Sulfonylureas Differ in Effects on Ischemic Preconditioning—Is it Time to Retire Glyburide?

Matthew C. Riddle

Division of Endocrinology, Diabetes, and Clinical Nutrition, Department of Medicine Oregon Health and Science, University Portland, Oregon 97201

Address all correspondence and requests for reprints to: Matthew C. Riddle, M.D., Section of Diabetes L-345, Oregon Health and Science University, 3181 SW Sam Jackson Park Road, Portland, Oregon 97201.

Sulfonylureas have been used for type 2 diabetes for over 50 yr and are still the leading class of oral antihyperglycemic agents. Their popularity is based on familiarity and habit, of course, but also their ease of administration (as once-daily tablets, in many cases), reliable effectiveness for recently diagnosed patients, lack of symptomatic side effects other than hypoglycemia, and low cost. However, concern about possible adverse cardiovascular effects has waxed and waned for several decades. It dates from publication of the findings of the University Group Diabetes Program (UGDP) study, in which cardiovascular mortality rates of type 2 diabetic patients treated with the sulfonylurea tolbutamide exceeded those of patients treated with placebo or insulin (1). An intense debate about the risks of using sulfonylureas ensued. Subsequently, criticism of the design of the UGDP study and lack of confirmation of risk from other studies led to a partial return of confidence in these agents (2). Although tolbutamide is no longer popular, other sulfonylureas including chlorpropamide, glipizide, gliclazide, glimepiride, and glyburide (known in Europe as glibenclamide) have had important roles. Glyburide is currently the most widely used sulfonylurea in the United States.

Recent data from large clinical studies have been reassuring about the safety of sulfonylureas. The United Kingdom Prospective Diabetes Study (UKPDS) found no enhancement of cardiovascular events or mortality by treatment with sulfonylureas, and instead a trend toward protection against myocardial infarction (3). Of the 1,573 subjects in the UKPDS who were randomized to begin therapy with a sulfonylurea, 50% used chlorpropramide, 39% glyburide, and 11% glipizide. In addition, a retrospective analysis of data from 25,035 older diabetic patients taking a sulfonylurea and 17,861 using no antihyperglycemic agent who were hospitalized for myocardial infarction compared morbidity and mortality rates for the two treatments (4). The patients using sulfonylurea, 68% of whom were taking glyburide, showed no tendency toward higher morbidity or mortality after the infarction.

Nevertheless, concern about cardiovascular risk has lingered. The package insert for sulfonylureas mandated by the U.S. Food and Drug Administration still includes a bold-print warning about possible cardiovascular risks. Also, a retrospective analysis of patients with diabetes who had balloon angioplasty after myocardial infarction reported increased early mortality (odds ratio 2.7 after adjustment for a number of covariates) in 67 persons taking sulfonylureas vs. 118 using insulin or lifestyle therapy alone (5). Although the sulfonylureas taken by patients in this study were not specified, it is likely that glyburide was most frequently used. Despite important limitations, such as the small number of subjects, lack of randomization, and imbalance between the groups (the patients using sulfonylureas were older and had worse myocardial function), this study attracted considerable attention.

Moreover, concern about cardiovascular risk associated with sulfonylureas has been supported by description of a plausible underlying mechanism: impairment of myocardial ischemic preconditioning. This is a process by which transitory ischemia "conditions" the myocardium in a protective fashion, allowing greater tolerance of subsequent ischemia (6, 7). This phenomenon has been shown experimentally to limit anginal pain, minimize irreversible tissue injury, and protect myocardial function. ATP-dependent potassium (KATP) channels in myocardial cells have an important role in this process (8, 9). Pharmacological agents that open KATP channels have a protective effect similar to that of prior ischemia, and agents closing the channels oppose preconditioning by ischemia. Similar (but not identical) KATP channels are also present in the pancreatic ß-cell (10, 11). The clinically useful actions of sulfonylureas derive from binding to a subunit of the ß-cell KATP channel complex (termed the sulfonylurea receptor), leading to closure of the channel and stimulation or potentiation of insulin secretion (12, 13). Sulfonylureas bind also to cardiovascular KATP channels, although less well than to ß-cell KATP channels (11), presumably promoting closure of the channel and opposing ischemic preconditioning. This property of sulfonylureas has the potential to increase cardiovascular risk in patients with diabetes.

