胸腰椎骨折经皮椎体后凸成形术后骨水泥渗漏的影响因素及与早期预后的关系

Factors influencing bone cement leakage after percutaneous kyphoplasty for thoracolumbar vertebral fractures and its association with early prognosis

摘要:

目的 探讨胸腰椎骨折经皮椎体后凸成形（PKP）术后骨水泥渗漏的影响因素及与早期预后的关系。方法 选取2020年5月—2023年8月在苏州市中医医院和太仓市中医医院行PKP术治疗的94例胸腰椎骨折患者。采用X射线片、MRI、CT检查统计患者术后骨水泥渗漏发生情况，依据骨水泥渗漏是否发生分为发生组和未发生组，对比两组患者基础资料，以及影响胸腰椎骨折患者PKP术后发生骨水泥渗漏的因素。结合患者术后3个月预后状况将其分为预后良好组和预后不良组，分析两组患者基础资料及影响胸腰椎骨折患者PKP术后预后状况的因素。结果 94例胸腰椎骨折患者，PKP术后骨水泥渗漏25例，渗漏率26.60%；PKP术后早期预后不良有20例，不良率为21.28%。发生组与未发生组患者性别构成、年龄、高血压、糖尿病、受伤原因、AO分型、伤椎位置、体质量指数（BMI）、病程时间比较，差异均无统计学意义（P >0.05）。发生组骨水泥量≥ 6 mL、椎体周壁破坏、椎体压缩程度≥ 50%和椎基静脉孔相通均高于未发生组（P <0.05）。多因素逐步Logistic回归分析结果显示：骨水泥量≥ 6 mL［O^R=3.333（95% CI：1.282，8.664）］、有椎体周壁破坏［O^R=3.893（95% CI：1.489，10.177）］、椎体压缩程度≥ 50%［O^R=3.606（95% CI：1.387，9.373）］、椎基静脉孔相通［O^R=3.900（95% CI：1.477，0.298）］是胸腰椎骨折患者PKP术后发生骨水泥渗漏的危险因素（P <0.05）。ROC曲线分析结果显示，骨水泥量≥6 mL、有椎体周壁破坏、椎体压缩程度≥ 50%、椎基静脉孔相通、联合预测胸腰椎骨折患者PKP术后发生骨水泥渗漏的敏感性分别为64.0%（95% CI：0.530，0.750）、56.0%（95% CI：0.449，0.671）、56.00%（95% CI：0.450，0.670）、52.0%（95% CI：0.410，0.630）、80.00%（95% CI：0.723，0.877）；特异性分别为65.2%（95% CI：0.543，0.763）、75.4%（95% CI：0.644，0.874）、73.90%（95% CI：0.629，0.849）、78.3%（95% CI：0.673，0.893）、97.1%（95% CI：0.821，1.000），联合预测的曲线下面积为0.886，有较高的预测价值。预后不良组与预后良好组患者性别构成、高血压、糖尿病、受伤原因、AO分型、伤椎位置、BMI、椎体周壁破坏、椎基静脉孔相通、病程时间比较，差异均无统计学意义（P >0.05）。不良组年龄、骨水泥量≥ 6 mL、骨水泥渗漏、椎体压缩程度≥ 50%均高于良好组（P <0.05）。多因素逐步Logistic回归分析结果显示：骨水泥量≥ 6 mL［O^R=3.333（95% CI：1.282，8.664）］、年龄较高［O^R=1.323（95% CI：1.105，1.584）］、椎体压缩程度≥ 50%［O^R=3.893（95% CI：1.489，10.177）］、有骨水泥渗漏［O^R=7.125（95% CI：2.585，19.637）］是胸腰椎骨折患者PKP术后预后不良的影响因素（P <0.05）。ROC曲线分析结果显示，骨水泥量≥ 6 mL、年龄较大、椎体压缩程度≥ 50%、有骨水泥渗漏、联合预测胸腰椎骨折患者PKP术后预后不良的敏感性分别为66.2%（95% CI：0.562，0.762）、73.0%（95% CI：0.620，0.840）、75.7%（95% CI：0.647，0.867）、82.4%（95% CI：0.714，0.934）、89.2%（95% CI：0.728，0.971）；特异性分别为75.0%（95% CI：0.650，0.850）、70.0%（95% CI：0.590，0.811）、65.0%（95% CI：0.581，0.729）、70.0%（95% CI：0.627，0.774）、90.0%（95% CI：0.829，0.971）。且联合预测的曲线下面积为0.905，有较高的预测价值。结论 骨水泥量≥ 6 mL、有椎体周壁破坏、椎体压缩程度≥ 50%、椎基静脉孔相通是胸腰椎骨折患者PKP术后发生骨水泥渗漏的影响因素，且骨水泥量≥6 mL、年龄较大、椎体压缩程度≥ 50%、有骨水泥渗漏是胸腰椎骨折患者PKP术后预后不良的影响因素，早期识别和纠正可逆因素有助于改善PKP术胸腰椎骨折患者骨水泥渗漏和预后状况。

Abstract:

