Contributor disclosures are reviewed for conflicts of interest by the editorial group. When found, these are addressed by vetting through a multi-level review process, and through requirements for references to be provided to support the content. Appropriately referenced content is required of all authors and must conform to UpToDate standards of evidence.
INTRODUCTION — Chronic kidney disease (CKD) is associated with insulin resistance and, in advanced CKD, decreased insulin degradation. The latter can lead to a marked decrease in insulin requirement or even the cessation of insulin therapy in patients with type 2 diabetes. Both of these abnormalities are at least partially reversed with the institution of dialysis. (See "Carbohydrate and insulin metabolism in chronic kidney disease".)
Because of the uncertainty in predicting insulin requirements, careful individualized therapy is essential among patients who have advanced CKD or are initiating dialysis.
The insulin requirement in any given patient depends upon the net balance between improving tissue sensitivity and restoring normal hepatic insulin metabolism. In addition, among patients on peritoneal dialysis, glucose contained in peritoneal dialysate tends to increase the need for diabetes therapy. Changes in dietary intake and exercise (ie, reduced intake due to anorexia prior to starting dialysis) can also affect the response to administered insulin. Furthermore, the uremic environment can affect methods used to assess glycemic control, and the metabolism of most oral diabetes agents is prolonged, making them more difficult to use.
This topic reviews glycemic targets, methods of monitoring glycemic control, and suggested treatment regimens for patients on hemodialysis and peritoneal dialysis. The treatment of diabetes in kidney transplant recipients is discussed elsewhere (see "New-onset diabetes after transplant (NODAT) in renal transplant recipients" and "New-onset diabetes after transplant (NODAT) in renal transplant recipients", section on 'Treatment'). Monitoring and maintaining glycemic control in the general population is discussed elsewhere. (See "Overview of medical care in adults with diabetes mellitus", section on 'Glycemic control'.)
MONITORING GLYCEMIC CONTROL
Glycated hemoglobin (A1C) — We monitor glycemic control in patients with diabetes and predialysis chronic kidney disease (CKD) or end-stage renal disease (ESRD) as we do in patients with diabetes and normal kidney function. Thus, we use serial measurements (two to four times yearly) of glycated hemoglobin (hemoglobin A1C, A1C) to assess chronic glycemic control in diabetic patients with predialysis CKD or ESRD. (See "Overview of medical care in adults with diabetes mellitus", section on 'Monitoring and target A1C'.)
However, A1C may not be as accurate among ESRD patients as in the general population due to biological and patient-specific factors. Some of the methods used to measure A1C, such as agar gel electrophoresis, are affected by ESRD. This is due, in part, to analytical interference from carbamylated hemoglobin formed in the presence of elevated concentrations of urea, leading to false elevations in the A1C level. The Boronate-agarose affinity chromatography [1,2] and the thiobarbituric acid method  are techniques for analyzing A1C that can be used reliably in ESRD. (See "Estimation of blood glucose control in diabetes mellitus", section on 'Glycated hemoglobin'.)
Other factors that affect the accuracy of these assays in ESRD include reduced red blood cell life span, recent transfusion, iron deficiency, accelerated erythropoiesis due to administration of erythropoietin, and metabolic acidosis. As a result, A1C values may be falsely elevated or decreased in those with CKD [1,3-9]. It is important to be aware of the specific assay used in each dialysis facility and the extent to which kidney disease and other factors affect the accuracy of A1C measurements. Laboratories should only use A1C assay methods certified by the National Glycohemoglobin Standardization Program (NGSP). (See "Estimation of blood glucose control in diabetes mellitus", section on 'Assay' and "Estimation of blood glucose control in diabetes mellitus", section on 'Sources of error'.)
Despite these limitations in the accuracy of some A1C assays, results in the range of 6 to 7 percent appear to estimate glycemic control similarly to that in patients without advanced kidney disease, while values >7.5 percent may overestimate the extent of hyperglycemia in these patients .
Some suggest that measurement of glycated albumin more accurately assesses glycemic control in this population [10-12]. Although glycated albumin also reflects mean glycemia, it reflects glycemic control over a much shorter interval (7 to 14 days, compared with 60 to 120 days for A1C). Glycated albumin measurements may not be reliable in patients with proteinuria or in those on peritoneal dialysis. In addition, in a small study in patients with type 2 diabetes and CKD undergoing intravenous iron or erythropoietin therapy, A1C (compared with glycated albumin and other markers of glycemic control) was most closely associated with mean blood glucose . There are no long-term clinical trials evaluating the relationship between glycated albumin and risk of chronic complications of diabetes [14,15]. For these reasons, we prefer monitoring glycemic control with A1C. (See "Estimation of blood glucose control in diabetes mellitus", section on 'Glycated hemoglobin'.)
Self-monitoring of blood glucose — All patients with diabetes mellitus who use insulin and some patients who take other glucose-lowering medications that can cause hypoglycemia should measure their blood glucose concentrations to help maintain safe, target-driven glucose control. The effectiveness of self-monitoring in patients with type 2 diabetes who do not use hypoglycemic agents is less certain. The indications for and practical points about blood glucose monitoring are reviewed elsewhere. (See "Blood glucose self-monitoring in management of adults with diabetes mellitus".)
Of particular importance to patients receiving peritoneal dialysis is the finding that some glucose monitors (those using the enzyme glucose dehydrogenase pyrroloquinoline quinone [GDH-PQQ]) will give falsely elevated readings in patients who have received treatments containing other sugars, including icodextrin in peritoneal dialysis fluids . This effect can persist for two weeks after stopping icodextrin (see "Peritoneal dialysis solutions"). New test strips have been designed to minimize interference with non-glucose sugars .
GOALS OF THERAPY
Nondialysis CKD patients — The A1C target that is associated with the best outcome in predialysis chronic kidney disease (CKD) patients has not been established. Target A1C levels should be tailored to the individual, balancing the improvement in microvascular complications with the risk of hypoglycemia. For most predialysis CKD patients, we suggest using an A1C target of approximately 7 percent, although the risks and benefits of targeting this goal are uncertain. Data supporting this goal are from studies of non-CKD patients and are discussed elsewhere. (See "Glycemic control and vascular complications in type 2 diabetes mellitus", section on 'Glycemic targets'.)
