These pages are best viewed with Netscape version 3.0 or higher or Internet Explorer version 3.0 or higher. When viewed with other browsers, some characters or attributes may not be rendered correctly.SUMMARY Overview of Current Therapeutic Options Session summary Matthew C. Riddle, MD The preceding sections have reviewed various pharmacotherapeutic tactics for type 2 diabetes. Dr. Feinglos described using oral antihyperglycemic agents singly or in combinations; Dr. Buse covered the combination of oral agents with insulin; and Dr. Berger concluded with a review of treatment with insulin alone. Much material was included, as no aspect of pharmacotherapy of hyperglycemia has remained unchanged in the last 5 years. More agents are available than ever before, and our ways of using them have multiplied. With such rapid change, it is no surprise that many questions were raised both in the lectures themselves and in the ensuing discussions. Dr. Feinglos began by listing four classes of oral agents and their
properties. Sulfonylureas, biguanides, Discussion of oral therapy focused on two main points. First, how can we best measure and discuss the therapeutic power of these agents relative to each other? Decline of glucose or HbA1c from baseline values, or compared with placebo, varies considerably from study to study and seems generally greatest when initial glycemic control is very poor. This suggests that some of the benefit is related to a decline of glucose toxicity following any reduction of glucose, with partial recovery of insulin secretion and action. Moreover, if only the patients with the highest levels of glucose in a population are selected for treatment, some regression toward the mean that is not a true treatment effect can confuse interpretation of uncontrolled studies. As an example of these two influences on treatment responses, patients in one uncontrolled study (1) who started treatment with acarbose when HbA1c was >12% showed a decline of 4.6% HbA1c, in contrast to a much smaller effect when HbA1c was lower initially. It was suggested that the final level of control achieved may sometimes be a more useful measure of treatment than the drop from baseline. However, this approach is limited in a different way. The proportion of patients reaching a target level such as HbA1c 8% will be strongly affected by the baseline level of control, so that a study started when glycemic control is already quite good may seem to show a better result than one beginning with very poor control, even if the same treatment is used. The other focus of discussion was the side effects of oral agents, notably the possible association of cardiovascular risk with sulfonylureas, and the weight gain accompanying treatment with thiazolidinediones. Concern about the increased cardiovascular mortality during tolbutamide treatment in the University Group Diabetes Program study (2) has been reawakened by recent evidence that vasodilation during myocardial ischemia may be impaired by tolbutamide or glyburide, through binding of sulfonylureas to the ATP-dependent potassium channel complex in cardiovascular tissues (3). On the other hand, other studies done at about the same time as the University Group Diabetes Program suggested a protective effect of lower doses of tolbutamide on cardiovascular risk (4,5). The evidence is still incomplete, but better understanding of this issue could change therapeutic tactics toward lower doses or toward use of sulfonylureas such as glimepiride that may have less effect on cardiovascular tissues. Several people commented on weight gain with troglitazone, the thiazolidinedione now in clinical use. A gain of 61 pounds in one patient was described, and a large study showed an average gain of 13 pounds when troglitazone was added to a sulfonylurea (6). Some of this gain is likely to be fluid, but increased visceral fat has been seen in animal studies. Hence, adding a thiazolidinedione may postpone the need for insulin but may cause as much weight gain as insulin. Better understanding of these and other unwanted effects of commonly used oral agents is needed. Dr. Buse's presentation covered the progressive nature of type 2 diabetes, leading to secondary failure of oral agents, singly or combined, and the eventual need to start insulin. He addressed the dilemma of having to choose between switching to insulin alone or using combinations of insulin with one or more oral agents, without enough objective comparisons of medical outcomes for the different regimens. Each of the classes of oral agents adds specific benefits when used together with insulin. Continuing a sulfonylurea while starting evening insulin allows a smooth and reliable transition with a simple regimen (7). Metformin often limits weight gain accompanying insulin. Acarbose should improve postprandial hyperglycemia in patients using simple insulin regimens. Troglitazone markedly reduces the insulin requirement in the most