Diabetes

Over-the-counter (OTC) insulin is chiefly purchased at Walmart pharmacies, likely because their brand is considerably less expensive than the OTC insulin sold elsewhere, according to a national survey of pharmacy employees.

The survey was undertaken by Jennifer N. Goldstein, MD, of Christiana Care Health System in Newark, Del., and her colleagues and published online Feb. 18 in JAMA Internal Medicine.

iStock/ThinkStock

As Walmart does not make its sales data public, Dr. Goldstein and her colleagues conducted a national telephone-based survey of Walmart and chain pharmacies in the 49 states where OTC insulin is available. They administered a five-item questionnaire that sought to determine the frequency of OTC insulin sales.

Of the 561 pharmacies that completed the questionnaire, 500 (89.1%) responded that they did sell OTC insulin; this included 284 of 292 Walmart pharmacies. Among those Walmart pharmacies, 247 (87%) said sales of OTC insulin occurred daily; 31 (10.9%) sold it weekly; and 3 (1.1%) sold it monthly.

The chains (CVS, Walgreens, Rite Aid) reported far less frequent sales of OTC insulin; 100 out of 216 (46.3%) pharmacies said they made sales only “a few times a year,” and none reported daily sales.

A majority of respondents (54.9%) also answered yes when asked if they believed patients purchased OTC insulin because they could not afford the copayment on prescription insulin; 70.1% of those positive responses came from Walmart pharmacies.

The coauthors acknowledged that their survey represents the impressions of pharmacy employees as opposed to actual sales data. The numbers, however, do “support an estimate of daily sales of more than 18,000 vials of over-the-counter insulin at Walmart pharmacies.” As for next steps, they noted that “further studies should explore clinical and safety outcomes related to the use of over-the-counter insulin.”

One author was supported by a grant from the National Institutes of Health. No conflicts of interest were reported.

SOURCE: Goldstein JN et al. JAMA Intern Med. 2019 Feb 18. doi: 10.1001/jamainternmed.2018.7279

Over-the-counter (OTC) insulin is chiefly purchased at Walmart pharmacies, likely because their brand is considerably less expensive than the OTC insulin sold elsewhere, according to a national survey of pharmacy employees.

The survey was undertaken by Jennifer N. Goldstein, MD, of Christiana Care Health System in Newark, Del., and her colleagues and published online Feb. 18 in JAMA Internal Medicine.

iStock/ThinkStock

As Walmart does not make its sales data public, Dr. Goldstein and her colleagues conducted a national telephone-based survey of Walmart and chain pharmacies in the 49 states where OTC insulin is available. They administered a five-item questionnaire that sought to determine the frequency of OTC insulin sales.

Of the 561 pharmacies that completed the questionnaire, 500 (89.1%) responded that they did sell OTC insulin; this included 284 of 292 Walmart pharmacies. Among those Walmart pharmacies, 247 (87%) said sales of OTC insulin occurred daily; 31 (10.9%) sold it weekly; and 3 (1.1%) sold it monthly.

The chains (CVS, Walgreens, Rite Aid) reported far less frequent sales of OTC insulin; 100 out of 216 (46.3%) pharmacies said they made sales only “a few times a year,” and none reported daily sales.

A majority of respondents (54.9%) also answered yes when asked if they believed patients purchased OTC insulin because they could not afford the copayment on prescription insulin; 70.1% of those positive responses came from Walmart pharmacies.

The coauthors acknowledged that their survey represents the impressions of pharmacy employees as opposed to actual sales data. The numbers, however, do “support an estimate of daily sales of more than 18,000 vials of over-the-counter insulin at Walmart pharmacies.” As for next steps, they noted that “further studies should explore clinical and safety outcomes related to the use of over-the-counter insulin.”

One author was supported by a grant from the National Institutes of Health. No conflicts of interest were reported.

SOURCE: Goldstein JN et al. JAMA Intern Med. 2019 Feb 18. doi: 10.1001/jamainternmed.2018.7279

Over-the-counter (OTC) insulin is chiefly purchased at Walmart pharmacies, likely because their brand is considerably less expensive than the OTC insulin sold elsewhere, according to a national survey of pharmacy employees.

The survey was undertaken by Jennifer N. Goldstein, MD, of Christiana Care Health System in Newark, Del., and her colleagues and published online Feb. 18 in JAMA Internal Medicine.

iStock/ThinkStock

As Walmart does not make its sales data public, Dr. Goldstein and her colleagues conducted a national telephone-based survey of Walmart and chain pharmacies in the 49 states where OTC insulin is available. They administered a five-item questionnaire that sought to determine the frequency of OTC insulin sales.

Of the 561 pharmacies that completed the questionnaire, 500 (89.1%) responded that they did sell OTC insulin; this included 284 of 292 Walmart pharmacies. Among those Walmart pharmacies, 247 (87%) said sales of OTC insulin occurred daily; 31 (10.9%) sold it weekly; and 3 (1.1%) sold it monthly.

The chains (CVS, Walgreens, Rite Aid) reported far less frequent sales of OTC insulin; 100 out of 216 (46.3%) pharmacies said they made sales only “a few times a year,” and none reported daily sales.

A majority of respondents (54.9%) also answered yes when asked if they believed patients purchased OTC insulin because they could not afford the copayment on prescription insulin; 70.1% of those positive responses came from Walmart pharmacies.

The coauthors acknowledged that their survey represents the impressions of pharmacy employees as opposed to actual sales data. The numbers, however, do “support an estimate of daily sales of more than 18,000 vials of over-the-counter insulin at Walmart pharmacies.” As for next steps, they noted that “further studies should explore clinical and safety outcomes related to the use of over-the-counter insulin.”

One author was supported by a grant from the National Institutes of Health. No conflicts of interest were reported.

SOURCE: Goldstein JN et al. JAMA Intern Med. 2019 Feb 18. doi: 10.1001/jamainternmed.2018.7279

The Food and Drug Administration has approved marketing of the Tandem Diabetes Care t:Slim X2 device, the first-ever interoperable, alternate controller–enabled infusion pump, for insulin delivery in children and adults with diabetes.

The pump delivers insulin under the skin at a variable or fixed rate. It can function on its own, or it can be digitally connected to automatically communicate with and receive drug-dosing commands from other diabetes management devices, such as automated insulin-dosing systems, the agency announced.

The approval was based on a review of performance data demonstrating that the device can deliver insulin accurately and reliably and at the rates and volumes programmed by the user. The agency also assessed the pump’s ability to connect reliably with other devices, as well as its cybersecurity and fail-safe modes.

Risks associated with the device were similar to those of other infusion pumps and include infection, bleeding, pain, or skin irritations. Blockages and air bubbles can occur in the tubing, which will affect drug delivery. Risks associated with incorrect drug delivery include hypo- and hyperglycemia as well as diabetic ketoacidosis.

“The marketing authorization of the [pump] has the potential to aid patients who seek more individualized diabetes therapy systems and opens the door for developers of future connected diabetes devices to get other safe and effective products to patients more efficiently,” FDA Commissioner Scott Gottlieb, MD, said in the announcement.

[email protected]

The Food and Drug Administration has approved marketing of the Tandem Diabetes Care t:Slim X2 device, the first-ever interoperable, alternate controller–enabled infusion pump, for insulin delivery in children and adults with diabetes.

The pump delivers insulin under the skin at a variable or fixed rate. It can function on its own, or it can be digitally connected to automatically communicate with and receive drug-dosing commands from other diabetes management devices, such as automated insulin-dosing systems, the agency announced.

The approval was based on a review of performance data demonstrating that the device can deliver insulin accurately and reliably and at the rates and volumes programmed by the user. The agency also assessed the pump’s ability to connect reliably with other devices, as well as its cybersecurity and fail-safe modes.

Risks associated with the device were similar to those of other infusion pumps and include infection, bleeding, pain, or skin irritations. Blockages and air bubbles can occur in the tubing, which will affect drug delivery. Risks associated with incorrect drug delivery include hypo- and hyperglycemia as well as diabetic ketoacidosis.

“The marketing authorization of the [pump] has the potential to aid patients who seek more individualized diabetes therapy systems and opens the door for developers of future connected diabetes devices to get other safe and effective products to patients more efficiently,” FDA Commissioner Scott Gottlieb, MD, said in the announcement.

[email protected]

The Food and Drug Administration has approved marketing of the Tandem Diabetes Care t:Slim X2 device, the first-ever interoperable, alternate controller–enabled infusion pump, for insulin delivery in children and adults with diabetes.

The pump delivers insulin under the skin at a variable or fixed rate. It can function on its own, or it can be digitally connected to automatically communicate with and receive drug-dosing commands from other diabetes management devices, such as automated insulin-dosing systems, the agency announced.

The approval was based on a review of performance data demonstrating that the device can deliver insulin accurately and reliably and at the rates and volumes programmed by the user. The agency also assessed the pump’s ability to connect reliably with other devices, as well as its cybersecurity and fail-safe modes.

Risks associated with the device were similar to those of other infusion pumps and include infection, bleeding, pain, or skin irritations. Blockages and air bubbles can occur in the tubing, which will affect drug delivery. Risks associated with incorrect drug delivery include hypo- and hyperglycemia as well as diabetic ketoacidosis.

“The marketing authorization of the [pump] has the potential to aid patients who seek more individualized diabetes therapy systems and opens the door for developers of future connected diabetes devices to get other safe and effective products to patients more efficiently,” FDA Commissioner Scott Gottlieb, MD, said in the announcement.

[email protected]

Click here to read the supplement

Topics Include:

• The cardiovascular (CV) safety and efficacy of the sodium-glucose cotransporter-2 inhibitor (SGLT-2i) dapagliflozin was evaluated in the DECLARE-TIMI 58 study.
• The study design and patient population of DECLARE-TIMI 58 differed from those of other CV outcomes trials of SGLT-2is (EMPA-REG OUTCOME and CANVAS)
• Key outcome results between the 3 CV outcomes trials of SGLT-2is may have been affected by variations in the study designs and patient populations, given that DECLARE-TIMI 58 evaluated a larger population of patients with type 2 diabetes (T2D) without prior CV events than the EMPA-REG OUTCOME or CANVAS studies, potentially allowing for a broader generalizability of the study results to patients with T2D in real-world clinical practice.

Click here to read the supplement

About the Author:

Bulent S. Atac, MD
Albert Einstein College of
Medicine and Montefiore Medical Center,
Bronx, New York

Click here to read the supplement

Topics Include:

• The cardiovascular (CV) safety and efficacy of the sodium-glucose cotransporter-2 inhibitor (SGLT-2i) dapagliflozin was evaluated in the DECLARE-TIMI 58 study.
• The study design and patient population of DECLARE-TIMI 58 differed from those of other CV outcomes trials of SGLT-2is (EMPA-REG OUTCOME and CANVAS)
• Key outcome results between the 3 CV outcomes trials of SGLT-2is may have been affected by variations in the study designs and patient populations, given that DECLARE-TIMI 58 evaluated a larger population of patients with type 2 diabetes (T2D) without prior CV events than the EMPA-REG OUTCOME or CANVAS studies, potentially allowing for a broader generalizability of the study results to patients with T2D in real-world clinical practice.

Click here to read the supplement

About the Author:

Bulent S. Atac, MD
Albert Einstein College of
Medicine and Montefiore Medical Center,
Bronx, New York

Click here to read the supplement

Topics Include:

• The cardiovascular (CV) safety and efficacy of the sodium-glucose cotransporter-2 inhibitor (SGLT-2i) dapagliflozin was evaluated in the DECLARE-TIMI 58 study.
• The study design and patient population of DECLARE-TIMI 58 differed from those of other CV outcomes trials of SGLT-2is (EMPA-REG OUTCOME and CANVAS)
• Key outcome results between the 3 CV outcomes trials of SGLT-2is may have been affected by variations in the study designs and patient populations, given that DECLARE-TIMI 58 evaluated a larger population of patients with type 2 diabetes (T2D) without prior CV events than the EMPA-REG OUTCOME or CANVAS studies, potentially allowing for a broader generalizability of the study results to patients with T2D in real-world clinical practice.

Click here to read the supplement

About the Author:

Bulent S. Atac, MD
Albert Einstein College of
Medicine and Montefiore Medical Center,
Bronx, New York

This week in MDedge Cardiocast: Most tPA-eligible stroke patients now get treated within an hour, atrial fib patients with prior hemorrhagic stroke who get their LAA closed can be safely treated, SGLT2 inhibitors are quickly morphing from diabetes to heart failure drugs, and the MESA cardiovascular disease risk calculator can be an improvement on the ACC/AHA version.

Amazon AlexaApple Podcasts

Google Podcasts

TuneIn

This week in MDedge Cardiocast: Most tPA-eligible stroke patients now get treated within an hour, atrial fib patients with prior hemorrhagic stroke who get their LAA closed can be safely treated, SGLT2 inhibitors are quickly morphing from diabetes to heart failure drugs, and the MESA cardiovascular disease risk calculator can be an improvement on the ACC/AHA version.

Amazon AlexaApple Podcasts

Google Podcasts

TuneIn

This week in MDedge Cardiocast: Most tPA-eligible stroke patients now get treated within an hour, atrial fib patients with prior hemorrhagic stroke who get their LAA closed can be safely treated, SGLT2 inhibitors are quickly morphing from diabetes to heart failure drugs, and the MESA cardiovascular disease risk calculator can be an improvement on the ACC/AHA version.

Amazon AlexaApple Podcasts

Google Podcasts

TuneIn

Diabetes mellitus (DM) was the third most common medical diagnosis in 2016.¹ Uncontrolled DM can lead to cardiovascular disease, nephropathy, neuropathy, and retinopathy. It is estimated that only 52.5% of patients with DM have achieved their goal hemoglobin A_1c (HbA_1c) level. The 2018 American Diabetes Association (ADA) clinical guidelines lack strong recommendations on sequential therapy for patients who have received a diagnosis of type 2 diabetes mellitus (T2DM) and have been unable to achieve their goal HbA_1c level with lifestyle changes and maximum-dose metformin.² Although those guidelines support treatment intensification with a glucagon-like peptide 1 receptor agonist (GLP-1 RA), prescribing patterns for T2DM most commonly include adding insulin to try to control blood glucose and reduce long-term comorbidities.^2,3

Related: 

Insulin therapy is known for its ability to effectively lower blood glucose and HbA_1c levels but comes with many limitations. Mealtime insulin has the highest risk of hypoglycemia, causes significant weight gain, requires several additional injections per day, and additional monitoring of blood glucose.^4,5 The 2018 ADA guidelines state that hypoglycemia is the major limiting factor in the management of insulin-treated T2DM.²

Compared with mealtime insulin, GLP-1 RAs have the benefit of reducing the risk of hypoglycemia, weight gain, and number of daily injections.⁵ In addition, compared with insulin alone, GLP-1 RAs have the advantage of reducing glycemic variability.⁶ These advantages are especially attractive in the treatment of geriatric patients. Given its mechanism of action, liraglutide is expected to have an effect on both fasting and postprandial blood glucose. There are no recommendations on how to empirically reduce the dose of insulin when starting liraglutide.⁷

Background

GLP-1 is an incretin hormone that is secreted in response to meal ingestion. GLP-1 stimulates insulin release, suppresses elevated glucagon levels, and delays gastric emptying. Patients with a DM diagnosis have impaired secretion of GLP-1.⁸

The GLP-1 RA liraglutide was approved by the FDA in January 2010 as a once-daily injection for patients with uncontrolled T2DM despite lifestyle changes and metformin monotherapy. Because of its intermediate half-life, liraglutide has an effect on both fasting and postprandial blood glucose.⁷ GLP-1 RAs are associated with reduced hypoglycemic episodes—an association attributable to the mechanism of action and potentially to improved pancreatic α-cell function.^3,4 In July 2016, results of the LEADER trial showed that liraglutide therapy had a cardiovascular benefit in high-risk patients.⁸ In October 2017, liraglutide was FDAapproved for reducing 3-point major adverse cardiac events.⁷

Xultophy (Novo Nordisk, Plainsboro, NJ) is a fixed-dose medication combining degludec, a long-acting basal insulin analog, with liraglutide. As seen in the DUAL trials, Xultophy was more beneficial in reducing HbA_1c levels than each component alone, and minimized hypoglycemic events, weight gain, and complexity of insulin treatment intensification.^9-11 Therapy that combines basal insulin and a GLP-1 RA may be more effective than either agent as monotherapy and may have a significant impact on cardiovascular risk because of the synergistic vasodilatory, anti-inflammatory, and antioxidant properties of insulin and GLP-1 RA.⁶ In addition, combination therapy offers many benefits over traditional basal and bolus insulin regimens. These benefits include fewer daily injections, additional weight reduction resulting from the reduced insulin requirement, and fewer episodes of hypoglycemia. Reported gastrointestinal adverse effects have been transient and were not augmented when a GLP-1 RA was used in combination with basal insulin.¹¹

Methods

We performed a retrospective chart analysis to quantify the benefit of using liraglutide as an add-on therapy to basal and bolus insulin regimens in veterans treated at VA Boston Healthcare System (VABHS). The analysis evaluated changes in insulin doses and HbA_1c levels when liraglutide was added to these regimens. Patients identified for the study had electronic medication orders for concurrent therapy with liraglutide, insulin glargine, and insulin aspart filled through outpatient VABHS campus pharmacies for at least 3 months between January 2010 and December 2016. Sixty-nine patients who were on basal-bolus insulin for T2DM and who were prescribed liraglutide for treatment intensification were screened for inclusion and exclusion criteria. Data were analyzed at baseline and 3 months after liraglutide treatment.

Study Protocol

The inclusion criteria were patients aged ≥ 18 years, T2DM diagnosis, and therapy with insulin glargine and insulin aspart for at least 3 months before treatment intensi fication with liraglutide. Exclusion criteria were diagnosis of type 1 DM. To accurately quantify mean change in number of insulin units used, the study included patients only if they had been prescribed insulin glargine and insulin aspart before starting liraglutide. All other insulin regimens were excluded. To detect the true change that occurs when liraglutide is added to basal-bolus insulin, the study also excluded patients if they had been previously prescribed another GLP-1 RA. Patients with contraindications to liraglutide, insulin aspart, or insulin glargine were excluded as well. In addition, patients were excluded from the exposed arm if they were injecting < 1.2 mg of liraglutide once daily or if they had been on liraglutide for < 3 months.

Study Outcomes

All 35 patients who met the inclusion and exclusion criteria were included in this retrospective chart review. The primary outcome was determined by changes in HbA_1c level and number of insulin doses 3 months after treatment with liraglutide. For each patient, a chart review was performed to determine the amount of insulin added or reduced during the study period. Data were collected at baseline and 3 months after initiation of liraglutide.

Statistical Analysis

Statistical analyses were performed with SPSS Version 20.0 (IBM, Armonk, NY). Population characteristics and study outcomes with normal distribution were compared using a paired t test and are reported as means with standard deviations. Nonnormally distributed variables (bolus insulin, HbA_1c level) were compared using the nonparametric Wilcoxon rank sum test and are reported as median values with interquartile ranges. Normality was tested with the Shapiro-Wilk test. The primary outcome evaluated was change in number of insulin units used. Secondary outcomes included change in HbA_1c level and change in body weight. A Bonferroni correction for multiple comparisons was used to prevent type I error. Significance at the Bonferroni-corrected level of .01 (.05/5 = .01) is indicated.

Results

Patients were included if they were previously on insulin glargine and insulin aspart before starting liraglutide for treatment intensification. 

As Table 1 indicates, 100% of patients were male, and mean (SD) age was 65.5 (9.3) years. 

Discussion

After 3 months of treatment with liraglutide, patients experienced a significant decrease in HbA_1c levels. Insulin doses also decreased, but this finding was not statistically significant after correcting for multiple testing. These results are similar with those in larger studies of the effectiveness of liraglutide and the addition of liraglutide to insulin therapy. ^6,8,12,13 Liraglutide has been shown to decrease HbA_1c levels, lower rates of progression of kidney failure, decrease weight, and provide cardiovascular benefit.

In a prospective, randomized controlled trial evaluating the effect of adding liraglutide to insulin therapy, 21 of the 37 patients who had T2DM and required more than 100 total units of basal-bolus insulin daily were initiated on liraglutide, and changes in HbA_1c level, body weight, and glycemic variability were compared. Results showed statistically significant improvement in all 3 outcomes in the group treated with liraglutide.⁶ Our findings, in conjunction with those of the larger studies, suggest that many of these results are generalizable to our local veteran population. Importantly, liraglutide was successfully started in pharmacy clinics—an indication that this treatment need not be initiated by an endocrine specialist.

Limitations

Given the lack of gender and racial diversity in this study population, our findings have limited generalizability to other populations. It is possible that, with a larger sample size, these results regarding reduced basal insulin doses would be significant. It has been hypothesized that patients experience fewer episodes of hypoglycemia when insulin doses are reduced, but we were unable to measure the frequency of these episodes. Other study limitations include inability to assess adherence and inability to account for concurrent regimens and/or for lifestyle changes that may have been made during the study period. Further, the study did not collect data on changes made to current DM medication regimens during the study period, and these changes may have influenced outcomes.

