Validation of the HULL Score clinical prediction rule for unsuspected pulmonary embolism in ambulatory cancer patients

Abstract

Background Clinical prediction rules (CPRs) developed to predict adverse outcomes of suspected pulmonary embolism (PE) and facilitate outpatient management have limitations in discriminating outcomes for ambulatory cancer patients with unsuspected PE (UPE). The HULL Score CPR uses a 5‑point scoring system incorporating performance status and self-reported new or recently evolving symptoms at UPE diagnosis. It stratifies patients into low, intermediate and high risk for proximate mortality. This study aimed to validate the HULL Score CPR in ambulatory cancer patients with UPE. Patients and methods 282 consecutive patients managed under the UPE-acute oncology service in Hull University Teaching Hospitals NHS Trust were included from January 2015 to March 2020. The primary end-point was all-cause mortality, and outcome measures were proximate mortality for the three risk categories of the HULL Score CPR. Results 30-day, 90-day and 180-day mortality rates for the whole cohort were 3.4% (n=7), 21.1% (n=43) and 39.2% (n=80), respectively. The HULL Score CPR stratified patients into low-risk (n=100, 35.5%), intermediate-risk (n=95, 33.7%) and high-risk (n=81, 28.7%) categories. Correlation of the risk categories with 30-day mortality (area under the curve (AUC) 0.717, 95% CI 0.522–0.912), 90-day mortality (AUC 0.772, 95% CI 0.707–0.838), 180-day mortality (AUC 0.751, 95% CI 0.692–0.809) and overall survival (AUC 0.749, 95% CI 0.686–0.811) was consistent with the derivation cohort. Conclusion This study validates the capacity of the HULL Score CPR to stratify proximate mortality risk in ambulatory cancer patients with UPE. The score uses immediately available clinical parameters and is easy to integrate into an acute outpatient oncology setting.