This goal is consistent with the 2012 Kidney Disease Outcomes Quality Initiative (K/DOQI)  and Kidney Disease: Improving Global Outcomes (KDIGO)  guidelines for patients with CKD. We also agree with K/DOQI and KDIGO that patients who are at risk for hypoglycemia should not be treated to an A1C <7 percent and that the target A1C may be higher than 7 percent in individuals who have comorbidities or limited life expectancy and who are at risk for hypoglycemia [18,19].
Dialysis patients — The A1C target that is associated with the best outcome in dialysis patients has not been established. Among dialysis patients, we target an A1C goal of 7 to 8 percent, with the specific goal in individual patients based upon the risk of hypoglycemia and presence of comorbid conditions. For patients who are relatively young (<50 years) and without significant comorbid conditions, we target an A1C goal that is close to 7 percent (ie, 7 to 7.5). However, among older patients with multiple comorbid conditions, the A1C target is closer to 8 percent (ie, 7.5 to 8).
Although benefits associated with better glycemic control in dialysis patients have been reported in several small observational studies [20-22], other larger observational studies have found no significant correlation between tight glycemic control and survival [23-27].
In a meta-analysis of observational studies of hemodialysis patients, compared with values of 6.5 to 7.4 percent, there was an increase in mortality associated with baseline and mean A1C values ≥8.5 percent (hazard ratios [HRs] of 1.14 and 1.29, respectively) . Among incident hemodialysis patients, there was an increase in mortality associated with A1C values ≤5.4 percent.
One observational study of 2798 diabetic peritoneal dialysis patients reported increased mortality in patients with A1C values ≥8 percent, compared with those achieving A1C values of 6 to 6.9 and 7 to 8 percent .
Such observational studies cannot establish a causal relationship between lower A1C levels and outcomes.
In addition, the risk of hypoglycemia is higher among dialysis patients, compared with nondialysis patients, especially with fluctuating dietary intake, such as when patients do not eat prior to dialysis .
TREATMENT — As in the non-chronic kidney disease (CKD) population, the treatment of nondialysis CKD and dialysis patients with diabetes involves both nonpharmacologic and pharmacologic therapies .
The nonpharmacologic therapies include dietary modification, exercise, and weight reduction. The additional burden of CKD dietary requirements (for example salt, protein, and volume restrictions) may further complicate diets in patients with diabetes. (See "Initial management of blood glucose in adults with type 2 diabetes mellitus", section on 'Diabetes education'.)
Pharmacologic therapies include insulin and oral agents. Our approach varies depending upon whether patients are on dialysis or not.
Nondialysis CKD patients — For nondialysis CKD patients with type 2 diabetes, the choice of initial agent depends upon glycemic goals (see 'Goals of therapy' above); the risk of medication-associated adverse events (hypoglycemia, lactic acidosis); and patient preferences and convenience. Nondialysis CKD patients with type 2 diabetes may be treated with an oral agent, although many patients end up on insulin therapy because it is more effective. The oral agents that are thought to be relatively safe in patients with nondialysis CKD include short-acting sulfonylureas (eg, glipizide, glimepiride) and repaglinide.
If an oral agent is used, the short-acting sulfonylureas, glipizide (initial dose 2.5 mg/day) or glimepiride (initial dose 1 mg/day), are the preferred agents among nondialysis CKD patients who have an estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2. Glyburide and other long-acting sulfonylureas are generally not recommended in any CKD patient with type 2 diabetes, because of the risk of hypoglycemia. Some clinicians recommend the use of the meglitinide repaglinide (starting with a dose of 0.5 mg) for nondialysis CKD patients since these agents are not renally cleared. (See "Initial management of blood glucose in adults with type 2 diabetes mellitus", section on 'Choice of initial therapy' and "Sulfonylureas and meglitinides in the treatment of diabetes mellitus".)
Metformin, which is a preferred agent in patients without kidney disease, should not be used among CKD patients with an eGFR of <30 mL/min/1.73 m2 because of an increased risk of lactic acidosis . However, we agree with the 2012 Kidney Disease Outcomes Quality Initiative (K/DOQI) guidelines that metformin may be used among patients with an eGFR >45 mL/min/1.73 . The use and dose of metformin among patients with an eGFR between 30 and 44 mL/min/1.73 are left up to the discretion of the clinician and are reviewed in detail separately. (See "Metformin in the treatment of adults with type 2 diabetes mellitus", section on 'Contraindications' and "Metformin in the treatment of adults with type 2 diabetes mellitus", section on 'Lactic acidosis'.)
Other agents including thiazolidinediones, alpha-glucosidase inhibitors, and dipeptidyl peptidase-4 (DPP-4) inhibitors are generally not considered first-line agents among CKD patients, because of limited data regarding long-term safety and efficacy. Sitagliptin and saxagliptin require dose adjustment in the setting of reduced GFR.
Patients who fail therapy with oral agents are treated with insulin. The indications for initiating insulin therapy and the principles underlying insulin therapy are the same for nondialysis CKD patients as for the general diabetic population. (See "General principles of insulin therapy in diabetes mellitus" and "Insulin therapy in type 2 diabetes mellitus".)
Among patients who are treated with insulin, the starting dose of insulin may need to be lower than would ordinarily be used for patients with normal kidney function. CKD is associated with decreased renal and ultimately hepatic metabolism of insulin (see "Carbohydrate and insulin metabolism in chronic kidney disease"). As a result, the following dose recommendations have been made for insulin dosing in this setting [31-33]:
●No dose adjustment is required if the GFR is >50 mL/min
●The insulin dose should be reduced to approximately 75 percent of baseline when the GFR is between 10 and 50 mL/min
●The dose should be reduced by as much as 50 percent when the GFR is <10 mL/min
The initial dose of intermediate- or long-acting insulin in patients without CKD is approximately 10 units or 0.2 units/kg (algorithm 1). Thus, in a patient with an eGFR between 10 and 50 mL/min, the initial dose would be approximately 7 units (75 percent of 10 units). The balance between altered insulin resistance and insulin clearance as renal dysfunction progresses is difficult to predict in any individual patient, so insulin adjustment is often largely empiric. Therefore, it is important that blood glucose levels be monitored closely and that individually appropriate dose adjustments in insulin therapy be made.
