退变性腰椎侧凸畸形出现冠状位一致性脊柱序列的原因及矫形术后失平衡的危险因素分析

The causes for consistent coronal alignment and the risk factors for coronal imbalance after corrective surgery in degenerative lumbar scoliosis

中文关键词:  退变性腰椎侧凸  冠状位失平衡  冠状位平衡距  相关因素  观测者操作特征曲线

英文关键词:Degenerative lumbar scoliosis  Coronal imbalance  Coronal balance distance  Correlated factors  Receiver operating characteristic curve

中文摘要:

  【摘要】 目的：探讨退变性腰椎侧凸（degenerative lumbar scoliosis，DLS）患者冠状位躯干偏移与侧凸同向性（冠状位一致性）脊柱序列的相关因素及矫形术后早期冠状位失平衡的危险因素。方法：纳入2015年5月~2020年1月在解放军总医院单一中心手术治疗的75例DLS患者，均行后路截骨矫形长节段固定融合术（固定椎体≥5）。根据术前C7铅垂线（C7 plumb line，C7PL）偏移与侧凸方向的关系，将患者分为两型：Ⅰ型，C7PL与侧凸同向；Ⅱ型，C7PL与侧凸相悖。定义C7PL到骶骨中点的距离为冠状位平衡距（coronal balance distance，CBD），CBD≥30mm即为冠状位失平衡（coronal imbalance，CIB）。根据术后CBD将Ⅰ型患者分为A组（CBD≥30mm）和B组（CBD<30mm）。记录所有患者性别、年龄及体重指数（body mass index，BMI），手术前后影像学参数包括CBD、主弯Cobb角（major curve Cobb，MCC）、腰骶弯Cobb角（fractional Cobb，FC）、L4及L5椎体倾斜度、顶椎侧方滑移度（滑向凸侧为+；滑向凹侧为-）、主弯累及椎体数；手术参数包括主弯矫正度及其矫正率、固定椎体数、上端固定椎（upper instrumented vertebra，UIV）及下端固定椎（lower instrumented vertebra，LIV）。应用观测者操作特征（receiver operating characteristic，ROC）曲线分析Ⅰ型患者顶椎侧方滑移度临界值及Ⅰ型患者主弯矫正率，根据曲线下面积（the area under ROC curve，AUC）得出相应临界值，并计算95%可信区间（confidence interval，CI）。组间正态分布参数采用独立样本t检验，非正态分布采用Mann-Whitney U检验。定性变量则采用卡方检验或Fisher检验，并计算相应比值比（odds ratio，OR）。结果：75例DLS患者中男15例，女60例；年龄62.93±8.42岁。Ⅰ型和Ⅱ型冠状位序列分别有33例和42例，其中Ⅰ型术前CIB患者12例、术后12例；Ⅱ型患者术前CIB有5例、术后6例，Ⅰ型患者手术前后CIB发生率均显著大于Ⅱ型患者（P<0.05）。Ⅰ型患者术前CBD（P=0.01）、顶椎侧方滑移度（P<0.001）及L4椎体倾斜度（P=0.015）显著大于Ⅱ型患者，其余均无异常（P>0.05）。ROC曲线分析结果显示Ⅰ型患者顶椎侧方滑移度临界值为6.5mm，敏感度=91.7%，特异度=85.7%；AUC=0.903，95%CI：0.793~1.000。Ⅰ型DLS患者分为A组12例、B组21例。A组患者术前顶椎侧方滑移度（P=0.037）、L4椎体倾斜度（P=0.001）及L5椎体倾斜度（P=0.038）均显著大于B组；两组患者固定节段、UIV及LIV均无统计学差异，但A组患者侧凸矫正度（P=0.001）及其矫正率（P<0.001）均显著大于B组。ROC曲线分析结果显示Ⅰ型患者主弯矫正率临界值为65.0%，敏感度=75.0%，特异度=81.0%；AUC=0.861，95%CI：0.739~0.984；即Ⅰ型患者主弯矫正率超过65%，术后CIB发生率显著增加（OR：9.6；95%CI：1.847~49.884；P=0.009）。结论：DLS患者术前CBD及主弯顶椎椎体向凸侧滑移可能为Ⅰ型冠状位序列的重要相关因素，且顶椎侧方滑移度≥6.5mm将导致此型患者发生术前CIB。此外，Ⅰ型DLS患者L4/L5椎体倾斜度越大，主弯矫正度过大，则术后即发生CIB几率越大，且主弯矫正率超过65%时，术后CIB发生率显著增加。

