White coat effect, blood pressure and mortality in men: prospective cohort study.

Abstract

Because long-term follow-up studies, which also included normotensive controls, have been lacking, the clinical significance of 'white coat' effect and of 'white coat' hypertension has remained controversial. Twenty-one-year prospective data was gathered in 536 men with cardiovascular risk factors at baseline. Blood pressure was measured both by a nurse and by a physician and 'white coat effect' was defined as the difference between the two measurements (physician minus nurse). In addition, four blood pressure groups were categorized: normotensive (n=259), white coat hypertensive (n=18), mildly hypertensive (n=150) and persistently hypertensive (n=109). Comparison of these groups at baseline showed that men with white coat hypertension had higher levels of metabolic risk factors. Sixty-eight men died during follow-up. The men with a white coat effect >30 mmHg (n=37) had significantly higher mortality than other men (relative risk 2.2, 95% confidence interval 1.1-4.2). Mortality was significantly higher in the white coat hypertensive group (33.3%) than in the normotensive group (9.5%, P=0.0005 between groups). Relative risk adjusted for baseline risk factors in the white coat hypertensive group was 3.3 (1.2-7.6) compared with the normotensive group. The development of drug-treated hypertension was also more common (27.8% vs 13.4% in the normotensive group, P<0.0001 between groups). The results suggest that white coat hypertension or a large white coat effect is not an innocent phenomenon. It tends to co-exist with metabolic risk factors and predicts total and cardiovascular mortality during long-term follow-up.