能谱CT征象和相关参数联合预测早期股骨头坏死患者股骨头塌陷风险的价值

Value of spectral CT signs combined with related parameters in predicting the risk of femoral head collapse in patients with early femoral head necrosis

目的探讨能谱CT征象和相关参数联合预测早期股骨头坏死患者股骨头塌陷风险的价值,以期为临床早期制定干预措施提供依据。方法前瞻性选取2022年3月至2023年2月平煤神马医疗集团总医院收治的早期股骨头坏死患者101例,根据确诊随访1年后是否发生股骨头塌陷分为塌陷组42例和非塌陷组59例。所有患者均行能谱CT检查,比较两组患者的能谱CT征象、CT值和能谱曲线斜率,采用受试者工作特征(ROC)曲线评价能谱CT征象、CT值、能谱曲线斜率对股骨头塌陷风险的预测价值。结果塌陷组患者的裂隙征、新月征、骨髓水肿率明显高于非塌陷组,差异均有统计学意义(P<0.05);塌陷组患者40~140kev能量下的CT值、能谱曲线斜率明显低于非塌陷组,差异均有统计学意义(P<0.05);经ROC曲线分析显示,能谱CT下裂隙征、新月征、骨髓水肿、能谱曲线斜率、40~140 kev能量下CT值预测股骨头塌陷风险的AUC分别为0.568、0.668、0.827、0.766、0.720、0.713、0.758、0.850、0.782、0.726、0.837、0.803、0.779、0.800、0.809;能谱CT下裂隙征、新月征、骨髓水肿、70 kev能量下CT值、能谱曲线斜率进行Logistic二元回归拟合,返回预测概率Logit(P)作为独立检验变量,获取联合预测的AUC为0.925(95%CI:0.877~0.974)。结论能谱CT征象、CT值、能谱曲线斜率联合预测股骨头塌陷风险具有较高价值,可为临床早期股骨头坏死塌陷制定针对性干预方案,改善预后提供参考依据。

Objective To explore the value of spectral CT signs and related parameters in combination for predicting the risk of femoral head collapse in patients with early femoral head necrosis,in order to provide a basis for the early clinical intervention measures.Methods A prospective study was conducted on 101 patients with early femoral head necrosis admitted to the General Hospital of Pingmei Shenma Group from Mar.2022 to Feb.2023.The patients were divided into a collapse group of 42 cases and a non-collapse group of 59 cases based on whether or not they had femoral head collapse after one year of follow-up.All patients underwent spectral CT examination.The spectral CT signs,CT values,and spectral curve slopes were compared between the two groups.The receiver operating characteristic(ROC)curve was used to evaluate the predictive value of spectral CT signs,CT values,and spectral curve slopes for the risk of femoral head collapse.Results The rates of fissure sign,crescent sign,and bone marrow edema in the collapse group were significantly higher than those in the non-collapse group(P<0.05).The CT values at 40-140 keV energy and spectral curve slopes in the collapse group were significantly lower than those in the non-collapse group(P<0.05).ROC curve analysis showed that the AUCs of the fracture sign,crescent sign,bone marrow edema,spectral curve slopes,CT values at 40-140 keV energy for predicting the risk of femoral head collapse were 0.568,0.668,0.827,0.766,0.720,0.713,0.758,0.850,0.782,0.726,0.837,0.803,0.779,0.800,and 0.809,respectively.Logistic binary regression fitting was performed on the fissure sign,crescent sign,bone marrow edema,CT value at 70 kev energy,and spectral curve slopes under spectral CT,and the predicted probability Logit(P)was returned as an independent test variable.The AUC of the combination of the indexes for predicting the risk of femoral head collapse was 0.925(95%CI:0.877-0.974).Conclusion The combination of spectral CT signs,CT values,and spectral curve slopes has significant value in predicting the risk of femoral head collapse.It can provide a reference for formulating targeted intervention plans for early femoral head necrosis collapse in clinical practice and improve the prognosis of patients.