Objective:To construct the risk prediction model for early-onset peritoneal dialysis-associated peritonitis (PDAP) in peritoneal dialysis patients based on six machine learning algorithms.Methods:This study was retrospective. Convenience sampling was used to select peritoneal dialysis patients who were regularly followed up in the Department of Nephrology of Guizhou Provincial People's Hospital from December 2009 to August 2023 to collect general information, primary diseases, and laboratory indicators of the study population. It was randomly divided into a modeling set and validation set in the ratio of 7∶3. With the occurrence of early-onset PDAP as the dependent variable, the risk prediction model of early-onset PDAP in peritoneal dialysis patients was constructed based on six machine learning algorithms, namely, Logistic regression, decision tree, support vector machine, random forest, extreme gradient boosting, and artificial neural network, respectively. Model performance was evaluated based on the area under the receiver operating characteristic curve ( AUC) , accuracy, and F1 score to select the optimal model. Results:The final data of 890 peritoneal dialysis patients were analyzed, of which 86 patients developed early-onset PDAP, and the incidence of early-onset PDAP was 9.66%. The four prediction models, Logistic regression, support vector machine, extreme gradient boosting, and random forest, had high accuracy with AUC values of 0.703, 0.729, 0.782, and 0.814, respectively, with the random forest model having higher AUC value, accuracy, and F1 score. Further ranking of the importance of risk factors for early-onset PDAP based on the random forest model showed that the top five characteristic variables were C-reactive protein, triglycerides, platelet, ferritin, and leukocyte, in that order. Conclusions:The risk prediction model for early-onset PDAP in peritoneal dialysis patients constructed based on the random forest model has optimal performance, which can help medical and nursing staff assess and prevent early-onset PDAP at an early stage.

Objective:To construct the risk prediction model for early-onset peritoneal dialysis-associated peritonitis (PDAP) in peritoneal dialysis patients based on six machine learning algorithms.Methods:This study was retrospective. Convenience sampling was used to select peritoneal dialysis patients who were regularly followed up in the Department of Nephrology of Guizhou Provincial People's Hospital from December 2009 to August 2023 to collect general information, primary diseases, and laboratory indicators of the study population. It was randomly divided into a modeling set and validation set in the ratio of 7∶3. With the occurrence of early-onset PDAP as the dependent variable, the risk prediction model of early-onset PDAP in peritoneal dialysis patients was constructed based on six machine learning algorithms, namely, Logistic regression, decision tree, support vector machine, random forest, extreme gradient boosting, and artificial neural network, respectively. Model performance was evaluated based on the area under the receiver operating characteristic curve ( AUC) , accuracy, and F1 score to select the optimal model. Results:The final data of 890 peritoneal dialysis patients were analyzed, of which 86 patients developed early-onset PDAP, and the incidence of early-onset PDAP was 9.66%. The four prediction models, Logistic regression, support vector machine, extreme gradient boosting, and random forest, had high accuracy with AUC values of 0.703, 0.729, 0.782, and 0.814, respectively, with the random forest model having higher AUC value, accuracy, and F1 score. Further ranking of the importance of risk factors for early-onset PDAP based on the random forest model showed that the top five characteristic variables were C-reactive protein, triglycerides, platelet, ferritin, and leukocyte, in that order. Conclusions:The risk prediction model for early-onset PDAP in peritoneal dialysis patients constructed based on the random forest model has optimal performance, which can help medical and nursing staff assess and prevent early-onset PDAP at an early stage.

Objective:To explore the association between platelet/lymphocyte ratio (PLR) and frequent peritoneal dialysis (PD) - associated peritonitis (PDAP) in PD patients.Methods:The data of PD patients with PDAP from Guizhou Provincial People's Hospital between January 2015 and June 2019 were analyzed retrospectively. The patients were divided into mono group (only once PDAP occurred in one year) and frequent group (2 or more PDAP occurred in one year) according to the frequency of PDAP. The demographic data including gender, age, height and weight, the clinical data including blood pressure, duration of PD, causes of peritonitis, the laboratory data at the first time of PDAP and the prognosis of PDAP were compared between two groups. Logistic regression analysis method was applied to analyze the relationship between PLR and frequent PDAP. The predictive power of PLR was evaluated by receiver operating characteristic curve (ROC).Results:A total of 78 PD patients with PDAP were enrolled, including 53 males and 25 females, with average age of 45.2 years. The total person-year was 765.1 person-years and the incidence of peritonitis was 0.10 case/person-year during the median follow-up of 16 months. All patients were divided into two groups: 53 patients in mono group and 25 patients in frequent group. Compared with mono group, the patients in frequent group had lower body mass index, longer dialysis duration, higher systolic blood pressure level, higher PLR level, lower uric acid level, and higher rate of drug-resistant bacteria in peritoneal effusion (all P<0.05). The extubation rate of the frequent group was 44.0%(11/25), which was significantly higher than that [15.1%(8/53)] of mono group ( P<0.05). Multivariate logistic regression analysis showed that higher PLR level was an independent related factor for frequent PDAP( OR=1.006, 95% CI 1.002-1.010, P=0.003), and the area under the ROC curve of PLR was 0.783(95% CI 0.663-0.904, P<0.001). Conclusions:High PLR level is an independent related factor of frequent PDAP for PD patients, and PLR can be a potential predictor of frequent PDAP.