成人退行性脊柱侧凸后路长节段腰椎体间融合术后发生近端交界性后凸的危险因素

Risk factors for proximal kyphosis after long segmental posterior lumbar interbody fusion of adult degenerative scoliosis

目的:分析成人退行性脊柱侧凸(ADS)经长节段后路腰椎体间融合术(PLIF)后近端交界性后凸(PJK)的发生率和相关危险因素。方法:回顾性分析2013年6月~2016年1月在我院行长节段PLIF治疗的144例ADS患者,术后43例发生PJK(PJK组),无PJK组101例。收集两组患者基本资料和临床功能评分,包括年龄、性别、随访时间、体重指数(BMI),术前合并贫血、糖尿病、高血压病和心脏病情况,吸烟情况,既往手术史,住院时间,术前与末次随访时的Oswestry功能障碍指数(Oswestry disability index,ODI)、JOA评分、VAS评分和末次随访时的腰椎僵硬性残疾指数(lumbar stiffness disability index,LSDI)评分;与手术相关的参数,包括术中出血量、手术用时、融合节段、减压节段、骶骨融合、后方韧带复合体(PLC)损伤、截骨术、螺钉松动、断棒、脑脊液漏、假关节形成、再手术、术前与末次随访Cobb角、术前骨盆入射角(PI)-腰椎前凸角(LL)、末次随访时的PI-LL、术前与末次随访时的矢状面垂直轴(SVA)。对比两组术后即刻影像学参数,包括骨盆倾斜角(PT)、PI、骶骨倾斜角(SS)、LL、PI-LL、胸椎后凸角(TK)、LL-TK、实际腰椎前凸角(RLL)、实际胸椎后凸角(RTK)、实际腰椎前凸角与实际胸椎后凸角匹配值(RLL-RTK)、上端椎坡度(UIVs)、上端椎倾斜角(UIVi)、SVA、UIV与C7垂线距离(UIVa)与UIV上终板前缘与椎体后缘垂线距离(UIVb)的比值(UIVa/b)。组间连续变量比较采用One-Way ANOVA检验,然后再用SNK法进行多重检验;分类变量采用χ2检验,而非连续数据的统计学显著性通过Pearson卡方检验进行比较。以可疑危险因素为自变量,以是否出现PJK作为因变量对其进行单因素分析,然后将有统计学意义的自变量代入Logistics模型进行多因素回归分析,最终确定PJK相关危险因素。Pearson相关检验(r)用于评估骨盆与脊柱参数的相关性。结果:患者一般资料中,性别、BMI、糖尿病、吸烟、末次随访ODI和LSDI评分两组间差异有统计学意义(P<0.05)。在手术相关参数中,术中出血量、融合节段数、减压节段数、骶骨融合、PLC损伤、再手术率、术前PI-LL、末次随访PI-LL和末次随访SVA两组比较有统计学差异(P<0.05)。术后即刻影像学参数中,PT、SS、LL、PI-LL、LL-TK、RLL、RLL-RTK、UIVs、UIVi和UIVa/b两组比较有统计学差异(P<0.05)。将组间比较结果有统计学差异的参数进行单因素分析后有意义的参数包括性别、BMI、糖尿病、吸烟、LL-TK、UIVs、UIVa/b、PI-LL、骶骨融合和PLC损伤。再经多因素Logistics回归分析,结果显示ADS术后发生PJK的独立危险因素包括BMI(OR 1.6,P=0.000)、吸烟(OR 3.1,P=0.016)、UIVs(OR 2.0,P=0.015)和PI-LL(OR 1.2,P=0.015)。在脊柱骨盆参数Pearson相关系数分析中发现,SVA与UIVa/b、UIVs和UIVi之间存在较高的相关性(r=-0.93,P<0.001)、(r=-0.66,P<0.001)和(r=-0.68,P<0.001);UIVa/b与UIVs及UIVi之间存在较高的相关性(r=-0.63,P<0.001)和(r=0.-68,P<0.001)。结论:接受长节段PLIF的ADS患者术后发生PJK的独立危险因素为BMI、吸烟、UIVs和PI-LL。UIVa/b可以反映SVA情况和UIV倾斜情况,可作为发生PJK的预测因子。

