Oral Antihyperglycemic Agents and Renal Disease: New Agents, New Concepts

Home , Acetohexamide, Chlorpropamide, Nateglinide, Repaglinide

Jean-Franc¸ois Yale Metabolic Day Centre, Royal Victoria Hospital, Montreal, Quebec, Canada

The results of the Diabetes Control and Complications Trial (DCCT) and UK Prospective Diabetes Study trials in type 1 and type 2 diabetes, respectively, have proved the importance of intensive glucose management in the prevention of microvascular complications (retinopathy, nephropathy, and neuropathy). Both trials showed encouraging trends for a decrease in macrovascular complications, and this is being pursued in new studies. These findings have led to more strict goals for glucose control. As glucose levels are aimed to be closer to the normal range, the risk for hypoglycemia also increases dramatically. The choice of the agent therefore is more influenced currently by the risk for hypoglycemia. There are presently four classes of oral antihyperglycemic agents. These agents differ greatly in terms of mechanisms of action, efficacy, side effect profiles, and cost. Except for Acarbose, all classes decrease the glycosylated hemoglobin by a similar magnitude: 1.0 to 1.5%. In chronic renal failure, the oral agents that can be used therefore include the insulin secretagogues repaglinide and nateglinide and the thiazolidinediones (rosiglitazone and pioglitazone) with caution. Insulin also can be used safely in renal failure. J Am Soc Nephrol 16: S7–S10, 2005. doi: 10.1681/ASN.2004110974

ype 2 diabetes occurs as a result of a complex interplay Therapy among multiple genetic and environmental factors that The diagnosis of type 2 diabetes is often delayed, and 20 to 50% T lead to both increased insulin resistance and impaired present with microvascular or macrovascular complications at the pancreatic insulin secretion. Epidemiologic studies have shown a time of diagnosis of type 2 diabetes. The management regimens of relationship between increasing levels of glucose and increased patients with type 2 diabetes should be tailored to the individual risk for both micro- and macrovascular complications. The thresh- patient, aiming for glycemic targets as close to normal as possible Ͻ old for increased cardiovascular risk occurs in the “nondiabetic” (A1c 6% when agents that do not cause hypoglycemia are used) range, and even minimal elevations of glucose are associated with and, in most people, as early as possible (5). increased cardiovascular risk (1). There is much less randomized It is known that both weight loss and increased physical controlled trial evidence on the value of glucose control in pre- activity can improve insulin resistance and thus improve hy- venting macrovascular disease. In the UK Prospective Diabetes perglycemia (6). It is recommended that lifestyle modification strategies be used in all patients with type 2 diabetes, whether Study, intensive treatment of individuals with newly diagnosed medication is used or not (5). type 2 diabetes reduced the risk for myocardial infarction by 16% (P ϭ 0.052), amputation or death from peripheral vascular disease Mechanisms of Action of Oral by 35% (P ϭ 0.15), fatal myocardial infarction by 6% (P ϭ 0.63), nonfatal myocardial infarction by 21% (P ϭ 0.057), fatal sudden Antihyperglycemic Agents The antihyperglycemic agents that are available include the death by 46% (P ϭ 0.047), and amputation by 39% (P ϭ 0.099) (2). insulin secretagogues (sulfonylureas and meglitinides), met- Every 1% reduction in glycosylated hemoglobin (HbA ) was 1c formin, ␣-glucosidase inhibitors, and the thiazolidinediones. associated with reductions in risk of 21% for any end point related These agents differ greatly in terms of mechanisms of action, to diabetes, 21% for diabetes-related deaths, 14% for myocardial efficacy, side effect profiles, and cost (Table 1). Ͻ infarction, and 37% for microvascular complications (all P Sulfonylureas and meglitinides increase insulin secretion by 0.0001) (3). No threshold of risk was observed for any end point; pancreatic ␤ cells. Metformin decreases hepatic gluconeogenesis. however, the lowest risk was in individuals with an HbA1c in the ␣-Glucosidase inhibitors delay the absorption of glucose from Ͻ normal range ( 6%). In the obese subset, metformin therapy was starch and sucrose, attenuating postprandial glucose increases. associated with a lower risk for diabetes-related end points (P ϭ Thiazolidinediones are potent, highly selective agonists for perox- 0.0034) and all-cause mortality (P ϭ 0.021) compared with the isome proliferator–activated receptor-␥. Thiazolidinediones de- other intensive therapies (4). crease insulin resistance, enhance peripheral disposal of glucose, and have some effect on hepatic production of glucose.

