LAUSR

BACKGROUND Diabetes was regarded as an independent risk factor for abnormal left heart remodeling. However, there was lacking population-based data on the relationship of glucose status with left ventricular hypertrophy (LVH) or left atrial enlargement (LAE). This study intended to clarify the influence of diabetes and prediabetes on the prevalence and incidence of LVH and LAE based on a northeast rural population of China. METHODS We analyzed clinical, laboratory and echocardiographic data of a total of 2824 participants aged over 35 years from a population-based prospective cohort NCRCHS study with 2 years of follow-up, which was carried out in rural areas of northeast China. All measurements were performed according to standardized protocols. RESULTS There were 2179 controls, 342 subjects with prediabetes and 303 ones with diabetes. The baseline distribution of LAD, IVSd, LVIDd, LVIDs, LVMI, E wave, A wave, E/A, E/e', diastolic dysfunction, LVEDV, LVESV and SV was significantly different among three groups (all Ptrend<0.05). After the adjustment for age, gender, BMI, waist circumference, heart rate, hypertension and dyslipidemia, glucose status remained associated with LVIDd and E/e' (all P < 0.05). At baseline, diabetes was independently related to the prevalence of LVH (OR = 1.53; 95%CI = 1.12-2.10; P < 0.01) and LAE (OR = 1.71; 95%CI = 1.19-2.43; P < 0.01) in the overall population, and the same significant results were also found in gender specific subgroups. During the 2-year follow-up, Cox regression models revealed that baseline diabetes had an independent association with the incidence of LAE in the total subjects (HR = 1.83; 95%CI = 1.10-3.06; P = 0.02) and females (HR = 1.90; 95%CI = 1.05-3.46; P = 0.04) after adjusting the potential confounders. CONCLUSION Diabetes, but not prediabetes, is an independent predictor for the prevalence of LVH and LAE, and for the new-onset LAE, it should be considered in the assessment of diabetes and cardiac structural remodeling.