摘要目的:探讨Mako机器人辅助直接前侧入路全髋关节置换术（DAA-THA）的学习曲线及短期临床疗效。方法:回顾性分析2018年12月至2020年12月在上海交通大学附属第六人民医院骨科进行的50例Mako机器人辅助下DAA-THA的患者术前基本资料、手术信息及术后功能情况，如手术时间、术中出血量、术后并发症、术后影像学参数（外展角、前倾角、下肢长度差、偏心距差）及术后Harris髋关节评分（HHS）。其中男16例，女34例；年龄50~79（67±10）岁。分析总结Mako机器人辅助全髋关节置换术的术后临床结果。以累积和分析法（CUSUM）分析手术时间。对CUSUM学习曲线进行拟合，以R2判断拟合优度。比较学习曲线不同阶段各观察指标的差异。结果:50例患者均得到6个月以上的随访，2例出现术后伤口愈合不良，5例出现股外侧皮神经损伤症状，表现为伤口周围的麻木感，均在1~2个月内消退。所有患者均未发生假体脱位、无菌性松动、假体周围感染或翻修等严重不良事件。患者手术时间为（81±16）min，术中出血量为（456±84）ml，术后末次随访时的Harris髋关节功能评分为（88.6±2.5）分。影像学结果显示髋臼杯位置均处于Lewinnnek安全区；下肢长度差为（0.15±0.50）cm，偏心距差为（-0.11±0.72）cm。手术时间随手术例数累积呈逐渐下降趋势。学习曲线最佳拟合为三次方曲线，拟合曲线在手术例数累积至第19例时达到顶点，以此为分界将学习曲线划分为两个阶段，其手术总时间、骨盆参考架安装时间、髋臼注册时间、髋臼磨锉时间差异均有统计学意义（均 P<0.05），但两组间Harris评分、髋臼假体前倾角和外展角差异均无统计学意义（均 P>0.05）。 结论:Mako机器人辅助DAA-THA的学习曲线在19例左右，Mako机器人辅助能确保DAA-THA假体植入精确性及学习曲线期间手术的安全性，术后短期临床效果优异。

abstractsObjective:To explore the learning curve and short-term clinical outcomes of Mako robotic-assisted direct anterior approach total hip arthroplasty (THA).Methods:The preoperative basic data, surgical information and postoperative rehabilitation of 50 patients who underwent Mako robotic-assisted THA for hip diseases in Department of Orthopedic Surgery of the 6th People′s Hospital Affiliated to Shanghai Jiao Tong University from December 2018 to December 2020 were analyzed retrospectively, included operation time, intraoperative blood loss, postoperative complications, postoperative imaging parameters (abduction angle, anteversion angle, lower limb length difference, eccentricity difference) and postoperative hip joint Harris score (hip Harris score, HHS). There were 16 males and 34 females, with a mean age of 50-79(67±10) years. The postoperative clinical results of Mako robotic-assisted total hip arthroplasty was analyzed. A cumulative sum analysis (CUSUM) was performed on the operation time (OT). The CUSUM learning curve was modeled by curve fitting and R2 was used to testify the goodness. The different phase of the learning curve was compared with several observation indicators.Results:All patients were followed up for more than 6 months. Two patients had poor wound healing and 5 patients had symptoms of lateral femoral cutaneous nerve injury, which disappeared within 1-2 months. No serious complications such as dislocation, aseptic loosening, periprosthetic infection or revision occurred in all the patients. The average operation time was (81±16) min, and the intraoperative blood loss was (456±84) ml. The average Harris hip score at the last follow-up was 88.6±2.5. The radiological evaluation showed that the positions of the acetabular cups were all in the Lewinnnek safety zone; the limb length discrepancy was (0.15±0.50) cm, the offset was (-0.11±0.72) cm. The OT decreased with the accumulation of the cases. The CUSUM learning curve was best modeled as cubic curve,the fitting curve reached the top at the 19th case. As a cut-off point, the 19th point divided the learing curve into two phases. There were statistical differences in OT, pelvic array installation time, acetabular registration time, acetabular reaming time (all P<0.05), but there was no significant differences in Harris hip score, acetabular prosthesis anteversion angle and abduction angle between the two groups (all P>0.05). Conclusions:The learning curve of Mako robot-assisted DAA-THA is about 19 cases. Mako robot-assisted DAA-THA can ensure the accuracy of prosthetic placement and the safety of the operation during the learning curve, and the short-term clinical results after surgery is excellent.