Clinical value of mNGS in autoimmune rheumatic disease-associated pneumonia infection in PJP

Objective: The aim of this study was to evaluate the efficacy of second-generation sequencing (mNGS) of the macrogenome in Pneumocystis jiroveci pneumonia (PJP) in patients with autoimmune rheumatic diseases with symptoms of infection or suspected infection. Methods: We analysed the data of 185 patients with autoimmune rheumatic diseases presenting with symptoms of infection or suspected infection in our hospital from 2019–2023, counted the patients' clinical manifestations, laboratory findings, treatments and mNGS results, compared the distribution and characteristics of viruses, bacteria and fungi in the PJP group and the No-PJP group, analysed the risk factors for the occurrence of PJP, and analysed the effect of each index on theAnalyse the risk factors for PJP and analyse the predictive efficacy of each index for PJP. Results: Females, high LDH and PCT values, mNGS test positive microorganisms, presence of fungal infections and glucocorticoid use were significantly associated with PJP (P < 0.05).Viruses and bacteria were the most common co-pathogens in patients with PJP.SIRI can be used to assess the degree or severity of inflammation.Patients using glucocorticoids were associated with a 61% increased risk of contracting PJP relative to those who did not, which was positively associated with PJP emergence (P < 0.05).LASSO analysis showed that LDH had the best performance in predicting PJP, and regression analyses found that the combination of mNGS, use of glucocorticoids, cDMARDs, LDH, and CTX, mNGS, use of glucocorticoids, cDMARDs, LDH, and CTXcorticosteroids, cDMARDs, CD4 + T, and CTX, a combination that had higher performance in predicting PJP. Conclusion: mNGS can be an important tool in the diagnosis of PJP and identification of co-infections.Factors such as female gender, high LDH and PCT values, positive microorganisms detected by mNGS, presence of fungal infections and use of glucocorticoids can be used as potential predictors of PJP, whereas the combination of mNGS, use of glucocorticoids, cDMARDs, LDH and CTX features had a very high efficacy in predicting infections in PJP.