Intra-arterial blood pressure during exercise and left ventricular indices in normotension and borderline and mild hypertension.

Our objective was to study the usefulness of BP responses to exercise compared to postural BP values for prediction of left ventricular (LV) indices in 28 normotensive (NT), 14 borderline hypertensive (BHT), and 24 mildly hypertensive (HT) men (aged 35-45 years) using intra-arterial BP, the "gold standard" measurement during a series of postures (10 min supine, 10 min sitting, 9 min standing), and exercise (isometric; hand grip, and dynamic; cycle ergometer). LV indices were studied by echocardiography. The highest postural BP correlation was obtained between LV mass index (LVMI) and standing systolic BP (SBP, r = .39, p < .01). Achieved exercise BP did not improve the result compared to standing BP (isometric SBP r = .41, p < .01, dynamic SBP r = .39, p < .01). When exercise responses were expressed as BP change (exercise BP-supine BP) they showed lower correlations with LVMI than achieved exercise BP readings. Using linear multiple regression including all the BP variables, the achieved SBP with isometric exercise explained 17% of LVMI variance (F = 12.9, p = .0006) without any additive value of other variable. In a forced linear regression using BP with various postures as the first variable, the achieved BP with dynamic exercise did not add to the explanation of LVMI variance. However the SBP during isometric exercise added 6% (F = 4.9, p < .05) for the explanation of LVMI variance with sitting SBP as baseline, and also 7% (F = 5.0, p < .05) with supine SBP and diastolic BP as baselines, but no additive value was seen with standing BP. We conclude, that BP levels during various postures were well related to LVMI. Achieved BP or BP change with dynamic exercise did not improve the prediction of the LVMI, and even the best single predictor, BP with isometric exercise, added only little to the power of prediction obtainable from postural values. With mild elevation of BP the clinical utility of dynamic exercise in evaluation of left ventricle is questionable whereas the utility of isometric exercise is marginal. Relatively low correlation values in this study suggest that BP is only one of many factors affecting the cardiac anatomy in the early phases of hypertension.