Postoperative alarm signs in the rapid response system and hospital mortality after non-cardiac surgery

BACKGROUND A variety of rapid response systems (RRSs) based on the systematic assessment of vital signs and laboratory tests have been developed to reduce hospital mortality through the early detection of alarm signs, while deterioration may still be reversible. This study aimed to determine the association between alarm signs and postoperative hospital mortality during postoperative days (POD) 0-3 in patients undergoing non-cardiac surgery. METHODS This retrospective observational study used data from the registry of a single tertiary academic hospital. The study population included patients who were ≥18 years old, admitted between January 1, 2013 and June 30, 2018 for non-cardiac surgery, and subsequently transferred to the general ward. RESULTS A total of 116,329 patients were included in the analysis. Among them, 10,099 patients (8.7%) showed positive alarm criteria and triggered the RRS in the postoperative ward during POD 0-3. In the multivariate logistic regression model, PaO2 <55 mmHg, SpO2 <90%, and total CO2 <15 mmol/L were associated with a 3.57-, 3.46-, and 12.53-fold increase in postoperative hospital mortality, respectively. Moreover, when compared to the no alarm signs group, patients with 1, 2, 3, and ≥4 alarm signs showed a 2.79-, 2.76-, 6.54-, and 20.02-fold increase in hospital mortality, respectively. CONCLUSION Increased postoperative hospital mortality was found to be associated with alarm signs detected by the RRS during POD 0-3. The postoperative alarm signs detected by the RRS may therefore be useful in determining high-risk patients who require medical interventions in the surgical ward.