RESUMO
BACKGROUND AND AIM: Cytomegalovirus (CMV) colitis is associated with negative outcomes in inflammatory bowel disease (IBD) and immunosuppressed cohorts and therefore requires timely recognition for appropriate management. We aimed to evaluate the diagnostic tools for CMV colitis and their associations with clinical outcomes. METHODS: A retrospective cohort study of patients in a metropolitan health service with colonic samples analysed for CMV between 2012 and 2022, stratified into IBD and non-IBD groups, was performed. The main outcome measures were the prevalence of positive and negative results for each CMV test, as well as need for colectomy, use of antiviral and hospital length of stay. RESULTS: Five hundred eighty-two biopsies from 418 patients were included; the median age was 36 years (interquartile range, 24-52 years) and 223 (53.3%) were men. Four hundred sixty-one (79.2%) biopsies were from patients with IBD and 121 (20.8%) were from those without IBD. There were similar proportions of positive CMV histology (IBD 5.9% and non-IBD 7.4%) and tissue CMV polymerase chain reaction (PCR) in the two groups (IBD 5.6% and non-IBD 5.0%), but within each group, results were discordant. Positive CMV histology was significantly associated with need for colectomy in the IBD group, while positive tissue CMV PCR was not. Positive CMV histology, and tissue and serum CMV PCR were all significantly associated with antiviral use. Positive serum CMV PCR was significantly associated with colectomy. CONCLUSIONS: Histopathology remains the most predictive tool in assessing CMV colitis, while qualitative tissue CMV PCR was found to have limited utility. Quantitative serum CMV PCR may be useful but requires further evaluation.