insulin-resistant patients, who in the past have had great difficulty achieving good glycemic control. Dr. Buse concluded by emphasizing that combinations of insulin with oral agents have become established options and that the combined regimen used should depend on the unique features of each patient and the aims of treatment. Discussion of this presentation raised the question of whether troglitazone used with insulin will, over time, cause enough weight gain to neutralize the benefits of improved sensitivity to insulin. No long-term data are available to answer this question at present. Also, the point was made that simply reducing the dosage of insulin needed for a given level of control is not a strong argument for adding or continuing an oral agent. Better glycemic control or some advantage in convenience or patient acceptance are more convincing benefits. One questioner asked what might lead to a decision to abandon two injections of insulin plus oral agents in favor of three or four injections of insulin alone for type 2 diabetes. This was acknowledged to be a therapeutic alternative chosen by some physicians and patients, allowing better dietary flexibility and perhaps greater opportunity for caloric restriction. Dr. Berger began his discussion of insulin treatment with concerns that sulfonylureas may enhance cardiovascular risk, as mentioned above, whereas evidence that treatment with injected insulin increases vascular risk is lacking. He then commented on the lack of firm evidence that treatment of type 2 diabetes, whether with insulin or with oral agents, can reduce either macrovascular or microvascular complications, in contrast to the excellent evidence for reduction of microvascular complications of type 1 diabetes in the Diabetes Control and Complications Trial. This led to the proposal that treatment goals be determined in part by the age of the patient. Good glycemic control (HbA1c 7–7.5%) might be pursued for younger type 2 patients, whereas higher levels (up to 8.5–9%) might be acceptable for older patients less likely to benefit because of shorter life expectancy. In either case, Dr. Berger argued forcefully for insulin as the best tested method of treatment, with good evidence for benefit and almost no evidence suggesting cardiovascular toxicity. The Diabetes Mellitus Insulin Glucose Infusion in Acute Myocardial Infarction Study provides support for this view. This 5-year Swedish trial of insulin infusion followed by aggressive multiple-injection insulin treatment after myocardial infarction in patients with type 2 diabetes showed clear reductions of both short- and long-term mortality (8). He described the systematic education of patients and primary care providers as a requirement for successful insulin treatment. He noted that twice-daily 70/30 insulin and before-meal regular insulin were the two main regimens. Lively discussion followed, with several themes. One person commented that the newer oral agents and insulins seem to have improved our ability to reach good glycemic control from ~40% 15 years ago to perhaps 60% now. Whether this will translate to reduced cardiovascular mortality and morbidity was discussed at length. At the time of the discussion the 10-year results of the U.K. Prospective Diabetes Study (UKPDS) had not been released, and several people speculated on their probable significance. One suggested that the UKPDS would show no benefit from the 1% reduction of HbA1c that has been achieved compared with the control group, because the study would have ended earlier if it had shown benefits. Another person responded that stopping the trial early would have required very large differences between groups, and that even smaller differences in outcomes might be both statistically and clinically significant. Another issue was whether the glycemic target should be determined by a given patient's age. On the one hand, older individuals have less time to live and therefore less chance of preventing complications with good glycemic control, as well as a possibly increased risk of harm from vigorous insulin treatment. On the other hand, many older individuals are quite healthy and will survive 20 years or more, and their duration of diabetes at the time of observation may be either short or long. If a person is well at age 65 but has already had diabetes for 10–15 years, as is very common, withholding treatment for hyperglycemia seems inadvisable, because the benefits of treatment should be considerable. In contrast, an older person with other serious medical problems who has just been found to have diabetes may not need such vigorous treatment for hyperglycemia. Dr. Berger was questioned about his scheme for systematically using insulin for most patients with type 2 diabetes in Germany. He acknowledged that not all physicians in Germany hold the same views, and that in addition to the use of insulin, education of physicians and patients was an essential