Circulation Journal. Vol. 80(2016) No. 11 2293-2294

 

Trends in the Impact of Diabetes on the Development of Cardiovascular Diseases

Kyoung Hwa Ha, Dae Jung Kim




Cardiovascular disease (CVD) is the leading cause of death and disability worldwide and was responsible
for nearly 17 million deaths and 329 million disabilityadjusted life-years lost in 2013. Despite a remarkable decrease in CVD mortality rates in high-income countries, CVD continues to affect people living in low- and middle-income countries. 1 Diabetes mellitus (DM), which is a notable risk factor for CVD, is one of the most common metabolic disorders worldwide. The prevalence of DM has increased significantly over the past several decades and is expected to continue to increase from 415 million in 2015 to 642 million by 2040.2

DM and CVD share many conventional risk factors, such as central obesity, dyslipidemia, and hypertension, and DM itself is a risk factor for CVD. Many studies have suggested that these common risk factors increase the risk of CVD not only in non-diabetic subjects but also in patients with DM. However, common risk factors in patients with DM cannot completely account for the increased risk of CVD. One study reported that type 2 DM is associated with a risk of incident peripheral arterial disease, ischemic stroke, stable angina, heart failure, and non-fatal myocardial infarction in a cohort of 1.92 million subjects in England.3 In Japan, DM has been also
identified as a risk factor for both ischemic stroke and coronary artery disease.4,5 Laakso and Kuusisto6 indicated that a “diabetic milieu” (insulin resistance and hyperglycemia) may explain the risk of CVD in patients with DM. Insulin resistance and hyperglycemia trigger the risk of vascular complications via impaired endothelial function, low-grade inflammation, advanced glycation endproduct levels, thrombosis, fibrinolysis, and modifications of lipoprotein particles.6 Estimating the population risk for CVD caused by DM is clinically important to establish public health strategies for managing patients with CVD. The population attributable fraction (PAF) is the proportional reduction in disease or death of a population that would occur if certain risk factors were unexposed. The range of PAFs caused by DM is 2–12% for coronary artery disease, 1–6% for hemorrhagic stroke, and 2–11% for ischemic stroke according to the Asia Pacific Cohort Studies Collaboration data.7 The Framingham Heart Study indicated that the age- and sex-adjusted PAF of CVD caused by DM
was 5.4% in 1952–1974 and 8.7% in 1975–1998. The attribut-able risk ratio for CVD associated with DM has increased, while the attributable risk for CVD associated with other risk factors, such as hypertension and obesity, decreased between 1952–1974 and 1975–1998.8 However, the Atherosclerosis Risk in Communities Study reported that the risk of CVD attributable to DM remained similar during 1987–1989 and 1996–1998; PAF was 15% in 1987–1989 and 17% in 1996–1998.9 The PAF of CVD caused by DM in Finland increased among men from 11.4% to 13.8% but decreased among women from 20.1% to 16.9% between 1992 and 2002, using data from 4 Finnish national registers.10 In this issue of the Journal, Hayama-Terada et al11 investigate the association between DM with a risk of CVD and the proportion of CVD risk attributable to DM in a large population-based sample of middle-aged Japanese. Although the incidence of CVD was stable at 2–3% in a subgroup of subjects with normal glucose tolerance or prediabetes, a dramatic increase in CVD incidence from 5% to 7% was shown in the DM group of the 3 cohorts. The multivariate hazard ratios of CVD for subjects with DM were 1.40 (95% confidence interval [CI]: 0.91–2.14) in 1992–1995, 1.93 (95% CI: 1.25–3.00) in 1996–1999, and 2.59 (95% CI: 1.77–3.81) in 2000–2003 compared with non-diabetic subjects after adjusting for traditional CVD risk factors such as obesity, triglycerides, systolic blood pressure, and antihypertensive medication use. The PAF
related to DM of CVD also increased even though glucose levels decreased in the DM subgroup over time; the PAF was 2.8% in 1992–1995, 5.6% in 1996–1999, and 12.4% in 2000–2003. It is well known that improved glycemic control reduces microvascular complications. However, the effects of improved glycemic control on macrovascular complications, such as CVD, are controversial, whereas controlling blood pressure and lipids definitely decreases the risk of CVD.12 Therefore, although glucose control has been improved with comprehensive care in patients with DM, it may be less effective in preventing CVD. Intensive lowering of blood glucose may cause severe hypoglycemia, which increases cardiovascular risk.13,14 In addition, the increased prevalence of DM itself may affect the increased PAF related to DM and CVD. This is the first study in Asia to examine temporal trends in the excess risk of CVD associated with DM over a longer follow-up. The strengths of this study are the precise determination of the incidence of CVD using death certificates, national health insurance claims, reports from local physicians, public health nurses and health volunteers, annual cardiovascular risk surveys, and a household questionnaire. However, some limitations should be mentioned. First, significant differences were found in major baseline risk factors for CVD, such as age, systolic blood pressure, total cholesterol, high-densitylipoprotein cholesterol, triglycerides, and use of hypertension and dyslipidemia medications for each time period. In particular, the mean age of subjects with diabetes was older in 2000–
2003 than that in 1992–1995. Older subjects may have a longer duration of DM. Previous studies have demonstrated that age at onset and duration of DM are associated with CVD.15,16 Although the age effect can be adjusted in the model, it does not consider DM duration. Therefore, the CVD risk in 2002–2003 may have been overestimated. Furthermore, use of antidiabetic medication was higher in 2002–2003 than during other
periods, which would lead to detection of asymptomatic CVD in patients with DM. Thus, a detection bias might also exist. Second, this study did not consider the definition of DM including glycated hemoglobin (HbA1c). The American Diabetes Association recommends measuring HbA1c to diagnose DM.17 Combined use of fasting glucose and HbA1c has a significantly higher sensitivity for diagnosing DM than either test alone.18 However, the possibility of misclassification of DM would be the same and lead to an underestimation of risk. In conclusion, morbidity and mortality from CVD have decreased markedly worldwide. However, the rising prevalence of DM may affect this reduction. The risk of CVD associated with DM and the proportion of CVD attributable to DM increased during 1992–2003 in Japan. These findings highlight that efforts to prevent DM and integrated management of DM are needed to prevent or delay CVD.