In search of optimal blood pressure

For the podcast associated with this article, please visit https://academic.oup.com/eurheartj/pages/Podcasts. The Framingham study established high blood pressure as a major cardiovascular risk factor,1 an observation that has been confirmed on numerous occasions to date.2 The seminal trials of Edward Freis using antihypertensive drugs in the late 1960s established hypertension as a cause of myocardial infarction, stroke, and death.3 However, what level of blood pressure should be achieved with antihypertensive therapy remained an ongoing controversy of the last decades.4,5 It started with 100 mmHg plus age in the systolic range, then 160/95 mmHg was recomended, later 140/80 mmHg was the target, and, most recently, the European6 and US Guidelines7 decreased the target even further, particularly in younger patients. The latter aspect is further analysed in the article ‘Reconsidering the cut-off diastolic blood pressure for predicting cardiovascular events: a nationwide population-based study in Korea’ by Si-Hyuck Kang and colleagues from the Seoul National University Bundang Hospital in Korea.8 They remind us that it is unclear whether a J-curve association exists in cardiovascular risk prediction and how independently systolic and diastolic blood pressure predict outcomes. To evaluate the association of systolic and diastolic blood pressure with major cardiovascular events (MACE), they studied 290 600 antihypertensive medication-naïve subjects with available blood pressure measurements and no history of MACE. The risk for MACE was lowest at systolic and diastolic blood pressures of 90‒99 mmHg and 40‒49 mmHg, respectively, above which blood pressures showed a log-linear risk prediction (Figure 1). The risk prediction of diastolic blood pressure was inconsistent when stratified by systolic blood pressure. A wider pulse pressure rather than a higher diastolic blood pressure was associated with cardiovascular outcomes among men aged ≥55 years. Thus, elevated blood pressure is a strong predictor of future cardiovascular events including cardiac death, myocardial infarction, stroke, and heart failure, with a log-linear relationship. Although hypertension is defined using a lower systolic blood pressure cut-off of ≥130 mmHg, the diastolic blood pressure component of ≥80 mmHg seems disproportionately low. Distribution of systolic and diastolic blood pressure and 10-year rates of major cardiovascular events. The scatter plot shows the distribution of systolic and diastolic blood pressure. The numbers on the scatter plot represent the rates of cardiovascular events per 1000 person-years of each group (from Choi Y-J, Kim S-H, Kang S-H, Yoon C-H, Lee H-Y, Youn T-J, Chae I-H, Kim C-H. Reconsidering the cut-off diastolic blood pressure for predicting cardiovascular events: a nationwide population-based study from Korea. See pages 724–731). This issue was further investigated in the MONICA/KORA Study in the article entitled ‘Association of hypertension cut-off values with 10-year cardiovascular mortality and clinical consequences: a real world perspective from the prospective MONICA/KORA study’ by Karl-Heinz Ladwig and colleagues from the Helmholtz Zentrum Munchen Deutsches Forschungszentrum fur Umwelt und Gesundheit in Neuherberg, Germany. They investigated the clinical value of lower blood pressure cut-offs of 130–139/80–89 mmHg for Stage 1 hypertension compared with Stage 2 cut-offs of ≥140/90 mmHg in 11 603 participants.9 The implementation of the new Stage 1 cut-offs increased the prevalence of hypertension from 34% to 63%. Only 24% of Stage 2 hypertensives were under treatment. Within 10 years, 370 fatal MACE occurred. The adjusted cardiovascular mortality was 1.61/1000 persons in Stage 2 and 1.07/1000 cases in Stage 1 hypertension compared with normal blood pressure. The hazard ratio for Stage 2 and cardiovascular mortality was 1.54, and in the presence of competing risks it was 1.47. However, significance for Stage 1 hypertension was not reached. Among Stage 2 participants, there was a higher prevalence of depressed mood in treated compared with non-treated patients. Thus, lower blood pressure cut-offs substantially increase hypertension prevalence, while capturing a population with lower cardiovascular mortality. The higher prevalence of depressed mood in treated participants might reflect a labelling effect. These provocative findings are put into context in a balanced Editorial by Ernesto L. Schiffrin from the Jewish General Hospital in Montreal, Canada.10 Renal denervation is a novel treatment modality for elevated blood pressure.11 Although modern ablation catheters using radiofrequency or ultrasound energy proved effective in lowering blood pressure,12 their effect on heart rate, a surrogate parameter of sympathetic activation, is less clear. In their article entitled ‘Ambulatory heart rate reduction after catheter-based renal denervation in hypertensive patients not receiving anti-hypertensive medications: data from SPYRAL HTN-OFF MED, a randomized, sham-controlled, proof-of-concept trial’, Michael Böhm and colleagues from the University Hospital Saarland in Homburg, Germany report the results of the SPYRAL HTN-OFF MED trial in 80 non-medicated hypertensives randomized 1:1 to renal denervation or a sham procedure.13 There was a significant reduction in 24-h heart rate by 2.5 bpm at 3 months with renal denervation compared with sham (–0.2 bpm). Mean baseline-adjusted treatment differences were significantly different between groups at 3 months for average morning heart rate (–4.4 bpm) and minimum morning heart rate (–3.0 bpm). Renal denervation patients with baseline 24-h heart rate above the median had significant reductions in ambulatory systolic blood pressure of 10.7/7.5 mmHg, whereas blood pressure changes in patients with below median heart rates were not significant (Figure 2). Thus, average and minimum morning heart rates were significantly reduced at 3 months with renal denervation compared with sham treatment. A baseline 24-h heart rate above the median predicted