Takayoshi Ohkubo

Thirty years of research on diagnostic and therapeutic thresholds for the self-measured blood pressure at home.

ObjectiveThe goal of this review study is to summarize 30 years of research on cut-off limits for the self-measured blood pressure. MethodsWe reviewed two meta-analyses, several prospective outcome studies in populations and hypertensive patients, studies in pregnant women, three clinical trials and the thresholds proposed in earlier and current hypertension guidelines. ResultsIn line with existing guidelines, prospective studies support that levels of the self-measured blood pressure at home of greater than or equal to 135u2009mmHg systolic or greater than or equal to 85u2009mmHg diastolic indicate hypertension. Circumstantial data suggest that levels of the self-measured blood pressure below 120/80 and 130/85u2009mmHg are optimal and normal, respectively. Therapeutic targets of the self-measured blood pressure to be attained on antihypertensive drug treatment are currently unknown, but should logically be lower (<135/85u2009mmHg) than those used to diagnose hypertension. Currently, there is no proof that therapeutic thresholds for the home blood pressure should be lower in high-risk compared with normal-risk patients. A large body of evidence, however, demonstrated that each millimetre of mercury of blood pressure lowering counts in the prevention of cardiovascular complications and that in high-risk patients even small decreases in blood pressure result in large absolute benefit. ConclusionThe thresholds to diagnose hypertension from self-measured blood pressure readings at home remain unaltered since the 2000 consensus conference, but are currently supported by outcome data. Further studies need to establish what values of the self-measured blood pressure are optimal and normal in terms of cardiovascular outcome.

Is blood pressure during the night more predictive of cardiovascular outcome than during the day

The objective of this study was to investigate the prognostic significance of the ambulatory blood pressure (BP) during night and day and of the night-to-day BP ratio (NDR). We studied 7458 participants (mean age 56.8 years; 45.8% women) enrolled in the International Database on Ambulatory BP in relation to Cardiovascular Outcome. Using Cox models, we calculated hazard ratios (HR) adjusted for cohort and cardiovascular risk factors. Over 9.6 years (median), 983 deaths and 943 cardiovascular events occurred. Nighttime BP predicted mortality outcomes (HR, 1.18–1.24; P<0.01) independent of daytime BP. Conversely, daytime systolic (HR, 0.84; P<0.01) and diastolic BP (HR, 0.88; P<0.05) predicted only noncardiovascular mortality after adjustment for nighttime BP. Both daytime BP and nighttime BP consistently predicted all cardiovascular events (HR, 1.11–1.33; P<0.05) and stroke (HR, 1.21–1.47; P<0.01). Daytime BP lost its prognostic significance for cardiovascular events in patients on antihypertensive treatment. Adjusted for the 24-h BP, NDR predicted mortality (P<0.05), but not fatal combined with nonfatal events. Participants with systolic NDR of at least 1 compared with participants with normal NDR (≥0.80 to <0.90) were older, at higher risk of death, but died at higher age. The predictive accuracy of the daytime and nighttime BP and the NDR depended on the disease outcome under study. The increased mortality in patients with higher NDR probably indicates reverse causality. Our findings support recording the ambulatory BP during the whole day.

7A.01: INCREASED RISK OF MORTALITY IN OBESE PATIENTS WITH HIGH NOCTURNAL BLOOD PRESSURE VARIABILITY. RESULTS FROM THE ABP-INTERNATIONAL STUDY.

Objective: The association between obesity and all-cause mortality is controversial and may differ according to subjects’ characteristics. Blood pressure variability (BPV) may be increased in obese individuals and thus impair prognosis. The purpose of this study was to evaluate whether the relationship between obesity and mortality is influenced by short-term ambulatory BPV. Design and method: The analysis was performed in 8724 participants (54% men) aged 51u200a±u200a15 years enrolled in 8 prospective studies in Australia, Italy, Japan, and U.S.A. The predictive power of obesity (BMI >=30u200akg/m2) for mortality was evaluated from multivariable Cox models in the subjects stratified by high or low nocturnal BPV (above or below the median). Results: Obese participants (Nu200a=u200a1286) had higher age-and-sex adjusted systolic and diastolic BPV than the non-obese participants (pu200a=u200a0.002/<0.001). Obese subjects with high systolic or diastolic BPV had higher nocturnal heart rate (pu200a=u200a0.01/<0.001) than obese subjects with low BPV and were more frequently diabetic (p<0.001) and heavy alcohol drinkers (pu200a<u200a0.001). During a median follow-up of 6.4 years there were 361 deaths, 4.7% in the obese and 4.0% in the non-obese individuals (Pu200a=u200aNS). However, the risk of mortality among the obese subjects greatly differed according to BPV level. In Cox models including age, sex, mean ambulatory BP, smoking, alcohol use, diabetes, cholesterol, creatinine, and nocturnal heart rate, the obese group with high systolic BPV had a doubled risk of mortality compared to the non-obese group (HR,2.0, 95%CI,1.4–2.9, pu200a<u200a0.001), whereas the risk was not increased in the obese group with low BPV (Pu200a=u200a0.81). Similar results were found for diastolic BPV, with a HR of 1.7 (1.2–2.5, pu200a=u200a0.002) in the high BPV group and no association at all with mortality (pu200a=u200a0.87) in the low BPV group. Inclusion of night-time BP dipping in the regressions did not change the strength of the associations. Conclusions: These data show that high nocturnal BPV greatly increases the risk of mortality related to obesity. High BPV is accompanied by increased heart rate and may reflect the influence of transient BP elevations related to sleep apnea and/or baroreflex dysfunction.

