Clinical impact of Endoscopic Surgical Skill Qualification System (ESSQS) by Japan Society for Endoscopic Surgery (JSES) for laparoscopic distal gastrectomy and low anterior resection based on the National Clinical Database (NCD) registry

Abstract Aim This study aimed to evaluate the association between surgeons certified via the Endoscopic Surgical Skill Qualification System (ESSQS) of the Japan Society for Endoscopic Surgery (JSES) and surgical outcomes of laparoscopic distal gastrectomy (LDG) and laparoscopic low anterior resection (LLAR). Methods Japanese National Clinical Database data on the patients undergoing LDG and LLAR between 2014‐2016 were analyzed retrospectively. The proportion of cases performed by ESSQS‐certified surgeons was calculated for each procedure, and clinicopathological factors with or without participation of ESSQS‐certified surgeons as an operator were assessed. Then, effects of operations performed by ESSQS‐certified surgeons on short‐term patient outcomes were analyzed using generalized estimating equations logistic regression analysis. Results There were 110 610 and 65 717 patients who underwent LDG and LLAR, respectively. The operations performed by ESSQS‐certified surgeons in each procedure totaled 28 467 (35.3%) and 12 866 (31.2%), respectively. A multivariable logistic regression model showed that odds ratios of mortality for LDG and LLAR performed by ESSQS‐certified surgeons were 0.774 (95% CI, 0.566‐1.060, P = 0.108) and 0.977 (0.591‐1.301, P = 0.514), respectively. Odds ratios for secondary endpoints of anastomotic leakage in LDG and LLAR performed by ESSQS‐certified surgeons were 0.835 (95% CI, 0.723‐0.964, P = 0.014) and 0.929 (0.860‐1.003, P = 0.059), respectively, whereas that of ileus/bowel obstruction for LLAR performed by ESSQS‐certified surgeons was 1.265 (1.132‐1.415, P < 0.001). There were no significant associations between the two operations performed by ESSQS‐certified surgeons and other factors such as mortality and overall complications. Conclusions ESSQS certification did not affect postoperative mortality following LDG and LLAR, but annual experience of laparoscopic surgery was associated with it. ESSQS certification may contribute to favorable outcomes regarding anastomotic leakage following LDG and LLAR.


| INTRODUC TI ON
The use of laparoscopic surgery has continued to increase worldwide ever since laparoscopic distal gastrectomy (LDG) for gastric cancer was introduced in 1994. 1 As a result of previous large randomized studies, laparoscopic surgery has now become a standard surgical treatment for cStage I gastric cancer. [2][3][4] Based on other randomized clinical trials, as with LDG, laparoscopic low anterior resection (LLAR) has also spread worldwide. [5][6][7] The National Survey of Endoscopic Surgery conducted by the Japan Society for Endoscopic Surgery (JSES) found that more than 75% of the low anterior resections performed in Japan in 2017 were done laparoscopically. 8 It is noteworthy, however, that these important data on the safety and superiority of laparoscopic surgery were mainly obtained from high-volume centers or hospitals specializing in gastric and rectal sur- Thus, in the present study, a large-scale retrospective cohort analysis of patient data in the Japanese NCD database was performed to evaluate the effectiveness of ESSQS by JSES for LDG and LLAR, which have developed into common and standardized operations in terms of lymph node dissection and reconstruction.

| ESSQS by JSES
The JSES established the ESSQS and began certification examinations in 2004. [9][10][11] Two judges assessed unedited videotapes of a procedure in a double-blinded fashion according to strict common and procedure-specific criteria. The common criteria were designed to evaluate the surgical set-up, operator autonomy, display of the surgical field, correct recognition of surgical anatomy, and surgical team co-operation. Procedure-specific criteria were used to assess the surgical procedure in a step-by-step manner. Of the 1114 surgeons assessed via this qualification system over a 4-year period, 537 (48.2%) were accredited. To date, the rate of qualification in each surgical field continues to remain at a similar level. During the first year, the inter-rater agreement of the two judges was low at 0.31, but with revisions of the criteria and the use of consensus meetings, agreement has improved. The surgeons assessed as being qualified by this ESSQS-certified surgeons on short-term patient outcomes were analyzed using generalized estimating equations logistic regression analysis.
Conclusions: ESSQS certification did not affect postoperative mortality following LDG and LLAR, but annual experience of laparoscopic surgery was associated with it. ESSQS certification may contribute to favorable outcomes regarding anastomotic leakage following LDG and LLAR.

