Joint application of multiplex drop-off digital PCR, droplet digital PCR, and metagenomic next-generation sequencing for the diagnosis of suspected infectious diseases: A retrospective cohort study

TL;DR

This study compares three molecular diagnostic methods for infectious diseases in ICU patients and finds that combining them improves diagnostic accuracy and guides treatment decisions.

Contribution

The study demonstrates the added value of combining MDO-dPCR, ddPCR, and mNGS for diagnosing infectious diseases in critically ill patients.

Findings

Joint detection using MDO-dPCR, ddPCR, and mNGS showed a high negative predictive value and improved clinical decision-making.

mNGS had the highest sensitivity (96.6%), while MDO-dPCR and ddPCR showed moderate sensitivity and higher specificity.

Pathogen copy numbers detected by MDO-dPCR correlated with clinical severity markers like procalcitonin and SOFA scores.

Abstract

Critically ill patients in ICUs are highly vulnerable to infectious diseases. Early and accurate identification of pathogens is vital for initiating appropriate antimicrobial therapy. To evaluate the diagnostic effectiveness in patients with suspected infectious diseases; three different molecular technologies and conventional microbiological tests were used.

A total of 97 individuals suspected of having infectious diseases were retrospectively enrolled from July 2023 to January 2024 at Shanghai General Hospital. Samples were collected for metagenomic next-generation sequencing (mNGS), droplet digital polymerase chain reaction (ddPCR), multiplex drop-off digital polymerase chain reaction (MDO-dPCR), and conventional microbiological tests (CMTs) for suspected pathogen detection. The diagnostic efficacies of the three molecular technologies and CMTs were compared, and the effects of their joint application on clinical outcomes were evaluated. Intergroup comparisons were performed using the Kruskal–Wallis test, with a P-value <0.05 considered statistically significant.

Joint detection exhibited a high negative predictive value. The sensitivity of MDO-dPCR, ddPCR, and mNGS was 52.6%, 48.5%, and 96.6%, respectively; and the corresponding specificity was 72.5%, 73.3%, and 50.0%. A positive correlation was observed between pathogen copies detected using MDO-dPCR and procalcitonin (Pearson’s ρ=0.21, P=0.039), acute physiology and chronic health evaluation II (Pearson’s ρ=0.24, P =0.018), and sequential organ failure assessment (Pearson’s ρ=0.25, P=0.012). Therapeutic regimens were adjusted in 51.5% of the patients (50/97) based on the results of the combination tests.

In the present study, we highlighted the significance of molecular technologies for the early diagnosis of patients with suspected infections. These technologies can serve as a complement to CMTs and should be implemented promptly to guide clinicians in providing timely and effective anti-infective treatments. Future studies should aim to confirm these findings in large-scale clinical trials to refine diagnostic protocols, while also incorporating cost-utility analyses.

Figures