AIMS Although left ventricular hypertrophy (LVH) defined by either standard 12-lead ECG or echocardiography strongly predicts cardiovascular mortality, its prevalence in Type 2 diabetes is largely unknown. We have assessed prevalence of ECG-LVH and its relationship with clinical and metabolic variables in an Italian population-based cohort of subjects with Type 2 diabetes. METHODS The study-base was 965 (61.3%) subjects with Type 2 diabetes of the population-based cohort living in Casale Monferrato (Italy). LVH was defined by ECG Cornell voltage-duration product. All measurements were centralized. RESULTS ECG-LVH was diagnosed in 165/965 subjects, giving a prevalence of 17.1% (95% CI 14.7-19.5). Large sex differences were found, with higher prevalence in women (23.5%, 19.9-27.0) than in men (8.4%, 5.6-11.0), even after adjustment for age, BMI and hypertension (OR 3.83, 95% CI 2.5-5.9). At the examination, subjects with ECG-LVH were older than those without it. Similar age- and sex-adjusted values of HbA(1c), plasma lipids, fibrinogen, uric acid and creatinine were found in the two subgroups. No differences in prevalence of hypertension, CHD, increased QT duration or dispersion, micro- and macro-albuminuria were found between subjects with ECG-LVH and those without it. In logistic regression analysis, variables independently associated with ECG-LVH, after age-adjustment, were sex and diastolic blood pressure.
Conclusions: This population-based study shows: (i) a high prevalence of ECG-LVH in Type 2 diabetic subjects; (ii) 3-fold higher risk in women than in men, independently of age, BMI, and blood pressure; (iii) an independent association between ECG-LVH and diastolic blood-pressure. Screening for ECG-LVH in diabetic subjects is therefore recommended, particularly in diabetic women.
Copyright 2004 Diabetes UK