摘要
Abstract
Objective To construct a risk prediction model for infectious peritonitis in peritoneal dialysis(PD)patients based on serum lipopolysaccharide(LPS)and 1,3-β-D glucan(BG).Methods Clinical data of 152 PD patients admitted to Huaian First People's Hospital from June 2023 to May 2024 were retrospectively collected.According to the occurrence of infectious peritonitis within one year,patients were divided into a peritonitis group(42 cases)and a non-peritonitis group(110 cases).General data and serum levels of LPS and BG were compared between the two groups;serum LPS and BG levels were also compared among infectious peritonitis patients with different pathogenic infections.Multivariate logistic regression analysis was used to analyze influencing factors for infectious peritonitis.A nomogram model was constructed,and calibration curves,receiver operating characteristic curves,and decision curves were plotted to validate the model's calibration,predictive value,and clinical applicability,respectively.Results Residual renal function(RRF)and albumin(Alb)levels in the peritonitis group were lower than those in the non-peritonitis group(P<0.05),while serum ferritin(SF),C-reactive protein(CRP),LPS,and BG levels were higher than those in the non-peritonitis group(P<0.05).There was no statistically significant difference in serum LPS and BG levels among infectious peritonitis patients with different pathogenic infections(P>0.05).Multivariate logistic regression analysis showed that elevated levels of SF(OR=1.007,95%CI:1.001-1.013),CRP(OR=1.110,95%CI:1.038-1.187),LPS(OR=86.770,95%CI:1.280-5 880.902),and BG(OR=1.029,95%CI:1.008-1.051)were risk factors for infectious peritonitis in PD patients(P<0.05);high levels of RRF(OR=0.746,95%CI:0.570-0.975)and Alb(OR=0.885,95%CI:0.799-0.979)were protective factors against infectious peritonitis in PD patients(P<0.05).After adjusting for confounding factors,LPS and BG remained independent risk factors for infectious peritonitis in PD patients.Furthermore,multicollinearity testing on the aforementioned variables showed that the variance inflation factor for all indicators was<5 and tolerance was>0.1,suggesting a low possibility of multicollinearity.The nomogram model based on serum LPS and BG showed a concordance index of 0.778(95%CI:0.706-0.850).The area under the curve of the nomogram model for predicting infectious peritonitis in PD patients was 0.814(95%CI:0.742-0.886).The decision curve showed a maximum net benefit of 0.223.Conclusion RRF,Alb,SF,CRP,LPS,and BG are influencing factors for infectious peritonitis in PD patients.The nomogram model constructed based on these factors has certain predictive value for infectious peritonitis.关键词
腹膜透析/感染性腹膜炎/脂多糖/1,3-β-D葡聚糖/列线图Key words
peritoneal dialysis/infectious peritonitis/lipopolysaccharide/1,3-β-D glucan/nomogram