In addition to animal experiments supporting this hypothesis (14, 15), a number of human studies testing the clinical relevance of interactions of sulfonylureas with the myocardium have been reported, most of them using glyburide. For example, an in vitro study of human atrial tissue compared samples obtained during surgery from six diabetic patients who had been taking glyburide and one taking glipizide with tissue from a group of four insulin-taking patients plus six without known diabetes (16). The atria from patients taking a sulfonylurea had less effective protection of contractility during severe hypoxia by prior exposure to hypoxia. Persons without diabetes have been studied during repeated balloon dilation done therapeutically for coronary disease. In one study of this kind, 20 patients were randomized to receive either a single oral dose of 10 mg glyburide or placebo just before the procedure (17). Protection against electrocardiographic changes and pain resulting during the second balloon dilation in those given placebo was abolished by glyburide pretreatment. In a second study accompanying angioplasty, pretreatment with a single iv dose of glyburide or glimepiride was compared with saline infusion (18). Electrocardiographic evidence of ischemia was more pronounced during the second period of ischemia after glyburide administration than after glimepiride or placebo. Patients with type 2 diabetes previously using a sulfonylurea have been studied at the time of dipyridamole stress testing, after randomization to pretreatment with insulin or glyburide (19). Dipyridamole stress led to greater worsening of echocardiographically determined myocardial function when glyburide was given than during insulin treatment. However, direct testing of the effect of glyburide on ischemic preconditioning in patients with diabetes using the best available model, repeated balloon dilation during angioplasty, has not been done until now.

A study by Lee and Chou (20), in this issue of JCEM, fills this void, and it is the most comprehensive and convincing demonstration of adverse cardiac effects of glyburide, to date. The investigators tested the effects of both glyburide and glimepiride in 20 nondiabetic persons and 23 patients with diabetes who had been using a sulfonylurea chronically, during the course of repeated therapeutic balloon dilation of coronary arteries. The doses of glyburide (10 mg) and glimepiride (2 mg) were selected as equivalent in antihyperglycemic effectiveness and were given an hour before the cardiac studies. The patients with diabetes had hemoglobin A1c values averaging about 8%, indicating a level of control that is relatively typical for patients in a primary care setting. Three different ways of assessing myocardial adaptation to prior ischemia were used: a clinical measure, subjectively reported anginal pain; an electrocardiographic measure, ST-segment changes; and a metabolic measure, lactate balance across the coronary bed. With no drug pretreatment or glimepiride pretreatment in the nondiabetic group, or with glimepiride pretreatment in the diabetic group, clear improvements of pain score, ST-segment shift, and net coronary lactate production were all seen after preconditioning. Quantitatively, the mean percentage reductions from the first ischemic challenge (for nondiabetic-no treatment, nondiabetic-glimepiride, and diabetic-glimepiride, respectively) were 59%, 48%, and 45% for pain; 59%, 49%, and 35% for ST-segment changes; and 76%, 83%, and 61% for lactate balance. In contrast, no improvement of any of these measures was seen after the second balloon inflation when glyburide had been administered. Thus, strong protection by preconditioning occurred in untreated nondiabetic subjects that was unaffected by glimepiride but much impaired by glyburide. In the diabetic patients, protection by preconditioning occurred with glimepiride but not in the presence of glyburide. In addition, the values for lactate balance suggested that diabetes itself may have impaired preconditioning.

How does this report advance our knowledge? The main way is by extending earlier reports that glyburide can impair ischemic preconditioning in persons without diabetes to patients with type 2 diabetes, with consistent findings in symptomatic, electrophysiologic, and metabolic endpoints. These findings deepen the shadow of suspicion over glyburide, but they do not prove that use of glyburide has done harm to patients, especially in light of the negative findings of the large clinical studies mentioned earlier. It is quite possible that whatever effect glyburide has on ischemic preconditioning is counterbalanced by other effects that are beneficial. The most obvious of these is improvement of glycemic control, as in the UKPDS. Another possible protective effect is an antiarrhythmic action of glyburide that seems to occur under certain circumstances (21). Thus, the effects of glyburide on cardiovascular outcomes may remain neutral or favorable despite an undesirable interaction with ischemic preconditioning. Properly designed and powered outcome trials will be required to settle this question decisively.

