P-1995. Diagnostic Yield of a Multiplex PCR Sample-to-Answer Array among Patients Admitted with Acute Febrile Illness, Uganda, 2019-2023

TL;DR

A new PCR test improved diagnosis of fever causes in Uganda, identifying more pathogens than standard tests, including dangerous ones like plague.

Contribution

Demonstrated that a multiplex PCR array increases diagnostic yield for acute febrile illness in a high-diversity pathogen setting.

Findings

mPCR identified pathogens in 44.3% of patients with negative standard tests.

Yersinia pestis and other high-consequence pathogens were detected.

mPCR detected more malaria cases than traditional methods.

Abstract

In settings with high pathogen diversity, identification of the etiology of acute febrile illness (AFI) is limited by available diagnostics. We sought to assess whether the addition of a multiplex sample-to-answer PCR-based array (mPCR) to standard microbiologic testing would increase pathogen identification among hospitalized febrile patients in Uganda.

Patients ≥18 years of age with fever (≥38.0°C) ≤48 hours after presentation to Mubende Regional Referral Hospital (RRH) and Arua RRH were eligible for enrollment (Table 1). Demographic, clinical, laboratory, and outcome data were collected. Protocolized standard microbiologic testing with blood cultures and rapid diagnostic tests (RDTs) was performed (Table 2). Whole blood was tested with the FilmArray Global Fever Panel – RUO (BioFire Defense, Salt Lake City, Utah), an mPCR sample-to-answer array for 9 viral, 6 bacterial, and Plasmodium spp.

From August 2019 to August 2023, 283 patients were enrolled; mPCR results were available for 191 (Table 1). 58.6% (n=112 of 191) had a positive microbiologic result on standard testing. Malaria was identified in 29.8% (n=57), M. tuberculosis in 13.1% (n=25), hepatitis A in 7.9% (n=15), hepatitis B in 8.9% (n=17), and positive blood culture in 4.8% (n=9; 2 of 9 were possible contaminants). 10.0% (n=19) had newly positive HIV; 29.8% (n=57 of 191) had HIV at discharge; 14.0% of these (n=8 of 57) had cryptococcal antigenemia.

A positive mPCR result was identified in 44.3% (n=35 of 79) of those without a positive microbiologic result on standard testing. Two patients were positive for Yersinia pestis; one had malaria co-infection. One participant each was positive for yellow fever, Leptospira spp, and Salmonella Typhimurium (Table 3). Of 96 malaria-positive on mPCR (50.2%), only 49.0% (n=47 of 96) were identified on smear and/or RDT; 17% (n=10 of 57) of malaria-positive patients by smear or RDT were mPCR negative (κ=0.38, SE 0.06).

Pathogen surveillance in AFI with mPCR increases identification of fever etiology and identifies high-consequence pathogens, enabling treatment of affected patients and intervention by public health systems. Further study of Plasmodium mPCR-positive febrile patients is warranted to understand clinical significance of these detections.

Yukari C. Manabe, MD, FIDSA, FRCP, bioMerieux: Research materials to JHU|Cepheid: Research materials to JHU|Chembio: Grant/Research Support|Chembio: Research materials to JHU|Hologic: Grant/Research Support|Hologic: Research materials to JHU|Roche: Research materials to JHU

Figures