英文摘要:

  【Abstract】 Objectives: To explore the parameters associated with trunk shifting and scoliosis in the same direction-consistent coronal alignment(CA), and to investigate the risk factors associated with early postoperative coronal imbalance(CIB) after corrective surgery in patients with degenerative lumbar scoliosis(DLS). Methods: A total of 75 DLS patients underwent the posterior osteotomy and orthopedics surgery of long-segment fusion(fixed vertebras≥5) with instrumentations from May 2015 to January 2020 were enrolled. According to relations of the C7 plumb line(C7PL) and major curve direction, the patients were divided into type Ⅰ(C7PL locating at the convex side of the major curve) and type Ⅱ(C7PL locating at the concave side of the major curve). The distance from C7PL to the midpoint of S1 was defined as the coronal balance distance(CBD), and coronal imbalance(CIB) was concerned if the CBD≥30mm. All the patients with type Ⅰ CIB were divided into group A(CBD≥30mm) and group B(CBD<30mm) according to the CBD postoperatively. The age, gender, and body mass index(BMI); radiographic parameters before and after surgery, including CBD, major curve Cobb(MCC), lumbosacral fractional Cobb(FC), L4 and L5 coronal tilt, coronal apical lateralisthesis degree(proximal vertebra sliding to the convex side was recorded as +； if not as -), and the vertebras including in the major curve; and surgical data including major curve correction degree and correction rate, number of fixed vertebras, upper instrumented vertebra(UIV), and lower instrumented vertebra(LIV) were recorded. The apical lateralisthesis degree and the MCC correction in patients with the type I CIB were analyzed using the receiver operating characteristic(ROC) curve analysis, the best cutoff value and 95% confidence interval(CI) were obtained by the area under the curve(AUC). The normal distribution parameters were analyzed using the independent sample t test, the Mann-Whitney U test was for non-normal distribution data, and qualitative data were tested using chi-square test or Fisher test, and the odds ratio (OR) was calculated subsequently. Results: The 75 DLS patients consisted 60 females and 15 males, averaged 62.93±8.42 years. There were 33 and 42 patients in type Ⅰ and Ⅱ respectively, and among them, 12 typeⅠ patients suffered from CIB before operation and 12 after operation, while 5 type Ⅱ patients suffered from CIB before operation and 6 after operation. The incidence of CIB in type Ⅰ was significantly higher than that in type Ⅱ(P<0.05). Preoperative CBD(P=0.01), the apical lateralisthesis degree(P<0.001) and L4 tilt(P=0.015) of type Ⅰ patients were much bigger than those of type Ⅱ patients. ROC analysis showed that the best cutoff value for the apical lateralisthesis was 6.5mm, the sensitivity=91.7%, specificity=85.7%; AUC=0.903, 95%CI: 0.793-1.000. There were 12 and 21 patients in groups A and B respectively. The preoperative apical lateralisthesis degree(P=0.037), L4 tilt(P=0.001), and L5 tilt(P=0.038) in group A were much greater than those in group B. There were no significant differences in number of fixed segments, UIV and LIV between groups, however, the CIB correction degree(P=0.001) and correction rate(P<0.001) in group A were much more than those in group B. ROC curve analysis showed that the best cutoff value of major curve correction rate of type Ⅰ patients was 65%, the sensitivity=75.0%, specificity=81.0%; AUC=0.861, 95%CI: 0.739-0.984. The correction rate of MCC over 65% in type Ⅰ patients, the incidence of postoperative CIB increased significantly(OR: 9.6; 95%CI: 1.847-49.884; P=0.009). Conclusions: In DLS patients, the preoperative CBD and the apical vertebra gliding towards convex side may correlate significantly with type Ⅰcoronal alignment, and a slippage of ≥6.5mm will lead to coronal imbalance. In addition, the greater L4/L5 tilt preoperatively and overcorrection of MCC in patients with type Ⅰmay result in bigger posibility of postoperative CIB, and the MCC correction rate surpassing 65% may increase postoperative CIB significantly.