However, such trials may never be done, in part because of the findings concerning glimepiride in the study by Lee and Chou (20). With the same experimental procedures that showed adverse effects of glyburide, ischemic preconditioning was unaffected by glimepiride, confirming earlier evidence that this sulfonylurea lacks deleterious cardiac effects (14, 18). Other sulfonylureas such as glipizide and gliclazide have not been studied as carefully but may also be free of such effects. If other sulfonylureas do not affect ischemic preconditioning, yet have equivalent ability to control glycemia, why should they be tarred with the same brush as glyburide? It would be very desirable to obtain the clinical advantages of sulfonylurea therapy without the suspicion that a cardiovascular price is being paid. Given the limited present use of tolbutamide, there seems no reason for intensive study to determine whether it shares the actions of glyburide on the human myocardium. However, it is worth mentioning that on theoretical grounds the design of the UGDP protocol might have uniquely favored an effect on cardiovascular KATP channels. Glycemic control in the tolbutamide arm of the UGDP study was scarcely better than in the placebo arm, and, therefore, little metabolic benefit would have been available to neutralize any adverse effect of treatment. In addition, the large (1000 mg) dose of tolbutamide taken in the morning by UGDP subjects assigned this treatment would result in a very high plasma level for several hours thereafter, followed by a rapid decline due to hepatic clearance. The peak level required for adequate duration of action from this short-acting agent might have been high enough to favor significant binding to the lower-affinity myocardial KATP channel complex as well as the ß-cell channels. Longer-acting sulfonylureas such as glimepiride, extended-release glipizide, and extended-release gliclazade lack such prominent peak blood levels, and thus may be more ß-cell specific for pharmacokinetic reasons.

In addition to having effects on the myocardium that set it apart, glyburide also may have a greater tendency to cause hypoglycemia than other sulfonylureas. In the UKPDS, glyburide caused twice as much hypoglycemia as did chlorpropamide (3), despite chlorpropamide’s undesirable dependence on good renal function for normal clearance. And in a recent population-based study from an area of Germany where the two leading sulfonylureas were glyburide and glimepiride, the usage-adjusted rate of severe hypoglycemia (requiring a visit to an emergency facility) was 10-fold higher with glyburide (22). As with the myocardial effects, the mechanism of glyburide’s tendency to cause hypoglycemia is not well defined. Differences in pharmacokinetics, binding properties, and formation and clearance of metabolites may all contribute.

A third finding of the study by Lee and Chou (20), that ischemic preconditioning may be impaired in diabetes independent of any pharmacotherapy, was less definitive but also of interest. If this is confirmed by future studies, it will be important to know whether this effect results from irreversible structural changes or reversible metabolic factors, which might include hyperglycemia, high levels of free fatty acids, or low levels of insulin. If metabolic factors are involved, as seems likely, recent reports of favorable effects of insulin on the heart in the setting of coronary disease take on additional meaning (23, 24, 25). Perhaps we should expand our thinking beyond avoiding factors that may impair ischemic preconditioning, to include more frequent use of insulin to restore normal adaptive mechanisms. In the acute care setting, all oral antihyperglycemic agents lack the ability reliably to maintain metabolic control, in large part because of suppression of endogenous insulin secretion by physiological stress. Also, each class of oral agents has distinctive disadvantages: besides the issue of sulfonylureas and ischemic preconditioning, there is concern about lactic acidosis associated with metformin and congestive heart failure with thiazolidinediones. Even with evidence already available, a strong case can be made not just for avoiding glyburide in the hospital setting but, generally, for replacing oral therapy with insulin on hospitalization.

In conclusion, this study suggests we may consider retiring glyburide, an old drug that has served well for many years, in favor of other antihyperglycemic agents that lack the potential to interfere with protective myocardial mechanisms. In the ambulatory setting these would include glimepiride and perhaps other sulfonylureas, metformin, thiazolidinediones, and, of course, insulin. Glyburide seems especially problematic for hospital use, and more study of the possible benefits of insulin therapy accompanying acute illness is desirable.

Acknowledgments

Footnotes

Abbreviations: KATP, ATP-dependent potassium; UGDP, University Group Diabetes Program; UKPDS, United Kingdom Prospective Diabetes Study.

Received December 15, 2002.

Accepted December 15, 2002.

