Prehospital Respiratory Early Warning Score for airway management in‐ambulance: A score comparison

• Published in: European journal of clinical investigation 2023-01, Vol.53 (1), p.e13875-n/a
• Main Authors: Martín‐Rodríguez, Francisco, López‐Izquierdo, Raúl, Sanz‐García, Ancor, Ortega, Guillermo J., Pozo Vegas, Carlos, Delgado‐Benito, Juan F., Castro Villamor, Miguel A., Soriano, Joan B.
• Format: Article
• Language: English
• Subjects: Adult; Airway Management; Ambulances; Blood pressure; Calibration; Decision making; Early Warning Score; Emergency medical care; Emergency medical services; Epidemiology; Heart rate; Humans; invasive respiratory support; non‐invasive respiratory support; Oximetry; Oxygen; Patients; prehospital; Prospective Studies; Pulse oximetry; Regression models; Respiration; Respiratory Rate; Retrospective Studies; risk scores; Saturation
• ISSN: 0014-2972; 1365-2362
• DOI: 10.1111/eci.13875

Background Prehospital Respiratory Early Warning Scores to estimate the requirement for advanced respiratory support is needed. To develop a prehospital Respiratory Early Warning Score to estimate the requirement for advanced respiratory support. Methods Multicentre, prospective, emergency medical services (EMS)‐delivered, longitudinal cohort derivationvalidation study carried out in 59 ambulances and five hospitals across five Spanish provinces. Adults with acute diseases evaluated, supported and discharged to the Emergency Department with high priority were eligible. The primary outcome was the need for invasive or non‐invasive respiratory support (NIRS or IRS) in the prehospital scope at the first contact with the patient. The measures included the following: epidemiological endpoints, prehospital vital signs (respiratory rate, pulse oximetry saturation, fraction of inspired oxygen, systolic and diastolic mean blood pressure, heart rate, tympanic temperature and consciousness level by the GCS). Results Between 26 Oct 2018 and 26 Oct 2021, we enrolled 5793 cases. For NIRS prediction, the final model of the logistic regression included respiratory rate and pulse oximetry saturation/fraction of inspired oxygen ratio. For the IRS case, the motor response from the Glasgow Coma Scale was also included. The REWS showed an AUC of 0.938 (95% CI: 0.918–0.958), a calibration‐in‐large of 0.026 and a higher net benefit as compared with the other scores. Conclusions Our results showed that REWS is a remarkably aid for the decision‐making process in the management of advanced respiratory support in prehospital care. Including this score in the prehospital scenario could improve patients' care and optimise the resources' management.