The Importance of Blood Pressure Variability

Hypertension is a leading modifiable risk factor for cardiovascular (CV) morbidity and mortality. Blood pressure (BP) levels vary during the day and overtime. Long-term visit-to-visit BP variability (BPV) represents fluctuations of BP over time around a true basal BP levels. de Havenon et al. analyzed data from the Atherosclerosis Risk in the Community (ARIC) study and demonstrated an association between long-term systolic BPV and mortality.1 However, they did not find an association between long-term BPV and CV events. The evaluation of BPV over time is reproducible particularly when subjects are not medically treated.2,3 Lim et al. suggested that 6 BP measurements are required to assess BPV over time.3 de Havenon et al. assessed BPV from 4 measurements and showed association only with mortality.1 We recently showed that systolic BPV assessed from 3 BP measurements was associated with long-term mortality and CV morbidity.1,4 These findings indicate that assessment of BPV can be based on less than 6 BP measurements. Several previous studies have shown that long-term BPV predicts CV events beyond the mean BP levels.4–6 However, the results regarding the association between long-term BPV and CV events are inconsistent. Mancia et al. showed in mildly to moderately hypertensive patients, that CV outcomes were related to the mean achieved clinic or ambulatory systolic BP but not to on-treatment visit-to-visit BPV.7 Similarly, Gao et al. showed in elderly subjects that BPV was not an independent predictor of mortality risk.8 The findings of de Havenon et al. support the observation that systolic BPV is associated with long-term mortality.1 Most studies assessed the association between systolic BPV and CV outcome. The importance of diastolic BPV as a risk factor for mortality and CV events was less evaluated. In a recent study, Dasa et al. showed in older patients with hypertension and coronary artery disease that only systolic and not diastolic BPV was associated with excess long-term mortality.9 Kaze et al. showed in diabetic patients that systolic BPV was associated with CV events and mortality whereas diastolic BPV was associated only with mortality.10 Unlike these studies Li et al.11 showed, in patients on hemodialysis, that diastolic BPV was associated with all-cause and CV mortality and de Havenon et al.12 showed in poststroke patients that diastolic BPV was as important as systolic BPV in prediction recurrent stroke. Thus, it seems that the importance of diastolic BPV as a risk factor is less clear and further studies are needed to evaluate the association between diastolic BPV and outcomes. Data on BPV have been derived primarily from selected populations, mostly secondary analyses of studies in hypertensive patients who were on antihypertensive treatment.13 Antihypertensive treatment and treatment adherence may have effect on BPV.14 Therefore, studies in treated hypertensive patients may be biased and may not represent the true BPV. de Havenon et al. evaluated data from the ARIC study and only 46.2% had at baseline hypertension, but for the hypertensive patients they did not have information on antihypertensive treatment and therefore could not exclude the possibility that long-term BPV was related to antihypertensive treatment.1 BPV may be related to variability in weight and it has been shown that weight variability is associated with mortality.15 We recently showed that among normotensive adults, who had routine annual examination modest weight changes of more than 2.5% were associated with a significant change in systolic BP.16 Unfortunately, de Havenon et al. did not relate the BPV to the weight variability.1 Apparently, it seems that fluctuations in hemodynamic and metabolic parameters are associated with poor outcome. It is not clear whether BPV increase CV morbidity and mortality or it is a marker of CV risk. This may have an impact on the approach to BPV. If it is a cause then treatment with calcium antagonists that reduces not only BP but also BPV should be superior to other treatments.17 Unfortunately, treatment with calcium antagonists is not superior to other treatments suggesting that BPV is a marker of CV risk and should be taken into consideration when assessing CV risk in hypertensive patients. A possible explanation for the association of BPV with increased CV risk is that the need to accommodate to BP fluctuations requires energy recruitment through activation of the sympathetic nervous system and the renin–angiotensin system and activation of these systems may increase the risk of CV events and mortality. The author declared no conflict of interest.