Hemodialysis patients — For most hemodialysis patients, we use insulin rather than oral agents. This is consistent with the 2005 K/DOQI guidelines, which suggest that, among dialysis patients, newer insulin regimens and insulin preparations should be used rather than oral agents for glycemic control . This is due to the lack of adequate data concerning the use of oral agents in dialysis patients and their inability to adequately excrete many such agents.
The principles of insulin therapy are the same for dialysis patients as for the general diabetic population. Several different insulin regimens can be used to achieve glycemic control. Examples include twice-daily intermediate-acting insulin, with regular insulin given before breakfast and before supper, or long-acting insulin, with two or three times daily supplemental regular insulin, given two or three times per day before meals [35,36]. For hemodialysis patients, the initial dose of insulin should be decreased by approximately 50 percent, as described above for nondialysis CKD patients with GFR <10 mL/min. The dose should be titrated upward, as indicated by blood glucose monitoring. Most patients will require more insulin than this initial dose. (See "General principles of insulin therapy in diabetes mellitus" and "Insulin therapy in type 2 diabetes mellitus".)
A consensus approach does not exist to the choice of insulin in patients with diabetes and end-stage renal disease (ESRD) . Some suggest that long-acting insulin preparations should be avoided, while others feel that such agents should be used.
Some clinicians prefer to use oral agents rather than insulin, especially among patients who are already on these agents and have achieved acceptable glycemic control. The preferred agents are glipizide or repaglinide since they are primarily metabolized by the liver, since inactive or only very weakly active metabolites are excreted in the urine, and since the risk of hypoglycemia is lower than with other oral agents . Although repaglinide drug concentration and elimination half-life are increased marginally in patients with reduced GFR, dose reductions are not necessary, and this agent may be an appropriate therapy for patients with ESRD .
Peritoneal dialysis patients
Treatment regimens — For patients who were already on an oral agent with good glycemic control prior to starting dialysis, we typically continue the oral agent. For patients who develop diabetes after starting dialysis, we generally treat first with an oral agent. However, over time, many peritoneal dialysis patients will require insulin.
As for nondialysis CKD patients, the preferred oral agent is either glipizide (initial dose 2.5 mg/day) or glimepiride (initial dose 1 mg/day). Repaglinide, which has minimal kidney clearance, is an alternative (starting with a dose of 0.5 mg/day). Metformin should not be used among peritoneal dialysis patients, because of an increased risk of lactic acidosis .
Most peritoneal dialysis patients require insulin to maintain good glycemic control. Patients on continuous ambulatory peritoneal dialysis (CAPD) or continuous cycler peritoneal dialysis (CCPD) who are being treated with insulin may be treated with subcutaneous or intraperitoneal insulin. We prefer subcutaneous insulin. The principles underlying subcutaneous insulin therapy are the same for nondialysis CKD patients as for the general diabetic population (see "General principles of insulin therapy in diabetes mellitus" and "Insulin therapy in type 2 diabetes mellitus"). The initial starting dose is similar to hemodialysis patients (decrease by approximately 50 percent), with upward titration based upon blood glucose monitoring.
We, and most other nephrologists, do not use intraperitoneal insulin, since it often does not adequately control blood sugars. Even among patients who achieve adequate glycemic control with intraperitoneal insulin alone, the required insulin regimen is very complex since patients often alter CAPD schedules, as well as the timing of meals, necessitating constant adjustment of the intraperitoneal insulin, which is burdensome for the patient.
Other disadvantages of intraperitoneal insulin include the risk of bacterial contamination of dialysate during injection of insulin into the bags ; the requirement for a higher total insulin dose due to losses into spent dialysate and to binding to the plastics in bags and tubing [40-43]; and an associated risk of peritoneal fibroblastic proliferation  and, perhaps, of hepatic subcapsular steatosis . Another potential concern is that the absorption of insulin may significantly vary among patients or may decline over time in a single individual due to acquired abnormalities in the peritoneal membrane . One study of seven patients suggested that the latter is not a common problem, as there was no difference in insulin absorption after 30 months of CAPD .
There are no long-term studies evaluating use of intraperitoneal insulin. A meta-analysis of three trials demonstrated better glycemic control, as assessed by A1C, with intraperitoneal compared with subcutaneous insulin in CAPD patients, but the insulin dose was more than double with intraperitoneal use . Patients treated with intraperitoneal insulin had lower high-density lipoprotein (HDL) cholesterol levels and higher triglyceride levels compared with those treated with subcutaneous .
Abnormal or variations in peritoneal kinetics — The above recommendations regarding the treatment of peritoneal dialysis patients assume that peritoneal transfer kinetics are relatively normal. Diabetic patients on CAPD who have uncontrolled hyperglycemia should undergo a peritoneal equilibration test. (See "Peritoneal equilibration test".)
High transporters, who can have enormous glucose loads from rapid peritoneal glucose absorption (figure 1), will usually benefit from transfer to nocturnal automated peritoneal dialysis [40,48]. In addition to raising the blood glucose, the rapid glucose absorption lowers the osmotic gradient between dialysate and blood, resulting in reduced ultrafiltration, diminished urea removal, and fluid retention. A vicious cycle is then created, with generalized edema requiring frequent use of 2.5 and 4.25 percent dextrose dialysis solutions, leading to further hyperglycemia.
Carbohydrate-sparing dialytic regimens — Adjusting the peritoneal dialysis solution may improve glycemic control in some patients. Glucose polymers were introduced to replace glucose-containing solutions by offering the possible advantages of decreased absorption of solute and increased ultrafiltration for a longer period of time. The use of a glucose polymer as an osmotic agent is particularly appealing as a substitute for glucose solutions, particularly in patients with diabetes .
Carbohydrate-sparing dialytic regimens include:
●Amino acid dialysate solutions, which are available in countries outside the United States
This topic is reviewed in detail elsewhere. (See "Peritoneal dialysis solutions".)
ORAL AGENTS — In general, a large number of oral agents are available to manage type 2 diabetes mellitus. Knowledge of the metabolism of these agents in patients with kidney dysfunction is essential, given that significant toxicity, including prolonged hypoglycemia, can be associated with some of the drugs [31,32,51]. (See "Initial management of blood glucose in adults with type 2 diabetes mellitus", section on 'Initial pharmacologic therapy' and "Management of persistent hyperglycemia in type 2 diabetes mellitus", section on 'Treatment options'.)
