTER OF BALANCE

Another factor to consider is the patient’s physiologic age compared with his or her chronologic age. If a patient has multiple comorbidities, frailty, limited life expectancy, or poor renal function, we may consider her older than her chronologic age. In this case, a drug’s risks may outweigh its benefit, which is something to be discussed. On the other hand, a high-functioning and relatively healthy elderly patient may be a candidate for medications known to reduce the risk of death or control a chronic disease better. Incorporating a patient’s goals of care and using shared decision-making are also likely to yield an optimal medication regimen.

Smartphone apps and resources embedded in electronic health records provide additional decision support. Used when prescribing or reconciling medications, these supplemental brains offer instant feedback and information on dose adjustments, drug interactions, clinical guidelines, and even potentially inappropriate medications. While the impacts of these electronic tools on prescribing patterns and outcomes in geriatric populations remain unclear, new ones are being developed and studied.⁶ This may be the most promising way to translate knowledge into practice, as it is more easily integrated with existing clinician workflows.

AN OPPORTUNITY TO IMPROVE

There is significant opportunity to reduce polypharmacy and optimize medication prescribing practices for older adults. Awareness of potentially inappropriate medications and clinical situations in which the use of certain classes of medications should be minimized is the first step in addressing this problem. Using tools such as the STOPP/START criteria, reviewing medications at critical transition points, prioritizing patient function and goals, and using electronic clinical decision support should aid prescribing decisions.

Whenever possible, collaborating with other care team members such as pharmacists may increase efficiency and effectiveness of medication management. Ultimately, inclusion of more older adults in clinical trials may provide data-driven guidance for weighing risks and benefits. Finally, further study of the effects of deprescribing on clinical outcomes may be the missing piece to help clinicians and patients find balance in prescription management.

When assessing and attempting to modify the risk of cardiovascular disease in older patients, physicians should consider incorporating the concept of frailty. The balance of risk and benefit may differ considerably for 2 patients of the same age if one is fit and the other is frail. Because the aging population is a diverse group, a one-size-fits-all approach to cardiovascular disease prevention and risk-factor management is not appropriate.

See related editorial

A GROWING, DIVERSE GROUP

The number of older adults with multiple cardiovascular risk factors is increasing as life expectancy improves. US residents who are age 65 today can expect to live to an average age of 84 (men) or 87 (women).¹

However, the range of life expectancy for people reaching these advanced ages is wide, and chronologic age is no longer sufficient to determine a patient’s risk profile. Furthermore, the prevalence of cardiovascular disease rises with age, and age itself is the strongest predictor of cardiovascular risk.²

Current risk calculators have not been validated in people over age 80,² making them inadequate for use in older patients. Age alone cannot identify who will benefit from preventive strategies, except in situations when a dominant disease such as metastatic cancer, end-stage renal disease, end-stage dementia, or end-stage heart failure is expected to lead to mortality within a year. Guidelines for treating common risk factors such as elevated cholesterol³ in the general population have generally not focused on adults over 75 or recognized their diversity in health status.⁴ In order to generate an individualized prescription for cardiovascular disease prevention for older adults, issues such as frailty, cognitive and functional status, disability, and comorbidity must be considered.

WHAT IS FRAILTY?

Clinicians have recognized frailty for decades, but to date there remains a debate on how to define it.

Clegg et al⁵ described frailty as “a state of increased vulnerability to poor resolution of homeostasis after a stressor event,”⁵ a definition generally agreed upon, as frailty predicts both poor health outcomes and death.

Indeed, in a prospective study of 5,317 men and women ranging in age from 65 to 101, those identified as frail at baseline were 6 times more likely to have died 3 years later (mortality rates 18% vs 3%), and the difference persisted at 7 years.⁶ After adjusting for comorbidities, those identified as frail were also more likely to fall, develop limitations in mobility or activities of daily living, or be hospitalized.

The two current leading theories of frailty were defined by Fried et al⁶ and by Rockwood and Mitnitski.⁷

Fried et al⁶ have operationalized frailty as a “physical phenotype,” defined as 3 or more of the following:

• Unintentional weight loss of 10 pounds in the past year
• Self-reported exhaustion
• Weakness as measured by grip strength
• Slow walking speed
• Decreased physical activity.⁶

Rockwood and Mitnitski⁷ define frailty as an accumulation of health-related deficits over time. They recommend that 30 to 40 possible deficits that cover a variety of health systems be included such as cognition, mood, function, and comorbidity. These are added and divided by the total possible number of variables to generate a score between 0 and 1.⁸

The difficulty in defining frailty has led to varying estimates of its prevalence, ranging from 25% to 50% in adults over 65 who have cardiovascular disease.⁹

CAUSE AND CONSEQUENCE OF CARDIOVASCULAR DISEASE

Studies have highlighted the bidirectional connection between frailty and cardiovascular disease.¹⁰ Frailty may predict cardiovascular disease, while cardiovascular disease is associated with an increased risk of incident frailty.^9,11

Frail adults with cardiovascular disease have a higher risk of poor outcomes, even after correcting for age, comorbidities, disability, and disease severity. For example, frailty is associated with a twofold higher mortality rate in individuals with cardiovascular disease.⁹

A prospective cohort study¹² of 3,895 middle-aged men and women demonstrated that those with an elevated cardiovascular risk score were at increased risk of frailty over 10 years (odds ratio [OR] 1.35, 95% confidence interval [CI] 1.21–1.51) and incident cardiovascular events (OR 1.36, 95% CI 1.15–1.61). This suggests that modification of cardiovascular risk factors earlier in life may reduce the risk of subsequently becoming frail.

Biologic mechanisms that may explain the connection between frailty and cardiovascular disease include derangements in inflammatory, hematologic, and endocrine pathways. People who are found to be clinically frail are more likely to have insulin resistance and elevated biomarkers such as C-reactive protein, D-dimer, and factor VIII.¹³ The inflammatory cytokine interleukin 6 is suggested as a common link between inflammation and thrombosis, perhaps contributing to the connection between cardiovascular disease and frailty. Many of these biomarkers have been linked to the pathophysiologic changes of aging, so-called “inflamm-aging” or immunosenescence, including sarcopenia, osteoporosis, and cardiovascular disease.¹⁴

For adults over age 70, frailty assessment is an important first step in managing cardiovascular disease risk.¹⁵ Frailty status will better identify those at risk of adverse outcomes in the short term and those who are most likely to benefit from long-term cardiovascular preventive strategies. Additionally, incorporating frailty assessment into traditional risk factor evaluation may permit appropriate intervention and prevention of a potentially modifiable risk factor.

Gait speed is a quick, easy, inexpensive, and sensitive way to assess frailty status, with excellent inter-rater and test-retest reliability, even in those with cognitive impairment.¹⁶ Slow gait speed predicts limitations in mobility, limitations in activities of daily living, and death.^8,17

In a prospective study¹⁸ of 1,567 men and women, mean age 74, slow gait speed was the strongest predictor of subsequent cardiovascular events.¹⁸

Gait speed is usually measured over a distance of 4 meters (13.1 feet),¹⁷ and the patient is asked to walk comfortably in an unobstructed, marked area. An assistive walking device can be used if needed. If possible, this is repeated once after a brief recovery period, and the average is recorded.

The FRAIL scale^19,20 is a simple, validated questionnaire that combines the Fried and Rockwood concepts of frailty and can be given over the phone or to patients in a waiting room. One point is given for each of the following, and people who have 3 or more are considered frail:

• Fatigue
• Resistance (inability to climb 1 flight of stairs)
• Ambulation (inability to walk 1 block)
• Illnesses (having more than 5)
• Loss of more than 5% of body weight.

Other measures of physical function such as grip strength (using a dynamometer), the Timed Up and Go test (assessing the ability to get up from a chair and walk a short distance), and Short Physical Performance Battery (assessing balance, chair stands, and walking speed) can be used to screen for frailty, but are more time-intensive than gait speed alone, and so are not always practical to use in a busy clinic.²¹

MANAGEMENT OF RISK FACTORS

Management of cardiovascular risk factors is best individualized as outlined below.

LOWERING HIGH BLOOD PRESSURE

The incidence of ischemic heart disease and stroke increases with age across all levels of elevated systolic and diastolic blood pressure.²² Hypertension is also associated with increased risk of cognitive decline. However, a J-shaped relationship has been observed in older adults, with increased cardiovascular events for both low and elevated blood pressure, although the clinical relevance remains controversial.²³

Odden et al²⁴ performed an observational study and found that high blood pressure was associated with an increased mortality rate in older adults with normal gait speed, while in those with slow gait speed, high blood pressure neither harmed nor helped. Those who could not walk 6 meters appeared to benefit from higher blood pressure.

HYVET (the Hypertension in the Very Elderly Trial),²⁵ a randomized controlled trial in 3,845 community-dwelling people age 80 or older with sustained systolic blood pressure higher than 160 mm Hg, found a significant reduction in rates of stroke and all-cause mortality (relative risk [RR] 0.76, P = .007) in the treatment arm using indapamide with perindopril if necessary to reach a target blood pressure of 150/80 mm Hg.

Frailty was not assessed during the trial; however, in a reanalysis, the results did not change in those identified as frail using a Rockwood frailty index (a count of health-related deficits accumulated over the lifespan).²⁶

SPRINT (the Systolic Blood Pressure Intervention Trial)²⁷ randomized participants age 50 and older with systolic blood pressure of 130 to 180 mm Hg and at increased risk of cardiovascular disease to intensive treatment (goal systolic blood pressure ≤ 120 mm Hg) or standard treatment (goal systolic blood pressure ≤ 140 mm Hg). In a prespecified subgroup of 2,636 participants over age 75 (mean age 80), hazard ratios and 95% confidence intervals for adverse outcomes with intensive treatment were:

• Major cardiovascular events: HR 0.66, 95% CI 0.51–0.85
• Death: HR 0.67, 95% CI 0.49–0.91.

Over 3 years of treatment this translated into a number needed to treat of 27 to prevent 1 cardiovascular event and 41 to prevent 1 death.

Within this subgroup, the benefit was similar regardless of level of frailty (measured both by a Rockwood frailty index and by gait speed).

However, the incidence of serious adverse treatment effects such as hypotension, orthostasis, electrolyte abnormalities, and acute kidney injury was higher with intensive treatment in the frail group. Although the difference was not statistically significant, it is cause for caution. Further, the exclusion criteria (history of diabetes, heart failure, dementia, stroke, weight loss of > 10%, nursing home residence) make it difficult to generalize the SPRINT findings to the general aging population.²⁷

Tinetti et al²⁸ performed an observational study using a nationally representative sample of older adults. They found that receiving any antihypertensive therapy was associated with an increased risk of falls with serious adverse outcomes. The risks of adverse events related to antihypertensive therapy increased with age.

Managing hypertension in frail patients at risk of cardiovascular disease requires balancing the benefits vs the risks of treatment, such as polypharmacy, falls, and orthostatic hypotension.

The Eighth Joint National Committee suggests a blood pressure goal of less than 150/90 mm Hg for all adults over age 60, and less than 140/90 mm Hg for those with a history of cardiovascular disease or diabetes.²⁹

The American College of Cardiology/American Heart Association (ACC/AHA) guidelines on hypertension, recently released, recommend a new blood pressure target of <120/<80 as normal, with 120–129/<80 considered elevated, 130–139/80–89 stage 1 hypertension, and ≥140/≥90 as stage 2 hypertension.³⁰ An important caveat to these guidelines is the recommendation to measure blood pressure accurately and with accurate technique, which is often not possible in many busy clinics. These guidelines are intended to apply to older adults as well, with a note that those with multiple morbidities and limited life expectancy will benefit from a shared decision that incorporates patient preferences and clinical judgment. Little guidance is given on how to incorporate frailty, although note is made that older adults who reside in assisted living facilities and nursing homes have not been represented in randomized controlled trials.³⁰

American Diabetes Association guidelines on hypertension in patients with diabetes recommend considering functional status, frailty, and life expectancy to decide on a blood pressure goal of either 140/90 mm Hg (if fit) or 150/90 mm Hg (if frail). They do not specify how to diagnose frailty.³¹

Canadian guidelines say that in those with advanced frailty (ie, entirely dependent for personal care and activities of daily living) and short life expectancy (months), it is reasonable to liberalize the systolic blood pressure goal to 160 to 190 mm Hg.³²

Our recommendations. In both frail and nonfrail individuals without a limited life expectancy, it is reasonable to aim for a blood pressure of at least less than 140/90 mm Hg. For those at increased risk of cardiovascular disease and able to tolerate treatment, careful lowering to 130/80 mm Hg may be considered, with close attention to side effects.

Treatment should start with the lowest possible dose, be titrated slowly, and may need to be tailored to standing blood pressure to avoid orthostatic hypotension.

Home blood pressure measurements may be beneficial in monitoring treatment.

MANAGING LIPIDS

For those over age 75, data on efficacy of statins are mixed due to the small number of older adults enrolled in randomized controlled trials of these drugs. To our knowledge, no statin trial has examined the role of frailty.

The PROSPER trial (Prospective Study of Pravastatin in the Elderly at Risk)³³ randomized 5,804 patients ages 70 to 82 to receive either pravastatin or placebo. Overall, the incidence of a composite end point of major cardiovascular events was 15% lower with active treatment (P = .014). However, the mean age was 75, which does little to address the paucity of evidence for those over age 75; follow-up time was only 3 years, and subgroup analysis did not show benefit in those who did not have a history of cardiovascular disease or in women.

The JUPITER trial (Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin)³⁴ randomized 5,695 people over age 70 without cardiovascular disease to receive either rosuvastatin or placebo. Exploratory analysis showed a significant 39% reduction in all-cause mortality and major cardiovascular events with active treatment (HR 0.61, 95% CI 0.46–0.82). Over 5 years of treatment, this translates to a number needed to treat of 19 to prevent 1 major cardiovascular event and 29 to prevent 1 cardiovascular death.

The benefit of statins for primary prevention in these trials began to be apparent 2 years after treatment was initiated.

The Women’s Health Initiative,³⁵ an observational study, found no difference in incident frailty in women older than 65 taking statins for 3 years compared with those who did not take statins

Odden et al³⁶ found that although statin use is generally well tolerated, the risks of statin-associated functional and cognitive decline may outweigh the benefits in those older than 75. The ongoing Statin in Reducing Events in the Elderly (STAREE) trial may shed light on this issue.

Recommendations on lipid management

The ACC/AHA,³ in their 2013 guidelines, do not recommend routine statin treatment for primary prevention in those over age 75, given a lack of evidence from randomized controlled trials. For secondary prevention, ie, for those who have a history of atherosclerotic cardiovascular disease, they recommend moderate-intensity statin therapy in this age group.

Our recommendations. For patients over age 75 without cardiovascular disease or frailty and with a life expectancy of at least 2 years, consider offering a statin for primary prevention of cardiovascular disease as part of shared decision-making.

In those with known cardiovascular disease, it is reasonable to continue statin therapy except in situations where the life expectancy is less than 6 months.³⁷

Although moderate- or high-intensity statin therapy is recommended in current guidelines, for many older adults it is prudent to consider the lowest tolerable dose to improve adherence, with close monitoring for side effects such as myalgia and weakness.

TYPE 2 DIABETES

Evidence suggests that tight glycemic control in type 2 diabetes is harmful for adults ages 55 to 79 and does not provide clear benefits for cardiovascular risk reduction, and controlling hemoglobin A_1c to less than 6.0% is associated with increased mortality in older adults.³⁸

The American Diabetes Association³¹ and the American Geriatrics Society³⁹ recommend hemoglobin A_1c goals of:

• 7.5% or less for older adults with 3 or more coexisting chronic illnesses requiring medical intervention (eg, arthritis, hypertension, and heart failure) and with intact cognition and function
• 8.0% or less for those identified as frail, or with multiple chronic illnesses or moderate cognitive or functional impairment
• 8.5% or 9.0% or less for those with very complex comorbidities, in long-term care, or with end-stage chronic illnesses (eg, end-stage heart failure), or with moderate to severe cognitive or functional limitation.

These guidelines do not endorse a specific frailty assessment, although the references allude to the Fried phenotype criteria, which include gait speed. An update from the American Diabetes Association provides a patient-centered approach to tailoring treatment regimens, taking into consideration the risk of hypoglycemia for each class of drugs, side effects, and cost.⁴⁰

Our recommendations. Hyperglycemia remains a risk factor for cardiovascular disease in older adults and increases the risk of many geriatric conditions including delirium, dementia, frailty, and functional decline. The goal in individualizing hemoglobin A_1c goals should be to avoid both hyper- and hypoglycemia.

Sulfonylureas and insulins should be used with caution, as they have the highest associated incidence of hypoglycemia of the diabetes medications.

For secondary prevention in older adults with a history of cardiovascular disease, pooled trials have consistently demonstrated a long-term benefit for aspirin use that exceeds bleeding risks, although age and frailty status were not considered.⁴¹

Aspirin for primary prevention?

The evidence for aspirin for primary prevention in older adults is mixed. Meta-analysis suggests a modest decrease in risk of nonfatal myocardial infarction but no appreciable effects on nonfatal stroke and cardiovascular death.⁴²

The Japanese Primary Prevention Project,⁴³ a randomized trial of low-dose aspirin for primary prevention of cardiovascular disease in adults ages 60 to 85, showed no reduction in major cardiovascular events. However, the event rate was lower than expected, the crossover rates were high, the incidence of hemorrhagic strokes was higher than in Western studies, and the trial may have been underpowered to detect the benefits of aspirin.

The US Preventive Services Task Force⁴⁴ in 2016 noted that among individuals with a 10-year cardiovascular disease risk of 10% or higher based on the ACC/AHA pooled cohort equation,³ the greatest benefit of aspirin was in those ages 50 to 59. In this age group, 225 nonfatal myocardial infarctions and 84 nonfatal strokes were prevented per 10,000 men treated, with a net gain of 333 life-years. Similar findings were noted in women.

However, in those ages 60 to 69, the risks of harm begin to rise and the benefit of starting daily aspirin necessitates individualized clinical decision-making, with particular attention to bleeding risk and life expectancy.⁴⁴

In those age 70 and older, data on benefit and harm are mixed. The bleeding risk of aspirin increases with age, predominantly due to gastrointestinal bleeding.⁴⁴

The ongoing Aspirin in Reducing Events in Elderly trial will add to the evidence.

Aspirin recommendations for primary prevention

The American Geriatrics Society Beers Criteria do not routinely recommend aspirin use for primary prevention in those over age 80, even in those with diabetes.⁴⁵

Our recommendations. In adults over age 75 who are not frail but are identified as being at moderate to high risk of cardiovascular disease using either the ACC/AHA calculator or any other risk estimator, and without a limited life expectancy, we believe it is reasonable to consider low-dose aspirin (75–100 mg daily) for primary prevention. However, there must be careful consideration particularly for those at risk of major bleeding. One approach to consider would be the addition of a proton pump inhibitor along with aspirin, though this requires further study.⁴⁶

For those who have been on aspirin for primary prevention and are now older than age 80 without an adverse bleeding event, it is reasonable to stop aspirin, although risks and benefits of discontinuing aspirin should be discussed with the patient as part of shared decision-making.

In frail individuals the risks of aspirin therapy likely outweigh any benefit for primary prevention, and aspirin cannot be routinely recommended.

EXERCISE AND WEIGHT MANAGEMENT

A low body mass index is often associated with frailty, and weight loss may be a marker of underlying illness, which increases the risk of poor outcomes. However, those with an elevated body mass index and increased adiposity are in fact more likely to be frail (using the Fried physical phenotype definition) than those with a low body mass index,⁴⁷ due in part to unrecognized sarcopenic obesity, ie, replacement of lean muscle with fat.

