Diagnostic accuracy of pulmonary embolism rule-out criteria: A systematic review and meta-analysis

Abstract

Study objective: To perform a systematic review and meta-analysis to define the diagnostic performance of pulmonary embolism rule-out criteria (PERC) in deferring the need for D-dimer testing to rule out pulmonary embolism in the emergency department (ED). Methods: We searched EMBASE, MEDLINE, Scopus, Web of Knowledge, and all the evidence-based medicine reviews that included the Cochrane Database of Systematic Reviews through August 14, 2011, and hand searched references in potentially eligible articles and conference proceedings of major emergency medicine organizations for the previous 2 years. We selected studies that reported diagnostic performance of PERC, reported original research, and were conducted in the ED, with no language restrictions. Two investigators independently identified eligible studies and extracted data. We used contingency tables to calculate sensitivity, specificity, and likelihood ratios. Results: We found 12 qualifying cohorts (studying 13,885 patients with 1,391 pulmonary embolism diagnoses), 10 prospective and 2 retrospective, from 6 countries. Pooled sensitivity, specificity, positive likelihood ratios, and negative likelihood ratios for 10 included studies were 0.97 (95% confidence interval [CI] 0.96 to 0.98), 0.23 (95% CI 0.22 to 0.24), 1.24 (95% CI 1.18 to 1.30), and 0.17 (95% CI 0.13 to 0.23), respectively. Significant heterogeneity was observed in specificity (I2=97.2%) and positive likelihood ratio (I2=84.2%). Conclusion: The existing literature suggests consistently high sensitivity and low but acceptable specificity of the PERC to rule out pulmonary embolism in patients with low pretest probability. Copyright © 2012 by the American College of Emergency Physicians.