How well do office and exercise blood pressures predict sustained hypertension? A Dundee Step Test Study

Exercise systolic blood pressure (BP) appears to be a better predictor of c ardiac mortality than casual office BP. We tested whether this could be exp lained by exercise systolic BP being a better predictor of sustained hypert ension than casual office BP. Exercise systolic BP was measured using the l ightweight 9-min single stage, submaximal Dundee Step Test in 191 consecuti ve subjects (102 male, age 52 (s.d. 13) years) who were referred to a speci alist hypertension clinic for assessment. Exercise systolic BP was compared with office BP and daytime ambulatory BP (ABP). Sustained hypertension was defined as a mean daytime systolic and/or diastolic ABP of greater than or equal to 140/90 mm Hg. Receiver operating characteristic (ROC) curves of e xercise systolic BP and office BP in predicting sustained hypertension were compared. The positive predictive value of office diastolic BP greater tha n or equal to 90 mmHg and office systolic BP greater than or equal to 140 m m Hg for sustained hypertension were 64% and 67% respectively. However, exe rcise systolic BP greater than or equal to 180 mm Hg had a positive predict ive value of 76%, Twenty-two percent (42/191) of subjects had an exercise s ystolic BP rise to greater than or equal to 270 mm Hg, and 93% of this grou p had sustained hypertension on ABP. Whilst exercise systolic BP was a bett er predictor of sustained hypertension using currently recommended office B P treatment thresholds, the ROC curves of these indices were not different. In a multiple regression analysis, exercise systolic BP was an independent predictor of sustained hypertension, accounting for 36% of the variance of daytime systolic ABP after adjusting for age, gender and antihypertensive drug treatment. In conclusion, exercise systolic BP was a marginally better predictor of sustained hypertension than office BP. This may partly explai n why exercise systolic BP is a potent predictor of cardiac mortality.