Evaluation of the Learning Curve for Laparoscopic Pyloromyotomy

Background: Hypertrophic pyloric stenosis is one of the most common surgical diseases of the infant. Surgical treatment is via a pyloromyotomy, and laparoscopy has become the preferred surgical approach in most pediatric surgical centers. The goal of this study was to analyze the learning curve for laparoscopic pyloromyotomy using the experience of five fellowship-trained pediatric surgeons who started their surgical practice at the same institution.

Methods: A retrospective review of consecutive patients undergoing laparoscopic pyloromyotomy between 2008 and 2020 was performed. All cases were performed with general surgery residents, not pediatric surgery fellows. Demographic, preoperative, intraoperative, and postoperative data were collected. Cumulative sum (CUSUM) analysis for operating time was performed for the first 150 consecutive cases, using the average operating time for surgeons as the dependent variable. Outcomes were then compared across the different phases of learning.

Results: A total of 414 patients were included in the analysis, as not all five surgeons have reached 150 cases. The mean age was 38 days, and the majority of patients were male (n=346, 84%). Median length and thickness of the pylorus, as measured on preoperative ultrasound, was 19 mm and 4.2 mm, respectively. 409 (98.8%) cases were completed laparoscopically; of the 5 conversions to open, 4 were due to perforation and 1 was due to difficulty completing the myotomy laparoscopically. Postoperatively, 21 (5.1%) patients experience a complication and there were 9 (2.2%) reoperations. The most common complication was omental evisceration (n=7, 1.7%). 2 (0.5%) patients had an incomplete myotomy and there were 5 (1.2%) surgical site infections. The mean length of postoperative stay was 1.3 days and there were 9 (2.2%) readmissions. The mean operating time over the study period was 29.2 minutes, with the raw data shown in Figure 1. Figure 2 shows the CUSUM analysis for mean operating time, with three phases of learning identified by inflection points in the curve: Learning phase (cases 1-16), Plateau phase (cases 17-87), and a Proficiency phase (cases 88-150). The mean operating time during the three phases was 34.1, 29.0, and 29.2 minutes, respectively (P=0.005). There were no differences in complication rates, reoperations, length of stay, or readmissions across the three phases of learning.

Conclusion: Using CUSUM analysis for operative time, we identified three distinct phases of learning for laparoscopic pyloromyotomy, with no differences in complication rates or outcomes across the three phases. The operating time differed only during the learning phase, suggesting some degree of proficiency likely occurs after about 16 cases. To our knowledge, this is the first learning curve analysis done for laparoscopic pyloromyotomy and may inform decisions on fellowship training and mentorship during early surgical practice.