Independent predictors of isolated clinic ('white-coat') hypertension

Background Hypertension guidelines recommend 24 h ambulatory blood pressure (ABP) monitoring in hypertensive subjects with suspected isolated clinic h ypertension (ICH). However, the pre-test probability of ICH based on the di stribution of its independent predictors has not yet been estimated in hype rtensive subjects with mildly elevated blood pressure. Objective To ascertain the independent predictors of ICH in mildly hyperten sive subjects. Methods In the setting of the HARVEST-PIUMA collaboration, we studied 1564 subjects with hypertension stage I. At entry, all subjects were untreated a nd all underwent ABP monitoring and echocardiography. Diabetes, hypertensio n grade > I, renal failure or previous cardiovascular morbid events were ex clusion criteria. Clinic BP was 143/92 mmHg (SD 9/5) and 24 h ABP was 128/8 1 mmHg (SD 10/8). Results Prevalence of ICH (daytime ABP < 130 mmHg systolic and 80 mmHg dias tolic) was 10.4%. In a multivariate logistic regression analysis, sex (P= 0 .002), smoking (P= 0.038) and clinic diastolic BP (P= 0.0002) were the sole independent predictors of ICH according to the following equation: Y= 2.64 38 + 0.5128 x sex (0 = men; 1 = women)+ 0.4543 x current smoking (0 = yes; 1 = no) - 0.0531 x clinic diastolic BP (mmHg) and P (probability of ICH)= e xp( Y)/[1 + (exp(Y)]. Left ventricular (LV) mass at echocardiography was a further independent predictor(P= 0.002) of ICH according to the following e quation: Y= 3.4343 + 0.4603 x sex + 0.5989 x current smoking - 0.0482 x cli nic diastolic BP - 0.0312 x LV mass [g/height (m)(2.7)]. LV mass was greate r(P < 0.01) in the group with ambulatory hypertension [42.3 g/height (m)(2. 7)] than in that with ICH [39.2 g/height (m)(2.7)] and not dissimilar betwe en the ICH group and a control group of 370 healthy normotensive subjects [ 38.1 g/height(m)(2.7)]. Conclusions In untreated subjects with stage I hypertension, ICH is most fr equent among women, nonsmokers and subjects with low clinic BP and smaller LV mass. These findings allow identification of subjects with indication to ABP monitoring because of suspected ICH.