References

  1. Meinert CL, Knatterud GL, Prout TE, Klimt CR 1970 A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. II. Mortality results. Diabetes 19(Suppl 2):789–830
  2. Feinglos MN, Bethel MA 1999 Therapy of type 2 diabetes, cardiovascular death, and the UGDP. Am Heart J 138:S346–S352
  3. United Kingdom Prospective Diabetes Study (UKPDS) Group 1998 Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352:837–853[CrossRef][Medline]
  4. Jollis JG, Simpson Jr RJ, Cascio WE, Chowdhury MK, Crouse JR, Smith Jr SC 1999 Relation between sulfonylurea therapy, complications, and outcome for elderly patients with acute myocardial infarction. Am Heart J 138:S376–S380
  5. Garratt KN, Brady PA, Hassinger NL, Grill DE, Terzic A, Holmes Jr DR 1999 Sulfonylurea drugs increase early mortality in patients with diabetes mellitus after direct angioplasty for acute myocardial infarction. J Am Coll Cardiol 33:119–124[CrossRef][Medline]
  6. Deutsch E, Berger M, Kussmaul WG, Hirschfeld Jr JW, Herrmann HC, Laskey WK 1990 Adaptation to ischemia during percutaneous transluminal coronary angioplasty. Clinical, hemodynamic, and metabolid features. Circulation 82:2044–2051[Abstract]
  7. Kloner RA, Yellon D 1994 Does ischemic preconditioning occur in patients? J Am Coll Cardiol 24:1133–1142[Medline]
  8. Terzic A, Jahangir A, Kurachi Y 1995 Cardiac ATP-sensitive K+ channels: regulation by intracellular nucleotides and K+ channel-opening drugs. Am J Physiol 269:C525–C545
  9. O’Rourke B 2000 Myocardial KATP channels in preconditioning. Circ Res 87:845–855[Abstract/Free Full Text]
  10. Aguilar-Bryan L, Clement JP, Gonzalez G, Kunjilwar K, Babenko A, Bryan J 1998 Toward understanding the assembly and structure of KATP channels. Physiol Rev 78:227–245[Abstract/Free Full Text]
  11. Lazdunski M 1996 Ion channel effects of antidiabetic sulfonylureas. Horm Metab Res 28:488–495[Medline]
  12. Ashcroft FM 1996 Mechanisms of the glycaemic effects of sulfonylureas. Horm Metab Res 28:456–463[Medline]
  13. Kramer W, Muller G, Geisen K 1996 Characterization of the molecular mode of action of the sulfonylurea, glimepiride, at ß-cells. Horm Metab Res 28:464–468[Medline]
  14. Geisen K, Vagh A, Krause E, Papp JG 1996 Cardiovascular effects of conventional sulfonylureas and glimepiride. Horm Metab Res 28:496–507[Medline]
  15. Mocanu MM, Maddock HL, Baxter GF, Lawrence CL, Standen NB, Yellon DM 2001 Glimepiride, a novel sulfonylurea, does not abolish myocardial protection afforded by either ischemic preconditioning or diazoxide. Circulation 103:3111–3116[Abstract/Free Full Text]
  16. Cleveland JC, Meldrum DR, Cain BS, Banerjee A, Harken AH 1997 Oral sulfonylurea hypoglycemic agents prevent ischemic preconditioning in human myocardium. Two paradoxes revisited. Circulation 96:29–32[Abstract/Free Full Text]
  17. Tomai F, Crea F, Gaspardone A, Versaci F, DePaulis R, Penta de Peppo A, Chiariello L, Gioffre PA 1994 Ischemic preconditioning during coronary angioplasty is prevented by glibenclamide, a selective ATP-sensitive K+ channel blocker. Circulation 90:700–705[Abstract]
  18. Klepzig H, Kober G, Matter C, Luus H, Schneider H, Boedeker KH, Kiowski W, Amann FW, Gruber D, Harris S, Burger W 1999 Sulfonylureas and ischemic preconditioning. A double-blind, placebo-controlled evaluation of glimepiride and glibenclamide. Eur Heart J 20:439–446[Abstract/Free Full Text]
  19. Scognamiglio R, Avogara A, Vigili de Kreutzenberg S, Negut C, Palisi M, Bagolin E, Tiengo A 2002 Effects of treatment with sulfonylurea drugs or insulin on ischemia-induced myocardial dysfunction in type 2 diabetes. Diabetes 51:808–812[Abstract/Free Full Text]
  20. Lee T-M, Chou T-F 2003 Impairment of myocardial protection in type 2 diabetic patients. J Clin Endocrinol Metab 88:531–537[Abstract/Free Full Text]
  21. Tosaki A, Szerdahelyi P, Engelman RM, Das DK 1993 Potassium channel openers and blockers: do they possess proarrthymic or antiarrhythmic activity in ischemic and reperfused rat hearts? J Pharmacol Exp Ther 267:1355–1362[Abstract]
  22. Holstein A, Plaschke A, Egberts E-H 2001 Lower incidence of severe hypoglycemia in patients with type 2 diabetes treated with glimepiride versus glibenclamide. Diabetes Metab Res Rev 17:467–473[CrossRef][Medline]
  23. McNulty PH, Pfau S, Deckelbaum LI 2000 Effect of plasma insulin level on myocardial blood flow and its mechanism of action. Am J Cardiol 85:161–165[CrossRef][Medline]
  24. Malmberg K for the DIGAMI Study Group 1997 Prospective randomized study of intensive insulin treatment on long-term survival after acute myocardial infarction in patients with diabetes mellitus. BMJ 314:1512–1552[Abstract/Free Full Text]
  25. Malmberg K, McGuire DK 1999 Diabetes and acute myocardial infarction: the role of insulin therapy. Am Heart J 138:S381–S386