Nephrology

People with acute kidney injury receiving renal replacement therapy have an increased long-term risk of dying regardless of the type of RRT they receive, according to a new study.

Intensity of RRT performed in intensive care units, researchers also found, made no difference in the likelihood of a patient needing maintenance dialysis or having protein in the urine within 4 years after the intervention.

The results, published Feb. 11 in PLoS Medicine (2014 Feb. 11 [doi: 10.1371/journal.pmed.1001601]), come from extended follow-up in a trial of 1,464 AKI patients in ICUs who were randomized to receive RRT of higher or lower intensity. Dr. Martin Gallagher of the George Institute for Global Health in Sydney, Australia, led the study.

The researchers did not see high-intensity RRT associated with any improvements in long-term survival: At a median of 43.9 months after randomization, 63% of patients in the lower-intensity group and 62% of patients in the higher-intensity group had died (risk ratio, 1.04; 95% CI, 0.96-1.12; P = .49).

Of the 810 patients who survived more than 90 days past randomization, rates of maintenance dialysis were similarly low: 5.1% for the lower-intensity RRT group and 5.8% for the higher-intensity group (RR, 1.12; 95% CI, 0.63-2.00; P = .69). Both groups, however, saw high rates of albuminuria: 40% and 44%, respectively (P = .48). Chronic proteinuria is an established risk factor for death, cardiovascular disease, and additional dialysis.

Dr. Gallagher and his colleagues wrote in their analysis that the findings "support the view that survivors of AKI are at increased risk and that closer surveillance may be justified. In addition, our findings suggest that chronic proteinuria reduction strategies, which have shown benefit in some patient groups with proteinuria, may warrant investigation as a therapeutic intervention."

The study was supported by the Australian government. One coauthor, Dr. Rinaldo Bellomo, disclosed receiving financial support from Eli Lilly, Cardinal Health, and CSL Bioplasma. The George Institute for Global Health, Dr. Gallagher’s institution, has received research funding from Servier, Novartis, and other companies.

[email protected]

People with acute kidney injury receiving renal replacement therapy have an increased long-term risk of dying regardless of the type of RRT they receive, according to a new study.

Intensity of RRT performed in intensive care units, researchers also found, made no difference in the likelihood of a patient needing maintenance dialysis or having protein in the urine within 4 years after the intervention.

The results, published Feb. 11 in PLoS Medicine (2014 Feb. 11 [doi: 10.1371/journal.pmed.1001601]), come from extended follow-up in a trial of 1,464 AKI patients in ICUs who were randomized to receive RRT of higher or lower intensity. Dr. Martin Gallagher of the George Institute for Global Health in Sydney, Australia, led the study.

The researchers did not see high-intensity RRT associated with any improvements in long-term survival: At a median of 43.9 months after randomization, 63% of patients in the lower-intensity group and 62% of patients in the higher-intensity group had died (risk ratio, 1.04; 95% CI, 0.96-1.12; P = .49).

Of the 810 patients who survived more than 90 days past randomization, rates of maintenance dialysis were similarly low: 5.1% for the lower-intensity RRT group and 5.8% for the higher-intensity group (RR, 1.12; 95% CI, 0.63-2.00; P = .69). Both groups, however, saw high rates of albuminuria: 40% and 44%, respectively (P = .48). Chronic proteinuria is an established risk factor for death, cardiovascular disease, and additional dialysis.

Dr. Gallagher and his colleagues wrote in their analysis that the findings "support the view that survivors of AKI are at increased risk and that closer surveillance may be justified. In addition, our findings suggest that chronic proteinuria reduction strategies, which have shown benefit in some patient groups with proteinuria, may warrant investigation as a therapeutic intervention."

The study was supported by the Australian government. One coauthor, Dr. Rinaldo Bellomo, disclosed receiving financial support from Eli Lilly, Cardinal Health, and CSL Bioplasma. The George Institute for Global Health, Dr. Gallagher’s institution, has received research funding from Servier, Novartis, and other companies.

[email protected]

People with acute kidney injury receiving renal replacement therapy have an increased long-term risk of dying regardless of the type of RRT they receive, according to a new study.

Intensity of RRT performed in intensive care units, researchers also found, made no difference in the likelihood of a patient needing maintenance dialysis or having protein in the urine within 4 years after the intervention.

The results, published Feb. 11 in PLoS Medicine (2014 Feb. 11 [doi: 10.1371/journal.pmed.1001601]), come from extended follow-up in a trial of 1,464 AKI patients in ICUs who were randomized to receive RRT of higher or lower intensity. Dr. Martin Gallagher of the George Institute for Global Health in Sydney, Australia, led the study.

The researchers did not see high-intensity RRT associated with any improvements in long-term survival: At a median of 43.9 months after randomization, 63% of patients in the lower-intensity group and 62% of patients in the higher-intensity group had died (risk ratio, 1.04; 95% CI, 0.96-1.12; P = .49).

Of the 810 patients who survived more than 90 days past randomization, rates of maintenance dialysis were similarly low: 5.1% for the lower-intensity RRT group and 5.8% for the higher-intensity group (RR, 1.12; 95% CI, 0.63-2.00; P = .69). Both groups, however, saw high rates of albuminuria: 40% and 44%, respectively (P = .48). Chronic proteinuria is an established risk factor for death, cardiovascular disease, and additional dialysis.

Dr. Gallagher and his colleagues wrote in their analysis that the findings "support the view that survivors of AKI are at increased risk and that closer surveillance may be justified. In addition, our findings suggest that chronic proteinuria reduction strategies, which have shown benefit in some patient groups with proteinuria, may warrant investigation as a therapeutic intervention."

The study was supported by the Australian government. One coauthor, Dr. Rinaldo Bellomo, disclosed receiving financial support from Eli Lilly, Cardinal Health, and CSL Bioplasma. The George Institute for Global Health, Dr. Gallagher’s institution, has received research funding from Servier, Novartis, and other companies.

[email protected]

SNOWMASS, COLO. – The presence of preoperative dysfunction in more than any one of four key organ systems profoundly reduces survival in patients undergoing surgical aortic valve replacement, a study showed.

"If you have two or more dysfunctional organ systems, you really need to think about what you’re doing for this patient. At 5 years, only about 40% of these patients are alive. It makes a lot of sense to me to say that if you have a patient with severe COPD [chronic obstructive pulmonary disease] and renal dysfunction, that patient should probably never get a surgical valve," Dr. Vinod H. Thourani said at the Annual Cardiovascular Conference at Snowmass.

In a retrospective analysis of a registry with prospectively entered data, 29% of 1,759 patients who underwent surgical aortic valve replacement (SAVR) with or without coronary artery bypass grafting at Emory University during 2002-2010 had preoperative dysfunction of one or more of four organ systems under scrutiny. Eighty-five patients had severe COPD, as defined by a forced expiratory volume in 1 second (FEV₁) that was less than 50% of predicted, 140 had chronic renal failure, 149 had a prior stroke, and 241 had heart failure with a left ventricular ejection less than 35%.

Patients with chronic renal failure had far and away the worst 30-day and long-term outcomes. Half were dead within 3 years. The 7-year survival rate was just 11.7%.

The second-worst outcomes were seen in patients with severe COPD preoperatively. Their 7-year survival rate was 30.8%.

"Anyone with an FEV₁ below about 40% becomes a higher-risk surgical candidate; think instead of TAVR [transcatheter aortic valve replacement],"advised Dr. Thourani of the division of cardiothoracic surgery at Emory University, Atlanta.

In contrast, outcomes in patients with either heart failure or prior stroke "were not that bad," he said, pointing to 7-year survival rates of 55.9% and 48.6%, respectively.

Ninety-five patients (5.4%) in this recently published study (Ann. Thorac. Surg. 2013;95:838-45) had more than one dysfunctional organ system prior to SAVR. Median survival in patients without dysfunction in any of the four organ systems was 8.2 years and counting. With one dysfunctional organ, it was still good at 7.2 years. However, with two dysfunctional organ systems, the median survival dropped precipitously to 4.1 years. With three dysfunctional organ systems, it was 5.9 years.

Dr. Thourini serves as a consultant to Edwards Lifesciences, Sorin, and St. Jude Medical.

[email protected]

SNOWMASS, COLO. – The presence of preoperative dysfunction in more than any one of four key organ systems profoundly reduces survival in patients undergoing surgical aortic valve replacement, a study showed.

"If you have two or more dysfunctional organ systems, you really need to think about what you’re doing for this patient. At 5 years, only about 40% of these patients are alive. It makes a lot of sense to me to say that if you have a patient with severe COPD [chronic obstructive pulmonary disease] and renal dysfunction, that patient should probably never get a surgical valve," Dr. Vinod H. Thourani said at the Annual Cardiovascular Conference at Snowmass.

In a retrospective analysis of a registry with prospectively entered data, 29% of 1,759 patients who underwent surgical aortic valve replacement (SAVR) with or without coronary artery bypass grafting at Emory University during 2002-2010 had preoperative dysfunction of one or more of four organ systems under scrutiny. Eighty-five patients had severe COPD, as defined by a forced expiratory volume in 1 second (FEV₁) that was less than 50% of predicted, 140 had chronic renal failure, 149 had a prior stroke, and 241 had heart failure with a left ventricular ejection less than 35%.

Patients with chronic renal failure had far and away the worst 30-day and long-term outcomes. Half were dead within 3 years. The 7-year survival rate was just 11.7%.

The second-worst outcomes were seen in patients with severe COPD preoperatively. Their 7-year survival rate was 30.8%.

"Anyone with an FEV₁ below about 40% becomes a higher-risk surgical candidate; think instead of TAVR [transcatheter aortic valve replacement],"advised Dr. Thourani of the division of cardiothoracic surgery at Emory University, Atlanta.

In contrast, outcomes in patients with either heart failure or prior stroke "were not that bad," he said, pointing to 7-year survival rates of 55.9% and 48.6%, respectively.

Ninety-five patients (5.4%) in this recently published study (Ann. Thorac. Surg. 2013;95:838-45) had more than one dysfunctional organ system prior to SAVR. Median survival in patients without dysfunction in any of the four organ systems was 8.2 years and counting. With one dysfunctional organ, it was still good at 7.2 years. However, with two dysfunctional organ systems, the median survival dropped precipitously to 4.1 years. With three dysfunctional organ systems, it was 5.9 years.

Dr. Thourini serves as a consultant to Edwards Lifesciences, Sorin, and St. Jude Medical.

[email protected]

SNOWMASS, COLO. – The presence of preoperative dysfunction in more than any one of four key organ systems profoundly reduces survival in patients undergoing surgical aortic valve replacement, a study showed.

"If you have two or more dysfunctional organ systems, you really need to think about what you’re doing for this patient. At 5 years, only about 40% of these patients are alive. It makes a lot of sense to me to say that if you have a patient with severe COPD [chronic obstructive pulmonary disease] and renal dysfunction, that patient should probably never get a surgical valve," Dr. Vinod H. Thourani said at the Annual Cardiovascular Conference at Snowmass.

In a retrospective analysis of a registry with prospectively entered data, 29% of 1,759 patients who underwent surgical aortic valve replacement (SAVR) with or without coronary artery bypass grafting at Emory University during 2002-2010 had preoperative dysfunction of one or more of four organ systems under scrutiny. Eighty-five patients had severe COPD, as defined by a forced expiratory volume in 1 second (FEV₁) that was less than 50% of predicted, 140 had chronic renal failure, 149 had a prior stroke, and 241 had heart failure with a left ventricular ejection less than 35%.

Patients with chronic renal failure had far and away the worst 30-day and long-term outcomes. Half were dead within 3 years. The 7-year survival rate was just 11.7%.

The second-worst outcomes were seen in patients with severe COPD preoperatively. Their 7-year survival rate was 30.8%.

"Anyone with an FEV₁ below about 40% becomes a higher-risk surgical candidate; think instead of TAVR [transcatheter aortic valve replacement],"advised Dr. Thourani of the division of cardiothoracic surgery at Emory University, Atlanta.

In contrast, outcomes in patients with either heart failure or prior stroke "were not that bad," he said, pointing to 7-year survival rates of 55.9% and 48.6%, respectively.

Ninety-five patients (5.4%) in this recently published study (Ann. Thorac. Surg. 2013;95:838-45) had more than one dysfunctional organ system prior to SAVR. Median survival in patients without dysfunction in any of the four organ systems was 8.2 years and counting. With one dysfunctional organ, it was still good at 7.2 years. However, with two dysfunctional organ systems, the median survival dropped precipitously to 4.1 years. With three dysfunctional organ systems, it was 5.9 years.

Dr. Thourini serves as a consultant to Edwards Lifesciences, Sorin, and St. Jude Medical.