part of his program (9). Applying experience in Germany to the U.S. may not be straightforward, because of the greater cultural diversity in the U.S. People of African, Asian, Hispanic, or Native American heritage are much more common; the mean level of adiposity is greater; and the age of onset of diabetes is earlier in the U.S. compared with European countries. It was suggested that oral combinations with insulin may be more attractive under these conditions. Looking at the discussion of pharmacotherapy as a whole, there was a level of consensus that surprised some participants. For example, there seemed to be general agreement that effective glycemic control should limit the microvascular complications of type 2 diabetes, despite the lack of definitive results like those provided for type 1 diabetes by the Diabetes Control and Complications Trial. Moreover, even lacking proof that glycemic control can reduce the macrovascular complications that are the leading hazard for type 2 patients, most participants seemed convinced that vigorous efforts to achieve good glycemic control are worthwhile. With a few exceptions, the group favored seeking HbA1c in the 7–8% range for a majority of patients. On the risk side of the question, the degree of concern expressed by most participants about sulfonylureas as potentiators of cardiovascular risk was low. Also, almost no support was voiced for the view that treatment with injected insulin may increase cardiovascular risks. Much of the time, the speakers and the group focused directly on tactics to translate these beliefs into action and to improve glycemic control in the large population of people with type 2 diabetes in all parts of the world. Many comments addressed the problems posed by the progressive nature of type 2 diabetes, requiring step-wise intensification of treatment and the need to individualize treatment. Perhaps the most recurring theme was individualization. Although the application of management principles to medical practice has fostered the use of standardized regimens, or algorithms, the diversity among patients with type 2 diabetes does not lend itself to a one-size-fits-all approach. Therapeutic tactics in different countries and different ethnic groups clearly must differ. For individual patients, the age of onset, duration of diabetes, presence of diabetic complications or other illnesses, the abilities and preferences of the patient, and the pathophysiological features of the case should all be considered. The problem posed by the need for individualization is considerable. It is among the main obstacles to improving glycemic treatment at a time when standardized practices are emphasized in an effort to contain costs. It is entwined with debate about the role of endocrinologists and other specialists in management of diabetes and the costs incurred or saved by providing or not providing individualized care. Some topics did not arise or received less attention than they deserve. Three in particular merit comment. The risks and benefits of thiazolidinediones, which have the potential to improve metabolic control and change patterns of use of insulin, could not be adequately addressed because of lack of data. Meanwhile, troglitazone is widely used and the weight gain, fluid retention, and especially the liver toxicity (10,11) associated with its use, as well as its hypothesized cardiovascular benefits, are active concerns in the medical community. We need much more information on how to identify patients who are at risk of liver toxicity and on whether or not this can occur with other thiazolidinediones. In the same way, surprisingly little guidance is available for physicians on the practical problems associated with treating type 2 diabetes with insulin. Among many unanswered questions, consider the following: how can we best identify patients with adult-onset type 1 diabetes who need immediate treatment with insulin? How great is the risk of hypoglycemia in typical type 2 patients? What can be done to predict and limit weight gain? When are three- or four-injection regimens advantageous? Another enormous question is the cost-effectiveness of treatment of type 2 diabetes. Reliable information on both medical outcomes and the costs of various forms of treatment is in very short supply. The financial benefits and costs of treatment are especially hard to estimate, being partly direct, partly indirect, and partly related to the perceived value of quality-adjusted years of life. The results of the UKPDS presented in Barcelona in September 1998 turned out to be quite helpful (12,13). Notably, microvascular complications were reduced 25% by intensified treatment. When adjusted for the smaller improvement of HbA1c in this trial, this effect is equivalent to the benefit of intensive treatment of type 1 diabetes shown in the Diabetes Control and Complications Trial. Intensive treatment with insulin or sulfonylureas had no adverse