Sulfonylureas — Glipizide and glimepiride are the sulfonylureas of choice in patients with chronic kidney disease (CKD) [52,53]. (See "Sulfonylureas and meglitinides in the treatment of diabetes mellitus".)
The administration of sulfonylureas in end-stage renal disease (ESRD) requires careful attention to dosing and routes of elimination [32,33,54,55]. The sulfonylureas are strongly protein bound, particularly to albumin . Thus, elevated plasma drug levels cannot be efficiently reversed by hemodialysis. Furthermore, displacement of these drugs from albumin by beta blockers, salicylates, and warfarin can lead to hypoglycemia due to increased plasma concentration of the free sulfonylurea. Older sulfonylureas (acetohexamide, chlorpropamide, tolazamide, and tolbutamide) have been largely replaced by newer agents such as glyburide, glipizide, and glimepiride.
The basic principles of sulfonylurea metabolism can be summarized as follows:
●Glyburide has weak active metabolites that are excreted in the urine and accumulate in patients with impaired kidney function.
Meglitinides — Meglitinides, such as repaglinide or nateglinide, are sulfonylurea-like agents that stimulate insulin secretion . (See "Sulfonylureas and meglitinides in the treatment of diabetes mellitus".)
Repaglinide is principally metabolized by the liver, with less than 10 percent renally excreted . In one study of patients with diabetes and various degrees of kidney impairment who were initiating repaglinide, the proportion of patients with hypoglycemia was not significantly different with increasing severity of kidney impairment . Maintenance doses were lower in patients with advanced CKD. In patients with an eGFR between 20 and 40 mL/min, initiation of treatment should be with 0.5 mg before the largest meal and then advanced to 0.5 mg prior to other meals, as needed. Close, careful monitoring of blood glucose levels is essential as the dose is titrated. In patients with an eGFR of ≥40 mL/min, no dose adjustment is necessary, and repaglinide can be initiated at 0.5 mg prior to each meal with dose titration based upon blood glucose levels. Repaglinide has not been studied in patients with an eGFR of <20 mL/min. (See "Sulfonylureas and meglitinides in the treatment of diabetes mellitus", section on 'Dosing and monitoring'.)
Nateglinide is hepatically metabolized, with renal excretion of active metabolites. With decreased kidney function, the accumulation of active metabolites and hypoglycemia has occurred [57,58]. This drug must therefore be used cautiously in this setting, if at all.
Dipeptidyl peptidase 4 (DPP-4) inhibitors — Although some DPP-4 inhibitors have been studied in patients with kidney dysfunction, the data are limited, and, therefore, we suggest not using DPP-4 inhibitors among such patients. However, limited data suggest that these agents are effective and relatively safe in CKD and ESRD patients [59,60]. Dose adjustments are needed for some agents in this class.
DPP-4 is an enzyme expressed on the surface of most cell types that deactivates the incretin hormones glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). DPP-4 inhibitors cause a glucose-dependent increase in insulin release. The use of DPP-4 inhibitors in patients without ESRD is discussed elsewhere. (See "Dipeptidyl peptidase-4 (DPP-4) inhibitors for the treatment of type 2 diabetes mellitus", section on 'Candidates'.)
Linagliptin is only minimally excreted in the urine (<10 percent) and does not require dose adjustment in patients on dialysis, but its use in ESRD patients is limited . Sitagliptin is largely excreted in the urine, with 70 to 80 percent of an oral dose appearing unchanged in the urine . If sitagliptin is used among ESRD patients, a dose reduction to 25 mg daily (usual dose 100 mg daily) is recommended. Although sitagliptin is partially cleared by hemodialysis, it may be given without regard to the timing of dialysis . Information on the use of sitagliptin in ESRD patients is limited , but it appears to have equal efficacy to glipizide [59,60]. Saxagliptin and its primary active metabolite are excreted in the urine (total urinary excretion approximately 60 to 75 percent); a daily dose of 2.5 mg is the recommended dose for patients with ESRD. Saxagliptin is removed by hemodialysis and should be administered after dialysis . Saxagliptin was reported to be well tolerated in a single small study . Vildagliptin also appears to be effective in this population in limited studies .
Thiazolidinediones — Thiazolidinediones (rosiglitazone, pioglitazone) should be avoided in patients with advanced CKD, especially those with preexisting heart failure, given the risk of edema and heart failure. The use of rosiglitazone is not recommended because of concern about its atherogenic lipid profiles and a potential increased risk for cardiovascular events. In an observational study, the use of rosiglitazone was associated with increased all-cause and cardiovascular mortality in hemodialysis patients .
Thiazolidinediones enhance tissue sensitivity to insulin and suppress hepatic glucose production via binding to peroxisome proliferator-activated receptor (PPAR) gamma . (See "Thiazolidinediones in the treatment of diabetes mellitus".)
Rosiglitazone and pioglitazone are highly protein bound, primarily to albumin, and almost completely metabolized by the liver . Rosiglitazone has inactive metabolites, and less than 1 percent of the original compound is excreted in the urine; pioglitazone has three active metabolites. With both agents, accumulation of the parent drug or the major metabolites does not occur in the setting of CKD. Hemodialysis does not affect the pharmacokinetics of these drugs.
These agents are associated with heart failure and edema, which may be more frequent in patients also receiving insulin . The mechanism of edema formation with these agents appears to be related to stimulation of PPAR gamma-mediated sodium reabsorption by renal epithelial sodium channels in the collecting duct . Although unproven clinically, amiloride, spironolactone, or similar agents may therefore be effective for managing this fluid retention.
These agents slow carbohydrate absorption from the gastrointestinal tract and reduce postprandial blood sugar peaks. (See "Alpha-glucosidase inhibitors and lipase inhibitors for treatment of diabetes mellitus".)
With acarbose, increased levels of the parent drug and metabolites are observed with CKD, although an increased risk of hypoglycemia has not been documented. Miglitol is absorbed to a greater extent than acarbose and is largely renally excreted, with increased accumulation in patients with decreased kidney function.
