Ethnicity-specific blood pressure thresholds based on cardiovascular and renal complications: a prospective study in the UK Biobank.

BACKGROUND: The appropriateness of hypertension thresholds for triggering action to prevent cardiovascular and renal complications among non-White populations in the UK is subject to question. Our objective was to establish ethnicity-specific systolic blood pressure (SBP) cutoffs for ethnic minority populations and assess the efficacy of these ethnicity-specific cutoffs in predicting adverse outcomes. METHODS: We analyzed data from UK Biobank, which included 444,418 participants from White, South Asian, Black Caribbean, and Black African populations with no history of cardiorenal complications. We fitted Poisson regression models with continuous SBP and ethnic groups, using Whites as the referent category, for the composite outcome of atherosclerotic cardiovascular disease, heart failure, and chronic kidney disease. We determined ethnicity-specific thresholds equivalent to the risks observed in Whites at SBP levels of 120, 130, and 140 mm Hg. We adjusted models for clinical characteristics, sociodemographic factors, and behavioral factors. The performance of ethnicity-specific thresholds for predicting adverse outcomes and associated population-attributable fraction (PAF) was assessed in ethnic minority groups. RESULTS: After a median follow-up of 12.5 years (interquartile range, 11.7-13.2), 32,662 (7.4%) participants had incident composite outcomes. At any given SBP, the predicted incidence rate of the composite outcome was the highest for South Asians, followed by White, Black Caribbean, and Black African. For an equivalent risk of outcomes observed in the White population at an SBP level of 140 mm Hg, the SBP threshold was lower for South Asians (123 mm Hg) and higher for Black Caribbean (156 mm Hg) and Black African (165 mm Hg). Furthermore, hypertension defined by ethnicity-specific thresholds was a stronger predictor and resulted in a larger PAF for composite outcomes in South Asians (21.5% [95% CI, 2.4,36.9] vs. 11.3% [95% CI, 2.6,19.1]) and Black Africans (7.1% [95% CI, 0.2,14.0] vs. 5.7 [95% CI, -16.2,23.5]) compared to hypertension defined by guideline-recommended thresholds. CONCLUSIONS: Guideline-recommended blood pressure thresholds may overestimate risks for the Black population and underestimate risks for South Asians. Using ethnicity-specific SBP thresholds may improve risk estimation and optimize hypertension management toward the goal of eliminating ethnic disparities in cardiorenal complications.

Hypertension in stroke survivors and associations with national premature stroke mortality: data for 2·5 million participants from multinational screening campaigns.

BACKGROUND: Blood pressure control has a pivotal role in reducing the incidence and recurrence of stroke. May Measurement Month (MMM), which was initiated in 2017 by the International Society of Hypertension, is the largest global blood pressure screening campaign. We aim to compare MMM participants with and without a previous history of stroke and to investigate associations between national-level estimates of blood pressure management from MMM and premature stroke mortality. METHODS: In this annual, global, cross-sectional survey, more than 2·5 million volunteers (≥18 years) from 92 countries were screened in May, 2017, and May, 2018. Three seated blood pressure readings and demographic, lifestyle, and cardiovascular disease data were collected. Associations between risk factors and stroke history were analysed with mixed-effects logistic regression, and associations between national-level estimates of blood pressure management and premature stroke mortality based on Global Burden of Disease data were investigated with linear regression. FINDINGS: 2 222 399 (88·4%) of 2 515 365 participants had recorded data on a history of stroke, of whom 62 639 (2·8%) reported a previous stroke. Participants with a history of stroke had higher rates of hypertension (77·0% vs 32·9%, p<0·0001) and of treated (90·2% vs 57·0%, p<0·0001) and controlled (55·9% vs 32·4%, p<0·0001) hypertension than those without a history of stroke. A third of participants with a history of stroke had either untreated hypertension or treated but uncontrolled hypertension (blood pressure ≥140/90 mm Hg). Strong positive associations were found between national premature stroke mortality and mean systolic blood pressure (84·3 [95% CI 38·8 to 129·9] years of life lost [YLL] per 100 000 people per mm Hg increase) and the percentage of participants with raised blood pressure (49·1 [22·6 to 75·6] YLL per 100 000 people per 1% increase). Strong negative associations were found between national premature stroke mortality and the percentage of participants with hypertension on treatment (-21·0 [-33·0 to -8·9] YLL per 100 000 people per 1% increase) and with controlled blood pressure (-31·6 [-43·8 to -19·4] YLL per 100 000 people per 1% increase). INTERPRETATION: Blood pressure control remains suboptimal worldwide among people with a history of stroke. National estimates of blood pressure management reflect national premature stroke mortality sufficiently to prompt policy makers to promote blood pressure screening and management. FUNDING: International Society of Hypertension and Servier Pharmaceuticals.