
Background. South Africa (SA) presents a unique epidemiological context characterised by a high prevalence of both HIV infection and rising non-communicable diseases (NCDs), which may interact to adversely affect COVID-19 outcomes. Objectives. To investigate the determinants of in-hospital mortality among patients with COVID-19, with particular emphasis on HIV status and coexisting comorbid conditions. Methods. A retrospective cohort study was conducted among adult patients with COVID-19 admitted to a tertiary hospital in KwaZulu-Natal Province, SA, between January 2021 and April 2022. Demographic and clinical data were extracted from patients’ medical records. Independent predictors of in-hospital mortality were identified using a multivariable Cox proportional hazards regression model. Results. Among the 753 patients included in the cohort, 22.9% were HIV positive, of whom 58.2% had an unsuppressed HIV viral load. In addition, 48.3% of participants presented with one or more comorbid conditions: 21.4% had a single comorbidity, 16.7% had two, and 10.2% had three or more. Most of the cohort (90.6%) were not vaccinated against COVID-19. The overall in-hospital COVID-19 case fatality rate was 21.0%. Mortality was significantly higher among patients who required advanced respiratory support (adjusted hazard ratio (aHR) 5.51, 95% confidence interval (CI) 3.40 - 8.96, p=0.001) and those who developed acute kidney injury (AKI) (aHR 1.62, 95% CI 1.09 - 2.42, p=0.017) than among those who did not. Age <45 years was independently associated with a reduced risk of death (aHR 0.55, 95% CI 0.34 - 0.88, p=0.013). The presence of any comorbidity significantly increased the risk of mortality (aHR 1.60, 95% CI 1.12 - 2.30, p=0.011). Neither HIV status nor HIV virological control was independently associated with mortality (p=0.819). Conclusion. In this SA cohort, advancing age, the need for supplemental oxygen or more advanced respiratory support, AKI and comorbidities burden emerged as key determinants of COVID-19-related mortality. These findings underscore the importance of addressing NCDs as part of an integrated care strategy for reducing COVID-19-related mortality in high-risk populations.
Background. Sepsis is a leading cause of intensive care unit (ICU) morbidity and mortality, particularly in low- and middle-income countries. Data from mixed adult and paediatric ICUs in such settings remain limited. Objectives. To determine the prevalence and timing of ICU sepsis and to describe infection sites, microbiological profile, HIV status, length of stay and mortality. Methods. A retrospective analysis of ICU admissions at a tertiary academic hospital in South Africa between 1 January and 31 December 2020 was done. Sepsis was defined using Sepsis-3 criteria. Organ dysfunction was defined using Sequential Organ Failure Assessment (SOFA) and paediatric SOFA scores. Early sepsis was defined as occurring at or shortly after admission (days 0 - 4). Results. The prevalence of sepsis was 56.3% (n=279/496), 54.1% in adults and 63.0% in children. Sepsis was present on the day of ICU admission in 86.8% of adults and 85.3% of children, and 96.1% and 97.3%, respectively, were identified within the first 4 ICU days. The most common infection site was the respiratory tract in both adults (43.6%) and children (48.0%). The odds of isolating a multidrug-resistant or extensively drug-resistant organism were 2.4-fold higher during the final episode of sepsis than during the initial episode (95% confidence interval 1.42 - 4.30). HIV prevalence was 24.0% in adults and 9.3% in children. Overall ICU mortality was 13.6%. In adults, ICU length of stay and the Simplified Acute Physiology Score II independently predicted mortality. Conclusion. We observed a high burden of sepsis, with increasing antimicrobial resistance after 96 hours of admission. Among adult patients, severity of illness and length of ICU stay predicted mortality. Paediatric severity-of-illness scoring systems appeared poorly calibrated in this high-infection setting.
Respiratory syncytial virus (RSV) is the leading cause of lower respiratory tract infection and hospitalisation in young infants. The maternal RSV prefusion F protein vaccine administered to pregnant women, and the long-acting monoclonal antibody administered to infants, have been approved by regulatory authorities for prevention of RSV in infancy in South Africa (SA). This statement, supported by national pulmonology, paediatrics, obstetrics and gynaecology and infectious diseases societies, details the clinical efficacy, safety, dosing and cost-effectiveness of RSV prevention strategies relevant to SA. It also provides a practical clinical algorithm and implementation recommendations for practitioners, health system considerations, ethical counselling guidance, and a framework for surveillance and safety monitoring to optimise programme impact and equity.