Objectives: Emergency medical services (EMS) personnel commonly use systolic blood pressure (sBP) to triage and treat acutely ill patients. The definition of prehospital hypotension and its associated outcomes are poorly defined. The authors sought to determine the discrimination of prehospital sBP thresholds for 30-day mortality and to compare patient classification by best-performing thresholds to traditional cutoffs.
Methods: In a community-based cohort of adult, nontrauma, noncardiac arrest patients transported by EMS between 2002 and 2006, entries to state hospital discharge data and death certificates were linked. Prehospital sBP thresholds between 40 and 140 mm Hg in derivation (n = 132,624) and validation (n = 22,020) cohorts and their discrimination for 30-day mortality, were examined. Cutoffs were evaluated using the 0/1 distance, Youden index, and adjusted Z-statistics from multivariable logistic regression models.
Results: In the derivation cohort, 1,594 (1.2%) died within 24 hours, 7,404 (6%) were critically ill during hospitalization, and 6,888 (5%) died within 30 days. The area under the receiver operating characteristic (ROC) curve for sBP was 0.60 (95% confidence interval [CI] = 0.59, 0.61) for 30-day mortality and 0.64 (95% CI = 0.62 0.66) for 24-hour mortality. The 0/1 distance, Youden index, and adjusted Z-statistics found best-performing sBP thresholds between 110 and 120 mm Hg. When compared to an sBP ≤ 90 mm Hg, a cutoff of 110 mm Hg would identify 17% (n = 137) more deaths at 30 days, while overtriaging four times as many survivors.
Conclusions: Prehospital sBP is a modest discriminator of clinical outcomes, yet no threshold avoids substantial misclassification of 30-day mortality among noninjured patients.
© 2013 by the Society for Academic Emergency Medicine.