[email protected]

CE/CME No: CR-1402

PROGRAM OVERVIEW
Earn credit by reading this article and successfully completing the posttest. Successful completion is defined as a cumulative score of at least 70% correct.

EDUCATIONAL OBJECTIVES
• Discuss the incidence and staging of chronic kidney disease (CKD).
• Enumerate the treatment goals for the CKD patient with diabetes.
• Describe the hypoglycemic medications that can be used at each stage of CKD.

FACULTY
Cheryl Gilmartin is a Clinical Pharmacist in Nephrology and a Clinical Assistant Professor in the College of Pharmacy at the University of Illinois at Chicago. Jane S. Davis, a member of the Clinician Reviews editorial board, is a nurse practitioner in the Division of Nephrology at the University of Alabama at Birmingham and is the communications chairperson for the National Kidney Foundation’s Council of Advanced Practitioners (NKF-CAP). Kim Zuber, past chair of the NKF-CAP, is a physician assistant with Metropolitan Nephrology in Alexandria, Virginia, and Clinton, Maryland.

ACCREDITATION STATEMENT

Article begins on next page >>

Because the vast majority of patients with chronic kidney disease (CKD) are diabetic and thus take hypoglycemic medications, knowledge of renal dosing for these medications, their mechanisms of action, and their safety profiles, as well as consideration of A1C goals, is vital for the practicing clinician. Management of the diabetic CKD patient, identified by stage of kidney disease, is outlined, with dosing regimens as determined by the glomerular filtration rate. Special attention is given to insulin management.

Diabetes is the most common cause of chronic kidney disease (CKD) throughout the world and leaves in its wake huge financial and social burdens.¹ Diabetes has been the main cause of kidney failure in the United States for a number of years; as of 2012, diabetes became the most common cause of kidney failure in the world.²

The cost of caring for the rapidly increasing diabetes population in the US has been enormous. In 2012, the price tag for treating diabetes was $245 billion for medical care alone.³ The societal cost—loss in life years, loss of productivity caused by an increase in early retirement and disability, and burden of caregiving on families—is immeasurable.

In addition to CKD, diabetes is a major risk factor for heart disease and can lead to blindness and amputations. The rising incidence of diabetes has resulted in a new sense of urgency and has prompted the health care community to find new ways to reduce the burden on the patient, as well as encourage more aggressive research on halting progression of the disease, preventing end-stage kidney failure, and avoiding the need for dialysis.

Nearly two out of three (59%) adults in the US will be diagnosed with CKD stages 3 to 5 during their lifetimes.⁴ Some of this can be attributed to normal loss of kidney function associated with aging. However, much is due to the double burden of diabetes and hypertension.⁵ Patients with CKD require ongoing medical care, and much of it will be provided in primary care practices.⁶ This article discusses the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 staging guidelines and explains which oral and injectable diabetes medications are acceptable at each stage of CKD (as defined by the glomerular filtration rate [GFR]) and how to simplify dosing.

On the next page: Staging CKD >>

STAGING CHRONIC KIDNEY DISEASE
The increasing incidence of CKD was first recognized in the 1990s. The need to care for the patient with CKD was understood, but there was a dearth of standardized management practices. Research was hindered by the many names used to identify kidney disease in the literature: from mildly elevated serum creatinine to low clearance to renal insufficiency.⁷ To rectify this, the National Kidney Foundation (NKF) convened an expert panel to develop a standard language and units of measurement to define kidney disease. The panel was tasked with developing classifications and an evaluation system for renal disease; identifying ways to attain more timely referrals to nephrology; clarifying research objectives; and delineating how to improve medication dosing. Ultimately, the main objective was to improve management of the patient with CKD.

In 2002, the NKF’s Dialysis Outcomes Quality Initiative (K/DOQI) defined the stages of kidney disease based on assessment of GFR and other kidney disease markers.⁸ GFR is a mathematical calculation for determining kidney function that corrects for loss of kidney function and incorporates the patient’s gender, race, age, and body size—the nonmodifiable risk factors known in 2001. GFR is expressed in mL/min/1.73 m². Classifying patients with CKD according to the stage of kidney disease by GFR provides a mechanism for studying the efficacy and adverse reactions of medications while correlating them to each stage. This is of particular importance when treating patients with diabetes and kidney disease, as it can help to avoid adverse effects associated with inappropriate renal dosing. Many medications are renally eliminated; as kidney function diminishes, the pharmacokinetics of drugs are altered, potentially causing toxicity and an increased risk for drug interactions. As the GFR falls, even drugs not eliminated by the kidney may have altered pharmacokinetics and pharmacodynamics that may cause renal or systemic toxicity.¹

Determining the precise method for renally dosing medication has been one of the most vigorously debated topics in CKD. Many medications were prescribed using the patient’s serum creatinine (SCr) concentration, a marker but not an exact measure of kidney disease.⁸ The Cockcroft-Gault equation, which provides an estimate of creatinine clearance (eCrCl; an alternate GFR measurement), was primarily used to determine renal drug dosing. ⁹

In 1999, a multisite study group worked to develop an equation to measure loss of kidney function in the patient with CKD.^10,11 The new formula, known as the MDRD (Modification of Diet in Renal Disease) equation, standardized the measurement of loss of kidney function and provided an estimated GFR (eGFR) that was more accurate then eCrCl. At that time, the National Kidney Disease Education Program (NKDEP) decided to use the MDRD equation to assess renal function and use the eCrCl for dosing medications. The latter was chosen because most drugs were analyzed using eCrCl.⁹ More recently, the NKDEP stated that either the eCrCl or the eGFR is acceptable for renal drug dosing. Hudson and Nyman⁹ advise that when discrepancies occur between eGFR and eCrCl, clinical judgment should be applied in the elderly and in those with extremes in body weight. In those cases, therapeutic outcomes and risk versus benefit need to be considered when determining renal dosing.

The publication of CKD stages by K/DOQI in 2002 sparked research that has uncovered additional ­criteria that might be useful in slowing the progression and improving the management of CKD. Thus, in 2013, new guidelines by KDIGO incorporating the latest international findings were published.¹ KDIGO expands the definition of CKD to include albuminuria (ALB) as an added risk factor (see Figure) and has modified K/DOQI stages 3 and 5. ALB was included because it has predictive value for progression of CKD to kidney failure. In the new classification, there are three ranges defined for ALB measured in mg/g (see Table 1). The KDIGO guidelines, like K/DOQI, classify kidney disease using a 1-to-5 staging system; however, the stage numbers in the KDIGO classification are prefaced by a G (ie, G1, G2). Also, as mentioned above, stages G3 and G5 in the KDIGO classification have been further categorized.

CKD stage 3 was divided for practical reasons. It was found that the original CKD stage 3 was too large and diverse a classification to use for research studies, tracking of hospitalizations, and precise dosing of medications.¹² K/DOQI stage 3 encompassed a GFR from 30 to 59 mL/min/1.73 m². The KDIGO guideline divides K/DOQI stage 3 into G3a (GFR of 45 to 59 mL/min/1.73 m²) and G3b (GFR of 30 to 44 mL/min/1.73 m²). The staging is then further classified by the presence or absence of ALB. The patient with a GFR of 56 mL/min/1.73 m² with ALB is in CKD stage 3aA3 (very high and nephrotic ALB) and has a greater risk for disease progression than the patient with a GFR of 56 mL/min/1.73 m² without ALB. The latter patient is in CKD stage 3aA1(optimal and mildly high ALB). Finally, many medication dosages that were acceptable at a GFR of 56 mL/min/1.73 m² had to be further adjusted for patients with a GFR of 30 mL/min/1.73 m² to avoid toxicity. This further justified the KDIGO reclassification of stage 3.

Stage 5 from K/DOQI only reflected a GFR < 15 mL/min/1.73 m². KDIGO utilizes the same GFR but splits the stage into patients not receiving dialysis and those receiving dialysis, G5 and G5D, respectively. The dosing and timing of medications due to drug dialyzability—including renal-specific medications—may be quite different for patients on dialysis as compared to those with a GFR < 15 mL/min/1.73 m² not on dialysis, making the distinction beneficial.

On the next page: Diabetes management >>

DIABETES MANAGEMENT
The treatment of patients early in diabetic CKD is often the responsibility of primary care providers, who are faced with the daunting task of addressing the renal dosing of medications. The eGFR or eCrCl needs to be utilized to adjust diabetic medications, largely to avoid hypoglycemia. To alleviate some of the difficulty, the KDIGO CKD stage and the eCrCl and eGFR for diabetic medications are delineated in Table 2.¹

Note that the guidelines are not meant to be strictly adhered to but rather are intended as a tool for managing the patient with CKD. Clinically significant patient factors need to be considered to adequately adjust drugs in CKD. Also, as new data become available from postmarketing reports, dosing adjustments, added precautions, or updated risk factors in these fragile patients may need to be incorporated into the guidelines. As new diabetic medications are approved by the FDA, kidney-specific dosing will continue to be a moving target.

While making lifestyle changes is very important early in diabetes, by the time diabetic nephropathy manifests, more aggressive action is warranted.¹² A part of the diabetic microvascular triad (along with retinopathy and neuropathy), nephropathy signals existing and irreversible organ damage. Before instituting treatment to delay diabetic nephropathy progression, an A1C goal needs to be established. The long-standing goal has been an A1C < 7%, according to guidelines established by the American Diabetes Association, or 6.5%, as recognized by the American Association of Clinical Endocrinologists.¹³ The Diabetes Control and Complications Trial (DCCT) demonstrated microvascular benefit of tight glycemic control in type 1 diabetic patients when the A1C was maintained at an average of 7.2% (versus 9.1% in conventional therapy).¹⁴ The United Kingdom Prospective Diabetes Study similarly demonstrated microvascular benefit with intensive blood glucose control in type 2 diabetic patients who were maintained at a median A1C of 7% compared to the ­conventional treatment group maintained at an A1C of 7.9%.¹⁵

In January 2013, Perkovic and colleagues, in a post hoc analysis of the ADVANCE Trial, showed that maintaining an A1C between 6.5% and 7% in a patient with diabetes and macroalbuminuria (UACR > 300 μg/mg) slows progression to kidney failure.¹⁶ The time needed to treat with intensive glucose control was five years, and the number needed to treat was 41. Thus, for every 41 patients with diabetes, CKD, and ALB whose A1C is below 7%, one will not progress to kidney failure in five years of treatment.

The five-year lead time is troublesome, however. It discourages many patients from taking steps to achieve strict glycemic control; a significant number fail to follow their diabetic diets or take their medications until the damage is done. For the elderly kidney patient with diabetes and multiple other comorbidities, aggressively managing diabetes may result in hypoglycemia.¹² On balance, the slight loss of kidney function five years hence is less problematic than the present risk for a fall and resulting hip fracture due to hypoglycemia. For this reason, the KDOQI glycemic guidelines suggest an A1C between 7% to 7.5% as the goal for those with significant comorbidities.¹²

On the next page: Diabetic medication dosing for CKD >>

DIABETIC MEDICATION DOSING FOR CKD
Glycemic control is important in delaying the progression of kidney failure in the patient with CKD. Hypoglycemic medications by definition may cause low blood glucose levels. This is of particular concern in patients with diminishing renal function, particularly the elderly and CKD patients. Antidiabetes drugs for patients in CKD stages 1 and 2 have few renal precautions, although care must be taken with metformin (see Table 3). Metformin is the drug of choice for the patient newly diagnosed with diabetes. It is inexpensive and effective, causes hypoglycemia only with intensive exercise, is taken orally, and is generally well tolerated. However, the package insert (PI) indicates stopping it in patients whose kidney disease has progressed to a GFR < 60 mL/min/1.73 m² because it causes an increased risk for lactic acidosis.^17,18

Metformin was developed prior to 1998, when the approval of the dosage regimen was tied to SCr levels rather than the presently accepted eGFR.¹⁹ As a result, the PI states that metformin should not be used in women with an SCr > 1.4 mg/dL or in men with an SCr > 1.5 mg/dL.¹⁷ However, as noted, SCr concentration alone is a very poor indicator of kidney disease (see Table 4).

