DOI: 10.47582/jompac.2017430 ISSN: 2717-7505

Admission predictors and clinical patterns of continued intensive care requirement at 48 hours in postoperative patients: a retrospective cohort study

Mustafa Ay, Serap Avcı Ay, Dilara Tüfek Öztan, Şule Asri
Show PDF Cite
Aims: We investigated whether clinical and bedside physiological variables available at intensive care unit (ICU) admission discriminate postoperative patients who continue to require ICU-level care at 48 hours, and characterized the reasons for this need. Methods: Consecutive adults admitted to the ICU directly after operating-room surgery between January 1 and July 1, 2026 were retrospectively reviewed. The primary outcome was failure to meet at least one of five prespecified clinical domains at 48 hours, or death within 48 hours; patients remaining solely for logistical reasons were not classified as having the outcome. The primary model comprised American Society of Anesthesiologists (ASA) physical status III–V, emergency surgery, intubation at ICU admission and shock index. Results: Of 143 patients, the primary outcome occurred in 39 (27.3%). The primary model yielded an area under the receiver operating characteristic curve (AUC) of 0.820 (95% confidence interval, 0.748–0.891; optimism-corrected, 0.799). Emergency surgery, intubation at ICU admission and a higher shock index were independently associated with the outcome, and adding bedside physiology and level of support improved discrimination (ΔAUC, 0.050; p=0.017). Of the 37 patients with an ongoing requirement, 6 (16.2%) met all four organ-domain criteria and 23 (62.2%) had no early adverse event.Conclusion: A four-variable bedside admission model showed good discrimination for continued ICU-level care requirement at 48 hours or death within 48 hours. This need frequently reflected unresolved acute problems requiring intensive monitoring rather than new organ support. External validation is required.

More from our Archive