Objectives:To analyze the incidence of PJK and related risk factors after posterior lumbar interbody fusion(PLIF)in adult degenerative scoliosis.Methods:A retrospective analysis was performed on 144 patients with adult degenerative lumbar scoliosis treated with long segmental PLIF from June 2013 to January 2016.There were PJK group with 43 cases and non-PJK group with 101 cases.Collected patient basic data and clinical function scores included age,gender,follow-up time,BMI,preoperative anemia,diabetes,hypertension history,smoking status,heart disease complications,previous surgical history,hospital stay,preoperative Oswestry disability index(ODI),last follow-up ODI,preoperative JOA score,last follow-up JOA score,preoperative VAS score,last follow-up VAS score and last follow-up lumbar stiffness disability index(LSDI)score.Surgery related parameters included intraoperative bleeding,surgical time,fusion segments,decompression segments,fusion to sacrum,posterior ligament complex(PLC),osteotomy,screw loosening,broken rod,cerebrospinal fluid leakage,pseudarthrosis,reoperation,preoperative Cobb angle,last follow-up Cobb angle,preoperative pelvic incidence minus lumbar lordosis mismatch(PI-LL),last follow-up PI-LL,preoperative sagittal vertical axis(SVA)and last follow-up SVA.The immediate imaging parameters of the two groups included pelvic tilt(PT),PI,sacral slope(SS),LL,PI-LL,thoracic kyphosis(TK),lumbar lordosis minus thoracic kyphosis mismatch(LL-TK),real lumbar lordosis(RLL),real thoracic kyphosis(RTK),real lumbar lordosisminus really thoracic kyphosis mismatch(RLL-RTK),upper instrumented vertebra slope(UIVs),UIV tilt angle(UIVi),SVA and C7 perpendicular distance to UIV upper end plate anterior point divided by posterior distance to UIV upper end plate anterior point(UIVa/b).The Pearson correlation test(r)was used to assess different correlations among PI,PT,SS,LL,TK,LL-TK,RLL-RTK,SVA,UIVs,UIVi and UIVa/b immediately after surgery.Results:There were significant differences in gender,BMI,diabetes,smoking,ODI and LSDI score between the two groups(P<0.05).In the operation-related parameters,intraoperative blood lost,fusion segments decompression segments,S1 fusion,PLC,reoperation,preoperative PI-LL,last follow-up PI-LL and last follow-up SVA were statistically different.PT,PI-LL and UIVa/b values immediately after surgery were significantly higher in the non-PJK group(P<0.05);but SS,LL,LL-TK,RLL,RLL-RTK,UIVs and UIVi immediately after surgery were significantly higher in PJK group(P<0.05).The multivariate Logistic regression model showed perioperative independent risk factors for PJK after long-segment PLIF in adult spinal deformity surgery including:BMI(OR 1.6,P=0.000),smoking(OR 3.1,P=0.016),and UIV slope(OR=2.0,P=0.015)and PI-LL(OR=1.2,P=0.015).There was a high correlation between SVA and UIVa/b,UIV slope and UIV tilt(r=-0.93;P<0.001),(r=-0.66;P<0.001)and(r=-0.68;P<0.001).There was a high correlation between UIVa/b and UIV slope and UIV tilt(r=-0.63;P<0.001)and(r=-0.68;P<0.001).Conclusions:In this study,the independent risk factors for PJK after posterior lumbar interbody fusion with adult degenerative scoliosis were analyzed by multivariate logistic regression,smoking,UIV slope and PI-LL.UIVa/b can reflect the SVA situation and UIV tilt,which can be used as a predictor of PJK.