Currently, the KDIGO guidelines recommend assessing the GFR for metformin dosing. When the GFR is < 60 mL/min/1.73 m², KDIGO recommends metformin be discontinued in patients with low muscle mass or serious concurrent illness, or those who are concurrently receiving renally eliminated or nephrotoxic drugs (this includes, but is not limited to, renin-angiotensin-aldosterone system [RAAS] inhibitors, NSAIDs, and diuretics).¹ In those patients who do not have the exclusions delineated above, KDIGO recommends that metformin may be continued when the GFR is > 45 mL/min/1.73 m² (stages G1 to G3a), closely monitored and reconsidered when the GFR is 30 to 44 mL/min/1.73 m² (stage G3b), and discontinued when the GFR is < 30 mL/min/1.73 m² (stages G4 to G5). The KDIGO differs from other sources that recommend metformin be stopped when the GFR is < 50 to 70 ml/min/1.73 m².^20-22 The metformin PI also recommends stopping metformin the day of or day before and two days after procedures in which radiographic iodinated contrast dye is administered, to avoid acute kidney failure. Finally, it is important to note that any combination medication formulated with metformin should be stopped when the GFR is 50 mL/min/1.73 m² because lactic acidosis has also been seen in patients taking these medications.²¹

At stage 3a (GFR, 45 to 60 mL/min/1.73 m²), the sulfonylureas require dosing changes (see Table 2). Glipizide is often the oral medication of choice during CKD, and it may be used in all stages, including dialysis, but does require renal dosing and monitoring.²³ Although the insulins are more effective than glipizide in glycemic control, many practitioners prescribe it for patients who will not accept an injectable medication. Glyburide is not used at this stage, as it can lead to complications in patients with a GFR < 60 mL/min/1.73 m², particularly hypoglycemia.

Like the sulfonylureas, the meglitinides enhance insulin secretion.^13,18 Repaglinide and nateglinide are fast-acting and need to be taken with meals. Because it is recommended that the dose of the meglitinides be omitted when a meal is skipped, they are good for patients who eat meals irregularly. The meglitinides require initiation at a low dose in severe renal insufficiency.

The thiazolidinediones (TDZs) are used with caution in patients with CKD, not only because of controversial adverse cardiac effects, but also for peripheral edema that interferes with CKD management.¹⁸ The TDZs (rosiglitazone and pioglitazone) require no dose adjustment and appear to have a beneficial effect in obese patients and in patients who experienced weight gain with other hypoglycemic agents.

The newest of the diabetic drugs, canagliflozin, requires a working kidney to be effective.^24,25 It is used with precaution when GFR is 45 to 59 mL/min/1.73 m² at stage G3a and is contraindicated when GFR is < 45 mL/min/1.73 m² in stage G3b. Canagliflozin's effects are felt in early segment of proximal convoluted tubule where it blocks sodium-glucose co-transporter 2, thereby lowering reabsorption of glucose and increasing excretion of glucose in the urine. In practical terms, it can affect kidney function and, due to the inhibition of the glucose-sodium pathway, induce hyperkalemia.²⁵ Monitoring of both the GFR and potassium is vital when administering this drug. As data from postmarketing reports come in and practitioners gain experience, use of canaglifozin in the patient with CKD will be further delineated.

At a GFR between 30 and 45 mL/min/1.73 m², CKD stage 3b, alpha-glucosidase inhibitors acarbose and miglitol are contraindicated. The GLP-1 mimetic liraglutide needs to be avoided when the GFR is < 60 mL/min/1.73 m² (stage G3a). While exenatide requires close monitoring and caution when the GFR is < 50 mL/min/1.73 m², it needs to be discontinued at a GFR
< 30 mL/min/1.73 m² (stage 4). DPP-4 inhibitors become problematic at a GFR < 60 mL/min/
1.73 m² beginning at stage G3a. All except the DPP-4 inhibitor linagliptin require dosing changes as the loss of GFR continues. Exact dosing adjustments for the drug categories are found in Table 2.

In stage 4 CKD (GFR, 15 to 30 mL/min/1.73 m²), more medications are eliminated from consideration. The amylinomimetic pramlintide cannot be used for a patient with a GFR < 15 mL/min/1.73 m² (see Table 2). Both GLP-1 mimetics, exenatide and liraglutide, must be stopped when the GFR drops to 30 mL/min/1.73 m². DDP-4 inhibitors need to be dosed according to the GFR but are still acceptable at CKD stage 4. Ideally, patients with diabetes should be referred to nephrology at stage 3, but referral is vital by stage 4.

Although oral medications are not as commonly prescribed as insulins at CKD stages 4 and 5, in stage 5 CKD (GFR, < 15 mL/min/1.73 m²), a limited number of oral diabetic agents are available (see Table 2). Glipizide is inexpensive and generic; therefore, it is the most commonly used oral medication in this stage. A recent study found linagliptin was safe to use throughout all stages of kidney disease without adjustment for GFR.²⁶ While nateglinide is still acceptable in stage 5, it is used less frequently than glipizide.

As the kidney fails, the need for diabetic medications also decreases and dosing needs to take into account any residual renal function. Because glucose is produced in the proximal tubules of the kidney (known as gluconeogenesis), the level of glucose falls as the kidney fails. Excretion of insulin and other diabetes medications decreases, and thus the amount of insulin remaining in circulation is increased. This, in combination with the reduced glucose, increases the patient’s risk for hypoglycemia.¹³ Although protocols for medication adjustment are tied to GFR levels, close follow-up by the practitioner is required because GFR is not a reliable measure of kidney function when creatinine levels change rapidly.²⁷

On the next page: Insulins >>

INSULINS
Many patients start insulin prior to reaching CKD stage 4 or 5 because it is the best choice for managing diabetes through all stages of CKD.²⁸ Insulin can slow the progression to kidney failure by providing better A1C control, and all patients should be encouraged to start insulin as soon as it is appropriate.²⁹ Many patients are very reluctant to use injectables, but by stage 5 they are willing to use insulin if it will slow progression of disease and help prevent the need for dialysis.²

The long-acting basal insulins (detemir and glargine) or a combination of basal and oral medication can work quite well at stage 5. Many patients will require a low dose of the long-acting basal insulins and short-acting insulin with meals. As the kidney fails, the short-acting doses can often be discontinued.²⁹ A protocol that mixes long- and short-acting insulin with the largest meal is very effective at this stage. The long-acting insulin dose will need to be decreased as the kidney continues to fail. The American College of Physicians protocol for insulin suggests a 25% decrease at stage 4 and a 50% decrease at stage 5. Each patient is different, however, and dosing must be determined individually.²³

On the next page: Patient education >>

PATIENT EDUCATION
There is a correlation between patients who participate in kidney disease education programs and a decrease in the progression of kidney disease (see “Case Study: The Recalcitrant Patient With Diabetes"). These classes can be taught by a physician, physician assistant, nurse practitioner, or clinical nurse specialist. To alleviate the burden of end-stage kidney disease in the US, Medicare Part B pays for six outpatient kidney disease patient education classes per lifetime of the beneficiary.³⁰

Since 2010, when kidney disease education programs were rolled out, over 10,000 classes have been taught.³¹ Many practitioners report that patients are more compliant and understand their disease better when they attend classes.³² Data are being gathered to determine the effectiveness of the classes on patient outcomes. 

On the next page: Conclusion >>

CONCLUSION
Knowing how to manage the patient with both diabetes and kidney disease is increasingly vital as this patient population grows. Much of the management of these patients will fall to the primary care practitioner.⁶ The most effective way to start treatment is to identify which CKD stage the patient fits into.

Once the patient is properly categorized, safe and effective diabetic medications can be selected and dosed according to stage. Although the new classification system may be difficult to incorporate into some electronic medical record systems and practitioner behavior, ultimately, it will allow safer management of the patient with CKD and a better predictive power of outcome. 

CE/CME No: CR-1402

PROGRAM OVERVIEW
Earn credit by reading this article and successfully completing the posttest. Successful completion is defined as a cumulative score of at least 70% correct.

EDUCATIONAL OBJECTIVES
• Discuss the incidence and staging of chronic kidney disease (CKD).
• Enumerate the treatment goals for the CKD patient with diabetes.
• Describe the hypoglycemic medications that can be used at each stage of CKD.

FACULTY
Cheryl Gilmartin is a Clinical Pharmacist in Nephrology and a Clinical Assistant Professor in the College of Pharmacy at the University of Illinois at Chicago. Jane S. Davis, a member of the Clinician Reviews editorial board, is a nurse practitioner in the Division of Nephrology at the University of Alabama at Birmingham and is the communications chairperson for the National Kidney Foundation’s Council of Advanced Practitioners (NKF-CAP). Kim Zuber, past chair of the NKF-CAP, is a physician assistant with Metropolitan Nephrology in Alexandria, Virginia, and Clinton, Maryland.

ACCREDITATION STATEMENT

Article begins on next page >>

Because the vast majority of patients with chronic kidney disease (CKD) are diabetic and thus take hypoglycemic medications, knowledge of renal dosing for these medications, their mechanisms of action, and their safety profiles, as well as consideration of A1C goals, is vital for the practicing clinician. Management of the diabetic CKD patient, identified by stage of kidney disease, is outlined, with dosing regimens as determined by the glomerular filtration rate. Special attention is given to insulin management.

Diabetes is the most common cause of chronic kidney disease (CKD) throughout the world and leaves in its wake huge financial and social burdens.¹ Diabetes has been the main cause of kidney failure in the United States for a number of years; as of 2012, diabetes became the most common cause of kidney failure in the world.²

The cost of caring for the rapidly increasing diabetes population in the US has been enormous. In 2012, the price tag for treating diabetes was $245 billion for medical care alone.³ The societal cost—loss in life years, loss of productivity caused by an increase in early retirement and disability, and burden of caregiving on families—is immeasurable.

In addition to CKD, diabetes is a major risk factor for heart disease and can lead to blindness and amputations. The rising incidence of diabetes has resulted in a new sense of urgency and has prompted the health care community to find new ways to reduce the burden on the patient, as well as encourage more aggressive research on halting progression of the disease, preventing end-stage kidney failure, and avoiding the need for dialysis.

Nearly two out of three (59%) adults in the US will be diagnosed with CKD stages 3 to 5 during their lifetimes.⁴ Some of this can be attributed to normal loss of kidney function associated with aging. However, much is due to the double burden of diabetes and hypertension.⁵ Patients with CKD require ongoing medical care, and much of it will be provided in primary care practices.⁶ This article discusses the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 staging guidelines and explains which oral and injectable diabetes medications are acceptable at each stage of CKD (as defined by the glomerular filtration rate [GFR]) and how to simplify dosing.

On the next page: Staging CKD >>

STAGING CHRONIC KIDNEY DISEASE
The increasing incidence of CKD was first recognized in the 1990s. The need to care for the patient with CKD was understood, but there was a dearth of standardized management practices. Research was hindered by the many names used to identify kidney disease in the literature: from mildly elevated serum creatinine to low clearance to renal insufficiency.⁷ To rectify this, the National Kidney Foundation (NKF) convened an expert panel to develop a standard language and units of measurement to define kidney disease. The panel was tasked with developing classifications and an evaluation system for renal disease; identifying ways to attain more timely referrals to nephrology; clarifying research objectives; and delineating how to improve medication dosing. Ultimately, the main objective was to improve management of the patient with CKD.

In 2002, the NKF’s Dialysis Outcomes Quality Initiative (K/DOQI) defined the stages of kidney disease based on assessment of GFR and other kidney disease markers.⁸ GFR is a mathematical calculation for determining kidney function that corrects for loss of kidney function and incorporates the patient’s gender, race, age, and body size—the nonmodifiable risk factors known in 2001. GFR is expressed in mL/min/1.73 m². Classifying patients with CKD according to the stage of kidney disease by GFR provides a mechanism for studying the efficacy and adverse reactions of medications while correlating them to each stage. This is of particular importance when treating patients with diabetes and kidney disease, as it can help to avoid adverse effects associated with inappropriate renal dosing. Many medications are renally eliminated; as kidney function diminishes, the pharmacokinetics of drugs are altered, potentially causing toxicity and an increased risk for drug interactions. As the GFR falls, even drugs not eliminated by the kidney may have altered pharmacokinetics and pharmacodynamics that may cause renal or systemic toxicity.¹

Determining the precise method for renally dosing medication has been one of the most vigorously debated topics in CKD. Many medications were prescribed using the patient’s serum creatinine (SCr) concentration, a marker but not an exact measure of kidney disease.⁸ The Cockcroft-Gault equation, which provides an estimate of creatinine clearance (eCrCl; an alternate GFR measurement), was primarily used to determine renal drug dosing. ⁹

In 1999, a multisite study group worked to develop an equation to measure loss of kidney function in the patient with CKD.^10,11 The new formula, known as the MDRD (Modification of Diet in Renal Disease) equation, standardized the measurement of loss of kidney function and provided an estimated GFR (eGFR) that was more accurate then eCrCl. At that time, the National Kidney Disease Education Program (NKDEP) decided to use the MDRD equation to assess renal function and use the eCrCl for dosing medications. The latter was chosen because most drugs were analyzed using eCrCl.⁹ More recently, the NKDEP stated that either the eCrCl or the eGFR is acceptable for renal drug dosing. Hudson and Nyman⁹ advise that when discrepancies occur between eGFR and eCrCl, clinical judgment should be applied in the elderly and in those with extremes in body weight. In those cases, therapeutic outcomes and risk versus benefit need to be considered when determining renal dosing.

