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.ORIGINAL ARTICLE Natural History of Macrovascular Disease in Type 2 Diabetes Role of insulin resistance Michael Stern, MD In this article, I attempt to provide a theoretical framework for understanding the relationship between diabetes and heart disease. In other articles, individual risk factors for macrovascular disease are covered in greater detail, so my contribution is more summary in nature. The classical concept has been that atherosclerosis is a complication of diabetes. But an alternative possibility is that both diseases "spring from the same soil," i.e., that they have common antecedents, both genetic and environmental. In fact, as I shall attempt to demonstrate, I believe that both of these concepts are correct. I will begin by discussing the "common soil" hypothesis, after which I will discuss macrovascular disease as a more conventional complication of diabetes. My perspective is that the cluster of risk factors that constitute the insulin resistance syndrome (IRS) forms part of the common soil from which both diabetes and cardiovascular disease spring. I approach this concept from the standpoint of an epidemiologist and will attempt to demonstrate that each of the elements of the IRS is, in fact, a risk factor not only for diabetes but for cardiovascular disease, as well. However, before we cast our eyes forward to the consequences of this cluster of risk factors, let us take just a moment to look backward and ask what produces the syndrome. There is considerable evidence that early life experiences, probably relating to fetal and infant nutrition, have long-term effects on adult diseases, notably cardiovascular disease and diabetes. But it is also interesting to note that the IRS itself is determined by these early life nutritional experiences. Data from the U.K. indicate that the risk of developing IRS in adult life is a function of birth weight (1). Those in the lowest birth-weight category have an 18-fold increased odds of developing IRS in adult life compared with those in the highest birth-weight category (1). A progressive decline in risk was observed as birth weight increased, suggesting that those with low birth weights, perhaps reflecting some adverse fetal nutritional experience, have, for whatever reason, a higher risk of developing the IRS in adult life. We have confirmed these findings in San Antonio and have shown that the same phenomenon occurs not only in European Americans but also in Mexican Americans (2). In the San Antonio study, the prevalence of IRS declined progressively across tertiles of birth weight in both ethnic groups (2). Thus, the scheme that I will present is that these fetal and early life nutritional deficiencies produce IRS, including abdominal obesity, and that IRS constitutes the "common soil" that leads to both type 2 diabetes and cardiovascular disease. Abundant data from prospective cohort studies indicate that virtually all of the elements of the IRS are in fact risk factors for the future development of type 2 diabetes (3). For example, 8-year follow-up data from San Antonio indicate that BMI, ratio of subscapular to triceps skinfold (centrality index), waist circumference, ratio of waist to hip circumference, plasma glucose concentration, impaired glucose tolerance, fasting insulin concentration, dyslipidemia of the high-triglyceride low-HDL cholesterol type, blood pressure, and clinically diagnosed hypertension are all risk factors for type 2 diabetes (4,5). The question then arises: Are these same IRS elements also risk factors for cardiovascular disease? The situation here is a bit more complicated than in the case of diabetes. Certain of the elements of IRS are unquestioned cardiovascular risk factors, e.g., blood pressure and low HDL cholesterol concentration. As these are well-recognized cardiovascular risk factors, I will not dwell on them. Instead, I will discuss the roles of the more controversial risk factors—namely, triglyceride, plasma glucose, and serum insulin concentrations. If it can be established that the majority of these elements of IRS are cardiovascular risk factors, then that would support the idea that IRS is a common antecedent for both conditions. With respect to triglycerides, it is probably best to view this lipid not as a risk factor in its own right, but primarily as a marker for other genuine cardiovascular risk factors. For example, there is a close relationship between triglycerides and small, dense LDL particles (6), which are thought to be more atherogenic than native LDL (7). Triglycerides are also associated with increased levels of plasminogen activator inhibitor-1, which is also widely considered to be a cardiovascular risk factor. Turning to insulin, I believe that the studies that purport to show insulin to be a cardiovascular risk factor can be divided into three categories: the classic, the neoclassic, and the postmodern. The "classic" studies were published in 1979 and 1980. These studies from Helsinki, Finland; Paris, France; and Busselton, Australia, all showed that hyperinsulinemia was predictive of the future development of cardiovascular disease in men (8). (The Busselton study was the