Conclusion

Patients who require treatment intensification for insulin-dependent T2DM may benefit from having liraglutide added to their basal-bolus insulin regimen. Liraglutide may prove to be more favorable than bolus insulin when choosing add-on therapy to basal insulin. Benefits include reductions in insulin doses, HbA_1c levels, number of daily injections, and body weight. Therefore, we suggest that empirically reducing basal insulin by 10% to 25% and bolus insulin by 25% to 50% will avoid relative hypoglycemia. Prescribers must keep in mind patient-specific factors when adjusting insulin doses, if these doses are adjusted at all. Follow-up of 2 to 4 weeks is recommended for review of home monitoring of glucose for further insulin adjustments.

This study has important clinical implications. First, the finding of a reduction in HbA_1c levels supports use of liraglutide therapy for HbA_1c reduction in veterans. Second, the number of veterans who were successfully initiated on liraglutide therapy by nonphysician providers indicates that liraglutide can be effectively and safely started in primary care pharmacy clinics, increasing access to the medication.

Diabetes mellitus (DM) was the third most common medical diagnosis in 2016.¹ Uncontrolled DM can lead to cardiovascular disease, nephropathy, neuropathy, and retinopathy. It is estimated that only 52.5% of patients with DM have achieved their goal hemoglobin A_1c (HbA_1c) level. The 2018 American Diabetes Association (ADA) clinical guidelines lack strong recommendations on sequential therapy for patients who have received a diagnosis of type 2 diabetes mellitus (T2DM) and have been unable to achieve their goal HbA_1c level with lifestyle changes and maximum-dose metformin.² Although those guidelines support treatment intensification with a glucagon-like peptide 1 receptor agonist (GLP-1 RA), prescribing patterns for T2DM most commonly include adding insulin to try to control blood glucose and reduce long-term comorbidities.^2,3

Related: 

Insulin therapy is known for its ability to effectively lower blood glucose and HbA_1c levels but comes with many limitations. Mealtime insulin has the highest risk of hypoglycemia, causes significant weight gain, requires several additional injections per day, and additional monitoring of blood glucose.^4,5 The 2018 ADA guidelines state that hypoglycemia is the major limiting factor in the management of insulin-treated T2DM.²

Compared with mealtime insulin, GLP-1 RAs have the benefit of reducing the risk of hypoglycemia, weight gain, and number of daily injections.⁵ In addition, compared with insulin alone, GLP-1 RAs have the advantage of reducing glycemic variability.⁶ These advantages are especially attractive in the treatment of geriatric patients. Given its mechanism of action, liraglutide is expected to have an effect on both fasting and postprandial blood glucose. There are no recommendations on how to empirically reduce the dose of insulin when starting liraglutide.⁷

Background

GLP-1 is an incretin hormone that is secreted in response to meal ingestion. GLP-1 stimulates insulin release, suppresses elevated glucagon levels, and delays gastric emptying. Patients with a DM diagnosis have impaired secretion of GLP-1.⁸

The GLP-1 RA liraglutide was approved by the FDA in January 2010 as a once-daily injection for patients with uncontrolled T2DM despite lifestyle changes and metformin monotherapy. Because of its intermediate half-life, liraglutide has an effect on both fasting and postprandial blood glucose.⁷ GLP-1 RAs are associated with reduced hypoglycemic episodes—an association attributable to the mechanism of action and potentially to improved pancreatic α-cell function.^3,4 In July 2016, results of the LEADER trial showed that liraglutide therapy had a cardiovascular benefit in high-risk patients.⁸ In October 2017, liraglutide was FDAapproved for reducing 3-point major adverse cardiac events.⁷

Xultophy (Novo Nordisk, Plainsboro, NJ) is a fixed-dose medication combining degludec, a long-acting basal insulin analog, with liraglutide. As seen in the DUAL trials, Xultophy was more beneficial in reducing HbA_1c levels than each component alone, and minimized hypoglycemic events, weight gain, and complexity of insulin treatment intensification.^9-11 Therapy that combines basal insulin and a GLP-1 RA may be more effective than either agent as monotherapy and may have a significant impact on cardiovascular risk because of the synergistic vasodilatory, anti-inflammatory, and antioxidant properties of insulin and GLP-1 RA.⁶ In addition, combination therapy offers many benefits over traditional basal and bolus insulin regimens. These benefits include fewer daily injections, additional weight reduction resulting from the reduced insulin requirement, and fewer episodes of hypoglycemia. Reported gastrointestinal adverse effects have been transient and were not augmented when a GLP-1 RA was used in combination with basal insulin.¹¹

Methods

We performed a retrospective chart analysis to quantify the benefit of using liraglutide as an add-on therapy to basal and bolus insulin regimens in veterans treated at VA Boston Healthcare System (VABHS). The analysis evaluated changes in insulin doses and HbA_1c levels when liraglutide was added to these regimens. Patients identified for the study had electronic medication orders for concurrent therapy with liraglutide, insulin glargine, and insulin aspart filled through outpatient VABHS campus pharmacies for at least 3 months between January 2010 and December 2016. Sixty-nine patients who were on basal-bolus insulin for T2DM and who were prescribed liraglutide for treatment intensification were screened for inclusion and exclusion criteria. Data were analyzed at baseline and 3 months after liraglutide treatment.

Study Protocol

The inclusion criteria were patients aged ≥ 18 years, T2DM diagnosis, and therapy with insulin glargine and insulin aspart for at least 3 months before treatment intensi fication with liraglutide. Exclusion criteria were diagnosis of type 1 DM. To accurately quantify mean change in number of insulin units used, the study included patients only if they had been prescribed insulin glargine and insulin aspart before starting liraglutide. All other insulin regimens were excluded. To detect the true change that occurs when liraglutide is added to basal-bolus insulin, the study also excluded patients if they had been previously prescribed another GLP-1 RA. Patients with contraindications to liraglutide, insulin aspart, or insulin glargine were excluded as well. In addition, patients were excluded from the exposed arm if they were injecting < 1.2 mg of liraglutide once daily or if they had been on liraglutide for < 3 months.

Study Outcomes

All 35 patients who met the inclusion and exclusion criteria were included in this retrospective chart review. The primary outcome was determined by changes in HbA_1c level and number of insulin doses 3 months after treatment with liraglutide. For each patient, a chart review was performed to determine the amount of insulin added or reduced during the study period. Data were collected at baseline and 3 months after initiation of liraglutide.

Statistical Analysis

Statistical analyses were performed with SPSS Version 20.0 (IBM, Armonk, NY). Population characteristics and study outcomes with normal distribution were compared using a paired t test and are reported as means with standard deviations. Nonnormally distributed variables (bolus insulin, HbA_1c level) were compared using the nonparametric Wilcoxon rank sum test and are reported as median values with interquartile ranges. Normality was tested with the Shapiro-Wilk test. The primary outcome evaluated was change in number of insulin units used. Secondary outcomes included change in HbA_1c level and change in body weight. A Bonferroni correction for multiple comparisons was used to prevent type I error. Significance at the Bonferroni-corrected level of .01 (.05/5 = .01) is indicated.

Results

Patients were included if they were previously on insulin glargine and insulin aspart before starting liraglutide for treatment intensification. 

As Table 1 indicates, 100% of patients were male, and mean (SD) age was 65.5 (9.3) years. 

Discussion

After 3 months of treatment with liraglutide, patients experienced a significant decrease in HbA_1c levels. Insulin doses also decreased, but this finding was not statistically significant after correcting for multiple testing. These results are similar with those in larger studies of the effectiveness of liraglutide and the addition of liraglutide to insulin therapy. ^6,8,12,13 Liraglutide has been shown to decrease HbA_1c levels, lower rates of progression of kidney failure, decrease weight, and provide cardiovascular benefit.

In a prospective, randomized controlled trial evaluating the effect of adding liraglutide to insulin therapy, 21 of the 37 patients who had T2DM and required more than 100 total units of basal-bolus insulin daily were initiated on liraglutide, and changes in HbA_1c level, body weight, and glycemic variability were compared. Results showed statistically significant improvement in all 3 outcomes in the group treated with liraglutide.⁶ Our findings, in conjunction with those of the larger studies, suggest that many of these results are generalizable to our local veteran population. Importantly, liraglutide was successfully started in pharmacy clinics—an indication that this treatment need not be initiated by an endocrine specialist.

Limitations

Given the lack of gender and racial diversity in this study population, our findings have limited generalizability to other populations. It is possible that, with a larger sample size, these results regarding reduced basal insulin doses would be significant. It has been hypothesized that patients experience fewer episodes of hypoglycemia when insulin doses are reduced, but we were unable to measure the frequency of these episodes. Other study limitations include inability to assess adherence and inability to account for concurrent regimens and/or for lifestyle changes that may have been made during the study period. Further, the study did not collect data on changes made to current DM medication regimens during the study period, and these changes may have influenced outcomes.

Conclusion

Patients who require treatment intensification for insulin-dependent T2DM may benefit from having liraglutide added to their basal-bolus insulin regimen. Liraglutide may prove to be more favorable than bolus insulin when choosing add-on therapy to basal insulin. Benefits include reductions in insulin doses, HbA_1c levels, number of daily injections, and body weight. Therefore, we suggest that empirically reducing basal insulin by 10% to 25% and bolus insulin by 25% to 50% will avoid relative hypoglycemia. Prescribers must keep in mind patient-specific factors when adjusting insulin doses, if these doses are adjusted at all. Follow-up of 2 to 4 weeks is recommended for review of home monitoring of glucose for further insulin adjustments.

This study has important clinical implications. First, the finding of a reduction in HbA_1c levels supports use of liraglutide therapy for HbA_1c reduction in veterans. Second, the number of veterans who were successfully initiated on liraglutide therapy by nonphysician providers indicates that liraglutide can be effectively and safely started in primary care pharmacy clinics, increasing access to the medication.

Diabetes mellitus (DM) was the third most common medical diagnosis in 2016.¹ Uncontrolled DM can lead to cardiovascular disease, nephropathy, neuropathy, and retinopathy. It is estimated that only 52.5% of patients with DM have achieved their goal hemoglobin A_1c (HbA_1c) level. The 2018 American Diabetes Association (ADA) clinical guidelines lack strong recommendations on sequential therapy for patients who have received a diagnosis of type 2 diabetes mellitus (T2DM) and have been unable to achieve their goal HbA_1c level with lifestyle changes and maximum-dose metformin.² Although those guidelines support treatment intensification with a glucagon-like peptide 1 receptor agonist (GLP-1 RA), prescribing patterns for T2DM most commonly include adding insulin to try to control blood glucose and reduce long-term comorbidities.^2,3

Related: 

Insulin therapy is known for its ability to effectively lower blood glucose and HbA_1c levels but comes with many limitations. Mealtime insulin has the highest risk of hypoglycemia, causes significant weight gain, requires several additional injections per day, and additional monitoring of blood glucose.^4,5 The 2018 ADA guidelines state that hypoglycemia is the major limiting factor in the management of insulin-treated T2DM.²

Compared with mealtime insulin, GLP-1 RAs have the benefit of reducing the risk of hypoglycemia, weight gain, and number of daily injections.⁵ In addition, compared with insulin alone, GLP-1 RAs have the advantage of reducing glycemic variability.⁶ These advantages are especially attractive in the treatment of geriatric patients. Given its mechanism of action, liraglutide is expected to have an effect on both fasting and postprandial blood glucose. There are no recommendations on how to empirically reduce the dose of insulin when starting liraglutide.⁷

Background

GLP-1 is an incretin hormone that is secreted in response to meal ingestion. GLP-1 stimulates insulin release, suppresses elevated glucagon levels, and delays gastric emptying. Patients with a DM diagnosis have impaired secretion of GLP-1.⁸

The GLP-1 RA liraglutide was approved by the FDA in January 2010 as a once-daily injection for patients with uncontrolled T2DM despite lifestyle changes and metformin monotherapy. Because of its intermediate half-life, liraglutide has an effect on both fasting and postprandial blood glucose.⁷ GLP-1 RAs are associated with reduced hypoglycemic episodes—an association attributable to the mechanism of action and potentially to improved pancreatic α-cell function.^3,4 In July 2016, results of the LEADER trial showed that liraglutide therapy had a cardiovascular benefit in high-risk patients.⁸ In October 2017, liraglutide was FDAapproved for reducing 3-point major adverse cardiac events.⁷

Xultophy (Novo Nordisk, Plainsboro, NJ) is a fixed-dose medication combining degludec, a long-acting basal insulin analog, with liraglutide. As seen in the DUAL trials, Xultophy was more beneficial in reducing HbA_1c levels than each component alone, and minimized hypoglycemic events, weight gain, and complexity of insulin treatment intensification.^9-11 Therapy that combines basal insulin and a GLP-1 RA may be more effective than either agent as monotherapy and may have a significant impact on cardiovascular risk because of the synergistic vasodilatory, anti-inflammatory, and antioxidant properties of insulin and GLP-1 RA.⁶ In addition, combination therapy offers many benefits over traditional basal and bolus insulin regimens. These benefits include fewer daily injections, additional weight reduction resulting from the reduced insulin requirement, and fewer episodes of hypoglycemia. Reported gastrointestinal adverse effects have been transient and were not augmented when a GLP-1 RA was used in combination with basal insulin.¹¹

Methods

We performed a retrospective chart analysis to quantify the benefit of using liraglutide as an add-on therapy to basal and bolus insulin regimens in veterans treated at VA Boston Healthcare System (VABHS). The analysis evaluated changes in insulin doses and HbA_1c levels when liraglutide was added to these regimens. Patients identified for the study had electronic medication orders for concurrent therapy with liraglutide, insulin glargine, and insulin aspart filled through outpatient VABHS campus pharmacies for at least 3 months between January 2010 and December 2016. Sixty-nine patients who were on basal-bolus insulin for T2DM and who were prescribed liraglutide for treatment intensification were screened for inclusion and exclusion criteria. Data were analyzed at baseline and 3 months after liraglutide treatment.

Study Protocol

The inclusion criteria were patients aged ≥ 18 years, T2DM diagnosis, and therapy with insulin glargine and insulin aspart for at least 3 months before treatment intensi fication with liraglutide. Exclusion criteria were diagnosis of type 1 DM. To accurately quantify mean change in number of insulin units used, the study included patients only if they had been prescribed insulin glargine and insulin aspart before starting liraglutide. All other insulin regimens were excluded. To detect the true change that occurs when liraglutide is added to basal-bolus insulin, the study also excluded patients if they had been previously prescribed another GLP-1 RA. Patients with contraindications to liraglutide, insulin aspart, or insulin glargine were excluded as well. In addition, patients were excluded from the exposed arm if they were injecting < 1.2 mg of liraglutide once daily or if they had been on liraglutide for < 3 months.

Study Outcomes

All 35 patients who met the inclusion and exclusion criteria were included in this retrospective chart review. The primary outcome was determined by changes in HbA_1c level and number of insulin doses 3 months after treatment with liraglutide. For each patient, a chart review was performed to determine the amount of insulin added or reduced during the study period. Data were collected at baseline and 3 months after initiation of liraglutide.

Statistical Analysis

Statistical analyses were performed with SPSS Version 20.0 (IBM, Armonk, NY). Population characteristics and study outcomes with normal distribution were compared using a paired t test and are reported as means with standard deviations. Nonnormally distributed variables (bolus insulin, HbA_1c level) were compared using the nonparametric Wilcoxon rank sum test and are reported as median values with interquartile ranges. Normality was tested with the Shapiro-Wilk test. The primary outcome evaluated was change in number of insulin units used. Secondary outcomes included change in HbA_1c level and change in body weight. A Bonferroni correction for multiple comparisons was used to prevent type I error. Significance at the Bonferroni-corrected level of .01 (.05/5 = .01) is indicated.

Results

Patients were included if they were previously on insulin glargine and insulin aspart before starting liraglutide for treatment intensification. 

As Table 1 indicates, 100% of patients were male, and mean (SD) age was 65.5 (9.3) years. 

Discussion

After 3 months of treatment with liraglutide, patients experienced a significant decrease in HbA_1c levels. Insulin doses also decreased, but this finding was not statistically significant after correcting for multiple testing. These results are similar with those in larger studies of the effectiveness of liraglutide and the addition of liraglutide to insulin therapy. ^6,8,12,13 Liraglutide has been shown to decrease HbA_1c levels, lower rates of progression of kidney failure, decrease weight, and provide cardiovascular benefit.

In a prospective, randomized controlled trial evaluating the effect of adding liraglutide to insulin therapy, 21 of the 37 patients who had T2DM and required more than 100 total units of basal-bolus insulin daily were initiated on liraglutide, and changes in HbA_1c level, body weight, and glycemic variability were compared. Results showed statistically significant improvement in all 3 outcomes in the group treated with liraglutide.⁶ Our findings, in conjunction with those of the larger studies, suggest that many of these results are generalizable to our local veteran population. Importantly, liraglutide was successfully started in pharmacy clinics—an indication that this treatment need not be initiated by an endocrine specialist.

Limitations

Given the lack of gender and racial diversity in this study population, our findings have limited generalizability to other populations. It is possible that, with a larger sample size, these results regarding reduced basal insulin doses would be significant. It has been hypothesized that patients experience fewer episodes of hypoglycemia when insulin doses are reduced, but we were unable to measure the frequency of these episodes. Other study limitations include inability to assess adherence and inability to account for concurrent regimens and/or for lifestyle changes that may have been made during the study period. Further, the study did not collect data on changes made to current DM medication regimens during the study period, and these changes may have influenced outcomes.

Conclusion

Patients who require treatment intensification for insulin-dependent T2DM may benefit from having liraglutide added to their basal-bolus insulin regimen. Liraglutide may prove to be more favorable than bolus insulin when choosing add-on therapy to basal insulin. Benefits include reductions in insulin doses, HbA_1c levels, number of daily injections, and body weight. Therefore, we suggest that empirically reducing basal insulin by 10% to 25% and bolus insulin by 25% to 50% will avoid relative hypoglycemia. Prescribers must keep in mind patient-specific factors when adjusting insulin doses, if these doses are adjusted at all. Follow-up of 2 to 4 weeks is recommended for review of home monitoring of glucose for further insulin adjustments.

This study has important clinical implications. First, the finding of a reduction in HbA_1c levels supports use of liraglutide therapy for HbA_1c reduction in veterans. Second, the number of veterans who were successfully initiated on liraglutide therapy by nonphysician providers indicates that liraglutide can be effectively and safely started in primary care pharmacy clinics, increasing access to the medication.

SNOWMASS, COLO. – The Multi-Ethnic Study of Atherosclerosis–based 10-year coronary heart disease risk calculator offers significant advantages over the far more widely used Atherosclerotic Cardiovascular Disease risk estimator based upon the pooled cohort equations recommended in the American College of Cardiology/American Heart Association guidelines, Robert A. Vogel, MD, said at the Annual Cardiovascular Conference at Snowmass.

Bruce Jancin/MDedge News

“I don’t use the PCE very much. I use the MESA [Multi-Ethnic Study of Atherosclerosis]. I like it because it gives me options I don’t have with the PCE [pooled cohort equations],” Dr. Vogel, a preventive cardiology expert at the University of Colorado at Denver, Aurora, said at the meeting sponsored by the American College of Cardiology.

Those added options include, importantly, the ability to plug in a patient’s coronary artery calcium score. MESA, a landmark longitudinal National Institutes of Health–sponsored study, generated a great deal of the data that established the prognostic value of measuring coronary artery calcium.

“You can do the MESA worksheet with or without coronary artery calcium, either way,” Dr. Vogel noted.

Another big advantage for the MESA risk calculator: It asks a yes/no question about family history of heart attack in first-degree relatives. That’s truly risk-altering information, and yet it’s absent from the ACC/AHA ASCVD (Atherosclerotic Cardiovascular Disease) risk calculator, the cardiologist continued.

Plus, the MESA risk calculator includes four ethnic options: Caucasian, African American, Chinese, or Hispanic, while the ACC/AHA’s PCE-based tool includes only the categories of white, African American, or other.

Dr. Vogel offered a case example for purposes of comparison and contrast of the two risk calculators: a 48-year-old white man with no history or symptoms of cardiovascular disease whose father had an MI at age 52. He wants to know whether he should start taking a statin. The patient is a former smoker who quit 3 years ago. His physical exam is normal. He has a body mass index of 29 kg/m^2, an LDL cholesterol of 134 mg/dL, a total cholesterol of 194 mg/dL, a triglyceride level of 92 mg/dL, blood pressure of 128/78 mm Hg, an HDL cholesterol of 42 mg/dL, and a hemoglobin A_1c of 5.9%.

Using the PCE-based risk calculator, the man’s 10-year risk of cardiovascular disease is only 3.3%, which falls below the ACC/AHA guideline-recommended threshold for statin therapy for primary prevention. But with the MESA score, as a result of the additional information regarding the family history of premature cardiovascular disease, the patient’s risk jumps to 4.9%, even without including a coronary artery calcium score.

“This shows you how big a factor the family history is. I think it’s unfortunate that the PCE doesn’t put it in,” Dr. Vogel said.

Indeed, in an analysis from MESA, a 60-year-old man with a lipid and blood pressure profile similar to that of Dr. Vogel’s patient would have a 10-year cardiovascular disease risk of 6% with a negative family history and a 9% risk with a positive history, he noted.