The publication of CKD stages by K/DOQI in 2002 sparked research that has uncovered additional ­criteria that might be useful in slowing the progression and improving the management of CKD. Thus, in 2013, new guidelines by KDIGO incorporating the latest international findings were published.¹ KDIGO expands the definition of CKD to include albuminuria (ALB) as an added risk factor (see Figure) and has modified K/DOQI stages 3 and 5. ALB was included because it has predictive value for progression of CKD to kidney failure. In the new classification, there are three ranges defined for ALB measured in mg/g (see Table 1). The KDIGO guidelines, like K/DOQI, classify kidney disease using a 1-to-5 staging system; however, the stage numbers in the KDIGO classification are prefaced by a G (ie, G1, G2). Also, as mentioned above, stages G3 and G5 in the KDIGO classification have been further categorized.

CKD stage 3 was divided for practical reasons. It was found that the original CKD stage 3 was too large and diverse a classification to use for research studies, tracking of hospitalizations, and precise dosing of medications.¹² K/DOQI stage 3 encompassed a GFR from 30 to 59 mL/min/1.73 m². The KDIGO guideline divides K/DOQI stage 3 into G3a (GFR of 45 to 59 mL/min/1.73 m²) and G3b (GFR of 30 to 44 mL/min/1.73 m²). The staging is then further classified by the presence or absence of ALB. The patient with a GFR of 56 mL/min/1.73 m² with ALB is in CKD stage 3aA3 (very high and nephrotic ALB) and has a greater risk for disease progression than the patient with a GFR of 56 mL/min/1.73 m² without ALB. The latter patient is in CKD stage 3aA1(optimal and mildly high ALB). Finally, many medication dosages that were acceptable at a GFR of 56 mL/min/1.73 m² had to be further adjusted for patients with a GFR of 30 mL/min/1.73 m² to avoid toxicity. This further justified the KDIGO reclassification of stage 3.

Stage 5 from K/DOQI only reflected a GFR < 15 mL/min/1.73 m². KDIGO utilizes the same GFR but splits the stage into patients not receiving dialysis and those receiving dialysis, G5 and G5D, respectively. The dosing and timing of medications due to drug dialyzability—including renal-specific medications—may be quite different for patients on dialysis as compared to those with a GFR < 15 mL/min/1.73 m² not on dialysis, making the distinction beneficial.

On the next page: Diabetes management >>

DIABETES MANAGEMENT
The treatment of patients early in diabetic CKD is often the responsibility of primary care providers, who are faced with the daunting task of addressing the renal dosing of medications. The eGFR or eCrCl needs to be utilized to adjust diabetic medications, largely to avoid hypoglycemia. To alleviate some of the difficulty, the KDIGO CKD stage and the eCrCl and eGFR for diabetic medications are delineated in Table 2.¹

Note that the guidelines are not meant to be strictly adhered to but rather are intended as a tool for managing the patient with CKD. Clinically significant patient factors need to be considered to adequately adjust drugs in CKD. Also, as new data become available from postmarketing reports, dosing adjustments, added precautions, or updated risk factors in these fragile patients may need to be incorporated into the guidelines. As new diabetic medications are approved by the FDA, kidney-specific dosing will continue to be a moving target.

While making lifestyle changes is very important early in diabetes, by the time diabetic nephropathy manifests, more aggressive action is warranted.¹² A part of the diabetic microvascular triad (along with retinopathy and neuropathy), nephropathy signals existing and irreversible organ damage. Before instituting treatment to delay diabetic nephropathy progression, an A1C goal needs to be established. The long-standing goal has been an A1C < 7%, according to guidelines established by the American Diabetes Association, or 6.5%, as recognized by the American Association of Clinical Endocrinologists.¹³ The Diabetes Control and Complications Trial (DCCT) demonstrated microvascular benefit of tight glycemic control in type 1 diabetic patients when the A1C was maintained at an average of 7.2% (versus 9.1% in conventional therapy).¹⁴ The United Kingdom Prospective Diabetes Study similarly demonstrated microvascular benefit with intensive blood glucose control in type 2 diabetic patients who were maintained at a median A1C of 7% compared to the ­conventional treatment group maintained at an A1C of 7.9%.¹⁵

In January 2013, Perkovic and colleagues, in a post hoc analysis of the ADVANCE Trial, showed that maintaining an A1C between 6.5% and 7% in a patient with diabetes and macroalbuminuria (UACR > 300 μg/mg) slows progression to kidney failure.¹⁶ The time needed to treat with intensive glucose control was five years, and the number needed to treat was 41. Thus, for every 41 patients with diabetes, CKD, and ALB whose A1C is below 7%, one will not progress to kidney failure in five years of treatment.

The five-year lead time is troublesome, however. It discourages many patients from taking steps to achieve strict glycemic control; a significant number fail to follow their diabetic diets or take their medications until the damage is done. For the elderly kidney patient with diabetes and multiple other comorbidities, aggressively managing diabetes may result in hypoglycemia.¹² On balance, the slight loss of kidney function five years hence is less problematic than the present risk for a fall and resulting hip fracture due to hypoglycemia. For this reason, the KDOQI glycemic guidelines suggest an A1C between 7% to 7.5% as the goal for those with significant comorbidities.¹²

On the next page: Diabetic medication dosing for CKD >>

DIABETIC MEDICATION DOSING FOR CKD
Glycemic control is important in delaying the progression of kidney failure in the patient with CKD. Hypoglycemic medications by definition may cause low blood glucose levels. This is of particular concern in patients with diminishing renal function, particularly the elderly and CKD patients. Antidiabetes drugs for patients in CKD stages 1 and 2 have few renal precautions, although care must be taken with metformin (see Table 3). Metformin is the drug of choice for the patient newly diagnosed with diabetes. It is inexpensive and effective, causes hypoglycemia only with intensive exercise, is taken orally, and is generally well tolerated. However, the package insert (PI) indicates stopping it in patients whose kidney disease has progressed to a GFR < 60 mL/min/1.73 m² because it causes an increased risk for lactic acidosis.^17,18

Metformin was developed prior to 1998, when the approval of the dosage regimen was tied to SCr levels rather than the presently accepted eGFR.¹⁹ As a result, the PI states that metformin should not be used in women with an SCr > 1.4 mg/dL or in men with an SCr > 1.5 mg/dL.¹⁷ However, as noted, SCr concentration alone is a very poor indicator of kidney disease (see Table 4).

Currently, the KDIGO guidelines recommend assessing the GFR for metformin dosing. When the GFR is < 60 mL/min/1.73 m², KDIGO recommends metformin be discontinued in patients with low muscle mass or serious concurrent illness, or those who are concurrently receiving renally eliminated or nephrotoxic drugs (this includes, but is not limited to, renin-angiotensin-aldosterone system [RAAS] inhibitors, NSAIDs, and diuretics).¹ In those patients who do not have the exclusions delineated above, KDIGO recommends that metformin may be continued when the GFR is > 45 mL/min/1.73 m² (stages G1 to G3a), closely monitored and reconsidered when the GFR is 30 to 44 mL/min/1.73 m² (stage G3b), and discontinued when the GFR is < 30 mL/min/1.73 m² (stages G4 to G5). The KDIGO differs from other sources that recommend metformin be stopped when the GFR is < 50 to 70 ml/min/1.73 m².^20-22 The metformin PI also recommends stopping metformin the day of or day before and two days after procedures in which radiographic iodinated contrast dye is administered, to avoid acute kidney failure. Finally, it is important to note that any combination medication formulated with metformin should be stopped when the GFR is 50 mL/min/1.73 m² because lactic acidosis has also been seen in patients taking these medications.²¹

At stage 3a (GFR, 45 to 60 mL/min/1.73 m²), the sulfonylureas require dosing changes (see Table 2). Glipizide is often the oral medication of choice during CKD, and it may be used in all stages, including dialysis, but does require renal dosing and monitoring.²³ Although the insulins are more effective than glipizide in glycemic control, many practitioners prescribe it for patients who will not accept an injectable medication. Glyburide is not used at this stage, as it can lead to complications in patients with a GFR < 60 mL/min/1.73 m², particularly hypoglycemia.

Like the sulfonylureas, the meglitinides enhance insulin secretion.^13,18 Repaglinide and nateglinide are fast-acting and need to be taken with meals. Because it is recommended that the dose of the meglitinides be omitted when a meal is skipped, they are good for patients who eat meals irregularly. The meglitinides require initiation at a low dose in severe renal insufficiency.

The thiazolidinediones (TDZs) are used with caution in patients with CKD, not only because of controversial adverse cardiac effects, but also for peripheral edema that interferes with CKD management.¹⁸ The TDZs (rosiglitazone and pioglitazone) require no dose adjustment and appear to have a beneficial effect in obese patients and in patients who experienced weight gain with other hypoglycemic agents.

The newest of the diabetic drugs, canagliflozin, requires a working kidney to be effective.^24,25 It is used with precaution when GFR is 45 to 59 mL/min/1.73 m² at stage G3a and is contraindicated when GFR is < 45 mL/min/1.73 m² in stage G3b. Canagliflozin's effects are felt in early segment of proximal convoluted tubule where it blocks sodium-glucose co-transporter 2, thereby lowering reabsorption of glucose and increasing excretion of glucose in the urine. In practical terms, it can affect kidney function and, due to the inhibition of the glucose-sodium pathway, induce hyperkalemia.²⁵ Monitoring of both the GFR and potassium is vital when administering this drug. As data from postmarketing reports come in and practitioners gain experience, use of canaglifozin in the patient with CKD will be further delineated.

At a GFR between 30 and 45 mL/min/1.73 m², CKD stage 3b, alpha-glucosidase inhibitors acarbose and miglitol are contraindicated. The GLP-1 mimetic liraglutide needs to be avoided when the GFR is < 60 mL/min/1.73 m² (stage G3a). While exenatide requires close monitoring and caution when the GFR is < 50 mL/min/1.73 m², it needs to be discontinued at a GFR
< 30 mL/min/1.73 m² (stage 4). DPP-4 inhibitors become problematic at a GFR < 60 mL/min/
1.73 m² beginning at stage G3a. All except the DPP-4 inhibitor linagliptin require dosing changes as the loss of GFR continues. Exact dosing adjustments for the drug categories are found in Table 2.

In stage 4 CKD (GFR, 15 to 30 mL/min/1.73 m²), more medications are eliminated from consideration. The amylinomimetic pramlintide cannot be used for a patient with a GFR < 15 mL/min/1.73 m² (see Table 2). Both GLP-1 mimetics, exenatide and liraglutide, must be stopped when the GFR drops to 30 mL/min/1.73 m². DDP-4 inhibitors need to be dosed according to the GFR but are still acceptable at CKD stage 4. Ideally, patients with diabetes should be referred to nephrology at stage 3, but referral is vital by stage 4.

Although oral medications are not as commonly prescribed as insulins at CKD stages 4 and 5, in stage 5 CKD (GFR, < 15 mL/min/1.73 m²), a limited number of oral diabetic agents are available (see Table 2). Glipizide is inexpensive and generic; therefore, it is the most commonly used oral medication in this stage. A recent study found linagliptin was safe to use throughout all stages of kidney disease without adjustment for GFR.²⁶ While nateglinide is still acceptable in stage 5, it is used less frequently than glipizide.

As the kidney fails, the need for diabetic medications also decreases and dosing needs to take into account any residual renal function. Because glucose is produced in the proximal tubules of the kidney (known as gluconeogenesis), the level of glucose falls as the kidney fails. Excretion of insulin and other diabetes medications decreases, and thus the amount of insulin remaining in circulation is increased. This, in combination with the reduced glucose, increases the patient’s risk for hypoglycemia.¹³ Although protocols for medication adjustment are tied to GFR levels, close follow-up by the practitioner is required because GFR is not a reliable measure of kidney function when creatinine levels change rapidly.²⁷

On the next page: Insulins >>

INSULINS
Many patients start insulin prior to reaching CKD stage 4 or 5 because it is the best choice for managing diabetes through all stages of CKD.²⁸ Insulin can slow the progression to kidney failure by providing better A1C control, and all patients should be encouraged to start insulin as soon as it is appropriate.²⁹ Many patients are very reluctant to use injectables, but by stage 5 they are willing to use insulin if it will slow progression of disease and help prevent the need for dialysis.²

The long-acting basal insulins (detemir and glargine) or a combination of basal and oral medication can work quite well at stage 5. Many patients will require a low dose of the long-acting basal insulins and short-acting insulin with meals. As the kidney fails, the short-acting doses can often be discontinued.²⁹ A protocol that mixes long- and short-acting insulin with the largest meal is very effective at this stage. The long-acting insulin dose will need to be decreased as the kidney continues to fail. The American College of Physicians protocol for insulin suggests a 25% decrease at stage 4 and a 50% decrease at stage 5. Each patient is different, however, and dosing must be determined individually.²³

On the next page: Patient education >>

PATIENT EDUCATION
There is a correlation between patients who participate in kidney disease education programs and a decrease in the progression of kidney disease (see “Case Study: The Recalcitrant Patient With Diabetes"). These classes can be taught by a physician, physician assistant, nurse practitioner, or clinical nurse specialist. To alleviate the burden of end-stage kidney disease in the US, Medicare Part B pays for six outpatient kidney disease patient education classes per lifetime of the beneficiary.³⁰

Since 2010, when kidney disease education programs were rolled out, over 10,000 classes have been taught.³¹ Many practitioners report that patients are more compliant and understand their disease better when they attend classes.³² Data are being gathered to determine the effectiveness of the classes on patient outcomes. 

