探讨影响初次髋关节置换术后持续疼痛的危险因素，并探索一种预测其风险的模型。

纳入随访时间≥6个月初次行髋关节置换术且资料齐全的患者，并排除术后发生深部感染或合并骨盆骨折或其他多关节损伤，或存在其他疾病可导致髋关节周围疼痛的患者，最后一共选择2017年1月至2019年12月在3家医院初次行髋关节置换术标准的病例475例作为研究对象。通过电子病历系统收集患者临床资料，检索国内外相关文献并结合临床，选取可能与初次髋关节置换术后发生持续疼痛相关的危险因素纳入研究。选取其中342例病例作为研究对象，进行相关风险因素分析及风险预测模型的建立，其中单因素分析采用卡方检验，多因素分析采用logistic分析。之后将模型应用于临床实际，对剩下的133例患者进行评估，结果采用受试者工作特征(ROC)和曲线下面积(AUC)进行统计分析，对该风险预测模型的鉴别效度和预测能力进行评价。

多因素logistic分析结果显示年龄≥75岁、身体质量指数(BMI)、糖尿病史、置换类型、术中出血量、下肢静脉血栓、术后C反应蛋白(CRP)≥10 mg/L是影响初次髋关节置换术后持续疼痛的独立危险因素，具有统计学意义(P<0.05)。对这些因素使用logistic回归分析构建出风险预测模型：y＝e^Logit(P)/[1＋e^Logit(P)]，其中y为发生持续疼痛的概率，e为指数函数。用模型对患者进行评估后显示模型总正确率为72.18%，并通过ROC曲线评价该模型的鉴别效度，结果显示：AUC＝0.863，敏感度为92.86%，特异度为62.86%。

初次髋关节置换术后持续性疼痛的影响因素主要为年龄≥75岁、BMI、糖尿病史、置换类型、术中出血量、下肢静脉血栓、术后CRP≥10 mg/L；根据上述影响因素构建出的风险预测模型对初次髋关节置换术后发生持续疼痛风险预测具有良好的判别能力，但是仍然需要通过前瞻性研究进行进一步的确认。

To explore the risk factors of persistent pain after primary hip replacement and to develop a model for predicting the risk.

According to the inclusion criteria (follow-up time≥six months and completed data of primary hip replace ment) and exclusion criteria(postoperative deep infection, combining pelvic fractures or other multiple joints injuries and diseases), 475 patients receiving primary hip arthroplasty in three hospitals from January 2017 to December 2019 were selected as the research subjects. The clinical data of patients was collected through the electronic medical record system, and the relevant literatures at home and abroad were retrieved and combined with clinical practice. The risk factors that might be associated with persistent pain after primary hip arthroplasty were selected for the study. Among them, 342 cases were selected, and related risk factor analysis and risk prediction model were established. Chi-square test was used for univariate analysis, and logistic analysis was used for multivariate analysis. The model was applied to clinical practice and the remaining 133 patients were evaluated. The results were analyzed using the area under the receiver operating characteristic (ROC) and area under the curve (AUC) to evaluate the differential validity and predictive power of the risk prediction model.

Multi-factor logistic analysis showed that age≥75 years, body mass index (BMI), history of diabetes, displacement type, intraoperative blood loss, lower limb vein thrombosis and postoperative C-reactive protein (CRP)≥10 mg/L were the independent risk factors for the development of continuous pain after total hip arthroplasty (P<0.05). The risk prediction model for these factors was as follows: y=e^Logit(P)/[1+ e^Logit(P))], in which y is the probability of persistent pain and e is an exponential function. After evaluating the patients with the model, the total accuracy of the model was 72.18%. The differential validity of the model was evaluated by ROC curve, and the results showed that the AUC was 0.863, the sensitivity was 92.86%, and the specificity was 62.86%.

The major influencing factors for persistent pain after primary total hip arthroplasty are the age≥75 years, BMI, diabetes mellitus, the displacement type, intraoperative blood loss, lower limb vein thrombosis and postoperative CRP≥10 mg/L. The risk prediction model constructed according to the above influencing factors has good discriminant ability for the risk prediction of persistent pain after primary hip replacement, but still needs further confirmation by prospective study.