In the 48-year-old patient used as an example by Dr. Vogel, plugging into the MESA risk calculator a coronary artery calcium score of, say, 50, 100, or 150 Agatston units would boost the 10-year cardiovascular disease risk to 7.5%, 8.9%, and 9.9%, respectively.

Both the MESA and the ACC/AHA risk calculators incorporate diabetes as a simple yes/no item. It’s either present or absent. This is too crude a dichotomy for Dr. Vogel’s liking in light of data showing that individuals with impaired glucose tolerance have a cumulative risk of cardiovascular mortality that’s intermediate between diabetic and normoglycemic individuals. And of course, the patient in his case example had a Hb_A1c of 5.9%. So what is a physician to do?

“I fudge the numbers,” he said. “I move the dial a little bit one way or the other depending on whether a patient falls into that impaired glucose tolerance category.”

He also adjusts a former smoker’s estimated 10-year cardiovascular risk based upon how long ago the smoker quit. An ex-smoker’s hazard ratio for coronary heart disease doesn’t drop to that of a never-smoker until 8 years after quitting. And since the patient in this example has been tobacco-free for only 3 years, Dr. Vogel bumps up that individual’s risk level once again. So at this point, even without the additional information that would be provided by a coronary calcium score, the patient’s adjusted MESA 10-year risk is in the 10% range, compared with the 3.3% figure derived using the PCE-based risk calculator.

He reported serving as a paid consultant to the National Football League and the Pritikin Longevity Center, receiving research grants from Sanofi, and serving on speakers bureaus for Regeneron and Sanofi.

[email protected]

SNOWMASS, COLO. – The Multi-Ethnic Study of Atherosclerosis–based 10-year coronary heart disease risk calculator offers significant advantages over the far more widely used Atherosclerotic Cardiovascular Disease risk estimator based upon the pooled cohort equations recommended in the American College of Cardiology/American Heart Association guidelines, Robert A. Vogel, MD, said at the Annual Cardiovascular Conference at Snowmass.

Bruce Jancin/MDedge News

“I don’t use the PCE very much. I use the MESA [Multi-Ethnic Study of Atherosclerosis]. I like it because it gives me options I don’t have with the PCE [pooled cohort equations],” Dr. Vogel, a preventive cardiology expert at the University of Colorado at Denver, Aurora, said at the meeting sponsored by the American College of Cardiology.

Those added options include, importantly, the ability to plug in a patient’s coronary artery calcium score. MESA, a landmark longitudinal National Institutes of Health–sponsored study, generated a great deal of the data that established the prognostic value of measuring coronary artery calcium.

“You can do the MESA worksheet with or without coronary artery calcium, either way,” Dr. Vogel noted.

Another big advantage for the MESA risk calculator: It asks a yes/no question about family history of heart attack in first-degree relatives. That’s truly risk-altering information, and yet it’s absent from the ACC/AHA ASCVD (Atherosclerotic Cardiovascular Disease) risk calculator, the cardiologist continued.

Plus, the MESA risk calculator includes four ethnic options: Caucasian, African American, Chinese, or Hispanic, while the ACC/AHA’s PCE-based tool includes only the categories of white, African American, or other.

Dr. Vogel offered a case example for purposes of comparison and contrast of the two risk calculators: a 48-year-old white man with no history or symptoms of cardiovascular disease whose father had an MI at age 52. He wants to know whether he should start taking a statin. The patient is a former smoker who quit 3 years ago. His physical exam is normal. He has a body mass index of 29 kg/m^2, an LDL cholesterol of 134 mg/dL, a total cholesterol of 194 mg/dL, a triglyceride level of 92 mg/dL, blood pressure of 128/78 mm Hg, an HDL cholesterol of 42 mg/dL, and a hemoglobin A_1c of 5.9%.

Using the PCE-based risk calculator, the man’s 10-year risk of cardiovascular disease is only 3.3%, which falls below the ACC/AHA guideline-recommended threshold for statin therapy for primary prevention. But with the MESA score, as a result of the additional information regarding the family history of premature cardiovascular disease, the patient’s risk jumps to 4.9%, even without including a coronary artery calcium score.

“This shows you how big a factor the family history is. I think it’s unfortunate that the PCE doesn’t put it in,” Dr. Vogel said.

Indeed, in an analysis from MESA, a 60-year-old man with a lipid and blood pressure profile similar to that of Dr. Vogel’s patient would have a 10-year cardiovascular disease risk of 6% with a negative family history and a 9% risk with a positive history, he noted.

In the 48-year-old patient used as an example by Dr. Vogel, plugging into the MESA risk calculator a coronary artery calcium score of, say, 50, 100, or 150 Agatston units would boost the 10-year cardiovascular disease risk to 7.5%, 8.9%, and 9.9%, respectively.

Both the MESA and the ACC/AHA risk calculators incorporate diabetes as a simple yes/no item. It’s either present or absent. This is too crude a dichotomy for Dr. Vogel’s liking in light of data showing that individuals with impaired glucose tolerance have a cumulative risk of cardiovascular mortality that’s intermediate between diabetic and normoglycemic individuals. And of course, the patient in his case example had a Hb_A1c of 5.9%. So what is a physician to do?

“I fudge the numbers,” he said. “I move the dial a little bit one way or the other depending on whether a patient falls into that impaired glucose tolerance category.”

He also adjusts a former smoker’s estimated 10-year cardiovascular risk based upon how long ago the smoker quit. An ex-smoker’s hazard ratio for coronary heart disease doesn’t drop to that of a never-smoker until 8 years after quitting. And since the patient in this example has been tobacco-free for only 3 years, Dr. Vogel bumps up that individual’s risk level once again. So at this point, even without the additional information that would be provided by a coronary calcium score, the patient’s adjusted MESA 10-year risk is in the 10% range, compared with the 3.3% figure derived using the PCE-based risk calculator.

He reported serving as a paid consultant to the National Football League and the Pritikin Longevity Center, receiving research grants from Sanofi, and serving on speakers bureaus for Regeneron and Sanofi.

[email protected]

SNOWMASS, COLO. – The Multi-Ethnic Study of Atherosclerosis–based 10-year coronary heart disease risk calculator offers significant advantages over the far more widely used Atherosclerotic Cardiovascular Disease risk estimator based upon the pooled cohort equations recommended in the American College of Cardiology/American Heart Association guidelines, Robert A. Vogel, MD, said at the Annual Cardiovascular Conference at Snowmass.

Bruce Jancin/MDedge News

“I don’t use the PCE very much. I use the MESA [Multi-Ethnic Study of Atherosclerosis]. I like it because it gives me options I don’t have with the PCE [pooled cohort equations],” Dr. Vogel, a preventive cardiology expert at the University of Colorado at Denver, Aurora, said at the meeting sponsored by the American College of Cardiology.

Those added options include, importantly, the ability to plug in a patient’s coronary artery calcium score. MESA, a landmark longitudinal National Institutes of Health–sponsored study, generated a great deal of the data that established the prognostic value of measuring coronary artery calcium.

“You can do the MESA worksheet with or without coronary artery calcium, either way,” Dr. Vogel noted.

Another big advantage for the MESA risk calculator: It asks a yes/no question about family history of heart attack in first-degree relatives. That’s truly risk-altering information, and yet it’s absent from the ACC/AHA ASCVD (Atherosclerotic Cardiovascular Disease) risk calculator, the cardiologist continued.

Plus, the MESA risk calculator includes four ethnic options: Caucasian, African American, Chinese, or Hispanic, while the ACC/AHA’s PCE-based tool includes only the categories of white, African American, or other.

Dr. Vogel offered a case example for purposes of comparison and contrast of the two risk calculators: a 48-year-old white man with no history or symptoms of cardiovascular disease whose father had an MI at age 52. He wants to know whether he should start taking a statin. The patient is a former smoker who quit 3 years ago. His physical exam is normal. He has a body mass index of 29 kg/m^2, an LDL cholesterol of 134 mg/dL, a total cholesterol of 194 mg/dL, a triglyceride level of 92 mg/dL, blood pressure of 128/78 mm Hg, an HDL cholesterol of 42 mg/dL, and a hemoglobin A_1c of 5.9%.

Using the PCE-based risk calculator, the man’s 10-year risk of cardiovascular disease is only 3.3%, which falls below the ACC/AHA guideline-recommended threshold for statin therapy for primary prevention. But with the MESA score, as a result of the additional information regarding the family history of premature cardiovascular disease, the patient’s risk jumps to 4.9%, even without including a coronary artery calcium score.

“This shows you how big a factor the family history is. I think it’s unfortunate that the PCE doesn’t put it in,” Dr. Vogel said.

Indeed, in an analysis from MESA, a 60-year-old man with a lipid and blood pressure profile similar to that of Dr. Vogel’s patient would have a 10-year cardiovascular disease risk of 6% with a negative family history and a 9% risk with a positive history, he noted.

In the 48-year-old patient used as an example by Dr. Vogel, plugging into the MESA risk calculator a coronary artery calcium score of, say, 50, 100, or 150 Agatston units would boost the 10-year cardiovascular disease risk to 7.5%, 8.9%, and 9.9%, respectively.

Both the MESA and the ACC/AHA risk calculators incorporate diabetes as a simple yes/no item. It’s either present or absent. This is too crude a dichotomy for Dr. Vogel’s liking in light of data showing that individuals with impaired glucose tolerance have a cumulative risk of cardiovascular mortality that’s intermediate between diabetic and normoglycemic individuals. And of course, the patient in his case example had a Hb_A1c of 5.9%. So what is a physician to do?

“I fudge the numbers,” he said. “I move the dial a little bit one way or the other depending on whether a patient falls into that impaired glucose tolerance category.”

He also adjusts a former smoker’s estimated 10-year cardiovascular risk based upon how long ago the smoker quit. An ex-smoker’s hazard ratio for coronary heart disease doesn’t drop to that of a never-smoker until 8 years after quitting. And since the patient in this example has been tobacco-free for only 3 years, Dr. Vogel bumps up that individual’s risk level once again. So at this point, even without the additional information that would be provided by a coronary calcium score, the patient’s adjusted MESA 10-year risk is in the 10% range, compared with the 3.3% figure derived using the PCE-based risk calculator.

He reported serving as a paid consultant to the National Football League and the Pritikin Longevity Center, receiving research grants from Sanofi, and serving on speakers bureaus for Regeneron and Sanofi.

[email protected]

SNOWMASS, COLO. – The oral sodium-glucose cotransporter-2 (SGLT2) inhibitors are the focus of a slew of ongoing phase 3 clinical trials in patients with symptomatic heart failure but no diabetes.

Bruce Jancin/MDedge News

“We have a wide array of exciting opportunities to modify cardiovascular risk with agents that were initially developed for the therapy of diabetes. I think we’re increasingly moving to an age where these agents are actually cardiovascular drugs that happen to lower blood glucose, rather than the other way around, which is how they were initially conceived,” Akshay S. Desai, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

These are each multi-thousand-patient trials, variously due to be completed in 2019-2021. Of note, several of them are restricted to nondiabetic patients with heart failure with preserved ejection fraction (HFpEF), a common, serious, understudied, extremely high-cost disease sorely in need of effective pharmacotherapies, added Dr. Desai, director of the cardiomyopathy and heart failure program at Brigham and Women’s Hospital and a cardiologist at Harvard Medical School, Boston.

All of these placebo-controlled trials have as their composite primary endpoint cardiovascular death and heart failure hospitalization.

EMPEROR-Preserved has randomized 4,126 patients with HFpEF to empagliflozin (Jardiance) or placebo, while EMPEROR-Reduced involves 2,850 patients with heart failure with reduced ejection fraction (HFrEF). Both are due to be completed in 2020.

In addition, the DELIVER trial is focused on 4,700 HFpEF patients randomized to dapagliflozin (Farxiga) or placebo, while Dapa-HF employs the SGLT2 inhibitor in a study of 4,500 patients with HFrEF. Dapa-HF will be completed by late 2019. DELIVER wraps up in mid-2021.

Again, remarkably, none of the participants in these trials has diabetes. All have symptomatic heart failure with elevated N-terminal pro b-type natriuretic peptide levels. The impetus for this ongoing round of studies was the impressive reduction in the risk of hospitalization for heart failure seen in the pivotal trials that earned the SGLT2 inhibitors empagliflozin, canagliflozin (Invokana), and dapagliflozin marketing approval for treatment of type 2 diabetes from the Food and Drug Administration.

Dr. Desai called attention to a new systematic review and meta-analysis of cardiovascular outcomes in randomized, placebo-controlled trials of SGLT2 inhibitors in more than 34,000 patients with type 2 diabetes. The conclusion: These drugs impressively reduced the risk of heart failure hospitalization by 32% in patients with a baseline history of heart failure and similarly by 29% in those with no such history. Also notable was the 45% reduction in the risk of progression of renal disease regardless of whether patients had atherosclerotic cardiovascular disease (Lancet. 2019 Jan 5;393[10166]:31-9).

Only one of the ongoing round of phase 3 trials of SGLT2 inhibitors in heart failure is being conducted in patients with comorbid type 2 diabetes: the 4,000-subject SOLOIST-WHF trial. This study features the investigational dual inhibitor of SGLT1 and 2, sotagliflozin, with a primary outcome of cardiovascular death or heart failure hospitalization. Results are expected in early 2021.


 

What the latest guidelines say

The 2018 American Diabetes Association/European Association for the Study of Diabetes joint consensus statement on management of hyperglycemia in type 2 diabetes reflects an appreciation of the cardiovascular benefits of the SGLT2 inhibitors as well as the injectable glucagon-like peptide-1 receptor (GLP-1) agonists, which have shown significant reductions in major adverse cardiovascular events in pivotal trials including LEADER, HARMONY, and REWIND, albeit without the impressive reduction in heart failure hospitalizations documented with the SGLT2 inhibitors.

The consensus statement emphasizes that aggressive lifestyle modification advice is step No. 1, with the first-line medication being metformin titrated to a target of 1,000 mg twice daily. For patients with clinical heart failure or chronic kidney disease and atherosclerotic cardiovascular heart disease, the next drug recommended is an SGLT2 inhibitor with proven cardiovascular benefit. A GLP-1 agonist is recommended as the first injectable medication, ahead of insulin.
 

Who will take the lead in this new treatment strategy?

Dr. Desai presented data showing that overall utilization of SGLT2 inhibitors and GLP-1 agonists is going up, but not as steeply as it should.

“Cardiologists need to take a more active role,” he declared.

“It’s increasingly clear that, if we’re interested in modifying cardiovascular outcomes, we need to take ownership of this problem, much as we’ve done for lipids and hypertension, because modulating cardiovascular risk is our job,” Dr. Desai asserted. “These drugs may have modest influence on glycemic control, but the primary goal with these agents is to influence cardiovascular outcomes – and if we leave that job to our colleagues, then it often is just a can that gets kicked down the road.”

As a practical matter in prescribing SGLT2 inhibitors and GLP-1 agonists, he emphasized the value of partnering with a primary care physician, endocrinologist, and/or pharmacist by creating pathways for accelerated referral for pharmacologic teaching and, in the case of GLP-1 agonists, injection-related instruction. Pharmacists are often particularly helpful in obtaining prior authorization and financial approval for these medications, and they are familiar with drug discounts and vouchers.

“A great way to jump start collaboration is to provide the patient with a prescription before leaving your office. I think often what we do is just suggest it to the patient, and then a year later they come back and nothing has changed,” the cardiologist said.

Dr. Desai reported serving as a paid consultant to more than half a dozen pharmaceutical or medical device companies.

[email protected]

SNOWMASS, COLO. – The oral sodium-glucose cotransporter-2 (SGLT2) inhibitors are the focus of a slew of ongoing phase 3 clinical trials in patients with symptomatic heart failure but no diabetes.

Bruce Jancin/MDedge News

“We have a wide array of exciting opportunities to modify cardiovascular risk with agents that were initially developed for the therapy of diabetes. I think we’re increasingly moving to an age where these agents are actually cardiovascular drugs that happen to lower blood glucose, rather than the other way around, which is how they were initially conceived,” Akshay S. Desai, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

These are each multi-thousand-patient trials, variously due to be completed in 2019-2021. Of note, several of them are restricted to nondiabetic patients with heart failure with preserved ejection fraction (HFpEF), a common, serious, understudied, extremely high-cost disease sorely in need of effective pharmacotherapies, added Dr. Desai, director of the cardiomyopathy and heart failure program at Brigham and Women’s Hospital and a cardiologist at Harvard Medical School, Boston.

All of these placebo-controlled trials have as their composite primary endpoint cardiovascular death and heart failure hospitalization.

EMPEROR-Preserved has randomized 4,126 patients with HFpEF to empagliflozin (Jardiance) or placebo, while EMPEROR-Reduced involves 2,850 patients with heart failure with reduced ejection fraction (HFrEF). Both are due to be completed in 2020.

In addition, the DELIVER trial is focused on 4,700 HFpEF patients randomized to dapagliflozin (Farxiga) or placebo, while Dapa-HF employs the SGLT2 inhibitor in a study of 4,500 patients with HFrEF. Dapa-HF will be completed by late 2019. DELIVER wraps up in mid-2021.

Again, remarkably, none of the participants in these trials has diabetes. All have symptomatic heart failure with elevated N-terminal pro b-type natriuretic peptide levels. The impetus for this ongoing round of studies was the impressive reduction in the risk of hospitalization for heart failure seen in the pivotal trials that earned the SGLT2 inhibitors empagliflozin, canagliflozin (Invokana), and dapagliflozin marketing approval for treatment of type 2 diabetes from the Food and Drug Administration.

Dr. Desai called attention to a new systematic review and meta-analysis of cardiovascular outcomes in randomized, placebo-controlled trials of SGLT2 inhibitors in more than 34,000 patients with type 2 diabetes. The conclusion: These drugs impressively reduced the risk of heart failure hospitalization by 32% in patients with a baseline history of heart failure and similarly by 29% in those with no such history. Also notable was the 45% reduction in the risk of progression of renal disease regardless of whether patients had atherosclerotic cardiovascular disease (Lancet. 2019 Jan 5;393[10166]:31-9).

Only one of the ongoing round of phase 3 trials of SGLT2 inhibitors in heart failure is being conducted in patients with comorbid type 2 diabetes: the 4,000-subject SOLOIST-WHF trial. This study features the investigational dual inhibitor of SGLT1 and 2, sotagliflozin, with a primary outcome of cardiovascular death or heart failure hospitalization. Results are expected in early 2021.


 

What the latest guidelines say

The 2018 American Diabetes Association/European Association for the Study of Diabetes joint consensus statement on management of hyperglycemia in type 2 diabetes reflects an appreciation of the cardiovascular benefits of the SGLT2 inhibitors as well as the injectable glucagon-like peptide-1 receptor (GLP-1) agonists, which have shown significant reductions in major adverse cardiovascular events in pivotal trials including LEADER, HARMONY, and REWIND, albeit without the impressive reduction in heart failure hospitalizations documented with the SGLT2 inhibitors.

The consensus statement emphasizes that aggressive lifestyle modification advice is step No. 1, with the first-line medication being metformin titrated to a target of 1,000 mg twice daily. For patients with clinical heart failure or chronic kidney disease and atherosclerotic cardiovascular heart disease, the next drug recommended is an SGLT2 inhibitor with proven cardiovascular benefit. A GLP-1 agonist is recommended as the first injectable medication, ahead of insulin.
 

Who will take the lead in this new treatment strategy?

Dr. Desai presented data showing that overall utilization of SGLT2 inhibitors and GLP-1 agonists is going up, but not as steeply as it should.

“Cardiologists need to take a more active role,” he declared.

“It’s increasingly clear that, if we’re interested in modifying cardiovascular outcomes, we need to take ownership of this problem, much as we’ve done for lipids and hypertension, because modulating cardiovascular risk is our job,” Dr. Desai asserted. “These drugs may have modest influence on glycemic control, but the primary goal with these agents is to influence cardiovascular outcomes – and if we leave that job to our colleagues, then it often is just a can that gets kicked down the road.”

As a practical matter in prescribing SGLT2 inhibitors and GLP-1 agonists, he emphasized the value of partnering with a primary care physician, endocrinologist, and/or pharmacist by creating pathways for accelerated referral for pharmacologic teaching and, in the case of GLP-1 agonists, injection-related instruction. Pharmacists are often particularly helpful in obtaining prior authorization and financial approval for these medications, and they are familiar with drug discounts and vouchers.

“A great way to jump start collaboration is to provide the patient with a prescription before leaving your office. I think often what we do is just suggest it to the patient, and then a year later they come back and nothing has changed,” the cardiologist said.

Dr. Desai reported serving as a paid consultant to more than half a dozen pharmaceutical or medical device companies.

[email protected]

SNOWMASS, COLO. – The oral sodium-glucose cotransporter-2 (SGLT2) inhibitors are the focus of a slew of ongoing phase 3 clinical trials in patients with symptomatic heart failure but no diabetes.

Bruce Jancin/MDedge News

“We have a wide array of exciting opportunities to modify cardiovascular risk with agents that were initially developed for the therapy of diabetes. I think we’re increasingly moving to an age where these agents are actually cardiovascular drugs that happen to lower blood glucose, rather than the other way around, which is how they were initially conceived,” Akshay S. Desai, MD, observed at the Annual Cardiovascular Conference at Snowmass sponsored by the American College of Cardiology.

