The Influence of Surgeons' Gender on Surgical Outcomes in Laparoscopic Right Hemicolectomy: A Japanese Nationwide Retrospective Cohort Study.

1
Introduction
The proportion of female physicians has increased worldwide, reaching 50% in OECD countries in 2021 [1]. Although this trend is also increasing in Japan, the rate of female physicians has barely reached 23.6%, which is one of the lowest rates in OECD countries [1]. The rapid aging and reduction of surgeons in the field of gastroenterological surgery has become a serious problem. Very recently, the Japanese Society of Gastroenterological Surgery (JSGS) predicted that the number of gastroenterological surgeons will halve within 20 years, potentially affecting the medical system within a decade [2]. Furthermore, the rate of female surgeons in this field is only 8.0% [3], which is significantly lower than that in other Western industrialized countries (e.g., 22.0% in the United States) [4]. Increasing the number and participation of female surgeons is crucial for the survival and advancement of gastroenterological surgery in Japan. Despite this necessity, Japan lags in terms of gender equality within the surgical field. A Japan Surgical Society survey revealed that female surgeons reported higher perceived work‐ and promotion‐related discrimination than did male surgeons. Additionally, approximately one‐fifth of respondents believed that female surgeons generally underperform in the profession, revealing a major unconscious bias [5]. Consequently, there is an urgent need to recognize and eliminate these unconscious biases against female surgeons as well as to improve their working environments [6].
Recently, several studies comparing the surgical outcomes of male and female surgeons have emerged. In a study from Canada, Wallis et al. reported a slight but significant reduction in surgical 30‐day mortality in combination with 25 selected procedures performed by female surgeons compared with those performed by male surgeons (adjusted odds ratio (OR); 0.88) [7]. Although no studies have demonstrated inferior outcomes with female compared with male surgeons, it is worth noting that there are limitations in mixing and analyzing miscellaneous surgical procedures.
In the colorectal cancer (CRC) surgical field, Okoshi et al. were the first to use the Japanese National Clinical Database (NCD) to demonstrate no significant differences between male and female surgeons in the adjusted risks of surgical mortality and surgical mortality with severe morbidity in low anterior resection (in conjunction with distal and total gastrectomy) [8]. Conversely, right hemicolectomy, which is a typical CRC surgical procedure traditionally considered to be relatively safe, was found to have higher 30‐day postoperative mortality and surgery‐related mortality than low anterior resection in the Japanese NCD data (1.4% vs. 0.3% and 2.2% vs. 0.5%, respectively) [9]. Laparoscopic surgery for colon cancer is a common and accepted surgical approach in Japan; laparoscopic right hemicolectomy could therefore be a useful procedure with which to evaluate safety and surgical outcomes in the CRC surgical field.
This study aimed to compare surgical mortality and severe complications in colon cancer patients undergoing laparoscopic right hemicolectomy between male and female surgeons using NCD data. We hypothesized a priori that equivalent surgical outcomes would be observed and that the results could support the choice of surgical careers for women.

2
Methods
2.1
Data Source and Patients
This study was performed by analyzing essential data extracted from the Japanese NCD, which is a web‐based, data‐entry system linked to the surgical board certification system. Initiated in 2011, the NCD now covers more than 97% of all surgical procedures in Japan [10]. All surgical cases are registered in the database, which includes details such as morbidities, comorbidities, postoperative complications, and mortality. The eligibility criteria for the NCD are accessible online (http://www.ncd.or.jp/). Annual site visits and audits assess the NCD data, finding it to be of high quality [11]. This retrospective cohort study included adult patients (≥ 18 years) registered in the NCD between January 1, 2018 and December 31, 2022 who had primary colon carcinoma and had undergone laparoscopic right hemicolectomy. Patients were excluded from this study if they had missing data, received emergency surgery, or if their surgery was performed by non‐JSGS members. Information about surgeons' self‐reported gender and the number of years since they had registered as licensed doctors are unavailable from the NCD data, and were obtained from the JSGS members' profile data. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline [12].

2.2
Outcomes
The main outcome measures of this study were surgical mortality and severe postoperative complications within 30 days after surgery. Surgical mortality is defined as all‐cause death up to 30 days after surgery, including death after discharge or in‐hospital mortality during the index admission within 90 days of surgery. Secondary outcome measures were reoperation, anastomotic leakage, and pneumonia. Postoperative complications were categorized according to the Clavien‐Dindo (CD) grading [13], and severe complications were defined as CD grade III or more. Operating time and estimated blood loss during surgery are presented as crude intraoperative outcomes.

