Predictive performance of Metagenomic Next Generation Sequencing in early detection of post-liver transplantation infections

Objective: To evaluate the predictive performance of metagenomic next-generation sequencing (mNGS) in identifying and predicting pulmonary infections following liver transplantation and to investigate its association with patient outcomes within the initial four-week post-transplantation period. Methods: We retrospectively analyzed 41 liver transplant patients with suspected pulmonary infections from August 2022 to May 2023. Bronchoalveolar lavage fluid (BALF) samples were collected on the first postoperative day for metagenomic next generation sequencing (mNGS) and culture. The predictive capability of mNGS for subsequent infections was assessed by monitoring inflammatory biomarkers and comparing the detection rates with culture methods. Real-time Polymerase Chain Reaction (Rt-PCR) was used to monitor Human betaherpesvirus 5 (CMV) and Human parvovirus B19 (B19) weekly during a four-week postoperative period. Inflammatory biomarkers and blood coagulation function were evaluated on specific days throughout the first, third, fifth, and during four weeks following surgery. The study was conducted until August 2023 to evaluate the patients' prognostic survival outcome, classifying them into groups based on the mortality and survival. Results: The analysis included a total of 41 patients, comprising 32 males and 9 females, with an average age of 52 (47, 63) years. Within one week after liver transplantation, there were 7 cases of bacterial infections, 5 cases of fungal infections, 19 cases of mixed infections, 8 cases without any infection, and 2 cases with unidentified pathogen-associated infections. mNGS successfully predicted 39 (72 %) strains of pathogens, while culture-based methods only detected 28 (52 %) strains. Among the 8 patients diagnosed as non-infected, culture methods identified positive results in 4 cases (50 %), whereas mNGS yielded positive results in 7 cases (87.5 %). The detection rates of CMV and B19 by Rt-PCR within 4 weeks after liver transplantation were 61 % and 17 %, respectively (25/41, 7/41) among the patients. During the study period, a total of 9 patients succumbed while 32 patients survived. The death group and the survival group exhibited significant differences in CRP, HGB, and INR levels at specific monitoring time points. The proportion of CMV detection in blood was significantly higher in the death group compared to the surviving group. Elevated CRP level was identified as a prognostic risk factor. Conclusions: Despite the presence of false positives, mNGS still presents a potential advantage in predicting pulmonary infection pathogens following liver transplantation. Furthermore, the levels of CRP and CMV carrier status within four weeks post-surgery exhibit significant associations with patient survival and prognosis.