only "classic" study that included women, and insulinemia did not emerge as a significant predictor of cardiovascular disease in women in this study. Recently, however, data from the Atherosclerosis Risk in Communities Study (9) indicated the reverse pattern, i.e., that insulinemia was a cardiovascular risk factor in women, but not in men). Next, came the "neoclassic" studies, which were largely negative (8). So, at this point the evidence was mixed at best. The following argument has been offered as an explanation for how both the classic and the neoclassic studies might have given spurious results (10). Many of these studies, it is said, failed to adequately exclude diabetic subjects because oral glucose tolerance tests were not performed. Some of the studies did not exclude diabetics at all, and others excluded only diagnosed cases. As it is known that about half of diabetic individuals are undiagnosed, failure to perform glucose tolerance tests would result in occult diabetic subjects being included in some of these cohorts. Moreover, diabetic subjects have an elevated cardiovascular risk unrelated to insulinemia and insulin resistance, perhaps due to such phenomena as glycosylation of tissue proteins (see below). Finally, it was argued that diabetic subjects have "spurious" hyperinsulinemia, that what they actually have is hyperproinsulinemia, and that the nonspecific immunoassays in use when the classic and neoclassic studies were performed would have measured this as insulin. This combination of circumstances would lead one to falsely conclude that insulin was a risk factor, because one would be inadvertently measuring proinsulin in patients who were at increased risk of cardiovascular disease for unrelated reasons. Enter the "postmodern" studies. Two such studies have been published that deal with the proinsulin objection. The Quebec Cardiovascular Study is a prospective study in which a nested case-control design was used to study 91 individuals who developed cardiovascular disease and 105 matched control subjects (11). Baseline specimens from these individuals were available, and insulin was measured using a specific immunoassay that did not cross-react with proinsulin. This study also measured HDL, which, as it had not been measured in most of the earlier studies, has also been cited as a weakness of those studies. The results of the Quebec Cardiovascular Study showed that an increase of one standard deviation in specific insulin levels conferred a 70% increase in cardiovascular risk. This risk was only minimally reduced to ~60% after adjusting for other lipid risk factors, including HDL. The British Regional Heart Study (12) also addressed the specific insulin issue. In this prospective study, 5,000 individuals were followed for 11 years. A large number of end points—specifically, 260 nonfatal myocardial infarctions and 261 cardiovascular or coronary fatalities—were recorded. Again, baseline insulin was measured using a specific radioimmunoassay that did not cross-react with proinsulin. The results of this study suggested that there could be a threshold for the cardiovascular disease–enhancing effects of insulin. There was minimal evidence of a dose-response effect below the 95th percentile for serum specific insulin concentration, but above this percentile there was a marked increase in cardiovascular risk. Now let us to turn to glycemia as a cardiovascular risk factor. For many years, the role of glycemia was controversial. Some of this controversy, I believe, resulted from confusion with respect to two related, but separate questions. The first pertains to whether the level of glycemia in the normal range is a risk factor for cardiovascular disease in the general population. The second pertains to the role of glycemia as a risk factor specifically among diabetic subjects. Recently, Laakso and Kuusisto (13) reviewed this topic and summarized a number of studies that showed that glycemia was a risk factor for cardiovascular disease in diabetic subjects. It has been said that the availability of glycohemoglobin measurements, which provide an integrated measure of glycemia, has facilitated the demonstration of glycemia as a cardiovascular risk factor. In fact, in many of the positive studies cited by these authors, this phenomenon was readily demonstrated with single fasting glucose measurements at baseline. All of the studies reviewed by Laakso and Kuusisto involved Caucasian populations. Our group recently extended these findings to Hispanics (14). Again, a quite strong effect on cardiovascular mortality in diabetic subjects was demonstrated using a single fasting glucose value. Evidence of a role for glycemia in the development of atherosclerosis per se was reported by the Pathobiological Determinants of Atherosclerosis in Youth Study (15). In this study, young people who died from traumatic causes, e.g., automobile accidents, violence, etc., underwent autopsies. Samples of their coronary arteries were obtained and stained for fatty streaks and raised atherosclerotic lesions using a quantitative technique. Glycohemoglobin levels measured postmortem were strongly related to the extent of coronary artery surface area that was