Metformin — Biguanides, such as metformin, are primarily excreted unchanged in the urine. Thus, patients with kidney dysfunction are more susceptible to drug accumulation and lactic acidosis with these compounds. They should therefore be avoided in patients with severe CKD . (See "Metformin in the treatment of adults with type 2 diabetes mellitus".)
Sodium-glucose cotransporter-2 (SGLT-2) inhibitors — These agents, several of which have been approved for use alone or in combination with others agents in the US, including dapagliflozin, canagliflozin, and empagliflozin, inhibit glucose absorption in the proximal tubule, causing glucosuria, weight loss, and improved glycemic control. Use of these medications is not recommended with an estimated glomerular filtration rate (eGFR) of <45 to 60 mL/min/1.73 m2, and they are contraindicated with an eGFR of <30 mL/min/1.73 m2, including patients with ESRD who are on dialysis [71-73]. These drugs have also been implicated in causing diabetic ketoacidosis in individuals with type 1 and type 2 diabetes . (See "Sodium-glucose co-transporter 2 inhibitors for the treatment of type 2 diabetes mellitus", section on 'Contraindications and precautions'.)
In addition, there have been postmarketing reports of acute kidney injury (AKI), some requiring hospitalization and dialysis, in patients taking canagliflozin or dapagliflozin [75,76]. Among 101 cases of possible SGLT-2-associated AKI reported to the FDA, approximately one-half occurred within one month of initiating the drug, and most patients improved after the drug was discontinued. Some patients with AKI may have been volume depleted, hypotensive, or taking other medications that could affect the kidneys. It is unclear whether any of the patients in these reports had preexisting CKD. (See "Sodium-glucose co-transporter 2 inhibitors for the treatment of type 2 diabetes mellitus", section on 'Acute kidney injury'.)
APPROACH TO PROBLEM PATIENTS — Some diabetic patients have persistent hyperglycemia, severe hyperglycemia, diabetic ketoacidosis, frequent hypoglycemia, or alternating episodes of hyperglycemia and hypoglycemia.
Hyperglycemia — Inadequate insulin dose and noncompliance (with diet or the insulin regimen) are the most common causes of persistent hyperglycemia in diabetic dialysis patients (defined as an A1C level >9 percent) . An additional problem is that microvascular disease can cause erratic absorption of insulin from the subcutaneous tissue, particularly if the patient does not rotate injection sites . (See "General principles of insulin therapy in diabetes mellitus", section on 'Site of injection'.) The approach to these issues is the same for chronic kidney disease (CKD) patients as for non-CKD patients and is discussed elsewhere. (See "General principles of insulin therapy in diabetes mellitus", section on 'Determinants of insulin efficacy' and "Management of persistent hyperglycemia in type 2 diabetes mellitus", section on 'Indications for a second agent'.)
Severe hyperglycemia and ketoacidosis — Severe hyperglycemia, with serum glucose concentrations occasionally >1000 mg/dL (55 mmol/L), may be observed among dialysis patients with diabetes. Unlike those without end-stage renal disease (ESRD), however, hypovolemia and marked hypernatremia do not occur, since glucosuria is absent in anuric individuals. The net effect is minimal symptoms, even among those with extreme hyperglycemia .
However, these patients may have marked hyperkalemia due to potassium efflux from cells in response to extracellular fluid hypertonicity, as well as hyponatremia and acute intravascular volume expansion . Patients with type 1 diabetes may also develop diabetic ketoacidosis.
Instead of fluid replacement, management is principally dependent upon the administration of low doses of intravenous insulin (commonly beginning at a dose of 2 units/hour) . As with nondialysis patients with severe hyperglycemia and diabetic ketoacidosis, serum glucose and potassium concentrations must be closely monitored. (See "Diabetic ketoacidosis and hyperosmolar hyperglycemic state in adults: Treatment", section on 'Monitoring'.)
Hypoglycemia — Frequent or persistent hypoglycemia in diabetic dialysis patients is often due to severe underdialysis, with poor calorie intake, or occult disease, such as infection or malignancy. Frequent adjustment of insulin dose and evaluation of blood glucose diaries are essential in this setting, as is provision of an adequate dialysis dose (see "Prescribing and assessing adequate hemodialysis" and "Prescribing and assessing adequate peritoneal dialysis" and "Protein intake in maintenance hemodialysis patients" and "Assessment of nutritional status in hemodialysis patients"). Drugs that interfere with the counterregulatory response to hypoglycemia (such as beta blockers) and long-acting insulin and oral agents should be discontinued, if possible, until more stable glycemic control without hypoglycemia is achieved.
Alternating hypoglycemia and hyperglycemia — ESRD patients with diabetes often have gastroparesis , which complicates the timing of insulin injections. Gastric-emptying studies will confirm the diagnosis, which can often be effectively treated with metoclopramide or bethanechol (urecholine) . Improvement in glycemic control may also improve gastric motility. (See "Diabetic autonomic neuropathy of the gastrointestinal tract", section on 'Gastroparesis'.)
Other causes of brittle blood glucose include patient misunderstanding of the timing of insulin injections, poor compliance with dietary restrictions and insulin therapy, erratic eating habits, and poor timing of continuous ambulatory peritoneal dialysis (CAPD) exchanges. These problems can often be corrected with patient re-education. Noncompliance, impaired vision, and a depressive illness should also be sought. (See "The adult patient with brittle diabetes mellitus".)
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Chronic kidney disease in adults".)
SUMMARY AND RECOMMENDATIONS
●Chronic kidney disease (CKD) is associated with insulin resistance and, in advanced CKD, decreased insulin degradation. Careful, individualized therapy is essential among patients with advanced CKD or on dialysis. (See 'Introduction' above.)
●We use the glycated hemoglobin (A1C) concentration to monitor for hyperglycemia in all patients with diabetes and predialysis CKD or end-stage renal disease (ESRD), although A1C may not be as accurate among ESRD patients as in the general population. It is important to be aware of the specific assay used in each dialysis facility and the extent to which kidney disease and other factors affect the accuracy of A1C measurements. (See 'Monitoring glycemic control' above.)
