One of the most important questions in diabetes management

Special Feature and Commentary Does Lowering of Blood Glucose Improve Cardiovascular Morbidity and Mortality? Nayyar Iqbal* and Arthur H. Rubenstein *Section of Endocrinology, Philadelphia Veterans Affairs Medical Center, and University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania Clin J Am Soc Nephrol 3: 163 167, 2008. doi: 10.2215/CJN.05041107 One of the most important questions in diabetes management is whether long-term glycemic control can reduce the risk for micro- and macrovascular complications in diabetes. Dr. Hadler in this issue of the Clinical Journal of the American Society of Nephrology raises his concerns about the use of glycosylated hemoglobin (HbA 1c ) as a surrogate marker of diabetes control and questions whether lowering HbA 1c with the currently available oral hypoglycemic agents makes a clinically important long-term outcome difference for patients with type 2 diabetes. He is particularly concerned about the lack of evidence of cardiovascular risk reduction with HbA 1c reduction in patients with type 2 diabetes. He further comments that we were blinded by the results of Diabetes Control and Complications Trial (DCCT) that showed reduction in microvascular complications after meticulous blood glucose control. However, it took 17 yr before any discernible advantage in terms of macrovascular disease was documented and the results are far from a dramatic advance in patients with type 1 diabetes; hence, one should not expect any more benefit in the middle-aged population with insulin resistance. He further cited the United Kingdom Prospective Diabetes Study (UKPDS), in which, after several years of intervention, there was no significant improvement in macrovascular complications. He questions the validity of UKPDS secondary analysis and challenges the new screening guidelines for detection and treatment of diabetes. Dr. Hadler supports his argument of lack of beneficial effects of effectiveness of oral hypoglycemic agents by reviewing A Diabetes Outcomes Progression Trial (ADOPT) results that showed relatively modest overall beneficial effects and an unexpected finding of higher fracture rates in women. Finally, he admonishes that in our focus of bringing down HbA 1c, we are exposing our patients to higher risk for adverse effects such as weight gain, hypoglycemia, and fractures:...talk about missing the forest for the trees. We address some of the issues raised by Dr. Hadler and express our belief that improved glycemic control in patients with type 2 diabetes should be the therapeutic goal given the Published online ahead of print. Publication date available at www.cjasn.org. Correspondence: Dr. Nayyar Iqbal, 1 Maloney, 3400 Spruce Street, Philadelphia, PA 19104. Phone: 215-823-4190; Fax: 215-823-6374; E-mail: nayyar.iqbal2@ med.va.gov available evidence at this time. Let us take the question of HbA 1c as a surrogate marker first. We agree that surrogate markers are not perfect and that a surrogate predicted outcome will be reliable only if there is a validated causal connection between a change in the surrogate and the clinically important outcome and if the surrogate fully captures all of the effects of treatment on that outcome. That is precisely the reason that we need to assess far more than a single study to make a decision about the adequacy of any surrogate marker. We also agree that a randomized trial designed to determine the effect of an intervention on a well-defined clinical outcome of unequivocal importance to patients is the only definitive solution to the surrogate outcome dilemma. However, when patients risks for serious morbidity or mortality are high, this wait and see strategy may result in serious health issues. Surrogate markers are cost-effective, can be reliable, and, if used appropriately, can provide very useful information. Because even Dr. Hadler admits that definitive outcomes take a long time to develop, we have essentially two choices: Either we (1) wait for the perfect study to be done for the longest period of time and once incontrovertible outcome data become available only then change our practices or (2) use our clinical judgment and available evidence to support our clinical decisions. Despite its limitations, we believe that HbA 1c is the best available method of measuring glycemic status, and several studies have demonstrated its dose effect relationship with both micro- and macrovascular complications. Numerous studies have demonstrated unequivocally that higher HbA 1c is associated with an increased risk for complications (1 6). This has been reaffirmed by large, multicenter, prospective studies that have shown that lowering of HbA 1c is associated with lessening of complications (7 10). On the basis of these data, we believe that HbA 1c does fulfill the criteria of an excellent surrogate marker and should be used, within reason, as a measure of glycemic control to reduce longer term diabetes-related complications. Additional studies in different patient populations may help to define the age-, race-, and gender-specific normal values and thus may change the absolute goals, but trends and consequences associated with hyperglycemia can be reliably measured by changes in HbA 1c (11). Dr. Hadler did admit that there was significant reduction Copyright 2008 by the American Society of Nephrology ISSN: 1555-9041/301 0163