These are each multi-thousand-patient trials, variously due to be completed in 2019-2021. Of note, several of them are restricted to nondiabetic patients with heart failure with preserved ejection fraction (HFpEF), a common, serious, understudied, extremely high-cost disease sorely in need of effective pharmacotherapies, added Dr. Desai, director of the cardiomyopathy and heart failure program at Brigham and Women’s Hospital and a cardiologist at Harvard Medical School, Boston.

All of these placebo-controlled trials have as their composite primary endpoint cardiovascular death and heart failure hospitalization.

EMPEROR-Preserved has randomized 4,126 patients with HFpEF to empagliflozin (Jardiance) or placebo, while EMPEROR-Reduced involves 2,850 patients with heart failure with reduced ejection fraction (HFrEF). Both are due to be completed in 2020.

In addition, the DELIVER trial is focused on 4,700 HFpEF patients randomized to dapagliflozin (Farxiga) or placebo, while Dapa-HF employs the SGLT2 inhibitor in a study of 4,500 patients with HFrEF. Dapa-HF will be completed by late 2019. DELIVER wraps up in mid-2021.

Again, remarkably, none of the participants in these trials has diabetes. All have symptomatic heart failure with elevated N-terminal pro b-type natriuretic peptide levels. The impetus for this ongoing round of studies was the impressive reduction in the risk of hospitalization for heart failure seen in the pivotal trials that earned the SGLT2 inhibitors empagliflozin, canagliflozin (Invokana), and dapagliflozin marketing approval for treatment of type 2 diabetes from the Food and Drug Administration.

Dr. Desai called attention to a new systematic review and meta-analysis of cardiovascular outcomes in randomized, placebo-controlled trials of SGLT2 inhibitors in more than 34,000 patients with type 2 diabetes. The conclusion: These drugs impressively reduced the risk of heart failure hospitalization by 32% in patients with a baseline history of heart failure and similarly by 29% in those with no such history. Also notable was the 45% reduction in the risk of progression of renal disease regardless of whether patients had atherosclerotic cardiovascular disease (Lancet. 2019 Jan 5;393[10166]:31-9).

Only one of the ongoing round of phase 3 trials of SGLT2 inhibitors in heart failure is being conducted in patients with comorbid type 2 diabetes: the 4,000-subject SOLOIST-WHF trial. This study features the investigational dual inhibitor of SGLT1 and 2, sotagliflozin, with a primary outcome of cardiovascular death or heart failure hospitalization. Results are expected in early 2021.


 

What the latest guidelines say

The 2018 American Diabetes Association/European Association for the Study of Diabetes joint consensus statement on management of hyperglycemia in type 2 diabetes reflects an appreciation of the cardiovascular benefits of the SGLT2 inhibitors as well as the injectable glucagon-like peptide-1 receptor (GLP-1) agonists, which have shown significant reductions in major adverse cardiovascular events in pivotal trials including LEADER, HARMONY, and REWIND, albeit without the impressive reduction in heart failure hospitalizations documented with the SGLT2 inhibitors.

The consensus statement emphasizes that aggressive lifestyle modification advice is step No. 1, with the first-line medication being metformin titrated to a target of 1,000 mg twice daily. For patients with clinical heart failure or chronic kidney disease and atherosclerotic cardiovascular heart disease, the next drug recommended is an SGLT2 inhibitor with proven cardiovascular benefit. A GLP-1 agonist is recommended as the first injectable medication, ahead of insulin.
 

Who will take the lead in this new treatment strategy?

Dr. Desai presented data showing that overall utilization of SGLT2 inhibitors and GLP-1 agonists is going up, but not as steeply as it should.

“Cardiologists need to take a more active role,” he declared.

“It’s increasingly clear that, if we’re interested in modifying cardiovascular outcomes, we need to take ownership of this problem, much as we’ve done for lipids and hypertension, because modulating cardiovascular risk is our job,” Dr. Desai asserted. “These drugs may have modest influence on glycemic control, but the primary goal with these agents is to influence cardiovascular outcomes – and if we leave that job to our colleagues, then it often is just a can that gets kicked down the road.”

As a practical matter in prescribing SGLT2 inhibitors and GLP-1 agonists, he emphasized the value of partnering with a primary care physician, endocrinologist, and/or pharmacist by creating pathways for accelerated referral for pharmacologic teaching and, in the case of GLP-1 agonists, injection-related instruction. Pharmacists are often particularly helpful in obtaining prior authorization and financial approval for these medications, and they are familiar with drug discounts and vouchers.

“A great way to jump start collaboration is to provide the patient with a prescription before leaving your office. I think often what we do is just suggest it to the patient, and then a year later they come back and nothing has changed,” the cardiologist said.

Dr. Desai reported serving as a paid consultant to more than half a dozen pharmaceutical or medical device companies.

[email protected]

HONOLULU – In patients who were hyperglycemic following an acute ischemic stroke, intensive insulin control using a continuous insulin drip and an aggressive blood glucose target of 80-130 mg/dL provided no incremental benefit in clinical outcome, compared with a more standard approach of serial, subcutaneous insulin injections and a moderate blood glucose target in a multicenter, U.S. trial with more than 1,100 patients.

Mitchel L. Zoler/MDedge News

The results also highlighted the potential downside to aggressive insulin treatment, with an associated 2.6% incidence of severe hypoglycemia, defined as blood glucose falling below 40 mg/dL, Karen C. Johnston, MD, said at the International Stroke Conference sponsored by the American Heart Association.

“Our data suggest that subcutaneously administered insulin with a target blood glucose level of less than 180 mg/dL is the preferred treatment” because it produces similar efficacy without causing any episodes of severe hypoglycemia, concluded Dr. Johnston, professor and chair of neurology at the University of Virginia in Charlottesville. “There should be no further debate” over the potential superiority of a glucose target substantially below 180 mg/dL, she added in an interview.

Continuing to use a glucose target of less than 180 mg/dL and treating patients with subcutaneous insulin injections every 6 hours to achieve this will mean substantially less resource use and precludes the need for keeping patients in intensive care beds as is needed with an insulin drip, Dr. Johnston noted. A treatment target of less than 180 mg/dL is also consistent with the most recent American Heart Association stroke treatment guidelines, which listed a blood glucose target of 140-180 mg/dL as a class IIa recommendation (Stroke. 2018 March;49[3]:e66-99).

The SHINE (Stroke Hyperglycemia Insulin Network Effort) trial enrolled 1,151 adults diagnosed with an acute ischemic stroke at 63 U.S. centers during 2012-2018, excluding patients with type 1 diabetes. Patients had to enter the study within 12 hours of their last known well time, and with an elevated blood glucose level, above 110 mg/dL in patients with type 2 diabetes or at or above 150 mg/dL in other patients. The median glucose level of enrolled patients was about 188 mg/dL. Enrolled patients averaged 66 years old, and about 80% had type 2 diabetes. The median time from last known well to randomization was just over 7 hours. Almost two-thirds of the patients received thrombolytic treatment, and about 13% underwent thrombectomy.

During up to 72 hours of treatment following enrollment the patients in the standard-treatment arm showed a fairly steady average blood glucose level of 179 mg/dL; patients in the intensive arm showed a steady average of 118 mg/dL.

The study’s primary end point was the percentage of patients with a favorable outcome 90 days after enrollment based on their modified Rankin scale score at that time, with the scores that qualified for this end point varying depending on stroke severity at baseline. The percentage of patients achieving this was 20.5% among the intensive patients and 21.6% among those who received standard insulin treatment, a difference that was not statistically significant.

The findings left open the question of how to better manage acute ischemic stroke patients who present with hyperglycemia.

“Hyperglycemic stroke patients have worse outcomes than stroke patients without hyperglycemia. More aggressively treating the hyperglycemia did not help these patients, We need to figure out what will help them,” Dr. Johnson said.

[email protected]

SOURCE: Johnston KC et al. ISC 2019, Abstract LB1.

HONOLULU – In patients who were hyperglycemic following an acute ischemic stroke, intensive insulin control using a continuous insulin drip and an aggressive blood glucose target of 80-130 mg/dL provided no incremental benefit in clinical outcome, compared with a more standard approach of serial, subcutaneous insulin injections and a moderate blood glucose target in a multicenter, U.S. trial with more than 1,100 patients.

Mitchel L. Zoler/MDedge News

The results also highlighted the potential downside to aggressive insulin treatment, with an associated 2.6% incidence of severe hypoglycemia, defined as blood glucose falling below 40 mg/dL, Karen C. Johnston, MD, said at the International Stroke Conference sponsored by the American Heart Association.

“Our data suggest that subcutaneously administered insulin with a target blood glucose level of less than 180 mg/dL is the preferred treatment” because it produces similar efficacy without causing any episodes of severe hypoglycemia, concluded Dr. Johnston, professor and chair of neurology at the University of Virginia in Charlottesville. “There should be no further debate” over the potential superiority of a glucose target substantially below 180 mg/dL, she added in an interview.

Continuing to use a glucose target of less than 180 mg/dL and treating patients with subcutaneous insulin injections every 6 hours to achieve this will mean substantially less resource use and precludes the need for keeping patients in intensive care beds as is needed with an insulin drip, Dr. Johnston noted. A treatment target of less than 180 mg/dL is also consistent with the most recent American Heart Association stroke treatment guidelines, which listed a blood glucose target of 140-180 mg/dL as a class IIa recommendation (Stroke. 2018 March;49[3]:e66-99).

The SHINE (Stroke Hyperglycemia Insulin Network Effort) trial enrolled 1,151 adults diagnosed with an acute ischemic stroke at 63 U.S. centers during 2012-2018, excluding patients with type 1 diabetes. Patients had to enter the study within 12 hours of their last known well time, and with an elevated blood glucose level, above 110 mg/dL in patients with type 2 diabetes or at or above 150 mg/dL in other patients. The median glucose level of enrolled patients was about 188 mg/dL. Enrolled patients averaged 66 years old, and about 80% had type 2 diabetes. The median time from last known well to randomization was just over 7 hours. Almost two-thirds of the patients received thrombolytic treatment, and about 13% underwent thrombectomy.

During up to 72 hours of treatment following enrollment the patients in the standard-treatment arm showed a fairly steady average blood glucose level of 179 mg/dL; patients in the intensive arm showed a steady average of 118 mg/dL.

The study’s primary end point was the percentage of patients with a favorable outcome 90 days after enrollment based on their modified Rankin scale score at that time, with the scores that qualified for this end point varying depending on stroke severity at baseline. The percentage of patients achieving this was 20.5% among the intensive patients and 21.6% among those who received standard insulin treatment, a difference that was not statistically significant.

The findings left open the question of how to better manage acute ischemic stroke patients who present with hyperglycemia.

“Hyperglycemic stroke patients have worse outcomes than stroke patients without hyperglycemia. More aggressively treating the hyperglycemia did not help these patients, We need to figure out what will help them,” Dr. Johnson said.

[email protected]

SOURCE: Johnston KC et al. ISC 2019, Abstract LB1.

HONOLULU – In patients who were hyperglycemic following an acute ischemic stroke, intensive insulin control using a continuous insulin drip and an aggressive blood glucose target of 80-130 mg/dL provided no incremental benefit in clinical outcome, compared with a more standard approach of serial, subcutaneous insulin injections and a moderate blood glucose target in a multicenter, U.S. trial with more than 1,100 patients.

Mitchel L. Zoler/MDedge News

The results also highlighted the potential downside to aggressive insulin treatment, with an associated 2.6% incidence of severe hypoglycemia, defined as blood glucose falling below 40 mg/dL, Karen C. Johnston, MD, said at the International Stroke Conference sponsored by the American Heart Association.

“Our data suggest that subcutaneously administered insulin with a target blood glucose level of less than 180 mg/dL is the preferred treatment” because it produces similar efficacy without causing any episodes of severe hypoglycemia, concluded Dr. Johnston, professor and chair of neurology at the University of Virginia in Charlottesville. “There should be no further debate” over the potential superiority of a glucose target substantially below 180 mg/dL, she added in an interview.

Continuing to use a glucose target of less than 180 mg/dL and treating patients with subcutaneous insulin injections every 6 hours to achieve this will mean substantially less resource use and precludes the need for keeping patients in intensive care beds as is needed with an insulin drip, Dr. Johnston noted. A treatment target of less than 180 mg/dL is also consistent with the most recent American Heart Association stroke treatment guidelines, which listed a blood glucose target of 140-180 mg/dL as a class IIa recommendation (Stroke. 2018 March;49[3]:e66-99).

The SHINE (Stroke Hyperglycemia Insulin Network Effort) trial enrolled 1,151 adults diagnosed with an acute ischemic stroke at 63 U.S. centers during 2012-2018, excluding patients with type 1 diabetes. Patients had to enter the study within 12 hours of their last known well time, and with an elevated blood glucose level, above 110 mg/dL in patients with type 2 diabetes or at or above 150 mg/dL in other patients. The median glucose level of enrolled patients was about 188 mg/dL. Enrolled patients averaged 66 years old, and about 80% had type 2 diabetes. The median time from last known well to randomization was just over 7 hours. Almost two-thirds of the patients received thrombolytic treatment, and about 13% underwent thrombectomy.

During up to 72 hours of treatment following enrollment the patients in the standard-treatment arm showed a fairly steady average blood glucose level of 179 mg/dL; patients in the intensive arm showed a steady average of 118 mg/dL.

The study’s primary end point was the percentage of patients with a favorable outcome 90 days after enrollment based on their modified Rankin scale score at that time, with the scores that qualified for this end point varying depending on stroke severity at baseline. The percentage of patients achieving this was 20.5% among the intensive patients and 21.6% among those who received standard insulin treatment, a difference that was not statistically significant.

The findings left open the question of how to better manage acute ischemic stroke patients who present with hyperglycemia.

“Hyperglycemic stroke patients have worse outcomes than stroke patients without hyperglycemia. More aggressively treating the hyperglycemia did not help these patients, We need to figure out what will help them,” Dr. Johnson said.

[email protected]

SOURCE: Johnston KC et al. ISC 2019, Abstract LB1.

Around three-quarters of people with type 2 diabetes mellitus (T2DM) who undergo Roux-en-Y gastric bypass experience remission of their disease within a year of the surgery, according to published findings from a population-based observational study. However, one in four of those people will have relapsed by 5 years, the authors noted.

Researchers looked at the effect of Roux-en-Y gastric bypass (RYGB) in 1,111 individuals with T2DM, compared with 1,074 controls who also had T2DM but did not undergo gastric bypass.

By 6 months after surgery, 65% of those who had undergone RYGB met the criteria for remission – defined as no use of glucose-lowering drugs and an HbA_1c below 48 mmol/mol (less than 6.5%) or metformin monotherapy with HbA_1c below 42 mmol/mol (less than 6.0%).

By 1 year, 74% of those who had surgery had achieved remission, and 73% of those remained in remission 5 years after surgery. However, at 2 years, 6% of those who had achieved remission in the first year had already relapsed; by 3 years, 12% had relapsed; and by 4 years, 18% had relapsed. By 5 years after surgery, a total of 27% of those who originally achieved remission in the first year had relapsed.

The overall prevalence of remission remained at 70% for every 6-month period during the duration of the study, which suggests that, although some achieved remission early and then relapsed, others achieved remission later.

Individuals who were aged 50-60 years were 12% less likely to achieve remission, compared with those who were younger than 40 years, whereas those aged 60 years or more were 17% less likely to achieve remission.

A longer duration of diabetes was also associated with a lower likelihood of achieving remission after RYGB; individuals who had had diabetes for 8 years or more had a 27% lower likelihood of remission, compared with those who had had the disease for less than 2 years.

A higher HbA_1c (greater than 53 mmol/mol) was associated with a 19% lower likelihood of remission, and individuals using insulin had a 43% lower likelihood of remission.

“Overall, our findings add evidence to the importance of regular check-ups following RYGB, despite initial diabetes remission, and also suggest that timing of RYGB is important (i.e., consider RYGB while there are still functional pancreatic beta cells),” wrote Lene R. Madsen, MD, from the department of endocrinology and internal medicine at Aarhus (Denmark) University Hospital and her colleagues.

The study also examined the effect of RYGB on microvascular and macrovascular diabetes complications. This revealed that the incidence of diabetic retinopathy was nearly halved among individuals who had undergone gastric bypass, the incidence of hospital-coded diabetic kidney disease was 46% lower, and the incidence of diabetic neuropathy was 16% lower.

In particular, individuals who achieved remission in the first year after surgery had a 57% lower incidence of microvascular events, compared with those who did not have surgery.

The authors noted that individuals who did not reach the threshold for diabetes remission after surgery still showed signs of better glycemic control, compared with individuals who had not undergone surgery.

“This aligns with the theory of ‘metabolic memory’ introduced by Coleman et al. [Diabetes Care. 2016;39(8):1400-07], suggesting that time spent in diabetes remission after RYGB is not spent in vain when it comes to reducing the risk of subsequent microvascular complications,” they wrote.

The surgery was also associated with a 46% reduction in the incidence of ischemic heart disease. In the first 30 days after surgery, 7.5% of patients were readmitted to hospital for any surgical complication, but the 90-day mortality rate after surgery was less than 0.5%.

The study was supported by the Health Research Fund of Central Denmark, the Novo Nordisk Foundation, and the A.P. Møller Foundation. The authors reported no conflicts of interest.

SOURCE: Madsen LR et al. Diabetologia. 2019, Feb 6. doi: 10.1007/s00125-019-4816-2.

Around three-quarters of people with type 2 diabetes mellitus (T2DM) who undergo Roux-en-Y gastric bypass experience remission of their disease within a year of the surgery, according to published findings from a population-based observational study. However, one in four of those people will have relapsed by 5 years, the authors noted.

Researchers looked at the effect of Roux-en-Y gastric bypass (RYGB) in 1,111 individuals with T2DM, compared with 1,074 controls who also had T2DM but did not undergo gastric bypass.

By 6 months after surgery, 65% of those who had undergone RYGB met the criteria for remission – defined as no use of glucose-lowering drugs and an HbA_1c below 48 mmol/mol (less than 6.5%) or metformin monotherapy with HbA_1c below 42 mmol/mol (less than 6.0%).

By 1 year, 74% of those who had surgery had achieved remission, and 73% of those remained in remission 5 years after surgery. However, at 2 years, 6% of those who had achieved remission in the first year had already relapsed; by 3 years, 12% had relapsed; and by 4 years, 18% had relapsed. By 5 years after surgery, a total of 27% of those who originally achieved remission in the first year had relapsed.

The overall prevalence of remission remained at 70% for every 6-month period during the duration of the study, which suggests that, although some achieved remission early and then relapsed, others achieved remission later.

Individuals who were aged 50-60 years were 12% less likely to achieve remission, compared with those who were younger than 40 years, whereas those aged 60 years or more were 17% less likely to achieve remission.

A longer duration of diabetes was also associated with a lower likelihood of achieving remission after RYGB; individuals who had had diabetes for 8 years or more had a 27% lower likelihood of remission, compared with those who had had the disease for less than 2 years.

A higher HbA_1c (greater than 53 mmol/mol) was associated with a 19% lower likelihood of remission, and individuals using insulin had a 43% lower likelihood of remission.

“Overall, our findings add evidence to the importance of regular check-ups following RYGB, despite initial diabetes remission, and also suggest that timing of RYGB is important (i.e., consider RYGB while there are still functional pancreatic beta cells),” wrote Lene R. Madsen, MD, from the department of endocrinology and internal medicine at Aarhus (Denmark) University Hospital and her colleagues.

The study also examined the effect of RYGB on microvascular and macrovascular diabetes complications. This revealed that the incidence of diabetic retinopathy was nearly halved among individuals who had undergone gastric bypass, the incidence of hospital-coded diabetic kidney disease was 46% lower, and the incidence of diabetic neuropathy was 16% lower.

In particular, individuals who achieved remission in the first year after surgery had a 57% lower incidence of microvascular events, compared with those who did not have surgery.

The authors noted that individuals who did not reach the threshold for diabetes remission after surgery still showed signs of better glycemic control, compared with individuals who had not undergone surgery.

“This aligns with the theory of ‘metabolic memory’ introduced by Coleman et al. [Diabetes Care. 2016;39(8):1400-07], suggesting that time spent in diabetes remission after RYGB is not spent in vain when it comes to reducing the risk of subsequent microvascular complications,” they wrote.

The surgery was also associated with a 46% reduction in the incidence of ischemic heart disease. In the first 30 days after surgery, 7.5% of patients were readmitted to hospital for any surgical complication, but the 90-day mortality rate after surgery was less than 0.5%.

The study was supported by the Health Research Fund of Central Denmark, the Novo Nordisk Foundation, and the A.P. Møller Foundation. The authors reported no conflicts of interest.

SOURCE: Madsen LR et al. Diabetologia. 2019, Feb 6. doi: 10.1007/s00125-019-4816-2.

Around three-quarters of people with type 2 diabetes mellitus (T2DM) who undergo Roux-en-Y gastric bypass experience remission of their disease within a year of the surgery, according to published findings from a population-based observational study. However, one in four of those people will have relapsed by 5 years, the authors noted.