On the next page: Conclusion >>

CONCLUSION
Knowing how to manage the patient with both diabetes and kidney disease is increasingly vital as this patient population grows. Much of the management of these patients will fall to the primary care practitioner.⁶ The most effective way to start treatment is to identify which CKD stage the patient fits into.

Once the patient is properly categorized, safe and effective diabetic medications can be selected and dosed according to stage. Although the new classification system may be difficult to incorporate into some electronic medical record systems and practitioner behavior, ultimately, it will allow safer management of the patient with CKD and a better predictive power of outcome. 

CE/CME No: CR-1402

PROGRAM OVERVIEW
Earn credit by reading this article and successfully completing the posttest. Successful completion is defined as a cumulative score of at least 70% correct.

EDUCATIONAL OBJECTIVES
• Discuss the incidence and staging of chronic kidney disease (CKD).
• Enumerate the treatment goals for the CKD patient with diabetes.
• Describe the hypoglycemic medications that can be used at each stage of CKD.

FACULTY
Cheryl Gilmartin is a Clinical Pharmacist in Nephrology and a Clinical Assistant Professor in the College of Pharmacy at the University of Illinois at Chicago. Jane S. Davis, a member of the Clinician Reviews editorial board, is a nurse practitioner in the Division of Nephrology at the University of Alabama at Birmingham and is the communications chairperson for the National Kidney Foundation’s Council of Advanced Practitioners (NKF-CAP). Kim Zuber, past chair of the NKF-CAP, is a physician assistant with Metropolitan Nephrology in Alexandria, Virginia, and Clinton, Maryland.

ACCREDITATION STATEMENT

Article begins on next page >>

Because the vast majority of patients with chronic kidney disease (CKD) are diabetic and thus take hypoglycemic medications, knowledge of renal dosing for these medications, their mechanisms of action, and their safety profiles, as well as consideration of A1C goals, is vital for the practicing clinician. Management of the diabetic CKD patient, identified by stage of kidney disease, is outlined, with dosing regimens as determined by the glomerular filtration rate. Special attention is given to insulin management.

Diabetes is the most common cause of chronic kidney disease (CKD) throughout the world and leaves in its wake huge financial and social burdens.¹ Diabetes has been the main cause of kidney failure in the United States for a number of years; as of 2012, diabetes became the most common cause of kidney failure in the world.²

The cost of caring for the rapidly increasing diabetes population in the US has been enormous. In 2012, the price tag for treating diabetes was $245 billion for medical care alone.³ The societal cost—loss in life years, loss of productivity caused by an increase in early retirement and disability, and burden of caregiving on families—is immeasurable.

In addition to CKD, diabetes is a major risk factor for heart disease and can lead to blindness and amputations. The rising incidence of diabetes has resulted in a new sense of urgency and has prompted the health care community to find new ways to reduce the burden on the patient, as well as encourage more aggressive research on halting progression of the disease, preventing end-stage kidney failure, and avoiding the need for dialysis.

Nearly two out of three (59%) adults in the US will be diagnosed with CKD stages 3 to 5 during their lifetimes.⁴ Some of this can be attributed to normal loss of kidney function associated with aging. However, much is due to the double burden of diabetes and hypertension.⁵ Patients with CKD require ongoing medical care, and much of it will be provided in primary care practices.⁶ This article discusses the Kidney Disease: Improving Global Outcomes (KDIGO) 2012 staging guidelines and explains which oral and injectable diabetes medications are acceptable at each stage of CKD (as defined by the glomerular filtration rate [GFR]) and how to simplify dosing.

On the next page: Staging CKD >>

STAGING CHRONIC KIDNEY DISEASE
The increasing incidence of CKD was first recognized in the 1990s. The need to care for the patient with CKD was understood, but there was a dearth of standardized management practices. Research was hindered by the many names used to identify kidney disease in the literature: from mildly elevated serum creatinine to low clearance to renal insufficiency.⁷ To rectify this, the National Kidney Foundation (NKF) convened an expert panel to develop a standard language and units of measurement to define kidney disease. The panel was tasked with developing classifications and an evaluation system for renal disease; identifying ways to attain more timely referrals to nephrology; clarifying research objectives; and delineating how to improve medication dosing. Ultimately, the main objective was to improve management of the patient with CKD.

In 2002, the NKF’s Dialysis Outcomes Quality Initiative (K/DOQI) defined the stages of kidney disease based on assessment of GFR and other kidney disease markers.⁸ GFR is a mathematical calculation for determining kidney function that corrects for loss of kidney function and incorporates the patient’s gender, race, age, and body size—the nonmodifiable risk factors known in 2001. GFR is expressed in mL/min/1.73 m². Classifying patients with CKD according to the stage of kidney disease by GFR provides a mechanism for studying the efficacy and adverse reactions of medications while correlating them to each stage. This is of particular importance when treating patients with diabetes and kidney disease, as it can help to avoid adverse effects associated with inappropriate renal dosing. Many medications are renally eliminated; as kidney function diminishes, the pharmacokinetics of drugs are altered, potentially causing toxicity and an increased risk for drug interactions. As the GFR falls, even drugs not eliminated by the kidney may have altered pharmacokinetics and pharmacodynamics that may cause renal or systemic toxicity.¹

Determining the precise method for renally dosing medication has been one of the most vigorously debated topics in CKD. Many medications were prescribed using the patient’s serum creatinine (SCr) concentration, a marker but not an exact measure of kidney disease.⁸ The Cockcroft-Gault equation, which provides an estimate of creatinine clearance (eCrCl; an alternate GFR measurement), was primarily used to determine renal drug dosing. ⁹

In 1999, a multisite study group worked to develop an equation to measure loss of kidney function in the patient with CKD.^10,11 The new formula, known as the MDRD (Modification of Diet in Renal Disease) equation, standardized the measurement of loss of kidney function and provided an estimated GFR (eGFR) that was more accurate then eCrCl. At that time, the National Kidney Disease Education Program (NKDEP) decided to use the MDRD equation to assess renal function and use the eCrCl for dosing medications. The latter was chosen because most drugs were analyzed using eCrCl.⁹ More recently, the NKDEP stated that either the eCrCl or the eGFR is acceptable for renal drug dosing. Hudson and Nyman⁹ advise that when discrepancies occur between eGFR and eCrCl, clinical judgment should be applied in the elderly and in those with extremes in body weight. In those cases, therapeutic outcomes and risk versus benefit need to be considered when determining renal dosing.

The publication of CKD stages by K/DOQI in 2002 sparked research that has uncovered additional ­criteria that might be useful in slowing the progression and improving the management of CKD. Thus, in 2013, new guidelines by KDIGO incorporating the latest international findings were published.¹ KDIGO expands the definition of CKD to include albuminuria (ALB) as an added risk factor (see Figure) and has modified K/DOQI stages 3 and 5. ALB was included because it has predictive value for progression of CKD to kidney failure. In the new classification, there are three ranges defined for ALB measured in mg/g (see Table 1). The KDIGO guidelines, like K/DOQI, classify kidney disease using a 1-to-5 staging system; however, the stage numbers in the KDIGO classification are prefaced by a G (ie, G1, G2). Also, as mentioned above, stages G3 and G5 in the KDIGO classification have been further categorized.

CKD stage 3 was divided for practical reasons. It was found that the original CKD stage 3 was too large and diverse a classification to use for research studies, tracking of hospitalizations, and precise dosing of medications.¹² K/DOQI stage 3 encompassed a GFR from 30 to 59 mL/min/1.73 m². The KDIGO guideline divides K/DOQI stage 3 into G3a (GFR of 45 to 59 mL/min/1.73 m²) and G3b (GFR of 30 to 44 mL/min/1.73 m²). The staging is then further classified by the presence or absence of ALB. The patient with a GFR of 56 mL/min/1.73 m² with ALB is in CKD stage 3aA3 (very high and nephrotic ALB) and has a greater risk for disease progression than the patient with a GFR of 56 mL/min/1.73 m² without ALB. The latter patient is in CKD stage 3aA1(optimal and mildly high ALB). Finally, many medication dosages that were acceptable at a GFR of 56 mL/min/1.73 m² had to be further adjusted for patients with a GFR of 30 mL/min/1.73 m² to avoid toxicity. This further justified the KDIGO reclassification of stage 3.

Stage 5 from K/DOQI only reflected a GFR < 15 mL/min/1.73 m². KDIGO utilizes the same GFR but splits the stage into patients not receiving dialysis and those receiving dialysis, G5 and G5D, respectively. The dosing and timing of medications due to drug dialyzability—including renal-specific medications—may be quite different for patients on dialysis as compared to those with a GFR < 15 mL/min/1.73 m² not on dialysis, making the distinction beneficial.

On the next page: Diabetes management >>

DIABETES MANAGEMENT
The treatment of patients early in diabetic CKD is often the responsibility of primary care providers, who are faced with the daunting task of addressing the renal dosing of medications. The eGFR or eCrCl needs to be utilized to adjust diabetic medications, largely to avoid hypoglycemia. To alleviate some of the difficulty, the KDIGO CKD stage and the eCrCl and eGFR for diabetic medications are delineated in Table 2.¹

Note that the guidelines are not meant to be strictly adhered to but rather are intended as a tool for managing the patient with CKD. Clinically significant patient factors need to be considered to adequately adjust drugs in CKD. Also, as new data become available from postmarketing reports, dosing adjustments, added precautions, or updated risk factors in these fragile patients may need to be incorporated into the guidelines. As new diabetic medications are approved by the FDA, kidney-specific dosing will continue to be a moving target.

While making lifestyle changes is very important early in diabetes, by the time diabetic nephropathy manifests, more aggressive action is warranted.¹² A part of the diabetic microvascular triad (along with retinopathy and neuropathy), nephropathy signals existing and irreversible organ damage. Before instituting treatment to delay diabetic nephropathy progression, an A1C goal needs to be established. The long-standing goal has been an A1C < 7%, according to guidelines established by the American Diabetes Association, or 6.5%, as recognized by the American Association of Clinical Endocrinologists.¹³ The Diabetes Control and Complications Trial (DCCT) demonstrated microvascular benefit of tight glycemic control in type 1 diabetic patients when the A1C was maintained at an average of 7.2% (versus 9.1% in conventional therapy).¹⁴ The United Kingdom Prospective Diabetes Study similarly demonstrated microvascular benefit with intensive blood glucose control in type 2 diabetic patients who were maintained at a median A1C of 7% compared to the ­conventional treatment group maintained at an A1C of 7.9%.¹⁵

In January 2013, Perkovic and colleagues, in a post hoc analysis of the ADVANCE Trial, showed that maintaining an A1C between 6.5% and 7% in a patient with diabetes and macroalbuminuria (UACR > 300 μg/mg) slows progression to kidney failure.¹⁶ The time needed to treat with intensive glucose control was five years, and the number needed to treat was 41. Thus, for every 41 patients with diabetes, CKD, and ALB whose A1C is below 7%, one will not progress to kidney failure in five years of treatment.

The five-year lead time is troublesome, however. It discourages many patients from taking steps to achieve strict glycemic control; a significant number fail to follow their diabetic diets or take their medications until the damage is done. For the elderly kidney patient with diabetes and multiple other comorbidities, aggressively managing diabetes may result in hypoglycemia.¹² On balance, the slight loss of kidney function five years hence is less problematic than the present risk for a fall and resulting hip fracture due to hypoglycemia. For this reason, the KDOQI glycemic guidelines suggest an A1C between 7% to 7.5% as the goal for those with significant comorbidities.¹²

On the next page: Diabetic medication dosing for CKD >>

DIABETIC MEDICATION DOSING FOR CKD
Glycemic control is important in delaying the progression of kidney failure in the patient with CKD. Hypoglycemic medications by definition may cause low blood glucose levels. This is of particular concern in patients with diminishing renal function, particularly the elderly and CKD patients. Antidiabetes drugs for patients in CKD stages 1 and 2 have few renal precautions, although care must be taken with metformin (see Table 3). Metformin is the drug of choice for the patient newly diagnosed with diabetes. It is inexpensive and effective, causes hypoglycemia only with intensive exercise, is taken orally, and is generally well tolerated. However, the package insert (PI) indicates stopping it in patients whose kidney disease has progressed to a GFR < 60 mL/min/1.73 m² because it causes an increased risk for lactic acidosis.^17,18

Metformin was developed prior to 1998, when the approval of the dosage regimen was tied to SCr levels rather than the presently accepted eGFR.¹⁹ As a result, the PI states that metformin should not be used in women with an SCr > 1.4 mg/dL or in men with an SCr > 1.5 mg/dL.¹⁷ However, as noted, SCr concentration alone is a very poor indicator of kidney disease (see Table 4).