Objective To analyze the influencing factors of bone cement leakage after percutaneous kyphoplasty (PKP) for thoracolumbar vertebral fractures and its relationship with early prognosis.Methods Ninety-four patients with thoracolumbar vertebral fractures who underwent PKP treatment in Suzhou Hospital of Traditional Chinese Medicine and Taicang Hospital of Traditional Chinese Medicine from May 2020 to August 2023 were selected. X-ray, MRI and CT scans were used to analyze the incidence of postoperative bone cement leakage, and patients were divided into the occurrence group and the non-occurrence group based on whether bone cement leakage occurred. The general characteristics of the two groups were compared, and the factors affecting bone cement leakage after PKP in patients with thoracolumbar vertebral fractures were analyzed. The patients were further divided into the good prognosis group and the poor prognosis group based on their outcomes three months after surgery. The general characteristics of the two subgroups were compared, and factors influencing postoperative prognosis after PKP in patients with thoracolumbar vertebral fractures were analyzed.Results In 94 patients with thoracolumbar vertebral fractures, there were 25 cases of bone cement leakage after PKP, with a leakage rate of 26.60%. There were 20 patients with a poor prognosis after PKP, with a poor prognosis rate of 21.28%. There were no statistically significant differences between the occurrence group and the non-occurrence group in terms of sex composition, age, history of hypertension and diabetes, cause of injury, AO classification, location of fractures, BMI, and disease duration (P > 0.05). The proportion of bone cement volume ≥ 6 mL, the rate of vertebral wall destruction, the proportion of the vertebral compression degree ≥ 50%, and the rate of connections between the basivertebral foramen and the intravertebral cleft were all higher in the occurrence group than those in the non-occurrence group. Multivariable Logistic regression analysis revealed that bone cement volume ≥ 6 mL [O^R = 3.333 (95% CI: 1.282, 8.664) ], vertebral wall destruction [O^R = 3.893 (95% CI: 1.489, 10.177) ], vertebral compression degree ≥ 50% [O^R = 3.606 (95% CI: 1.387, 9.373) ], and connections between the basivertebral foramen and the intravertebral cleft [O^R = 3.900 (95% CI: 1.477, 0.298) ] were risk factors for bone cement leakage after PKP in patients with thoracolumbar vertebral fractures (P < 0.05). According to the ROC curve analysis, the sensitivities of bone cement volume ≥ 6 mL, vertebral wall destruction, vertebral compression degree ≥ 50%, connections between the basivertebral foramen and the intravertebral cleft, and their combination for predicting bone cement leakage after PKP in patients with thoracolumbar vertebral fractures were 64.0% (95% CI: 0.530, 0.750) and 56.0% (95% CI: 95% CI: 0.449, 0.671), 56.0% (95% CI: 0.450, 0.670), 52.0% (95% CI: 0.410, 0.630), and 80.0% (95% CI: 0.723, 0.877), with the specificities being 65.2% (95% CI: 0.543, 0.763), 75.4% (95% CI: 0.644, 0.874), 73.9% (95% CI: 0.629, 0.849), 78.3% (95% CI: 0.673, 0.893), and 97.1% (95% CI: 0.821, 1.000). The area under the curve (AUC) of the combined detection was 0.886, demonstrating a high predictive value. There were no statistically significant differences between the poor prognosis group and the good prognosis group in terms of sex composition, history of hypertension and diabetes, cause of injury, AO classification, location of fractures, BMI, vertebral wall destruction, connections between the basivertebral foramen and the intravertebral cleft, and disease duration (P > 0.05). The age, the proportion of bone cement volume ≥ 6 mL, the incidence of bone cement leakage, and the proportion of the vertebral compression degree ≥50% in the poor prognosis group were higher than those in the good prognosis group (P < 0.05). Multivariable Logistic regression analysis showed that bone cement volume ≥ 6 mL [O^R = 3.333 (95% CI: 1.282, 8.664) ], older age [O^R = 1.323 (95% CI: 1.105, 1.584) ], vertebral compression degree ≥ 50% [O^R = 3.893 (95% CI: 1.489, 10.177) ], and bone cement leakage [O^R = 7.125 (95% CI: 2.585, 19.637) ] were factors affecting the poor prognosis after PKP in patients with thoracolumbar vertebral fractures (P < 0.05). ROC curve analysis demonstrated that the sensitivities of bone cement volume ≥ 6 mL, older age, vertebral compression degree ≥50%, bone cement leakage and their combination for predicting the poor prognosis after PKP in patients with thoracolumbar vertebral fractures were 66.2% (95% CI: 0.562, 0.762), 73.0% (95% CI: 0.620, 0.840), 75.7% (95% CI: 0.647, 0.867), 82.4% (95% CI: 0.714, 0.934) and 89.2% (95% CI: 0.728, 0.971), with the specificities being 75.0% (95% CI: 0.650, 0.850), 70.0% (95% CI: 0.590, 0.811), 65.0% (95% CI: 0.581, 0.729), 70.0% (95% CI: 0.627, 0.774) and 90.0% (95% CI: 0.829, 0.971). In addition, the combined detection of these indicators yielded an AUC of 0.905, with a high predictive value.Conclusions Bone cement volume ≥ 6 mL, vertebral wall destruction, vertebral compression degree ≥ 50%, and connections between the basivertebral foramen and the intravertebral cleft were the influencing factors for bone cement leakage after PKP in patients with thoracolumbar vertebral fractures. Besides, bone cement volume ≥ 6 mL, older age, vertebral compression degree ≥ 50%, and bone cement leakage were the influencing factors for poor prognosis after PKP in patients with thoracolumbar vertebral fractures. Early identification and correction of reversible factors are helpful to improve bone cement leakage and prognosis of patients with thoracolumbar vertebral fractures after PKP.