Microbial co-infections and super-infections in critical COVID-19: A Kenyan retrospective cohort analysis

ABSTRACT Objectives To outline the burden, risk factors and outcomes for critical COVID-19 patients with co-infections or superinfections. Methods This was a retrospective descriptive study of adults who were admitted with critical COVID-19 for ≥ 24 hours. Data collected included demographic profiles and other baseline characteristics, laboratory and radiological investigations, medical interventions, and clinical outcomes. Outcomes of interest included presence or absence of co-infections and superinfections and in-hospital mortality. Differences between those with and without co-infections and superinfections were compared for statistical significance. Results We reviewed 321 patient records. Baseline characteristics included a median age (IQR) of 61.4 (51.4-72.9) years, male (71.3%) and African/black predominance (66.4%). Death occurred in 132 (44.1%) patients with significant difference noted between those with added infections (58.2%) compared to those with none (36.6%) (p = 0.002, odds ratio (OR) = 2.41). One patient had co-infection with pulmonary tuberculosis. Approximately two-thirds of patients received broad-spectrum antimicrobial therapy. Conclusion Added infections in critically ill COVID-19 patients were relatively uncommon but where present, were associated with higher mortality. Empiric use of broad spectrum antimicrobials was common and may have led to selection of multidrug resistant organisms. More robust local data on antimicrobial susceptibility patterns may help in appropriate antibiotic selection to improve outcomes without driving up rates of drug resistant pathogens.