Diabetes Mellitus and Cardiovascular Risk : epidemiology, etiology and intervention

Abstract: Background: The Framingham Study from 1988 showed a heavy impact of diabetes mellitus (DM) on the risk and prognosis of cardiovascular disease (CVD). Several other studies have confirmed that DM is an independent risk factor for coronary heart disease (CHD) and that patients with DM have a poor prognosis. However, the strength of DM as a risk factor is debated. Some studies indicate that DM, as a risk factor for a coronary event, is comparable to already known or established CHD. Also, mechanisms of how diabetes increases the CVD risk are under intensive research. A novel risk factor such as altered fibrinolysis is one of the potential mechanisms explaining the heavy cardiovascular burden in diabetes. Hypofibrinolytic changes can be seen in individuals with metabolic syndrome, insulin resistance, and obesity as well as in patients with manifest diabetes or manifest CHD.Methods: This dissertation is divided in three parts; epidemiological, etiological and interventional. Papers I and II are epidemiological, population based retrospective studies which compare time trends in myocardial infarction and stroke morbidity and mortality between patients with or without diabetes. Papers III and IV have an etiological approach to the fibrinolytic system and CVD risk both in patients with or without diabetes. Paper V is the interventional part of this thesis studying if intensive insulin treatment can improve fibrinolysis in patients with high CVD risk.Results: The incidence and mortality from myocardial infarction and stroke have declined in the counties of Västerbotten and Norrbotten in Northern Sweden. Unfortunately, the subgroup with patients with diabetes and myocardial infarction (MI) in Paper I did not benefit from these favorable trends over the study period. For stroke, this is for the first time declining incidence has been reported in this population. Incidence of first-ever stroke decreased for non-diabetic men with a yearly change of -0.8 percent (p-value  <0.001), and for diabetic women with a yearly change of -1.5 percent (p-value=0.012). Recurrent stroke incidence declined highly significant, p-value  <0.001, for non-diabetic men and women and for diabetic women with a yearly change of -1.5, -2.7, and -5.4 percent, respectively. For diabetic men a non-significant decline of -1.0 percent (p-value=0.28) in stroke incidence was seen. Paper II showed that women with diabetes had decreased incidence for stroke but not men with diabetes. Also, reduced mortality from stroke was found in all patients except for diabetic women with first ever stroke.Patients with diabetes have altered fibrinolysis compared to individuals with normal glucose metabolism. Also, patients with prior MI manifest a hypofibrinolytic stage with low tissue-type plasminogen activator (tPA) activity and high levels of plasminogen activator inhibitor-1 (PAI-1) and fibrinogen. Paper III showed a significant association between decreased tPA activity and increased fibrinogen levels in patients with a first MI when examined 3 months after the event. The results persisted even after adjustment for traditional risk factors and variables mirroring the insulin resistance syndrome and were significant in the whole study group of subjects with MI as well as for men alone.Paper IV, a 10 year follow-up study, showed that in patients with diabetes, tPA-activity significantly predicted the presence of sign(s) of lower extremity arterial disease (LEAD) at the baseline and at the 10-year follow-up. In addition, tPA-mass at the 10-year follow-up was associated with signs of LEAD. Baseline age, hypertension, and HbA1c were independently associated with sign(s) of LEAD at 10 years. This long-term study supports previous findings of a significant association between asymptomatic LEAD and tPA-activity. Thus, tPA-activity may be an early marker of LEAD, although the mechanism of this relationship remains unclear.Several studies report conflicting results regarding benefits and disadvantages of intensive insulin treatment. ORIGIN trial was an international multicenter trial studying effects of intensive insulin treatment on CVD. This thesis reports a Swedish sub-study of ORIGIN trial examining effects of insulin treatment on fibrinolysis. The allocation to insulin treatment did not significantly affect the studied fibrinolytic markers or von Willebrand factor (VWF) compared to the standard treatment. Log mean delta values between baseline and end of study increased significantly for tissue plasminogen activator activity (tPAact) and tPA/PAI-1 complex. For plasminogen activator inhibitor-1 (PAI-1), tPA antigen (tPAag) and VWF no significant differences were found. Within-group analysis during the whole study period revealed significant changes for tPA/PAI-1 complex, tPA antigen, and VWF in the insulin treatment group but no significant changes in the standard treatment group.The hypothesis that allocation to insulin treatment would improve the levels of markers of fibrinolysis or VWF compared to standard glucose lowering treatment could not be verified. Conclusion: This dissertation shows that patients with diabetes still have a heavy cardiovascular disease burden with increased risk for MI and stroke compared to individuals with normal glucose metabolism. This cardiovascular burden also includes increased morbidity, mortality, and poorer long-term prognosis. The fibrinolytic system has an impact on cardiovascular disease and this thesis has shown that patients with diabetes have unfavorable changes in their fibrinolysis and that alterations in fibrinolysis can predict future peripheral artery disease. However, intensive insulin therapy did not appear to affect this system to the extent of resulting in reduced cardiovascular morbidity.