PROPER

PERC Rule to Exclude Pulmonary Embolism in the Emergency Department

Patient / Population Intervention / Exposition Comparaison Critère de jugement

In1916 ED patients at very low risk of pulmonary embolism,a PERC rule-out strategywas noninferior toconventional testingfor3-month thromboembolic events

1 916

Schéma

Cluster-randomized crossover noninferiority trial in 14 French emergency departments; each centre randomized to the sequence of PERC and control periods; patients with a low gestalt clinical probability of PE enrolled August 2015 to September 2016, follow-up to December 2016; in the PERC period PE was excluded without further testing when all 8 PERC items were negative; NCT02375919.

Critère

Primary: symptomatic thromboembolic event during 3-month follow-up that was not diagnosed at the index visit, with a prespecified noninferiority margin of 1.5% absolute. Secondary: CTPA rate, ED length of stay, hospital admission.

Résultat962 patients PERC vs 954 control (mean age 44 years, 51% women); 1749 completed the trial. Primary endpoint 1 event (0.1%) in the PERC group vs 0 (0%) in the control group (difference 0.1%, 95% CI −∞ to 0.8%), meeting the 1.5% noninferiority margin. PE at index presentation 26 (2.7%) control vs 14 (1.5%) PERC (difference 1.3%, 95% CI −0.1% to 2.7%; P = .052). CTPA 13% vs 23% (difference −10%, 95% CI −13% to −6%; P < .001), ED stay shorter by a mean 36 minutes and admission lower by 3.3%. Subsegmental PE was found in 5 control vs 1 PERC patients; the single false-negative was a young man discharged PERC-negative who returned the next day and had subsegmental defects on V/Q scan, anticoagulated for 6 months.

Freund Y, Cachanado M, Aubry A, et al. JAMA. 2018;319(6):559-566. 10.1001/jama.2017.21904