[PP.21.11] NOVEL IDENTIFICATION OF NOCTURNAL HYPERTENSION IMPROVES PREDICTION OF CARDIOVASCULAR DEATH IN A GENERAL POPULATION

Objective: Subjects with nocturnal hypertension often show a non-dipping pattern and have higher risk of cardiovascular mortality. Usually, time-based classification of nocturnal dipping is used to define risks but this is limited by a variability of dipping pattern amongst subjects (e.g. early risers and late risers). Design and method: We applied new method to classify non-dipper using a 6 parameter logistic equation to determine the exact magnitude of the dip irrespective of time, and defined as the range of mean blood pressure (BP) between the upper and lower plateaus. We examined the prognostic value of au200a<u200a10% reduction in range versus the conventional day-night difference of <10% using the data from the Ohasama 15-year outcome study. Results: Among 1522 subjects, 7.4% (nu200a=u200a112) was categorized non-dipper by our new range classification (R-ND), they had higher nocturnal mean BP (89u200a±u200a10 vs 84u200a±u200a9 mmHg, pu200a<u200a0.01), lower day-time mean BP (88u200a±u200a10 vs 93u200a±u200a10 mmHg, pu200a<u200a0.01), similar 24u200ah averaged mean BP (88u200a±u200a10 vs 89u200a±u200a9 mmHg), and were older (67.1u200a±u200a11 vs 61.3u200a±u200a11, pu200a<u200a0.01), compared with dippers. On the other hand, using convention classification by times, 56% (nu200a=u200a847) defined as non-dippers (C-ND), had higher nocturnal mean BP (87u200a±u200a10 vs 81u200a±u200a7 mmHg, pu200a<u200a0.01), lower day-time mean BP (90.1u200a±u200a9.6 vs 95.0u200a±u200a9.2 mmHg, pu200a<u200a0.01), similar 24u200ah averaged mean BP (88.6u200a±u200a9.4 vs 88.8u200a±u200a8.2 mmHg), and were older (63.5u200a±u200a0.4 vs 59.5u200a±u200a0.4, pu200a<u200a0.01), compared with dippers. Both R-ND and C-ND were correlated with the severe adverse cardiovascular events but R-ND predicted all caused deaths (odds ratio 2.72, pu200a<u200a0.01) and cardiovascular deaths (odds ratio 2.33, pu200a<u200a0.01) stronger than C-ND (odds ratio 2.2, pu200a<u200a0.01; odds ratio 1.91, pu200a<u200a0.01, respectively). Conclusions: In conclusion, our novel methods of analysis avoid the uncertainty of choosing the best time to predict nocturnal dipping and better identify the high risk patients with nocturnal hypertension. This leads to an overall improvement in the risk stratification for predicting cardiovascular deaths.

[OP.5C.06] MASKED TACHYCARDIA PREDICTS ADVERSE OUTCOME IN HYPERTENSION. THE ABP-INTERNATIONAL STUDY

Objective: Heart rate (HR) is a well known predictor of cardiovascular events (CVE) and mortality but the relative role of resting office HR and ambulatory HR for predicting these outcomes is not well known. Aim of this study was to investigate the association of white-coat tachycardia and masked tachycardia with CVE and all-cause mortality in hypertensive subjects. Design and method: We performed 24-hour ambulatory blood pressure and HR monitoring in 7602 hypertensive patients (4165 men) aged 52u200a±u200a16 years enrolled in 8 prospective studies in Italy, U.S.A., Japan, and Australia. None of the participants had atrial fibrillation and all were untreated at baseline examination. Office and ambulatory tachycardia were defined according to previously published criteria (J Hypertens 2014;32:2099). Participants were divided into 4 groups: 1) normal office and normal ambulatory HRs (Nu200a=u200a5238), 2) white-coat tachycardia (high office and normal ambulatory HR, Nu200a=u200a998), 3) masked tachycardia (normal office and high ambulatory HR, Nu200a=u200a796), and 4) sustained tachycardia (Nu200a=u200a570). Results: During a median follow-up of 5.0 years the rates of fatal plus nonfatal CVE and of all-cause mortality were 8.4% and 3.7%, respectively. In an age-and-sex-adjusted Cox model, using the normal HRs group as a reference, white-coat tachycardia was not a significant predictor of outcome. In contrast, both masked tachycardia (hazard ratio, 95% CI; 1.36, 1.07–1.72) and sustained tachycardia (1.94, 1.51–2.49) were associated with risk of CVE. In addition, masked tachycardia (hazard ratio, 95% CI; 1.59, 1.12–2.25) but not sustained tachycardia (1.22, 0.75–1.98) was a significant predictor of mortality. These relationships held true in multivariable parsimonious Cox models in which sex, age, diabetes, smoking, creatinine, total cholesterol, and ambulatory blood pressure were included. In these models, masked tachycardia remained an independent predictor of CVE (hazard ratio, 95% CI; 1.34, 1.06–1.71) and all-cause mortality (1.68, 1.17–2.43). Conclusions: This study confirms that measurement of HR adds to the risk stratification for CVE and mortality and shows that an elevated ambulatory HR confers an increased cardiovascular risk to hypertensive patients also in the presence of a normal office HR.