The November issue of the Journal of Hypertension includes two reviews and meta-analyses on the clinical value of ambulatory vs. office blood pressure (BP). Onakpoya et al. (pp. 2084–2094) performed a meta-analysis of available studies on the prognostic impact of office vs. 24-h SBP, using data from 13 studies for a total of more than 81 000 adults. In line with the conclusion of the PAMELA and other studies published several years ago, a 10 mmHg increase of 24-h SBP was associated with a greater risk of all-cause mortality than a 10 mmHg increase of office SBP. This does not ‘per se’ reflect a greater predictive ability because, as a mean, 24-h BP has a narrower distribution in the population or the analyzed cohorts, which can account for a steeper BP-outcome relationship. Like previous meta-analyses, however, Onakpoya et al. also showed that 24-h SBP maintained a significant relationship with mortality when data were adjusted for office BP whereas the reverse was not the case. Although adjustment for a variable does not mathematically exclude its contribution to the data of interest, this does suggest a greater predictive ability of total mortality by ambulatory over office BP. The meta-analysis also focused on some important methodological aspects of the studies comparing the prognostic value of office vs. out-of-office BP, one of them being the limitations of the meta-analyses that make use of average rather than individual data from the selected studies. Another is the study quality, which Onakpoya et al. define as only moderate. In my opinion, a number of these studies suffer from insufficient information on the accuracy and standardization of the BP measurements, office BP in particular, raising the question whether, at least in some studies, out-of-office BP is compared with poor quality and standardization of office BP measurements. Bo et al. (pp. 2095–2109) confirm, in a large meta-analysis of studies published until May 2019, that ambulatory 24-h, day or night BP values exhibit an excellent reproducibility when measured over approximately 1-month interval. As emphasized by the authors, this can be an important advantage for research on nonpharmacological or pharmacological antihypertensive treatment, as a good reproducibility reduces the chance of casual BP reductions to obscure the drug-related effects. It does not mean, on the other hand that ambulatory BP data are highly reproducible in all or most individuals. Contrary to what happened to mean values, individual ambulatory BP values exhibited over a 1-month interval, considerable changes in one direction or another. Bo et al. emphasize, in particular that this was the case for the nocturnal BP-dipping phenomenon, as reported by the SAMPLE study many years ago. This is not surprising as sleep quality (and, thus its influence on BP) varies from day to day but the poor dipping reproducibility may definitely represent a problem for an accurate determination of the prognostic importance of night-time BP.
The above two reviews are followed by a review on the relationship between chronic kidney disease and cardiovascular risk, a review on the BP effects of inorganic nitrates and an Editorial on the scientific value of network meta-analyses. Carmena et al. (pp. 2110–2121) provide an overview of the close relationship between chronic kidney disease and cardiovascular morbid or fatal events, with a detailed description of additional risk factors involved and the treatment strategies that most effectively protect nephropathic patients. Li et al. (pp. 2122–2140) report that, in more than 1100 individuals, repeated (≥3 days) administration of inorganic nitrates is accompanied by a modest SBP reduction (about 3 mmHg), which extends to central BP (about 2 mmHg). Thomopoulos and Mancia (pp. 2141–2143) argue that network meta-analyses of outcome trials have, as a main problem, loss of trial randomization. Namely, that to include the largest possible amount of data, network meta-analyses compare subgroups from one or more trials to subgroups from other trials, which can lead to major demographic and clinical imbalances between the compared groups that adjustment procedures can hardly cope with and the results obtained by this statistical approach should thus be taken with caution.
