External Validation of the SCARE Score

Chest pain is a very common reason for resorting to the call center. The etiologies are very varied, ranging from benign pathologies to some that may involve, in the short term, the vital prognosis such as Acute Coronary Syndrome (ACS). ACS is a partial or complete occlusion of a coronary artery that causes potentially irreversible myocardial pain unless prompt treatment is undertaken. ACS represents 120 000 cases per year in France and causes about 18 000 deaths. There is currently no support score for the assessment of chest pain. However, reducing the duration of management of ACS is essential in order to hope to reduce the associated morbidity and mortality. In 2016, SAMU45's team established a predictive ACS score for the assessement of chest pain in SAMU 45 (France) based on the prospective study of 1367 patients. Seven items significantly associated with this risk of ACS could be highlighted: age, sex, smoking, typicality (potentially constrictive chest pain radiating potentially to the shoulders and / or jaw) pain, inaugural character of pain (ie first episode of this type), presence of sweats and the physician's belief to be in the presence of an ACS. These seven variables make up the SCARE score. This had good internal discrimination (AUC at 0.81) and excellent calibration ("p" of Hosmer-Lemeshow at 0.74). This score makes it possible to stratify the risk of ACS, by using epidemiological elements but also the physician's belief, whose Negative Predictive Value (VPN) proved excellent. The objective of this new project is to confirm by an external validation via a multicentric study the robustness of this score and thus be able to consider its use in front of any chest pain regulated in France by a call center.

The main objective is to validate the predictive SCARE score of acute coronary syndrome during the medical assessment of chest pain. The primary endpoint will be SCARE score analysis (pre-established in 2016) with assessment of its calibration (Hosmer Lemeshow) and discrimination (AUC) in a multicenter population of chest pain with a potential diagnosis of Acute Coronary Syndrome established according to the European Society of Cardiology criteria. This is a multi-center study including any patient over 18 years of age calling call center for chest pain over a period of six months. It will exclude post-traumatic chest pain, patients under 18 years old, patients who do not speak French, patients refusing to participate in the study or refusing treatment, patients not affiliated to social security, patients incarcerated in a penitentiary center, patients under tutorship, curatorship or safeguard of justice. The collection of data will be carried out thanks to files filled prospectively by the medical physician for each call for the reason of a chest pain. These cards will list the epidemiological data (age, sex, weight, height) and clinical data of each patient, as well as the decision and the resources committed by the regulating physician (hospital care via SMUR or ambulance, treatment in city medicine). For hospitalized patients, the diagnosis of ACS will be retained if the patient meets the criteria defined by the European Society of Cardiology. For patients managed in ambulatory medicine, a telephone call to the patient will be made at one month to obtain the diagnosis. Then, the SCARE score will be analized in this multicenter population with evaluation of its calibration (Hosmer Lemeshow) and discrimination (AUC). The characteristic performances of the score (sensitivity, specificity, PPV, NPV, positive and negative likelihood ratios)will also be analized.