2.3
Clinical Factors
The clinical factors included age at time of surgery (< 65, ≥ 65 < 75, and ≥ 75 years), sex (male or female), body mass index (BMI; < 18.5, ≥ 18.5 < 25, and ≥ 25 kg/m2), diabetes mellitus, smoking history, habitual drinking, dependence in activity of daily living, chronic obstructive pulmonary disease, need for preoperative dialysis, history of ischemic heart disease, congestive heart failure within 30 days, long‐term steroid use, previous cerebrovascular disease, weight loss, preoperative blood transfusion, hemoglobin (male: < 13.5, ≥ 13.5 g/dL, female; < 11.5, ≥ 11.5 g/dL), albumin (< 3.5, ≥ 3.5 g/dL), blood urea nitrogen (< 8.0, ≥ 8.0 mg/dL), creatinine (< 1.2, ≥ 1.2 mg/dL), preoperative chemotherapy, clinical T factor (T1‐2, T3–4), N factor (N0, N1–3), distant metastasis (M) (M0, M1) (the 7th edition of the American Joint Committee on Cancer TNM classification), and the American Society of Anesthesiologists' physical status (ASA‐PS: 1, 2, and 3–5). These variables and categorization were based on the results of previous studies [14, 15].
Surgeons' characteristics included gender and years since medical license registration in 5‐year increments. Each category of years since license registration—including 5 years or less, 6–10 years, 11–15 years, 16–20 years, and ≥ 21 years—was assumed to reflect phased surgical skill acquisition, representing incomplete general surgery training program, board‐certified general surgeons, board‐certified gastroenterological surgeons, board‐certified trainers, and holding the position of Director or similar in a surgical department, respectively.
Institutions were divided into four categories by the annual number of laparoscopic right hemicolectomy cases so that each category contained approximately the same number of cases to increase statistical power, as follows: very low, low, high, and very high (< 7, ≥ 7 < 12, ≥ 12 < 19, and ≥ 19).
The average number of laparoscopic right hemicolectomies per surgeon was calculated by dividing the number of relevant surgeries during the study period by the number of surgeons who performed the surgeries.

2.4
Statistical Analysis
This study was approved by the Nippon Medical School Central Review Board (M‐2023‐163) [16]. Continuous variables are presented with medians and interquartile ranges (IQR), and categorical variables are presented as numbers and proportions. The multilevel multivariate logistic regression model, with adjustment for characteristics of the patients, surgeons, and hospitals, was used to examine the association between surgeons' gender and surgery‐related outcomes. A two‐sided p < 0.05 was considered statistically significant. All statistical analyses were performed using R software version 4.3.0 (www.r‐project.org).

3
Results
3.1
Study Population
A flow diagram of included patients is presented in Figure S1. A total of 109 893 patients who underwent right hemicolectomy were registered in the NCD between 2018 and 2022. Patients were excluded if they were < 18 years (n = 81), did not have colon carcinoma (n = 11 637), had undergone emergency surgery (n = 4184) or open or robot‐assisted surgery (n = 28 094), were operated on by non‐JSGS members (n = 6177), or lacked sufficient data (n = 1217), leaving 58 503 patients. A total of 53 387 (91.3%) eligible surgeries were performed by male surgeons and 5116 (8.7%) by female surgeons. Median male and female years since medical license registration at the time of each surgery were 14 and 10, respectively. The proportion of patients operated on by female surgeons gradually decreased from 14.3% to 2.8% in 5‐year increments over the years since medical license registration (Table 1). There were 7111 (90.2%) male and 770 (9.8%) female surgeons, who underwent at least one laparoscopic right hemicolectomy. Median male and female surgeons' years since medical license registration from the perspective of each surgeon were 14 and 8, respectively (Table 2).

3.2
Characteristics of Institutions, Patients, and Surgeons
Institutional characteristics, preoperative factors, intraoperative outcomes, and postoperative outcomes by surgeon gender are shown in Tables 3 and 4.

3.3
Institutional Characteristics
Operations by female surgeons occurred at institutions with a higher number of laparoscopic right hemicolectomies per year than operations by male surgeons (median: 13.4 vs. 11.8, respectively). In particular, there was a higher proportion of patients who underwent their surgery at institutions performing 19 or more laparoscopic right hemicolectomies per year in the female surgeon group than in the male surgeon group (32.8% vs. 24.6%) (Table 3). The average number of laparoscopic right hemicolectomies per surgeon in the female surgeon group was fewer but not notably lower compared to the male surgeon group (Table S1).

3.4
Preoperative Factors
Patients' age and distribution were similar between male and female surgeon groups. With respect to the preoperative variables related to surgical risks, in the female surgeon group, the proportions of patients with previous cardiovascular diseases (5.9% vs. 5.0%) and ASA‐PS [3, 4, 5] (17.4% vs. 16.8%) were higher, but those of abnormal albumin (< 3.5 g/dL) (23.9% vs. 24.5%) and creatinine (> 1.2 mg/dL) (8.1% vs. 8.9%) were lower than in the male surgeon group (Table 3).

3.5
Intraoperative Outcomes
The female surgeon group had a longer median operating time (232 vs. 225 min) and less estimated blood loss during surgery (22 vs. 25 mL) than the male surgeon group (Table 4).

