Background The pulmonary embolism rule-out criteria (PERC) rule is an eight-factor decision rule to support the decision not to order a diagnostic test when the gestalt-based clinical suspicion on pulmonary embolism (PE) is low.

Methods In a retrospective cohort study, we determined the accuracy of a negative PERC (0) in patients with a low Wells score (<2) to rule-out PE, and compared this to the accuracy of the default algorithm used in our hospital (a low Wells score in combination with a negative D-dimer).

Results During the study period, 377 patients with a Wells score <2 were included. CT pulmonary angiography (CTPA) was performed in 86 patients, and V/Q scintigraphy in one patient. PE was diagnosed in 18 patients. 78 patients (21%) had a negative PERC score. When further diagnostic studies would have been omitted in these patients, two (subsegmental) PEs would have been missed, resulting in a sensitivity of 89% (64%–98%) and a negative likelihood ratio (LR−) of 0.52 (0.14–1.97). The default algorithm missed one (subsegmental) PE, resulting in a sensitivity of 95% (71%–99%) and an LR− of 0.25 (0.04–1.73).

Conclusions The combination of a Wells score <2 and a PERC rule of 0 had a suboptimal sensitivity for excluding PE in our sample of patients presenting in the ED. Further studies are warranted to test this algorithm in larger populations.