Physical activity is currently the only intervention known to improve frailty.⁵

Physical activity and a balanced diet are just as important in older adults, including those with reduced functional ability and multiple comorbid conditions, as in younger individuals.

A trial in frail long-term care residents (mean age 87) found that high-intensity resistance training improved muscle strength and mobility.⁴⁸ The addition of a nutritional supplement with or without exercise did not affect frailty status. In community-dwelling older adults, physical activity has also been shown to improve sarcopenia and reduce falls and hip fractures.⁴⁹

Progressive resistance training has been shown to improve strength and gait speed even in those with dementia.⁵⁰

Tai chi has shown promising results in reducing falls and improving balance and function in both community-dwelling older adults and those in assisted living.^51,52

Exercise recommendations

The US Department of Health and Human Services⁵³ issued physical activity guidelines in 2008 with specific recommendations for older adults that include flexibility and balance training, which have been shown to reduce falls, in addition to aerobic activities and strength training.

Our recommendations. For all older adults, particularly those who are frail, we recommend a regimen of general daily activity, balance training such as tai chi, moderate-intensity aerobics such as cycling, resistance training such as using light weights, and stretching. Sessions lasting as little as 10 minutes are beneficial.

Gait speed can be monitored in the clinic to assess improvement in function over time.

SMOKING CESSATION

Although rates of smoking are decreasing, smoking remains one of the most important cardiovascular risk factors. Smoking has been associated with increased risk of frailty and significantly increased risk of death compared with never smoking.⁵⁴ Smoking cessation is beneficial even for those who quit later in life.

The US Department of Health and Human Services in 2008 released an update on tobacco use and dependence,⁵⁵ with specific attention to the benefit of smoking cessation for older adults.

All counseling interventions have been shown to be effective in older adults, as has nicotine replacement. Newer medications such as varenicline should be used with caution, as the risk of side effects is higher in older patients.

Samieri et al,⁵⁶ in an observational study of 10,670 nurses, found that those adhering to Mediterranean-style diets during midlife had 46% increased odds of healthy aging.

The PREDIMED study (Primary Prevention of Cardiovascular Disease With a Mediterranean Diet)⁵⁷ in adults ages 55 to 80 showed the Mediterranean diet supplemented with olive oil and nuts reduced the incidence of major cardiovascular disease.

Leon-Munoz et al.⁵⁸ A prospective study of 1,815 community-dwelling older adults followed for 3.5 years in Spain demonstrated that adhering to a Mediterranean diet was associated with a lower incidence of frailty (P = .002) and a lower risk of slow gait speed (OR 0.53, 95% CI 0.35–0.79). Interestingly, this study also found a protective association between fish and fruit consumption and frailty.

Our recommendations. A well-balanced, diverse diet rich in whole grains, fruits, vegetables, nuts, fish, and healthy fats (polyunsaturated fatty acids), with a moderate amount of lean meats, is recommended to prevent heart disease. However, poor dental health may limit the ability of older individuals to adhere to such diets, and modifications may be needed. Additionally, age-related changes in taste and smell may contribute to poor nutrition and unintended weight loss.⁵⁹ Involving a nutritionist and social worker in the patient care team should be considered especially as poor nutrition may be a sign of cognitive impairment, functional decline, and frailty.

SPECIAL CONSIDERATIONS

Special considerations when managing cardiovascular risk in the older adult include polypharmacy, multimorbidity, quality of life, and the patient’s personal preferences.

Polypharmacy, defined as taking more than 5 medications, is associated with an increased risk of adverse drug events, falls, fractures, decreased adherence, and “prescribing cascade”— prescribing more drugs to treat side effects of the first drug (eg, adding hypertensive medications to treat hypertension induced by nonsteroidal anti-inflammatory drugs).⁶⁰ This is particularly important when considering adding additional medications. If a statin will be the 20th pill, it may be less beneficial and more likely to lead to additional adverse effects than if it is the fifth medication.

Patient preferences are critically important, particularly when adding or removing medications. Interventions should include a detailed medication review for appropriate prescribing and deprescribing, referral to a pharmacist, and engaging the patient’s support system.

Multimorbidity. Many older individuals have multiple chronic illnesses. The interaction of multiple conditions must be considered in creating a comprehensive plan, including prognosis, patient preference, available evidence, treatment interactions, and risks and benefits.

Quality of life. Outlook on life and choices made regarding prolongation vs quality of life may be different for the older patient than the younger patient.

Personal preferences. Although interventions such as high-intensity statins for a robust 85-year-old may be appropriate, the individual can choose to forgo any treatment. It is important to explore the patient’s goals of care and advanced directives as part of shared decision-making when building a patient-centered prevention plan.⁶¹

ONE SIZE DOES NOT FIT ALL

The heterogeneity of aging rules out a one-size-fits-all recommendation for cardiovascular disease prevention and management of cardiovascular risk factors in older adults.

There is significant overlap between cardiovascular risk status and frailty.

Incorporating frailty into the creation of a cardiovascular risk prescription can aid in the development of an individualized care plan for the prevention of cardiovascular disease in the aging population.

When assessing and attempting to modify the risk of cardiovascular disease in older patients, physicians should consider incorporating the concept of frailty. The balance of risk and benefit may differ considerably for 2 patients of the same age if one is fit and the other is frail. Because the aging population is a diverse group, a one-size-fits-all approach to cardiovascular disease prevention and risk-factor management is not appropriate.

See related editorial

A GROWING, DIVERSE GROUP

The number of older adults with multiple cardiovascular risk factors is increasing as life expectancy improves. US residents who are age 65 today can expect to live to an average age of 84 (men) or 87 (women).¹

However, the range of life expectancy for people reaching these advanced ages is wide, and chronologic age is no longer sufficient to determine a patient’s risk profile. Furthermore, the prevalence of cardiovascular disease rises with age, and age itself is the strongest predictor of cardiovascular risk.²

Current risk calculators have not been validated in people over age 80,² making them inadequate for use in older patients. Age alone cannot identify who will benefit from preventive strategies, except in situations when a dominant disease such as metastatic cancer, end-stage renal disease, end-stage dementia, or end-stage heart failure is expected to lead to mortality within a year. Guidelines for treating common risk factors such as elevated cholesterol³ in the general population have generally not focused on adults over 75 or recognized their diversity in health status.⁴ In order to generate an individualized prescription for cardiovascular disease prevention for older adults, issues such as frailty, cognitive and functional status, disability, and comorbidity must be considered.

WHAT IS FRAILTY?

Clinicians have recognized frailty for decades, but to date there remains a debate on how to define it.

Clegg et al⁵ described frailty as “a state of increased vulnerability to poor resolution of homeostasis after a stressor event,”⁵ a definition generally agreed upon, as frailty predicts both poor health outcomes and death.

Indeed, in a prospective study of 5,317 men and women ranging in age from 65 to 101, those identified as frail at baseline were 6 times more likely to have died 3 years later (mortality rates 18% vs 3%), and the difference persisted at 7 years.⁶ After adjusting for comorbidities, those identified as frail were also more likely to fall, develop limitations in mobility or activities of daily living, or be hospitalized.

The two current leading theories of frailty were defined by Fried et al⁶ and by Rockwood and Mitnitski.⁷

Fried et al⁶ have operationalized frailty as a “physical phenotype,” defined as 3 or more of the following:

• Unintentional weight loss of 10 pounds in the past year
• Self-reported exhaustion
• Weakness as measured by grip strength
• Slow walking speed
• Decreased physical activity.⁶

Rockwood and Mitnitski⁷ define frailty as an accumulation of health-related deficits over time. They recommend that 30 to 40 possible deficits that cover a variety of health systems be included such as cognition, mood, function, and comorbidity. These are added and divided by the total possible number of variables to generate a score between 0 and 1.⁸

The difficulty in defining frailty has led to varying estimates of its prevalence, ranging from 25% to 50% in adults over 65 who have cardiovascular disease.⁹

CAUSE AND CONSEQUENCE OF CARDIOVASCULAR DISEASE

Studies have highlighted the bidirectional connection between frailty and cardiovascular disease.¹⁰ Frailty may predict cardiovascular disease, while cardiovascular disease is associated with an increased risk of incident frailty.^9,11

Frail adults with cardiovascular disease have a higher risk of poor outcomes, even after correcting for age, comorbidities, disability, and disease severity. For example, frailty is associated with a twofold higher mortality rate in individuals with cardiovascular disease.⁹

A prospective cohort study¹² of 3,895 middle-aged men and women demonstrated that those with an elevated cardiovascular risk score were at increased risk of frailty over 10 years (odds ratio [OR] 1.35, 95% confidence interval [CI] 1.21–1.51) and incident cardiovascular events (OR 1.36, 95% CI 1.15–1.61). This suggests that modification of cardiovascular risk factors earlier in life may reduce the risk of subsequently becoming frail.

Biologic mechanisms that may explain the connection between frailty and cardiovascular disease include derangements in inflammatory, hematologic, and endocrine pathways. People who are found to be clinically frail are more likely to have insulin resistance and elevated biomarkers such as C-reactive protein, D-dimer, and factor VIII.¹³ The inflammatory cytokine interleukin 6 is suggested as a common link between inflammation and thrombosis, perhaps contributing to the connection between cardiovascular disease and frailty. Many of these biomarkers have been linked to the pathophysiologic changes of aging, so-called “inflamm-aging” or immunosenescence, including sarcopenia, osteoporosis, and cardiovascular disease.¹⁴

For adults over age 70, frailty assessment is an important first step in managing cardiovascular disease risk.¹⁵ Frailty status will better identify those at risk of adverse outcomes in the short term and those who are most likely to benefit from long-term cardiovascular preventive strategies. Additionally, incorporating frailty assessment into traditional risk factor evaluation may permit appropriate intervention and prevention of a potentially modifiable risk factor.

Gait speed is a quick, easy, inexpensive, and sensitive way to assess frailty status, with excellent inter-rater and test-retest reliability, even in those with cognitive impairment.¹⁶ Slow gait speed predicts limitations in mobility, limitations in activities of daily living, and death.^8,17

In a prospective study¹⁸ of 1,567 men and women, mean age 74, slow gait speed was the strongest predictor of subsequent cardiovascular events.¹⁸

Gait speed is usually measured over a distance of 4 meters (13.1 feet),¹⁷ and the patient is asked to walk comfortably in an unobstructed, marked area. An assistive walking device can be used if needed. If possible, this is repeated once after a brief recovery period, and the average is recorded.

The FRAIL scale^19,20 is a simple, validated questionnaire that combines the Fried and Rockwood concepts of frailty and can be given over the phone or to patients in a waiting room. One point is given for each of the following, and people who have 3 or more are considered frail:

• Fatigue
• Resistance (inability to climb 1 flight of stairs)
• Ambulation (inability to walk 1 block)
• Illnesses (having more than 5)
• Loss of more than 5% of body weight.

Other measures of physical function such as grip strength (using a dynamometer), the Timed Up and Go test (assessing the ability to get up from a chair and walk a short distance), and Short Physical Performance Battery (assessing balance, chair stands, and walking speed) can be used to screen for frailty, but are more time-intensive than gait speed alone, and so are not always practical to use in a busy clinic.²¹

MANAGEMENT OF RISK FACTORS

Management of cardiovascular risk factors is best individualized as outlined below.

LOWERING HIGH BLOOD PRESSURE

The incidence of ischemic heart disease and stroke increases with age across all levels of elevated systolic and diastolic blood pressure.²² Hypertension is also associated with increased risk of cognitive decline. However, a J-shaped relationship has been observed in older adults, with increased cardiovascular events for both low and elevated blood pressure, although the clinical relevance remains controversial.²³

Odden et al²⁴ performed an observational study and found that high blood pressure was associated with an increased mortality rate in older adults with normal gait speed, while in those with slow gait speed, high blood pressure neither harmed nor helped. Those who could not walk 6 meters appeared to benefit from higher blood pressure.

HYVET (the Hypertension in the Very Elderly Trial),²⁵ a randomized controlled trial in 3,845 community-dwelling people age 80 or older with sustained systolic blood pressure higher than 160 mm Hg, found a significant reduction in rates of stroke and all-cause mortality (relative risk [RR] 0.76, P = .007) in the treatment arm using indapamide with perindopril if necessary to reach a target blood pressure of 150/80 mm Hg.

Frailty was not assessed during the trial; however, in a reanalysis, the results did not change in those identified as frail using a Rockwood frailty index (a count of health-related deficits accumulated over the lifespan).²⁶

SPRINT (the Systolic Blood Pressure Intervention Trial)²⁷ randomized participants age 50 and older with systolic blood pressure of 130 to 180 mm Hg and at increased risk of cardiovascular disease to intensive treatment (goal systolic blood pressure ≤ 120 mm Hg) or standard treatment (goal systolic blood pressure ≤ 140 mm Hg). In a prespecified subgroup of 2,636 participants over age 75 (mean age 80), hazard ratios and 95% confidence intervals for adverse outcomes with intensive treatment were:

• Major cardiovascular events: HR 0.66, 95% CI 0.51–0.85
• Death: HR 0.67, 95% CI 0.49–0.91.

Over 3 years of treatment this translated into a number needed to treat of 27 to prevent 1 cardiovascular event and 41 to prevent 1 death.

Within this subgroup, the benefit was similar regardless of level of frailty (measured both by a Rockwood frailty index and by gait speed).

However, the incidence of serious adverse treatment effects such as hypotension, orthostasis, electrolyte abnormalities, and acute kidney injury was higher with intensive treatment in the frail group. Although the difference was not statistically significant, it is cause for caution. Further, the exclusion criteria (history of diabetes, heart failure, dementia, stroke, weight loss of > 10%, nursing home residence) make it difficult to generalize the SPRINT findings to the general aging population.²⁷

Tinetti et al²⁸ performed an observational study using a nationally representative sample of older adults. They found that receiving any antihypertensive therapy was associated with an increased risk of falls with serious adverse outcomes. The risks of adverse events related to antihypertensive therapy increased with age.

Managing hypertension in frail patients at risk of cardiovascular disease requires balancing the benefits vs the risks of treatment, such as polypharmacy, falls, and orthostatic hypotension.

The Eighth Joint National Committee suggests a blood pressure goal of less than 150/90 mm Hg for all adults over age 60, and less than 140/90 mm Hg for those with a history of cardiovascular disease or diabetes.²⁹

The American College of Cardiology/American Heart Association (ACC/AHA) guidelines on hypertension, recently released, recommend a new blood pressure target of <120/<80 as normal, with 120–129/<80 considered elevated, 130–139/80–89 stage 1 hypertension, and ≥140/≥90 as stage 2 hypertension.³⁰ An important caveat to these guidelines is the recommendation to measure blood pressure accurately and with accurate technique, which is often not possible in many busy clinics. These guidelines are intended to apply to older adults as well, with a note that those with multiple morbidities and limited life expectancy will benefit from a shared decision that incorporates patient preferences and clinical judgment. Little guidance is given on how to incorporate frailty, although note is made that older adults who reside in assisted living facilities and nursing homes have not been represented in randomized controlled trials.³⁰

American Diabetes Association guidelines on hypertension in patients with diabetes recommend considering functional status, frailty, and life expectancy to decide on a blood pressure goal of either 140/90 mm Hg (if fit) or 150/90 mm Hg (if frail). They do not specify how to diagnose frailty.³¹

Canadian guidelines say that in those with advanced frailty (ie, entirely dependent for personal care and activities of daily living) and short life expectancy (months), it is reasonable to liberalize the systolic blood pressure goal to 160 to 190 mm Hg.³²

Our recommendations. In both frail and nonfrail individuals without a limited life expectancy, it is reasonable to aim for a blood pressure of at least less than 140/90 mm Hg. For those at increased risk of cardiovascular disease and able to tolerate treatment, careful lowering to 130/80 mm Hg may be considered, with close attention to side effects.

Treatment should start with the lowest possible dose, be titrated slowly, and may need to be tailored to standing blood pressure to avoid orthostatic hypotension.

Home blood pressure measurements may be beneficial in monitoring treatment.

MANAGING LIPIDS

For those over age 75, data on efficacy of statins are mixed due to the small number of older adults enrolled in randomized controlled trials of these drugs. To our knowledge, no statin trial has examined the role of frailty.

The PROSPER trial (Prospective Study of Pravastatin in the Elderly at Risk)³³ randomized 5,804 patients ages 70 to 82 to receive either pravastatin or placebo. Overall, the incidence of a composite end point of major cardiovascular events was 15% lower with active treatment (P = .014). However, the mean age was 75, which does little to address the paucity of evidence for those over age 75; follow-up time was only 3 years, and subgroup analysis did not show benefit in those who did not have a history of cardiovascular disease or in women.

The JUPITER trial (Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin)³⁴ randomized 5,695 people over age 70 without cardiovascular disease to receive either rosuvastatin or placebo. Exploratory analysis showed a significant 39% reduction in all-cause mortality and major cardiovascular events with active treatment (HR 0.61, 95% CI 0.46–0.82). Over 5 years of treatment, this translates to a number needed to treat of 19 to prevent 1 major cardiovascular event and 29 to prevent 1 cardiovascular death.

The benefit of statins for primary prevention in these trials began to be apparent 2 years after treatment was initiated.

The Women’s Health Initiative,³⁵ an observational study, found no difference in incident frailty in women older than 65 taking statins for 3 years compared with those who did not take statins

Odden et al³⁶ found that although statin use is generally well tolerated, the risks of statin-associated functional and cognitive decline may outweigh the benefits in those older than 75. The ongoing Statin in Reducing Events in the Elderly (STAREE) trial may shed light on this issue.

Recommendations on lipid management

The ACC/AHA,³ in their 2013 guidelines, do not recommend routine statin treatment for primary prevention in those over age 75, given a lack of evidence from randomized controlled trials. For secondary prevention, ie, for those who have a history of atherosclerotic cardiovascular disease, they recommend moderate-intensity statin therapy in this age group.

Our recommendations. For patients over age 75 without cardiovascular disease or frailty and with a life expectancy of at least 2 years, consider offering a statin for primary prevention of cardiovascular disease as part of shared decision-making.

In those with known cardiovascular disease, it is reasonable to continue statin therapy except in situations where the life expectancy is less than 6 months.³⁷

Although moderate- or high-intensity statin therapy is recommended in current guidelines, for many older adults it is prudent to consider the lowest tolerable dose to improve adherence, with close monitoring for side effects such as myalgia and weakness.

TYPE 2 DIABETES

Evidence suggests that tight glycemic control in type 2 diabetes is harmful for adults ages 55 to 79 and does not provide clear benefits for cardiovascular risk reduction, and controlling hemoglobin A_1c to less than 6.0% is associated with increased mortality in older adults.³⁸

The American Diabetes Association³¹ and the American Geriatrics Society³⁹ recommend hemoglobin A_1c goals of:

• 7.5% or less for older adults with 3 or more coexisting chronic illnesses requiring medical intervention (eg, arthritis, hypertension, and heart failure) and with intact cognition and function
• 8.0% or less for those identified as frail, or with multiple chronic illnesses or moderate cognitive or functional impairment
• 8.5% or 9.0% or less for those with very complex comorbidities, in long-term care, or with end-stage chronic illnesses (eg, end-stage heart failure), or with moderate to severe cognitive or functional limitation.

These guidelines do not endorse a specific frailty assessment, although the references allude to the Fried phenotype criteria, which include gait speed. An update from the American Diabetes Association provides a patient-centered approach to tailoring treatment regimens, taking into consideration the risk of hypoglycemia for each class of drugs, side effects, and cost.⁴⁰

Our recommendations. Hyperglycemia remains a risk factor for cardiovascular disease in older adults and increases the risk of many geriatric conditions including delirium, dementia, frailty, and functional decline. The goal in individualizing hemoglobin A_1c goals should be to avoid both hyper- and hypoglycemia.

