Prediction of blood pressure level and need for antihypertensive medication:: 10 years of follow-up

Objective To evaluate the usefulness of blood pressure and its variability in the prediction of future blood pressure and need for antihypertensive medication. Methods We used WHO criteria to classify, at baseline, 97 healthy untreated male volunteers as normotensive (n = 34), borderline hypertensive (n = 29) or mild hypertensive (n = 34), with casual measurements before intra-arterial 24 h ambulatory blood pressure monitoring, After 10 years of follow-up, 87 of the men (90%) were available and their blood pressure was recorded using casual measurements and non-invasive ambulatory 24 h monitoring. Results During the follow-up, the blood pressure classification deteriorated in 35 individuals (40%) and improved in six (7%) (McNemar test, P <0.0001). In the borderline hypertensive group, 77% became hypertensive (P=0.03). The 24 h mean systolic blood pressure was the best predictor of follow-up casual systolic (adj.R-2 = 0.420) and 24 h systolic (adj.R-2 = 0.540) blood pressure. The 24 h mean diastolic blood pressure was the best predictor of follow-up casual diastolic (adj.R-2 = 0.301) and 24 h diastolic (adj.R-2 = 0.292) blood pressure, The baseline casual systolic blood pressure also predicted the follow-up casual systolic blood pressure relatively well (adj.R-2 = 0.356), but was clearly weaker for the follow-up 24 h systolic (adj.R-2 = 0.275) blood pressure. The prediction of follow-up casual diastolic (adj.R-2 = 0.259) and follow-up 24 h diastolic (adj.R-2 = 0.214) blood pressure by baseline casual blood pressure was even weaker. The means and variabilities of the 24 h, daytime, and night-time blood pressures were the best predictors of the need for antihypertensive medication (Cox-Snell R-2 = 0.399). The characteristics of the individual did not significantly predict future blood pressure and the need for antihypertensive medication. Conclusions The 24 h mean blood pressure was an excellent predictor of the future blood pressure and the need for antihypertensive medication. Prediction of antihypertensive medication was further improved by also using blood pressure variability. Systolic blood pressure was more predictable than diastolic blood (C) 2001 Lippincott Williams & Wilkins.