The following five papers deal with BP measurements. Gotzmann et al. (pp. 2154–2160) report the results of calibrating oscillometric aortic BP measurements via mean/diastolic rather than systo/diastolic BP, and conclude that differences in the accuracy of the calibration between these two approaches were minor. The problems associated with accuracy of noninvasive central BP estimates are discussed, more in general, in the Editorial Commentary of Pucci et al. (pp. 2146–2147). Grillo et al. (pp. 2161–2168) focus on the shortcomings of calculating mean BP by adding one-third of pulse pressure to the diastolic value. The authors show that in 1526 patients, the pulse pressure value to be added to DBP in order to obtain the mean varied markedly (23--58%) among individuals when the addition was based on the pulse waveform rather than on a fixed pulse pressure portion. The differences were related to a number of factors, such as the level of DBP, sex, age, mental stress, and so on, which might serve as correction factors, although, according to the authors, regular use of pulse waveform measurement would be preferable. Jang et al. (pp. 2169–2175) examined the duration of the orthostatic test necessary to properly identify orthostatic hypotension and conclude that early (i.e. within 1 min) BP measurements may be necessary in the elderly while in younger people (<40 years of age), BP needs to be checked over several minutes. These are important observations as orthostatic hypotension is by no means rare and data are available on its short (injurious falls) and long-term adverse prognostic significance. Stabouli et al. (pp. 2176–2184) describe the association of executive function performance with central and ambulatory BP whereas Ikemiyagi et al. (pp. 2185–2191) show an association between the ankle--brachial index and electrocardiographic left ventricular hypertrophy in a large number of participants (n = 13 396; age 19–89 years) in the Okinawa Peripheral Arterial Disease Study. From a pathophysiological perspective, this is not surprising as an alteration of the ankle--brachial index is likely to reflect an increase of large artery stiffness, which is an important determinant of arterial impedance and cardiac afterload.
The next eight articles address epidemiological aspects of hypertension. Ma et al. (pp. 2192–2197) show that, in the Chinese cohort of the Comprehensive Geriatric Assessment Study, frailty (assessed by the frailty index) was frequently correlated with age and affected almost 14% of the elderly fraction of the study population. Frailty was less common in normotensive than in hypertensive individuals in whom, however, awareness of the problems posed by high BP and the extent of antihypertensive treatment was better than in nonfrail individuals. The problems posed by studies and interpretation of the data in frail patients are discussed in the Editorial Commentary of Ungar and Ravasi (pp. 2148–2149). Bika Lele et al. (pp. 2198–2204) report that in Bantus from southern Cameroon, age-adjusted prevalence of hypertension was greater than in Pygmies and that in both ethnic groups high BP (identified by a BP ≥140/90 mmHg or antihypertensive drug use) was substantially more common in urban than in rural residents. Al Sami et al. (pp. 2205–2214) provide yet another set of data on the inverse association between birth weight and BP values in the adult life. In a cohort of more than 4000 Australian participants, the evidence was that each kilogram of birthweight was independently associated with an about 2 mmHg lower SBP and that a birthweight lower than 2.5 kg was accompanied by an increased risk of hypertension. Zhang et al. (pp. 2215–2222) report that, in 7225 primary school children, an increase in the 1-year average concentration of SO2, O3 and of particulte matter particles with an aerodynamic diameter less than 2.5 μm (PM 2.5) had a measurable pressor effect and that this was more evident in children with a higher intake of sugar-sweetened beverages. Han et al. (pp. 2223–2229) show an association of low relative skeletal muscle mass and the incidence of hypertension in men, an observation made also in other studies which is still devoid of an understanding of the responsible mechanisms. The Editorial Commentary of Buckley et al. (pp. 2150–2151) provides an in-depth view of the problem. Hisamatsu et al. (pp. 2230–2236) show that in 4776 individuals with a mean age of 39.4 years, those with isolated systolic hypertension (ISH) were 389 (8.1%). Over a 29-year follow-up, 115 ISH patients died from a cardiovascular disease, which made their cardiovascular risk markedly greater (odds ratio 4.10, 95% confidence interval 1.87–9.03) than that of individuals with normal BP. Thus, evidence is growing that ISH at a relatively young age is characterized by an increased risk, rather than being an innocent condition as maintained some time ago. This conclusion finds strong support in the data of Hisamatsu et al. both because of the respectable number of relatively young ISH patients and because of the long follow-up. Del Pinto et al. (pp. 2237–2244) analyzed the SPRINT data and found that visit-to-visit SBP variability was associated with the risk of cardiovascular outcomes in a U fashion, that is both high and low SBP variability were accompanied by an increased risk compared with intermediate values. An adverse prognostic significance of visit-to-visit BP variability has been reported in a large number of, although not in all, studies. Reports on a parallel increase in cardiovascular risk associated with low variability are more rare and difficult to be mechanistically explained. This is the case, more in general, for the entire visit-to-visit BP variability phenomenon, on which studies addressing more in depth their determinants would be highly desirable. Finally, Tadic et al. (pp. 2245–2251) show that in untreated hypertensive patients both nondipping and reversed dipping were related to an impaired function of the left atrium, reverse dipping exhibiting the most adverse effect. Since a few years, measurement of atrial structure and function has become a routine component of echocardiographic examinations, and evidence is now strong that this information significantly contributes to the assessment of the effects of BP and other risk factors on the heart.