Researchers looked at the effect of Roux-en-Y gastric bypass (RYGB) in 1,111 individuals with T2DM, compared with 1,074 controls who also had T2DM but did not undergo gastric bypass.

By 6 months after surgery, 65% of those who had undergone RYGB met the criteria for remission – defined as no use of glucose-lowering drugs and an HbA_1c below 48 mmol/mol (less than 6.5%) or metformin monotherapy with HbA_1c below 42 mmol/mol (less than 6.0%).

By 1 year, 74% of those who had surgery had achieved remission, and 73% of those remained in remission 5 years after surgery. However, at 2 years, 6% of those who had achieved remission in the first year had already relapsed; by 3 years, 12% had relapsed; and by 4 years, 18% had relapsed. By 5 years after surgery, a total of 27% of those who originally achieved remission in the first year had relapsed.

The overall prevalence of remission remained at 70% for every 6-month period during the duration of the study, which suggests that, although some achieved remission early and then relapsed, others achieved remission later.

Individuals who were aged 50-60 years were 12% less likely to achieve remission, compared with those who were younger than 40 years, whereas those aged 60 years or more were 17% less likely to achieve remission.

A longer duration of diabetes was also associated with a lower likelihood of achieving remission after RYGB; individuals who had had diabetes for 8 years or more had a 27% lower likelihood of remission, compared with those who had had the disease for less than 2 years.

A higher HbA_1c (greater than 53 mmol/mol) was associated with a 19% lower likelihood of remission, and individuals using insulin had a 43% lower likelihood of remission.

“Overall, our findings add evidence to the importance of regular check-ups following RYGB, despite initial diabetes remission, and also suggest that timing of RYGB is important (i.e., consider RYGB while there are still functional pancreatic beta cells),” wrote Lene R. Madsen, MD, from the department of endocrinology and internal medicine at Aarhus (Denmark) University Hospital and her colleagues.

The study also examined the effect of RYGB on microvascular and macrovascular diabetes complications. This revealed that the incidence of diabetic retinopathy was nearly halved among individuals who had undergone gastric bypass, the incidence of hospital-coded diabetic kidney disease was 46% lower, and the incidence of diabetic neuropathy was 16% lower.

In particular, individuals who achieved remission in the first year after surgery had a 57% lower incidence of microvascular events, compared with those who did not have surgery.

The authors noted that individuals who did not reach the threshold for diabetes remission after surgery still showed signs of better glycemic control, compared with individuals who had not undergone surgery.

“This aligns with the theory of ‘metabolic memory’ introduced by Coleman et al. [Diabetes Care. 2016;39(8):1400-07], suggesting that time spent in diabetes remission after RYGB is not spent in vain when it comes to reducing the risk of subsequent microvascular complications,” they wrote.

The surgery was also associated with a 46% reduction in the incidence of ischemic heart disease. In the first 30 days after surgery, 7.5% of patients were readmitted to hospital for any surgical complication, but the 90-day mortality rate after surgery was less than 0.5%.

The study was supported by the Health Research Fund of Central Denmark, the Novo Nordisk Foundation, and the A.P. Møller Foundation. The authors reported no conflicts of interest.

SOURCE: Madsen LR et al. Diabetologia. 2019, Feb 6. doi: 10.1007/s00125-019-4816-2.

From Temple University School of Medicine, Philadelphia, PA.

Abstract

• Objective: To outline the use and utility of gastric electric stimulation (GES) as a therapeutic intervention for gastroparesis.
• Methods: Review of the literature.
• Results: Gastroparesis is characterized by delayed gastric emptying, with symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Some patients with gastroparesis do not respond to medical intervention, and for these patients surgical intervention may be warranted. GES utilizes high-frequency gastric neurostimulation to facilitate gastric emptying and reduce symptoms of gastroparesis. It is indicated for patients with idiopathic and diabetic gastroparesis who have nausea and vomiting as their primary symptoms and who have not responded to medical therapy. GES has also been used in postsurgical and pediatric gastroparesis patients. Optimizing the outcome of this surgical treatment through proper patient selection and meticulous surgical technique is essential as there are inherent risks to the procedure. Nonblinded studies of GES for medically refractory gastroparesis have demonstrated therapeutic symptomatic benefit, whereas randomized controlled trials have not. New interventions such as pyloromyotomy and pyloroplasty are reasonable alternatives or addendums to GES.
• Conclusion: GES may be considered among the therapies available for treating patients with refractory symptoms of gastroparesis. More studies, specifically those comparing GES, pyloromyotomy, GES combined with pyloromyotomy, and placebo, are needed to help guide therapy selection for refractory gastroparesis.

Keywords: diabetes; gastroparesis; dysmotility; gastric emptying; electric stimulation.

Gastroparesis is a chronic dysmotility disorder characterized by delayed gastric emptying with associated symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Medical treatments for gastroparesis include dietary modifications, glucose control in those with diabetes, prokinetic medications, antiemetic medications, and symptom modulators, but unfortunately patients frequently do not respond to these treatments. In patients refractory to medical therapy, surgical treatments can be considered.

Gastric electric stimulation (GES; Enterra [Medtronic, Minneapolis, MN]) was approved via a Food and Drug Administration (FDA) Humanitarian Use Device (HUD) exemption for the treatment of medically refractory gastroparesis in 2000. Understanding the indications, risks, outcomes, and alternatives to GES is essential to providing appropriate care for patients with medically refractory gastroparesis. This article outlines the use and utility of GES as a therapeutic intervention for gastroparesis.

Types of Gastroparesis

Gastroparesis is a chronic symptomatic disorder of the stomach manifested by delayed gastric emptying without evidence of gastric outlet obstruction or ulceration.¹ The pathophysiology of gastroparesis appears to involve abnormalities in functioning of several elements including the autonomic nervous system, especially the vagus nerve, smooth muscle cells, enteric neurons, and interstitial cells of Cajal.

Idiopathic gastroparesis and diabetic gastroparesis are the 2 most common types of gastroparesis.² Symptomatic delayed gastric emptying with no primary underlying abnormality predisposing to gastroparesis is categorized as idiopathic gastroparesis.³ A small subset of patients with idiopathic gastroparesis report an initial infectious prodrome such as gastroenteritis or respiratory infection. It has been suggested that this postinfectious gastroparesis results from viral injury to the neural innervation of the stomach or the interstitial cells of Cajal in the stomach.⁴ Viruses that have been implicated in the development of gastroparesis include cytomegalovirus, Epstein-Barr virus, Norwalk virus, rotavirus, herpes zoster, and varicella zoster.^5-9

Diabetic gastroparesis is characterized as onset of symptoms of gastroparesis in patients with diabetes, with concomitant delayed gastric emptying. It is often attributed to chronic hyperglycemia-induced damage to the vagus nerve, and is frequently observed in association with other diabetic complications such as neuropathy, retinopathy, and nephropathy.¹⁰

Gastroparesis that develops following surgery is classified as postsurgical gastroparesis. In the past, this form of gastroparesis most commonly occurred after ulcer surgery, often performed with vagotomy. These types of surgeries are performed less frequently in the era of proton pump inhibitor therapy and treatments for Helicobacter pylori. Presently, Nissen fundoplication and bariatric surgery are the more common surgical procedures associated with gastroparesis.³ Long-term use of medications that delay gastric emptying, such as opiate narcotic medications, can lead to gastroparesis and represent another form of iatrogenic gastroparesis. Other forms of gastroparesis (atypical gastroparesis) arise due to various underlying etiologies, including neurological disorders (eg, Parkinson disease, multiple sclerosis), metabolic or endocrine conditions (eg, hypothyroidism), autoimmune disorders, connective tissue and collagen vascular disorders (eg, systemic lupus erythematosus, scleroderma, Sjögren syndrome, Ehlers-Danlos syndrome), or eating disorders (eg, anorexia, bulimia).³

Epidemiology

There is a female preponderance in patients with gastroparesis. Data from the Rochester Epidemiology Project, a database of linked medical records for residents of Olmsted County, MN, showed that the age-adjusted prevalence of definite gastroparesis per 100,000 inhabitants was 37.8 for women and 9.6 for men.¹¹ More recent estimates have suggested a much higher prevalence of probable gastroparesis (approximately 1.8%) in the general population using symptoms suggestive of gastroparesis.¹² Hospitalization rates for gastroparesis have increased since 2000, which could reflect rising prevalence and/or the effects of heightened awareness about and better identification of gastroparesis.¹³ This increase may also be due in part to the rising rate of diabetes leading to more cases of diabetic gastroparesis; withdrawal of some gastroparesis treatments from the market (cisapride, tegaserod) leading to hospitalizations for symptoms not adequately being treated; and hospitalizations needed for insertion of the gastric electric stimulator.

Gastroparesis Symptoms

The main symptoms of gastroparesis are early satiety, postprandial fullness, bloating, nausea, and vomiting.¹⁴ Nausea (> 90% of patients) and early satiety (60% of patients) are the most common symptoms.¹⁵ Abdominal pain is often present in patients with gastroparesis but is usually not the predominant symptom. The pain can be multifactorial, with somatic, visceral, and neuropathic components.^16-18 Moderate to severe abdominal pain has been found more often in patients with idiopathic gastroparesis and in association with opiate use.¹⁶ Symptoms of gastroparesis may be persistent or present as episodic flares. Due to the symptoms, some patients will experience weight loss and malnutrition and, in severe cases, dehydration.¹⁹

Although the definition of gastroparesis is a delay in gastric emptying along with symptoms, symptoms correlate poorly with the degree of delayed gastric emptying. The symptoms that appear to have the strongest correlation with gastric emptying are nausea, vomiting, early satiety, and postprandial fullness, whereas symptoms such as abdominal pain and bloating have little correlation. Furthermore, improving gastric emptying does not necessarily lead to improved symptoms, and symptom improvement does not always lead to improved gastric emptying times.²⁰ Between 5% and 12% of patients with diabetes report symptoms consistent with gastroparesis, though many of these patients have normal gastric emptying. The symptoms of gastroparesis overlap with those of functional dyspepsia, as both may have motor and sensory alterations.²¹

The Gastroparesis Cardinal Symptom Index (GCSI), a subset of the Patient Assessment of Gastrointestinal Disorders Symptom Severity Index (PAGI-SYM), is a questionnaire that is commonly used to establish symptom severity in patients with gastroparesis. It is comprised of 3 subscales—nausea and vomiting, postprandial fullness and early satiety, and bloating—which are averaged to provide a total GCSI score. Symptoms over the 2 weeks prior to administration of the questionnaire are assessed and rated from 0 (none) to 5 (very severe).²² Grading the severity of gastroparesis may take into account symptoms, quality of life, and gastric emptying. One commonly used grading system assigns a grade from 1 to 3, with grade 1 being mild gastroparesis, grade 2 being compensated gastroparesis, and grade 3 being gastric failure with refractory symptoms that are uncontrolled.^18,23 Quality-of-life surveys also suggest that gastroparesis independent of other factors leads to a worse quality of life.²⁴

Indications for GES

Gastric electric stimulator implantation is a surgical procedure with inherent risks and complications and is reserved for patients with intractable symptoms of gastroparesis who remain symptomatic despite treatment attempts with dietary management, antiemetic agents (eg, compazine, phenergan, and ondansetron), and prokinetic agents (eg, metoclopramide, erythromycin, and domperidone). Symptom modulators such as nortriptyline and mirtazapine are occasionally tried.

Surgical intervention can be considered upon failure of medical treatment measures. At least a year of documented care provided by a physician specializing in gastroparesis is suggested for surgical consideration. The gastric electric neurostimulator is approved by the FDA as a HUD for the care of patients with idiopathic and diabetic gastroparesis, performed on a compassionate basis. GES implantation requires Institutional Review Board approval at the institution, and patients are required to have documented delayed gastric emptying.

It is important to remember that the GES device is incompatible with magnetic resonance imaging (MRI) and explantation of the device is necessary prior to MRI. As such, in patients with anticipated need of frequent MRI, such as those with multiple sclerosis, serious consideration should be given to alternative strategies prior to focusing on this modality.

Device Placement

GES was devised to improve gastric emptying. The Enterra GES system uses high-frequency, low-energy electric stimulation. An alternative method is true gastric pacing that uses high-energy, low-frequency stimulation to entrain the gastric slow waves and subsequent contractions at 3 cycles per minute (cpm). Gastric pacing has greater energy requirements than GES, which makes the size of the stimulator too large to be practical. In pilot animal studies, GES produced an accelerating effect on gastric emptying, but in human studies GES had an inconsistent effect on gastric emptying. Studies have suggested that GES influences the proximal stomach, with a reduction of gastric tone,²⁵ and also that GES has an afferent modulatory mechanism.²⁶

The Enterra GES is placed surgically under general anesthesia, commonly via laparotomy or minimal access surgical techniques (laparoscopically or robotically assisted). Preoperative intravenous antibiotics are given. The system consists of a pair of electrodes connected to a pulse generator. The 2 stimulation leads are inserted into the gastric muscularis propria 1 cm apart along the greater curvature 10 cm proximal to the pylorus. Upper endoscopy is performed to ensure that the leads do not penetrate through the mucosa into the stomach lumen; if this occurs, repositioning of the lead is necessary. A horizontal incision through the skin is made, and the distal ends of the stimulating wires are tunneled through the abdominal wall and connected to the pulse generator. The impedance (resistance) between the wires is measured to ensure the appropriate range (200-800 Ohms). The neurostimulator with the distal ends of the stimulating wires is then placed into the subcutaneous pocket and sutured to the underlying fascia. The pulse generator delivers a high-frequency, low-energy, 0.1-second train of pulses at a frequency of 12 cpm. Within each pulse train, individual pulses oscillate at a frequency of 14 cycles per second. The voltage of the stimulations is set to provide a current of 5 milliamps (mA; remembering that voltage = current × resistance).

Patients are often hospitalized with a recovery time of 1 to 3 days. Immediate postoperative care usually includes intravenous fluids, controlling any postoperative ileus, advancing diet, and providing analgesic pain medications. Hospital length of stay can be impacted by surgical technique.²⁵ Patients are seen several weeks after discharge for assessment of the incision and toleration of diet. Medications for gastroparesis that patients were taking prior to the GES implantation are usually continued postoperatively, with a goal of reducing these medications over time. Patients are then followed every 3 to 12 months, depending on their clinical condition.

At follow-up visits, medications are reviewed and new treatments can be added if appropriate. The gastric stimulator is interrogated to determine if changes in resistance occurred; if necessary, minor readjustments can be made to keep the current at desired levels (5 mA). For persistent symptoms with GES treatment, the stimulator parameters can be adjusted after 3 months of follow up, typically first increasing the current from 5 to 7.5 mA and then to 10 mA. After this, the frequency can be increased from 14 Hz to 28 Hz, and then to 55 Hz. Rarely, the ON duration is increased from 0.1 to 1 second. Increasing the ON time can worsen symptoms in some patients, cause abdominal pain, and decrease the battery life from the usual 7 years.

Complications of GES

In an analysis of the Manufacturer and User Facility Device Experience (MAUDE) databank, Bielefeldt identified 1587 reports of adverse effects related to the gastric electric stimulator from January 2001 to October 2015.²⁷ The most common adverse effects are reviewed here.

Skin erosion/wound dehiscence is one of the most common reported complications; it may be related to superficial placement or inadequate securing of the device to the fascia. Abscess can develop postoperatively due to hematogenous seeding or may be a sign of lead erosion into the lumen, tracking along the leads into subcutaneous tissue.²⁸ It is important to warn patients to protect the area over the device from needle injections as this also can lead to hematoma formation and direct contamination of the device. If the device gets infected, it cannot be salvaged and requires explantation. Implantation of a new device can be attempted once all wound issues resolve.

Device migration/flipping most often occurs because the device is inadequately fixed to the underlying fascia, but occasionally it can occur from patients flipping the device around. Flipping can occur due to superficial pocket location within subcutaneous tissue, especially in obese patients. Migration/flipping can lead to prominence of the contour of the device and discomfort, ultimately requiring surgical correction.

Perforation and erosion of the leads. With time, leads can erode into the stomach, although this is rare. Usually erosion is associated with loss of device function. Endoscopy confirms this finding. In rare cases, infection can track proximally along the lead and present as a surgical site infection at the pulse generator. This complication often requires explantation of the neurostimulator leads and pulse generator.

Intestinal obstruction. Although rare, the intestines can get wrapped around the leads of the device, causing different degrees of obstruction (Figure 1). Positioning the device in the left upper quadrant minimizes the intraabdominal length of the leads and pulls them maximally out, coiling under the device (Figure 2). In cases where other locations are used either due to a hostile upper abdominal region (skin infection, presence of gastrostomy or other devices) or surgeon’s preference, the GES device can be implanted in the lower abdomen (Figure 3). In these circumstances, carefully draping the omentum over the bowels might help to prevent this complication. Tacking of the leads to the parietal peritoneum with sutures can also be preventative. In cases of obstruction requiring intervention by laparotomy or minimal access techniques (laparoscopy or robotic assisted surgery), all efforts are made to preserve the neurostimulator leads. In cases that require bowel resection, lead contamination is a serious concern, but lead explantation is not mandatory. Close postoperative monitoring for the development of lead infection is required.

Hematoma and seroma. Postoperative hematomas can occur from inadequate hemostasis, and seromas can occur in the stimulator pocket. Small hematomas may be observed if not complicated (Figure 4). In cases of large hematomas with skin compromise or dehiscence, prompt washout and drainage is required. In ideal cases, the device can be preserved. Relocation to another site might be required if skin necrosis develops. The possibility of device contamination also must be considered; after resolution of wound issues, implantation of a new device may be tried. Seromas at the generator pocket site are a frequent occurrence but are often benign, self-limiting, and generally resolve over 4 to 6 weeks.

Incisional hernia. Hernias can develop after any abdominal surgery and are not unique to GES implantation. Use of minimally invasive technique for the GES implantation minimizes this complication.

Electric shock sensations may occur from breakage of the plastic lining covering the stimulator wires or from fluid buildup around the insertion of the wires into the stimulator. Shocks can also occur due to shortening of the leads on the muscles of the abdominal wall. Patients describe periodic muscle cramps with the frequency of the device (every 5 seconds). To prevent this complication, freshly implanted leads should be covered by an omental flap to isolate them from the abdominal wall. In patients who continue to feel shocks despite all efforts, the possibility of visceral hypersensitivity should be considered. A trial of symptom modulators such as nortriptyline and lowering of the output amperage below the minimal recommended setting of 5 mA can be undertaken. If these interventions do not work, the device must be turned off for a period of time. Occasionally, replacement of the leads or explantation of the device must be considered.

Lack of effect/persistent symptoms. If a patient presents with lack of improvement after device implantation, a thorough workup should be undertaken to ensure that the device is functioning properly. In the case of abnormal impedance values, an abdominal x-ray study can be performed to rule out lead migration (Figure 5). If no abnormalities are detected, the output of the device can be increased. After adjusting device settings, the patient should be assessed for improvement over at least a 1- to 3-month period. One report suggests that in patients not responding to GES, repositioning the location of the stimulator leads on the stomach can be helpful.²⁹

Outcomes of GES

Study results of investigative GES models in animals and select patients were published in 1997.^30,31 Following these reports, 2 large multicenter studies were conducted to demonstrate the efficacy of GES for the treatment of refractory gastroparesis. The Gastric Electrical Mechanical Stimulation Study (GEMS) was an open-label, multicenter study of 38 patients who received percutaneous and later permanent GES devices.³² Marked reduction in weekly vomiting and nausea was observed at 4 weeks, with a 90% reduction in nausea and vomiting frequency at 11 months. Following this, a second multicenter study (Worldwide Anti-Vomiting Electrical Stimulation Study [WAVES]) involving a double-blind sham stimulation controlled trial with 33 idiopathic and diabetic gastroparesis patients was performed.³³ During the blinded portion of this study, there was a noticeable decrease in vomiting frequency, particularly in the patients with diabetic gastroparesis. Patient preference was for the stimulator ON as compared to OFF. The FDA’s HUD exemption for the Enterra GES device in 2000 was based on these studies.

Four independent double-blind studies of GES have been conducted (Table 1).^33-37 It has been difficult to demonstrate improvement during the double-blind period with gastric stimulation compared to no stimulation. Despite total symptom severity improvement and individual symptom improvements in these studies, a recent meta-analysis demonstrated a summative insignificant difference between the GES ON versus OFF states.³⁸

In contrast to the double-blind studies, numerous open-label studies have demonstrated clinical improvements in patients with diabetic and idiopathic gastroparesis (Table 2),^{32,33,35,36,39-55} leading some to question whether the demonstrable efficacy reflects a placebo effect or regression to the mean. Patients may perceive an operative, aggressive intervention as likely to be effectual in comparison to incremental medication efforts, thus creating a placebo effect. It should also be noted that not all open-label studies have demonstrated improvement with GES. Indeed, Jones et al reported no significant difference in nausea and vomiting at 6-month follow-up, and recommended that physicians exercise caution with GES as a therapeutic strategy given the cost and lack of confirmed demonstrable effect.⁵⁶ Thus, the clinical successes demonstrated in open-label studies must be weighed not only against the lack of unequivocal improvement, but also against the potential deleterious effects of the surgery.