effect on vascular events. Instead these treatments led to a 16% reduction of myocardial infarction. Whether this difference implies clinical benefit seems to depend on the temperament and prior opinions of the observer, since the statistical significance is equivocal (P = 0.052). In general, insulin, sulfonylureas, and metformin had similar effects, except that metformin caused no weight gain. Concurrent intensified treatment of hypertension (with either atenol or captopril) had an impressive beneficial effect on both microvascular and macrovascular events (14). Most observers interpret these findings as supporting earlier and more intensive treatment of both hyperglycemia and hypertension in type 2 diabetes, with insulin included in the regimen when needed. Positive as these findings are, they do not fully demonstrate the cost-effectiveness of the treatments used. What's more, newer agents such as the thiazolidinediones have not been tested in this way. Much as we want to practice "evidence-based" medicine, there may be too little relevant evidence. In this vacuum, decisions may be driven mainly by the inertia of past habits or by perceptions of cost. To take the next step toward rational use of treatments for diabetes, we will have to find the resources to test the improved options now at our disposal. References 2. Goldner MG, Knatterud GL, Prout TE, the University Group Diabetes Program: Effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. III. Clinical implications of the UGDP results. J Am Med Assoc 218:1400–1410, 1971 3. Bijlstra PJ, Russel FGM, Thien T, Lutterman JA, Smits P: Effects of tolbutamide on vascular ATP-sensitive potassium channels in humans: comparison with literature data on glibenclamide and glimepiride. Horm Metab Res 28:512–516, 1996 4. Paasikivi J, Wahlberg F: Preventive tolbutamide treatment and arterial disease in mild hyperglycemia. Diabetologia 7:323–327, 1971 5. Sartor G, Shersten B, Carlstrom S, Melander A, Norden A, Persson G: Ten-year follow-up of subjects with impaired glucose tolerance: prevention of diabetes by tolbutamide and diet regulation. Diabetes 29:41–49, 1980 6. Horton ES, Whitehouse F, Ghazzi MN, Vanable RC, the Troglitazone Study Group, Whitcomb RW: Troglitazone in combination with sulfonylurea restores glycemic control in patients with type 2 diabetes. Diabetes Care 21:1462–1469, 1998 7. Riddle MC, Schneider J, the Glimepiride Combination Group: Beginning insulin treatment of obese patients with evening 70/30 insulin plus glimepiride versus insulin alone. Diabetes Care 21:1052–1057, 1998 8. Malmberg K, the DIGAMI Study Group: Prospective randomised study of intensive insulin treatment on long-term survival after myocardial infarction in patients with diabetes mellitus. BMJ 314:1512–1515, 1997 9. Berger M, Jorgens V, Flatten G: Health care for persons with non-insulin-dependent diabetes mellitus: the German experience. Ann Intern Med 124:153–155, 1996 10. Gitlin N, Julie NL, Spurr CL, Lim KN, Juarbe HM: Two cases of severe clinical and histologic hepatotoxicity associated with troglitazone. Ann Intern Med 129:36–38, 1998 11. Neuschwander-Tetri BA, Isley WL, Oki JC, Ramrakhiani S, Quiason SG, Phillips NJ, Brunt EM: Troglitazone-induced hepatic failure leading to liver transplantation: a case report. Ann Intern Med 129:38–41, 1998 12. UK Prospective Diabetes Study Group: Intensive blood-glucose control with suphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352:837–853, 1998 13. UK Prospective Diabetes Study Group: Effect of intensive blood-glucose with metformin on complications in overweight patients with type 2 diabetes (UKPDS 34). Lancet 352:854–865, 1998 14. UK Prospective Diabetes Study Group: Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes (UKPDS 38). BMJ 317:703–713, 1998 From the Oregon Health Science University, Section of Diabetes, Portland, Oregon. Address correspondence and reprint requests to Matthew C. Riddle, MD, Oregon Health Science University, Section of Diabetes, 3181 SW Sam Jackson Park Rd., Portland, OR 97201. Received for publication 12 August 1998 and accepted in revised form 12 November 1998. Abbreviations: UKPDS, U.K. Prospective Diabetes Study. A table elsewhere in this issue shows conventional and Système International (SI) units and conversion factors for many substances. This article is based on presentations at a conference organized by the Indiana University Diabetes Research and Training Center. The conference and the publication of this article were made possible by an unrestricted educational grant from Eli Lilly and Company. Copyright © 1999 American Diabetes Association For Technical Issues contact webmaster@diabetes.org |
-glucosidase inhibitors, and thiazolidinediones all have different
mechanisms of action and different side effects, and so allow treatment of patients with
differing pathophysiology. Two or more may also be used in combination with additive
benefits. Because biguanides,