Currently, the KDIGO guidelines recommend assessing the GFR for metformin dosing. When the GFR is < 60 mL/min/1.73 m², KDIGO recommends metformin be discontinued in patients with low muscle mass or serious concurrent illness, or those who are concurrently receiving renally eliminated or nephrotoxic drugs (this includes, but is not limited to, renin-angiotensin-aldosterone system [RAAS] inhibitors, NSAIDs, and diuretics).¹ In those patients who do not have the exclusions delineated above, KDIGO recommends that metformin may be continued when the GFR is > 45 mL/min/1.73 m² (stages G1 to G3a), closely monitored and reconsidered when the GFR is 30 to 44 mL/min/1.73 m² (stage G3b), and discontinued when the GFR is < 30 mL/min/1.73 m² (stages G4 to G5). The KDIGO differs from other sources that recommend metformin be stopped when the GFR is < 50 to 70 ml/min/1.73 m².^20-22 The metformin PI also recommends stopping metformin the day of or day before and two days after procedures in which radiographic iodinated contrast dye is administered, to avoid acute kidney failure. Finally, it is important to note that any combination medication formulated with metformin should be stopped when the GFR is 50 mL/min/1.73 m² because lactic acidosis has also been seen in patients taking these medications.²¹

At stage 3a (GFR, 45 to 60 mL/min/1.73 m²), the sulfonylureas require dosing changes (see Table 2). Glipizide is often the oral medication of choice during CKD, and it may be used in all stages, including dialysis, but does require renal dosing and monitoring.²³ Although the insulins are more effective than glipizide in glycemic control, many practitioners prescribe it for patients who will not accept an injectable medication. Glyburide is not used at this stage, as it can lead to complications in patients with a GFR < 60 mL/min/1.73 m², particularly hypoglycemia.

Like the sulfonylureas, the meglitinides enhance insulin secretion.^13,18 Repaglinide and nateglinide are fast-acting and need to be taken with meals. Because it is recommended that the dose of the meglitinides be omitted when a meal is skipped, they are good for patients who eat meals irregularly. The meglitinides require initiation at a low dose in severe renal insufficiency.

The thiazolidinediones (TDZs) are used with caution in patients with CKD, not only because of controversial adverse cardiac effects, but also for peripheral edema that interferes with CKD management.¹⁸ The TDZs (rosiglitazone and pioglitazone) require no dose adjustment and appear to have a beneficial effect in obese patients and in patients who experienced weight gain with other hypoglycemic agents.

The newest of the diabetic drugs, canagliflozin, requires a working kidney to be effective.^24,25 It is used with precaution when GFR is 45 to 59 mL/min/1.73 m² at stage G3a and is contraindicated when GFR is < 45 mL/min/1.73 m² in stage G3b. Canagliflozin's effects are felt in early segment of proximal convoluted tubule where it blocks sodium-glucose co-transporter 2, thereby lowering reabsorption of glucose and increasing excretion of glucose in the urine. In practical terms, it can affect kidney function and, due to the inhibition of the glucose-sodium pathway, induce hyperkalemia.²⁵ Monitoring of both the GFR and potassium is vital when administering this drug. As data from postmarketing reports come in and practitioners gain experience, use of canaglifozin in the patient with CKD will be further delineated.

At a GFR between 30 and 45 mL/min/1.73 m², CKD stage 3b, alpha-glucosidase inhibitors acarbose and miglitol are contraindicated. The GLP-1 mimetic liraglutide needs to be avoided when the GFR is < 60 mL/min/1.73 m² (stage G3a). While exenatide requires close monitoring and caution when the GFR is < 50 mL/min/1.73 m², it needs to be discontinued at a GFR
< 30 mL/min/1.73 m² (stage 4). DPP-4 inhibitors become problematic at a GFR < 60 mL/min/
1.73 m² beginning at stage G3a. All except the DPP-4 inhibitor linagliptin require dosing changes as the loss of GFR continues. Exact dosing adjustments for the drug categories are found in Table 2.

In stage 4 CKD (GFR, 15 to 30 mL/min/1.73 m²), more medications are eliminated from consideration. The amylinomimetic pramlintide cannot be used for a patient with a GFR < 15 mL/min/1.73 m² (see Table 2). Both GLP-1 mimetics, exenatide and liraglutide, must be stopped when the GFR drops to 30 mL/min/1.73 m². DDP-4 inhibitors need to be dosed according to the GFR but are still acceptable at CKD stage 4. Ideally, patients with diabetes should be referred to nephrology at stage 3, but referral is vital by stage 4.

Although oral medications are not as commonly prescribed as insulins at CKD stages 4 and 5, in stage 5 CKD (GFR, < 15 mL/min/1.73 m²), a limited number of oral diabetic agents are available (see Table 2). Glipizide is inexpensive and generic; therefore, it is the most commonly used oral medication in this stage. A recent study found linagliptin was safe to use throughout all stages of kidney disease without adjustment for GFR.²⁶ While nateglinide is still acceptable in stage 5, it is used less frequently than glipizide.

As the kidney fails, the need for diabetic medications also decreases and dosing needs to take into account any residual renal function. Because glucose is produced in the proximal tubules of the kidney (known as gluconeogenesis), the level of glucose falls as the kidney fails. Excretion of insulin and other diabetes medications decreases, and thus the amount of insulin remaining in circulation is increased. This, in combination with the reduced glucose, increases the patient’s risk for hypoglycemia.¹³ Although protocols for medication adjustment are tied to GFR levels, close follow-up by the practitioner is required because GFR is not a reliable measure of kidney function when creatinine levels change rapidly.²⁷

On the next page: Insulins >>

INSULINS
Many patients start insulin prior to reaching CKD stage 4 or 5 because it is the best choice for managing diabetes through all stages of CKD.²⁸ Insulin can slow the progression to kidney failure by providing better A1C control, and all patients should be encouraged to start insulin as soon as it is appropriate.²⁹ Many patients are very reluctant to use injectables, but by stage 5 they are willing to use insulin if it will slow progression of disease and help prevent the need for dialysis.²

The long-acting basal insulins (detemir and glargine) or a combination of basal and oral medication can work quite well at stage 5. Many patients will require a low dose of the long-acting basal insulins and short-acting insulin with meals. As the kidney fails, the short-acting doses can often be discontinued.²⁹ A protocol that mixes long- and short-acting insulin with the largest meal is very effective at this stage. The long-acting insulin dose will need to be decreased as the kidney continues to fail. The American College of Physicians protocol for insulin suggests a 25% decrease at stage 4 and a 50% decrease at stage 5. Each patient is different, however, and dosing must be determined individually.²³

On the next page: Patient education >>

PATIENT EDUCATION
There is a correlation between patients who participate in kidney disease education programs and a decrease in the progression of kidney disease (see “Case Study: The Recalcitrant Patient With Diabetes"). These classes can be taught by a physician, physician assistant, nurse practitioner, or clinical nurse specialist. To alleviate the burden of end-stage kidney disease in the US, Medicare Part B pays for six outpatient kidney disease patient education classes per lifetime of the beneficiary.³⁰

Since 2010, when kidney disease education programs were rolled out, over 10,000 classes have been taught.³¹ Many practitioners report that patients are more compliant and understand their disease better when they attend classes.³² Data are being gathered to determine the effectiveness of the classes on patient outcomes. 

On the next page: Conclusion >>

CONCLUSION
Knowing how to manage the patient with both diabetes and kidney disease is increasingly vital as this patient population grows. Much of the management of these patients will fall to the primary care practitioner.⁶ The most effective way to start treatment is to identify which CKD stage the patient fits into.

Once the patient is properly categorized, safe and effective diabetic medications can be selected and dosed according to stage. Although the new classification system may be difficult to incorporate into some electronic medical record systems and practitioner behavior, ultimately, it will allow safer management of the patient with CKD and a better predictive power of outcome. 

SAN DIEGO – Over the past 20 years, death rates from prostate cancer in the United States have declined by 39%, due largely to early detection and/or improved treatment, according to Dr. Peter R. Carroll.

"It’s important to realize that this accounts for 20% of the decrease in cancer-specific deaths in men," said Dr. Carroll, professor and chair of the department of urology at the University of California, San Francisco. "However, the Achilles’ heel of PSA [prostate-specific antigen] testing is that it does so at the risk of overdetection – detecting disease that would not have become clinically apparent over a patient’s lifetime if left untreated. In this country, detection and treatment are too tightly linked."

In addition, widespread use of serum PSA has resulted in a "dramatic stage and grade shift, with most cancers currently being detected of limited cancer grade and stage," he said. Data from his colleagues at UCSF found that between 1990 and 2012, age-adjusted death rates from prostate cancer decreased 3.96% in North America yet increased 41% worldwide. In the United States, men with nonpalpable PSA-driven cancer comprise the largest segment of prostate cancer patients. "We think that we’ve seen a leveling of overdetection, but I think we’ll see another round of overdetection, because of a lowering PSA threshold to prompt biopsy, aggressive rescreening, the use of PSA velocity at low PSA values to prompt biopsy, and the use of saturation biopsies," he said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

In 2012 the U.S. Preventive Services Task Force came out against prostate cancer screening, classifying it as a grade D recommendation (Ann. Intern. Med. 2012;157[2]:120-34). The magnitude of overdetection varies with time period, age, comorbidities, region, definition, and screening practices and is thought to range between 2% and 67%, Dr. Carroll said. "I think a good number is somewhere between 35% and 40%." A recently published nomogram for predicting overdiagnosis found that depending on a man’s age, Gleason score, and PSA level, the likelihood that his tumor has been overdiagnosed ranges from 2.9% to 88.1% (J. Natl. Cancer Inst. 2014 [doi:10.1093/jnci.djt367]).

If prostate cancer screening is to be undertaken, "it should only be done recognizing that selective, rather than indiscriminate, treatment should follow," Dr. Carroll said. "Such an approach has been shown to reduce mortality while managing many with active surveillance in lieu of immediate treatment" (Lancet Oncol. 2010;11[8]:725-32). At UCSF, where more than 1,000 men are on active surveillance, the 5-year treatment-free survival is 65%, the 5-year overall survival is 97%, and the 5-year prostate cancer–specific survival is 100%. "The window of opportunity for treatment appears to be open for a long period of time," he said.

Potential solutions Dr. Carroll proposed to decrease the rates of overdetection include:

• Reducing the treatment of low-risk tumors.

• Identifying high-risk populations and targeting prevention and screening efforts to those populations.

• Developing new screening markers.

• Developing clinical and patient tools to support informed decision making about prevention, screening, biopsy, and treatment.

• Changing screening guidelines.

"The single biggest predictor of risk is a baseline PSA. It trumps ethnicity and family history," Dr. Carroll said. "I think there’s a strong rationale for a baseline screening between ages 45 and 55. If you screen beyond age 70, you increase the risk of overdetection. But if you stop screening you also increase the mortality. So beyond age 70 you want to individualize, consider screening only in those with a long life expectancy, and perhaps change the rationale for biopsy. Digital rectal examination in my mind is optional as a primary screening maneuver. Screening can be done at 1- to 2-year intervals. One thing we need to get away from is using PSA velocity at low PSA levels. That drives overdetection quite a bit."

Dr. Carroll disclosed that he has received honoraria, research support, and/or consulting fees from Genomic Health, Intuitive, Janssen, and Myriad.

[email protected]

SAN DIEGO – Over the past 20 years, death rates from prostate cancer in the United States have declined by 39%, due largely to early detection and/or improved treatment, according to Dr. Peter R. Carroll.

"It’s important to realize that this accounts for 20% of the decrease in cancer-specific deaths in men," said Dr. Carroll, professor and chair of the department of urology at the University of California, San Francisco. "However, the Achilles’ heel of PSA [prostate-specific antigen] testing is that it does so at the risk of overdetection – detecting disease that would not have become clinically apparent over a patient’s lifetime if left untreated. In this country, detection and treatment are too tightly linked."