Sulfonylureas and insulins should be used with caution, as they have the highest associated incidence of hypoglycemia of the diabetes medications.

For secondary prevention in older adults with a history of cardiovascular disease, pooled trials have consistently demonstrated a long-term benefit for aspirin use that exceeds bleeding risks, although age and frailty status were not considered.⁴¹

Aspirin for primary prevention?

The evidence for aspirin for primary prevention in older adults is mixed. Meta-analysis suggests a modest decrease in risk of nonfatal myocardial infarction but no appreciable effects on nonfatal stroke and cardiovascular death.⁴²

The Japanese Primary Prevention Project,⁴³ a randomized trial of low-dose aspirin for primary prevention of cardiovascular disease in adults ages 60 to 85, showed no reduction in major cardiovascular events. However, the event rate was lower than expected, the crossover rates were high, the incidence of hemorrhagic strokes was higher than in Western studies, and the trial may have been underpowered to detect the benefits of aspirin.

The US Preventive Services Task Force⁴⁴ in 2016 noted that among individuals with a 10-year cardiovascular disease risk of 10% or higher based on the ACC/AHA pooled cohort equation,³ the greatest benefit of aspirin was in those ages 50 to 59. In this age group, 225 nonfatal myocardial infarctions and 84 nonfatal strokes were prevented per 10,000 men treated, with a net gain of 333 life-years. Similar findings were noted in women.

However, in those ages 60 to 69, the risks of harm begin to rise and the benefit of starting daily aspirin necessitates individualized clinical decision-making, with particular attention to bleeding risk and life expectancy.⁴⁴

In those age 70 and older, data on benefit and harm are mixed. The bleeding risk of aspirin increases with age, predominantly due to gastrointestinal bleeding.⁴⁴

The ongoing Aspirin in Reducing Events in Elderly trial will add to the evidence.

Aspirin recommendations for primary prevention

The American Geriatrics Society Beers Criteria do not routinely recommend aspirin use for primary prevention in those over age 80, even in those with diabetes.⁴⁵

Our recommendations. In adults over age 75 who are not frail but are identified as being at moderate to high risk of cardiovascular disease using either the ACC/AHA calculator or any other risk estimator, and without a limited life expectancy, we believe it is reasonable to consider low-dose aspirin (75–100 mg daily) for primary prevention. However, there must be careful consideration particularly for those at risk of major bleeding. One approach to consider would be the addition of a proton pump inhibitor along with aspirin, though this requires further study.⁴⁶

For those who have been on aspirin for primary prevention and are now older than age 80 without an adverse bleeding event, it is reasonable to stop aspirin, although risks and benefits of discontinuing aspirin should be discussed with the patient as part of shared decision-making.

In frail individuals the risks of aspirin therapy likely outweigh any benefit for primary prevention, and aspirin cannot be routinely recommended.

EXERCISE AND WEIGHT MANAGEMENT

A low body mass index is often associated with frailty, and weight loss may be a marker of underlying illness, which increases the risk of poor outcomes. However, those with an elevated body mass index and increased adiposity are in fact more likely to be frail (using the Fried physical phenotype definition) than those with a low body mass index,⁴⁷ due in part to unrecognized sarcopenic obesity, ie, replacement of lean muscle with fat.

Physical activity is currently the only intervention known to improve frailty.⁵

Physical activity and a balanced diet are just as important in older adults, including those with reduced functional ability and multiple comorbid conditions, as in younger individuals.

A trial in frail long-term care residents (mean age 87) found that high-intensity resistance training improved muscle strength and mobility.⁴⁸ The addition of a nutritional supplement with or without exercise did not affect frailty status. In community-dwelling older adults, physical activity has also been shown to improve sarcopenia and reduce falls and hip fractures.⁴⁹

Progressive resistance training has been shown to improve strength and gait speed even in those with dementia.⁵⁰

Tai chi has shown promising results in reducing falls and improving balance and function in both community-dwelling older adults and those in assisted living.^51,52

Exercise recommendations

The US Department of Health and Human Services⁵³ issued physical activity guidelines in 2008 with specific recommendations for older adults that include flexibility and balance training, which have been shown to reduce falls, in addition to aerobic activities and strength training.

Our recommendations. For all older adults, particularly those who are frail, we recommend a regimen of general daily activity, balance training such as tai chi, moderate-intensity aerobics such as cycling, resistance training such as using light weights, and stretching. Sessions lasting as little as 10 minutes are beneficial.

Gait speed can be monitored in the clinic to assess improvement in function over time.

SMOKING CESSATION

Although rates of smoking are decreasing, smoking remains one of the most important cardiovascular risk factors. Smoking has been associated with increased risk of frailty and significantly increased risk of death compared with never smoking.⁵⁴ Smoking cessation is beneficial even for those who quit later in life.

The US Department of Health and Human Services in 2008 released an update on tobacco use and dependence,⁵⁵ with specific attention to the benefit of smoking cessation for older adults.

All counseling interventions have been shown to be effective in older adults, as has nicotine replacement. Newer medications such as varenicline should be used with caution, as the risk of side effects is higher in older patients.

Samieri et al,⁵⁶ in an observational study of 10,670 nurses, found that those adhering to Mediterranean-style diets during midlife had 46% increased odds of healthy aging.

The PREDIMED study (Primary Prevention of Cardiovascular Disease With a Mediterranean Diet)⁵⁷ in adults ages 55 to 80 showed the Mediterranean diet supplemented with olive oil and nuts reduced the incidence of major cardiovascular disease.

Leon-Munoz et al.⁵⁸ A prospective study of 1,815 community-dwelling older adults followed for 3.5 years in Spain demonstrated that adhering to a Mediterranean diet was associated with a lower incidence of frailty (P = .002) and a lower risk of slow gait speed (OR 0.53, 95% CI 0.35–0.79). Interestingly, this study also found a protective association between fish and fruit consumption and frailty.

Our recommendations. A well-balanced, diverse diet rich in whole grains, fruits, vegetables, nuts, fish, and healthy fats (polyunsaturated fatty acids), with a moderate amount of lean meats, is recommended to prevent heart disease. However, poor dental health may limit the ability of older individuals to adhere to such diets, and modifications may be needed. Additionally, age-related changes in taste and smell may contribute to poor nutrition and unintended weight loss.⁵⁹ Involving a nutritionist and social worker in the patient care team should be considered especially as poor nutrition may be a sign of cognitive impairment, functional decline, and frailty.

SPECIAL CONSIDERATIONS

Special considerations when managing cardiovascular risk in the older adult include polypharmacy, multimorbidity, quality of life, and the patient’s personal preferences.

Polypharmacy, defined as taking more than 5 medications, is associated with an increased risk of adverse drug events, falls, fractures, decreased adherence, and “prescribing cascade”— prescribing more drugs to treat side effects of the first drug (eg, adding hypertensive medications to treat hypertension induced by nonsteroidal anti-inflammatory drugs).⁶⁰ This is particularly important when considering adding additional medications. If a statin will be the 20th pill, it may be less beneficial and more likely to lead to additional adverse effects than if it is the fifth medication.

Patient preferences are critically important, particularly when adding or removing medications. Interventions should include a detailed medication review for appropriate prescribing and deprescribing, referral to a pharmacist, and engaging the patient’s support system.

Multimorbidity. Many older individuals have multiple chronic illnesses. The interaction of multiple conditions must be considered in creating a comprehensive plan, including prognosis, patient preference, available evidence, treatment interactions, and risks and benefits.

Quality of life. Outlook on life and choices made regarding prolongation vs quality of life may be different for the older patient than the younger patient.

Personal preferences. Although interventions such as high-intensity statins for a robust 85-year-old may be appropriate, the individual can choose to forgo any treatment. It is important to explore the patient’s goals of care and advanced directives as part of shared decision-making when building a patient-centered prevention plan.⁶¹

ONE SIZE DOES NOT FIT ALL

The heterogeneity of aging rules out a one-size-fits-all recommendation for cardiovascular disease prevention and management of cardiovascular risk factors in older adults.

There is significant overlap between cardiovascular risk status and frailty.

Incorporating frailty into the creation of a cardiovascular risk prescription can aid in the development of an individualized care plan for the prevention of cardiovascular disease in the aging population.

When assessing and attempting to modify the risk of cardiovascular disease in older patients, physicians should consider incorporating the concept of frailty. The balance of risk and benefit may differ considerably for 2 patients of the same age if one is fit and the other is frail. Because the aging population is a diverse group, a one-size-fits-all approach to cardiovascular disease prevention and risk-factor management is not appropriate.

See related editorial

A GROWING, DIVERSE GROUP

The number of older adults with multiple cardiovascular risk factors is increasing as life expectancy improves. US residents who are age 65 today can expect to live to an average age of 84 (men) or 87 (women).¹

However, the range of life expectancy for people reaching these advanced ages is wide, and chronologic age is no longer sufficient to determine a patient’s risk profile. Furthermore, the prevalence of cardiovascular disease rises with age, and age itself is the strongest predictor of cardiovascular risk.²

Current risk calculators have not been validated in people over age 80,² making them inadequate for use in older patients. Age alone cannot identify who will benefit from preventive strategies, except in situations when a dominant disease such as metastatic cancer, end-stage renal disease, end-stage dementia, or end-stage heart failure is expected to lead to mortality within a year. Guidelines for treating common risk factors such as elevated cholesterol³ in the general population have generally not focused on adults over 75 or recognized their diversity in health status.⁴ In order to generate an individualized prescription for cardiovascular disease prevention for older adults, issues such as frailty, cognitive and functional status, disability, and comorbidity must be considered.

WHAT IS FRAILTY?

Clinicians have recognized frailty for decades, but to date there remains a debate on how to define it.

Clegg et al⁵ described frailty as “a state of increased vulnerability to poor resolution of homeostasis after a stressor event,”⁵ a definition generally agreed upon, as frailty predicts both poor health outcomes and death.

Indeed, in a prospective study of 5,317 men and women ranging in age from 65 to 101, those identified as frail at baseline were 6 times more likely to have died 3 years later (mortality rates 18% vs 3%), and the difference persisted at 7 years.⁶ After adjusting for comorbidities, those identified as frail were also more likely to fall, develop limitations in mobility or activities of daily living, or be hospitalized.

The two current leading theories of frailty were defined by Fried et al⁶ and by Rockwood and Mitnitski.⁷

Fried et al⁶ have operationalized frailty as a “physical phenotype,” defined as 3 or more of the following:

• Unintentional weight loss of 10 pounds in the past year
• Self-reported exhaustion
• Weakness as measured by grip strength
• Slow walking speed
• Decreased physical activity.⁶

Rockwood and Mitnitski⁷ define frailty as an accumulation of health-related deficits over time. They recommend that 30 to 40 possible deficits that cover a variety of health systems be included such as cognition, mood, function, and comorbidity. These are added and divided by the total possible number of variables to generate a score between 0 and 1.⁸

The difficulty in defining frailty has led to varying estimates of its prevalence, ranging from 25% to 50% in adults over 65 who have cardiovascular disease.⁹

CAUSE AND CONSEQUENCE OF CARDIOVASCULAR DISEASE

Studies have highlighted the bidirectional connection between frailty and cardiovascular disease.¹⁰ Frailty may predict cardiovascular disease, while cardiovascular disease is associated with an increased risk of incident frailty.^9,11

Frail adults with cardiovascular disease have a higher risk of poor outcomes, even after correcting for age, comorbidities, disability, and disease severity. For example, frailty is associated with a twofold higher mortality rate in individuals with cardiovascular disease.⁹

A prospective cohort study¹² of 3,895 middle-aged men and women demonstrated that those with an elevated cardiovascular risk score were at increased risk of frailty over 10 years (odds ratio [OR] 1.35, 95% confidence interval [CI] 1.21–1.51) and incident cardiovascular events (OR 1.36, 95% CI 1.15–1.61). This suggests that modification of cardiovascular risk factors earlier in life may reduce the risk of subsequently becoming frail.

Biologic mechanisms that may explain the connection between frailty and cardiovascular disease include derangements in inflammatory, hematologic, and endocrine pathways. People who are found to be clinically frail are more likely to have insulin resistance and elevated biomarkers such as C-reactive protein, D-dimer, and factor VIII.¹³ The inflammatory cytokine interleukin 6 is suggested as a common link between inflammation and thrombosis, perhaps contributing to the connection between cardiovascular disease and frailty. Many of these biomarkers have been linked to the pathophysiologic changes of aging, so-called “inflamm-aging” or immunosenescence, including sarcopenia, osteoporosis, and cardiovascular disease.¹⁴

For adults over age 70, frailty assessment is an important first step in managing cardiovascular disease risk.¹⁵ Frailty status will better identify those at risk of adverse outcomes in the short term and those who are most likely to benefit from long-term cardiovascular preventive strategies. Additionally, incorporating frailty assessment into traditional risk factor evaluation may permit appropriate intervention and prevention of a potentially modifiable risk factor.

Gait speed is a quick, easy, inexpensive, and sensitive way to assess frailty status, with excellent inter-rater and test-retest reliability, even in those with cognitive impairment.¹⁶ Slow gait speed predicts limitations in mobility, limitations in activities of daily living, and death.^8,17

In a prospective study¹⁸ of 1,567 men and women, mean age 74, slow gait speed was the strongest predictor of subsequent cardiovascular events.¹⁸

Gait speed is usually measured over a distance of 4 meters (13.1 feet),¹⁷ and the patient is asked to walk comfortably in an unobstructed, marked area. An assistive walking device can be used if needed. If possible, this is repeated once after a brief recovery period, and the average is recorded.

The FRAIL scale^19,20 is a simple, validated questionnaire that combines the Fried and Rockwood concepts of frailty and can be given over the phone or to patients in a waiting room. One point is given for each of the following, and people who have 3 or more are considered frail:

• Fatigue
• Resistance (inability to climb 1 flight of stairs)
• Ambulation (inability to walk 1 block)
• Illnesses (having more than 5)
• Loss of more than 5% of body weight.

Other measures of physical function such as grip strength (using a dynamometer), the Timed Up and Go test (assessing the ability to get up from a chair and walk a short distance), and Short Physical Performance Battery (assessing balance, chair stands, and walking speed) can be used to screen for frailty, but are more time-intensive than gait speed alone, and so are not always practical to use in a busy clinic.²¹

MANAGEMENT OF RISK FACTORS

Management of cardiovascular risk factors is best individualized as outlined below.

LOWERING HIGH BLOOD PRESSURE

The incidence of ischemic heart disease and stroke increases with age across all levels of elevated systolic and diastolic blood pressure.²² Hypertension is also associated with increased risk of cognitive decline. However, a J-shaped relationship has been observed in older adults, with increased cardiovascular events for both low and elevated blood pressure, although the clinical relevance remains controversial.²³

Odden et al²⁴ performed an observational study and found that high blood pressure was associated with an increased mortality rate in older adults with normal gait speed, while in those with slow gait speed, high blood pressure neither harmed nor helped. Those who could not walk 6 meters appeared to benefit from higher blood pressure.

HYVET (the Hypertension in the Very Elderly Trial),²⁵ a randomized controlled trial in 3,845 community-dwelling people age 80 or older with sustained systolic blood pressure higher than 160 mm Hg, found a significant reduction in rates of stroke and all-cause mortality (relative risk [RR] 0.76, P = .007) in the treatment arm using indapamide with perindopril if necessary to reach a target blood pressure of 150/80 mm Hg.

Frailty was not assessed during the trial; however, in a reanalysis, the results did not change in those identified as frail using a Rockwood frailty index (a count of health-related deficits accumulated over the lifespan).²⁶

SPRINT (the Systolic Blood Pressure Intervention Trial)²⁷ randomized participants age 50 and older with systolic blood pressure of 130 to 180 mm Hg and at increased risk of cardiovascular disease to intensive treatment (goal systolic blood pressure ≤ 120 mm Hg) or standard treatment (goal systolic blood pressure ≤ 140 mm Hg). In a prespecified subgroup of 2,636 participants over age 75 (mean age 80), hazard ratios and 95% confidence intervals for adverse outcomes with intensive treatment were:

• Major cardiovascular events: HR 0.66, 95% CI 0.51–0.85
• Death: HR 0.67, 95% CI 0.49–0.91.

Over 3 years of treatment this translated into a number needed to treat of 27 to prevent 1 cardiovascular event and 41 to prevent 1 death.

Within this subgroup, the benefit was similar regardless of level of frailty (measured both by a Rockwood frailty index and by gait speed).

However, the incidence of serious adverse treatment effects such as hypotension, orthostasis, electrolyte abnormalities, and acute kidney injury was higher with intensive treatment in the frail group. Although the difference was not statistically significant, it is cause for caution. Further, the exclusion criteria (history of diabetes, heart failure, dementia, stroke, weight loss of > 10%, nursing home residence) make it difficult to generalize the SPRINT findings to the general aging population.²⁷

Tinetti et al²⁸ performed an observational study using a nationally representative sample of older adults. They found that receiving any antihypertensive therapy was associated with an increased risk of falls with serious adverse outcomes. The risks of adverse events related to antihypertensive therapy increased with age.

Managing hypertension in frail patients at risk of cardiovascular disease requires balancing the benefits vs the risks of treatment, such as polypharmacy, falls, and orthostatic hypotension.

The Eighth Joint National Committee suggests a blood pressure goal of less than 150/90 mm Hg for all adults over age 60, and less than 140/90 mm Hg for those with a history of cardiovascular disease or diabetes.²⁹

The American College of Cardiology/American Heart Association (ACC/AHA) guidelines on hypertension, recently released, recommend a new blood pressure target of <120/<80 as normal, with 120–129/<80 considered elevated, 130–139/80–89 stage 1 hypertension, and ≥140/≥90 as stage 2 hypertension.³⁰ An important caveat to these guidelines is the recommendation to measure blood pressure accurately and with accurate technique, which is often not possible in many busy clinics. These guidelines are intended to apply to older adults as well, with a note that those with multiple morbidities and limited life expectancy will benefit from a shared decision that incorporates patient preferences and clinical judgment. Little guidance is given on how to incorporate frailty, although note is made that older adults who reside in assisted living facilities and nursing homes have not been represented in randomized controlled trials.³⁰

American Diabetes Association guidelines on hypertension in patients with diabetes recommend considering functional status, frailty, and life expectancy to decide on a blood pressure goal of either 140/90 mm Hg (if fit) or 150/90 mm Hg (if frail). They do not specify how to diagnose frailty.³¹

Canadian guidelines say that in those with advanced frailty (ie, entirely dependent for personal care and activities of daily living) and short life expectancy (months), it is reasonable to liberalize the systolic blood pressure goal to 160 to 190 mm Hg.³²

Our recommendations. In both frail and nonfrail individuals without a limited life expectancy, it is reasonable to aim for a blood pressure of at least less than 140/90 mm Hg. For those at increased risk of cardiovascular disease and able to tolerate treatment, careful lowering to 130/80 mm Hg may be considered, with close attention to side effects.

Treatment should start with the lowest possible dose, be titrated slowly, and may need to be tailored to standing blood pressure to avoid orthostatic hypotension.

Home blood pressure measurements may be beneficial in monitoring treatment.

MANAGING LIPIDS

For those over age 75, data on efficacy of statins are mixed due to the small number of older adults enrolled in randomized controlled trials of these drugs. To our knowledge, no statin trial has examined the role of frailty.

The PROSPER trial (Prospective Study of Pravastatin in the Elderly at Risk)³³ randomized 5,804 patients ages 70 to 82 to receive either pravastatin or placebo. Overall, the incidence of a composite end point of major cardiovascular events was 15% lower with active treatment (P = .014). However, the mean age was 75, which does little to address the paucity of evidence for those over age 75; follow-up time was only 3 years, and subgroup analysis did not show benefit in those who did not have a history of cardiovascular disease or in women.