●The A1C target that is associated with the best outcome in predialysis CKD patients has not been established. For most predialysis CKD patients, we suggest using an A1C goal of approximately 7 percent, rather than lower values (Grade 2C). (See "Glycemic control and vascular complications in type 2 diabetes mellitus", section on 'Glycemic targets' and 'Nondialysis CKD patients' above.)
●The A1C goal that is associated with the best outcome in dialysis patients has not been established, and most studies have suggested that rigorous glycemic control does not lead to better outcomes. The A1C goal we use depends upon the age of the patients, the risk of hypoglycemia, and upon comorbid conditions (see 'Dialysis patients' above):
•For patients who are relatively young (≤50 years) and have no other significant comorbid conditions, we suggest using an A1C goal of 7 to 7.5, rather than higher values (Grade 2C).
•For older patients (ie, >50 years) who have multiple comorbid conditions, we suggest using an A1C goal of 7.5 to 8, rather than lower values (Grade 2C).
●The treatment of nondialysis CKD and dialysis patients with diabetes involves both nonpharmacologic and pharmacologic therapies. The nonpharmacologic therapies include dietary modification, exercise, and weight reduction. Pharmacologic therapies include insulin and oral agents. Our pharmacologic approach varies depending upon whether patients have predialysis CKD or are on dialysis.
●For nondialysis CKD patients with type 2 diabetes, we suggest initial treatment with an oral agent, rather than insulin (Grade 2B). The preferred agents and initial dosing are glipizide (2.5 mg/day) or glimepiride (1 mg/day); an alternative agent is repaglinide, starting with a dose of 0.5 mg/day. Metformin should not be used among CKD patients with an estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2, because of an increased risk of lactic acidosis.
Patients who fail therapy with oral agents are treated with insulin. Among patients who are treated with insulin, the starting dose of insulin may need to be lower than would ordinarily be used for patients with normal kidney function. (See 'Nondialysis CKD patients' above.)
●For most hemodialysis patients with type 2 diabetes, we suggest initial treatment with insulin, rather than oral agents (Grade 2C). Several different insulin regimens can be used to achieve glycemic control. Examples include: twice-daily intermediate-acting insulin, with regular insulin given before breakfast and before supper, or long-acting insulin, with two or three times daily supplemental regular insulin, given two or three times per day before meals. The initial dose of insulin should be decreased by approximately 50 percent. (See 'Hemodialysis patients' above.)
Some clinicians prefer to use oral agents rather than insulin, especially among patients who have already achieved acceptable glycemic control on these agents. If an oral agent is used, the preferred agents are glipizide or repaglinide.
●For peritoneal dialysis patients with type 2 diabetes who were already on an oral agent with good glycemic control prior to starting dialysis and for patients who develop diabetes after starting dialysis, we suggest using an oral agent, rather than insulin (Grade 2C). The preferred agents and suggested initial doses are glipizide (2.5 mg/day) or glimepiride (1 mg/day). Repaglinide, starting with a dose of 0.5 mg/day, can also be considered. Metformin should not be used among peritoneal dialysis patients, because of an increased risk of lactic acidosis. (See 'Peritoneal dialysis patients' above.)
Many peritoneal dialysis patients will require insulin therapy to achieve adequate glycemic control. For such patients, we suggest the use of subcutaneous, rather than intraperitoneal, insulin (Grade 2C). The initial dose of insulin should be decreased by approximately 50 percent. (See 'Peritoneal dialysis patients' above.)
●Severe hyperglycemia may be observed among dialysis patients with diabetes. Unlike those without ESRD, hypovolemia and marked hypernatremia do not occur, since glucosuria is absent in anuric individuals. Patients have few symptoms, but may have marked hyperkalemia, hyponatremia, and acute intravascular volume expansion. Instead of fluid replacement, management is principally dependent upon the administration of low doses of intravenous insulin. (See 'Severe hyperglycemia and ketoacidosis' above.)
- Scott MG, Hoffmann JW, Meltzer VN, et al. Effects of azotemia on results of the boronate-agarose affinity and ion-exchange methods for glycated hemoglobin. Clin Chem 1984; 30:896.
- Bruns DE, Lobo PI, Savory J, Wills MR. Specific affinity-chromatographic measurement of glycated hemoglobins in uremic patients. Clin Chem 1984; 30:569.
- Paisey R, Banks R, Holton R, et al. Glycosylated haemoglobin in uraemia. Diabet Med 1986; 3:445.
- Ansari A, Thomas S, Goldsmith D. Assessing glycemic control in patients with diabetes and end-stage renal failure. Am J Kidney Dis 2003; 41:523.
- de Boer MJ, Miedema K, Casparie AF. Glycosylated haemoglobin in renal failure. Diabetologia 1980; 18:437.
- Wettre S, Lundberg M. Kinetics of glycosylated haemoglobin in uraemia determined on ion-exchange and affinity chromatography: no increase in the rate of glycosylation. Diabetes Res 1986; 3:107.
- Joy MS, Cefalu WT, Hogan SL, Nachman PH. Long-term glycemic control measurements in diabetic patients receiving hemodialysis. Am J Kidney Dis 2002; 39:297.
- Freedman BI, Shenoy RN, Planer JA, et al. Comparison of glycated albumin and hemoglobin A1c concentrations in diabetic subjects on peritoneal and hemodialysis. Perit Dial Int 2010; 30:72.
- Robinson TW, Freedman BI. Assessing glycemic control in diabetic patients with severe nephropathy. J Ren Nutr 2013; 23:199.
- Peacock TP, Shihabi ZK, Bleyer AJ, et al. Comparison of glycated albumin and hemoglobin A(1c) levels in diabetic subjects on hemodialysis. Kidney Int 2008; 73:1062.
- Freedman BI, Andries L, Shihabi ZK, et al. Glycated albumin and risk of death and hospitalizations in diabetic dialysis patients. Clin J Am Soc Nephrol 2011; 6:1635.
- Freedman BI, Shihabi ZK, Andries L, et al. Relationship between assays of glycemia in diabetic subjects with advanced chronic kidney disease. Am J Nephrol 2010; 31:375.
- Konya J, Ng JM, Cox H, et al. Use of complementary markers in assessing glycaemic control in people with diabetic kidney disease undergoing iron or erythropoietin treatment. Diabet Med 2013; 30:1250.
