Objective To investigate risk factors for surgical site infection after total hip arthroplasty based on exhaustive CHAID decision tree and logistic regression model, so as to provide the basis for the development of control strategy of surgical site infection (SSI). MethodsA retrospective analysis was performed on the patients undergoing total hip arthroplasty from January 2012 to December 2018 in Wenzhou People's Hospital. Twenty four patients with surgical site infection after total hip arthroplasty were selected as the case group, and 576 patients who did not have surgical site infection aftertotal hip arthroplasty in the same period were selected as the control group. Logistic regression and decision tree model were used to determine influencing factors for surgical site infection. ResultsFourriskfactorswerefoundbylogisticregressionmodelanddecisiontreemodel, including the operation time no less than 180 min, the amount of intraoperative bleeding more than 200 mL, using at least 2 kinds of antibiotics, and serum albumin less than 35 `g/L. The results of CHAID decision tree analysis showed that the operation time was the main factor affecting the SSI. ConclusionIn order to provide evidence and reference for prevention and control measure of surgical site infection, it is essential to combine these two different methods to describe influencing factors of surgical site infection from different aspects.
周佩敏,苏忠良,徐青,郑舒. 采用决策树和logistic回归模型分析全髋关节置换术后手术部位感染的影响因素[J]. 中国医院统计, 2019, 26(6): 404-407.
Zhou Peimin, Su Zhongliang, Xu Qing, Zheng Shu. Influencing factors of surgical site infection after total hip arthroplasty using decision tree and logistic regression model. journal1, 2019, 26(6): 404-407.
[1]SEAGRAVE KG,TROELSEN A,MALCHAU H,et al.Acetabular cup position and risk of dislocation in primary total hip arthroplasty[J].Acta Orthop,2017,88(1):10-17.
[3]WESTBERG M,FRIHAGEN F,BRUN OC,et al.Effectiveness of gentamicincontaining collagen sponges for prevention of surgical site infection after hip arthroplasty:A multicenter randomized trial[J].Clin Infect Dis,2015,60(12):1752-1759.
[4]MEROLLINI KM,CRAWFORD RW,WHITEHOUSE SL,et al.Surgical site infection prevention following total hip arthroplasty in Australia:A costeffectiveness analysis[J].Am J Infect Control,2013,41(9):803-809.
[5]BERROSTORRES SI,UMSCHEID CA,BRATZLER DW,et al.Centers for disease control and prevention guideline for the prevention of surgical site infection,2017[J].JAMA Surg,2017,152(8):784-791.
[12]TRIANTAFYLLOPOULOS G,STUNDNER O,MEMTSOUDIS S,et al.Patient,surgery,and hospital related risk factors for surgical site infections following total hip arthroplasty[J].Sci World J,2015:979560.
[13]RASOULI MR,RESTREPO C,MALTENFORT MG,et al.Risk factors for surgical site infection following total joint arthroplasty[J].J Bone Jo SurgAm,2014,96(18):e158.