Continuous physiological monitoring for the detection of postoperative deterioration: a protocol for a multistage, multicentre, international, prospective cohort study

Résumé fourni par la source

INTRODUCTION: Intermittent physiological monitoring and early warning scores (EWS) are limited in their ability to detect deteriorating patients in a timely manner. Wearable physiological sensors allow continuous remote monitoring and may be more timely and accurate in the identification of those at risk, compared with manual collection. This study aims to determine if wearable physiological sensors can be used for the early detection of postoperative deterioration, while being acceptable to patients and healthcare staff. METHODS AND ANALYSIS: This is a prospective observational cohort study that will recruit adults undergoing major surgery in Benin, India, Ghana, Guatemala, Mexico, Nigeria, Rwanda and the UK. Participants will wear wearable physiological chest and limb sensors before, during and after surgery for up to 10 days or until discharge. In this 'shadow-mode' study, continuous physiological observations collected using the devices will not be made available to clinical teams. No changes in participant care will result. Standard of care clinical data will be collected contemporaneously. Continuous sensor data will be used to design algorithms to predict deterioration and specific complications in this population. Usability and feasibility testing, through focus groups, interviews and questionnaires, will be undertaken with healthcare professionals and people undergoing surgery. ETHICS AND DISSEMINATION: Our stakeholder panel are directly involved in all aspects of this study, which will be conducted in accordance with the principles of the International Conference on Harmonisation Tripartite Guideline for Good Clinical Practice (ICH GCP) in addition to the principles of the ethics committee(s)/Institutional Review Boards (IRBs) who have reviewed and approved this study. Artificial intelligence (AI) prediction models will be reported in accordance with the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis+Artificial Intelligence (TRIPOD+AI) and Developmental and Exploratory Clinical Investigations of DEcision support systems driven by Artificial Intelligence (DECIDE-AI) reporting guidelines frameworks. TRIAL REGISTRATION NUMBER: NCT06565559.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.