# Analytical performance of species-targeted quantitative PCR for intra-abdominal candidiasis in critically ill patients: a proof-of-concept post hoc analysis from the pBDG2 multicenter study

## TL;DR

A new PCR test for detecting Candida species in critically ill patients with suspected intra-abdominal candidiasis shows high sensitivity and potential for rapid diagnosis.

## Contribution

The study evaluates a species-targeted quantitative PCR assay for Candida detection in peritoneal fluid, demonstrating its diagnostic potential in critically ill patients.

## Key findings

- The PCR assay achieved 100% sensitivity and 60% specificity compared to fungal culture.
- ROC-derived cycle threshold cutoffs improved the interpretation of PCR results for C. albicans and C. tropicalis.
- Discordant PCR-positive/culture-negative results were linked to high Ct values, possibly indicating low fungal burden or non-viable DNA.

## Abstract

Intra-abdominal candidiasis (IAC) is a severe form of invasive candidiasis that affects critically ill patients and is associated with a high mortality rate and persistent diagnostic delays. Conventional fungal culture from peritoneal fluid remains slow and has insufficient sensitivity. Molecular assays promote early diagnosis, but data from critically ill populations remain scarce.

We evaluated the diagnostic performance of the OLM CandID Real-Time PCR assay directly applied to peritoneal fluid samples obtained from a previously established multicenter cohort of critically ill patients with suspected or confirmed IAC.

OLM CandID Real-Time PCR is a quantitative PCR assay for detecting C. albicans, C. glabrata, C. parapsilosis, C. tropicalis, C. krusei, and C. dubliniensis. Binary PCR results were compared with those of fungal culture as the reference standard. Secondary analyses analyzed quantitative PCR values to better understand culture-negative/PCR-positive discordances using ROC-derived cycle threshold (Ct) cutoffs.

Fifty-seven patients, mainly with postoperative peritonitis, were included. Among the culture-positive samples, C. albicans was predominant. Using binary interpretation, the PCR achieved 100% sensitivity (95% CI 100–100) and 60% specificity (95% CI 38–81); the positive and negative predictive values were 82% (95% CI 71–93) and 100% (95% CI 100–100), respectively. In the secondary Ct-based analysis, ROC-derived threshold determination was feasible for C. albicans (n = 35) and C. tropicalis (n = 12). For C. albicans and C. tropicalis, the ROC AUC values were 0.92 (95% CI 0.75-1) and 0.90 (95% CI 0.66-1), respectively, and applying the optimal Ct cutoffs (30.75 and 34.63 respectively) yielded 100% specificity and 100% positive predictive value for both species.

OLM CandID Real-Time PCR applied directly to peritoneal fluid demonstrated strong analytical performance for the detection of Candida in critically ill patients with suspected IAC. The assay showed high concordance with culture results, supporting its feasibility for rapid microbiological assessment. Discordant PCR-positive/culture-negative results were mainly associated with high Ct values, likely reflecting low fungal burden, detection of non-viable DNA, or analytical over-detection. These findings are exploratory and highlight the need for prospective studies to determine the clinical relevance of quantitative PCR results and their role in guiding antifungal management.

The study was registered with ClinicalTrials.gov (ID number 07005258, first registered on March 6, 2025).

The online version contains supplementary material available at 10.1186/s13054-026-05882-5.