In addition, widespread use of serum PSA has resulted in a "dramatic stage and grade shift, with most cancers currently being detected of limited cancer grade and stage," he said. Data from his colleagues at UCSF found that between 1990 and 2012, age-adjusted death rates from prostate cancer decreased 3.96% in North America yet increased 41% worldwide. In the United States, men with nonpalpable PSA-driven cancer comprise the largest segment of prostate cancer patients. "We think that we’ve seen a leveling of overdetection, but I think we’ll see another round of overdetection, because of a lowering PSA threshold to prompt biopsy, aggressive rescreening, the use of PSA velocity at low PSA values to prompt biopsy, and the use of saturation biopsies," he said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

In 2012 the U.S. Preventive Services Task Force came out against prostate cancer screening, classifying it as a grade D recommendation (Ann. Intern. Med. 2012;157[2]:120-34). The magnitude of overdetection varies with time period, age, comorbidities, region, definition, and screening practices and is thought to range between 2% and 67%, Dr. Carroll said. "I think a good number is somewhere between 35% and 40%." A recently published nomogram for predicting overdiagnosis found that depending on a man’s age, Gleason score, and PSA level, the likelihood that his tumor has been overdiagnosed ranges from 2.9% to 88.1% (J. Natl. Cancer Inst. 2014 [doi:10.1093/jnci.djt367]).

If prostate cancer screening is to be undertaken, "it should only be done recognizing that selective, rather than indiscriminate, treatment should follow," Dr. Carroll said. "Such an approach has been shown to reduce mortality while managing many with active surveillance in lieu of immediate treatment" (Lancet Oncol. 2010;11[8]:725-32). At UCSF, where more than 1,000 men are on active surveillance, the 5-year treatment-free survival is 65%, the 5-year overall survival is 97%, and the 5-year prostate cancer–specific survival is 100%. "The window of opportunity for treatment appears to be open for a long period of time," he said.

Potential solutions Dr. Carroll proposed to decrease the rates of overdetection include:

• Reducing the treatment of low-risk tumors.

• Identifying high-risk populations and targeting prevention and screening efforts to those populations.

• Developing new screening markers.

• Developing clinical and patient tools to support informed decision making about prevention, screening, biopsy, and treatment.

• Changing screening guidelines.

"The single biggest predictor of risk is a baseline PSA. It trumps ethnicity and family history," Dr. Carroll said. "I think there’s a strong rationale for a baseline screening between ages 45 and 55. If you screen beyond age 70, you increase the risk of overdetection. But if you stop screening you also increase the mortality. So beyond age 70 you want to individualize, consider screening only in those with a long life expectancy, and perhaps change the rationale for biopsy. Digital rectal examination in my mind is optional as a primary screening maneuver. Screening can be done at 1- to 2-year intervals. One thing we need to get away from is using PSA velocity at low PSA levels. That drives overdetection quite a bit."

Dr. Carroll disclosed that he has received honoraria, research support, and/or consulting fees from Genomic Health, Intuitive, Janssen, and Myriad.

[email protected]

SAN DIEGO – Over the past 20 years, death rates from prostate cancer in the United States have declined by 39%, due largely to early detection and/or improved treatment, according to Dr. Peter R. Carroll.

"It’s important to realize that this accounts for 20% of the decrease in cancer-specific deaths in men," said Dr. Carroll, professor and chair of the department of urology at the University of California, San Francisco. "However, the Achilles’ heel of PSA [prostate-specific antigen] testing is that it does so at the risk of overdetection – detecting disease that would not have become clinically apparent over a patient’s lifetime if left untreated. In this country, detection and treatment are too tightly linked."

In addition, widespread use of serum PSA has resulted in a "dramatic stage and grade shift, with most cancers currently being detected of limited cancer grade and stage," he said. Data from his colleagues at UCSF found that between 1990 and 2012, age-adjusted death rates from prostate cancer decreased 3.96% in North America yet increased 41% worldwide. In the United States, men with nonpalpable PSA-driven cancer comprise the largest segment of prostate cancer patients. "We think that we’ve seen a leveling of overdetection, but I think we’ll see another round of overdetection, because of a lowering PSA threshold to prompt biopsy, aggressive rescreening, the use of PSA velocity at low PSA values to prompt biopsy, and the use of saturation biopsies," he said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

In 2012 the U.S. Preventive Services Task Force came out against prostate cancer screening, classifying it as a grade D recommendation (Ann. Intern. Med. 2012;157[2]:120-34). The magnitude of overdetection varies with time period, age, comorbidities, region, definition, and screening practices and is thought to range between 2% and 67%, Dr. Carroll said. "I think a good number is somewhere between 35% and 40%." A recently published nomogram for predicting overdiagnosis found that depending on a man’s age, Gleason score, and PSA level, the likelihood that his tumor has been overdiagnosed ranges from 2.9% to 88.1% (J. Natl. Cancer Inst. 2014 [doi:10.1093/jnci.djt367]).

If prostate cancer screening is to be undertaken, "it should only be done recognizing that selective, rather than indiscriminate, treatment should follow," Dr. Carroll said. "Such an approach has been shown to reduce mortality while managing many with active surveillance in lieu of immediate treatment" (Lancet Oncol. 2010;11[8]:725-32). At UCSF, where more than 1,000 men are on active surveillance, the 5-year treatment-free survival is 65%, the 5-year overall survival is 97%, and the 5-year prostate cancer–specific survival is 100%. "The window of opportunity for treatment appears to be open for a long period of time," he said.

Potential solutions Dr. Carroll proposed to decrease the rates of overdetection include:

• Reducing the treatment of low-risk tumors.

• Identifying high-risk populations and targeting prevention and screening efforts to those populations.

• Developing new screening markers.

• Developing clinical and patient tools to support informed decision making about prevention, screening, biopsy, and treatment.

• Changing screening guidelines.

"The single biggest predictor of risk is a baseline PSA. It trumps ethnicity and family history," Dr. Carroll said. "I think there’s a strong rationale for a baseline screening between ages 45 and 55. If you screen beyond age 70, you increase the risk of overdetection. But if you stop screening you also increase the mortality. So beyond age 70 you want to individualize, consider screening only in those with a long life expectancy, and perhaps change the rationale for biopsy. Digital rectal examination in my mind is optional as a primary screening maneuver. Screening can be done at 1- to 2-year intervals. One thing we need to get away from is using PSA velocity at low PSA levels. That drives overdetection quite a bit."

Dr. Carroll disclosed that he has received honoraria, research support, and/or consulting fees from Genomic Health, Intuitive, Janssen, and Myriad.

[email protected]

SAN DIEGO – Now is the time for researchers and clinicians to examine strategies and interventions for preventing prostate cancer, in the opinion of Dr. Peter H. Gann.

"In the last 12 years or so we have seen a litany of failure with regard to prostate cancer prevention in trials with clinical endpoints," Dr. Gann, professor and director of pathology research at the University of Illinois at Chicago, said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

"With vitamin E supplementation, for example, we see a possible increased risk of prostate cancer, no effect with selenium and possibly and increased risk of diabetes, no effect from soy, and the potential impact of green tea polyphenols is unresolved."

Moreover, it’s conceivable that you might need to screen 1,500 average-risk patients to prevent one prostate cancer from occurring. "That means that preventive agents don’t have to be safe; they have to be incredibly safe if they’re going to be used in this way," Dr. Gann said.

He proposed five ways to advance prostate cancer prevention efforts:

• Develop better preclinical models. Canadian investigators have reported success with generating high fidelity patient-derived xenografts for accelerating prostate cancer discovery and drug development (Cancer Res. 2013 Dec. 19 [doi:10.1158/0008-5472.CAN-13-2921-T]). "One of the incredible things about this is that the transplantable cells can be obtained from needle biopsies," said Dr. Gann, who was not involved with the study. "Architecture and protein marker expression is preserved in these xenografts, as are other molecular characteristics of the tumor, which is fantastic."

An unrelated Australian study demonstrated that it’s possible to generate xenografts of the earlier low-to-moderate grade, localized tumors (Nat. Protoc. 2013;8:836-48). "Why is this exciting for us in prevention? More and more we’re thinking about not how to prevent things from the initiation stage but rather in terms of progression," Dr. Gann explained. "Being able to have individualized samples that we can study versus agents that may inhibit progression is a major opportunity, I think."

• Improve clinical trial design and infrastructure. The existing networks for prostate cancer prevention trials are largely undeveloped, unlike cooperative group networks available for therapeutic trials, according to Dr. Gann. "Many investigators with promising ideas do not have access to the clinical infrastructure or funding sources they need for translational research," he noted. "Moreover, the pros and cons of various available designs for phase II and III trials have not been adequately debated, amidst a shifting landscape in which some designs become less feasible as others become more feasible."

• Develop better risk stratification and patient targeting. What if clinicians could do a better job of sorting out who is truly at risk for prostate cancer? "Active surveillance cohorts are the most promising opportunity we have for prevention trials involving low-risk interventions," Dr. Gann said. Cumulative results from genome-wide association studies and the expected results from sequencing studies capable of identifying rare genetic variants with high penetrance hold promise for identifying populations for whom preventive strategies would have the most benefit, he added.

• Develop better interventional agents. To date, "I think we’ve taken a haphazard approach to identifying agents for prostate cancer prevention," Dr. Gann said. "With preclinical models lacking, sometimes they’re not vetted very well, either. We’re not going to be able to develop a rational approach to preventing prostate cancer until we have a better idea of how it all comes about. One idea is to do high throughput cell assay based drug screening, which we do for therapeutics but not for chemopreventive agents. The question is, do we have the right libraries of compounds and do we have the right readouts? I suggest that we can start with compounds, especially dietary agents that could inhibit 5-alpha reductase. We also need to think beyond a pharmacologic approach: studying the effects of diet and physical activity, for example."

• Establish better intermediate endpoints for phase II trials. A lack of intermediate endpoint biomarkers (IEBs) for phase II trials is creating a "phase II bottleneck," he said. "There is an ongoing explosion of opportunities provided by new biotechnology and computational methods, but converting these opportunities into validated IEBs for trials will take thought and planning."

Dr. Gann disclosed that he has received grant support from GlaxoSmithKline.

[email protected]

SAN DIEGO – Now is the time for researchers and clinicians to examine strategies and interventions for preventing prostate cancer, in the opinion of Dr. Peter H. Gann.

"In the last 12 years or so we have seen a litany of failure with regard to prostate cancer prevention in trials with clinical endpoints," Dr. Gann, professor and director of pathology research at the University of Illinois at Chicago, said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

"With vitamin E supplementation, for example, we see a possible increased risk of prostate cancer, no effect with selenium and possibly and increased risk of diabetes, no effect from soy, and the potential impact of green tea polyphenols is unresolved."

Moreover, it’s conceivable that you might need to screen 1,500 average-risk patients to prevent one prostate cancer from occurring. "That means that preventive agents don’t have to be safe; they have to be incredibly safe if they’re going to be used in this way," Dr. Gann said.

He proposed five ways to advance prostate cancer prevention efforts:

• Develop better preclinical models. Canadian investigators have reported success with generating high fidelity patient-derived xenografts for accelerating prostate cancer discovery and drug development (Cancer Res. 2013 Dec. 19 [doi:10.1158/0008-5472.CAN-13-2921-T]). "One of the incredible things about this is that the transplantable cells can be obtained from needle biopsies," said Dr. Gann, who was not involved with the study. "Architecture and protein marker expression is preserved in these xenografts, as are other molecular characteristics of the tumor, which is fantastic."

An unrelated Australian study demonstrated that it’s possible to generate xenografts of the earlier low-to-moderate grade, localized tumors (Nat. Protoc. 2013;8:836-48). "Why is this exciting for us in prevention? More and more we’re thinking about not how to prevent things from the initiation stage but rather in terms of progression," Dr. Gann explained. "Being able to have individualized samples that we can study versus agents that may inhibit progression is a major opportunity, I think."

• Improve clinical trial design and infrastructure. The existing networks for prostate cancer prevention trials are largely undeveloped, unlike cooperative group networks available for therapeutic trials, according to Dr. Gann. "Many investigators with promising ideas do not have access to the clinical infrastructure or funding sources they need for translational research," he noted. "Moreover, the pros and cons of various available designs for phase II and III trials have not been adequately debated, amidst a shifting landscape in which some designs become less feasible as others become more feasible."

• Develop better risk stratification and patient targeting. What if clinicians could do a better job of sorting out who is truly at risk for prostate cancer? "Active surveillance cohorts are the most promising opportunity we have for prevention trials involving low-risk interventions," Dr. Gann said. Cumulative results from genome-wide association studies and the expected results from sequencing studies capable of identifying rare genetic variants with high penetrance hold promise for identifying populations for whom preventive strategies would have the most benefit, he added.