- Abe M, Matsumoto K. Glycated hemoglobin or glycated albumin for assessment of glycemic control in hemodialysis patients with diabetes? Nat Clin Pract Nephrol 2008; 4:482.
- Sacks DB, Arnold M, Bakris GL, et al. Guidelines and recommendations for laboratory analysis in the diagnosis and management of diabetes mellitus. Diabetes Care 2011; 34:e61.
- Sloand JA. Dialysis patient safety: safeguards to prevent iatrogenic hypoglycemia in patients receiving icodextrin. Am J Kidney Dis 2012; 60:514.
- http://www.pharmacypracticenews.com/download/BGSystems.pdf (Accessed on May 28, 2015).
- National Kidney Foundation. KDOQI Clinical Practice Guideline for Diabetes and CKD: 2012 Update. Am J Kidney Dis 2012; 60:850.
- KDIGO. Chapter 1: Definition and classification of CKD. Kidney Int Suppl 2013; 3:19. http://www.kdigo.org/clinical_practice_guidelines/pdf/CKD/KDIGO_2012_CKD_GL.pdf (Accessed on March 04, 2013).
- Tzamaloukas AH, Murata GH, Zager PG, et al. The relationship between glycemic control and morbidity and mortality for diabetics on dialysis. ASAIO J 1993; 39:880.
- McMurray SD, Johnson G, Davis S, McDougall K. Diabetes education and care management significantly improve patient outcomes in the dialysis unit. Am J Kidney Dis 2002; 40:566.
- Oomichi T, Emoto M, Tabata T, et al. Impact of glycemic control on survival of diabetic patients on chronic regular hemodialysis: a 7-year observational study. Diabetes Care 2006; 29:1496.
- Williams ME, Lacson E Jr, Teng M, et al. Hemodialyzed type I and type II diabetic patients in the US: Characteristics, glycemic control, and survival. Kidney Int 2006; 70:1503.
- Shurraw S, Majumdar SR, Thadhani R, et al. Glycemic control and the risk of death in 1,484 patients receiving maintenance hemodialysis. Am J Kidney Dis 2010; 55:875.
- Williams ME, Lacson E Jr, Wang W, et al. Glycemic control and extended hemodialysis survival in patients with diabetes mellitus: comparative results of traditional and time-dependent Cox model analyses. Clin J Am Soc Nephrol 2010; 5:1595.
- Kovesdy CP, Park JC, Kalantar-Zadeh K. Glycemic control and burnt-out diabetes in ESRD. Semin Dial 2010; 23:148.
- Duong U, Mehrotra R, Molnar MZ, et al. Glycemic control and survival in peritoneal dialysis patients with diabetes mellitus. Clin J Am Soc Nephrol 2011; 6:1041.
- Hill CJ, Maxwell AP, Cardwell CR, et al. Glycated hemoglobin and risk of death in diabetic patients treated with hemodialysis: a meta-analysis. Am J Kidney Dis 2014; 63:84.
- Tzamaloukas AH. The use of glycosylated hemoglobin in dialysis patients. Semin Dial 1998; 11:143.
- Garg R, Williams ME. Diabetes management in the kidney patient. Med Clin North Am 2013; 97:135.
- Charpentier G, Riveline JP, Varroud-Vial M. Management of drugs affecting blood glucose in diabetic patients with renal failure. Diabetes Metab 2000; 26 Suppl 4:73.
- Snyder RW, Berns JS. Use of insulin and oral hypoglycemic medications in patients with diabetes mellitus and advanced kidney disease. Semin Dial 2004; 17:365.
- Aronoff GR, Berns JS, Brier ME, et al. Drug Prescribing in Renal Failure: Dosing Guidelines for Adults, 4th ed, American College of Physicians, Philadelphia 1999.
- K/DOQI Workgroup. K/DOQI clinical practice guidelines for cardiovascular disease in dialysis patients. Am J Kidney Dis 2005; 45:S1.
- Tunbridge FK, Newens A, Home PD, et al. A comparison of human ultralente- and lente-based twice-daily injection regimens. Diabet Med 1989; 6:496.
- Freeman SL, O'Brien PC, Rizza RA. Use of human ultralente as the basal insulin component in treatment of patients with IDDM. Diabetes Res Clin Pract 1991; 12:187.
- Tzamaloukas AH, Friedman EA. Diabetes. In: Handbook of Dialysis, 3rd ed, Daugirdas JT, Blake PG, Ing TS (Eds), Lippincott, Williams & Wilkins, Philadelphia 2001. p.453.
- Hasslacher C, Multinational Repaglinide Renal Study Group. Safety and efficacy of repaglinide in type 2 diabetic patients with and without impaired renal function. Diabetes Care 2003; 26:886.
- Selgas R, Diez JJ, Muñoz J, et al. Comparative study of two different routes for insulin administration in CAPD diabetic patients. A multicenter study. Adv Perit Dial 1989; 5:181.
- Diaz-Buxo JA. Blood glucose control in diabetics: I. Semin Dial 1993; 6:392.
- Daniels ID, Markell MS. Blood glucose control in diabetics: II. Semin Dial 1993; 6:394.
- Maxwell DR, Prince MJ. Blood glucose control in diabetics: III. Semin Dial 1993; 6:397.
- Quellhorst E. Insulin therapy during peritoneal dialysis: pros and cons of various forms of administration. J Am Soc Nephrol 2002; 13 Suppl 1:S92.
- Selgas R, Lopez-Riva A, Alvaro F, et al. Insulin influence on the mitogenic-induced effect of the peritoneal effluent in CAPD patients. In: Adv Peritoneal Dial, Khanna R, Nolph KD, Prowant B, et al (Eds), University of Toronto Press, Toronto 1991. Vol 7, p.161.
- Fine A, Parry D, Ariano R, Dent W. Marked variation in peritoneal insulin absorption in peritoneal dialysis. Perit Dial Int 2000; 20:652.
- Scavini M, Pincelli A, Petrella G, et al. Intraperitoneal insulin absorption after long-term intraperitoneal insulin therapy. Diabetes Care 1995; 18:56.
- Almalki MH, Altuwaijri MA, Almehthel MS, et al. Subcutaneous versus intraperitoneal insulin for patients with diabetes mellitus on continuous ambulatory peritoneal dialysis: meta-analysis of non-randomized clinical trials. Clin Invest Med 2012; 35:E132.
