Low systolic blood pressure for predicting all-cause mortality in patients hospitalised with heart failure: a systematic review and meta-analysis

Abstract

Heart failure (HF) is still a major global health concern, contributing to high mortality and morbidity. Despite the advances in medical treatment, the prognosis of acute HF patients remains poor. Therefore, early risk stratification is of greater importance in the management of patients hospitalised with HF. Higher systolic blood pressure (SBP) may confer a better prognosis in these patients. A well-designed meta-analysis has summarised that each 10mmHg increase in SBP was associated with a 13% reduction in death risk among chronic HF patients. Increasing attention has focused on the prognostic role of SBP on acute HF. However, no previous meta-analysis has summarised the prognostic value of SBP in acute HF settings. This meta-analysis aimed to investigate quantitatively the association between low SBP on admission or discharge and all-cause mortality risk in hospitalised HF patients. Two reviewers independently searched the PubMed and Embase databases to identify relevant studies published from their inception to March 2018. The search terms used were: systolic blood pressure AND heart failure AND mortality OR death AND admission OR hospitalisation OR hospitalised OR discharge. The inclusion criteria were: (a) observational studies including post hoc analyses of randomised clinical trials; (b) enrollment of hospitalised HF patients; (c) low SBP on admission or discharge as exposure; (d) all-cause mortality as outcome measures; and (e) providing multivariable adjusted risk estimate for the lowest versus the reference higher SBP category and per unit SBP decrease. Studies with stable HF, aortic/ mitral valve disease population were excluded. Moreover, studies with unclear timing measurement of SBP or inhospital death as an outcome measure were also excluded. The Newcastle–Ottawa scale was applied to examine the methodological quality of the included studies. Data analyses were performed using STATA 12.0. For the categorical analysis, we pooled risk estimates for the lowest versus the reference higher SBP category. For analysing SBP as continuous data, we recalculated the risk estimate by per 10mmHg decrease of SBP using the following formula: HR101⁄4 exp (ln (HR1) 10). Fifteen studies involving 133,549 HF patients were ultimately included in this meta-analysis (Table 1). Meta-analysis from six studies showed that the lowest SBP on admission increased the risk of all-cause mortality (hazard ratio (HR) 2.22; 95% confidence interval (CI) 1.72–2.86) when compared with the reference higher SBP category. There was no evidence of publication bias according to the Begg’s test (P1⁄4 0.210) or Egger’s test (P1⁄4 0.119). However, the lowest SBP on discharge was not associated with an increased all-cause mortality (HR 1.35; 95% CI 0.66–2.73) in pooling three studies. Meta-analysis from seven studies indicated that the pooled HR of all-cause mortality was 1.10 (95% CI 1.06–1.15) per 10mmHg admission SBP decrease in a random effect model. Egger’s test (P1⁄4 0.061) but not Begg’s test (P1⁄4 0.266) revealed the presence of publication bias. Subgroup analyses indicated that both categorical and continuous low admission SBP were consistently associated with a higher risk of all-cause mortality in the study design, follow-up period, study quality or whether with reduced left ventricular ejection fraction subgroups. This meta-analysis shows that low SBP on admission is independently associated with an increased risk of death among patients hospitalised with HF. Patients with the lowest admission SBP increases 2.22-fold risk