带状疱疹后神经痛发生的影响因素及临床预测模型构建

Influence factors of occurrence of postherpetic neuralgia and establishment of clinical prediction model

目的:探讨带状疱疹(herpeszoster,HZ)病人发生带状疱疹后神经痛(postherpetic neuralgia,PHN)的影响因素并建立临床预测模型。方法:选取2020年8月至2021年3月中日友好医院疼痛门诊就诊的HZ病人进行前瞻性研究,采用自拟《带状疱疹后神经痛病例报告表》详细记录病人临床资料,采用门诊复查、电话随访等方式进行12周的随访,比较发生与未发生PHN病人上述调查表资料的差异,多因素回归分析筛选影响PHN发病的影响因素后,应用Logistic回归方法建立合适的模型,并绘制预测HZ病人发生PHN的列线图。最后计算C指数、绘制模型的校准曲线、接收者操作特征曲线和临床决策曲线进一步验证预测模型的准确性。结果:共纳入174例HZ病人,按PHN发生情况分为PHN组(52例)和非PHN组(122例)。多因素回归分析筛选出女性、年龄>50岁、有前驱痛、急性期皮损面积大、急性期疼痛程度重共5个预测PHN发生的危险因素。根据多因素回归分析筛选的5个因素构建预测模型列线图,进一步制得评分系统。检验评分系统的准确性显示,C指数为0.81,95%置信区间为0.77-0.95,模型校准曲线与实际曲线重合较好,验证模型准确性的ROC曲线下面积为0.81。临床决策曲线分析病人发生概率在>6%和<92%这个范围内,这种模型在临床上应用的准确性、净获益最高,超出该范围此模型的准确性有限,净获益将大幅度下降。结论:女性、年龄>50岁、有前驱痛、急性期皮损面积大、急性期疼痛程度重这5个危险因素均会影响PHN的发生概率,根据ROC曲线下面积0.81、模型校准曲线与实际曲线重合较好说明本研究所制模型准确性较高。

Objective:To explore influence factors for the occurrence of postherpetic neuralgia(PHN)in herpes zoster(HZ)patients and establish a clinical prediction model.Methods:A prospective study was conducted in the pain clinic of China-Japan Friendship Hospital from August 2020 to March 2021.The self-made case report form of postherpetic neuralgia was used to record various clinical data of the patients in detail.The patients were followed up for 3 months through outpatient reexamination and telephone follow-up,and the differences between the above questionnaire data of patients with and without PHN were compared.Multiple logistic regression analysis was used to screen the predictors of occurrence of PHN in HZ patients.Logistic regression method was used to establish a suitable model,and the nomogram was drawn to predict the occurrence of PHN in HZ patients.The calibration curve,C-index,Receiver operating characteristic(ROC)curve and decision curve analysis were drawn to further verify the accuracy of the prediction model.Results:A total of 174 HZ patients were collected.According to the occurrence of PHN,they were divided into PHN group(52 cases)and nonPHN group(122 cases).Multiple logistic regression analysis finally screened out 5 risk factors for predicting PHN,including female,age>50 years old,prodromal pain,large lesion area and high VAS score in acute stage.According to the 5 risk factors screened by multiple logistic regression analysis,the prediction model nomogram was constructed and the scoring system was further developed.The accuracy of the test scoring system shows that the C index is 0.81,the 95%confidence interval is 0.77-0.95,the model calibration curve coincides with the actual curve,and the area under the ROC curve to verify the accuracy of the model is 0.81.The analysis of clinical decision curve shows that the recurrence probability of patients is in the range of>6%and<92%,and this model has the highest accuracy and net benefit in clinical application.Beyond this range,the accuracy of this model is limited,and the net benefit will be greatly reduced.Conclusion:Female,age>50 years old,prodromal pain,large lesion area and high VAS score in acute stage will affect the incidence of postherpetic neuralgia.According to the under ROC curve area of 0.81 and the good coincidence between the model calibration curve and the actual curve,the accuracy of the model established in this study is high.