- Twardowski ZJ, Nolph KD, Khanna R, et al. Daily clearances with continuous ambulatory peritoneal dialysis and nightly peritoneal dialysis. ASAIO Trans 1986; 32:575.
- Holmes C, Mujais S. Glucose sparing in peritoneal dialysis: implications and metrics. Kidney Int Suppl 2006; :S104.
- Burkart J. Metabolic consequences of peritoneal dialysis. Semin Dial 2004; 17:498.
- Flynn C, Bakris GL. Noninsulin glucose-lowering agents for the treatment of patients on dialysis. Nat Rev Nephrol 2013; 9:147.
- Alsahli M, Gerich JE. Hypoglycemia in Patients with Diabetes and Renal Disease. J Clin Med 2015; 4:948.
- Rosenkranz B, Profozic V, Metelko Z, et al. Pharmacokinetics and safety of glimepiride at clinically effective doses in diabetic patients with renal impairment. Diabetologia 1996; 39:1617.
- Mujais SK, Fadda G. Carbohydrate metabolism in end-stage renal disease. Semin Dial 1989; 2:46.
- Harrower AD. Pharmacokinetics of oral antihyperglycaemic agents in patients with renal insufficiency. Clin Pharmacokinet 1996; 31:111.
- Skillman TG, Feldman JM. The pharmacology of sulfonylureas. Am J Med 1981; 70:361.
- Inoue T, Shibahara N, Miyagawa K, et al. Pharmacokinetics of nateglinide and its metabolites in subjects with type 2 diabetes mellitus and renal failure. Clin Nephrol 2003; 60:90.
- Nagai T, Imamura M, Iizuka K, Mori M. Hypoglycemia due to nateglinide administration in diabetic patient with chronic renal failure. Diabetes Res Clin Pract 2003; 59:191.
- St Peter WL, Weinhandl ED, Flessner MF. Sitagliptin--another option for managing type 2 diabetes in dialysis patients? Am J Kidney Dis 2013; 61:532.
- Arjona Ferreira JC, Corry D, Mogensen CE, et al. Efficacy and safety of sitagliptin in patients with type 2 diabetes and ESRD receiving dialysis: a 54-week randomized trial. Am J Kidney Dis 2013; 61:579.
- Gallwitz B. Safety and efficacy of linagliptin in type 2 diabetes patients with common renal and cardiovascular risk factors. Ther Adv Endocrinol Metab 2013; 4:95.
- Bergman AJ, Stevens C, Zhou Y, et al. Pharmacokinetic and pharmacodynamic properties of multiple oral doses of sitagliptin, a dipeptidyl peptidase-IV inhibitor: a double-blind, randomized, placebo-controlled study in healthy male volunteers. Clin Ther 2006; 28:55.
- Bergman AJ, Cote J, Yi B, et al. Effect of renal insufficiency on the pharmacokinetics of sitagliptin, a dipeptidyl peptidase-4 inhibitor. Diabetes Care 2007; 30:1862.
- Chan JC, Scott R, Arjona Ferreira JC, et al. Safety and efficacy of sitagliptin in patients with type 2 diabetes and chronic renal insufficiency. Diabetes Obes Metab 2008; 10:545.
- Boulton DW, Li L, Frevert EU, et al. Influence of renal or hepatic impairment on the pharmacokinetics of saxagliptin. Clin Pharmacokinet 2011; 50:253.
- Nowicki M, Rychlik I, Haller H, et al. Long-term treatment with the dipeptidyl peptidase-4 inhibitor saxagliptin in patients with type 2 diabetes mellitus and renal impairment: a randomised controlled 52-week efficacy and safety study. Int J Clin Pract 2011; 65:1230.
- Russo E, Penno G, Del Prato S. Managing diabetic patients with moderate or severe renal impairment using DPP-4 inhibitors: focus on vildagliptin. Diabetes Metab Syndr Obes 2013; 6:161.
- Ramirez SP, Albert JM, Blayney MJ, et al. Rosiglitazone is associated with mortality in chronic hemodialysis patients. J Am Soc Nephrol 2009; 20:1094.
- Lubowsky ND, Siegel R, Pittas AG. Management of glycemia in patients with diabetes mellitus and CKD. Am J Kidney Dis 2007; 50:865.
- Guan Y, Hao C, Cha DR, et al. Thiazolidinediones expand body fluid volume through PPARgamma stimulation of ENaC-mediated renal salt absorption. Nat Med 2005; 11:861.
- Kalra S. Sodium Glucose Co-Transporter-2 (SGLT2) Inhibitors: A Review of Their Basic and Clinical Pharmacology. Diabetes Ther 2014; 5:355.
- Brietzke SA. Oral antihyperglycemic treatment options for type 2 diabetes mellitus. Med Clin North Am 2015; 99:87.
- Vlotides G, Mertens PR. Sodium-glucose cotransport inhibitors: mechanisms, metabolic effects and implications for the treatment of diabetic patients with chronic kidney disease. Nephrol Dial Transplant 2015; 30:1272.
- Peters AL, Buschur EO, Buse JB, et al. Euglycemic Diabetic Ketoacidosis: A Potential Complication of Treatment With Sodium-Glucose Cotransporter 2 Inhibition. Diabetes Care 2015; 38:1687.
- Invokana (canagliflozin). US FDA approved product information; Titusville, NJ: Janssen Pharmaceuticals, Inc; May 2016. (Available online at www.accessdata.fda.gov/drugsatfda_docs/label/2016/204042s015s019lbl.pdf (accessed May 31, 2016)).
- http://www.fda.gov/Drugs/DrugSafety/ucm505860.htm (Accessed on June 17, 2016).
- Tzamaloukas AH, Murata GH, Eisenberg B, et al. Hypoglycemia in diabetics on dialysis with poor glycemic control: hemodialysis versus continuous ambulatory peritoneal dialysis. Int J Artif Organs 1992; 15:390.
- Mak RH. Impact of end-stage renal disease and dialysis on glycemic control. Semin Dial 2000; 13:4.
- Montoliu J, Revert L. Lethal hyperkalemia associated with severe hyperglycemia in diabetic patients with renal failure. Am J Kidney Dis 1985; 5:47.