The JUPITER trial (Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin)³⁴ randomized 5,695 people over age 70 without cardiovascular disease to receive either rosuvastatin or placebo. Exploratory analysis showed a significant 39% reduction in all-cause mortality and major cardiovascular events with active treatment (HR 0.61, 95% CI 0.46–0.82). Over 5 years of treatment, this translates to a number needed to treat of 19 to prevent 1 major cardiovascular event and 29 to prevent 1 cardiovascular death.

The benefit of statins for primary prevention in these trials began to be apparent 2 years after treatment was initiated.

The Women’s Health Initiative,³⁵ an observational study, found no difference in incident frailty in women older than 65 taking statins for 3 years compared with those who did not take statins

Odden et al³⁶ found that although statin use is generally well tolerated, the risks of statin-associated functional and cognitive decline may outweigh the benefits in those older than 75. The ongoing Statin in Reducing Events in the Elderly (STAREE) trial may shed light on this issue.

Recommendations on lipid management

The ACC/AHA,³ in their 2013 guidelines, do not recommend routine statin treatment for primary prevention in those over age 75, given a lack of evidence from randomized controlled trials. For secondary prevention, ie, for those who have a history of atherosclerotic cardiovascular disease, they recommend moderate-intensity statin therapy in this age group.

Our recommendations. For patients over age 75 without cardiovascular disease or frailty and with a life expectancy of at least 2 years, consider offering a statin for primary prevention of cardiovascular disease as part of shared decision-making.

In those with known cardiovascular disease, it is reasonable to continue statin therapy except in situations where the life expectancy is less than 6 months.³⁷

Although moderate- or high-intensity statin therapy is recommended in current guidelines, for many older adults it is prudent to consider the lowest tolerable dose to improve adherence, with close monitoring for side effects such as myalgia and weakness.

TYPE 2 DIABETES

Evidence suggests that tight glycemic control in type 2 diabetes is harmful for adults ages 55 to 79 and does not provide clear benefits for cardiovascular risk reduction, and controlling hemoglobin A_1c to less than 6.0% is associated with increased mortality in older adults.³⁸

The American Diabetes Association³¹ and the American Geriatrics Society³⁹ recommend hemoglobin A_1c goals of:

• 7.5% or less for older adults with 3 or more coexisting chronic illnesses requiring medical intervention (eg, arthritis, hypertension, and heart failure) and with intact cognition and function
• 8.0% or less for those identified as frail, or with multiple chronic illnesses or moderate cognitive or functional impairment
• 8.5% or 9.0% or less for those with very complex comorbidities, in long-term care, or with end-stage chronic illnesses (eg, end-stage heart failure), or with moderate to severe cognitive or functional limitation.

These guidelines do not endorse a specific frailty assessment, although the references allude to the Fried phenotype criteria, which include gait speed. An update from the American Diabetes Association provides a patient-centered approach to tailoring treatment regimens, taking into consideration the risk of hypoglycemia for each class of drugs, side effects, and cost.⁴⁰

Our recommendations. Hyperglycemia remains a risk factor for cardiovascular disease in older adults and increases the risk of many geriatric conditions including delirium, dementia, frailty, and functional decline. The goal in individualizing hemoglobin A_1c goals should be to avoid both hyper- and hypoglycemia.

Sulfonylureas and insulins should be used with caution, as they have the highest associated incidence of hypoglycemia of the diabetes medications.

For secondary prevention in older adults with a history of cardiovascular disease, pooled trials have consistently demonstrated a long-term benefit for aspirin use that exceeds bleeding risks, although age and frailty status were not considered.⁴¹

Aspirin for primary prevention?

The evidence for aspirin for primary prevention in older adults is mixed. Meta-analysis suggests a modest decrease in risk of nonfatal myocardial infarction but no appreciable effects on nonfatal stroke and cardiovascular death.⁴²

The Japanese Primary Prevention Project,⁴³ a randomized trial of low-dose aspirin for primary prevention of cardiovascular disease in adults ages 60 to 85, showed no reduction in major cardiovascular events. However, the event rate was lower than expected, the crossover rates were high, the incidence of hemorrhagic strokes was higher than in Western studies, and the trial may have been underpowered to detect the benefits of aspirin.

The US Preventive Services Task Force⁴⁴ in 2016 noted that among individuals with a 10-year cardiovascular disease risk of 10% or higher based on the ACC/AHA pooled cohort equation,³ the greatest benefit of aspirin was in those ages 50 to 59. In this age group, 225 nonfatal myocardial infarctions and 84 nonfatal strokes were prevented per 10,000 men treated, with a net gain of 333 life-years. Similar findings were noted in women.

However, in those ages 60 to 69, the risks of harm begin to rise and the benefit of starting daily aspirin necessitates individualized clinical decision-making, with particular attention to bleeding risk and life expectancy.⁴⁴

In those age 70 and older, data on benefit and harm are mixed. The bleeding risk of aspirin increases with age, predominantly due to gastrointestinal bleeding.⁴⁴

The ongoing Aspirin in Reducing Events in Elderly trial will add to the evidence.

Aspirin recommendations for primary prevention

The American Geriatrics Society Beers Criteria do not routinely recommend aspirin use for primary prevention in those over age 80, even in those with diabetes.⁴⁵

Our recommendations. In adults over age 75 who are not frail but are identified as being at moderate to high risk of cardiovascular disease using either the ACC/AHA calculator or any other risk estimator, and without a limited life expectancy, we believe it is reasonable to consider low-dose aspirin (75–100 mg daily) for primary prevention. However, there must be careful consideration particularly for those at risk of major bleeding. One approach to consider would be the addition of a proton pump inhibitor along with aspirin, though this requires further study.⁴⁶

For those who have been on aspirin for primary prevention and are now older than age 80 without an adverse bleeding event, it is reasonable to stop aspirin, although risks and benefits of discontinuing aspirin should be discussed with the patient as part of shared decision-making.

In frail individuals the risks of aspirin therapy likely outweigh any benefit for primary prevention, and aspirin cannot be routinely recommended.

EXERCISE AND WEIGHT MANAGEMENT

A low body mass index is often associated with frailty, and weight loss may be a marker of underlying illness, which increases the risk of poor outcomes. However, those with an elevated body mass index and increased adiposity are in fact more likely to be frail (using the Fried physical phenotype definition) than those with a low body mass index,⁴⁷ due in part to unrecognized sarcopenic obesity, ie, replacement of lean muscle with fat.

Physical activity is currently the only intervention known to improve frailty.⁵

Physical activity and a balanced diet are just as important in older adults, including those with reduced functional ability and multiple comorbid conditions, as in younger individuals.

A trial in frail long-term care residents (mean age 87) found that high-intensity resistance training improved muscle strength and mobility.⁴⁸ The addition of a nutritional supplement with or without exercise did not affect frailty status. In community-dwelling older adults, physical activity has also been shown to improve sarcopenia and reduce falls and hip fractures.⁴⁹

Progressive resistance training has been shown to improve strength and gait speed even in those with dementia.⁵⁰

Tai chi has shown promising results in reducing falls and improving balance and function in both community-dwelling older adults and those in assisted living.^51,52

Exercise recommendations

The US Department of Health and Human Services⁵³ issued physical activity guidelines in 2008 with specific recommendations for older adults that include flexibility and balance training, which have been shown to reduce falls, in addition to aerobic activities and strength training.

Our recommendations. For all older adults, particularly those who are frail, we recommend a regimen of general daily activity, balance training such as tai chi, moderate-intensity aerobics such as cycling, resistance training such as using light weights, and stretching. Sessions lasting as little as 10 minutes are beneficial.

Gait speed can be monitored in the clinic to assess improvement in function over time.

SMOKING CESSATION

Although rates of smoking are decreasing, smoking remains one of the most important cardiovascular risk factors. Smoking has been associated with increased risk of frailty and significantly increased risk of death compared with never smoking.⁵⁴ Smoking cessation is beneficial even for those who quit later in life.

The US Department of Health and Human Services in 2008 released an update on tobacco use and dependence,⁵⁵ with specific attention to the benefit of smoking cessation for older adults.

All counseling interventions have been shown to be effective in older adults, as has nicotine replacement. Newer medications such as varenicline should be used with caution, as the risk of side effects is higher in older patients.

Samieri et al,⁵⁶ in an observational study of 10,670 nurses, found that those adhering to Mediterranean-style diets during midlife had 46% increased odds of healthy aging.

The PREDIMED study (Primary Prevention of Cardiovascular Disease With a Mediterranean Diet)⁵⁷ in adults ages 55 to 80 showed the Mediterranean diet supplemented with olive oil and nuts reduced the incidence of major cardiovascular disease.

Leon-Munoz et al.⁵⁸ A prospective study of 1,815 community-dwelling older adults followed for 3.5 years in Spain demonstrated that adhering to a Mediterranean diet was associated with a lower incidence of frailty (P = .002) and a lower risk of slow gait speed (OR 0.53, 95% CI 0.35–0.79). Interestingly, this study also found a protective association between fish and fruit consumption and frailty.

Our recommendations. A well-balanced, diverse diet rich in whole grains, fruits, vegetables, nuts, fish, and healthy fats (polyunsaturated fatty acids), with a moderate amount of lean meats, is recommended to prevent heart disease. However, poor dental health may limit the ability of older individuals to adhere to such diets, and modifications may be needed. Additionally, age-related changes in taste and smell may contribute to poor nutrition and unintended weight loss.⁵⁹ Involving a nutritionist and social worker in the patient care team should be considered especially as poor nutrition may be a sign of cognitive impairment, functional decline, and frailty.

SPECIAL CONSIDERATIONS

Special considerations when managing cardiovascular risk in the older adult include polypharmacy, multimorbidity, quality of life, and the patient’s personal preferences.

Polypharmacy, defined as taking more than 5 medications, is associated with an increased risk of adverse drug events, falls, fractures, decreased adherence, and “prescribing cascade”— prescribing more drugs to treat side effects of the first drug (eg, adding hypertensive medications to treat hypertension induced by nonsteroidal anti-inflammatory drugs).⁶⁰ This is particularly important when considering adding additional medications. If a statin will be the 20th pill, it may be less beneficial and more likely to lead to additional adverse effects than if it is the fifth medication.

Patient preferences are critically important, particularly when adding or removing medications. Interventions should include a detailed medication review for appropriate prescribing and deprescribing, referral to a pharmacist, and engaging the patient’s support system.

Multimorbidity. Many older individuals have multiple chronic illnesses. The interaction of multiple conditions must be considered in creating a comprehensive plan, including prognosis, patient preference, available evidence, treatment interactions, and risks and benefits.

Quality of life. Outlook on life and choices made regarding prolongation vs quality of life may be different for the older patient than the younger patient.

Personal preferences. Although interventions such as high-intensity statins for a robust 85-year-old may be appropriate, the individual can choose to forgo any treatment. It is important to explore the patient’s goals of care and advanced directives as part of shared decision-making when building a patient-centered prevention plan.⁶¹

ONE SIZE DOES NOT FIT ALL

The heterogeneity of aging rules out a one-size-fits-all recommendation for cardiovascular disease prevention and management of cardiovascular risk factors in older adults.

There is significant overlap between cardiovascular risk status and frailty.

Incorporating frailty into the creation of a cardiovascular risk prescription can aid in the development of an individualized care plan for the prevention of cardiovascular disease in the aging population.

Despite a marked increase in awareness in recent years surrounding the prevalence and prognosis of frailty in our aging population and its association with cardiovascular disease, itself highly prevalent in elderly cohorts, the exact pathobiological links between the 2 conditions have not been fully elucidated. As a consequence, this has led to difficulty not only in accurately defining cardiovascular risk in vulnerable elderly patients, but also in adequately mitigating against it.

See related article

It is well accepted that cardiovascular disease, whether clinical or subclinical, is associated with an increased risk of developing the frail phenotype.^1,2 Frailty, in turn, has been consistently identified as a universal marker of adverse outcomes in patients at risk of, and in patients with already manifest, cardiovascular disease.^2,3 However, whether or not frailty is its own unique risk factor for cardiovascular disease, independent of co-associated risk markers, or is merely a downstream byproduct indicating a more advanced disease state, has yet to be determined. Furthermore, the question of whether modification of frail status may impact the development and progression of cardiovascular disease has not yet been established.

The article by Orkaby et al⁴ in this issue delves deeper into this question by looking specifically at the interaction between frailty and standard risk factors as they relate to the prevention of cardiovascular disease.

NEEDED: A UNIVERSAL DEFINITION OF FRAILTY

It is important to acknowledge up front that before we can truly examine frailty as a novel risk entity in the assessment and management of cardiovascular risk in older-age patients, we need to agree on an accepted, validated definition of the phenotype as it relates to this population. As acknowledged by Orkaby et al,⁴ lack of such a standardized definition has resulted in highly variable estimates of the prevalence of frailty, ranging from 6.9% in a community-dwelling population in the original Cardiovascular Health Study to as high as 50% in older adults with manifest cardiovascular disease.^1,2

The ideal frailty assessment tool should be a simple, quantitative, objective, and universally accepted method, capable of providing a consistent, valid, reproducible definition that can then be used in real time by the clinician to determine the absolute presence or absence of the phenotype, much like hypertension or diabetes. Whether this optimal tool will turn out to be the traditional or modified version of the Fried Scale,¹ an alternative multicomponent measure such as the Deficit Index,⁵ or even the increasingly popular single-item measures such as gait speed or grip strength, remains to be determined.

Exact choice of tool is perhaps less important than the singular adoption of a universal method that can then be rigorously tried and tested in multicenter studies. Given the bulk of data to date for the original Fried phenotype and its development in an older-age community setting with a typical prevalence of cardiovascular risk factors, the Fried Scale appears a particularly suitable tool to use for this domain of disease prevention. Single-item spin-off measures from this phenotype, including gait speed, may also be useful for their increased feasibility and practicality in certain situations.

A TWO-WAY STREET

Given what we know about the pathophysiological, immunological, and inflammatory processes underlying advancing age that have also been implicated in both frailty and cardiovascular disease syndromes, how can we determine if frailty truly is an independent risk factor for cardiovascular disease or merely an epiphenomenon of the aging process?

We do know that older age is not a prerequisite for frailty, as is evident in studies of the phenotype in middle-aged (and younger) patients with advanced heart failure.⁶ We also know not only that frail populations have a higher age-adjusted prevalence of cardiovascular risk factors including diabetes and hypertension,¹ but also that community-dwellers with prefrailty (as defined in studies using the Fried criteria as 1 or 2 vs 3 present criteria) at baseline have a significantly increased risk of developing incident cardiovascular disease compared with those defined as nonfrail, even after adjustment for traditional risk factors and other biomarkers.³ Exploring the differences between these subgroups at baseline revealed that prefrailty was significantly associated with several subclinical insults that may serve as adverse vascular mediators, including insulin resistance, elevated inflammatory markers, and central adiposity.³

A substudy of the Cardiovascular Health Study also found that in over 1,200 participants without a prior history of a cardiovascular event, the presence of frailty was associated with multiple noninvasive measures of subclinical cardiovascular disease, including electrocardiographic and echocardiographic markers of left ventricular hypertrophy, carotid stenosis, and silent cerebrovascular infarcts on magnetic resonance imaging.⁷

These findings support a mechanistic link between evolving stages of frailty and a gradient of progressive cardiovascular risk, with a multifaceted dysregulation of metabolic processes known to underpin the pathogenesis of the frailty phenotype likely also triggering risk pathways (altered insulin metabolism, inflammation) involved in incident cardiovascular disease. Although the exact pathobiological pathways underlying these complex interlinked relationships between aging, frailty, and cardiovascular disease have yet to be fully elucidated, awareness of the bidirectional relationship between both morbid conditions highlights the absolute importance of modifying risk factors and subclinical conditions that are common to both.

Orkaby et al⁴ nicely lay out the guidelines for standard cardiovascular risk factor modification viewed in light of what is currently known—or not known—about how these recommendations should be interpreted for the older, frail, at-risk population. It is important to note at the outset that clinical trial data both inclusive of this population and incorporating the up-front assessment of frailty to predefine frail-or-not subgroups are sparse, and thereby evidence for how to optimize cardiovascular disease prevention in this important cohort is largely based on smaller observational studies and expert consensus.

Hypertension

However, important subanalyses derived from 2 large randomized controlled trials (Hypertension in the Very Elderly Trial [HYVET] and Systolic Blood Pressure Intervention Trial [SPRINT]) looking specifically at the impact of frail status on blood pressure treatment targets and related outcomes in elderly adults have recently been published.^8,9 Notably, both studies showed the beneficial outcomes of more intensive treatment (to 150/80 mm Hg or 120 mm Hg systolic, respectively) persisted in those characterized as frail (via Rockwood frailty index or slow gait speed).^8,9 Importantly, in the SPRINT analysis, higher event rates were seen with increasing frailty in both treatment groups; across each frailty stratum, absolute event rates were lower for the intensive treatment arm.⁹ These results were evident without a significant difference in the overall rate of serious adverse events⁹ or withdrawal rates⁸ between treatment groups.

Hypertension is the primary domain in which up-to-date clinical trial data have shown benefit for continued aggressive treatment for cardiovascular disease prevention regardless of the presence of frailty. Despite these data, in the real world, the “eyeball” frailty test often leads us to err on the side of caution regarding blood pressure management in the frail older adult. Certainly, the use of antihypertensive therapy in this population requires balanced consideration of the risk for adverse effects; the SPRINT analysis also found higher absolute rates of hypotension, falls, and acute kidney injury in the more intensively treated group.⁹ These adverse effects may be ameliorated not necessarily by modifying the target goal that is required, but by employing alternative strategies in achieving this goal, such as starting with lower doses, uptitrating more slowly, and monitoring with more frequent laboratory testing.

Currently, consensus guidelines in Canada have recommended liberalizing blood pressure treatment goals in those with “advanced frailty” associated with a shorter life expectancy.¹⁰

Dyslipidemia

Regarding the other major vascular risk factors, trials looking at the role of frailty in the targeted treatment of hyperlipidemia with statins in older patients for primary prevention of cardiovascular disease are lacking, although the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial showed a significant positive benefit for statin therapy in adults over age 70 (number needed to treat of 19 to prevent 1 major cardiovascular event, and 29 to prevent 1 cardiovascular death).¹¹ This again may be counterbalanced by the purported increased risk of cognitive and potential adverse functional effects of statins in this age group; however, trial data specific to frail status or not is required to truly assess the benefit-risk ratio in this population.

Hyperglycemia

Meanwhile, recent clinical trials looking at the impact of age, functional impairment, and burden of comorbidities (rather than specific frailty measures) on glucose-lowering targets and cardiovascular outcomes have failed to show a benefit from intensive glycemic control strategies, leading guideline societies to endorse less-stringent hemoglobin A_1c goals in this population.¹² Given the well-documented association between hyperglycemia and cardiovascular disease, as well as the purported dysregulated glucose metabolism underlying the frail phenotype, it is important that future trials looking at optimal hemoglobin A_1c targets incorporate the presence or absence of frailty to better inform specific recommendations for this population.

ONE SIZE MAY NOT FIT ALL

Overall, if both prefrailty and frailty are independent risk factors for, and a consequence of, clinical cardiovascular disease, it is worth bearing in mind that the modification of “intensive” or best practice therapies based on qualitatively assessed frailty may actually contribute to the problem. With best intentions, the negative impact of frailty on cardiovascular outcomes may be augmented by automatically assuming it to reflect a need for “therapy-light.” The adverse downstream consequences of inadequately treated cardiovascular risk factors are not in doubt, and it is important as the role of frailty becomes an increasingly recognized cofactor in the management of older adults with these risk factors that the vicious cycle underlying both syndromes is kept in mind, in order to avoid frailty becoming a harbinger of undertreatment in older, geriatric populations.