164 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 3: 163 167, 2008 in microvascular complications with reduction in HbA 1c, but he focused primarily on the relatively modest macrovascular effects. Let us examine the microvascular effects first. Type 1 diabetes is associated with long-term complications that affect the eyes, kidneys, and peripheral and autonomic nervous systems (1 8). Hyperglycemia seems to play a central role in the pathophysiology of these complications (12). Epidemiologic studies have demonstrated a strong association between the level of hyperglycemia and the occurrence of these diabetic complications. The Wisconsin Epidemiologic Study of Diabetic Retinopathy (WESDR) enrolled a total of 10,135 patients who had diabetes, were receiving primary care in an 11-county area of southwestern Wisconsin, and were identified by chart review during 1979 to 1980. The study demonstrated a higher prevalence of retinopathy and nephropathy in patients with type 1 diabetes (2 4). Prevalence of retinopathy increased with duration of diabetes, from 6% at 4 yr to 73% at 14 yr. Risk for developing retinopathy increased with increasing duration, worse glycemic control, and age of onset up to 20 yr. The authors commented that improvements in diabetes care leading to better glycemic control may have contributed to the much lower prevalence and reduced severity of retinopathy observed than what would have been expected on the basis of a previous report from the same region of Wisconsin (13). The data obtained from DCCT were the first long-term prospective evidence of reduction in complications associated with diabetes. The study demonstrated a 56 to 70% reduction in retinopathy, neuropathy, and nephropathy with approximately a 2% reduction in HbA 1c during a mean intervention period of 6.5 yr (conventional versus intensive) (7). This outcome was associated with increased hypoglycemic episodes but clearly demonstrated that hyperglycemia was associated with significant microvascular complications and that reduction of hyperglycemia was preventive. These effects are not trivial, and they were associated with a significantly reduced morbidity and an improved quality of life. The same findings have been reproduced by other prospective trials, and, in general, a 1% reduction in HbA 1c is associated with an approximately 20 to 37% reduction in risk for microvascular complications (7 10). The evidence described indicates a microvascular risk benefit with HbA 1c reduction in both type 1 and type 2 diabetes; however, emerging evidence suggests a macrovascular risk reduction as well. Epidemiologically, Balkau et al. (14) combined the results of three large European studies and came to the conclusions that men in the upper 20% of the 2-h glucose distributions and those in the upper 2.5% for fasting glucose had a significantly higher risk for all-cause mortality in comparison with men in the lower 80% of these distributions, with age-adjusted hazard ratios of 1.6 (95% confidence interval [CI] 1.4 to 1.9) and 2.0 (95% CI 1.6 to 2.6), respectively, for the upper 2.5%. For death from cardiovascular and coronary heart disease, men in the upper 2.5% of the 2-h and fasting glucose distributions were at higher risk, with age-adjusted hazard ratios for coronary heart disease of 1.8 (95% CI 1.4 to 2.4) and 2.7 (95% CI 1.7 to 4.4), respectively. They concluded that if early intervention aimed at lowering blood glucose concentrations could be shown to reduce mortality, then it would be justified to lower aggressively the levels of both 2-h and fasting glucose, which define diabetes. Recently, Shankar et al. (15) reported that elevated HbA 1c is associated with higher risk for all-cause and cardiovascular mortality in patients with type 1 diabetes. The DCCT/Epidemiology of Diabetes Interventions and Complications (DCCT/EDIC) Study demonstrated that intensive insulin treatment reduces the risk for cardiovascular disease (CVD) among patients with type 1 diabetes. One of the most important goals of the EDIC trial was the prospective evaluation of the long-term effects of differences in previous diabetes treatment (conventional versus intensive) during the DCCT on the subsequent development and progression of CVD in patients with type 1 diabetes. The EDIC trial demonstrated that, as compared with conventional therapy, intensive insulin therapy reduced the risk for any CVD by 42% and for nonfatal myocardial infarction, stroke, or death from cardiovascular causes by 57%. Furthermore, improved glycemic control, as assessed by