Systolic blood pressure and short-term mortality in the emergency department and prehospital setting: a hospital-based cohort study

INTRODUCTION: Systolic blood pressure is a widely used tool to assess circulatory function in acutely ill patients. The systolic blood pressure limit where a given patient should be considered hypotensive is the subject of debate and recent studies have advocated higher systolic blood pressure thresholds than the traditional 90 mmHg. The aim of this study was to identify the best performing systolic blood pressure thresholds with regards to predicting 7-day mortality and to evaluate the applicability of these in the emergency department as well as in the prehospital setting. METHODS: A retrospective, hospital-based cohort study was performed at Odense University Hospital that included all adult patients in the emergency department between 1995 and 2011, all patients transported to the emergency department in ambulances in the period 2012 to 2013, and all patients serviced by the physician-staffed mobile emergency care unit (MECU) in Odense between 2007 and 2013. We used the first recorded systolic blood pressure and the main outcome was 7-day mortality. Best performing thresholds were identified with methods based on receiver operating characteristics (ROC) and multivariate regression. The performance of systolic blood pressure thresholds was evaluated with standard summary statistics for diagnostic tests. RESULTS: Seven-day mortality rates varied from 1.8 % (95 % CI (1.7, 1.9)) of 112,727 patients in the emergency department to 2.2 % (95 % CI (2.0, 2.5)) of 15,862 patients in the ambulance and 5.7 % (95 % CI (5.3, 6.2)) of 12,270 patients in the mobile emergency care units. Best performing thresholds ranged from 95 to 119 mmHg in the emergency department, 103 to 120 mmHg in the ambulance, and 101 to 115 mmHg in the MECU but area under the ROC curve indicated poor overall discriminatory performance of SBP thresholds in all cohorts. CONCLUSIONS: Systolic blood pressure alone is not sufficient to identify patients at risk regardless of the defined threshold for hypotension. If, however, a threshold is to be defined, a systolic blood pressure threshold of 100 to 110 mmHg is probably more relevant than the traditional 90 mmHg. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13054-015-0884-y) contains supplementary material, which is available to authorized users.