Screening Of Diagnostic And Prognostic Markers In Exhaled Air Condensate Of Chronic Sinusitis With Nasal Polyps Based On Metabonomics

OBJECTIVEThe upper airway exhaled air condensate(EBC)of chronic sinusitis with nasal polyps(CRSw NP)and control patients was analyzed by metabonomics technique,and a database of metabolites was established.After qualitative and quantitative analysis of quantized metabolites,differential markers were sought,and the metabolic pathway of differential metabolites in disease occurrence and development was discussed through bioinformatics analysis.Metabolites of CRSw NP group,control group and patients with different prognosis were analyzed by machine learning method,and diagnostic and prognostic biomarkers were screened.It is used for early diagnosis and prognosis evaluation of chronic sinusitis with nasal polyps.METHODS1.According to the exclusion and inclusion criteria,this study prospectively included69 patients diagnosed with chronic sinusitis with nasal polyps in Yantai Yuhuangding Hospital affiliated to Qingdao University from June 2021 to March 2022 as the CRSw NP group,and selected 61 patients diagnosed with chronic sinusitis with nasal polyps as the control group during the same period.Samples of exhaled air condensate from the upper airway were collected from both groups.2.Biological samples of exhaled air condensate were analyzed using LC-MS/MS technology and analyzed by metabonomics to obtain relevant data and metabolic map.3.Bioinformation analysis was used to screen significant differential metabolites and their metabolic pathways.4.The machine learning method was used to screen biomarkers,and the panel diagnostic and prognostic models were constructed with biomarkers,and the ROC curve was used to analyze and judge the discriminant effectiveness of the models.RESULTS1.There were no significant differences in gender,age and BMI between the CRSw NP group and the control group(P>0.05).There was significant difference in peripheral blood eosinophils(%)(P < 0.05).2.In this study,a total of 757 metabolites were identified,and 35 different metabolites were identified between the CRSw NP group and the control group(13 were up-regulated and 22 were down-regulated).3.Biological information analysis showed that differential metabolites were enriched in 11 metabolic pathways,mainly concentrated in the metabolic pathways of arginine and proline.4.A total of 8 diagnostic biomarkers were screened out by machine learning method,and the panel diagnostic model was established,whose diagnostic efficacy was 0.981 AUC in the training set and 0.975 AUC in the test set.5.Five prognostic biomarkers were screened out by machine learning method,and the panel prognostic model was established.The predictive efficacy was 0.796 in the test set and 0.934 in the test set.CONCLUSIONS1.EBC can be used as a biological sample for the study of CRSw NP diseases,and its potential value will be explored with the deepening of research.2.The metabolic pathways of arginine and proline may play an important role in the occurrence and development of CRSw NP diseases.3.panel diagnostic model composed of 8 different metabolites has good diagnostic efficacy and is expected to be an auxiliary tool to assist clinicians in decision-making.4.The panel prognostic model established by using 5 biomarkers has certain predictive performance,which can evaluate the prognosis of patients before surgery,and implement targeted treatment measures for patients with poor prognosis during the postoperative treatment.