Abstract 2935: Blood Pressure Variability At Night Predicts Cardiovascular Events In Type 2 Diabetics

Background It is not known whether BP variability is a risk factor for future incidence of cardiovascular disease (CVD) in type 2 diabetics (DM). Objective To test the hypotheses that short-term BP variability, including the morning BP surge, evaluated by ambulatory BP, predicts risk of incident CVD in DM. Methods Ambulatory BP monitoring was performed in uncomplicated 300 diabetics without BP medications, and they were followed for 54 ± 20 months. Morning BP was defined as the mean BP during the two hours after awaking, and morning BP surge as morning BP minus mean BP during the 1 hour that included the lowest BP at night. Incident CVD was defined as either myocardial infarction, stroke, or sudden cardiac death, and its relationship to different measures of BP variability, including morning BP surge and the standard deviation (SD) of daytime and nighttime BP was investigated. Cox proportional hazard models were used to estimate hazard ratios (HR) and their 95% CI, before and after controlling for age, sex, BMI, current smoking, antihypertensive medication, total cholesterol, serum creatinine, and 24-h SBP. Results The mean age was 67.8±9.6 years, 48% were male, 253 (84%) had a diagnosis of hypertension, and the mean values of the SD of awake SBP/DBP were 18±6/11±4 mmHg, and those of sleep SBP/DBP were 13±5/9±3 mmHg. During follow-up, there were 29 cardiovascular events. In univariable analyses, morning SBP (p=0.001), SD of awake DBP (p=0.002), and SDs of sleep SBP/DBP (both ps<0.01) were positively associated with incident CVD, but the morning surge of SBP/DBP was not. In multivariable analyses, the SDs of sleep SBP (HR=1.08; 95%CI, 1.01–1.16, p<0.05) and sleep DBP (HR=1.16; 1.07–1.26, p<0.001) were independently associated with incident CVD. Morning BP and morning BP surge were not associated with incident CVD. High nighttime variability of SBP and DBP (15.5 and 10.0 mmHg, each 75 percentile of SD) were associated with 1.6 and 3.0 times the risk of incident CVD, compared to individuals at the 25 percentile. Conclusions The hypothesis that morning BP surge is a predictor of CVD was not supported in our diabetic patients. However, BP variability at night appears to predict incident CVD, independent of ambulatory BP level and other traditional risk factors, in type 2 diabetics.