Glucose-Lowering Efficacy of Address correspondence to: Dr. Jean-Franc¸ois Yale, McGill Nutrition and Food Antihyperglycemic Agents Science Centre, 687 Pine Avenue West, Room H685, Montreal, Quebec, H3A 1A1, Canada. Phone: 514-843-1665; Fax: 514-843-1706; E-mail: jean-francois.yale@ A large number of clinical trials comparing the efficacy of the mcgill.ca oral antihyperglycemic agents have been completed. The com-

Table 1. Summary of oral antihyperglycemic agents

Weight Cost Generic Name HbA1c Drop Hypoglycemia Initial Dose Maximum Dose per Use in Renal failure Gain Month

Biguanides 1.0–1.5 No No Contraindicated when creatinine clearance is Ͻ60 ml/min because of the risk for lactic acidosis metformin No 250 mg twice daily 850 mg three times a day $20 Thiazolidinediones 1.0–1.5 Yes Metabolism not affected in renal failure, but fluid retention may be a problem, with heightened risk for congestive heart failure rosiglitazone 4 mg/d 8 mg/d or 4 mg twice daily $50 pioglitazone 15 mg/d 45 mg/d $60 Sulfonylureas 1.0–1.5 Yes Yes The metabolism of all sulfonylureas is affected by renal failure; this will initially require decreases in dosages, glyburide 2.5–5 mg/d 10 mg twice daily $35 and eventually avoidance of these agents gliclazide 40–80 mg twice daily 160 mg twice daily $40 or MR 30 mg or MR 120 mg/d glimepiride 1 mg/d 8 mg/d $50 glipizide 5 mg/d or XL 5 mg/d 20 mg twice daily or XL 20 mg/d $30 tolbutamide 500 mg twice daily 500 mg four times a day $15 chlorpropamide 100 mg/d 500 mg/d $10 acetohexamide 250 mg/d 500 mg three times a day $10 Meglitinides 1.0–1.5 Yes Yes Can be used in the presence of renal failure as the pharmacokinetics are unaffected repaglinide 0.5 mg three times a day 4 mg three times a day $0 nateglinide 120 mg three times a day 180 mg three times a day $0 ␣-Glucosidase 0.5–1.0 No No Contraindicated in renal failure inhibitors 2005 S7–S10, 16: Nephrol Soc Am J acarbose 25 mg three times a day 100 mg three times a day $40 miglitol 25 mg three times a day 100 mg three times a day $60 J Am Soc Nephrol 16: S7–S10, 2005 Oral Antihyperglycemic Agents and Renal Disease S9 parative glycemic effect of some of these agents are well known Other Side Effects when used as monotherapy and in combination with other oral Metformin is associated with a high frequency of nausea and antihyperglycemic agents or insulin (Table 1). In general, met- diarrhea. This side effect can be reduced by taking the pills in formin, the thiazolidinediones, and the insulin secretagogues the middle of the meal. Acarbose, by inhibiting proximal ab- (sulfonylureas and repaglinide) have approximately equivalent sorption of starch and sucrose, can cause flatulence. This side Ͼ efficacy (reductions in HbA1c of 1.0 to 1.5% compared with effect is present in 70% of patients in the first months, but placebo) (7–11). Higher reductions are generally seen in treat- some adaptation occurs and the magnitude of this side effect ment-naı¨ve patients and those with higher baseline glycemic decreases over subsequent months. The major risk of met- values (9,11). Treatment with acarbose seems somewhat less formin is lactic acidosis. It is a rare side effect, occurring par- effective with reductions in HbA1c of 0.5 to 1% compared with ticularly in the presence of renal failure, hepatic dysfunction, or placebo in previously untreated patients (12–14). tissue ischemia. Most of the oral antihyperglycemic agents can be combined The thiazolidinediones rosiglitazone and pioglitazone have with each other and insulin therapy with additive effects. The been associated with small decreases in hemoglobin in patients initial use of combinations of submaximal doses of oral antihy- with type 2 diabetes, likely explained by a modest increase in perglycemic agents produces more rapid and improved glyce- plasma volume. Edema was noted with greater frequency in mic control compared with monotherapy with the maximal patients who were treated with pioglitazone or rosiglitazone dose of one agent, without a significant increase in side effects compared with those who were treated with placebo in clinical (15). trials. A recent American Diabetes Association/American Therapy with exogenous insulin is recommended when in- Heart Association position statement recommended avoiding dividuals have not achieved glucose targets with oral agents the use of thiazolidinediones in the presence of class III or IV