Objectives

This study determines the predictive value of brachial pulse pressure for cardiovascular stroke and coronary mortality in a large population categorized by blood pressure level.

Methods

The population was composed of 69 989 subjects aged more than 50 years and divided into ‘optimal’,‘normal’and ‘normal high’normotensive subjects, and grades 1, 2, 3 hypertensive subjects as defined in the 2003 European Guidelines for Management of Hypertension. Hazard ratios for mortality were evaluated using Cox regression models before and after adjustment for pulse pressure. To further assess the role of pulse pressure in mortality, subjects were also classified as ‘normotensive’and ‘hypertensive’whether they were with or without elevated pulse pressure (≥ 60 mmHg).

Results

By comparison with ‘optimal’values, before adjustment for pulse pressure, the hazard ratio for cardiovascular, coronary and stroke mortality increased markedly for blood pressure groups as defined by the European blood pressure classification. For cardiovascular and coronary mortality, the hazard ratio was significant both in the ‘high normal’and hypertensive ranges. For stroke mortality, hazard ratio was significant only in the hypertensive ranges. After adjustment to pulse pressure, hazard ratio remained unmodified for stroke mortality. For coronary mortality, the risk was attenuated in ‘high normal’and all hypertensive groups. A study of the pulse pressure classification in subjects dichotomized as normotensive and hypertensive subjects indicated that pulse pressure was an independent risk factor for cardiovascular mortality.

Conclusion

Increased pulse pressure predicts cardiovascular mortality, acting more on coronary than cerebral vessels. This finding involves all blood pressure ranges, including subjects with low diastolic but normal systolic blood pressure.