In an open-label study that employed the GCSI to follow symptoms of gastroparesis, 29 patients underwent GES implantation over an 18-month period, with follow-up in 28 patients.⁴⁴ GES resulted in clinical improvement in 50% of patients with refractory gastroparesis. The overall GCSI significantly decreased, with improvement in the nausea/vomiting subscore and the post-prandial fullness subscore, but no improvement in the bloating subscore or abdominal pain. The decrease in GCSI was greater for patients with diabetic versus idiopathic gastroparesis. Patients with the main symptom of nausea/vomiting had a greater improvement than patients with the main symptom of abdominal pain. Patients taking narcotic analgesics at the time of implant had a poorer response compared to patients who were not. In this study, 3 clinical parameters were associated with a favorable clinical response: (1) diabetic rather than idiopathic gastroparesis, (2) nausea/vomiting rather than abdominal pain as the primary symptom, and (3) independence from narcotic analgesics prior to stimulator implantation. Knowledge of these 3 factors may allow improved patient selection for GES.

A large prospective study by Heckert et al detailed marked improvements with GES and the patterns of those improvements.⁵⁵ Nausea, vomiting, loss of appetite, and early satiety improved significantly with stimulator use, with a greater improvement in vomiting in patients with diabetic gastroparesis than in those with the idiopathic form. Although GES improved symptoms in 75% of all patients, patients with diabetes had a post-GES Clinical Patient Grading Assessment score that was statistically higher than the score among patients with idiopathic gastroparesis. This difference is thought to be due to the neuromolecular mechanism of diabetic gastroparesis, where blunting of the enteric nervous system may contribute to symptomatology.

Several studies have demonstrated a clinical response to GES in patients with postsurgical gastroparesis. A study by Oubre et al showed that GES led to weekly vomiting improvements as well as a reduction in total symptom severity score.⁵⁷ A study by McCallum et al further demonstrated improved symptoms, quality of life, nutritional status, and hospitalization requirements.⁵⁸ GES has also been shown to improve gastroparesis symptoms in pediatric populations.^47,59 Thus, although not a direct indication, GES has been shown to be beneficial in various subtypes of gastroparesis.

Additionally, irrespective of gastroparesis type, the improved symptomatology with GES appears to be durable, with one study showing persistent clinical improvements up to 8 years after device placement.⁶⁰ The improvements were persistent and incremental. Likewise, McCallum et al showed that continued reductions in total symptom severity scores were evident in all gastroparesis types up to 10 years after stimulator implantation.⁶¹ The success of the procedures in part comes from careful selection of patients. Clinical parameters that are associated with favorable clinical response include diabetic gastroparesis subtype, nausea/vomiting predominance, and independence from narcotic analgesics prior to stimulator placement.⁶²

GES has also been noted to improve other patient care metrics besides symptomatology, including nutritional status, reduced need for nutritional supplementation, and improved HbA1c.^63-65 Additionally, a study by Cutts et al established that health care resource utilization significantly improved at 12, 24, and 36 months following GES placement, as compared to patients receiving standard medical therapy.⁶⁶ This decreased resource utilization was also reflected in decreased costs in the GES group compared with the standard care group.

Surgical Alternatives to GES

Pyloric interventions such as pyloroplasty and pyloromyotomy are other surgical treatment modalities offered for gastroparesis. Whereas GES uses neurostimulation to facilitate gastric emptying and potentially improve fundic accommodation, pyloric interventions are intended to increase gastric emptying by reducing outflow resistance from the pyloric sphincter.

Pyloric Interventions

Various studies have shown significant improvements with pyloric interventions, similar to the improvements seen with GES. One such study involving 177 patients demonstrated an 86% improvement in gastric emptying, with symptom severity scores for nausea, vomiting, bloating, abdominal pain, and early satiety decreasing significantly at 3 months following pyloroplasty.⁶⁷ A significant advantage of pyloric interventions is that pyloromyotomy can be performed endoscopically (gastric peroral endoscopic pyloromyotomy [G-POEM] or peroral pyloromyotomy [POP]), thus minimizing the risks of open surgery. A recent review that included a pooled analysis of 7 studies of G-POEM for gastroparesis demonstrated 100% technical success, with clinical efficacy in 81.5% of the procedures as assessed by the GCSI.⁶⁸ Additionally, the intraoperative and perioperative complication rates were 6.6% and 7.6%, respectively, suggesting that G-POEM is a safe and clinically beneficial therapeutic option. Few studies comparing the outcomes of pyloric interventions to GES have been performed.

Recently, GES has been combined with pyloric interventions to maximize therapeutic potential. This allows simultaneous neurologic and functional interventions to expedite gastric emptying and improve patient symptomatology. Davis et al demonstrated significant improvement in 21 patients who underwent GES placement and pyloroplasty, with 71% improvement in total symptom severity.⁶⁹ Notably, dual surgery did not increase the incidence of infection or adverse surgical outcomes. Although this study did not directly compare dual surgery to GES alone, the results are nonetheless favorable. GES provides a strong antiemetic and anti-nausea effect, whereas the pyloromyotomy provides improvement in gastric emptying.

Feeding/Venting Tubes

Feeding jejunostomy tubes and venting gastrostomy tubes can be used alone or in combination with GES. Feeding jejunostomy is performed for malnutrition and weight loss that accompanies the refractory symptoms of early satiety, nausea, and vomiting. Venting gastrostomy tubes allow for removal of retained gastric contents that may cause distension, nausea, and vomiting. Gastrojejunostomy tubes can also be placed endoscopically or by interventional radiology.

Gastrectomy

Gastrectomy can provide therapeutic benefit through elimination of the gastric reservoir function and consequent removal of afferent neural impulses. In select patient populations, outcomes of gastrectomy have compared favorably with those of GES. For example, one study demonstrated favorable outcomes of Roux-en-Y gastrectomy in morbidly obese patients with gastroparesis.⁷⁰ In another study, favorable outcomes were reported in a cohort of 103 patients, with gastrectomy demonstrating 87% symptom improvement (nausea, vomiting, epigastric pain) compared to just 63% improvement with GES.⁷¹ However, the dramatic impact on anatomy and physiology and the invasiveness of the procedure need to be weighed against the therapeutic benefit. For example, in the same study, the 30-day morbidity was 23% for gastrectomy versus just 8% for the GES implant.⁷¹

 

 

When to Use GES

The gastric electrical neurostimulator (Enterra; Medtronic, Inc.) is approved for treatment of idiopathic and diabetic gastroparesis that is refractory to medical treatment, performed on a compassionate basis. Patients with diabetic gastroparesis respond to GES better than do patients with the idiopathic form. Of the symptoms of gastroparesis, primarily nausea and vomiting improve. Thus, GES favors patients with diabetic gastroparesis who have primarily nausea and vomiting, rather than, for instance, patients with idiopathic gastroparesis who have primarily abdominal pain and may be taking narcotics. Some centers provide GES for postsurgical patients and children with gastroparesis.

The 3 main surgical interventions for medically refractory gastroparesis are GES, pyloric intervention (pyloroplasty or pyloromyotomy), and gastrectomy. Of the 3 interventions, gastrectomy is the most radical given its dramatic effect on anatomy and is thus not preferred. The clinical decision then becomes: GES, pyloric intervention, or both? There are limited data to support a definitive answer to this question.

In a single-center retrospective analysis of prospective data (electronic medical record), Arthur et al compared outcomes of GES patients with medically refractory gastroparesis who received various surgical interventions.⁷² In total, 33 stimulator, 7 pyloroplasty, 2 gastrectomy, and 16 combined stimulator and pyloroplasty patients were analyzed for postoperative symptom improvement. Pyloroplasty alone demonstrated the least symptom improvement, combination GES and pyloroplasty demonstrated increased improvement, and GES alone demonstrated the most improvement. The results of this study suggest that barring contraindication, placement of a gastric stimulator as the initial treatment is best, with pyloroplasty reserved for patients who do not achieve adequate symptom control. Limitations of the study include its single-center design and low patient numbers for pyloroplasty in isolation.

In contrast, a recent retrospective systematic review synthesized the outcomes of various studies of GES and pyloric interventions for medically refractory gastroparesis.⁷³ A therapeutic effect was found for each surgical intervention, with pyloric surgery patients demonstrating a greater response to intervention than GES patients. Unfortunately, attempts to analyze combination interventions were hindered by a lack of power.

Conclusion

Initial management of gastroparesis is medical (lifestyle and diet changes), with antiemetic and prokinetic agents used in refractory cases. Following failure of this therapy, placement of a GES device is a surgical intervention that has been approved under FDA humanitarian device exemption to help ameliorate symptomatology. Improvement with GES has been demonstrated in nonblinded studies, but the lack of randomized controlled trials demonstrating benefit suggests the possibility of an underlying placebo effect. Additionally, new medical procedures such as G-POEM complicate the decision of which intervention should be attempted first. More studies, specifically comparing GES, pyloric interventions, and combined GES with pyloric intervention to placebo, are needed to fully understand what therapy is best for refractory gastroparesis.

Corresponding author: Henry P. Parkman, MD, Gastroenterology Section, Temple University School of Medicine, 3401 North Broad Street, Philadelphia, PA 19140; [email protected].

Financial disclosures: None.

From Temple University School of Medicine, Philadelphia, PA.

Abstract

• Objective: To outline the use and utility of gastric electric stimulation (GES) as a therapeutic intervention for gastroparesis.
• Methods: Review of the literature.
• Results: Gastroparesis is characterized by delayed gastric emptying, with symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Some patients with gastroparesis do not respond to medical intervention, and for these patients surgical intervention may be warranted. GES utilizes high-frequency gastric neurostimulation to facilitate gastric emptying and reduce symptoms of gastroparesis. It is indicated for patients with idiopathic and diabetic gastroparesis who have nausea and vomiting as their primary symptoms and who have not responded to medical therapy. GES has also been used in postsurgical and pediatric gastroparesis patients. Optimizing the outcome of this surgical treatment through proper patient selection and meticulous surgical technique is essential as there are inherent risks to the procedure. Nonblinded studies of GES for medically refractory gastroparesis have demonstrated therapeutic symptomatic benefit, whereas randomized controlled trials have not. New interventions such as pyloromyotomy and pyloroplasty are reasonable alternatives or addendums to GES.
• Conclusion: GES may be considered among the therapies available for treating patients with refractory symptoms of gastroparesis. More studies, specifically those comparing GES, pyloromyotomy, GES combined with pyloromyotomy, and placebo, are needed to help guide therapy selection for refractory gastroparesis.

Keywords: diabetes; gastroparesis; dysmotility; gastric emptying; electric stimulation.

Gastroparesis is a chronic dysmotility disorder characterized by delayed gastric emptying with associated symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Medical treatments for gastroparesis include dietary modifications, glucose control in those with diabetes, prokinetic medications, antiemetic medications, and symptom modulators, but unfortunately patients frequently do not respond to these treatments. In patients refractory to medical therapy, surgical treatments can be considered.

Gastric electric stimulation (GES; Enterra [Medtronic, Minneapolis, MN]) was approved via a Food and Drug Administration (FDA) Humanitarian Use Device (HUD) exemption for the treatment of medically refractory gastroparesis in 2000. Understanding the indications, risks, outcomes, and alternatives to GES is essential to providing appropriate care for patients with medically refractory gastroparesis. This article outlines the use and utility of GES as a therapeutic intervention for gastroparesis.

Types of Gastroparesis

Gastroparesis is a chronic symptomatic disorder of the stomach manifested by delayed gastric emptying without evidence of gastric outlet obstruction or ulceration.¹ The pathophysiology of gastroparesis appears to involve abnormalities in functioning of several elements including the autonomic nervous system, especially the vagus nerve, smooth muscle cells, enteric neurons, and interstitial cells of Cajal.

Idiopathic gastroparesis and diabetic gastroparesis are the 2 most common types of gastroparesis.² Symptomatic delayed gastric emptying with no primary underlying abnormality predisposing to gastroparesis is categorized as idiopathic gastroparesis.³ A small subset of patients with idiopathic gastroparesis report an initial infectious prodrome such as gastroenteritis or respiratory infection. It has been suggested that this postinfectious gastroparesis results from viral injury to the neural innervation of the stomach or the interstitial cells of Cajal in the stomach.⁴ Viruses that have been implicated in the development of gastroparesis include cytomegalovirus, Epstein-Barr virus, Norwalk virus, rotavirus, herpes zoster, and varicella zoster.^5-9

Diabetic gastroparesis is characterized as onset of symptoms of gastroparesis in patients with diabetes, with concomitant delayed gastric emptying. It is often attributed to chronic hyperglycemia-induced damage to the vagus nerve, and is frequently observed in association with other diabetic complications such as neuropathy, retinopathy, and nephropathy.¹⁰

Gastroparesis that develops following surgery is classified as postsurgical gastroparesis. In the past, this form of gastroparesis most commonly occurred after ulcer surgery, often performed with vagotomy. These types of surgeries are performed less frequently in the era of proton pump inhibitor therapy and treatments for Helicobacter pylori. Presently, Nissen fundoplication and bariatric surgery are the more common surgical procedures associated with gastroparesis.³ Long-term use of medications that delay gastric emptying, such as opiate narcotic medications, can lead to gastroparesis and represent another form of iatrogenic gastroparesis. Other forms of gastroparesis (atypical gastroparesis) arise due to various underlying etiologies, including neurological disorders (eg, Parkinson disease, multiple sclerosis), metabolic or endocrine conditions (eg, hypothyroidism), autoimmune disorders, connective tissue and collagen vascular disorders (eg, systemic lupus erythematosus, scleroderma, Sjögren syndrome, Ehlers-Danlos syndrome), or eating disorders (eg, anorexia, bulimia).³

Epidemiology

There is a female preponderance in patients with gastroparesis. Data from the Rochester Epidemiology Project, a database of linked medical records for residents of Olmsted County, MN, showed that the age-adjusted prevalence of definite gastroparesis per 100,000 inhabitants was 37.8 for women and 9.6 for men.¹¹ More recent estimates have suggested a much higher prevalence of probable gastroparesis (approximately 1.8%) in the general population using symptoms suggestive of gastroparesis.¹² Hospitalization rates for gastroparesis have increased since 2000, which could reflect rising prevalence and/or the effects of heightened awareness about and better identification of gastroparesis.¹³ This increase may also be due in part to the rising rate of diabetes leading to more cases of diabetic gastroparesis; withdrawal of some gastroparesis treatments from the market (cisapride, tegaserod) leading to hospitalizations for symptoms not adequately being treated; and hospitalizations needed for insertion of the gastric electric stimulator.

Gastroparesis Symptoms

The main symptoms of gastroparesis are early satiety, postprandial fullness, bloating, nausea, and vomiting.¹⁴ Nausea (> 90% of patients) and early satiety (60% of patients) are the most common symptoms.¹⁵ Abdominal pain is often present in patients with gastroparesis but is usually not the predominant symptom. The pain can be multifactorial, with somatic, visceral, and neuropathic components.^16-18 Moderate to severe abdominal pain has been found more often in patients with idiopathic gastroparesis and in association with opiate use.¹⁶ Symptoms of gastroparesis may be persistent or present as episodic flares. Due to the symptoms, some patients will experience weight loss and malnutrition and, in severe cases, dehydration.¹⁹

Although the definition of gastroparesis is a delay in gastric emptying along with symptoms, symptoms correlate poorly with the degree of delayed gastric emptying. The symptoms that appear to have the strongest correlation with gastric emptying are nausea, vomiting, early satiety, and postprandial fullness, whereas symptoms such as abdominal pain and bloating have little correlation. Furthermore, improving gastric emptying does not necessarily lead to improved symptoms, and symptom improvement does not always lead to improved gastric emptying times.²⁰ Between 5% and 12% of patients with diabetes report symptoms consistent with gastroparesis, though many of these patients have normal gastric emptying. The symptoms of gastroparesis overlap with those of functional dyspepsia, as both may have motor and sensory alterations.²¹

The Gastroparesis Cardinal Symptom Index (GCSI), a subset of the Patient Assessment of Gastrointestinal Disorders Symptom Severity Index (PAGI-SYM), is a questionnaire that is commonly used to establish symptom severity in patients with gastroparesis. It is comprised of 3 subscales—nausea and vomiting, postprandial fullness and early satiety, and bloating—which are averaged to provide a total GCSI score. Symptoms over the 2 weeks prior to administration of the questionnaire are assessed and rated from 0 (none) to 5 (very severe).²² Grading the severity of gastroparesis may take into account symptoms, quality of life, and gastric emptying. One commonly used grading system assigns a grade from 1 to 3, with grade 1 being mild gastroparesis, grade 2 being compensated gastroparesis, and grade 3 being gastric failure with refractory symptoms that are uncontrolled.^18,23 Quality-of-life surveys also suggest that gastroparesis independent of other factors leads to a worse quality of life.²⁴

Indications for GES

Gastric electric stimulator implantation is a surgical procedure with inherent risks and complications and is reserved for patients with intractable symptoms of gastroparesis who remain symptomatic despite treatment attempts with dietary management, antiemetic agents (eg, compazine, phenergan, and ondansetron), and prokinetic agents (eg, metoclopramide, erythromycin, and domperidone). Symptom modulators such as nortriptyline and mirtazapine are occasionally tried.

Surgical intervention can be considered upon failure of medical treatment measures. At least a year of documented care provided by a physician specializing in gastroparesis is suggested for surgical consideration. The gastric electric neurostimulator is approved by the FDA as a HUD for the care of patients with idiopathic and diabetic gastroparesis, performed on a compassionate basis. GES implantation requires Institutional Review Board approval at the institution, and patients are required to have documented delayed gastric emptying.

It is important to remember that the GES device is incompatible with magnetic resonance imaging (MRI) and explantation of the device is necessary prior to MRI. As such, in patients with anticipated need of frequent MRI, such as those with multiple sclerosis, serious consideration should be given to alternative strategies prior to focusing on this modality.

Device Placement

GES was devised to improve gastric emptying. The Enterra GES system uses high-frequency, low-energy electric stimulation. An alternative method is true gastric pacing that uses high-energy, low-frequency stimulation to entrain the gastric slow waves and subsequent contractions at 3 cycles per minute (cpm). Gastric pacing has greater energy requirements than GES, which makes the size of the stimulator too large to be practical. In pilot animal studies, GES produced an accelerating effect on gastric emptying, but in human studies GES had an inconsistent effect on gastric emptying. Studies have suggested that GES influences the proximal stomach, with a reduction of gastric tone,²⁵ and also that GES has an afferent modulatory mechanism.²⁶

The Enterra GES is placed surgically under general anesthesia, commonly via laparotomy or minimal access surgical techniques (laparoscopically or robotically assisted). Preoperative intravenous antibiotics are given. The system consists of a pair of electrodes connected to a pulse generator. The 2 stimulation leads are inserted into the gastric muscularis propria 1 cm apart along the greater curvature 10 cm proximal to the pylorus. Upper endoscopy is performed to ensure that the leads do not penetrate through the mucosa into the stomach lumen; if this occurs, repositioning of the lead is necessary. A horizontal incision through the skin is made, and the distal ends of the stimulating wires are tunneled through the abdominal wall and connected to the pulse generator. The impedance (resistance) between the wires is measured to ensure the appropriate range (200-800 Ohms). The neurostimulator with the distal ends of the stimulating wires is then placed into the subcutaneous pocket and sutured to the underlying fascia. The pulse generator delivers a high-frequency, low-energy, 0.1-second train of pulses at a frequency of 12 cpm. Within each pulse train, individual pulses oscillate at a frequency of 14 cycles per second. The voltage of the stimulations is set to provide a current of 5 milliamps (mA; remembering that voltage = current × resistance).

Patients are often hospitalized with a recovery time of 1 to 3 days. Immediate postoperative care usually includes intravenous fluids, controlling any postoperative ileus, advancing diet, and providing analgesic pain medications. Hospital length of stay can be impacted by surgical technique.²⁵ Patients are seen several weeks after discharge for assessment of the incision and toleration of diet. Medications for gastroparesis that patients were taking prior to the GES implantation are usually continued postoperatively, with a goal of reducing these medications over time. Patients are then followed every 3 to 12 months, depending on their clinical condition.

At follow-up visits, medications are reviewed and new treatments can be added if appropriate. The gastric stimulator is interrogated to determine if changes in resistance occurred; if necessary, minor readjustments can be made to keep the current at desired levels (5 mA). For persistent symptoms with GES treatment, the stimulator parameters can be adjusted after 3 months of follow up, typically first increasing the current from 5 to 7.5 mA and then to 10 mA. After this, the frequency can be increased from 14 Hz to 28 Hz, and then to 55 Hz. Rarely, the ON duration is increased from 0.1 to 1 second. Increasing the ON time can worsen symptoms in some patients, cause abdominal pain, and decrease the battery life from the usual 7 years.

Complications of GES

In an analysis of the Manufacturer and User Facility Device Experience (MAUDE) databank, Bielefeldt identified 1587 reports of adverse effects related to the gastric electric stimulator from January 2001 to October 2015.²⁷ The most common adverse effects are reviewed here.

