目的 比较开放前后路手术与混杂手术(微创极外侧入路+开放后路)治疗成人脊柱侧凸的并发症及临床疗效.方法 选取符合筛选标准的30例原发性成人脊柱侧凸手术患者,其中开放前后路手术18例(开放组),混杂手术12例(混杂组).开放前路融合需行胸廓切开或取胸腹联合入路,极外侧椎间融合术则采用可扩张通道辅助经腰大肌入路到达目标间隙.所有患者随访超过2年,统计手术时间、出血量和输血量,评估术前及末次随访影像学参数变化,记录脊柱侧凸研究学会(SRS)-22评分、Oswestry功能障碍指数(ODI)及围手术期并发症发生情况.结果 两组患者年龄、性别、术前侧凸严重程度及融合节段比较,差异无统计学意义.开放组出血量明显多于混杂组(3176 vs 2304 mL,P=0.04);混杂组手术时间相对较短(493 vs 528 min,P=0.057);两组术后影像学改善程度比较,差异无统计学意义(P>0.05).并发症方面,混杂组和开放组围手术期并发症分别为4、11例,远期并发症分别为10、12例,两组无统计学差异(P>0.05).其中开放组伤口感染3例,混杂组无伤口感染(P=0.26);开放组早期返修手术3例,混杂组无返修病例(P=0.26);开放组1例、混杂组3例出现假关节(P=0.27);近端交界性后凸(PJK)开放组7例、混杂组5例;末次随访两组患者SRS-22评分、ODI均较术前明显改善,但两组比较无统计学差异(P>0.05).结论 开放手术与混杂手术治疗成人脊柱侧凸皆能取得良好疗效;但是混杂手术能减少出血量,倾向于减少手术时间,在围手术期并发症、伤口感染及早期返修手术发生率方面呈现下降趋势,而假关节发生率似乎更高,2年随访时两种术式远期并发症发生率和临床结果无异.“,”Objectives To determine the differences of clinical outcome and complications between combined extreme lateral/posterior (X/P) approach and open anterior/posterior (A/P) approach for treatment of adult scoliosis. Methods Twelve patients who had X/P fusion for primary adult scoliosis were matched with a cohort of 18 patients who underwent A/P treatment. Open anterior fusion was performed via a thoracotomy/thoracoabdominal approach, and extreme lateral interbody fusion (XLIF) was performed using an expandable tubular retractor through a transpsoas access to the target disc space. All of the patients were followed up at least 2 years. Operating time, estimate blood loss (EBL) and transfusion volume were noted, radiographic parameters were evaluated before surgery and at fi nal follow-up, and Scoliosis Research Society (SRS)-22 score, Oswestry disability index (ODI) as well as complications were recorded. Results There were no significant differences between two groups for age, gender, diagnosis, preoperative curve magnitudes and fusion levels. EBL was statistically less in X/P than A/P group (2304 vs 3176 mL, P = 0.04), and there was a trend toward shorter surgical time (493 vs 528 min, P = 0.057). Postoperative radiographic coronal correction were almost the same between two groups. There were 14 complications including 4 perioperative and 10 late in X/P group, comparing to A/P group with 11 perioperative and 12 late complications, which showed no significant differences between two groups (P >0.05). Wound infection was found in 3 A/P patients while no X/P patients was infected (P =0.26). Early revision surgery was performed in 0 X/P patients and 3 A/P patients without significant difference (P = 0.26). Pseudarthrosis rates found in X/P versus A/P group (3/12 vs 1/18) were not significantly different (P =0.27). The occurrence rate of proximal junctional kyphosis (PJK) was the same (5/12 vs 7/18) in two groups. SRS-22 score and ODI in two groups reflected a similar patient assessment before surgery, and improvement at the last follow-up. Conclusions Combined X/P approach surgery in the treatment of adult scoliosis can achieve similar significant coronal correction to A/P surgery with decreased EBL, and a tendency of shorter surgical time. It also presented a declining trend in the rates of perioperative complications, wound infection and requirement for early reoperation. There was a tendency of higher lumbosacral pseudarthrosis rate in X/P group. Late complications and clinical outcomes were similar at over 2-year follow-up.