The September issue of the Journal of Hypertension includes a meta-analysis on a topic that has gained much interest after the demonstration, years ago, that hypertension is an important risk factor for the development of cognitive dysfunction and dementia. Since then, individual studies and meta-analyses have looked at whether there is a reduction in the risk of dementia when blood pressure (BP) is reduced by treatment. The meta-analysis published by van Middelaar et al. (pp. 1780–1787) in the present issue of the Journal shows that, compared with a control group, dementia was not affected by BP-lowering treatment, mostly via antihypertensive medicaments. It also shows, on a smaller database, that this seems to be the case for either vascular dementia and Alzheimer disease, both of which have been related to hypertension in epidemiological studies. The results of van Middelaar and colleagues appear to be solid because the number of patients (almost 60 000) was huge and generated a large number of dementia cases (n = 2131). Yet, in my opinion, the possibility that reductions of an elevated BP protect against dementia deserves an appeal. In the present meta-analysis, the average follow-up was 3.9 years, and it cannot be excluded that longer time periods are needed to allow a favourable effect of BP-lowering treatment to surface. It would be also desirable to quantify cognitive function by more sensitive means than those so far used in many studies. One can hardly think of a more important topic for future research, given the steep progressive senescence of the population, especially in industrialized countries.
The original articles start with two interesting reports on BP measurements. In one report, Gazzola et al. (pp. 1798–1802) found that self-measurement of BP led to an about 7 mmHg SBP increase, a change that was paralleled by a tachycardic response and exhibited an attenuation after multiple BP readings with no, however, complete disappearance. This questions the widespread notion that, at variance from office BP, home BP does not elicit an alerting reaction. Based on Gazzola et al.'s observations, it appears that home BP may be affected by an emotionally related BP rise, although quantitatively less marked than that affecting conventional office BP measurements. In the other report, Pressler et al. (pp. 1803–1809) produce an impressive set of results on the BP response to maximal exercise in athletes (n = 2419). On average, the exercise-induced increase in SBP was extremely large, with higher exercise-related absolute values (>200 mmHg) in men than in women and a maximal increase in endurance athletes. Forty-three percent of men and 28% of women exceeded the pressor limits indicated by guidelines and could thus be categorized as having an exaggerated response. The prognostic significance of the BP responses to exercise is under debate, and in this context an additional finding of Pressler et al is that the pressor effect of exercise showed a relationship with the left ventricular mass value of the athletes suggesting that it may have an adverse implication. Further considerations will be found in the accompanying editorial commentary of de Buyzere and Rietzschel (pp. 1788–1790).
Several other articles deal with various aspects of the epidemiology of hypertension. Palatini et al. (pp. 1810–1815) report the results of an analysis of the HARVEST study, which focused on the risk of younger participants (<45 years of age) with isolated systolic hypertension at office visits to develop sustained hypertension as identified by ambulatory BP monitoring. The results show that, compared with a normotensive group, individuals with isolated office systolic hypertension did not exhibit an increased risk of developing ambulatory hypertension, at variance from the increase seen in those with office diastolic or systo-diastolic hypertension. Although the question of the prognostic significance of isolated systolic hypertension of the youth remains open, these results score in favour of the clinical innocence of this condition. Further information can be found in an expert consensus article recently published in the Journal of Hypertension and in the accompanying editorial of Bursztyn (pp. 1791–1792). Ebrahimi et al. (pp. 1816–1824) report the results of a survey on the prevalence of hypertension in children (6–12 years of age) living in northeastern Iran. On the basis of conventional diagnostic criteria for that age, about 7% of the children were diagnosed as having a BP elevation, and more than 7% were found to belong to the high BP range category, with a surprisingly greater risk of having a BP abnormality in rural rather than in urban areas. BP studies in children and adolescents are submitted to (and published in) the Journal of Hypertension with greater frequency, expanding information on a crucial area for mechanistic and clinical BP research. Humbert et al. (pp. 1825–1832) and Cuspidi et al. (pp. 1833–1839) provide interesting observations on the cardiovascular risk associated with the white-coat effect and white-coat hypertension. In the former article, individuals from the Gubbio cohort showed an increased risk of cardiovascular events if the decrease of SBP from the first to the third measurement made during the visit (taken as to reflect a white-coat effect) was greater than 10 mmHg. In the latter article, participants of the PAMELA cohort showed white-coat (and masked) hypertension to be accompanied by a greater long-term risk of an incident metabolic syndrome. In both instances, the message appears to be that alerting-induced BP alterations during an office visit should not be regarded as artefacts to get rid of by modifications of the BP measurement approach (e.g. unattended BP measurements), but that they have a predictive value for patients’ outcome. Finally, Lurbe et al. (pp. 1840–1846) show that in 333 Caucasian children and adolescents, serum uric acid has a positive association with office and ambulatory BP as well as with plasma insulin and triglycerides. This extends to a younger age the evidence that uric acid coexists with hypertension and dysmetabolic risk factors repeatedly reported for adult individuals, supporting uric acid as a routine added measure for an appropriate quantification of the risk factor profile at all ages.
