Blood pressure control in the hypertensive population. Is the trend favourable?

In the last 20 years a large number of studies has been devoted to the rate of blood pressure (BP) control in hypertensive populations from different countries. The results have shown that, although between-country differences exist [1], the number of hypertensive patients in whom BP is reduced to the levels recommended by guidelines because of their association with substantial cardiovascular and renal protection [2,3] is disappointingly low [4,5]. This has devastating consequences for public health because, in absence or during antihypertensive treatment, persistency of high BP values is accompanied by much higher cardiovascular and renal morbidity and mortality compared with well treated hypertensive individuals [6]. This maintains hypertension as the major cause of death worldwide [7]. In this context one of the most important pieces of information to obtain is whether the percentage of patients at the recommended BP targets has increased in the recent past and it is thus moving in a direction that can improve cardiovascular prevention in the population. This is, however, not an easy goal to achieve because the studies serving for comparison are often based on hypertensive cohorts with different demographic and clinical characteristics, which means that their response to antihypertensive treatment can be different [8,9]. Furthermore, the procedures through which BP is obtained can often differ, the in-study values being provided by a doctor or a nurse, a sphygmomanometer or a semi-automatic device, a single measurement or multiple measurements. This may affect the alerting reaction to use BP measuring procedures so variably [10,11] as to make between study comparisons meaningless except when differences in the rate of BP control are substantial. Despite the above-mentioned limitations, the available data (those from more adequately standardized studies, in particular [12]) suggest that following early major improvements [12,13] the rate of BP control exhibited by hypertensive populations has more recently shown only limited further increases, with a trend that in the last 10–15 years can more properly be visualized as a plateau [13]. This receives support from the results of the study by Carrington et al.[14] published in this issue of the Journal of Hypertension. The study analysed retrospectively a large number of electronic medical records from more than a half a million Australian individuals under general practitioner care, with results of considerable interest and practical implications. First, the number of adult (>18 years) individuals with an elevated BP (taken via automatic devices according to national standards) was quite high, that is no less than one-quarter of the overall study population. Second, a noticeable percentage of individuals with an elevated BP were not under antihypertensive treatment. Third, there were also many hypertensive individuals (about 50%) under antihypertensive treatment whose BP was above target values. Fourth, over the 6-year period covered by the data analysis (2005–2010) there was no trend towards a reduction in the number of patients with an elevated BP, this being the case both in absence and in presence of antihypertensive drug treatment. Assuming that the results are not just peculiar for Australia, the unequivocal conclusion is that, although many effective antihypertensive drugs and drug combinations are available, in affluent countries the low rate of BP control that characterizes the hypertensive population is not moving towards a satisfactory result. Thus, the perception that, after many years of therapeutic progress, hypertension is less of a problem cannot be more wrong. The data provided by Carrington et al.[14] do not clarify the reasons for the persistently high number of individuals with a BP elevation in Australian clinical practice. It has been repeatedly shown that, in clinical practice, hypertensive patients have a very low adherence to lifestyle changes that reduce BP, and that this is the case also for antihypertensive drugs [15], the level of adherence exhibiting a clearcut inverse relationship with the risk of hospitalization for cardiovascular morbid or fatal events [16,17]. It has also been shown that another important factor is ‘clinical inertia’ [18], that is, the failure of physicians to start or modify treatment when BP is above control. Although Carrington et al.[14] could not provide information on patients’ adherence to treatment, their database sheds light on the importance of clinical inertia, given the overall high number of patients in whom detection of elevated BP values was not accompanied by antihypertensive treatment measures; and average BP values were not further reduced after the second visit despite remaining above control. Clinical inertia is a disappointing phenomenon which betrays physicians’ unawareness or distrust of scientific data. It is also, however, a phenomenon which can perhaps stand a greater chance of being favourably modified by well designed educational programs, with potentially important results. The study of Carrington et al.[14] has limitations, including few technical problems which the authors openly disclose and critically address. It has the additional merit, however, to provide large-scale evidence on visit-to-visit BP variability which was correctly assessed by the SD of the mean BP value throughout four to six visits in the same individual. Visit-to-visit BP variability has been recently reported to independently predict the risk of cardiovascular morbidity and mortality [19,20] and although the posthoc nature of the evidence so far available suggests that the results need to be interpreted with caution [21], it represents a promising additional tool to quantify patients’ protection by treatment. The merit of the study of Carrington et al.[14] is to provide data on visit-to-visit BP variability in a real life setting rather than in trials, as mostly done so far. It is also to describe conditions which make a greater visit-to-visit BP variability more likely. Although some results, for example greater visit-to-visit BP variability in aged patients and patients with higher BP values, are predictable (because BP variability bears a direct relationship with mean BP) others are not. The most interesting finding (perhaps not sufficiently emphasized by the authors) is that visit-to-visit BP variability did not show significant differences in patients under different treatments. This does not confirm the conclusion drawn by a meta-analysis of a large number of trials that visit-to-visit BP variability differs markedly between drug classes, with a minimum for calcium antagonists and a maximum for β-blockers. In that meta-analysis, however, visit-to-visit variability was quantified by a questionable method, that is, interindividual variability was taken as a proxy of the intraindividual one [22]. This makes the contribution of the paper by Carrington et al. [14] on this specific issue particularly important. ACKNOWLEDGEMENTS Conflicts of interest There are no conflicts of interest.