What is clear is that more prospective clinical trial data in this population are urgently needed in order to better delineate the exact interactions between frail status and these risk factors and the potential downstream consequences, using prespecified and robust frailty assessment methods.

Perhaps frailty should be seen as a series of stages rather than simply as a binary “there or not there” biomarker; through initial and established stages of the syndrome, which have been independently associated with both clinical events and subclinical surrogates of cardiovascular disease, risk factors should continue to be treated aggressively and according to best available evidence. However, as guideline societies are already beginning to endorse as highlighted above, once the phenotype becomes tethered with a certain threshold burden of comorbidity, cognitive or functional impairment, or more end-stage disease status, then goals for cardiovascular disease prevention may need to be readdressed and modified. If frailty is truly confirmed as a cardiovascular disease equivalent, not only appropriately treating associated cardiovascular risk factors but also seeking therapies that actively target the frailty syndrome itself should be an important goal of future studies seeking to impact the development of both clinical and subclinical cardiovascular disease in this population.

Despite a marked increase in awareness in recent years surrounding the prevalence and prognosis of frailty in our aging population and its association with cardiovascular disease, itself highly prevalent in elderly cohorts, the exact pathobiological links between the 2 conditions have not been fully elucidated. As a consequence, this has led to difficulty not only in accurately defining cardiovascular risk in vulnerable elderly patients, but also in adequately mitigating against it.

See related article

It is well accepted that cardiovascular disease, whether clinical or subclinical, is associated with an increased risk of developing the frail phenotype.^1,2 Frailty, in turn, has been consistently identified as a universal marker of adverse outcomes in patients at risk of, and in patients with already manifest, cardiovascular disease.^2,3 However, whether or not frailty is its own unique risk factor for cardiovascular disease, independent of co-associated risk markers, or is merely a downstream byproduct indicating a more advanced disease state, has yet to be determined. Furthermore, the question of whether modification of frail status may impact the development and progression of cardiovascular disease has not yet been established.

The article by Orkaby et al⁴ in this issue delves deeper into this question by looking specifically at the interaction between frailty and standard risk factors as they relate to the prevention of cardiovascular disease.

NEEDED: A UNIVERSAL DEFINITION OF FRAILTY

It is important to acknowledge up front that before we can truly examine frailty as a novel risk entity in the assessment and management of cardiovascular risk in older-age patients, we need to agree on an accepted, validated definition of the phenotype as it relates to this population. As acknowledged by Orkaby et al,⁴ lack of such a standardized definition has resulted in highly variable estimates of the prevalence of frailty, ranging from 6.9% in a community-dwelling population in the original Cardiovascular Health Study to as high as 50% in older adults with manifest cardiovascular disease.^1,2

The ideal frailty assessment tool should be a simple, quantitative, objective, and universally accepted method, capable of providing a consistent, valid, reproducible definition that can then be used in real time by the clinician to determine the absolute presence or absence of the phenotype, much like hypertension or diabetes. Whether this optimal tool will turn out to be the traditional or modified version of the Fried Scale,¹ an alternative multicomponent measure such as the Deficit Index,⁵ or even the increasingly popular single-item measures such as gait speed or grip strength, remains to be determined.

Exact choice of tool is perhaps less important than the singular adoption of a universal method that can then be rigorously tried and tested in multicenter studies. Given the bulk of data to date for the original Fried phenotype and its development in an older-age community setting with a typical prevalence of cardiovascular risk factors, the Fried Scale appears a particularly suitable tool to use for this domain of disease prevention. Single-item spin-off measures from this phenotype, including gait speed, may also be useful for their increased feasibility and practicality in certain situations.

A TWO-WAY STREET

Given what we know about the pathophysiological, immunological, and inflammatory processes underlying advancing age that have also been implicated in both frailty and cardiovascular disease syndromes, how can we determine if frailty truly is an independent risk factor for cardiovascular disease or merely an epiphenomenon of the aging process?

We do know that older age is not a prerequisite for frailty, as is evident in studies of the phenotype in middle-aged (and younger) patients with advanced heart failure.⁶ We also know not only that frail populations have a higher age-adjusted prevalence of cardiovascular risk factors including diabetes and hypertension,¹ but also that community-dwellers with prefrailty (as defined in studies using the Fried criteria as 1 or 2 vs 3 present criteria) at baseline have a significantly increased risk of developing incident cardiovascular disease compared with those defined as nonfrail, even after adjustment for traditional risk factors and other biomarkers.³ Exploring the differences between these subgroups at baseline revealed that prefrailty was significantly associated with several subclinical insults that may serve as adverse vascular mediators, including insulin resistance, elevated inflammatory markers, and central adiposity.³

A substudy of the Cardiovascular Health Study also found that in over 1,200 participants without a prior history of a cardiovascular event, the presence of frailty was associated with multiple noninvasive measures of subclinical cardiovascular disease, including electrocardiographic and echocardiographic markers of left ventricular hypertrophy, carotid stenosis, and silent cerebrovascular infarcts on magnetic resonance imaging.⁷

These findings support a mechanistic link between evolving stages of frailty and a gradient of progressive cardiovascular risk, with a multifaceted dysregulation of metabolic processes known to underpin the pathogenesis of the frailty phenotype likely also triggering risk pathways (altered insulin metabolism, inflammation) involved in incident cardiovascular disease. Although the exact pathobiological pathways underlying these complex interlinked relationships between aging, frailty, and cardiovascular disease have yet to be fully elucidated, awareness of the bidirectional relationship between both morbid conditions highlights the absolute importance of modifying risk factors and subclinical conditions that are common to both.

Orkaby et al⁴ nicely lay out the guidelines for standard cardiovascular risk factor modification viewed in light of what is currently known—or not known—about how these recommendations should be interpreted for the older, frail, at-risk population. It is important to note at the outset that clinical trial data both inclusive of this population and incorporating the up-front assessment of frailty to predefine frail-or-not subgroups are sparse, and thereby evidence for how to optimize cardiovascular disease prevention in this important cohort is largely based on smaller observational studies and expert consensus.

Hypertension

However, important subanalyses derived from 2 large randomized controlled trials (Hypertension in the Very Elderly Trial [HYVET] and Systolic Blood Pressure Intervention Trial [SPRINT]) looking specifically at the impact of frail status on blood pressure treatment targets and related outcomes in elderly adults have recently been published.^8,9 Notably, both studies showed the beneficial outcomes of more intensive treatment (to 150/80 mm Hg or 120 mm Hg systolic, respectively) persisted in those characterized as frail (via Rockwood frailty index or slow gait speed).^8,9 Importantly, in the SPRINT analysis, higher event rates were seen with increasing frailty in both treatment groups; across each frailty stratum, absolute event rates were lower for the intensive treatment arm.⁹ These results were evident without a significant difference in the overall rate of serious adverse events⁹ or withdrawal rates⁸ between treatment groups.

Hypertension is the primary domain in which up-to-date clinical trial data have shown benefit for continued aggressive treatment for cardiovascular disease prevention regardless of the presence of frailty. Despite these data, in the real world, the “eyeball” frailty test often leads us to err on the side of caution regarding blood pressure management in the frail older adult. Certainly, the use of antihypertensive therapy in this population requires balanced consideration of the risk for adverse effects; the SPRINT analysis also found higher absolute rates of hypotension, falls, and acute kidney injury in the more intensively treated group.⁹ These adverse effects may be ameliorated not necessarily by modifying the target goal that is required, but by employing alternative strategies in achieving this goal, such as starting with lower doses, uptitrating more slowly, and monitoring with more frequent laboratory testing.

Currently, consensus guidelines in Canada have recommended liberalizing blood pressure treatment goals in those with “advanced frailty” associated with a shorter life expectancy.¹⁰

Dyslipidemia

Regarding the other major vascular risk factors, trials looking at the role of frailty in the targeted treatment of hyperlipidemia with statins in older patients for primary prevention of cardiovascular disease are lacking, although the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial showed a significant positive benefit for statin therapy in adults over age 70 (number needed to treat of 19 to prevent 1 major cardiovascular event, and 29 to prevent 1 cardiovascular death).¹¹ This again may be counterbalanced by the purported increased risk of cognitive and potential adverse functional effects of statins in this age group; however, trial data specific to frail status or not is required to truly assess the benefit-risk ratio in this population.

Hyperglycemia

Meanwhile, recent clinical trials looking at the impact of age, functional impairment, and burden of comorbidities (rather than specific frailty measures) on glucose-lowering targets and cardiovascular outcomes have failed to show a benefit from intensive glycemic control strategies, leading guideline societies to endorse less-stringent hemoglobin A_1c goals in this population.¹² Given the well-documented association between hyperglycemia and cardiovascular disease, as well as the purported dysregulated glucose metabolism underlying the frail phenotype, it is important that future trials looking at optimal hemoglobin A_1c targets incorporate the presence or absence of frailty to better inform specific recommendations for this population.

ONE SIZE MAY NOT FIT ALL

Overall, if both prefrailty and frailty are independent risk factors for, and a consequence of, clinical cardiovascular disease, it is worth bearing in mind that the modification of “intensive” or best practice therapies based on qualitatively assessed frailty may actually contribute to the problem. With best intentions, the negative impact of frailty on cardiovascular outcomes may be augmented by automatically assuming it to reflect a need for “therapy-light.” The adverse downstream consequences of inadequately treated cardiovascular risk factors are not in doubt, and it is important as the role of frailty becomes an increasingly recognized cofactor in the management of older adults with these risk factors that the vicious cycle underlying both syndromes is kept in mind, in order to avoid frailty becoming a harbinger of undertreatment in older, geriatric populations.

What is clear is that more prospective clinical trial data in this population are urgently needed in order to better delineate the exact interactions between frail status and these risk factors and the potential downstream consequences, using prespecified and robust frailty assessment methods.

Perhaps frailty should be seen as a series of stages rather than simply as a binary “there or not there” biomarker; through initial and established stages of the syndrome, which have been independently associated with both clinical events and subclinical surrogates of cardiovascular disease, risk factors should continue to be treated aggressively and according to best available evidence. However, as guideline societies are already beginning to endorse as highlighted above, once the phenotype becomes tethered with a certain threshold burden of comorbidity, cognitive or functional impairment, or more end-stage disease status, then goals for cardiovascular disease prevention may need to be readdressed and modified. If frailty is truly confirmed as a cardiovascular disease equivalent, not only appropriately treating associated cardiovascular risk factors but also seeking therapies that actively target the frailty syndrome itself should be an important goal of future studies seeking to impact the development of both clinical and subclinical cardiovascular disease in this population.

Despite a marked increase in awareness in recent years surrounding the prevalence and prognosis of frailty in our aging population and its association with cardiovascular disease, itself highly prevalent in elderly cohorts, the exact pathobiological links between the 2 conditions have not been fully elucidated. As a consequence, this has led to difficulty not only in accurately defining cardiovascular risk in vulnerable elderly patients, but also in adequately mitigating against it.

See related article

It is well accepted that cardiovascular disease, whether clinical or subclinical, is associated with an increased risk of developing the frail phenotype.^1,2 Frailty, in turn, has been consistently identified as a universal marker of adverse outcomes in patients at risk of, and in patients with already manifest, cardiovascular disease.^2,3 However, whether or not frailty is its own unique risk factor for cardiovascular disease, independent of co-associated risk markers, or is merely a downstream byproduct indicating a more advanced disease state, has yet to be determined. Furthermore, the question of whether modification of frail status may impact the development and progression of cardiovascular disease has not yet been established.

The article by Orkaby et al⁴ in this issue delves deeper into this question by looking specifically at the interaction between frailty and standard risk factors as they relate to the prevention of cardiovascular disease.

NEEDED: A UNIVERSAL DEFINITION OF FRAILTY

It is important to acknowledge up front that before we can truly examine frailty as a novel risk entity in the assessment and management of cardiovascular risk in older-age patients, we need to agree on an accepted, validated definition of the phenotype as it relates to this population. As acknowledged by Orkaby et al,⁴ lack of such a standardized definition has resulted in highly variable estimates of the prevalence of frailty, ranging from 6.9% in a community-dwelling population in the original Cardiovascular Health Study to as high as 50% in older adults with manifest cardiovascular disease.^1,2

The ideal frailty assessment tool should be a simple, quantitative, objective, and universally accepted method, capable of providing a consistent, valid, reproducible definition that can then be used in real time by the clinician to determine the absolute presence or absence of the phenotype, much like hypertension or diabetes. Whether this optimal tool will turn out to be the traditional or modified version of the Fried Scale,¹ an alternative multicomponent measure such as the Deficit Index,⁵ or even the increasingly popular single-item measures such as gait speed or grip strength, remains to be determined.

Exact choice of tool is perhaps less important than the singular adoption of a universal method that can then be rigorously tried and tested in multicenter studies. Given the bulk of data to date for the original Fried phenotype and its development in an older-age community setting with a typical prevalence of cardiovascular risk factors, the Fried Scale appears a particularly suitable tool to use for this domain of disease prevention. Single-item spin-off measures from this phenotype, including gait speed, may also be useful for their increased feasibility and practicality in certain situations.

A TWO-WAY STREET

Given what we know about the pathophysiological, immunological, and inflammatory processes underlying advancing age that have also been implicated in both frailty and cardiovascular disease syndromes, how can we determine if frailty truly is an independent risk factor for cardiovascular disease or merely an epiphenomenon of the aging process?

We do know that older age is not a prerequisite for frailty, as is evident in studies of the phenotype in middle-aged (and younger) patients with advanced heart failure.⁶ We also know not only that frail populations have a higher age-adjusted prevalence of cardiovascular risk factors including diabetes and hypertension,¹ but also that community-dwellers with prefrailty (as defined in studies using the Fried criteria as 1 or 2 vs 3 present criteria) at baseline have a significantly increased risk of developing incident cardiovascular disease compared with those defined as nonfrail, even after adjustment for traditional risk factors and other biomarkers.³ Exploring the differences between these subgroups at baseline revealed that prefrailty was significantly associated with several subclinical insults that may serve as adverse vascular mediators, including insulin resistance, elevated inflammatory markers, and central adiposity.³

A substudy of the Cardiovascular Health Study also found that in over 1,200 participants without a prior history of a cardiovascular event, the presence of frailty was associated with multiple noninvasive measures of subclinical cardiovascular disease, including electrocardiographic and echocardiographic markers of left ventricular hypertrophy, carotid stenosis, and silent cerebrovascular infarcts on magnetic resonance imaging.⁷

These findings support a mechanistic link between evolving stages of frailty and a gradient of progressive cardiovascular risk, with a multifaceted dysregulation of metabolic processes known to underpin the pathogenesis of the frailty phenotype likely also triggering risk pathways (altered insulin metabolism, inflammation) involved in incident cardiovascular disease. Although the exact pathobiological pathways underlying these complex interlinked relationships between aging, frailty, and cardiovascular disease have yet to be fully elucidated, awareness of the bidirectional relationship between both morbid conditions highlights the absolute importance of modifying risk factors and subclinical conditions that are common to both.

Orkaby et al⁴ nicely lay out the guidelines for standard cardiovascular risk factor modification viewed in light of what is currently known—or not known—about how these recommendations should be interpreted for the older, frail, at-risk population. It is important to note at the outset that clinical trial data both inclusive of this population and incorporating the up-front assessment of frailty to predefine frail-or-not subgroups are sparse, and thereby evidence for how to optimize cardiovascular disease prevention in this important cohort is largely based on smaller observational studies and expert consensus.

Hypertension

However, important subanalyses derived from 2 large randomized controlled trials (Hypertension in the Very Elderly Trial [HYVET] and Systolic Blood Pressure Intervention Trial [SPRINT]) looking specifically at the impact of frail status on blood pressure treatment targets and related outcomes in elderly adults have recently been published.^8,9 Notably, both studies showed the beneficial outcomes of more intensive treatment (to 150/80 mm Hg or 120 mm Hg systolic, respectively) persisted in those characterized as frail (via Rockwood frailty index or slow gait speed).^8,9 Importantly, in the SPRINT analysis, higher event rates were seen with increasing frailty in both treatment groups; across each frailty stratum, absolute event rates were lower for the intensive treatment arm.⁹ These results were evident without a significant difference in the overall rate of serious adverse events⁹ or withdrawal rates⁸ between treatment groups.

Hypertension is the primary domain in which up-to-date clinical trial data have shown benefit for continued aggressive treatment for cardiovascular disease prevention regardless of the presence of frailty. Despite these data, in the real world, the “eyeball” frailty test often leads us to err on the side of caution regarding blood pressure management in the frail older adult. Certainly, the use of antihypertensive therapy in this population requires balanced consideration of the risk for adverse effects; the SPRINT analysis also found higher absolute rates of hypotension, falls, and acute kidney injury in the more intensively treated group.⁹ These adverse effects may be ameliorated not necessarily by modifying the target goal that is required, but by employing alternative strategies in achieving this goal, such as starting with lower doses, uptitrating more slowly, and monitoring with more frequent laboratory testing.

Currently, consensus guidelines in Canada have recommended liberalizing blood pressure treatment goals in those with “advanced frailty” associated with a shorter life expectancy.¹⁰

Dyslipidemia

Regarding the other major vascular risk factors, trials looking at the role of frailty in the targeted treatment of hyperlipidemia with statins in older patients for primary prevention of cardiovascular disease are lacking, although the Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin (JUPITER) trial showed a significant positive benefit for statin therapy in adults over age 70 (number needed to treat of 19 to prevent 1 major cardiovascular event, and 29 to prevent 1 cardiovascular death).¹¹ This again may be counterbalanced by the purported increased risk of cognitive and potential adverse functional effects of statins in this age group; however, trial data specific to frail status or not is required to truly assess the benefit-risk ratio in this population.

Hyperglycemia

Meanwhile, recent clinical trials looking at the impact of age, functional impairment, and burden of comorbidities (rather than specific frailty measures) on glucose-lowering targets and cardiovascular outcomes have failed to show a benefit from intensive glycemic control strategies, leading guideline societies to endorse less-stringent hemoglobin A_1c goals in this population.¹² Given the well-documented association between hyperglycemia and cardiovascular disease, as well as the purported dysregulated glucose metabolism underlying the frail phenotype, it is important that future trials looking at optimal hemoglobin A_1c targets incorporate the presence or absence of frailty to better inform specific recommendations for this population.

ONE SIZE MAY NOT FIT ALL

Overall, if both prefrailty and frailty are independent risk factors for, and a consequence of, clinical cardiovascular disease, it is worth bearing in mind that the modification of “intensive” or best practice therapies based on qualitatively assessed frailty may actually contribute to the problem. With best intentions, the negative impact of frailty on cardiovascular outcomes may be augmented by automatically assuming it to reflect a need for “therapy-light.” The adverse downstream consequences of inadequately treated cardiovascular risk factors are not in doubt, and it is important as the role of frailty becomes an increasingly recognized cofactor in the management of older adults with these risk factors that the vicious cycle underlying both syndromes is kept in mind, in order to avoid frailty becoming a harbinger of undertreatment in older, geriatric populations.

What is clear is that more prospective clinical trial data in this population are urgently needed in order to better delineate the exact interactions between frail status and these risk factors and the potential downstream consequences, using prespecified and robust frailty assessment methods.