• Develop better interventional agents. To date, "I think we’ve taken a haphazard approach to identifying agents for prostate cancer prevention," Dr. Gann said. "With preclinical models lacking, sometimes they’re not vetted very well, either. We’re not going to be able to develop a rational approach to preventing prostate cancer until we have a better idea of how it all comes about. One idea is to do high throughput cell assay based drug screening, which we do for therapeutics but not for chemopreventive agents. The question is, do we have the right libraries of compounds and do we have the right readouts? I suggest that we can start with compounds, especially dietary agents that could inhibit 5-alpha reductase. We also need to think beyond a pharmacologic approach: studying the effects of diet and physical activity, for example."

• Establish better intermediate endpoints for phase II trials. A lack of intermediate endpoint biomarkers (IEBs) for phase II trials is creating a "phase II bottleneck," he said. "There is an ongoing explosion of opportunities provided by new biotechnology and computational methods, but converting these opportunities into validated IEBs for trials will take thought and planning."

Dr. Gann disclosed that he has received grant support from GlaxoSmithKline.

[email protected]

SAN DIEGO – Now is the time for researchers and clinicians to examine strategies and interventions for preventing prostate cancer, in the opinion of Dr. Peter H. Gann.

"In the last 12 years or so we have seen a litany of failure with regard to prostate cancer prevention in trials with clinical endpoints," Dr. Gann, professor and director of pathology research at the University of Illinois at Chicago, said during a conference sponsored by the American Association for Cancer Research and the Prostate Cancer Foundation.

"With vitamin E supplementation, for example, we see a possible increased risk of prostate cancer, no effect with selenium and possibly and increased risk of diabetes, no effect from soy, and the potential impact of green tea polyphenols is unresolved."

Moreover, it’s conceivable that you might need to screen 1,500 average-risk patients to prevent one prostate cancer from occurring. "That means that preventive agents don’t have to be safe; they have to be incredibly safe if they’re going to be used in this way," Dr. Gann said.

He proposed five ways to advance prostate cancer prevention efforts:

• Develop better preclinical models. Canadian investigators have reported success with generating high fidelity patient-derived xenografts for accelerating prostate cancer discovery and drug development (Cancer Res. 2013 Dec. 19 [doi:10.1158/0008-5472.CAN-13-2921-T]). "One of the incredible things about this is that the transplantable cells can be obtained from needle biopsies," said Dr. Gann, who was not involved with the study. "Architecture and protein marker expression is preserved in these xenografts, as are other molecular characteristics of the tumor, which is fantastic."

An unrelated Australian study demonstrated that it’s possible to generate xenografts of the earlier low-to-moderate grade, localized tumors (Nat. Protoc. 2013;8:836-48). "Why is this exciting for us in prevention? More and more we’re thinking about not how to prevent things from the initiation stage but rather in terms of progression," Dr. Gann explained. "Being able to have individualized samples that we can study versus agents that may inhibit progression is a major opportunity, I think."

• Improve clinical trial design and infrastructure. The existing networks for prostate cancer prevention trials are largely undeveloped, unlike cooperative group networks available for therapeutic trials, according to Dr. Gann. "Many investigators with promising ideas do not have access to the clinical infrastructure or funding sources they need for translational research," he noted. "Moreover, the pros and cons of various available designs for phase II and III trials have not been adequately debated, amidst a shifting landscape in which some designs become less feasible as others become more feasible."

• Develop better risk stratification and patient targeting. What if clinicians could do a better job of sorting out who is truly at risk for prostate cancer? "Active surveillance cohorts are the most promising opportunity we have for prevention trials involving low-risk interventions," Dr. Gann said. Cumulative results from genome-wide association studies and the expected results from sequencing studies capable of identifying rare genetic variants with high penetrance hold promise for identifying populations for whom preventive strategies would have the most benefit, he added.

• Develop better interventional agents. To date, "I think we’ve taken a haphazard approach to identifying agents for prostate cancer prevention," Dr. Gann said. "With preclinical models lacking, sometimes they’re not vetted very well, either. We’re not going to be able to develop a rational approach to preventing prostate cancer until we have a better idea of how it all comes about. One idea is to do high throughput cell assay based drug screening, which we do for therapeutics but not for chemopreventive agents. The question is, do we have the right libraries of compounds and do we have the right readouts? I suggest that we can start with compounds, especially dietary agents that could inhibit 5-alpha reductase. We also need to think beyond a pharmacologic approach: studying the effects of diet and physical activity, for example."

• Establish better intermediate endpoints for phase II trials. A lack of intermediate endpoint biomarkers (IEBs) for phase II trials is creating a "phase II bottleneck," he said. "There is an ongoing explosion of opportunities provided by new biotechnology and computational methods, but converting these opportunities into validated IEBs for trials will take thought and planning."

Dr. Gann disclosed that he has received grant support from GlaxoSmithKline.

[email protected]

DALLAS – High-potency statin therapy given post acute coronary syndrome did not raise serum creatinine or cause more risk of acute kidney injury than low-potency statin regimens did in a new analysis of two published landmark randomized clinical trials.

"Considering the recently updated AHA/American College of Cardiology lipid guidelines, which call for the use of high-potency statins in millions more patients, these findings provide important reassurance that a high-potency statin regimen will not increase the incidence of adverse renal events," Dr. Amy Sarma said in presenting the study results at the American Heart Association scientific sessions.

She noted that a recent Canadian observational study utilizing Canadian and U.S. administrative databases totaling more than 2 million patients over age 40 who were newly placed on statin therapy showed an adjusted 1.34-fold increased rate of hospitalization for acute kidney injury within the first 120 days in those on a high- as compared to a low-potency regimen, and a 1.11-fold increased risk beyond 120 days through the end of the first year (BMJ 2013;346:f880). Both risk elevations were statistically significant.

However, observational studies such as this are prone to bias in the form of potentially crucial differences between the patients given a prescription for statins and those who aren’t. For this reason, Dr. Sarma and her coinvestigators turned for guidance to two randomized trials of high- versus low-dose statins, since this study design obviates the risks of confounding. The trials were PROVE IT-TIMI 22 (N. Engl. J. Med. 2004;350:1495-504) and the A-to-Z trial (JAMA 2004;292:1307-16).

PROVE IT included 4,122 patients randomized within 10 days post ACS to standard background therapy plus either pravastatin at 40 mg/day or atorvastatin at 80 mg/day. A-to-Z involved 4,497 patients who were placed on simvastatin at either 20 or 80 mg/day within 5 days post ACS. Both trials had a median follow-up of 2 years, and both featured serial measurements of serum creatinine. Two-thirds of subjects in PROVE IT had a baseline estimated glomerular filtration rate below 90 mL/min per 1.73 m², while two-thirds of those in A-to-Z had a baseline eGFR less than 60, noted Dr. Sarma of Brigham and Women’s Hospital, Boston.

In both studies, mean serum creatinine rose equally during the first 30 days of statin therapy, regardless of treatment potency, and then levels declined. In PROVE IT, for example, serum creatinine in the pravastatin and atorvastatin arms rose by 0.96% and 0.97% above baseline, respectively, at 30 days. Values then dropped by 2.88% and 3.85% from baseline at 4 months, and by 3.88% and 5.83% at 16 months.

In both studies, there was no difference between the high- and low-potency statin groups in the incidence of any increase in serum creatinine of at least 1.5-fold, 2.0-fold, or 3.0-fold greater than baseline. In other words, there was no hint of a safety signal, she added.

Dr. Sarma reported having no financial conflicts of interest.

[email protected]

DALLAS – High-potency statin therapy given post acute coronary syndrome did not raise serum creatinine or cause more risk of acute kidney injury than low-potency statin regimens did in a new analysis of two published landmark randomized clinical trials.

"Considering the recently updated AHA/American College of Cardiology lipid guidelines, which call for the use of high-potency statins in millions more patients, these findings provide important reassurance that a high-potency statin regimen will not increase the incidence of adverse renal events," Dr. Amy Sarma said in presenting the study results at the American Heart Association scientific sessions.

She noted that a recent Canadian observational study utilizing Canadian and U.S. administrative databases totaling more than 2 million patients over age 40 who were newly placed on statin therapy showed an adjusted 1.34-fold increased rate of hospitalization for acute kidney injury within the first 120 days in those on a high- as compared to a low-potency regimen, and a 1.11-fold increased risk beyond 120 days through the end of the first year (BMJ 2013;346:f880). Both risk elevations were statistically significant.

However, observational studies such as this are prone to bias in the form of potentially crucial differences between the patients given a prescription for statins and those who aren’t. For this reason, Dr. Sarma and her coinvestigators turned for guidance to two randomized trials of high- versus low-dose statins, since this study design obviates the risks of confounding. The trials were PROVE IT-TIMI 22 (N. Engl. J. Med. 2004;350:1495-504) and the A-to-Z trial (JAMA 2004;292:1307-16).

PROVE IT included 4,122 patients randomized within 10 days post ACS to standard background therapy plus either pravastatin at 40 mg/day or atorvastatin at 80 mg/day. A-to-Z involved 4,497 patients who were placed on simvastatin at either 20 or 80 mg/day within 5 days post ACS. Both trials had a median follow-up of 2 years, and both featured serial measurements of serum creatinine. Two-thirds of subjects in PROVE IT had a baseline estimated glomerular filtration rate below 90 mL/min per 1.73 m², while two-thirds of those in A-to-Z had a baseline eGFR less than 60, noted Dr. Sarma of Brigham and Women’s Hospital, Boston.

In both studies, mean serum creatinine rose equally during the first 30 days of statin therapy, regardless of treatment potency, and then levels declined. In PROVE IT, for example, serum creatinine in the pravastatin and atorvastatin arms rose by 0.96% and 0.97% above baseline, respectively, at 30 days. Values then dropped by 2.88% and 3.85% from baseline at 4 months, and by 3.88% and 5.83% at 16 months.

In both studies, there was no difference between the high- and low-potency statin groups in the incidence of any increase in serum creatinine of at least 1.5-fold, 2.0-fold, or 3.0-fold greater than baseline. In other words, there was no hint of a safety signal, she added.

Dr. Sarma reported having no financial conflicts of interest.

[email protected]

DALLAS – High-potency statin therapy given post acute coronary syndrome did not raise serum creatinine or cause more risk of acute kidney injury than low-potency statin regimens did in a new analysis of two published landmark randomized clinical trials.

"Considering the recently updated AHA/American College of Cardiology lipid guidelines, which call for the use of high-potency statins in millions more patients, these findings provide important reassurance that a high-potency statin regimen will not increase the incidence of adverse renal events," Dr. Amy Sarma said in presenting the study results at the American Heart Association scientific sessions.

She noted that a recent Canadian observational study utilizing Canadian and U.S. administrative databases totaling more than 2 million patients over age 40 who were newly placed on statin therapy showed an adjusted 1.34-fold increased rate of hospitalization for acute kidney injury within the first 120 days in those on a high- as compared to a low-potency regimen, and a 1.11-fold increased risk beyond 120 days through the end of the first year (BMJ 2013;346:f880). Both risk elevations were statistically significant.

However, observational studies such as this are prone to bias in the form of potentially crucial differences between the patients given a prescription for statins and those who aren’t. For this reason, Dr. Sarma and her coinvestigators turned for guidance to two randomized trials of high- versus low-dose statins, since this study design obviates the risks of confounding. The trials were PROVE IT-TIMI 22 (N. Engl. J. Med. 2004;350:1495-504) and the A-to-Z trial (JAMA 2004;292:1307-16).

PROVE IT included 4,122 patients randomized within 10 days post ACS to standard background therapy plus either pravastatin at 40 mg/day or atorvastatin at 80 mg/day. A-to-Z involved 4,497 patients who were placed on simvastatin at either 20 or 80 mg/day within 5 days post ACS. Both trials had a median follow-up of 2 years, and both featured serial measurements of serum creatinine. Two-thirds of subjects in PROVE IT had a baseline estimated glomerular filtration rate below 90 mL/min per 1.73 m², while two-thirds of those in A-to-Z had a baseline eGFR less than 60, noted Dr. Sarma of Brigham and Women’s Hospital, Boston.

In both studies, mean serum creatinine rose equally during the first 30 days of statin therapy, regardless of treatment potency, and then levels declined. In PROVE IT, for example, serum creatinine in the pravastatin and atorvastatin arms rose by 0.96% and 0.97% above baseline, respectively, at 30 days. Values then dropped by 2.88% and 3.85% from baseline at 4 months, and by 3.88% and 5.83% at 16 months.

In both studies, there was no difference between the high- and low-potency statin groups in the incidence of any increase in serum creatinine of at least 1.5-fold, 2.0-fold, or 3.0-fold greater than baseline. In other words, there was no hint of a safety signal, she added.

Dr. Sarma reported having no financial conflicts of interest.

[email protected]