either alone or in combination (5). Oral agents may be contin- NYHA congestive heart failure (CHF) and to use with caution ued or added on to insulin therapy as necessary. in those who have less severe CHF or are at risk for CHF (history of heart failure, previous myocardial infarction or an- gina, hypertension, left ventricular hypertrophy, significant Side Effects of Antihyperglycemic Agents aortic or mitral valve disease, age Ͼ70 yr, diabetes duration Hypoglycemia Ͼ10 yr, preexisting edema or treatment with loop diuretics, Within the oral antihyperglycemic agents, insulin secreta- development of edema or weight gain on thiazolidinediones, gogues are associated with the highest occurrence of hypogly- insulin co-administration, and chronic renal failure) (16). cemic episodes, ranging from 10 to 35% (4,7,10). However, severe episodes that require intervention are relatively rare (0 Use of Antihyperglycemic Agents in Patients to 1.3%). Metformin, thiazolidinediones, and ␣-glucosidase inhibitors do not usually cause hypoglycemia when used alone with Renal Failure Metformin is contraindicated in renal failure because of the but can potentiate the hypoglycemic potency of insulin secre- associated risk for lactic acidosis. It can be used at low dosages tagogues (4,7,9,12,14). Repaglinide and nateglinide are partic- up to a creatinine clearance of 30 to 60 ml/min and should be ular in having a rapid onset and short duration of action and avoided with clearances Ͻ30 (17). Although the metabolism of can be given at mealtimes. They have been shown to decrease thiazolidinediones is unaffected by renal failure, they must be hypoglycemia when given to patients with irregular mealtimes. used with caution in this context because of their volume- Insulin therapy in patients with type 2 diabetes is associated retaining effect with a risk for heart failure (18). with the highest frequency of hypoglycemia (16 to 34%) The sulfonylureas (glyburide, gliclazide, glipizide, gliben- (4,7,12), although this frequency is much lower than that seen clamide, tolbutamide, and chlorpropamide) have increased po- with insulin therapy of patients with type 1 diabetes. tency as the renal function decreases and are contraindicated in Although ␣-glucosidase inhibitors do not cause hypoglyce- severe renal failure (19). The nonsulfonylurea insulin secreta- mia, they may prevent sucrose or starch from being absorbed in gogues repaglinide and nateglinide can be used in renal failure a timely manner for the treatment of hypoglycemia caused by without dose adjustments (20). ␣-Glucosidase inhibitors (acar- other treatments. Patients who take ␣-glucosidase inhibitors bose and miglitol) are contraindicated in renal failure. therefore must use glucose (dextrose tablets), grape juice, or honey to treat hypoglycemia. Conclusion In the absence of contraindications, metformin should be Body Weight preferred over other agents for a number of reasons. Compared Insulin secretagogues, both sulfonylureas and repaglinide, with insulin secretagogues in general, metformin has equal are associated with an increase in body weight compared with potency and a low risk for hypoglycemia and causes less placebo of up to approximately 4.5 kg over 3 yr (7). The use of weight gain. In obese patients, there is strong clinical evidence rosiglitazone and pioglitazone in the treatment of type 2 dia- of reduced microvascular and macrovascular outcomes. betes has been associated with weight gain of 1 to 3 kg. ␣-Glu- In the presence of contraindications or intolerance to met- cosidase inhibitors demonstrate neutral effects on weight, formin or when metformin alone does not result in optimal whereas metformin is usually associated with no weight gain control, thiazolidinediones should be used. They should be and occasionally weight loss. favored over insulin secretagogues because they are not asso- S10 Journal of the American Society of Nephrology J Am Soc Nephrol 16: S7–S10, 2005 ciated with hypoglycemia. Compared with acarbose, thiazo- 10. Goldberg, Einhorn D, Lucas C, Rendell MS, Damsbo P, lidinediones have more potent antihyperglycemic effects. Sul- Huang WC, Strange P, Brodows RG: A randomized, pla- fonylureas and other insulin secretagogues should be reserved cebo controlled trial of repaglinide in the treatment of type for combination therapy because of the risk for hypoglycemia. II diabetes. Diabetes Care 21: 