Five articles of the November issue describe the relationships between BP, other risk factors and organ damage in specific clinical conditions. Xie et al. (pp. 2252–2260) report the association between BP and mortality in an extremely large number of patients under peritoneal dialysis. During a median follow-up of about 3 years, the relationship between BP and cardiovascular or all-cause mortality exhibited a U-shaped pattern, the risk of death being higher both at a BP greater than 141/85 mmHg and at a BP less than 119/69 mmHg. Although with the caution due to their observation nature, these findings are important in an area in which information and controlled investigations are scanty. Chen et al. (pp. 2261–2269) found an impairment of right ventricular function (speckle tracking echocardiography) in primary aldosteronism, an observation that is addressed in the Editorial Commentary of Petramela and Letizia (pp. 2152–2153). Wang et al. (pp. 2270–2278) found orthostatic hypotension in one out of four patients with Parkinson's disease without a history of BP abnormalities. Interestingly, a reverse dipping pattern was present in more than 50% of these patients in whom it helped identification of the orthostatic hypotension phenomenon. Viazzi et al. (pp. 2279–2286) show that, in about 9000 patients with type 2 diabetes and normal estimated glomerular filtration rates, visit-to-visit BP variability predicted both incident hypertension and worsening of albuminuria. This addresses an area in which evidence is still limited and not entirely univocal. In the study of Viazzi et al., however, the impact of an increased visit-to-visit BP variability was clear, that is, a variability in the upper quartile increased the risk of incident hypertension by more than 50%, and the risk of worsening albuminuria by 20%. Magnusdottir et al. (pp. 2287–2294) show that in patients with obstructive sleep apnoea and high cardiovascular risk participating in the Heart Biomarker Evaluation in Apnea Treatment (HeartBEAT) study, several interventions were accompanied by favourable results, including a reduction of sleep BP by continuous positive airway pressure (CPAP) treatment.
The last four articles are devoted to experimental hypertension (two articles) and antihypertensive drug treatment (two articles). Palei et al. (pp. 2295-2304) report that, in a genetic model of rat obesity, hypertension in pregnancy was associated with lower levels of the placental growth factor, whose administration in the human recombinant form led to a BP reduction. Caniffi et al. (pp. 2305–2317) show that, in SHR and normotensive rats, a 14-day administration of C-type natriuretic peptide attenuated hypertension and diminished the inflammatory and structural (fibrosis) cardiac alterations with a regression of the associated left ventricular hypertrophy. Zanisi et al. (pp. 2318–2324) show that BP-measuring errors as well as insufficient information on the variability of the BP response to antihypertensive drugs have an important impact on the assessment of patients failing to achieve BP control by treatment. Larger use of repeated BP measurements or alternative BP measurement techniques may significantly reduce this phenomenon. Gosse et al. (pp. 2325–2330) review the historical background and the pathophysiological aspects of malignant hypertension, a condition that in Europe has been drastically reduced by early preventive measures and more modern pharmacological treatments. Drastic reduction does not mean disappearance, however, which means that the level of attention should not decline. Malignant hypertension remains relatively prevalent in low-income countries.
The last contribution to the November issue of the Journal is a description of the HyperChildNET COST-Action project by the Chairman, Professor Empar Lurbe (Valencia University), Vice-President of the European Society of Hypertension (ESH). HyperChildNET COST-Action is a research project, which has been funded by the European Community to improve prevention and treatment of hypertension and cardiovascular disease in the young age. The study will involve a large number of interdisciplinary European centres over the next 4 years. It was designed based on the indications of Professor Lurbe, the ESH Working Group on Hypertension in Children and Adolescents, and the ESH Hypertension Guidelines in Children and Adolescents. To Professor Lurbe, the core investigators of the project, and the ESH, the congratulations of the Journal of Hypertension for this important achievement.
Conflicts of interest
There are no conflicts of interest.