# Clinical Characteristics and Predictors of Antiviral Treatment Duration in Hospitalized Patients with Ulcerative Colitis-Associated Cytomegalovirus Colitis, Including Biologic Therapy

## TL;DR

This study identifies factors affecting antiviral treatment duration in ulcerative colitis patients with CMV colitis, including biologic therapy use and viral load.

## Contribution

The study introduces a tissue CMV PCR cut-off to predict prolonged antiviral therapy in ulcerative colitis patients with CMV colitis.

## Key findings

- A tissue CMV PCR cut-off of 162,000 IU/mg predicts prolonged antiviral therapy.
- Biologic therapy patients were younger and had higher C-reactive protein levels.
- Tissue CMV PCR viral load weakly correlated with lower serum albumin levels.

## Abstract

Cytomegalovirus (CMV) colitis is a significant entity in hospitalized patients with ulcerative colitis, particularly during immunosuppressive therapy. The factors associated with antiviral treatment duration remain incompletely defined. This retrospective cohort study included hospitalized adult patients with ulcerative colitis and immunohistochemically confirmed CMV colitis. Baseline demographic, clinical, endoscopic, and laboratory characteristics were evaluated for the cohort and stratified by antiviral treatment duration of ≤14 days and >14 days. Correlation analyses were performed between tissue CMV polymerase chain reaction (PCR) viral load and laboratory parameters. Receiver operating characteristic analysis identified a tissue CMV PCR cut-off associated with prolonged antiviral therapy. The study included 52 patients (median age, 41.5 years; 65.4% male). Fourteen patients received biologic therapy and were younger and had higher C-reactive protein levels than those who did not receive biologics. Tissue CMV PCR viral load was higher in patients who received antiviral therapy for >14 days. The analysis identified a tissue CMV PCR cut-off value of 162,000 IU/mg, with an area under the curve of 0.69, sensitivity of 70.4%, and specificity of 76.0%. Tissue CMV PCR viral load showed a weak negative correlation with serum albumin levels (Spearman ’s r = −0.34, p < 0.05). Tissue CMV PCR viral load is associated with antiviral treatment duration and may help identify patients with ulcerative colitis–associated CMV colitis who require prolonged therapy.