A multivariable model of the association between postoperative complications and 30-day in-hospital mortality in the African surgical outcome study (ASOS) cohort

Simons, Caroline

A multivariable model of the association between postoperative complications and 30-day in-hospital mortality in the African surgical outcome study (ASOS) cohort

Thesis / Dissertation

2026

Publisher

University of Cape Town

Department

Department of Public Health and Family Medicine

Faculty

Faculty of Health Sciences

Abstract

BACKGROUND: The perioperative mortality rate in Africa is twice the global average. Postoperative complications independently associated with postoperative mortality has not been explored in this region. The aim of this study was to determine which severe postoperative complications are independently associated with postoperative mortality in Africa. METHODS: Data from an African prospective observational study of perioperative adult patients conducted in 2016 was analysed. We used a logistic mixed-effects model to assess the association between severe postoperative complications and 30-day postoperative mortality, adjusting for known patient and surgical factors. This model was used to calculate the attributable fraction (AF) of death. RESULTS: We included 10 748 patients of whom 222 (2.07%) died in hospital within the 30day postoperative period. The severe postoperative complications with the highest adjusted odds ratios (OR) for mortality were pulmonary embolism (OR 349, 95% confidence interval (CI) 61.6 – 1980, p < 0.001), pulmonary oedema (OR 44.8, 95% CI 5.34 – 375, p < 0.001) and bloodstream infections (OR 33.5, 95% CI 15.2 – 73.7, p < 0.001). The severe complications with the highest AF to 30-day postoperative mortality were bloodstream infections (AF 51.33%, 95% CI 40.1 – 60.51), acute kidney injury or AKI (AF 27.64%, 95% CI 17.98 – 36.22), and pneumonia (AF 26.55%, 95% CI 16.98 – 35.01). CONCLUSIONS: The complications with the highest AF for postoperative mortality in Africa identify the complications with the greatest contribution to the postoperative mortality which can guide priorities for future interventional strategies to decrease postoperative mortality in Africa.