involved with fatty streaks and raised lesions. There are a number of theoretical mechanisms whereby glycemia could produce atherosclerosis. These include glycation of collagen matrix in the vascular wall, which could lead to cross-linking of collagen fibers and trapping of LDL (8). Also, LDL itself can become glycated, which increases its half-life and its susceptibility to oxidation (8). Thus, there are both empirical data from epidemiological studies implicating glycemia as a cardiovascular risk factor and a number of theoretical mechanisms that could be postulated to explain this relationship. One possibility is that there is a threshold for these glycation effects such that, among people with diabetes, there is a correlation between the degree of glycemia and large-vessel atherosclerosis, whereas among people without diabetes there is no such correlation, because they fall below the glycemic threshold. This concept would offer an explanation for why studies of diabetic subjects have been quite consistent in showing a relationship between level of glycemia, whether measured by plasma glucose concentration or glycated hemoglobin, and cardiovascular end points, whereas efforts to document a similar relationship in nondiabetic subjects have been much more inconsistent. A further point is that if hyperglycemia, the hallmark of the diabetic state, directly causes atherosclerosis, we may regard macrovascular disease as a genuine complication of diabetes, not merely the result of shared genetic and environmental antecedents. These relationships are summarized in Fig. 1, which depicts both the "common soil" hypothesis and macrovascular disease as a true diabetic complication.
Figure 1 also suggests the possibility that insulin, even if it is atherogenic in nondiabetic individuals, may nevertheless be of benefit in reducingatherosclerotic complications in diabetic individuals. It is important to keep in mind that the epidemiological studies implicating hyperinsulinemia as a cardiovascular risk factor pertain to the effects of endogenousinsulin in nondiabetic subjects, not to the administration of exogenousinsulin to diabetic subjects. The latter at least have a theoretical opportunity to benefit from insulin, whereas the former do not. Figure 1 focuses on plasma glucose lowering as the mechanism whereby exogenous insulin therapy reduces cardiovascular risk among diabetic subjects. But the possibility also exists that insulin could improve cardiovascular risk by other mechanisms as well, e.g., by improving the lipid profile. In fact, there is a large body of literature regarding the benefits of insulin therapy on lipid profiles. Most of these studies, however, are uncontrolled, have small numbers of subjects, and/or are of relatively short duration, e.g., from 6 weeks to 3 months. There are, however, at least two large randomized long-term controlled studies on this topic. The first is the Diabetes Control and Complications Trial (DCCT), which included a randomized control group and lasted for many years. The DCCT showed that intensive insulin management led to a modest lowering of both total and LDL cholesterol concentration, a moderate lowering of triglyceride concentration, and essentially no effect on HDL cholesterol (16). Thus, the DCCT represents a long-term randomized study that supports the concept that insulin has beneficial effects on lipid profiles. The other randomized clinical trial on this topic is the VA Cooperative Study on Glycemic Control and Complications in Type II Diabetes (17). In this study, 153 individuals were followed for 30 months. A marked difference in glycohemoglobin and in glucose concentrations between the two randomized groups was produced. However, there was virtually no effect on lipid concentrations, although this may have been a consequence in part of differential lipid-lowering therapy in the two groups. It is not necessary, however, to postulate a beneficial effect of exogenous insulin on lipid profiles to explain a beneficial effect on cardiovascular risk. If glycemia is a risk factor for cardiovascular disease, then controlling glycemia itself ought to be of benefit with respect to this end point. And such a benefit could offset any disadvantages that insulin might have, even if it were atherogenic. What about clinical trials on the effects of exogenous insulin treatment on atherosclerotic risk in patients with diabetes? Unfortunately, the data are quite limited. The familiar University Group Diabetes Program study compared groups given a variable dose of insulin, a fixed dose of insulin, and a placebo (18). The cardiovascular mortality was essentially identical in all three groups. One way to interpret these results is that insulin is neutral insofar as cardiovascular disease is concerned. This could happen if the potential benefits of exogenous insulin therapy on macrovascular disease, as outlined above, were exactly offset by its atherogenic potential. But even if that were the case, tight control of glycemia, involving aggressive insulin treatment when required, would still be strongly indicated because of the known beneficial effects of tight control on microvascular complications. It is sufficient to conclude