Skin erosion/wound dehiscence is one of the most common reported complications; it may be related to superficial placement or inadequate securing of the device to the fascia. Abscess can develop postoperatively due to hematogenous seeding or may be a sign of lead erosion into the lumen, tracking along the leads into subcutaneous tissue.²⁸ It is important to warn patients to protect the area over the device from needle injections as this also can lead to hematoma formation and direct contamination of the device. If the device gets infected, it cannot be salvaged and requires explantation. Implantation of a new device can be attempted once all wound issues resolve.

Device migration/flipping most often occurs because the device is inadequately fixed to the underlying fascia, but occasionally it can occur from patients flipping the device around. Flipping can occur due to superficial pocket location within subcutaneous tissue, especially in obese patients. Migration/flipping can lead to prominence of the contour of the device and discomfort, ultimately requiring surgical correction.

Perforation and erosion of the leads. With time, leads can erode into the stomach, although this is rare. Usually erosion is associated with loss of device function. Endoscopy confirms this finding. In rare cases, infection can track proximally along the lead and present as a surgical site infection at the pulse generator. This complication often requires explantation of the neurostimulator leads and pulse generator.

Intestinal obstruction. Although rare, the intestines can get wrapped around the leads of the device, causing different degrees of obstruction (Figure 1). Positioning the device in the left upper quadrant minimizes the intraabdominal length of the leads and pulls them maximally out, coiling under the device (Figure 2). In cases where other locations are used either due to a hostile upper abdominal region (skin infection, presence of gastrostomy or other devices) or surgeon’s preference, the GES device can be implanted in the lower abdomen (Figure 3). In these circumstances, carefully draping the omentum over the bowels might help to prevent this complication. Tacking of the leads to the parietal peritoneum with sutures can also be preventative. In cases of obstruction requiring intervention by laparotomy or minimal access techniques (laparoscopy or robotic assisted surgery), all efforts are made to preserve the neurostimulator leads. In cases that require bowel resection, lead contamination is a serious concern, but lead explantation is not mandatory. Close postoperative monitoring for the development of lead infection is required.

Hematoma and seroma. Postoperative hematomas can occur from inadequate hemostasis, and seromas can occur in the stimulator pocket. Small hematomas may be observed if not complicated (Figure 4). In cases of large hematomas with skin compromise or dehiscence, prompt washout and drainage is required. In ideal cases, the device can be preserved. Relocation to another site might be required if skin necrosis develops. The possibility of device contamination also must be considered; after resolution of wound issues, implantation of a new device may be tried. Seromas at the generator pocket site are a frequent occurrence but are often benign, self-limiting, and generally resolve over 4 to 6 weeks.

Incisional hernia. Hernias can develop after any abdominal surgery and are not unique to GES implantation. Use of minimally invasive technique for the GES implantation minimizes this complication.

Electric shock sensations may occur from breakage of the plastic lining covering the stimulator wires or from fluid buildup around the insertion of the wires into the stimulator. Shocks can also occur due to shortening of the leads on the muscles of the abdominal wall. Patients describe periodic muscle cramps with the frequency of the device (every 5 seconds). To prevent this complication, freshly implanted leads should be covered by an omental flap to isolate them from the abdominal wall. In patients who continue to feel shocks despite all efforts, the possibility of visceral hypersensitivity should be considered. A trial of symptom modulators such as nortriptyline and lowering of the output amperage below the minimal recommended setting of 5 mA can be undertaken. If these interventions do not work, the device must be turned off for a period of time. Occasionally, replacement of the leads or explantation of the device must be considered.

Lack of effect/persistent symptoms. If a patient presents with lack of improvement after device implantation, a thorough workup should be undertaken to ensure that the device is functioning properly. In the case of abnormal impedance values, an abdominal x-ray study can be performed to rule out lead migration (Figure 5). If no abnormalities are detected, the output of the device can be increased. After adjusting device settings, the patient should be assessed for improvement over at least a 1- to 3-month period. One report suggests that in patients not responding to GES, repositioning the location of the stimulator leads on the stomach can be helpful.²⁹

Outcomes of GES

Study results of investigative GES models in animals and select patients were published in 1997.^30,31 Following these reports, 2 large multicenter studies were conducted to demonstrate the efficacy of GES for the treatment of refractory gastroparesis. The Gastric Electrical Mechanical Stimulation Study (GEMS) was an open-label, multicenter study of 38 patients who received percutaneous and later permanent GES devices.³² Marked reduction in weekly vomiting and nausea was observed at 4 weeks, with a 90% reduction in nausea and vomiting frequency at 11 months. Following this, a second multicenter study (Worldwide Anti-Vomiting Electrical Stimulation Study [WAVES]) involving a double-blind sham stimulation controlled trial with 33 idiopathic and diabetic gastroparesis patients was performed.³³ During the blinded portion of this study, there was a noticeable decrease in vomiting frequency, particularly in the patients with diabetic gastroparesis. Patient preference was for the stimulator ON as compared to OFF. The FDA’s HUD exemption for the Enterra GES device in 2000 was based on these studies.

Four independent double-blind studies of GES have been conducted (Table 1).^33-37 It has been difficult to demonstrate improvement during the double-blind period with gastric stimulation compared to no stimulation. Despite total symptom severity improvement and individual symptom improvements in these studies, a recent meta-analysis demonstrated a summative insignificant difference between the GES ON versus OFF states.³⁸

In contrast to the double-blind studies, numerous open-label studies have demonstrated clinical improvements in patients with diabetic and idiopathic gastroparesis (Table 2),^{32,33,35,36,39-55} leading some to question whether the demonstrable efficacy reflects a placebo effect or regression to the mean. Patients may perceive an operative, aggressive intervention as likely to be effectual in comparison to incremental medication efforts, thus creating a placebo effect. It should also be noted that not all open-label studies have demonstrated improvement with GES. Indeed, Jones et al reported no significant difference in nausea and vomiting at 6-month follow-up, and recommended that physicians exercise caution with GES as a therapeutic strategy given the cost and lack of confirmed demonstrable effect.⁵⁶ Thus, the clinical successes demonstrated in open-label studies must be weighed not only against the lack of unequivocal improvement, but also against the potential deleterious effects of the surgery.

In an open-label study that employed the GCSI to follow symptoms of gastroparesis, 29 patients underwent GES implantation over an 18-month period, with follow-up in 28 patients.⁴⁴ GES resulted in clinical improvement in 50% of patients with refractory gastroparesis. The overall GCSI significantly decreased, with improvement in the nausea/vomiting subscore and the post-prandial fullness subscore, but no improvement in the bloating subscore or abdominal pain. The decrease in GCSI was greater for patients with diabetic versus idiopathic gastroparesis. Patients with the main symptom of nausea/vomiting had a greater improvement than patients with the main symptom of abdominal pain. Patients taking narcotic analgesics at the time of implant had a poorer response compared to patients who were not. In this study, 3 clinical parameters were associated with a favorable clinical response: (1) diabetic rather than idiopathic gastroparesis, (2) nausea/vomiting rather than abdominal pain as the primary symptom, and (3) independence from narcotic analgesics prior to stimulator implantation. Knowledge of these 3 factors may allow improved patient selection for GES.

A large prospective study by Heckert et al detailed marked improvements with GES and the patterns of those improvements.⁵⁵ Nausea, vomiting, loss of appetite, and early satiety improved significantly with stimulator use, with a greater improvement in vomiting in patients with diabetic gastroparesis than in those with the idiopathic form. Although GES improved symptoms in 75% of all patients, patients with diabetes had a post-GES Clinical Patient Grading Assessment score that was statistically higher than the score among patients with idiopathic gastroparesis. This difference is thought to be due to the neuromolecular mechanism of diabetic gastroparesis, where blunting of the enteric nervous system may contribute to symptomatology.

Several studies have demonstrated a clinical response to GES in patients with postsurgical gastroparesis. A study by Oubre et al showed that GES led to weekly vomiting improvements as well as a reduction in total symptom severity score.⁵⁷ A study by McCallum et al further demonstrated improved symptoms, quality of life, nutritional status, and hospitalization requirements.⁵⁸ GES has also been shown to improve gastroparesis symptoms in pediatric populations.^47,59 Thus, although not a direct indication, GES has been shown to be beneficial in various subtypes of gastroparesis.

Additionally, irrespective of gastroparesis type, the improved symptomatology with GES appears to be durable, with one study showing persistent clinical improvements up to 8 years after device placement.⁶⁰ The improvements were persistent and incremental. Likewise, McCallum et al showed that continued reductions in total symptom severity scores were evident in all gastroparesis types up to 10 years after stimulator implantation.⁶¹ The success of the procedures in part comes from careful selection of patients. Clinical parameters that are associated with favorable clinical response include diabetic gastroparesis subtype, nausea/vomiting predominance, and independence from narcotic analgesics prior to stimulator placement.⁶²

GES has also been noted to improve other patient care metrics besides symptomatology, including nutritional status, reduced need for nutritional supplementation, and improved HbA1c.^63-65 Additionally, a study by Cutts et al established that health care resource utilization significantly improved at 12, 24, and 36 months following GES placement, as compared to patients receiving standard medical therapy.⁶⁶ This decreased resource utilization was also reflected in decreased costs in the GES group compared with the standard care group.

Surgical Alternatives to GES

Pyloric interventions such as pyloroplasty and pyloromyotomy are other surgical treatment modalities offered for gastroparesis. Whereas GES uses neurostimulation to facilitate gastric emptying and potentially improve fundic accommodation, pyloric interventions are intended to increase gastric emptying by reducing outflow resistance from the pyloric sphincter.

Pyloric Interventions

Various studies have shown significant improvements with pyloric interventions, similar to the improvements seen with GES. One such study involving 177 patients demonstrated an 86% improvement in gastric emptying, with symptom severity scores for nausea, vomiting, bloating, abdominal pain, and early satiety decreasing significantly at 3 months following pyloroplasty.⁶⁷ A significant advantage of pyloric interventions is that pyloromyotomy can be performed endoscopically (gastric peroral endoscopic pyloromyotomy [G-POEM] or peroral pyloromyotomy [POP]), thus minimizing the risks of open surgery. A recent review that included a pooled analysis of 7 studies of G-POEM for gastroparesis demonstrated 100% technical success, with clinical efficacy in 81.5% of the procedures as assessed by the GCSI.⁶⁸ Additionally, the intraoperative and perioperative complication rates were 6.6% and 7.6%, respectively, suggesting that G-POEM is a safe and clinically beneficial therapeutic option. Few studies comparing the outcomes of pyloric interventions to GES have been performed.

Recently, GES has been combined with pyloric interventions to maximize therapeutic potential. This allows simultaneous neurologic and functional interventions to expedite gastric emptying and improve patient symptomatology. Davis et al demonstrated significant improvement in 21 patients who underwent GES placement and pyloroplasty, with 71% improvement in total symptom severity.⁶⁹ Notably, dual surgery did not increase the incidence of infection or adverse surgical outcomes. Although this study did not directly compare dual surgery to GES alone, the results are nonetheless favorable. GES provides a strong antiemetic and anti-nausea effect, whereas the pyloromyotomy provides improvement in gastric emptying.

Feeding/Venting Tubes

Feeding jejunostomy tubes and venting gastrostomy tubes can be used alone or in combination with GES. Feeding jejunostomy is performed for malnutrition and weight loss that accompanies the refractory symptoms of early satiety, nausea, and vomiting. Venting gastrostomy tubes allow for removal of retained gastric contents that may cause distension, nausea, and vomiting. Gastrojejunostomy tubes can also be placed endoscopically or by interventional radiology.

Gastrectomy

Gastrectomy can provide therapeutic benefit through elimination of the gastric reservoir function and consequent removal of afferent neural impulses. In select patient populations, outcomes of gastrectomy have compared favorably with those of GES. For example, one study demonstrated favorable outcomes of Roux-en-Y gastrectomy in morbidly obese patients with gastroparesis.⁷⁰ In another study, favorable outcomes were reported in a cohort of 103 patients, with gastrectomy demonstrating 87% symptom improvement (nausea, vomiting, epigastric pain) compared to just 63% improvement with GES.⁷¹ However, the dramatic impact on anatomy and physiology and the invasiveness of the procedure need to be weighed against the therapeutic benefit. For example, in the same study, the 30-day morbidity was 23% for gastrectomy versus just 8% for the GES implant.⁷¹

 

 

When to Use GES

The gastric electrical neurostimulator (Enterra; Medtronic, Inc.) is approved for treatment of idiopathic and diabetic gastroparesis that is refractory to medical treatment, performed on a compassionate basis. Patients with diabetic gastroparesis respond to GES better than do patients with the idiopathic form. Of the symptoms of gastroparesis, primarily nausea and vomiting improve. Thus, GES favors patients with diabetic gastroparesis who have primarily nausea and vomiting, rather than, for instance, patients with idiopathic gastroparesis who have primarily abdominal pain and may be taking narcotics. Some centers provide GES for postsurgical patients and children with gastroparesis.

The 3 main surgical interventions for medically refractory gastroparesis are GES, pyloric intervention (pyloroplasty or pyloromyotomy), and gastrectomy. Of the 3 interventions, gastrectomy is the most radical given its dramatic effect on anatomy and is thus not preferred. The clinical decision then becomes: GES, pyloric intervention, or both? There are limited data to support a definitive answer to this question.

In a single-center retrospective analysis of prospective data (electronic medical record), Arthur et al compared outcomes of GES patients with medically refractory gastroparesis who received various surgical interventions.⁷² In total, 33 stimulator, 7 pyloroplasty, 2 gastrectomy, and 16 combined stimulator and pyloroplasty patients were analyzed for postoperative symptom improvement. Pyloroplasty alone demonstrated the least symptom improvement, combination GES and pyloroplasty demonstrated increased improvement, and GES alone demonstrated the most improvement. The results of this study suggest that barring contraindication, placement of a gastric stimulator as the initial treatment is best, with pyloroplasty reserved for patients who do not achieve adequate symptom control. Limitations of the study include its single-center design and low patient numbers for pyloroplasty in isolation.

In contrast, a recent retrospective systematic review synthesized the outcomes of various studies of GES and pyloric interventions for medically refractory gastroparesis.⁷³ A therapeutic effect was found for each surgical intervention, with pyloric surgery patients demonstrating a greater response to intervention than GES patients. Unfortunately, attempts to analyze combination interventions were hindered by a lack of power.

Conclusion

Initial management of gastroparesis is medical (lifestyle and diet changes), with antiemetic and prokinetic agents used in refractory cases. Following failure of this therapy, placement of a GES device is a surgical intervention that has been approved under FDA humanitarian device exemption to help ameliorate symptomatology. Improvement with GES has been demonstrated in nonblinded studies, but the lack of randomized controlled trials demonstrating benefit suggests the possibility of an underlying placebo effect. Additionally, new medical procedures such as G-POEM complicate the decision of which intervention should be attempted first. More studies, specifically comparing GES, pyloric interventions, and combined GES with pyloric intervention to placebo, are needed to fully understand what therapy is best for refractory gastroparesis.

Corresponding author: Henry P. Parkman, MD, Gastroenterology Section, Temple University School of Medicine, 3401 North Broad Street, Philadelphia, PA 19140; [email protected].

Financial disclosures: None.

From Temple University School of Medicine, Philadelphia, PA.

Abstract

• Objective: To outline the use and utility of gastric electric stimulation (GES) as a therapeutic intervention for gastroparesis.
• Methods: Review of the literature.
• Results: Gastroparesis is characterized by delayed gastric emptying, with symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Some patients with gastroparesis do not respond to medical intervention, and for these patients surgical intervention may be warranted. GES utilizes high-frequency gastric neurostimulation to facilitate gastric emptying and reduce symptoms of gastroparesis. It is indicated for patients with idiopathic and diabetic gastroparesis who have nausea and vomiting as their primary symptoms and who have not responded to medical therapy. GES has also been used in postsurgical and pediatric gastroparesis patients. Optimizing the outcome of this surgical treatment through proper patient selection and meticulous surgical technique is essential as there are inherent risks to the procedure. Nonblinded studies of GES for medically refractory gastroparesis have demonstrated therapeutic symptomatic benefit, whereas randomized controlled trials have not. New interventions such as pyloromyotomy and pyloroplasty are reasonable alternatives or addendums to GES.
• Conclusion: GES may be considered among the therapies available for treating patients with refractory symptoms of gastroparesis. More studies, specifically those comparing GES, pyloromyotomy, GES combined with pyloromyotomy, and placebo, are needed to help guide therapy selection for refractory gastroparesis.

Keywords: diabetes; gastroparesis; dysmotility; gastric emptying; electric stimulation.

Gastroparesis is a chronic dysmotility disorder characterized by delayed gastric emptying with associated symptoms of nausea, vomiting, early satiety, postprandial fullness, and abdominal pain. Medical treatments for gastroparesis include dietary modifications, glucose control in those with diabetes, prokinetic medications, antiemetic medications, and symptom modulators, but unfortunately patients frequently do not respond to these treatments. In patients refractory to medical therapy, surgical treatments can be considered.

Gastric electric stimulation (GES; Enterra [Medtronic, Minneapolis, MN]) was approved via a Food and Drug Administration (FDA) Humanitarian Use Device (HUD) exemption for the treatment of medically refractory gastroparesis in 2000. Understanding the indications, risks, outcomes, and alternatives to GES is essential to providing appropriate care for patients with medically refractory gastroparesis. This article outlines the use and utility of GES as a therapeutic intervention for gastroparesis.

Types of Gastroparesis

Gastroparesis is a chronic symptomatic disorder of the stomach manifested by delayed gastric emptying without evidence of gastric outlet obstruction or ulceration.¹ The pathophysiology of gastroparesis appears to involve abnormalities in functioning of several elements including the autonomic nervous system, especially the vagus nerve, smooth muscle cells, enteric neurons, and interstitial cells of Cajal.

Idiopathic gastroparesis and diabetic gastroparesis are the 2 most common types of gastroparesis.² Symptomatic delayed gastric emptying with no primary underlying abnormality predisposing to gastroparesis is categorized as idiopathic gastroparesis.³ A small subset of patients with idiopathic gastroparesis report an initial infectious prodrome such as gastroenteritis or respiratory infection. It has been suggested that this postinfectious gastroparesis results from viral injury to the neural innervation of the stomach or the interstitial cells of Cajal in the stomach.⁴ Viruses that have been implicated in the development of gastroparesis include cytomegalovirus, Epstein-Barr virus, Norwalk virus, rotavirus, herpes zoster, and varicella zoster.^5-9

Diabetic gastroparesis is characterized as onset of symptoms of gastroparesis in patients with diabetes, with concomitant delayed gastric emptying. It is often attributed to chronic hyperglycemia-induced damage to the vagus nerve, and is frequently observed in association with other diabetic complications such as neuropathy, retinopathy, and nephropathy.¹⁰

Gastroparesis that develops following surgery is classified as postsurgical gastroparesis. In the past, this form of gastroparesis most commonly occurred after ulcer surgery, often performed with vagotomy. These types of surgeries are performed less frequently in the era of proton pump inhibitor therapy and treatments for Helicobacter pylori. Presently, Nissen fundoplication and bariatric surgery are the more common surgical procedures associated with gastroparesis.³ Long-term use of medications that delay gastric emptying, such as opiate narcotic medications, can lead to gastroparesis and represent another form of iatrogenic gastroparesis. Other forms of gastroparesis (atypical gastroparesis) arise due to various underlying etiologies, including neurological disorders (eg, Parkinson disease, multiple sclerosis), metabolic or endocrine conditions (eg, hypothyroidism), autoimmune disorders, connective tissue and collagen vascular disorders (eg, systemic lupus erythematosus, scleroderma, Sjögren syndrome, Ehlers-Danlos syndrome), or eating disorders (eg, anorexia, bulimia).³

Epidemiology

There is a female preponderance in patients with gastroparesis. Data from the Rochester Epidemiology Project, a database of linked medical records for residents of Olmsted County, MN, showed that the age-adjusted prevalence of definite gastroparesis per 100,000 inhabitants was 37.8 for women and 9.6 for men.¹¹ More recent estimates have suggested a much higher prevalence of probable gastroparesis (approximately 1.8%) in the general population using symptoms suggestive of gastroparesis.¹² Hospitalization rates for gastroparesis have increased since 2000, which could reflect rising prevalence and/or the effects of heightened awareness about and better identification of gastroparesis.¹³ This increase may also be due in part to the rising rate of diabetes leading to more cases of diabetic gastroparesis; withdrawal of some gastroparesis treatments from the market (cisapride, tegaserod) leading to hospitalizations for symptoms not adequately being treated; and hospitalizations needed for insertion of the gastric electric stimulator.