K E Y W O R D S
endoscopic surgical skill qualification system, laparoscopic distal gastrectomy, laparoscopic low anterior resection, National Clinical Database, short-term outcome system experience less frequent complications compared with those who failed the assessment. This system has positively improved and brought standardization to laparoscopic surgery in Japan. 10-12

| NCD registration
Details of the registration of data in the Japanese NCD system were described previously. 15,16 Briefly, the NCD began as a nationwide registry system in Japan that was linked with the surgical board certification system in 2011. As of 2018, over 5000 institutions have participated in this system, with approximately 1 400 000 surgical cases being registered annually.
The Japanese Society of Gastroenterological Surgery (JSGS) designated eight main surgical procedures in the gastroenterological section of the NCD as being especially important in terms of medical standards for improving surgical quality: esophagectomy, distal gastrectomy, total gastrectomy, right hemicolectomy, low anterior resection, hepatectomy, pancreaticoduodenectomy, and surgery for acute diffuse peritonitis. All surgical cases are registered in the NCD, and details including morbidities, comorbidities, postoperative complications, and mortalities are input into the system.

| Study population
In total, 110 610 cases of LDG and 65 717 cases of LLAR were registered in the NCD between 2014 and 2016. Among them, benign disease, malignant disease of organs other than the stomach and rectum, open surgery, and emergent surgeries were excluded from this study. Also excluded were 24 cases of LDG and 21 cases of LLAR with data deficits (Figure 1).

| Study endpoints
The primary endpoint of the study was mortality (within 30 days after surgery or within 90 days in hospital, respectively). The To determine a trend in operation volumes in terms of various outcomes, we used cut-off values that divided the study population into three parts, and we created three groups. Finally, the impact of the operations performed by ESSQS-certified surgeons on postoperative complications was evaluated by multivariable regression modeling.

| Statistical analysis
We used the χ 2 test for statistical comparisons between groups. We accounted for clustering at the hospital level by use of generalized Additionally, a comparative analysis was conducted between groups in terms of operation time, intraoperative blood loss, and the necessity of transfusion. A two-sided probability level < 0.05 was considered to indicate a significant difference. The software package R version 3.6.0 (2019; R Foundation for Statistical Computing, Vienna, Austria) was used for statistical analysis. The study protocol was approved by the institutional review board of Oita University (approval number B190275).

| Surgical outcomes
The operative outcomes are shown in Tables 2A and 2B (Table 2A: Laparoscopic distal gastrectomy).  (Table 3A:  There was no correlation between the operations performed by ESSQS-certified surgeons and mortality. The regression analysis showed that a higher laparoscopic operation volume per year (>42) was an independent predictor of mortality, overall

| D ISCUSS I ON
To the best of our knowledge, the present study is the first and largest to investigate the effectiveness of ESSQS for LDG and LLAR through a large-scale retrospective cohort analysis of patient data in the Japanese NCD. Although the present study did not reveal a contribution of ESSQS to lower mortality, it did show that ESSQS might contribute to a favorable outcome in terms of anastomotic leakage following LDG and LLAR. In terms of the primary endpoint of the present study, the mortality from LDG and LLAR was not significantly affected regardless of whether non-ESSQ-or ESSQ-certified surgeons performed the procedures. We speculated that the very low mortality rate for LDG and LLAR might have obscured the impact of ESSQ between the two procedures. Although the P value for the incidence of anastomotic leakage with LLAR in the ESSQS-certified surgeons group was 0.059 by multivariate analysis, we thought that at least the ESSQS-certified surgeons group showed a tendency for a lower incidence of anastomotic leakage, which was considered a meaningful indicator clinically.
In the 8-year period since the NCD was initiated in 2011, over 5 million cases from more than 4200 facilities have been registered. The efforts of Japanese surgeons have clearly resulted in the successful establishment of this nationwide surgical database.
Risk models for 30-day and operative mortality for several procedures were created, and outcomes for each procedure have been retrospectively published. 18    There are several limitations in the present study. First, it is a retrospective, observational study. Potential bias due to heterogeneity of the surgical quality or performance by each hospital cannot be excluded. Second, this analysis did not overcome all of the uncertainties associated with the details of the surgical procedures, such as the degree of lymphadenectomy, methods or technique of reconstruction, type of approach (e.g. robotic surgery, single-port surgery, or transanal total mesorectal excision), or types of energy devices used. Third, the oncological and long-term outcomes are not available from the Japanese NCD. Only data regarding baseline characteristics and short-term outcomes were allowed to be input.
A definitive conclusion as to the oncological validity of this surgery must also depend on the data from other clinical trials. Fourth, the clinical impact of operations in which ESSQS-certified surgeons participated as an assistant could not be evaluated.
In conclusion, ESSQS certification did not affect postoperative mortality following LDG and LLAR, but annual experience of laparoscopic surgery was associated with it. ESSQS certification may contribute to favorable outcomes regarding anastomotic leakage following LDG and LLAR.

D I SCLOS U R E
Funding: Author YK received lecture fees and was supported by