the decrease in HbA 1c values during the DCCT, seemed to account for much of the cardiovascular benefit attributed to intensive insulin therapy. The results strongly suggest that the mean of 6.5 yr of intensive diabetes therapy during the DCCT had a sustained effect on ameliorating the risk for CVD (8). Although Dr. Hadler pointed out that it took 17 yr to show this difference, this conclusion is neither entirely true nor unexpected. The cumulative incidence for any predefined cardiovascular outcome curves for intensive versus conventional therapy started separating within 2 yr of follow-up, favoring intensive control. Development of macrovascular disease takes time, and one would expect a prolonged time course before developing a clinically measurable outcome. Also, one must bear in mind that this was not a straightforward comparison of a conventional with an intensive therapy group. Once patients started follow-up, both groups received the intensive insulin regimen. This study design could have been the cause of the occurrence of fewer overall events and the reason for the delay in expression of macrovascular complications, thus supporting the notion that intensive therapy is beneficial. Similarly, DCCT/EDIC (16) showed that after 6 yr, the carotid intima-media thickness was significantly greater in patients with diabetes than in control subjects. The mean progression of the intimamedia thickness was significantly less in the group that had received intensive therapy during the DCCT than in the group that had received conventional therapy. Dr. Hadler s expressed his disappointment with the UKPDS findings that despite its a priori aim to induce cardiovascular risk reduction, it failed to show a statistically significant difference at the end of the follow-up period. We draw attention to the fact that the UKPDS clearly showed a significant reduction in microvascular disease but also showed a trend toward reduction in cardiovascular events as well. The P value did not reach the 0.05 cutoff, but the direction was clear with a 16% reduction in CVD (P 0.052). This was achieved despite a relatively modest difference of

Clin J Am Soc Nephrol 3: 163 167, 2008 Blood Glucose and Cardiovascular Morbidity and Mortality 165 HbA 1c between the two arms of the study (median HbA 1c 7% in intensive versus 7.9% in the conventional arm). The UKPDS published their follow-up observational analysis that showed highly significant associations between the development of each of the complications of diabetes, including mortality, across the wide range of exposure to glycemia that occurred in patients with type 2 diabetes (17). This association remained after adjustment for other known risk factors, including age at diagnosis, gender, ethnic group, systolic BP, lipid concentrations, smoking, and albuminuria. Each 1% reduction in HbA 1c was associated with a 37% decrease in risk for microvascular complications and a 21% decrease in the risk for any end point or death related to diabetes. These results suggest that the effect of hyperglycemia itself may account for at least part of the excess cardiovascular risk observed in patients with diabetes compared with individuals without diabetes beyond that explained by the conventional risk factors of dyslipidemia, hypertension, and smoking. The rate of increase of relative risk for microvascular disease with hyperglycemia was greater than that for myocardial infarction, which emphasizes the crucial role of hyperglycemia in the cause of small-vessel disease and may explain the greater rate of microvascular complications seen in populations with less satisfactory control of glycemia. Although Dr. Handler was not very impressed with the degree of association between glycemic control and macrovascular events, we believe that the study does emphasize the strong relationship between glycemic control and all diabetic complications and thus underscores the need for tighter control. The increased risk for cardiovascular events in type 2 diabetes cannot be explained simply on the basis of traditional cardiovascular risk factors (17) and may in part be related to the clustering of metabolic and inflammatory factors (metabolic syndrome) in patients with diabetes (18,19). This syndrome is believed to have a genetic basis, but the precise genetic/molecular pathways involved are unknown (20). A number of key molecular signals that may modulate accelerated atherosclerosis in type 2 diabetes have been identified. Functional single-nucleotide polymorphisms have been discovered in genes that regulate these pathways, and many are associated with an increased risk for atherosclerosis. Whether hyperglycemia plays a role in the pathophysiology of atherosclerosis mediated through activation of inflammatory molecules is an area of active investigation. However, this association, if it turns out to be true, would explain the cause effect relationship and support the importance of early and aggressive glucose lowering in all