Gastroparesis Symptoms

The main symptoms of gastroparesis are early satiety, postprandial fullness, bloating, nausea, and vomiting.¹⁴ Nausea (> 90% of patients) and early satiety (60% of patients) are the most common symptoms.¹⁵ Abdominal pain is often present in patients with gastroparesis but is usually not the predominant symptom. The pain can be multifactorial, with somatic, visceral, and neuropathic components.^16-18 Moderate to severe abdominal pain has been found more often in patients with idiopathic gastroparesis and in association with opiate use.¹⁶ Symptoms of gastroparesis may be persistent or present as episodic flares. Due to the symptoms, some patients will experience weight loss and malnutrition and, in severe cases, dehydration.¹⁹

Although the definition of gastroparesis is a delay in gastric emptying along with symptoms, symptoms correlate poorly with the degree of delayed gastric emptying. The symptoms that appear to have the strongest correlation with gastric emptying are nausea, vomiting, early satiety, and postprandial fullness, whereas symptoms such as abdominal pain and bloating have little correlation. Furthermore, improving gastric emptying does not necessarily lead to improved symptoms, and symptom improvement does not always lead to improved gastric emptying times.²⁰ Between 5% and 12% of patients with diabetes report symptoms consistent with gastroparesis, though many of these patients have normal gastric emptying. The symptoms of gastroparesis overlap with those of functional dyspepsia, as both may have motor and sensory alterations.²¹

The Gastroparesis Cardinal Symptom Index (GCSI), a subset of the Patient Assessment of Gastrointestinal Disorders Symptom Severity Index (PAGI-SYM), is a questionnaire that is commonly used to establish symptom severity in patients with gastroparesis. It is comprised of 3 subscales—nausea and vomiting, postprandial fullness and early satiety, and bloating—which are averaged to provide a total GCSI score. Symptoms over the 2 weeks prior to administration of the questionnaire are assessed and rated from 0 (none) to 5 (very severe).²² Grading the severity of gastroparesis may take into account symptoms, quality of life, and gastric emptying. One commonly used grading system assigns a grade from 1 to 3, with grade 1 being mild gastroparesis, grade 2 being compensated gastroparesis, and grade 3 being gastric failure with refractory symptoms that are uncontrolled.^18,23 Quality-of-life surveys also suggest that gastroparesis independent of other factors leads to a worse quality of life.²⁴

Indications for GES

Gastric electric stimulator implantation is a surgical procedure with inherent risks and complications and is reserved for patients with intractable symptoms of gastroparesis who remain symptomatic despite treatment attempts with dietary management, antiemetic agents (eg, compazine, phenergan, and ondansetron), and prokinetic agents (eg, metoclopramide, erythromycin, and domperidone). Symptom modulators such as nortriptyline and mirtazapine are occasionally tried.

Surgical intervention can be considered upon failure of medical treatment measures. At least a year of documented care provided by a physician specializing in gastroparesis is suggested for surgical consideration. The gastric electric neurostimulator is approved by the FDA as a HUD for the care of patients with idiopathic and diabetic gastroparesis, performed on a compassionate basis. GES implantation requires Institutional Review Board approval at the institution, and patients are required to have documented delayed gastric emptying.

It is important to remember that the GES device is incompatible with magnetic resonance imaging (MRI) and explantation of the device is necessary prior to MRI. As such, in patients with anticipated need of frequent MRI, such as those with multiple sclerosis, serious consideration should be given to alternative strategies prior to focusing on this modality.

Device Placement

GES was devised to improve gastric emptying. The Enterra GES system uses high-frequency, low-energy electric stimulation. An alternative method is true gastric pacing that uses high-energy, low-frequency stimulation to entrain the gastric slow waves and subsequent contractions at 3 cycles per minute (cpm). Gastric pacing has greater energy requirements than GES, which makes the size of the stimulator too large to be practical. In pilot animal studies, GES produced an accelerating effect on gastric emptying, but in human studies GES had an inconsistent effect on gastric emptying. Studies have suggested that GES influences the proximal stomach, with a reduction of gastric tone,²⁵ and also that GES has an afferent modulatory mechanism.²⁶

The Enterra GES is placed surgically under general anesthesia, commonly via laparotomy or minimal access surgical techniques (laparoscopically or robotically assisted). Preoperative intravenous antibiotics are given. The system consists of a pair of electrodes connected to a pulse generator. The 2 stimulation leads are inserted into the gastric muscularis propria 1 cm apart along the greater curvature 10 cm proximal to the pylorus. Upper endoscopy is performed to ensure that the leads do not penetrate through the mucosa into the stomach lumen; if this occurs, repositioning of the lead is necessary. A horizontal incision through the skin is made, and the distal ends of the stimulating wires are tunneled through the abdominal wall and connected to the pulse generator. The impedance (resistance) between the wires is measured to ensure the appropriate range (200-800 Ohms). The neurostimulator with the distal ends of the stimulating wires is then placed into the subcutaneous pocket and sutured to the underlying fascia. The pulse generator delivers a high-frequency, low-energy, 0.1-second train of pulses at a frequency of 12 cpm. Within each pulse train, individual pulses oscillate at a frequency of 14 cycles per second. The voltage of the stimulations is set to provide a current of 5 milliamps (mA; remembering that voltage = current × resistance).

Patients are often hospitalized with a recovery time of 1 to 3 days. Immediate postoperative care usually includes intravenous fluids, controlling any postoperative ileus, advancing diet, and providing analgesic pain medications. Hospital length of stay can be impacted by surgical technique.²⁵ Patients are seen several weeks after discharge for assessment of the incision and toleration of diet. Medications for gastroparesis that patients were taking prior to the GES implantation are usually continued postoperatively, with a goal of reducing these medications over time. Patients are then followed every 3 to 12 months, depending on their clinical condition.

At follow-up visits, medications are reviewed and new treatments can be added if appropriate. The gastric stimulator is interrogated to determine if changes in resistance occurred; if necessary, minor readjustments can be made to keep the current at desired levels (5 mA). For persistent symptoms with GES treatment, the stimulator parameters can be adjusted after 3 months of follow up, typically first increasing the current from 5 to 7.5 mA and then to 10 mA. After this, the frequency can be increased from 14 Hz to 28 Hz, and then to 55 Hz. Rarely, the ON duration is increased from 0.1 to 1 second. Increasing the ON time can worsen symptoms in some patients, cause abdominal pain, and decrease the battery life from the usual 7 years.

Complications of GES

In an analysis of the Manufacturer and User Facility Device Experience (MAUDE) databank, Bielefeldt identified 1587 reports of adverse effects related to the gastric electric stimulator from January 2001 to October 2015.²⁷ The most common adverse effects are reviewed here.

Skin erosion/wound dehiscence is one of the most common reported complications; it may be related to superficial placement or inadequate securing of the device to the fascia. Abscess can develop postoperatively due to hematogenous seeding or may be a sign of lead erosion into the lumen, tracking along the leads into subcutaneous tissue.²⁸ It is important to warn patients to protect the area over the device from needle injections as this also can lead to hematoma formation and direct contamination of the device. If the device gets infected, it cannot be salvaged and requires explantation. Implantation of a new device can be attempted once all wound issues resolve.

Device migration/flipping most often occurs because the device is inadequately fixed to the underlying fascia, but occasionally it can occur from patients flipping the device around. Flipping can occur due to superficial pocket location within subcutaneous tissue, especially in obese patients. Migration/flipping can lead to prominence of the contour of the device and discomfort, ultimately requiring surgical correction.

Perforation and erosion of the leads. With time, leads can erode into the stomach, although this is rare. Usually erosion is associated with loss of device function. Endoscopy confirms this finding. In rare cases, infection can track proximally along the lead and present as a surgical site infection at the pulse generator. This complication often requires explantation of the neurostimulator leads and pulse generator.

Intestinal obstruction. Although rare, the intestines can get wrapped around the leads of the device, causing different degrees of obstruction (Figure 1). Positioning the device in the left upper quadrant minimizes the intraabdominal length of the leads and pulls them maximally out, coiling under the device (Figure 2). In cases where other locations are used either due to a hostile upper abdominal region (skin infection, presence of gastrostomy or other devices) or surgeon’s preference, the GES device can be implanted in the lower abdomen (Figure 3). In these circumstances, carefully draping the omentum over the bowels might help to prevent this complication. Tacking of the leads to the parietal peritoneum with sutures can also be preventative. In cases of obstruction requiring intervention by laparotomy or minimal access techniques (laparoscopy or robotic assisted surgery), all efforts are made to preserve the neurostimulator leads. In cases that require bowel resection, lead contamination is a serious concern, but lead explantation is not mandatory. Close postoperative monitoring for the development of lead infection is required.

Hematoma and seroma. Postoperative hematomas can occur from inadequate hemostasis, and seromas can occur in the stimulator pocket. Small hematomas may be observed if not complicated (Figure 4). In cases of large hematomas with skin compromise or dehiscence, prompt washout and drainage is required. In ideal cases, the device can be preserved. Relocation to another site might be required if skin necrosis develops. The possibility of device contamination also must be considered; after resolution of wound issues, implantation of a new device may be tried. Seromas at the generator pocket site are a frequent occurrence but are often benign, self-limiting, and generally resolve over 4 to 6 weeks.

Incisional hernia. Hernias can develop after any abdominal surgery and are not unique to GES implantation. Use of minimally invasive technique for the GES implantation minimizes this complication.

Electric shock sensations may occur from breakage of the plastic lining covering the stimulator wires or from fluid buildup around the insertion of the wires into the stimulator. Shocks can also occur due to shortening of the leads on the muscles of the abdominal wall. Patients describe periodic muscle cramps with the frequency of the device (every 5 seconds). To prevent this complication, freshly implanted leads should be covered by an omental flap to isolate them from the abdominal wall. In patients who continue to feel shocks despite all efforts, the possibility of visceral hypersensitivity should be considered. A trial of symptom modulators such as nortriptyline and lowering of the output amperage below the minimal recommended setting of 5 mA can be undertaken. If these interventions do not work, the device must be turned off for a period of time. Occasionally, replacement of the leads or explantation of the device must be considered.

Lack of effect/persistent symptoms. If a patient presents with lack of improvement after device implantation, a thorough workup should be undertaken to ensure that the device is functioning properly. In the case of abnormal impedance values, an abdominal x-ray study can be performed to rule out lead migration (Figure 5). If no abnormalities are detected, the output of the device can be increased. After adjusting device settings, the patient should be assessed for improvement over at least a 1- to 3-month period. One report suggests that in patients not responding to GES, repositioning the location of the stimulator leads on the stomach can be helpful.²⁹

Outcomes of GES

Study results of investigative GES models in animals and select patients were published in 1997.^30,31 Following these reports, 2 large multicenter studies were conducted to demonstrate the efficacy of GES for the treatment of refractory gastroparesis. The Gastric Electrical Mechanical Stimulation Study (GEMS) was an open-label, multicenter study of 38 patients who received percutaneous and later permanent GES devices.³² Marked reduction in weekly vomiting and nausea was observed at 4 weeks, with a 90% reduction in nausea and vomiting frequency at 11 months. Following this, a second multicenter study (Worldwide Anti-Vomiting Electrical Stimulation Study [WAVES]) involving a double-blind sham stimulation controlled trial with 33 idiopathic and diabetic gastroparesis patients was performed.³³ During the blinded portion of this study, there was a noticeable decrease in vomiting frequency, particularly in the patients with diabetic gastroparesis. Patient preference was for the stimulator ON as compared to OFF. The FDA’s HUD exemption for the Enterra GES device in 2000 was based on these studies.

Four independent double-blind studies of GES have been conducted (Table 1).^33-37 It has been difficult to demonstrate improvement during the double-blind period with gastric stimulation compared to no stimulation. Despite total symptom severity improvement and individual symptom improvements in these studies, a recent meta-analysis demonstrated a summative insignificant difference between the GES ON versus OFF states.³⁸

In contrast to the double-blind studies, numerous open-label studies have demonstrated clinical improvements in patients with diabetic and idiopathic gastroparesis (Table 2),^{32,33,35,36,39-55} leading some to question whether the demonstrable efficacy reflects a placebo effect or regression to the mean. Patients may perceive an operative, aggressive intervention as likely to be effectual in comparison to incremental medication efforts, thus creating a placebo effect. It should also be noted that not all open-label studies have demonstrated improvement with GES. Indeed, Jones et al reported no significant difference in nausea and vomiting at 6-month follow-up, and recommended that physicians exercise caution with GES as a therapeutic strategy given the cost and lack of confirmed demonstrable effect.⁵⁶ Thus, the clinical successes demonstrated in open-label studies must be weighed not only against the lack of unequivocal improvement, but also against the potential deleterious effects of the surgery.

In an open-label study that employed the GCSI to follow symptoms of gastroparesis, 29 patients underwent GES implantation over an 18-month period, with follow-up in 28 patients.⁴⁴ GES resulted in clinical improvement in 50% of patients with refractory gastroparesis. The overall GCSI significantly decreased, with improvement in the nausea/vomiting subscore and the post-prandial fullness subscore, but no improvement in the bloating subscore or abdominal pain. The decrease in GCSI was greater for patients with diabetic versus idiopathic gastroparesis. Patients with the main symptom of nausea/vomiting had a greater improvement than patients with the main symptom of abdominal pain. Patients taking narcotic analgesics at the time of implant had a poorer response compared to patients who were not. In this study, 3 clinical parameters were associated with a favorable clinical response: (1) diabetic rather than idiopathic gastroparesis, (2) nausea/vomiting rather than abdominal pain as the primary symptom, and (3) independence from narcotic analgesics prior to stimulator implantation. Knowledge of these 3 factors may allow improved patient selection for GES.

A large prospective study by Heckert et al detailed marked improvements with GES and the patterns of those improvements.⁵⁵ Nausea, vomiting, loss of appetite, and early satiety improved significantly with stimulator use, with a greater improvement in vomiting in patients with diabetic gastroparesis than in those with the idiopathic form. Although GES improved symptoms in 75% of all patients, patients with diabetes had a post-GES Clinical Patient Grading Assessment score that was statistically higher than the score among patients with idiopathic gastroparesis. This difference is thought to be due to the neuromolecular mechanism of diabetic gastroparesis, where blunting of the enteric nervous system may contribute to symptomatology.

Several studies have demonstrated a clinical response to GES in patients with postsurgical gastroparesis. A study by Oubre et al showed that GES led to weekly vomiting improvements as well as a reduction in total symptom severity score.⁵⁷ A study by McCallum et al further demonstrated improved symptoms, quality of life, nutritional status, and hospitalization requirements.⁵⁸ GES has also been shown to improve gastroparesis symptoms in pediatric populations.^47,59 Thus, although not a direct indication, GES has been shown to be beneficial in various subtypes of gastroparesis.

Additionally, irrespective of gastroparesis type, the improved symptomatology with GES appears to be durable, with one study showing persistent clinical improvements up to 8 years after device placement.⁶⁰ The improvements were persistent and incremental. Likewise, McCallum et al showed that continued reductions in total symptom severity scores were evident in all gastroparesis types up to 10 years after stimulator implantation.⁶¹ The success of the procedures in part comes from careful selection of patients. Clinical parameters that are associated with favorable clinical response include diabetic gastroparesis subtype, nausea/vomiting predominance, and independence from narcotic analgesics prior to stimulator placement.⁶²

GES has also been noted to improve other patient care metrics besides symptomatology, including nutritional status, reduced need for nutritional supplementation, and improved HbA1c.^63-65 Additionally, a study by Cutts et al established that health care resource utilization significantly improved at 12, 24, and 36 months following GES placement, as compared to patients receiving standard medical therapy.⁶⁶ This decreased resource utilization was also reflected in decreased costs in the GES group compared with the standard care group.

Surgical Alternatives to GES

Pyloric interventions such as pyloroplasty and pyloromyotomy are other surgical treatment modalities offered for gastroparesis. Whereas GES uses neurostimulation to facilitate gastric emptying and potentially improve fundic accommodation, pyloric interventions are intended to increase gastric emptying by reducing outflow resistance from the pyloric sphincter.

Pyloric Interventions

Various studies have shown significant improvements with pyloric interventions, similar to the improvements seen with GES. One such study involving 177 patients demonstrated an 86% improvement in gastric emptying, with symptom severity scores for nausea, vomiting, bloating, abdominal pain, and early satiety decreasing significantly at 3 months following pyloroplasty.⁶⁷ A significant advantage of pyloric interventions is that pyloromyotomy can be performed endoscopically (gastric peroral endoscopic pyloromyotomy [G-POEM] or peroral pyloromyotomy [POP]), thus minimizing the risks of open surgery. A recent review that included a pooled analysis of 7 studies of G-POEM for gastroparesis demonstrated 100% technical success, with clinical efficacy in 81.5% of the procedures as assessed by the GCSI.⁶⁸ Additionally, the intraoperative and perioperative complication rates were 6.6% and 7.6%, respectively, suggesting that G-POEM is a safe and clinically beneficial therapeutic option. Few studies comparing the outcomes of pyloric interventions to GES have been performed.

Recently, GES has been combined with pyloric interventions to maximize therapeutic potential. This allows simultaneous neurologic and functional interventions to expedite gastric emptying and improve patient symptomatology. Davis et al demonstrated significant improvement in 21 patients who underwent GES placement and pyloroplasty, with 71% improvement in total symptom severity.⁶⁹ Notably, dual surgery did not increase the incidence of infection or adverse surgical outcomes. Although this study did not directly compare dual surgery to GES alone, the results are nonetheless favorable. GES provides a strong antiemetic and anti-nausea effect, whereas the pyloromyotomy provides improvement in gastric emptying.

Feeding/Venting Tubes

Feeding jejunostomy tubes and venting gastrostomy tubes can be used alone or in combination with GES. Feeding jejunostomy is performed for malnutrition and weight loss that accompanies the refractory symptoms of early satiety, nausea, and vomiting. Venting gastrostomy tubes allow for removal of retained gastric contents that may cause distension, nausea, and vomiting. Gastrojejunostomy tubes can also be placed endoscopically or by interventional radiology.

Gastrectomy

Gastrectomy can provide therapeutic benefit through elimination of the gastric reservoir function and consequent removal of afferent neural impulses. In select patient populations, outcomes of gastrectomy have compared favorably with those of GES. For example, one study demonstrated favorable outcomes of Roux-en-Y gastrectomy in morbidly obese patients with gastroparesis.⁷⁰ In another study, favorable outcomes were reported in a cohort of 103 patients, with gastrectomy demonstrating 87% symptom improvement (nausea, vomiting, epigastric pain) compared to just 63% improvement with GES.⁷¹ However, the dramatic impact on anatomy and physiology and the invasiveness of the procedure need to be weighed against the therapeutic benefit. For example, in the same study, the 30-day morbidity was 23% for gastrectomy versus just 8% for the GES implant.⁷¹

 

 

When to Use GES

The gastric electrical neurostimulator (Enterra; Medtronic, Inc.) is approved for treatment of idiopathic and diabetic gastroparesis that is refractory to medical treatment, performed on a compassionate basis. Patients with diabetic gastroparesis respond to GES better than do patients with the idiopathic form. Of the symptoms of gastroparesis, primarily nausea and vomiting improve. Thus, GES favors patients with diabetic gastroparesis who have primarily nausea and vomiting, rather than, for instance, patients with idiopathic gastroparesis who have primarily abdominal pain and may be taking narcotics. Some centers provide GES for postsurgical patients and children with gastroparesis.

The 3 main surgical interventions for medically refractory gastroparesis are GES, pyloric intervention (pyloroplasty or pyloromyotomy), and gastrectomy. Of the 3 interventions, gastrectomy is the most radical given its dramatic effect on anatomy and is thus not preferred. The clinical decision then becomes: GES, pyloric intervention, or both? There are limited data to support a definitive answer to this question.

In a single-center retrospective analysis of prospective data (electronic medical record), Arthur et al compared outcomes of GES patients with medically refractory gastroparesis who received various surgical interventions.⁷² In total, 33 stimulator, 7 pyloroplasty, 2 gastrectomy, and 16 combined stimulator and pyloroplasty patients were analyzed for postoperative symptom improvement. Pyloroplasty alone demonstrated the least symptom improvement, combination GES and pyloroplasty demonstrated increased improvement, and GES alone demonstrated the most improvement. The results of this study suggest that barring contraindication, placement of a gastric stimulator as the initial treatment is best, with pyloroplasty reserved for patients who do not achieve adequate symptom control. Limitations of the study include its single-center design and low patient numbers for pyloroplasty in isolation.

In contrast, a recent retrospective systematic review synthesized the outcomes of various studies of GES and pyloric interventions for medically refractory gastroparesis.⁷³ A therapeutic effect was found for each surgical intervention, with pyloric surgery patients demonstrating a greater response to intervention than GES patients. Unfortunately, attempts to analyze combination interventions were hindered by a lack of power.

Conclusion

Initial management of gastroparesis is medical (lifestyle and diet changes), with antiemetic and prokinetic agents used in refractory cases. Following failure of this therapy, placement of a GES device is a surgical intervention that has been approved under FDA humanitarian device exemption to help ameliorate symptomatology. Improvement with GES has been demonstrated in nonblinded studies, but the lack of randomized controlled trials demonstrating benefit suggests the possibility of an underlying placebo effect. Additionally, new medical procedures such as G-POEM complicate the decision of which intervention should be attempted first. More studies, specifically comparing GES, pyloric interventions, and combined GES with pyloric intervention to placebo, are needed to fully understand what therapy is best for refractory gastroparesis.

Corresponding author: Henry P. Parkman, MD, Gastroenterology Section, Temple University School of Medicine, 3401 North Broad Street, Philadelphia, PA 19140; [email protected].

Financial disclosures: None.