The Pulmonary Embolism Rule-out Criteria (PERC) to Rule out Acute Pulmonary Embolism in Patients with Sickle Cell Disease: A Retrospective Cohort Study

Introduction: Patients with Sickle cell disease (SCD) are in a hypercoagulable state and are associated with an increased risk of venous thromboembolism (VTE). They frequently present at the emergency department (ED) due to vaso-occlusive painful episodes (VOE). However, these patients are often suspected of acute pulmonary embolism (PE) due to overlapping symptoms, resulting in over-diagnosis with Computed Tomography Pulmonary Angiography (CTPA). Currently, there is no safe decision algorithm to rule out PE in patients with SCD. D-dimer levels are chronically elevated in SCD, rendering D-dimer based decision algorithms such as the YEARS criteria not useful for ruling out PE in patients with SCD. Pulmonary Embolism Rule-out Criteria (PERC) are developed to rule out acute PE in patients with low risk (<15%), based on clinical gestalt or low Wells score.

Aim: This study aimed to retrospectively evaluate the diagnostic accuracy of the PERC score in patients with SCD who were suspected of PE and had undergone CTPA.

Methods: This was a single-center retrospective cohort study of patients with SCD in whom a CTPA was performed to confirm or refute PE at the ED. All information was collected from electronic patient records. We calculated sensitivity, specificity, efficiency and failure rate of the PERC criteria and a dichotomized Wells score without D-dimer (positive if ≥4), with 95% confidence intervals (CI) calculated with the Wilson score interval. Based on PERC score, PE is considered excluded if none of the following criteria are present: age ≥50, heart rate ≥100, SpO₂ ≤95%, unilateral leg swelling, hemoptysis, recent trauma or surgery, prior VTE, or hormone use. We assessed the association of chest pain not recognized as VOE with PE diagnosis.

Results: A total of 164 CTPAs were performed in 67 patients (median age 34 years [IQR 29-40], 60.4% females). Out of the 164 CTPAs, 119 (72.5%) were performed in patients with HbSS/HbSβ⁰, and 45 (27.5%) in patients with HbSC/HbSβ⁺. Only 12 CTPAs were positive for PE, resulting in a test positivity rate of 7.3%. Of the 12 PE cases, 1 was localized centrally, 5 were segmental, and 6 were sub segmental. PERC criteria showed a sensitivity and specificity of 100% (95% CI: 76%-100%) and 38% (95% CI: 31%-46%) respectively. While 12/12 CTPAs with PE met the PERC criteria, 93 of 150 CTPAs without PE had also met the criteria (at least 1 positive criterion). The efficiency of the PERC score (proportion of patients in whom PE could have been considered excluded based on PERC) was 35% (95% CI: 28%-43%). The failure rate (proportion of patients with a negative PERC who were diagnosed with PE, also known as safety or 1 - negative predictive value) was 0% (95% CI 0%-6%). A Wells score of ≥4 without D-dimer measurement, as currently advised by expert opinion, showed sensitivity and specificity of 41.7% (95% CI: 19%-68%) and 93.3% (95% CI: 88%-96%) respectively, with 5/12 patients with PE meeting the score and 10/150 of patients without PE meeting the score. The efficiency and failure rate of Wells ≥4 were 91% (95% CI: 80%-91%) and 5% (95% CI: 2%-10%) respectively. Chest pain not recognized as VOE by the patient, seemed to be predictive of PE, with a sensitivity and specificity of 83% (95% CI 55%-95%) and 74% respectively (95% CI 66%-80%). Efficiency and failure rate were 70% (95% CI 62%-76%), and 2% respectively (95% CI 1%-6%). The odds ratio was 14.23 (95% CI: 3.0-67.8).

Conclusion: Based on our findings, CTPA could be withheld in 35% of the cases with the PERC score as compared to the previously reported 7% with the YEARS algorithm in SCD. Since the failure rate does not exceed 1.7% (generally accepted safety threshold, based on pulmonary angiography), the PERC score is a potentially safe score to rule out PE in SCD. Chest pain not recognized as VOE had a potentially strong predictive value in diagnosing PE in patients with SCD, and adding it to a potential SCD-specific PERC score in a larger cohort might further improve the performance of the PERC score. Larger prospective studies are needed to determine the diagnostic value of PERC or potentially SCD-specific PERC in ruling out PE in patients with SCD.

Biemond:novartis: Research Funding; novo nordisk: Honoraria; BMS: Consultancy, Research Funding; pfizer: Consultancy, Research Funding; sanofi: Honoraria. Nur:vertex: Speakers Bureau; Novartis: Research Funding.