patients with type 2 diabetes. Finally, the issue of safety of the currently available hypoglycemic drugs is raised, and we agree that this is a very important question. Does lowering of blood glucose with the currently available oral hypoglycemic agents result in more harm than benefits? Dr. Hadler s arguments remind us of the editorial by Siperstein et al. (21) published in the New England Journal of Medicine in 1977 that raised the question of whether the benefits of tight control are sufficiently proved to warrant the attendant risks for hypoglycemia. Now, 30 yr later, we believe that maintenance of euglycemia is associated with reduction of microvascular complications in patients with both type 1 and type 2 diabetes and that the risk for hypoglycemia can be treated with better and more sophisticated monitoring and treatment modalities. Siperstein and his colleagues had never questioned the importance of glucose control but wanted clinicians to balance the risks associated with achieving euglycemia with the known adverse effects of the therapeutic interventions available in the mid-1970s. The same argument may apply today. There are substantial data supporting the association between hyperglycemia and CVD, and there are emerging data that cardiovascular outcomes may improve with reduction in HbA 1c. Achieving this outcome with minimum adverse effects is the challenge that we face with currently available oral hypoglycemic agents. Nevertheless, the availability of strong evidence supporting the microvascular risk reduction in itself is a sufficient reason to aim for better glycemic control. As more data about the cardiovascular outcomes become available, we should reassess our position. In this regard, the discussion in the scientific community between the Food and Drug Administration (FDA) and GlaxoSmithKline regarding rosiglitazone and its risk for cardiovascular morbidity is instructive (22,23). The Food and Drug Administration panel pointed out that currently all oral hypoglycemic agents are approved for glycemic control. Furthermore, there is progressive worsening of glycemic control over time in diabetic patients and thus long-term trials will probably need to compare one drug within a multi-drug regimen with other available therapies, making demonstration of the effect of any single drug a formidable task. The GlaxoSmithKline response raised the issue that the higher risk for myocardial ischemic events with rosiglitazone was observed in only placebo-controlled trials; in studies comparing different therapeutic interventions, there was no excess risk with rosiglitazone as compared with metformin and sulfonylureas, and there was no statistically significant difference among the three treatment groups. Does this mean that all three of these hypoglycemic agents may cause an increase in ischemic heart disease? These analyses were done on trials reported in Nissen s meta-analysis (24), and those trials differed widely in their aims, study populations, duration of exposures, etc. Thus, at this time, there are no high-quality data that would substantiate this claim. The UKPDS has demonstrated that there is a natural progression of disease in patients with type 2 diabetes, thus requiring the use of more agents to control hyperglycemia over time. Because we believe that hyperglycemia is associated with higher risks for complications, we argue that the best current approach is to use the most effective and safest available agents with the goal of achieving euglycemia for the longest period of time with the least number of adverse effects. The search for safe and effective oral hypoglycemics should continue, but on the basis of the current evidence, the benefits of using hypoglycemic agents far outweigh the potential risks of drug-associated complications. We agree with Ness, who noted that uncertainty along the

166 Clinical Journal of the American Society of Nephrology Clin J Am Soc Nephrol 3: 163 167, 2008 way is the hallmark of science and the price of progress (25). Concato (26) rightly pointed out that progress in science is iterative and incremental...putting what we know into practice would prevent more disease than worshipping at the altar of randomized trials. In summary, we believe that hyperglycemia is a risk factor in a dose-dependent manner for both micro- and macrovascular complications in both type 1 and type 2 diabetes, and HbA 1c is an excellent measure of glycemic control. The HbA 1c value is the best surrogate marker that we have right now. Clearly, clinical judgment should be used when treating individual patients, such as the example cited in Dr. Hadler s article, but the importance of glycemic control should not be minimized. In the absence of any threshold of blood glucose at which the complications occur, we believe that lowering of blood glucose to as close to the normal range as possible with safest drugs available should be the goal of both patients and their physicians. Disclosures None. References 1. Nathan DM: Long-term complications of diabetes mellitus. N Engl J Med 328: 1676 1685, 1993 2. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL: Glycosylated hemoglobin predicts the incidence and progression of diabetic retinopathy. JAMA 260: 2864 2871, 1988 3. Klein R, Klein BE, Moss SE, Davis MD, DeMets DL: The Wisconsin Epidemiologic Study of Diabetic Retinopathy. IX. Four- year incidence and progression of diabetic retinopathy when age at diagnosis is less than 30 years. Arch Ophthalmol 107: 237 243, 1989 4. Klein R, Klein BE, Moss S, DeMets DL: Proteinuria in diabetes. Arch Intern Med 148: 181 186, 1988 5. Klein R, Klein BE, Moss SE, Cruickshanks KJ: Ten-year incidence of gross proteinuria in people with diabetes. Diabetes 44: 916 923, 1995 6. Klein R, Klein BE, Moss SE, Cruickshanks KJ, Brazy PC: The 10-year incidence of renal insufficiency in people with type 1 diabetes. Diabetes Care 22: 743 751, 1999 7. 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: 977 979, 1993 8. Diabetes Control and Complications Trial/Epidemiology of Diabetes Interventions and Complications Research Group: Retinopathy and nephropathy in patients with type 1 diabetes four years after a trial of intensive therapy. N Engl J Med 342: 381 389, 2000 9. UK Prospective Diabetes Study (UKPDS) Group: Intensive blood-glucose control with sulphonylurea or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 352: 837 853, 1998 10. Ohkubo Y, Kishikawa h, Araki E, Miyata T, Isami S, Motoyoshi M, Kojima Y, Furuyoshi N, Shichiri M: Intensive insulin therapy prevents the progression of diabetic microvascular complications in Japanese patients with non-insulin-dependent diabetes mellitus: A randomized prospective 6-year study. Diabetes Res Clin Pract 28: 103 117, 1995 11. Herman WH, Ma Y, Uwaifo G, Haffner S, Kahn SE, Horton ES, Lachin JM, Montez MG, Brenneman T, Barrett-Connor E, Diabetes Prevention Program Research Group: Differences in A1c by race and ethnicity among patients with impaired glucose tolerance in the Diabetes Prevention Program. Diabetes Care 30: 2453 2457, 2007 12. Sheetz MJ, King GL: Molecular understanding of hyperglycemia s adverse effects for diabetic complications. JAMA 288: 2579 2588, 2002 13. LeCaire T, Palta M, Zhang H, Allen C, Klein R, D Alessio D: Lower-than-expected prevalence and severity of retinopathy in an incident cohort followed during the first 4 14 years of type 1 diabetes. Am J Epidemiol 164: 143 150, 2006 14. Balkau B, Shipley M, Jarrett RJ, Pyorala K, Pyorala M, Forhan A, Eschwege E: High blood glucose concentration is a risk factor for mortality in middleaged nondiabetic men: 20-Year follow-up in the Whitehall Study, the Paris Prospective Study, and the Helsinki Policemen Study. Diabetes Care 21: 360 367, 1998 15. Shankar A, Klein R, Klein BE, Moss SE: Association between glycosylated hemoglobin level and cardiovascular and all-cause mortality in type 1 diabetes. Am J Epidemiol 166: 393 402, 2007 16. The Diabetes Control and Complications Trial, Epidemiology of Diabetes Interventions and Complications Research Group: Intensive diabetes therapy and carotid intima-media thickness in type 1 diabetes mellitus. N Engl J Med 348: 2294 2303, 2003 17. Stratton IM, Adler AI, Neil HA, Matthews DR, Manley SE, Cull CA, Hadden D, Turner RC, Holman RR, on behalf of the UK Prospective Diabetes Study Group: Association of glycemia with macrovascular and microvascular complications of type 2 diabetes (UKPDS 35): Prospective observational study. BMJ 321: 405 412, 2000 18. Uusitupa M, Siitonen O, Pyörälä K, Aro A, Hersio K, Penttilä I, Voutilainen E: The relationship of cardiovascular risk factors to the prevalence of coronary heart disease in newly diagnosed type 2 (non-insulin-dependent) diabetes. Diabetologia 28: 653 659, 1985 19. Reaven GM: Banting lecture 1988: Role of insulin resistance in human disease. Diabetes 37: 1595 1607, 1988 20. Saito I, Folsom AR, Brancati FL, Duncan BB, Chambless LE, McGovern PG: Nontraditional risk factors for coronary heart disease incidence among persons with diabetes: The Atherosclerosis Risk in Communities (ARIC) Study. Ann Intern Med 33: 81 91, 2000 21. Siperstein MD, Foster DW, Knowles HC Jr, Levine R, Madison LL, Roth J: Control of blood glucose and diabetic vascular disease. N Engl J Med 296: 1060 1063, 1977

Clin J Am Soc Nephrol 3: 163 167, 2008 Blood Glucose and Cardiovascular Morbidity and Mortality 167 22. Joffe HV, Parks MH, Meyer RJ, Jenkins JK, Temple R: Rosiglitazone and the FDA [Letter]. N Engl J Med 357: 1775 1776, 2007 23. Krall RL: Rosiglitazone and the FDA [Letter]. N Engl J Med 357: 1776, 2007 24. Nissen SE, Wolski K: Effect of rosiglitazone on the risk of myocardial infarction and death from cardiovascular causes. N Engl J Med 356: 2457 2471, 2007 25. Ness RB: Letters. New York Times Magazine September 30, 2007: 20 26. Concato J: Letters. New York Times Magazine September 30, 2007: 20 See related article, Oral Hypoglycemics and Diabetic Nephropathy, on pages 159 162.