Assessing antihypertensive treatment by real life data

In the past decade the perception has grown that information on antihypertensive treatment needs to be extended beyond that offered by clinical trials with a controlled design. This is because, although providing reliable data on the ability of a given antihypertensive drug or treatment regimen to lower an elevated blood pressure (BP) and achieve BP control with an acceptable safety and tolerability profile, controlled clinical trials are performed in conditions that are very different from those characterizing medical practice. To make few examples, in order to limit confounders and favour an appropriate data interpretation in clinical trials the demographic and clinical characteristics of the patients exposed to the treatment under investigation are much more homogeneous than the highly diversified ones typical of the population exposed to the same treatment after it enters the therapeutic market. Two, in clinical trials patients are followed closely and adherence to the antihypertensive treatment regimen is usually high and similar between the groups that are compared, a prerequisite for the results to be scientifically acceptable. In contrast, medical practice is associated with an extremely high rate of treatment discontinuation [1], not to mention other types of inadequate compliance [2], and achieved BP control is no more than half, and more frequently only one-third [3] of the control obtained throughout the therapeutic phase of clinical trials [4,5]. Three, physicians involved in clinical trials usually have an above-average expertise, which makes clinical judgement more sound and diagnostic or therapeutic errors less common than those made by practising physicians, with a nonmarginal contribution of the patients themselves who not infrequently self-modify the prescribed treatment regimen during the often long intervals from one visit to the next one. Finally, the evidence provided by controlled trials covers only a small fraction of the time over which antihypertensive treatment must be implemented, which extends over the entire life expectancy of the patient [6]. This is the case for event-based trials which for a variety of reasons (increasing rates of patients’ drop-out, changes of investigators’ position and working site, cost, etc.) do not last more than few years. It is even more the case for drugs approved for use only based on the demonstration of their BP-lowering efficacy, because in this instance the evidence is usually limited to trials of a few months duration and the long-term persistency of the BP reduction is inferred from an uncontrolled extension of the shorter controlled phase of the trials usually to 1 year. The above considerations explain current interest on data sources such as administrative databases and registries, which allow information to be obtained on real-life use of antihypertensive treatments, often from a huge number of patients examined over extended time periods. They also justify the increasing attention of medical journals to these results, which are more and more commonly regarded as a useful, if not necessary, complement to the results provided by controlled trials, rather than as products of second-rate research. The study by Volpe et al.[7], published in this issue of the Journal of Hypertension, is a pertinent example. In this study Volpe et al. made use of the Italian web-based drug monitoring system to retrospectically analyse what happened in a large number of hypertensive patients (n = 11 511) followed by a large number of physicians (n = 6464) when aliskiren was added to an existing antihypertensive drug treatment that had left BP values uncontrolled. The results show that systolic and diastolic BP underwent a sizeable reduction after 1 month which became greater after 6 months and was associated with a decrease in the concomitant use of other antihypertensive medications and a limited number of side effects. Although the uncontrolled nature of the study makes quantitative interpretation of the results inappropriate (a limitation acknowledged by the authors) there is little question that, despite its greater complexity, real life data on aliskiren are in line with those previously obtained by controlled trials, supporting the overall conclusion that this new drug-based strategy is a viable practical option. One final point deserves mention. Technical progress is improving the quality and extension of the information provided by registries which will presumably extend in a few years time to a large number of countries. This makes it likely that more and more data of this type will become available in the future and that this will result more and more frequently in between-country comparisons as well as in assessment of temporal trends. In this context, standardizing data collection and analysis of individual registries may represent a necessary goal. Acknowledgements Conflicts of interest G.M. has received speakers fee and honararia from Boehringer Ingelheim, Bayer, Daiichi Sankyo, Menarini Int. Novartis, Sanofi Aventis, Servier and Takeda.