老年患者腰椎融合术后发生邻近节段椎体压缩骨折的列线图预测模型

Establishment and validation of a nomogram for predicting adjacent vertebral compression fractures after lumbar fusion in elderly patients

目的:探究老年患者腰椎融合术后邻近节段椎体压缩骨折(AVCF)的危险因素并构建列线图预测模型。方法:回顾性分析2017年1月至2021年12月湖北省中西医结合医院收治的老年腰椎融合术患者297例。根据是否发生AVCF分为骨折组和非骨折组,记录两组患者的年龄、性别、体重指数、文化程度、合并基础疾病数量、病程时间、术前骨密度T值、术中出血量、术中手术融合节段数量、术后输血量、术后是否进入ICU监护、术后外固定种类及外固定时间,对术后发生AVCF的影响因素进行单因素和多因素logistic回归分析。利用R软件构建列线图预测模型,并评估模型预测效能。结果:根据患者是否发生AVCF分为骨折组67例和非骨折组230例,老年患者腰椎融合术后发生AVCF的概率为22.56%。老年患者腰椎融合术后邻近椎体压缩骨折的影响因素有年龄、文化程度、术前骨密度T值、术中出血量、术后输血量、术后进入ICU监护(P均<0.05);多因素logistic回归分析提示年龄、术前骨密度T值、术中出血量、术后输血量、术后进入ICU监护为老年患者腰椎融合术后邻近椎体压缩骨折的独立危险因素(P均<0.05)。受试者工作特征(ROC)曲线结果显示,预测老年患者腰椎融合术后邻近椎体压缩骨折发生风险的曲线下面积为0.769,Hosmer-Lemeshow拟合优度为χ^(2)=8.009,P=0.533;校准曲线斜率接近1,一致性良好。结论:基于年龄、术前骨密度T值、术中出血量、术后输血量、术后进入ICU监护构建了老年患者腰椎融合术后邻近椎体压缩骨折的列线图预测模型,该模型对于老年患者腰椎融合术后邻近椎体压缩骨折有较好的预测价值,可以对高风险患者制定针对性策略,以降低腰椎融合术后AVCF的发生风险。

Objective:To explore the risk factors for adjacent vertebral compression fractures(AVCF)after lumbar fusion in elderly patients and to construct a nomogram predictive model.Methods:A retrospective analysis was conducted on 297 elderly patients who underwent lumbar fusion surgery at the Hubei Provincial Hospital of Integrated Chinese and Western Medicine from January 2017 to December 2021.Patients were divided into fracture and non-fracture groups based on the development of AVCF.Demographic and clinical data were collected,including age,sex,body mass index,educational level,number of comorbidities,duration of disease,preoperative bone mineral density T score,intraoperative blood loss,number of fusion segments,postoperative blood transfusion volume,postoperative ICU monitoring,and type and duration of external fixation.Univariate and multivariate logistic regression analyses were conducted to identify the risk factors for postoperative AVCF.A nomogram predictive model was constructed using R software and its predictive performance was evaluated.Results:There were 67 cases in the fracture group and 230 cases in the nonfracture group,with a 22.56%incidence rate of AVCF.The risk factors for AVCF included age,educational level,preoperative bone mineral density T score,intraoperative blood loss,postoperative blood transfusion volume,and postoperative ICU monitoring(all P<0.05).Multivariate logistic regression analysis identified age,preoperative bone mineral density T score,intraoperative blood loss,postoperative blood transfusion volume,and postoperative ICU monitoring as independent risk factors for postoperative AVCF in elderly patients(all P<0.05).The area under the receiver operating characteristic curve was 0.769,and the Hosmer-Lemeshow goodness-of-fit test yielded χ^(2)=8.009,P=0.533.The calibration curve slope was close to 1,indicating good consistency.Conclusions:A nomogram predictive model for postoperative AVCF in elderly patients was successfully developed,incorporating age,preoperative bone mineral density,intraoperative blood loss,postoperative blood transfusion volume,and postoperative ICU monitoring.This model has good predictive value for postoperative AVC,enabling targeted interventions to reduce the risk of AVCF after lumbar fusion.