Background: It is unclear whether the prognostic value of blood pressure (BP) changes from day to night is modified by concomitant risk factors and organ damage.
Methods: We studied 3247 clinically hypertensive patients who underwent off-therapy 24-hour ambulatory BP monitoring during their initial diagnostic work-up. Over a mean 9.9-year follow-up, 292 patients developed a first primary outcome event (composite of nonfatal myocardial infarction, nonfatal stroke, cardiovascular death, hospitalization for heart failure) and 281 died. The day-night BP changes were analyzed as night-to-day systolic BP ratio and included in semiparametric Cox models.
Results: After allowance for age, sex, diabetes, cigarette smoking, low-density lipoprotein cholesterol, serum uric acid and average 24-hour systolic BP, the night-to-day systolic BP ratio maintained an independent association with outcome (hazard ratio, 1.17 [1.01-1.35]; P=0.041). When 2 markers of organ damage, chronic kidney disease and left ventricular hypertrophy, were added to the model, the night-to-day systolic BP ratio failed to maintain an independent association with outcome (hazard ratio, 1.19 [0.97-1.31]). The Akaike Information Criterion and the Schwarz-Bayes Information Criterion showed that, after allowance for other covariables, the night-to-day systolic BP ratio: (1) provided less information when compared with chronic kidney disease and left ventricular hypertrophy and (2) did not improve information provided by chronic kidney disease and left ventricular hypertrophy. Results were comparable on all-cause death as well as using categories of the night-day systolic BP ratio (extreme dippers, dippers, nondippers, reverse dippers).
Conclusions: Chronic kidney disease and left ventricular hypertrophy are potent modifiers of the prognostic value of the circadian BP changes.
Keywords: blood pressure; cardiovascular diseases; hypertrophy; kidney diseases; risk factors.