1897–1903, 1998 In chronic renal failure, the oral agents that can be used 11. Jovanovic L, Dailey G 3rd, Huang WC, Strange P, Gold- therefore include the insulin secretagogues repaglinide and stein BJ: Repaglinide in type 2 diabetes: A 24-week, fixed- nateglinide and the thiazolidinediones (rosiglitazone and pio- dose efficacy and safety study. J Clin Pharmacol 40: 49–57, glitazone) with caution. Insulin also can be used safely in renal 2000 12. Chiasson JL, Josse RG, Hunt JA, Palmason C, Rodger NW, failure. Ross SA, Ryan EA, Tan MH, Wolever TMS: The efficacy of acarbose in the treatment of patients with non-insulin- References dependent diabetes mellitus. Ann Intern Med 121: 928–935, 1. Gerstein HC: Dysglycaemia: A cardiovascular risk factor. 1994 Diabetes Res Clin Pract 97 40[Suppl]: S9–S14, 1998 13. Rodger N, Chiasson JL, Josse R, Hunt JA, Palmason C, Ross 2. UK Prospective Diabetes Study (UKPDS) Group: Intensive SA, Ryan EA, Tan MH, Wolever TM: Clinical experience blood-glucose control with sulphonylureas or insulin com- with acarbose: Results of a Canadian multicentre study. pared with conventional treatment and risk of complica- Clin Invest Med 18: 318–324, 1995 tions in patients with type 2 diabetes (UKPDS 33). Lancet 14. Fischer S, Hanefeld M, Spengler M, Boehme K, Temelkova- 352: 837–853, 1998 Kurktschiev T: European study on dose-response relation- 3. Stratton IM, Adler AI, Neil AW, UK Prospective Diabetes ship on acarbose as a first-line drug in insulin-dependent Study Group: Association of glycaemia with macrovascu- diabetes mellitus: Efficacy and safety of low and high lar and microvascular complications of type 2 diabetes doses. Acta Diabetol 35: 34–40, 1998 (UKPDS 35): Prospective observational study. BMJ 321: 15. Garber AJ, Larsen J, Schneider SH, Piper BA, Henry D; 405–412, 2000 Glyburide/Metformin Initial Therapy Study Group: Si- 4. UK Prospective Diabetes Study (UKPDS) Group: Effect of multaneous glyburide/metformin therapy is superior to intensive blood-glucose control with metformin on com- component monotherapy as an initial pharmacological plications in overweight patients with type 2 diabetes (UK- treatment for type 2 diabetes. Diabetes Obes Metab 4: 201– PDS 34). Lancet 352: 854–865, 1998 208, 2002 5. Canadian Diabetes Association Clinical Practice Guide- 16. Nesto RW, Bell D, Bonow RO, Fonseca V, Grundy SM, lines Expert Committee: Canadian Diabetes Association Horton ES, Le Winter M, Porte D, Semenkovich CF, Smith 2003 Clinical Practice Guidelines for the Prevention and S, Young LH, Kahn R: Thiazolidinedione use, fluid reten- Management of Diabetes in Canada. Can J Diabetes tion, and congestive heart failure: A consensus statement 27[Suppl 2], 2003 from the American Heart Association and American Dia- 6. Bogardus C, Ravussin E, Robbins DC, Wolfe RR, Horton betes Association. Diabetes Care 27: 256–263, 2004 ES, Sims EA: Effects of physical training and diet therapy 17. Gan SC, Barr J, Arieff AI, Pearl RG: Biguanide-associated on carbohydrate metabolism in patients with glucose in- lactic acidosis. Case report and review of the literature. tolerance and non-insulin-dependent diabetes mellitus. Di- Arch Intern Med 152: 2333–2336, 1992 abetes 33: 311–318, 1984 7. UKPDS 13: Relative efficacy of randomly allocated diet, 18. Chapelsky MC, Thompson-Culkin K, Miller AK, Sack M, sulphonylurea, insulin, or metformin in patients with Blum R, Freed MI: Pharmacokinetics of rosiglitazone in newly diagnosed non-insulin dependent diabetes followed patients with varying degrees of renal insufficiency. J Clin for 3 years. BMJ 310: 83–88, 1995 Pharmacol 43: 252–259, 2003 8. DeFronzo RA, Goodman AM, and the Multicentre Met- 19. Krepinsky J, Ingram AJ, Clase CM: Prolonged sulfonylu- formin Study Group: Efficacy of metformin in patients rea-induced hypoglycemia in diabetic patients with end- with non-insulin-dependent diabetes mellitus. N Engl stage renal disease. Am J Kidney Dis 35: 500–505, 2000 J Med 333: 541–549, 1995 20. Schumacher S, Abbasi I, Weise D, Hatorp V, Sattler K, 9. Garber A, Duncan T, Goodman A, Mills DJ, Rohlf JL: Sieber J, Hasslacher C: Single- and multiple-dose pharma- Efficacy of metformin in type II diabetes: Results of a cokinetics of repaglinide in patients with type 2 diabetes double-blind, placebo controlled, dose-response trial. Am J and renal impairment. Eur J Clin Pharmacol 57: 147–152, Med 103: 491–497, 1997 2001