Nocturnal blood pressure fall on ambulatory monitoring in a large international database. The "Ad Hoc' Working Group.

A wide range of definitions is used to distinguish subjects in whom blood pressure (BP) falls at night (dippers) from their counterparts (nondippers). In an attempt to standardize the definition of nondipping, we determined the nocturnal BP fall and night-day BP ratio by 24-hour ambulatory monitoring in 4765 normotensive and 2555 hypertensive subjects from 10 to 99 years old. In all subjects combined, the systolic/diastolic nocturnal fall and corresponding ratio averaged (+/- SD) -16.7 +/- 11.0/ -13.6 +/- 8.1 mm Hg and 87.2 +/- 8.0%/83.1 +/- 9.6%, respectively. In normotensive subjects, the 95th percentiles were -0.3/-1.1 mm Hg for the nocturnal fall and 99.7%/98.3% for the night-day ratio. Both the fall and ratio showed a curvilinear correlation with age. The smallest fall and largest ratio were observed in older (> or = 70 years) subjects. A higher BP on conventional sphygmomanometry was associated with a larger systolic (partial r = .11) and diastolic (r = .12) nocturnal BP fall. The diastolic (r = .08) but not the systolic night-day ratio increased with higher conventional BP. The nocturnal BP fall was larger and the corresponding night-day ratio smaller in oscillometric (n = 5884) than in auscultatory (n = 1436) recordings, in males (n = 3730) than in females (n = 3590), and in Europe (n = 4556) than in the other continents (n = 2764). The distributions of the nocturnal BP fall and night-day ratio showed considerable overlap among normotensive and hypertensive subjects, but the overlap tended to be larger for the ratio than for the fall. Of all subjects, 3.2% had systolic and diastolic ratios of 100% or more. With adjustments applied for confounders, the probability of being a nondipper increased 2.8 times (95% confidence interval, 2.0-4.0) from 30 to 60 years and 5.7 times (4.4-7.4) from 60 to 80 years. The odds ratios were 1.0 (0.8-1.4) for males versus females. 1.6 (1.2-2.1) for subjects with definite hypertension versus normotensive subjects, 2.4 (1.2-4.7) for Asians (n = 2213, 96% Japanese) versus inhabitants of the other continents, and 2.4 (1.5-3.8) for subjects examined with auscultatory versus oscillometric devices. In conclusion, the mathematical definition of nondipping, ie, having a night-day ratio of 100% or more for systolic and diastolic BPs, closely approximated the 95th percentiles of the night-day ratio in normotensive subjects. The ratio depends less on BP level than the nocturnal BP fall and is therefore to be preferred in the definition of dipping status. Notwithstanding the present findings, the reproducibility of nondipping and its prognostic significance need further clarification.