Blood pressure levels and variability, smoking, and left ventricular structure in normotension and in borderline and mild hypertension

The aims of this study were to determine the importance of ambulatory blood pressure measurement, diurnal blood pressure (BP) profile, and variability in the evaluation of left ventricular (LV) parameters, and the impact of smoking on these factors. We performed intraarterial ambulatory BP (IAMB) recording and echocardiography in 80 healthy, unmedicated men aged 35 to 45 years. Based on repeated casual (GAS) readings before the study, the subjects were classified as normotensive (NT, n = 32), borderline hypertensive (BHT, n = 21), or mildly hypertensive (HT, n = 27) according to WHO criteria. There were 19 (8 NT/5 BHT/6 HT) smokers and 48 (18 NT/13 BHT/17 HT) nonsmokers. Both BHT and HT had significantly greater LV mass index (LVMI) than NT, but LVMI did not differ between nonsmokers and smokers. For the whole group, 24-h BP correlated somewhat better with LVMI than CAS BP (24-h IAMB SEP r = 0.44, P < .001, DBP r = 0.36, P < .001, and CAS SEP r = 0.35, P < .01, DBP r = 0.37, P < .001). Casual SEP alone explained 12% of LVMI variance (F = 10.7, P < .01), whereas 24-h IAMB SEP alone explained 19% of LVMI variance (F = 18.4, P < .001). When comparing day and night SEP and DBP levels, night SEP showed the closest correlation with LVMI (r = 0.43, P < .001), and this alone explained 18% of LVMI variance (F = 18.1, P < .001). Smokers had higher correlations between night BP and LVMI (SBP and DBP r = 0.56, P < .05) than nonsmokers (SEP r = 0.37, P < .01 and DBP r = 0.30, P < .05). In a multiple linear regression including all BP variables, for smokers, night DBP (although only marginally better than night SEP) was the best predictor, explaining 32% of LVMI variance (F = 10.6, P < .01) and additionally night DBP standard deviation (SD) added 18% to the prediction of LVMI (F = 5.8, P < .05). For nonsmokers, day SEP had closest correlation with LVMI (r = .43, P < .01), but explained only 19% of LVMI variance (F = 10.5, P < .01), and other measures did not increase the explanation. We conclude that ambulatory BP was slightly better than CAS BP in predicting LVMI, but BP level, also when measured with the best method available, explained only a moderate fraction of LVMI variance in mild hypertension. However, among smokers, BP, especially nighttime and BP variability, explained LV changes better than among nonsmokers. Thus smoking may have an impact on the interaction of ambulatory BP and LVMI, and in future studies more attention should be paid to this toxic factor.