The next seven articles deal with mechanistic, diagnostic and prognostic aspects of organ damage. Zhao et al. (pp. 1847–1857) describe a gene that exerts an antihypertrophic influence in mouse angiotensin-II-induced models of cardiac hypertrophy and demonstrate that this influence is exerted by opposing the influence of PDE5A, that is, an important factor in the production of such cardiac abnormality. This represents a further example of the growing tendency of genetic research to elucidate the path through which genes operate and achieve an effect. Saeed et al. (pp. 1858–1864) show that not only middle age but also younger survivors of an ischemic stroke (age range 15 to <60 years) exhibit a high prevalence of structural abnormalities of the left ventricle, which may contribute to the high incidence not only of stroke recurrence but also of cardiac and other cardiovascular events that characterizes the post-stroke state. Van Kleef et al. (pp. 1865–1873) and Lehtonen et al. (pp. 1874–1881) address the prognostic potential of the electrocardiogram for cardiovascular outcomes and atrial fibrillation, respectively. In the article of van Kleef and colleagues, left ventricular hypertrophy was associated with an increased risk of cardiovascular outcomes regardless the ECG criterion employed to detect this cardiac abnormality. According to the authors, however, the Cornell/strain index had some advantages over the others (Sokolow–Lyon, Cornell product, Framingham) because of a greater diagnostic ability and an association with a higher outcome risk. In the article of Lehtonen et al., several abnormalities detectable by conventional 12-lead ECG predicted the risk of incident atrial fibrillation in a large cohort (n = 5813) of individuals followed for about 12 years, with some differences in the predictive ability of the individual abnormalities between normotensive and hypertensive patients. The interest of these observations for clinical practice is emphasized in the accompanying editorial by Shin et al. (pp. 1793–1794), which also reminds, however, that ECG changes contributing to the development of atrial fibrillation may often be too small to be visualized by conventional recordings, and that thus use of ECG as a biomarker of future atrial fibrillation would greatly benefit of more sensitive and accurate ECG measuring approaches There are then two articles on arterial stiffness and one on the effect of age on the relationship of BP with organ damage. Polonis et al. (pp. 1882–1888) provide evidence of an association of a multilocus genetic score with arterial stiffness in hypertensive patients from the CARE NORTH cohort whereas Tanaka et al. (pp. 1889–1894) show, from the data-base of the ARIC study, that arterial stiffness is lower in older adults with higher physical activity as well as in people who were physically active in mid to late-life. Thus, distensibility of the large arteries is under modulation of both genetic and environmental factors. Finally, Olesen et al. (pp. 1895–1901) report on the relationship between BP and various measures of organ damage (arterial stiffness, microvascular damage and left ventricular hypertrophy), which is known to be closer for the ambulatory than for the office BP values. The novelty of the study is that the association between organ damage and measures derived from ambulatory BP monitoring becomes closer with ageing, making ambulatory BP an even better marker (and perhaps cause) of the organ damage status in older as compared with younger individuals. The measures considered include BP variability, an issue further discussed in the editorial of Palatini (pp. 1795–1797).
The last five articles address treatment aspects of hypertension. Rubattu et al. (pp. 1902–1914) show that in the stroke-prone hypertensive rat, the combination of AT1 receptor blockade with neprilysin inhibition (sacupitril/valsartan) lowers BP and improves organ damage (proteinuria) more effectively than AT1 receptor blockade alone, its administration achieving a complete prevention of stroke over a long-term treatment period as well. Originally developed for the treatment of heart failure, the dual mechanism of action that characterizes this new medicament appears to be of potential advantage also for hypertension, in which collection of clinical evidence would be highly desirable. Laurent et al. (pp. 1915–1920) report an analysis of a large patient population (n = 5496) in which initial treatment with very small doses of perindopril and amlodipine (3.5/2.5 mg) led to a greater 1-month BP reduction than that seen with various blockers of the renin–angiotensin system (perindopril 5 mg, irbesartan 150 mg, valsartan 80 mg) given in monotherapy, with no inconvenience for the tolerability profile. In patients at high cardiovascular risk, a prompter BP reduction may result in a more timely protection against the risk of early BP events. In addition, as emphasized by the recent European guidelines, an initial two-drug combination treatment may be associated with long-term advantages that make this treatment strategy the preferable one in most hypertensive individuals. Finally, three interesting studies address various aspects of the relationships between BP-lowering treatment and events. Liu et al. (pp. 1921–1928) report the results observed by retrospectively analyzing the post-trial data from the CASE-J study, for a total follow-up of 10 years. No difference in the risk of cardiovascular events was seen between patients treated with candesartan or amlodipine, although patients in the candesartan group showed a 29% significant reduction in the risk of incident diabetes, in line with what has been reported in a number of trials. Wändell et al. (pp. 1929–1935) show that in a very large primary care data-base from individuals with atrial fibrillation, the 5.7-year risk of heart failure was lower for treatments based on thiazide diuretics, vessel active calcium channel blockers, and nonselective beta-blockers (mostly sotalol), the risk reduction in fully adjusted models being 26, 24, and 16%, respectively. No benefit of antihypertensive treatment is on the other hand reported in the article of Park et al. (pp. 1936–1941) on patients with a subacute ischemic stroke because of a stenosis of internal carotid or middle cerebral arteries of at least 50% of the vessel lumen. After 7–42 days from the acute episode, patients were randomized to a more or less intense SBP reduction (<120 or <140 mmHg, actual on-treatment mean values 125 and 132 mmHg, respectively), which was continued for about 6 months. At the end of the treatment period the volume of the ischemic lesion was higher in the intense compared with the less intense treatment group, indicating that the widespread notion that the lower the BP achieved by treatment the greater is the protective effect on the brain cannot be generalized and it may definitively not apply to patients with significant stenosis of the intracranial arteries. The time may have come to suggest a more frequent search for this condition before deciding the target BP to aim for with antihypertensive treatment, particularly considering the lower target value now recommended by guidelines.
Conflicts of interest
There are no conflicts of interest.