探讨改良手助后腹腔镜活体供肾切取术(HARPLDN)的学习曲线。

回顾性分析首都医科大学附属北京友谊医院2015年5月至2021年2月81例行改良HARPLDN供者临床资料，所有手术均由同一名腹腔镜手术经验丰富的泌尿外科医师主刀。收集供者入院时的基本信息和围手术期相关资料。术后并发症按改良的Clavien系统分型进行分级。通过移动平均法(MAM)定性分析改良HARPLDN学习过程中手术时间(OT)随例数的变化趋势。使用累积和分析法(CUSUM)定量评估学习曲线，依照CUSUM曲线顶点对应的横坐标数值将81例供者划分为第1阶段的初始学习阶段组以及第2阶段的掌握阶段组。使用Kolmogorov-Smirnov检验明确数据是否符合正态分布，组间计量资料采用独立样本t检验或Mann-Whitney U检验进行比较，计数资料采用卡方检验或Fisher确切概率法比较，P<0.05为差异有统计学意义。

MAM拟合的学习曲线提示，随着改良HARPLDN手术例数的累积，OT有下降的趋势；约完成至40例后，OT明显缩短，此后波动在150 min上下。采用CUSUM拟合学习曲线，第1~41例的CUSUM值处于快速上升阶段，即学习阶段组(n＝41)，第41例的CUSUM值达到顶点，此后CUSUM值开始下降，第42~81例纳入掌握阶段组(n＝40)。综合分析手术例数-CUSUM学习曲线，可得到拟合曲线y＝－166.163+4.6951×x+0.228×x²－0.004×x³，(R²＝0.679，P<0.001)。学习阶段组供者腹部手术史、术前血清肌酐、OT、热缺血时间、术中出血量和住院时间分别为3例、(72±12)μmol/L、(166±36)min、120 s(90~300 s)、50 mL(10~300 mL)和(12.7±2.4)d，掌握阶段组分别为10例、(62±12)μmol/L、(149±33)min、90 s(60~180 s)、50 mL(20~200 mL)和(12.4±3.5)d，差异均有统计学意义(χ²＝4.699，t＝3.750、2.190和0.503，Z＝－4.276和－2.569，P均<0.05)；其余临床资料差异均无统计学意义(P均>0.05)。

改良HARPLDN是一种安全、有效的供肾切取术式。对于具有丰富腹腔镜手术经验的泌尿外科医师来说，约41例即可完成HARPLDN的初始学习阶段。

To explore the learning curve of modified hand-assisted retroperitoneoscopic living donor nephrectomy (HARPLDN).

The clinical data of 81 consecutive donors who underwent modified HARPLDN from May 2015 to February 2021 were retrospectively evaluated. All procedures were performed by an experienced urologist in laparoscopic surgery. Baseline demographic, laboratory, and clinical perioperative data of the donors were collected. Postoperative complications were classified according to the modified Clavien classification system. The trend was analyzed qualitatively in the operation time (OT) varied with the number of cases during the learning process of modified HARPLDN, which is obtained by the moving average method (MAM). Then the learning curve was quantitatively evaluated by the cumulative sum analysis (CUSUM). According to the x-coordinate corresponding to the vertex of the CUSUM curve, 81 donors were divided into two phases, namely the initial learning phase group and the mastery phase group. The Kolmogorov-Smirnov test was used to confirm whether the data were normal distribution. To compare the initial learning phase group and the mastery phase group, the continuous variables were analyzed with an independent samples t-test or Mann-Whitney U test, and the categorical variables were analyzed with chi-square test or Fisher′s exact test. Statistical significance was defined as P value <0.05.

According to the MAM learning curve, the OT decreased as the number of modified HARPLDN cases increased. After about 40 cases, OT was significantly shortened and thereafter fluctuated around 150 min. According to the CUSUM curve, the CUSUM value increased rapidly in the first 40 donors and reached the peak at the 41st case, the first 41 cases were brought into the initial learning phase group (n=41), and then started to decline. The remain cases were brought into the mastery phase group (n=40). The cases of surgeries-CUSUM learning curve in chronological order was comprehensively analyzed, and the following equation was obtained: y=-166.163+ 4.6951×x+ 0.228×x²-0.004×x³; the highest R² value was 0.679, and P<0.001. In the initial learning phase group, the history of abdominal operation, preoperative serum creatinine, OT, warm ischemia time, intraoperative blood loss and length of stay were 3 cases, (72±12) μmol/L, (166±36) min, 120 s (90-300 s), 50 mL(10-300 mL) and (12.7±2.4) d, respectively; While in the mastery phase group were 10 cases, (62±12) μmol/L, (149±33) min, 90 s(60-180 s), 50 mL(20-200 mL) and (12.4±3.5) d. The differences were statistically significant (χ²=4.699, t=3.750, 2.190 and 0.503, Z=-4.276 and -2.569, all P<0.05). There were no significant differences between the two phases in terms of other aspects (all P>0.05).

The modified HARPLDN used in our hospital is a safe and effective method for kidney transplantation. For urologists with rich experience in laparoscopic surgery, 41 cases of the surgery were needed to finish the initial learning phase of the modified HARPLDN.