Comparative epidemiology of diarrhoeal organisms in high-risk adults using multiplex PCR gastrointestinal panel testing: a multicentre observational study.

Abstract

Objectives: Infectious diarrhoea remains a significant cause of morbidity and mortality in immunocompromised and high-risk populations. However, pathogen-specific epidemiology and diagnostic utility of molecular gastrointestinal panels (GIPs) in these groups remain poorly defined. Our objective was to describe diarrhoeal pathogen distribution across 10 high-risk conditions and evaluate the diagnostic yield of GIPs in these populations to inform the development of improved diagnostic algorithms.

Methods: We conducted a retrospective observational study analysing all adults (age ≥18 years) tested with a BioFire FilmArray GIP at 12 U.S. hospitals from 2019-2024. We reported observed organism detection rates overall, risk status, and by each of 10 high-risk conditions identified by International Classification of Diseases 10 (ICD-10) codes. Mixed-effects multivariable logistic regression estimated the independent effect of each high-risk condition on organism detection, controlling for sex, location, setting, year, season, and 10 high-risk conditions.

Results: Among 16 570 patients, (median age 66 years; 65% [10 752/16 570] with ≥1 high-risk condition), pathogens were detected in 22% [3591/16 570]. Enteropathogenic Escherichia coli (EPEC) and norovirus were the most common organisms detected (6.4% [1059/16 570] and 6.1% [1006/16 570], respectively). The third most common organism varied based on high-risk comorbidity, typically Salmonella or Campylobacter. High-risk patients had significantly lower odds of detecting any pathogen compared with low-risk patients (19% [1996/10 752] vs. 27% [1595/5818]) (OR 0.60; 95% CI, 0.56-0.65). Patients with HIV had higher bacterial (adjusted OR 1.61; 95% CI, 1.05-2.38) and parasitic detection (2.94; 95% CI, 1.11-6.46), while transplant recipients had higher viral detection (adjusted OR 1.50; 95% CI, 1.08-2.07). Codetections most commonly involved EPEC with other organisms.

Discussion: Most GIP testing in high-risk patients did not identify a causative pathogen. When the GIP did detect an organism, EPEC and norovirus were the most common, with distinct pathogen profiles by comorbidity. Our findings support risk-stratified testing and highlight GIP limitations, including potential false positives.