that insulin is neutral with respect to macrovascular disease—to exonerate it, given its established benefits on microvascular complications as demonstrated in the DCCT (19). Moreover, the DCCT itself showed a benefit of intensive insulin management on macrovascular end points (16), although the number of cases was small because of the young age of this type 1 diabetic cohort. Finally, I conclude by emphasizing the inadequacies of these clinical trial data with respect to the question at hand, namely, the benefits or risks of exogenous insulin therapy on macrovascular disease. In my opinion, a convincing clinical trial directed specifically at this issue is desperately needed to resolve the present controversy. Just as the cholesterol controversy was never satisfactorily resolved until persuasive clinical trial data became available, and just as the controversy over tight control of diabetes was never satisfactorily resolved until the DCCT data became available, the controversy over the atherogenic potential of insulin will never be resolved until adequate clinical trials on this topic are performed. References 2. Valdez R, Athens MA, Thompson GH, Bradshaw BS, Stern MP: Birthweight and adult health outcomes in a biethnic U.S. population. Diabetologia 37:624–631, 1994 3. Stern MP: Do NIDDM and cardiovascular disease share common antecedents? Ann Intern Med 124:110–116, 1996 4. Stern MP, Morales PA, Valdez RA, Monterrosa A, Haffner SM, Mitchell BD, Hazuda HP: Predicting diabetes: moving beyond impaired glucose tolerance. Diabetes 42:706–714, 1993 5. Wei M, Gaskill SP, Haffner SM, Stern MP: Waist circumference as the best predictor of NIDDM compared to body mass index, waist/hip ratio and other anthropometric measurements in Mexican Americans: a 7-year prospective study. Obes Res 5:16–23, 1997 6. Haffner SM, Mykkänen L, Valdez RA, Paidi M, Stern MP, Howard BV: LDL size and subclass pattern in a biethnic population. Arterioscler Thromb 13:1623–1630, 1993 7. Austin MA, Breslow JL, Hennekens CH, Buring JE, Willett WC, Krauss RM: Low density lipoprotein subclass patterns and risk of myocardial infarction. JAMA 260:1917–1921, 1988 8. Stern MP: The effect of glycemic control on the incidence of macrovascular complications of type 2 diabetes. Arch Fam Med 7:155–162, 1998 9. Folsom AR, Szklo M, Stevens J, Liao F, Smith R, Eckfeldt JH: A prospective study of coronary heart disease in relation to fasting insulin, glucose, and diabetes: the Atherosclerosis Risk in Communities (ARIC) Study. Diabetes Care 20:935–942, 1997 10. Ferrara A, Barrett-Conner EL, Edelstein SL: Hyperinsulinemia does not increase the risk of fatal cardiovascular disease in elderly men or women without diabetes: the Rancho Bernardo Study, 1984–1991. Am J Epidemiol 140:857–869, 1994 11. Depres J-P, Lamarche B, Mauriege P, Cantin B, Dagenais GR, Moorjani S, Lupien P-J: Hyperinsulinemia as an independent risk factor for ischemic heart disease. N Engl J Med 334:952–957, 1996 12. Perry IJ, Wannamethee SG, Whincup PH, Shaper AG, Walker MK, Alberti KGMM: Serum insulin and incident coronary heart disease in middle-aged British men. Am J Epidemiol 144:224–234, 1996 13. Laakso M, Kuusisto J: Epidemiologic evidence for the association of hyperglycaemia and atherosclerotic vascular disease in non-insulin dependent diabetes mellitus. Ann Med 28:415–418, 1996 14. Wei M, Gaskill SP, Haffner SM, Stern MP: Effects of diabetes and level of glycemia on all-cause and cardiovascular mortality: the San Antonio Heart Study. Diabetes Care 21:1167–1172, 1998 15. McGill HC, McMahan A, Malcolm GT, Oalmann MC, Strong JP, and the Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group: Relation of glycohemoglobin and adiposity to atherosclerosis in youth. Arterioscler Thromb Vasc Biol 15:431–440, 1995 16. Diabetes Control and Complications Trial Research Group: Effect of intensive diabetes management on macrovascular events and risk factors in the Diabetes Control and Complications Trial. Am J Cardiol 75:894–903, 1995 17. Abraira C, Colwell JA, Nutall FQ, Sawin CT, Johnson-Nagel N, Comstock JP, Emanuele NV, Levin SR, Henderson W, Lee HS, VA CSDM Group: Veterans Affairs Cooperative Study on Glycemic Control and Complications in Type II Diabetes (VA CSDM): results of the feasibility trial. Diabetes Care 18:1113–1123, 1995 18. University Group Diabetes Program: A study of the effects of hypoglycemic agents on vascular complications in patients with adult-onset diabetes. II. Mortality results. Diabetes 19 (Suppl. 2):785–830, 1970 19. The Diabetes Control and Complications Trial Research Group: The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 329:977–986, 1993 From the Division of Clinical Epidemiology, the University of Texas Health Science Center, San Antonio, Texas. Address correspondence and reprint requests to Michael Stern, MD, Division of Clinical Epidemiology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr., San Antonio, TX 78284-7873. E-mail: stern@uthscsa.edu. Received for publication 6 July 1998 and accepted in revised form 1 October 1998. Abbreviations: DCCT, Diabetes Control and Complications Trial; IRS, insulin resistance syndrome. 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 |