Perhaps frailty should be seen as a series of stages rather than simply as a binary “there or not there” biomarker; through initial and established stages of the syndrome, which have been independently associated with both clinical events and subclinical surrogates of cardiovascular disease, risk factors should continue to be treated aggressively and according to best available evidence. However, as guideline societies are already beginning to endorse as highlighted above, once the phenotype becomes tethered with a certain threshold burden of comorbidity, cognitive or functional impairment, or more end-stage disease status, then goals for cardiovascular disease prevention may need to be readdressed and modified. If frailty is truly confirmed as a cardiovascular disease equivalent, not only appropriately treating associated cardiovascular risk factors but also seeking therapies that actively target the frailty syndrome itself should be an important goal of future studies seeking to impact the development of both clinical and subclinical cardiovascular disease in this population.

Hyperkalemia is common in patients with cardiovascular disease. Its consequences can be severe and life-threatening, and its management and prevention require a multidisciplinary approach that entails reducing intake of high-potassium foods, adjusting medications that cause hyperkalemia, and adding medications that reduce the plasma potassium concentration. With this approach, patients at high risk can receive the cardiorenal benefits of drugs that block the renin-angiotensin-aldosterone system without developing hyperkalemia.

98% OF POTASSIUM IS INSIDE CELLS

The body of a typical 70-kg man contains about 3,500 mmol of potassium, 98% of which is in the intracellular space; the remaining 2% is in the extracellular space. This large intracellular-to-extracellular gradient determines the cell voltage and explains why disorders in plasma potassium give rise to manifestations in excitable tissues such as the heart and nervous system.

The most important determinants of potassium distribution between the intracellular and extracellular space are insulin and beta-adrenergic receptor stimulation.

Maintenance of total-body potassium content is primarily the job of the kidneys, with a small contribution by the gastrointestinal tract.^1,2 Hyperkalemia is most commonly encountered in patients with decreased kidney function.

The normal kidney can secrete a large amount of potassium, making hyperkalemia uncommon in the absence of kidney disease. This large capacity may have evolved to handle the diet of Paleolithic humans, which contained 4 times as much potassium as contemporary diets.^3,4 With the onset of agriculture, dietary intake of potassium has progressively declined while sodium intake has risen. A popular theory suggests this mismatch between the modern diet and the nutritional requirements encoded in the human genome during evolution may contribute to chronic diseases such as hypertension, stroke, kidney stones, and bone disease.⁵

MANY POTENTIAL CAUSES OF HYPERKALEMIA

Causes of hyperkalemia are outlined in Table 1. Shifting of potassium from the cells to the extracellular space is a cause of transient hyperkalemia, while chronic hyperkalemia indicates an impairment in renal potassium secretion. The following discussion is a guide to the approach to the hyperkalemic patient.

Is the patient’s hyperkalemia really pseudohyperkalemia?

Pseudohyperkalemia, an artifact of measurement, occurs due to mechanical release of potassium from cells during phlebotomy or specimen processing.⁶ This diagnosis is made when the serum potassium concentration exceeds the plasma potassium concentration by more than 0.5 mmol/L, and should be considered when hyperkalemia occurs in the absence of a clinical risk factor. Fist-clenching, application of a tight-fitting tourniquet, or use of small-bore needles during phlebotomy can all cause pseudohyperkalemia.

Mechanism of pseudohyperkalemia. Since serum is the liquid part of blood remaining after coagulation, release of potassium from cells injured during the process of coagulation raises the potassium level in the serum. Plasma is the cell-free part of blood that has been treated with anticoagulants; it has no cells that can be injured and release potassium. Thus, the serum potassium level will be higher than that in the plasma.

Reverse pseudohyperkalemia, in contrast, occurs when the plasma potassium level is falsely elevated but the serum value is normal. This situation has been described in hematologic disorders characterized by pronounced leukocytosis in which malignant cells are prone to lysis with minimal mechanical stress due to increased fragility or altered sodium-potassium ATPase pump activity.⁷ This phenomenon is unusual but occurs because the cells are so fragile.

A spurious increase in plasma potassium concentration along with a low plasma calcium concentration raises the possibility of calcium chelation and release of potassium in a sample tube contaminated with the anticoagulant ethylenediaminetetraacetic acid.

Is there increased potassium intake?

Increased potassium intake is a potential cause of hyperkalemia in patients with decreased kidney function or adrenal disease.

Foods naturally rich in potassium include bananas (a medium-sized banana contains 451 mg or 12 mmol of potassium) and potatoes (844 mg or 22 mmol in a large baked potato with skin). Other potassium-rich foods are melons, citrus juice, and avocados. Less-obvious food sources include raw coconut juice (potassium concentration 44.3 mmol/L) and noni juice (56 mmol/L).

Salt substitutes, recommended to hypertensive patients with chronic kidney disease, can be a hidden source of dietary potassium.

Clay ingestion is a potential cause of dyskalemia. White clay consumption causes hypokalemia due to potassium binding in the gastrointestinal tract. Red clay or river bed clay, on the other hand, is enriched in potassium (100 mmol of potassium in 100 g of clay) and can cause life-threatening hyperkalemia in patients with chronic kidney disease.⁸

Eating burnt match heads. Some individuals chew and ingest burnt match heads, a condition called cautopyreiophagia. In one reported case,⁹ this activity contributed an additional 80 mmol of daily potassium intake in a dialysis patient, resulting in a plasma potassium concentration of 8 mmol/L.

Acute hyperkalemia can be the result of redistribution of cellular potassium. Shifting of as little as 2% of the body’s potassium from the intracellular to the extracellular space can double the plasma potassium concentration.

Tissue injury. Hyperkalemia frequently occurs in diseases that cause tissue injury such as rhabdomyolysis, trauma, massive hemolysis, and tumor lysis.

Insulin deficiency. Insulin and catecholamines are major regulators of potassium distribution within the body. After a meal, release of insulin not only regulates the plasma glucose concentration, it also causes potassium to move into cells until the kidneys have had sufficient time to excrete the dietary potassium load and reestablish total-body potassium content.

Exercise, beta-blockers. During exercise, potassium is released from skeletal muscle cells and accumulates in the interstitial compartment, where it exerts a vasodilatory effect. The simultaneous increase in circulating catecholamines regulates this release by promoting cell potassium uptake through beta-adrenergic receptor stimulation.

Metabolic acidosis can facilitate exit (ie, shift) of potassium from cells, but this effect depends on the type of acidosis. Hyperchloremic normal anion gap acidosis (mineral acidosis) most commonly causes this effect due to the relative impermeability of the cell membrane to the chloride anion. As hydrogen ions move into the cell due to accumulation of ammonium chloride or hydrogen chloride, electrical neutrality is maintained by potassium exit.

In contrast, organic acidosis (due to lactic, beta-hydroxybutyric, or methylmalonic acid) tends not to cause a potassium shift, since most organic anions readily cross the cell membrane along with hydrogen. Lactic acidosis is often associated with potassium shift, but this effect is due to loss of cell integrity as a result of cell ischemia. The hyperkalemia typically present on admission in patients with diabetic ketoacidosis is the result of insulin deficiency and hypertonicity and not the underlying organic acidosis.¹⁰

Hypertonic states can cause hyperkalemia due to cell shift. For example, hyperglycemia, as in diabetic ketoacidosis, pulls water from the intracellular into the extracellular compartment, thereby concentrating intracellular potassium and creating a more favorable gradient for potassium efflux through membrane channels. This same effect can occur in neurosurgical patients given large amounts of hypertonic mannitol. Repetitive doses of immunoglobulin can lead to extracellular accumulation of sorbitol, maltose, or sucrose, since these sugars are added to the preparations to prevent immunoglobulin aggregation.¹¹

Is a disturbance in renal potassium excretion present?

Sustained hyperkalemia is more commonly associated with decreases in renal potassium excretion than with a cellular shift. In most instances the clinician can distinguish between cell shift and impaired renal excretion based on the available clinical data.

The transtubular potassium gradient has been used to determine whether there is a disturbance in renal potassium excretion and to assess renal potassium handling.¹²

This calculation is based on the assumption that only water is reabsorbed past the cortical collecting duct, and not solutes. It has fallen out of favor since we have found this assumption to be incorrect; a large amount of urea is reabsorbed daily in the downstream medullary collecting duct as a result of intrarenal recycling of urea.

The one situation in which the transtubular potassium gradient may be of use is determining whether hyperkalemia is a result of low aldosterone levels as opposed to aldosterone resistance. One can compare the transtubular potassium gradient before and after a physiologic dose (0.05 mg) of 9-alpha fludrocortisone. An increase of more than 6 over a 4-hour period favors aldosterone deficiency, whereas smaller changes would indicate aldosterone resistance.

24-hour potassium excretion, spot urine potassium-creatinine ratio.  A better way to assess renal potassium handling is to measure the amount of potassium in a 24-hour urine collection or determine a spot urine potassium-creatinine ratio. A 24-hour urinary potassium excretion of less than 15 mmol or a potassium-creatinine ratio less than 1 suggests an extrarenal cause of hypokalemia. A ratio greater than 20 would be an appropriate renal response to hyperkalemia.

One or more of 3 abnormalities should be considered in the hyperkalemic patient with impaired renal excretion of potassium:

• Decreased distal delivery of sodium
• Mineralocorticoid deficiency
• Abnormal cortical collecting tubule function.¹³

Under normal circumstances, potassium is freely filtered across the glomerulus and then mostly reabsorbed in the proximal tubule and thick ascending limb. Potassium secretion begins in the distal convoluted tubule and increases in magnitude into the collecting duct. Tubular secretion is the component of potassium handling that varies and is regulated according to physiologic needs.

In acute kidney injury, the rapid decline in glomerular filtration rate and reduction in functioning nephron mass lead to decreased distal potassium secretion.

Hyperkalemia is a frequent problem when oliguria is present, since the reduction in distal delivery of sodium and water further impairs potassium secretion. Patients with oliguric acute kidney injury are more likely to have a more severe underlying disease state, and therefore tissue breakdown and catabolism further increase the risk of hyperkalemia.

In contrast, in nonoliguric patients, the renal injury tends to be less severe, and enough sodium and water are usually delivered distally to prevent hyperkalemia.

In chronic kidney disease, nephron dropout and reduction in collecting tubule mass also lead to a global decline in distal potassium secretion. However, this is countered by an increased capacity of the remaining individual nephrons for potassium secretion. High flow, increased distal sodium delivery, and increased activity and number of sodium-potassium ATPase pumps in the remaining nephrons account for this increased secretory capacity.¹⁴ As renal function declines over time, colonic potassium secretion progressively increases.¹⁵

These adaptive changes help to keep the plasma potassium concentration within the normal range until the glomerular filtration rate falls to less than 10 or 15 mL/min. Development of hyperkalemia with more modest reductions in the glomerular filtration rate suggest decreased mineralocorticoid activity or a specific lesion of the tubule.

Mineralocorticoid deficiency

Aldosterone deficiency can occur alone or in combination with decreased cortisol levels. Destruction of the adrenal glands is suggested when both hormones are reduced. Enzyme defects in cortisol metabolism can result in either isolated deficiency of aldosterone or adrenogenital syndromes associated with decreased mineralocorticoid activity.

Heparin administration leads to a reversible defect in adrenal synthesis of aldosterone. Drugs that block the stimulatory effect of angiotensin II on the zona glomerulosa cells of the adrenal gland will lower aldosterone.

Renin-angiotensin-aldosterone system blockers. Angiotensin-converting enzyme inhibitors block the formation of angiotensin II, whereas angiotensin II receptor blockers prevent angiotensin II from binding to its adrenal receptor. The direct renin inhibitor aliskiren lowers angiotensin II levels by blocking the enzymatic activity of renin and lowers the circulating levels of both angiotensin I and II.¹⁶

The syndrome of hyporeninemic hypoaldosteronism is a common cause of hyperkalemia in patients who have a glomerular filtration rate between 40 and 60 mL/min. Diabetic nephropathy and interstitial renal disease are the most common clinical entities associated with this syndrome.¹⁰ Other causes include analgesic nephropathy, urinary tract obstruction, sickle cell disease, systemic lupus erythematosus, and amyloidosis.

Nonsteroidal anti-inflammatory drugs can cause hyperkalemia by suppressing renin release and reducing delivery of sodium to the distal nephron.¹⁸

Calcineurin inhibitors impair potassium secretion by suppressing renin release and by direct tubular effects.¹⁹

Beta-blockers. Beta-1 and to a lesser extent beta-2 receptor blockade can also result in a hyporeninemic state.

Distal tubular defect

Hyperkalemia can result from interstitial renal diseases that specifically affect the distal nephron. In this setting, the glomerular filtration rate is only mildly reduced, and circulating aldosterone levels are normal.

Renal transplant, lupus erythematosus, amyloidosis, urinary obstruction, and sickle cell disease are conditions in which an impairment in renin release may coexist with a defect in tubular secretion.

Potassium-sparing diuretics impair the ability of the cortical collecting tubule to secrete potassium. Specifically, amiloride and triamterene inhibit sodium reabsorption mediated by the epithelial sodium channel located on the apical membrane of the principal cell. This effect abolishes the lumen’s negative potential and thereby removes a driving force for potassium secretion.

Trimethoprim and pentamidine cause similar effects.

Spironolactone and eplerenone compete with aldosterone at the level of the mineralocorticoid receptor and can result in hyperkalemia.

Drospirenone, a non-testosterone-derived progestin contained in certain oral contraceptives, possesses mineralocorticoid-blocking effects similar to those of spironolactone.

The plasma potassium level should be monitored when these drugs are prescribed in patients receiving potassium supplements, renin-angiotensin-aldosterone system blockers, or nonsteroidal anti-inflammatory drugs.²⁰

CLINICAL FEATURES OF HYPERKALEMIA

Neuromuscular manifestations of hyperkalemia include paresthesias and fasciculations in the arms and legs. Severe elevation in potassium can give rise to an ascending paralysis with eventual flaccid quadriplegia. Typically, the trunk, head, and respiratory muscles are spared, and respiratory failure is rare.

Cardiac signs

Hyperkalemia has depolarizing effects on the heart that are manifested by changes in the electrocardiogram (Figure 2). The progressive changes of hyperkalemia are classically listed as:

• Peaked T waves that are tall, narrow, and symmetrical and can occasionally be confused with the hyperacute T-wave change associated with an ST-segment elevation myocardial infarction.²¹ However, in the latter condition, the T waves tend to be more broad-based and asymmetric in shape.
• ST-segment depression
• Widening of the PR interval
• Widening of the QRS interval
• Loss of the P wave
• A sine-wave pattern—an ominous development and a harbinger of impending ventricular fibrillation and asystole.

The plasma potassium concentration often correlates poorly with cardiac manifestations. In a retrospective review, only 16 of 90 cases met strict criteria for electrocardiographic changes reflective of hyperkalemia (defined as new peaked and symmetric T waves that resolved on follow-up).²² In 13 of these cases, the electrocardiogram was interpreted as showing no T-wave changes even when read by a cardiologist. In addition, electrocardiographic criteria for hyperkalemia were noted in only 1 of 14 patients who manifested arrhythmias or cardiac arrest attributed to increased plasma potassium concentration.

The treatment of hyperkalemia depends on the magnitude of increase in the plasma potassium concentration and the presence or absence of electrocardiographic changes or neuromuscular symptoms.²³ Acute treatment is indicated for marked electrocardiographic changes and severe muscle weakness.

Intravenous calcium rapidly normalizes membrane excitability by antagonizing the potassium-induced decrease in membrane excitability but does not alter the plasma potassium concentration.

Insulin lowers the plasma potassium concentration by promoting its entry into cells. To avoid hypoglycemia, 10 units of short-acting insulin should be accompanied by a 50-g infusion of glucose, increased to 60 g if 20 units of insulin are given.²⁴

Beta-2 receptor agonists produce a similar effect. The shift of potassium into cells with insulin and beta-2-adrenergic receptor stimulation is brought about by increases in sodium-potassium ATPase pump activity, primarily in skeletal muscle cells.

Sodium bicarbonate, in the absence of acidosis, lowers the plasma potassium concentration only slightly. It should be reserved for hyperkalemic patients who have coexisting metabolic acidosis after the patient has received insulin and glucose, an adrenergic agent, and calcium.

These acute treatments need to be followed by therapies designed to lower the total body potassium content such as diuretics, potassium-binding drugs, and dialysis.

TREATMENT OF CHRONIC HYPERKALEMIA

Review medications. Once the diagnosis of hyperkalemia has been made, the initial approach should be to review the patient’s medications and make every effort to discontinue drugs that can impair renal potassium excretion.¹⁶ Patients should be asked about their use of over-the-counter nonsteroidal anti-inflammatory drugs and herbal remedies, since herbs may be a hidden source of dietary potassium.

Dietary counseling. Patients should be instructed to reduce their dietary intake of potassium and to avoid salt substitutes that contain potassium.

Diuretic therapy is beneficial in minimizing hyperkalemia in patients with chronic kidney disease. Thiazide and loop diuretics enhance renal potassium excretion by increasing flow and delivery of sodium to the collecting duct. Thiazide diuretics are effective when the estimated glomerular filtration rate is greater than 30 mL/min, while loop diuretics should be used in patients with more severe renal insufficiency (Table 2).

Sodium bicarbonate is an effective agent to minimize increases in the plasma potassium concentration in patients with chronic kidney disease and metabolic acidosis. This drug increases renal potassium excretion by increasing distal sodium delivery and shifts potassium into cells as the acidosis is corrected. The likelihood of developing volume overload as a complication of sodium bicarbonate administration can be minimized with effective diuretic therapy.

Avoiding hyperkalemia if renin-angiotensin-aldosterone system blockers are needed

Renin-angiotensin-aldosterone system blockers can be problematic, as these drugs cause hyperkalemia, often in the very patients who derive the greatest cardiovascular benefit from them.¹⁶ A number of steps can reduce the risk of hyperkalemia and allow these drugs to be used.

The initial dose should be low and the plasma potassium should be measured within 1 to 2 weeks after drug initiation. If the potassium level is normal, the dose can be titrated upwards with remeasurement of the plasma potassium after each dose titration. If the plasma potassium concentration rises to 5.5 mmol/L, in some cases lowering the dose will reduce the potassium concentration and allow the patient to remain on the drug.

In patients at risk of hyperkalemia, angiotensin II receptor blockers and direct renin inhibitors should be used with the same caution as angiotensin-converting enzyme inhibitors.

If the plasma potassium concentration exceeds 5.5 mmol/L despite the above precautions, one can consider using a potassium-binding drug (see below) before deciding to avoid renin-angiotensin-aldosterone system blockers.

Sodium polystyrene sulfonate binds potassium in the gastrointestinal tract in exchange for sodium and has been used to manage hyperkalemia. This drug is most commonly given along with sorbitol as a therapy for acute hyperkalemia. Although the drug is widely used, most of the potassium-lowering effect is due to an increase in stool volume caused by sorbitol.^25,26 In addition, long-term use is poorly tolerated, and the drug has been linked to gastrointestinal toxicity in rare cases.

Patiromer and sodium zirconium cyclosilicate are two new potassium-binding drugs that have been shown to be effective in reducing plasma potassium concentration in the setting of ongoing use of renin-angiotensin-aldosterone system blockers.

Patiromer is a nonabsorbed polymer approved for clinical use to treat hyperkalemia. The drug binds potassium in exchange for calcium in the gastrointestinal tract, predominantly in the colon, and lowers the plasma potassium concentration in a dose-dependent manner, with the greatest reduction in those with higher starting values.^27,28

Patiromer effectively controlled plasma potassium concentrations in a 1-year randomized trial in high-risk patients on renin-angiotensin-aldosterone system blockers.²⁹ The main adverse events in clinical trials have been constipation and hypomagnesemia, which required magnesium replacement in a small number of patients, but overall, the drug is well tolerated.