greater blood pressure reductions and may allow physicians to select patients likely to respond to the procedure. This observation is put into context in a thought-provoking Editorial by Murray David Esler from the Baker Heart Research Institute in Melbourne, Australia.14 Physiology of sympathetic activation by the interaction of the brain with the kidneys involved in hypertension and high heart rate as well as other sympathetically driven disturbances such as arrhythmias, metabolic disease, heart failure, and sleep apnoea. Renal denervation, by using contemporary approaches allowing renal artery main stem and branch treatments, exerts effects not only on blood pressure but also on ambulatory circadian heart rate parameters. (A) Sympathetic activation is generated in the brain and stimulates the heart, the vessels, and the kidney by efferent nerves, which mediate the physiological and pathophysiological effects. In turn, afferent nerve stimulation from the kidney to the sympathetic nervous system increases central sympathetic outflow, producing a vicious cycle effect on whole-body sympathetic activity. Renal denervation interrupts afferent and efferent signalling to the brain, reducing sympathetic activity. The contemporary technique with the SPYRAL catheter system allowing reliable circumferential as well as branch plus main stem treatments has been used. (B) The consequences are a reduction of 24-h heart rate, morning heart rate, day time heart rate, and night-time heart rate at 3 months, which was significantly different compared with placebo, in particular at a higher baseline heart rate. While the blood pressure-reducing effects have been reported previously and the heart rate effects are reported herein, the significance of experimental findings such as antiarrhythmic effects, metabolic effects, effects on heart failure models, and sleep apnoea need to be scrutinized in clinical studies (from Böhm M, Mahfoud F, Townsend RR, Kandzari DE, Pocock S, Ukena C, Weber MA, Hoshide S, Patel M, Tyson CC, Weil J, Agdirlioglu T, Fahy M, Kario K. Ambulatory heart rate reduction after catheter-based renal denervation in hypertensive patients not receiving anti-hypertensive medications: data from SPYRAL HTN-OFF MED, a randomized, sham-controlled, proof-of-concept trial. See pages 743–751). Ischaemic and haemorrhagic stroke is a well-known complication of hypertension with a clear pressure–response relationship.15–17 In their article entitled ‘Epidemiology report: trends in sex-specific cerebrovascular disease mortality in Europe based on WHO mortality data’, Nicholas Townsend and colleagues aimed to examine gender-specific trends in cerebrovascular disease and subtypes, i.e. ischaemic stroke, haemorrhagic stroke, and subarachnoid haemorrhage, in Europe between 1980 and 2014 using the WHO global mortality database.18 The last 35 years have seen large overall declines in cerebrovascular disease mortality rates in the majority of European countries. While these declines have continued steadily in more than half of countries, recently this trend plateaued and even increased in a number of countries. Analysis by stroke subtype revealed that recent plateauing was most common for ischaemic stroke, while increases were most common for subarachnoid haemorrhage. Thus, there is an ongoing need for continued research into the inequalities in both current stroke mortality outcomes and trends across Europe. These concerning findings are further discussed in an Editorial by Barbara Casadei from the John Radcliffe Hospital in Oxford, UK.19 Inflammation has also been implicated in the pathogenesis of hypertension20,21 and left ventricular remodelling.22 Recent data suggest that macrophages (Mφ) contribute to, and protect from, hypertension. Endothelin-1 (ET-1) is the most potent endogenous vasoconstrictor with additional pro-inflammatory properties. In their article ‘A novel role for myeloid endothelin-B receptors in hypertension’, Neeraj Dhaun and colleagues from the University of Edinburgh in the UK investigated the role of ET-1 in experimental, angiotensin II-induced, and clinical hypertension by modifying Mφ number and phenotype.23 Mouse Mφ and human monocytes expressed both ETA and ETB receptors, and displayed chemokinesis to ET-1. However, stimulation of Mφ with exogenous ET-1 did not polarize the Mφ phenotype. Both mouse and human Mφ cleared ET-1 through ETB receptor-mediated endocytosis. Mφ depletion resulted in an augmented acute and chronic hypertensive response to ET-1. Mice with myeloid cell-specific ETB receptor deletion displayed an exaggerated response to ET-1 and angiotensin II. Finally, in patients who received Mφ-depleting immunotherapy, blood ptressure was higher and endothelial function worse than in those receiving non-depleting therapies. Thus, Mφ and ET-1 appear to play an important role in blood pressure control and potentially have a critical role as a therapeutic target in hypertension. This issue is complemented by two Discussion Forum contributions. First, Stuart J. Pocock and colleagues from the London School of Hygiene and Tropical Medicine in the UK comment on the recently published article ‘Association between systolic blood pressure and dementia in the Whitehall II cohort study: role of age, duration, and threshold used to define hypertension’ by Jessica G. Abell and colleagues from University College London in the UK.24,25 In turn, Abell and colleagues respond to the comments by Pocock et al. 26 In a second Discussion Forum contribution, Christian Torp-Pedersen and colleagues from the Aalborg University and Aalborg University Hospital in Denmark comment on the recently published contribution entitled ‘Asleep blood pressure: significant prognostic marker of vascular risk and therapeutic target for prevention’ by Ramon C. Hermida and colleagues from the University of Vigo in Spain.27,28 Hermida et al. respond to this contribution.29 The editors hope that readers of this issue of the European Heart Journal will find it of interest. With thanks to Amelia Meier-Batschelet for help with compilation of this article.