Undifferentiated Prehospital Chest Pain: Aetiologies and a Proof-of-Concept Model for Significant Coronary Lesions

Background: Undifferentiated chest pain is one of the most common reasons for emergency medical service (EMS) activation, yet its aetiological spectrum remains poorly characterised in the prehospital setting. Furthermore, no clinical prediction model has been specifically developed to identify patients at risk of significant coronary lesions using only information available before hospital arrival. This study aimed to describe the aetiologies of undifferentiated prehospital chest pain and develop a proof-of-concept clinical prediction model. Methods: We conducted a retrospective, single-centre study including 409 consecutive patients managed by the Orléans Mobile Intensive Care Unit (MICU) for undifferentiated chest pain between January and June 2024. Predictors of significant coronary lesions requiring coronary revascularisation were identified using multivariable logistic regression. Model performance was assessed by discrimination and calibration, and internally validated using 1000 bootstrap resamples. Results: Cardiological aetiologies accounted for 19% of cases, including 53 patients (13%) with significant coronary lesions. Four independent predictors were identified: age (OR 6.7–8.9 according to category), male sex (OR 2.2), typical chest pain (OR 6.6), and a positive family history of cardiovascular disease (OR 3.4). These variables were combined to develop the HATS (History, Age, Typical chest pain, Sex) model. The model demonstrated good discrimination (AUC 0.81), excellent calibration (Hosmer–Lemeshow P=0.88), and satisfactory internal validity after bootstrap validation. Conclusions: This study characterises the aetiological spectrum of undifferentiated prehospital chest pain and proposes the HATS model as a proof-of-concept clinical prediction tool. Prospective multicentre external validation is required before routine clinical implementation.