Clinical value of droplet digital PCR in the diagnosis and dynamic monitoring of suspected bacterial bloodstream infections

Background
Bloodstream infections (BSIs) represent a significant public health challenge due to their high morbidity and mortality. The clinical prognosis of BSIs is closely related to the timely and accurate diagnosis and the rational use of initial antimicrobials. We aimed to evaluate the clinical value of droplet digital PCR (ddPCR) in rapid diagnosis and dynamic monitoring of BSIs.

Methods
In this prospective study, using a ddPCR-based approach which detects 18 common pathogens, we compared the detection results and clinical concordance rates of ddPCR with blood culture (BC) in 211 patients with suspected BSIs. Further, the inflammatory profile of BSIs with Gram-negative bacteria was analyzed by Olink proteomics platform.

Results
Our data showed that the positive detection rate of ddPCR was 48.82%, which was higher than that of BC (9.48%). For BC-validated BSIs, ddPCR had a sensitivity of 90.00% and a specificity of 55.50%. When considering clinically-validated BSIs, the diagnostic value of ddPCR improved with a sensitivity of 92.59% and a specificity of 78.46%.The bacterial load detected by ddPCR was correlated with traditional clinical inflammatory indicators such as interleukin-6 (IL-6) and C-reactive protein (CRP). In addition, using Olink proteomics platform, we revealed that serological osteoprotegerin (OPG), interleukin-8 (IL-8), interleukin-18 receptor 1 (IL-18R1), C–C motif chemokine 20 (CCL20) and IL-6 were substantially elevated in Gram-negative bacteria-associated BSIs, which could serve as novel auxiliary diagnostic indicators for Gram-negative bacteria BSIs.

Conclusion
ddPCR has the potential to provide early pathogen diagnosis, dynamic monitoring, and treatment regimen optimization for patients with BSIs.