Sodium zirconium cyclosilicate is a nonabsorbed microporous compound that binds potassium in exchange for sodium throughout the gastrointestinal tract. It has been found effective in lowering plasma potassium concentration in a dose-dependent fashion in high-risk patients, most of whom were receiving renin-angiotensin-aldosterone system blockers.^30–32 Adverse events were generally comparable to those with placebo in clinical trials; however, edema occurred more frequently when higher doses were used. This drug is not yet approved for clinical use.

Hyperkalemia is common in patients with cardiovascular disease. Its consequences can be severe and life-threatening, and its management and prevention require a multidisciplinary approach that entails reducing intake of high-potassium foods, adjusting medications that cause hyperkalemia, and adding medications that reduce the plasma potassium concentration. With this approach, patients at high risk can receive the cardiorenal benefits of drugs that block the renin-angiotensin-aldosterone system without developing hyperkalemia.

98% OF POTASSIUM IS INSIDE CELLS

The body of a typical 70-kg man contains about 3,500 mmol of potassium, 98% of which is in the intracellular space; the remaining 2% is in the extracellular space. This large intracellular-to-extracellular gradient determines the cell voltage and explains why disorders in plasma potassium give rise to manifestations in excitable tissues such as the heart and nervous system.

The most important determinants of potassium distribution between the intracellular and extracellular space are insulin and beta-adrenergic receptor stimulation.

Maintenance of total-body potassium content is primarily the job of the kidneys, with a small contribution by the gastrointestinal tract.^1,2 Hyperkalemia is most commonly encountered in patients with decreased kidney function.

The normal kidney can secrete a large amount of potassium, making hyperkalemia uncommon in the absence of kidney disease. This large capacity may have evolved to handle the diet of Paleolithic humans, which contained 4 times as much potassium as contemporary diets.^3,4 With the onset of agriculture, dietary intake of potassium has progressively declined while sodium intake has risen. A popular theory suggests this mismatch between the modern diet and the nutritional requirements encoded in the human genome during evolution may contribute to chronic diseases such as hypertension, stroke, kidney stones, and bone disease.⁵

MANY POTENTIAL CAUSES OF HYPERKALEMIA

Causes of hyperkalemia are outlined in Table 1. Shifting of potassium from the cells to the extracellular space is a cause of transient hyperkalemia, while chronic hyperkalemia indicates an impairment in renal potassium secretion. The following discussion is a guide to the approach to the hyperkalemic patient.

Is the patient’s hyperkalemia really pseudohyperkalemia?

Pseudohyperkalemia, an artifact of measurement, occurs due to mechanical release of potassium from cells during phlebotomy or specimen processing.⁶ This diagnosis is made when the serum potassium concentration exceeds the plasma potassium concentration by more than 0.5 mmol/L, and should be considered when hyperkalemia occurs in the absence of a clinical risk factor. Fist-clenching, application of a tight-fitting tourniquet, or use of small-bore needles during phlebotomy can all cause pseudohyperkalemia.

Mechanism of pseudohyperkalemia. Since serum is the liquid part of blood remaining after coagulation, release of potassium from cells injured during the process of coagulation raises the potassium level in the serum. Plasma is the cell-free part of blood that has been treated with anticoagulants; it has no cells that can be injured and release potassium. Thus, the serum potassium level will be higher than that in the plasma.

Reverse pseudohyperkalemia, in contrast, occurs when the plasma potassium level is falsely elevated but the serum value is normal. This situation has been described in hematologic disorders characterized by pronounced leukocytosis in which malignant cells are prone to lysis with minimal mechanical stress due to increased fragility or altered sodium-potassium ATPase pump activity.⁷ This phenomenon is unusual but occurs because the cells are so fragile.

A spurious increase in plasma potassium concentration along with a low plasma calcium concentration raises the possibility of calcium chelation and release of potassium in a sample tube contaminated with the anticoagulant ethylenediaminetetraacetic acid.

Is there increased potassium intake?

Increased potassium intake is a potential cause of hyperkalemia in patients with decreased kidney function or adrenal disease.

Foods naturally rich in potassium include bananas (a medium-sized banana contains 451 mg or 12 mmol of potassium) and potatoes (844 mg or 22 mmol in a large baked potato with skin). Other potassium-rich foods are melons, citrus juice, and avocados. Less-obvious food sources include raw coconut juice (potassium concentration 44.3 mmol/L) and noni juice (56 mmol/L).

Salt substitutes, recommended to hypertensive patients with chronic kidney disease, can be a hidden source of dietary potassium.

Clay ingestion is a potential cause of dyskalemia. White clay consumption causes hypokalemia due to potassium binding in the gastrointestinal tract. Red clay or river bed clay, on the other hand, is enriched in potassium (100 mmol of potassium in 100 g of clay) and can cause life-threatening hyperkalemia in patients with chronic kidney disease.⁸

Eating burnt match heads. Some individuals chew and ingest burnt match heads, a condition called cautopyreiophagia. In one reported case,⁹ this activity contributed an additional 80 mmol of daily potassium intake in a dialysis patient, resulting in a plasma potassium concentration of 8 mmol/L.

Acute hyperkalemia can be the result of redistribution of cellular potassium. Shifting of as little as 2% of the body’s potassium from the intracellular to the extracellular space can double the plasma potassium concentration.

Tissue injury. Hyperkalemia frequently occurs in diseases that cause tissue injury such as rhabdomyolysis, trauma, massive hemolysis, and tumor lysis.

Insulin deficiency. Insulin and catecholamines are major regulators of potassium distribution within the body. After a meal, release of insulin not only regulates the plasma glucose concentration, it also causes potassium to move into cells until the kidneys have had sufficient time to excrete the dietary potassium load and reestablish total-body potassium content.

Exercise, beta-blockers. During exercise, potassium is released from skeletal muscle cells and accumulates in the interstitial compartment, where it exerts a vasodilatory effect. The simultaneous increase in circulating catecholamines regulates this release by promoting cell potassium uptake through beta-adrenergic receptor stimulation.

Metabolic acidosis can facilitate exit (ie, shift) of potassium from cells, but this effect depends on the type of acidosis. Hyperchloremic normal anion gap acidosis (mineral acidosis) most commonly causes this effect due to the relative impermeability of the cell membrane to the chloride anion. As hydrogen ions move into the cell due to accumulation of ammonium chloride or hydrogen chloride, electrical neutrality is maintained by potassium exit.

In contrast, organic acidosis (due to lactic, beta-hydroxybutyric, or methylmalonic acid) tends not to cause a potassium shift, since most organic anions readily cross the cell membrane along with hydrogen. Lactic acidosis is often associated with potassium shift, but this effect is due to loss of cell integrity as a result of cell ischemia. The hyperkalemia typically present on admission in patients with diabetic ketoacidosis is the result of insulin deficiency and hypertonicity and not the underlying organic acidosis.¹⁰

Hypertonic states can cause hyperkalemia due to cell shift. For example, hyperglycemia, as in diabetic ketoacidosis, pulls water from the intracellular into the extracellular compartment, thereby concentrating intracellular potassium and creating a more favorable gradient for potassium efflux through membrane channels. This same effect can occur in neurosurgical patients given large amounts of hypertonic mannitol. Repetitive doses of immunoglobulin can lead to extracellular accumulation of sorbitol, maltose, or sucrose, since these sugars are added to the preparations to prevent immunoglobulin aggregation.¹¹

Is a disturbance in renal potassium excretion present?

Sustained hyperkalemia is more commonly associated with decreases in renal potassium excretion than with a cellular shift. In most instances the clinician can distinguish between cell shift and impaired renal excretion based on the available clinical data.

The transtubular potassium gradient has been used to determine whether there is a disturbance in renal potassium excretion and to assess renal potassium handling.¹²

This calculation is based on the assumption that only water is reabsorbed past the cortical collecting duct, and not solutes. It has fallen out of favor since we have found this assumption to be incorrect; a large amount of urea is reabsorbed daily in the downstream medullary collecting duct as a result of intrarenal recycling of urea.

The one situation in which the transtubular potassium gradient may be of use is determining whether hyperkalemia is a result of low aldosterone levels as opposed to aldosterone resistance. One can compare the transtubular potassium gradient before and after a physiologic dose (0.05 mg) of 9-alpha fludrocortisone. An increase of more than 6 over a 4-hour period favors aldosterone deficiency, whereas smaller changes would indicate aldosterone resistance.

24-hour potassium excretion, spot urine potassium-creatinine ratio.  A better way to assess renal potassium handling is to measure the amount of potassium in a 24-hour urine collection or determine a spot urine potassium-creatinine ratio. A 24-hour urinary potassium excretion of less than 15 mmol or a potassium-creatinine ratio less than 1 suggests an extrarenal cause of hypokalemia. A ratio greater than 20 would be an appropriate renal response to hyperkalemia.

One or more of 3 abnormalities should be considered in the hyperkalemic patient with impaired renal excretion of potassium:

• Decreased distal delivery of sodium
• Mineralocorticoid deficiency
• Abnormal cortical collecting tubule function.¹³

Under normal circumstances, potassium is freely filtered across the glomerulus and then mostly reabsorbed in the proximal tubule and thick ascending limb. Potassium secretion begins in the distal convoluted tubule and increases in magnitude into the collecting duct. Tubular secretion is the component of potassium handling that varies and is regulated according to physiologic needs.

In acute kidney injury, the rapid decline in glomerular filtration rate and reduction in functioning nephron mass lead to decreased distal potassium secretion.

Hyperkalemia is a frequent problem when oliguria is present, since the reduction in distal delivery of sodium and water further impairs potassium secretion. Patients with oliguric acute kidney injury are more likely to have a more severe underlying disease state, and therefore tissue breakdown and catabolism further increase the risk of hyperkalemia.

In contrast, in nonoliguric patients, the renal injury tends to be less severe, and enough sodium and water are usually delivered distally to prevent hyperkalemia.

In chronic kidney disease, nephron dropout and reduction in collecting tubule mass also lead to a global decline in distal potassium secretion. However, this is countered by an increased capacity of the remaining individual nephrons for potassium secretion. High flow, increased distal sodium delivery, and increased activity and number of sodium-potassium ATPase pumps in the remaining nephrons account for this increased secretory capacity.¹⁴ As renal function declines over time, colonic potassium secretion progressively increases.¹⁵

These adaptive changes help to keep the plasma potassium concentration within the normal range until the glomerular filtration rate falls to less than 10 or 15 mL/min. Development of hyperkalemia with more modest reductions in the glomerular filtration rate suggest decreased mineralocorticoid activity or a specific lesion of the tubule.

Mineralocorticoid deficiency

Aldosterone deficiency can occur alone or in combination with decreased cortisol levels. Destruction of the adrenal glands is suggested when both hormones are reduced. Enzyme defects in cortisol metabolism can result in either isolated deficiency of aldosterone or adrenogenital syndromes associated with decreased mineralocorticoid activity.

Heparin administration leads to a reversible defect in adrenal synthesis of aldosterone. Drugs that block the stimulatory effect of angiotensin II on the zona glomerulosa cells of the adrenal gland will lower aldosterone.

Renin-angiotensin-aldosterone system blockers. Angiotensin-converting enzyme inhibitors block the formation of angiotensin II, whereas angiotensin II receptor blockers prevent angiotensin II from binding to its adrenal receptor. The direct renin inhibitor aliskiren lowers angiotensin II levels by blocking the enzymatic activity of renin and lowers the circulating levels of both angiotensin I and II.¹⁶

The syndrome of hyporeninemic hypoaldosteronism is a common cause of hyperkalemia in patients who have a glomerular filtration rate between 40 and 60 mL/min. Diabetic nephropathy and interstitial renal disease are the most common clinical entities associated with this syndrome.¹⁰ Other causes include analgesic nephropathy, urinary tract obstruction, sickle cell disease, systemic lupus erythematosus, and amyloidosis.

Nonsteroidal anti-inflammatory drugs can cause hyperkalemia by suppressing renin release and reducing delivery of sodium to the distal nephron.¹⁸

Calcineurin inhibitors impair potassium secretion by suppressing renin release and by direct tubular effects.¹⁹

Beta-blockers. Beta-1 and to a lesser extent beta-2 receptor blockade can also result in a hyporeninemic state.

Distal tubular defect

Hyperkalemia can result from interstitial renal diseases that specifically affect the distal nephron. In this setting, the glomerular filtration rate is only mildly reduced, and circulating aldosterone levels are normal.

Renal transplant, lupus erythematosus, amyloidosis, urinary obstruction, and sickle cell disease are conditions in which an impairment in renin release may coexist with a defect in tubular secretion.

Potassium-sparing diuretics impair the ability of the cortical collecting tubule to secrete potassium. Specifically, amiloride and triamterene inhibit sodium reabsorption mediated by the epithelial sodium channel located on the apical membrane of the principal cell. This effect abolishes the lumen’s negative potential and thereby removes a driving force for potassium secretion.

Trimethoprim and pentamidine cause similar effects.

Spironolactone and eplerenone compete with aldosterone at the level of the mineralocorticoid receptor and can result in hyperkalemia.

Drospirenone, a non-testosterone-derived progestin contained in certain oral contraceptives, possesses mineralocorticoid-blocking effects similar to those of spironolactone.

The plasma potassium level should be monitored when these drugs are prescribed in patients receiving potassium supplements, renin-angiotensin-aldosterone system blockers, or nonsteroidal anti-inflammatory drugs.²⁰

CLINICAL FEATURES OF HYPERKALEMIA

Neuromuscular manifestations of hyperkalemia include paresthesias and fasciculations in the arms and legs. Severe elevation in potassium can give rise to an ascending paralysis with eventual flaccid quadriplegia. Typically, the trunk, head, and respiratory muscles are spared, and respiratory failure is rare.

Cardiac signs

Hyperkalemia has depolarizing effects on the heart that are manifested by changes in the electrocardiogram (Figure 2). The progressive changes of hyperkalemia are classically listed as:

• Peaked T waves that are tall, narrow, and symmetrical and can occasionally be confused with the hyperacute T-wave change associated with an ST-segment elevation myocardial infarction.²¹ However, in the latter condition, the T waves tend to be more broad-based and asymmetric in shape.
• ST-segment depression
• Widening of the PR interval
• Widening of the QRS interval
• Loss of the P wave
• A sine-wave pattern—an ominous development and a harbinger of impending ventricular fibrillation and asystole.

The plasma potassium concentration often correlates poorly with cardiac manifestations. In a retrospective review, only 16 of 90 cases met strict criteria for electrocardiographic changes reflective of hyperkalemia (defined as new peaked and symmetric T waves that resolved on follow-up).²² In 13 of these cases, the electrocardiogram was interpreted as showing no T-wave changes even when read by a cardiologist. In addition, electrocardiographic criteria for hyperkalemia were noted in only 1 of 14 patients who manifested arrhythmias or cardiac arrest attributed to increased plasma potassium concentration.

The treatment of hyperkalemia depends on the magnitude of increase in the plasma potassium concentration and the presence or absence of electrocardiographic changes or neuromuscular symptoms.²³ Acute treatment is indicated for marked electrocardiographic changes and severe muscle weakness.

Intravenous calcium rapidly normalizes membrane excitability by antagonizing the potassium-induced decrease in membrane excitability but does not alter the plasma potassium concentration.

Insulin lowers the plasma potassium concentration by promoting its entry into cells. To avoid hypoglycemia, 10 units of short-acting insulin should be accompanied by a 50-g infusion of glucose, increased to 60 g if 20 units of insulin are given.²⁴

Beta-2 receptor agonists produce a similar effect. The shift of potassium into cells with insulin and beta-2-adrenergic receptor stimulation is brought about by increases in sodium-potassium ATPase pump activity, primarily in skeletal muscle cells.

Sodium bicarbonate, in the absence of acidosis, lowers the plasma potassium concentration only slightly. It should be reserved for hyperkalemic patients who have coexisting metabolic acidosis after the patient has received insulin and glucose, an adrenergic agent, and calcium.

These acute treatments need to be followed by therapies designed to lower the total body potassium content such as diuretics, potassium-binding drugs, and dialysis.

TREATMENT OF CHRONIC HYPERKALEMIA

Review medications. Once the diagnosis of hyperkalemia has been made, the initial approach should be to review the patient’s medications and make every effort to discontinue drugs that can impair renal potassium excretion.¹⁶ Patients should be asked about their use of over-the-counter nonsteroidal anti-inflammatory drugs and herbal remedies, since herbs may be a hidden source of dietary potassium.

Dietary counseling. Patients should be instructed to reduce their dietary intake of potassium and to avoid salt substitutes that contain potassium.

Diuretic therapy is beneficial in minimizing hyperkalemia in patients with chronic kidney disease. Thiazide and loop diuretics enhance renal potassium excretion by increasing flow and delivery of sodium to the collecting duct. Thiazide diuretics are effective when the estimated glomerular filtration rate is greater than 30 mL/min, while loop diuretics should be used in patients with more severe renal insufficiency (Table 2).

Sodium bicarbonate is an effective agent to minimize increases in the plasma potassium concentration in patients with chronic kidney disease and metabolic acidosis. This drug increases renal potassium excretion by increasing distal sodium delivery and shifts potassium into cells as the acidosis is corrected. The likelihood of developing volume overload as a complication of sodium bicarbonate administration can be minimized with effective diuretic therapy.

Avoiding hyperkalemia if renin-angiotensin-aldosterone system blockers are needed

Renin-angiotensin-aldosterone system blockers can be problematic, as these drugs cause hyperkalemia, often in the very patients who derive the greatest cardiovascular benefit from them.¹⁶ A number of steps can reduce the risk of hyperkalemia and allow these drugs to be used.

The initial dose should be low and the plasma potassium should be measured within 1 to 2 weeks after drug initiation. If the potassium level is normal, the dose can be titrated upwards with remeasurement of the plasma potassium after each dose titration. If the plasma potassium concentration rises to 5.5 mmol/L, in some cases lowering the dose will reduce the potassium concentration and allow the patient to remain on the drug.

In patients at risk of hyperkalemia, angiotensin II receptor blockers and direct renin inhibitors should be used with the same caution as angiotensin-converting enzyme inhibitors.

If the plasma potassium concentration exceeds 5.5 mmol/L despite the above precautions, one can consider using a potassium-binding drug (see below) before deciding to avoid renin-angiotensin-aldosterone system blockers.

Sodium polystyrene sulfonate binds potassium in the gastrointestinal tract in exchange for sodium and has been used to manage hyperkalemia. This drug is most commonly given along with sorbitol as a therapy for acute hyperkalemia. Although the drug is widely used, most of the potassium-lowering effect is due to an increase in stool volume caused by sorbitol.^25,26 In addition, long-term use is poorly tolerated, and the drug has been linked to gastrointestinal toxicity in rare cases.

Patiromer and sodium zirconium cyclosilicate are two new potassium-binding drugs that have been shown to be effective in reducing plasma potassium concentration in the setting of ongoing use of renin-angiotensin-aldosterone system blockers.

Patiromer is a nonabsorbed polymer approved for clinical use to treat hyperkalemia. The drug binds potassium in exchange for calcium in the gastrointestinal tract, predominantly in the colon, and lowers the plasma potassium concentration in a dose-dependent manner, with the greatest reduction in those with higher starting values.^27,28

Patiromer effectively controlled plasma potassium concentrations in a 1-year randomized trial in high-risk patients on renin-angiotensin-aldosterone system blockers.²⁹ The main adverse events in clinical trials have been constipation and hypomagnesemia, which required magnesium replacement in a small number of patients, but overall, the drug is well tolerated.

Sodium zirconium cyclosilicate is a nonabsorbed microporous compound that binds potassium in exchange for sodium throughout the gastrointestinal tract. It has been found effective in lowering plasma potassium concentration in a dose-dependent fashion in high-risk patients, most of whom were receiving renin-angiotensin-aldosterone system blockers.^30–32 Adverse events were generally comparable to those with placebo in clinical trials; however, edema occurred more frequently when higher doses were used. This drug is not yet approved for clinical use.

