Utility of cytomegalovirus (CMV) qualitative polymerase chain reaction from gastrointestinal biopsies in diagnosis of CMV gastrointestinal disease: A 10-year retrospective study

Cytomegalovirus (CMV) gastrointestinal (GI) disease is traditionally diagnosed via histopathology of endoscopic tissue biopsies. The utility of tissue PCR for predicting CMV GI disease remains unclear. We conducted a 10-year retrospective single-center study comparing tissue PCR performance to histopathology for the diagnosis of CMV GI disease. Adult patients with GI tissue biopsy between February 2014 and January 2024 were included. Qualitative tissue PCR was compared to histopathology (gold standard) to determine sensitivity, specificity, and receiver operating characteristic (ROC) curves. Log-binomial regression models were performed to evaluate potential predictors of CMV GI disease. Our study comprised 533 patients with 635 endoscopies. Underlying diagnoses included solid organ transplant, hematopoietic stem cell transplantation (HSCT), inflammatory bowel disease, and others. Histopathologic evidence of CMV disease was found in 41/635 biopsies and in 37/533 patients. Compared to histopathology, tissue PCR sensitivity was 100% (95% CI 91.4–100%), and specificity was 71.7% (67.9–75.3%). Area under ROC curve (AUC) was 0.86 (0.84–0.88). Exclusion of specimens with cycle threshold >32 increased specificity to 82.7% (79.4-85.6%) but at the cost of decreased sensitivity (87.8%, 73.8–95.9%). Multivariable analyses demonstrated that plasma CMV DNAemia >1000 IU/mL increased risk of GI disease (risk ratio 10.9, 95% CI 5.32–22.49) while HSCT was negatively associated (risk ratio 0.18, 95% 0.05–0.78). CMV tissue PCR correctly identified CMV GI disease in 100% of histology-proven cases. Specificity was poor, likely reflecting detection of viral shedding. Overall, our study suggests that CMV tissue PCR should be reserved to rule out CMV GI disease in high pre-test probability settings (e.g. patients with known risk factors and compatible symptoms). • The utility of tissue PCR for predicting CMV gastrointestinal (GI) disease remains unclear. • Qualitative CMV tissue PCR was evaluated compared to the gold standard of histopathology for endoscopic tissue specimens from 533 patients. • Compared to histopathology, CMV tissue PCR has excellent sensitivity (100%) but poor specificity below 75% and very low positive predictive value • CMV GI disease was strongly associated with peripheral CMV viral loads > 1000 IU/mL • We suggest that tissue PCR should only be used clinically to rule out CMV GI disease in high pre-test probability settings (e.g. patients with known risk factors and compatible symptoms)

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.