Hyperkalemia is common in patients with cardiovascular disease. Its consequences can be severe and life-threatening, and its management and prevention require a multidisciplinary approach that entails reducing intake of high-potassium foods, adjusting medications that cause hyperkalemia, and adding medications that reduce the plasma potassium concentration. With this approach, patients at high risk can receive the cardiorenal benefits of drugs that block the renin-angiotensin-aldosterone system without developing hyperkalemia.

98% OF POTASSIUM IS INSIDE CELLS

The body of a typical 70-kg man contains about 3,500 mmol of potassium, 98% of which is in the intracellular space; the remaining 2% is in the extracellular space. This large intracellular-to-extracellular gradient determines the cell voltage and explains why disorders in plasma potassium give rise to manifestations in excitable tissues such as the heart and nervous system.

The most important determinants of potassium distribution between the intracellular and extracellular space are insulin and beta-adrenergic receptor stimulation.

Maintenance of total-body potassium content is primarily the job of the kidneys, with a small contribution by the gastrointestinal tract.^1,2 Hyperkalemia is most commonly encountered in patients with decreased kidney function.

The normal kidney can secrete a large amount of potassium, making hyperkalemia uncommon in the absence of kidney disease. This large capacity may have evolved to handle the diet of Paleolithic humans, which contained 4 times as much potassium as contemporary diets.^3,4 With the onset of agriculture, dietary intake of potassium has progressively declined while sodium intake has risen. A popular theory suggests this mismatch between the modern diet and the nutritional requirements encoded in the human genome during evolution may contribute to chronic diseases such as hypertension, stroke, kidney stones, and bone disease.⁵

MANY POTENTIAL CAUSES OF HYPERKALEMIA

Causes of hyperkalemia are outlined in Table 1. Shifting of potassium from the cells to the extracellular space is a cause of transient hyperkalemia, while chronic hyperkalemia indicates an impairment in renal potassium secretion. The following discussion is a guide to the approach to the hyperkalemic patient.

Is the patient’s hyperkalemia really pseudohyperkalemia?

Pseudohyperkalemia, an artifact of measurement, occurs due to mechanical release of potassium from cells during phlebotomy or specimen processing.⁶ This diagnosis is made when the serum potassium concentration exceeds the plasma potassium concentration by more than 0.5 mmol/L, and should be considered when hyperkalemia occurs in the absence of a clinical risk factor. Fist-clenching, application of a tight-fitting tourniquet, or use of small-bore needles during phlebotomy can all cause pseudohyperkalemia.

Mechanism of pseudohyperkalemia. Since serum is the liquid part of blood remaining after coagulation, release of potassium from cells injured during the process of coagulation raises the potassium level in the serum. Plasma is the cell-free part of blood that has been treated with anticoagulants; it has no cells that can be injured and release potassium. Thus, the serum potassium level will be higher than that in the plasma.

Reverse pseudohyperkalemia, in contrast, occurs when the plasma potassium level is falsely elevated but the serum value is normal. This situation has been described in hematologic disorders characterized by pronounced leukocytosis in which malignant cells are prone to lysis with minimal mechanical stress due to increased fragility or altered sodium-potassium ATPase pump activity.⁷ This phenomenon is unusual but occurs because the cells are so fragile.

A spurious increase in plasma potassium concentration along with a low plasma calcium concentration raises the possibility of calcium chelation and release of potassium in a sample tube contaminated with the anticoagulant ethylenediaminetetraacetic acid.

Is there increased potassium intake?

Increased potassium intake is a potential cause of hyperkalemia in patients with decreased kidney function or adrenal disease.

Foods naturally rich in potassium include bananas (a medium-sized banana contains 451 mg or 12 mmol of potassium) and potatoes (844 mg or 22 mmol in a large baked potato with skin). Other potassium-rich foods are melons, citrus juice, and avocados. Less-obvious food sources include raw coconut juice (potassium concentration 44.3 mmol/L) and noni juice (56 mmol/L).

Salt substitutes, recommended to hypertensive patients with chronic kidney disease, can be a hidden source of dietary potassium.

Clay ingestion is a potential cause of dyskalemia. White clay consumption causes hypokalemia due to potassium binding in the gastrointestinal tract. Red clay or river bed clay, on the other hand, is enriched in potassium (100 mmol of potassium in 100 g of clay) and can cause life-threatening hyperkalemia in patients with chronic kidney disease.⁸

Eating burnt match heads. Some individuals chew and ingest burnt match heads, a condition called cautopyreiophagia. In one reported case,⁹ this activity contributed an additional 80 mmol of daily potassium intake in a dialysis patient, resulting in a plasma potassium concentration of 8 mmol/L.

Acute hyperkalemia can be the result of redistribution of cellular potassium. Shifting of as little as 2% of the body’s potassium from the intracellular to the extracellular space can double the plasma potassium concentration.

Tissue injury. Hyperkalemia frequently occurs in diseases that cause tissue injury such as rhabdomyolysis, trauma, massive hemolysis, and tumor lysis.

Insulin deficiency. Insulin and catecholamines are major regulators of potassium distribution within the body. After a meal, release of insulin not only regulates the plasma glucose concentration, it also causes potassium to move into cells until the kidneys have had sufficient time to excrete the dietary potassium load and reestablish total-body potassium content.

Exercise, beta-blockers. During exercise, potassium is released from skeletal muscle cells and accumulates in the interstitial compartment, where it exerts a vasodilatory effect. The simultaneous increase in circulating catecholamines regulates this release by promoting cell potassium uptake through beta-adrenergic receptor stimulation.

Metabolic acidosis can facilitate exit (ie, shift) of potassium from cells, but this effect depends on the type of acidosis. Hyperchloremic normal anion gap acidosis (mineral acidosis) most commonly causes this effect due to the relative impermeability of the cell membrane to the chloride anion. As hydrogen ions move into the cell due to accumulation of ammonium chloride or hydrogen chloride, electrical neutrality is maintained by potassium exit.

In contrast, organic acidosis (due to lactic, beta-hydroxybutyric, or methylmalonic acid) tends not to cause a potassium shift, since most organic anions readily cross the cell membrane along with hydrogen. Lactic acidosis is often associated with potassium shift, but this effect is due to loss of cell integrity as a result of cell ischemia. The hyperkalemia typically present on admission in patients with diabetic ketoacidosis is the result of insulin deficiency and hypertonicity and not the underlying organic acidosis.¹⁰

Hypertonic states can cause hyperkalemia due to cell shift. For example, hyperglycemia, as in diabetic ketoacidosis, pulls water from the intracellular into the extracellular compartment, thereby concentrating intracellular potassium and creating a more favorable gradient for potassium efflux through membrane channels. This same effect can occur in neurosurgical patients given large amounts of hypertonic mannitol. Repetitive doses of immunoglobulin can lead to extracellular accumulation of sorbitol, maltose, or sucrose, since these sugars are added to the preparations to prevent immunoglobulin aggregation.¹¹

Is a disturbance in renal potassium excretion present?

Sustained hyperkalemia is more commonly associated with decreases in renal potassium excretion than with a cellular shift. In most instances the clinician can distinguish between cell shift and impaired renal excretion based on the available clinical data.

The transtubular potassium gradient has been used to determine whether there is a disturbance in renal potassium excretion and to assess renal potassium handling.¹²

This calculation is based on the assumption that only water is reabsorbed past the cortical collecting duct, and not solutes. It has fallen out of favor since we have found this assumption to be incorrect; a large amount of urea is reabsorbed daily in the downstream medullary collecting duct as a result of intrarenal recycling of urea.

The one situation in which the transtubular potassium gradient may be of use is determining whether hyperkalemia is a result of low aldosterone levels as opposed to aldosterone resistance. One can compare the transtubular potassium gradient before and after a physiologic dose (0.05 mg) of 9-alpha fludrocortisone. An increase of more than 6 over a 4-hour period favors aldosterone deficiency, whereas smaller changes would indicate aldosterone resistance.

24-hour potassium excretion, spot urine potassium-creatinine ratio.  A better way to assess renal potassium handling is to measure the amount of potassium in a 24-hour urine collection or determine a spot urine potassium-creatinine ratio. A 24-hour urinary potassium excretion of less than 15 mmol or a potassium-creatinine ratio less than 1 suggests an extrarenal cause of hypokalemia. A ratio greater than 20 would be an appropriate renal response to hyperkalemia.

One or more of 3 abnormalities should be considered in the hyperkalemic patient with impaired renal excretion of potassium:

• Decreased distal delivery of sodium
• Mineralocorticoid deficiency
• Abnormal cortical collecting tubule function.¹³

Under normal circumstances, potassium is freely filtered across the glomerulus and then mostly reabsorbed in the proximal tubule and thick ascending limb. Potassium secretion begins in the distal convoluted tubule and increases in magnitude into the collecting duct. Tubular secretion is the component of potassium handling that varies and is regulated according to physiologic needs.

In acute kidney injury, the rapid decline in glomerular filtration rate and reduction in functioning nephron mass lead to decreased distal potassium secretion.

Hyperkalemia is a frequent problem when oliguria is present, since the reduction in distal delivery of sodium and water further impairs potassium secretion. Patients with oliguric acute kidney injury are more likely to have a more severe underlying disease state, and therefore tissue breakdown and catabolism further increase the risk of hyperkalemia.

In contrast, in nonoliguric patients, the renal injury tends to be less severe, and enough sodium and water are usually delivered distally to prevent hyperkalemia.

In chronic kidney disease, nephron dropout and reduction in collecting tubule mass also lead to a global decline in distal potassium secretion. However, this is countered by an increased capacity of the remaining individual nephrons for potassium secretion. High flow, increased distal sodium delivery, and increased activity and number of sodium-potassium ATPase pumps in the remaining nephrons account for this increased secretory capacity.¹⁴ As renal function declines over time, colonic potassium secretion progressively increases.¹⁵

These adaptive changes help to keep the plasma potassium concentration within the normal range until the glomerular filtration rate falls to less than 10 or 15 mL/min. Development of hyperkalemia with more modest reductions in the glomerular filtration rate suggest decreased mineralocorticoid activity or a specific lesion of the tubule.

Mineralocorticoid deficiency

Aldosterone deficiency can occur alone or in combination with decreased cortisol levels. Destruction of the adrenal glands is suggested when both hormones are reduced. Enzyme defects in cortisol metabolism can result in either isolated deficiency of aldosterone or adrenogenital syndromes associated with decreased mineralocorticoid activity.

Heparin administration leads to a reversible defect in adrenal synthesis of aldosterone. Drugs that block the stimulatory effect of angiotensin II on the zona glomerulosa cells of the adrenal gland will lower aldosterone.

Renin-angiotensin-aldosterone system blockers. Angiotensin-converting enzyme inhibitors block the formation of angiotensin II, whereas angiotensin II receptor blockers prevent angiotensin II from binding to its adrenal receptor. The direct renin inhibitor aliskiren lowers angiotensin II levels by blocking the enzymatic activity of renin and lowers the circulating levels of both angiotensin I and II.¹⁶

The syndrome of hyporeninemic hypoaldosteronism is a common cause of hyperkalemia in patients who have a glomerular filtration rate between 40 and 60 mL/min. Diabetic nephropathy and interstitial renal disease are the most common clinical entities associated with this syndrome.¹⁰ Other causes include analgesic nephropathy, urinary tract obstruction, sickle cell disease, systemic lupus erythematosus, and amyloidosis.

Nonsteroidal anti-inflammatory drugs can cause hyperkalemia by suppressing renin release and reducing delivery of sodium to the distal nephron.¹⁸

Calcineurin inhibitors impair potassium secretion by suppressing renin release and by direct tubular effects.¹⁹

Beta-blockers. Beta-1 and to a lesser extent beta-2 receptor blockade can also result in a hyporeninemic state.

Distal tubular defect

Hyperkalemia can result from interstitial renal diseases that specifically affect the distal nephron. In this setting, the glomerular filtration rate is only mildly reduced, and circulating aldosterone levels are normal.

Renal transplant, lupus erythematosus, amyloidosis, urinary obstruction, and sickle cell disease are conditions in which an impairment in renin release may coexist with a defect in tubular secretion.

Potassium-sparing diuretics impair the ability of the cortical collecting tubule to secrete potassium. Specifically, amiloride and triamterene inhibit sodium reabsorption mediated by the epithelial sodium channel located on the apical membrane of the principal cell. This effect abolishes the lumen’s negative potential and thereby removes a driving force for potassium secretion.

Trimethoprim and pentamidine cause similar effects.

Spironolactone and eplerenone compete with aldosterone at the level of the mineralocorticoid receptor and can result in hyperkalemia.

Drospirenone, a non-testosterone-derived progestin contained in certain oral contraceptives, possesses mineralocorticoid-blocking effects similar to those of spironolactone.

The plasma potassium level should be monitored when these drugs are prescribed in patients receiving potassium supplements, renin-angiotensin-aldosterone system blockers, or nonsteroidal anti-inflammatory drugs.²⁰

CLINICAL FEATURES OF HYPERKALEMIA

Neuromuscular manifestations of hyperkalemia include paresthesias and fasciculations in the arms and legs. Severe elevation in potassium can give rise to an ascending paralysis with eventual flaccid quadriplegia. Typically, the trunk, head, and respiratory muscles are spared, and respiratory failure is rare.

Cardiac signs

Hyperkalemia has depolarizing effects on the heart that are manifested by changes in the electrocardiogram (Figure 2). The progressive changes of hyperkalemia are classically listed as:

• Peaked T waves that are tall, narrow, and symmetrical and can occasionally be confused with the hyperacute T-wave change associated with an ST-segment elevation myocardial infarction.²¹ However, in the latter condition, the T waves tend to be more broad-based and asymmetric in shape.
• ST-segment depression
• Widening of the PR interval
• Widening of the QRS interval
• Loss of the P wave
• A sine-wave pattern—an ominous development and a harbinger of impending ventricular fibrillation and asystole.

The plasma potassium concentration often correlates poorly with cardiac manifestations. In a retrospective review, only 16 of 90 cases met strict criteria for electrocardiographic changes reflective of hyperkalemia (defined as new peaked and symmetric T waves that resolved on follow-up).²² In 13 of these cases, the electrocardiogram was interpreted as showing no T-wave changes even when read by a cardiologist. In addition, electrocardiographic criteria for hyperkalemia were noted in only 1 of 14 patients who manifested arrhythmias or cardiac arrest attributed to increased plasma potassium concentration.

The treatment of hyperkalemia depends on the magnitude of increase in the plasma potassium concentration and the presence or absence of electrocardiographic changes or neuromuscular symptoms.²³ Acute treatment is indicated for marked electrocardiographic changes and severe muscle weakness.

Intravenous calcium rapidly normalizes membrane excitability by antagonizing the potassium-induced decrease in membrane excitability but does not alter the plasma potassium concentration.

Insulin lowers the plasma potassium concentration by promoting its entry into cells. To avoid hypoglycemia, 10 units of short-acting insulin should be accompanied by a 50-g infusion of glucose, increased to 60 g if 20 units of insulin are given.²⁴

Beta-2 receptor agonists produce a similar effect. The shift of potassium into cells with insulin and beta-2-adrenergic receptor stimulation is brought about by increases in sodium-potassium ATPase pump activity, primarily in skeletal muscle cells.

Sodium bicarbonate, in the absence of acidosis, lowers the plasma potassium concentration only slightly. It should be reserved for hyperkalemic patients who have coexisting metabolic acidosis after the patient has received insulin and glucose, an adrenergic agent, and calcium.

These acute treatments need to be followed by therapies designed to lower the total body potassium content such as diuretics, potassium-binding drugs, and dialysis.

TREATMENT OF CHRONIC HYPERKALEMIA

Review medications. Once the diagnosis of hyperkalemia has been made, the initial approach should be to review the patient’s medications and make every effort to discontinue drugs that can impair renal potassium excretion.¹⁶ Patients should be asked about their use of over-the-counter nonsteroidal anti-inflammatory drugs and herbal remedies, since herbs may be a hidden source of dietary potassium.

Dietary counseling. Patients should be instructed to reduce their dietary intake of potassium and to avoid salt substitutes that contain potassium.

Diuretic therapy is beneficial in minimizing hyperkalemia in patients with chronic kidney disease. Thiazide and loop diuretics enhance renal potassium excretion by increasing flow and delivery of sodium to the collecting duct. Thiazide diuretics are effective when the estimated glomerular filtration rate is greater than 30 mL/min, while loop diuretics should be used in patients with more severe renal insufficiency (Table 2).

Sodium bicarbonate is an effective agent to minimize increases in the plasma potassium concentration in patients with chronic kidney disease and metabolic acidosis. This drug increases renal potassium excretion by increasing distal sodium delivery and shifts potassium into cells as the acidosis is corrected. The likelihood of developing volume overload as a complication of sodium bicarbonate administration can be minimized with effective diuretic therapy.

Avoiding hyperkalemia if renin-angiotensin-aldosterone system blockers are needed

Renin-angiotensin-aldosterone system blockers can be problematic, as these drugs cause hyperkalemia, often in the very patients who derive the greatest cardiovascular benefit from them.¹⁶ A number of steps can reduce the risk of hyperkalemia and allow these drugs to be used.

The initial dose should be low and the plasma potassium should be measured within 1 to 2 weeks after drug initiation. If the potassium level is normal, the dose can be titrated upwards with remeasurement of the plasma potassium after each dose titration. If the plasma potassium concentration rises to 5.5 mmol/L, in some cases lowering the dose will reduce the potassium concentration and allow the patient to remain on the drug.

In patients at risk of hyperkalemia, angiotensin II receptor blockers and direct renin inhibitors should be used with the same caution as angiotensin-converting enzyme inhibitors.

If the plasma potassium concentration exceeds 5.5 mmol/L despite the above precautions, one can consider using a potassium-binding drug (see below) before deciding to avoid renin-angiotensin-aldosterone system blockers.

Sodium polystyrene sulfonate binds potassium in the gastrointestinal tract in exchange for sodium and has been used to manage hyperkalemia. This drug is most commonly given along with sorbitol as a therapy for acute hyperkalemia. Although the drug is widely used, most of the potassium-lowering effect is due to an increase in stool volume caused by sorbitol.^25,26 In addition, long-term use is poorly tolerated, and the drug has been linked to gastrointestinal toxicity in rare cases.

Patiromer and sodium zirconium cyclosilicate are two new potassium-binding drugs that have been shown to be effective in reducing plasma potassium concentration in the setting of ongoing use of renin-angiotensin-aldosterone system blockers.

Patiromer is a nonabsorbed polymer approved for clinical use to treat hyperkalemia. The drug binds potassium in exchange for calcium in the gastrointestinal tract, predominantly in the colon, and lowers the plasma potassium concentration in a dose-dependent manner, with the greatest reduction in those with higher starting values.^27,28

Patiromer effectively controlled plasma potassium concentrations in a 1-year randomized trial in high-risk patients on renin-angiotensin-aldosterone system blockers.²⁹ The main adverse events in clinical trials have been constipation and hypomagnesemia, which required magnesium replacement in a small number of patients, but overall, the drug is well tolerated.

Sodium zirconium cyclosilicate is a nonabsorbed microporous compound that binds potassium in exchange for sodium throughout the gastrointestinal tract. It has been found effective in lowering plasma potassium concentration in a dose-dependent fashion in high-risk patients, most of whom were receiving renin-angiotensin-aldosterone system blockers.^30–32 Adverse events were generally comparable to those with placebo in clinical trials; however, edema occurred more frequently when higher doses were used. This drug is not yet approved for clinical use.