3.6
Postoperative Outcomes
As main outcomes, surgical mortality and severe postoperative complications were 0.6% vs. 0.5% and 3.9% vs. 4.2% in the female and male surgeon groups, respectively (Table 4).

3.7
Risk‐Adjusted Model for Main and Secondary Outcomes
With respect to main outcomes, the risk‐adjusted ORs for surgical mortality and severe complications in the female surgeon group compared with the male surgeon group were 1.29 (95% CI: 0.87–1.92, p = 0.20) and 0.96 (95% CI: 0.83–1.13, p = 0.65), respectively. For secondary outcomes, the risk‐adjusted ORs for reoperation, anastomotic leakage, and pneumonia were 0.79 (95% CI: 0.62–0.99, p = 0.049), 0.94 (95% CI: 0.72–1.22, p = 0.62), and 1.18 (95% CI: 0.91–1.54, p = 0.22), respectively (Figure 1).

3.8
Interactions Between Surgeon Gender and Years Since Medical License Registration
For the sub‐analysis, we compared main outcomes (i.e., surgical mortality and severe complications) between the male and female surgeon groups in categories representing 5‐year increments of the total years since medical license registration. For those with 5 years or less experience after registration, the risk‐adjusted OR for surgical mortality was significantly higher in the female surgeon group than in the male surgeon group (2.22, 95% CI: 1.04–4.74, p = 0.04) (Figure 2a). However, the risk‐adjusted ORs for surgical mortality did not differ significantly between the genders for the other categories of years since registration. As for severe complications, the risk‐adjusted ORs revealed no significant differences in any years since licensing category (Figure 2b).

4
Discussion
To our knowledge, this is the first comparative study encompassing 58 503 patients from a nationwide database to examine differences in the surgical outcomes of laparoscopic right hemicolectomy by surgeons' gender. The major finding of this study was that no significant differences were observed in the adjusted risks for surgical mortality and severe complications between male and female surgeons. This result suggests that the performance of female surgeons is equivalent to that of male surgeons with respect to patients' clinical outcomes after laparoscopic right hemicolectomy.
There is increasing evidence to suggest that surgical outcomes can be attributed to both patient‐ and surgeon‐related factors. Surgeon‐related factors, which include not only specialty and technical proficiency but also individual personality characteristics such as communication skills, clinical judgment, and the ability to facilitate teamwork, may also be important determinants of patient outcomes [17, 18]. In this vein, research comparing differences in treatment outcomes by surgeons' gender has gained increased attention.
This study clearly demonstrated equivalent surgical outcomes (i.e., surgical mortality and severe complications) in patients who underwent laparoscopic right hemicolectomy for male and female surgeons. Previous comparisons of surgical outcomes by surgeon gender have reported better outcomes for female surgeons [7, 19, 20]. Saka et al.'s recent meta‐analysis, which included 5 448 121 patients from 15 cohort studies, found that patients treated by female surgeons had a lower postoperative mortality (adjusted OR: 0.93, 95% CI: 0.88–0.97, I
2 = 27%) [21]. Another recent large study by Scali et al., involving over 4 000 000 patients from the United States, reported a small statistical and marginal superiority in the female surgeon group in terms of the composite outcome of mortality and morbidity. However, they concluded that the association with surgeon gender is unlikely to be causal for the observed differences but is probably due to outcome variations by surgical specialty and procedures [22]. The results of previous studies incorporating a variety of surgical techniques have been inconclusive, but thus far, no studies have reported inferior outcomes for female surgeons.
In addition to limiting the procedure to right hemicolectomy, the primary reasons for limiting elective laparoscopic surgery were (i) to exclude cases of laparotomy often performed on highly advanced cancer patients and high‐risk patients, as well as cases of robot‐assisted surgery, which are performed in a limited number of cases; laparoscopic surgery has already become a standard procedure, (ii) because open or emergency colon surgery is associated with a significantly higher frequency of postoperative complications and mortality than laparoscopic or elective surgery [23, 24], we wanted to remove as much variation in surgical procedure as possible to ensure a more homogeneous cohort.
In this study, female surgeons had significantly lower reoperation rates than male surgeons (OR: 0.79, 95% CI: 0.62–0.99, p = 0.049) despite there being no significant difference in the incidence of severe complications. One possible explanation is that female surgeons may adopt a more cautious and meticulous intraoperative technique, which could help reduce the likelihood of complications that require reintervention. Additionally, female surgeons may be more attentive to early signs of postoperative deterioration, leading to timely conservative management before surgical reoperation becomes necessary. While it has been reported that female physicians have better communication skills than their male counterparts and provide more guideline‐compliant care [25, 26, 27], these behavioral differences may also influence postoperative management strategies. However, given the observational nature of this study, no causal relationship can be established, and this interpretation remains speculative.
One of the major limitations of our study is that the nature of the data precludes evaluation of long‐term prognosis. A recent single‐institution study by Engdahl et al. reported that there was no difference not only in short‐term outcomes—including postoperative complications and 30‐day mortality—but also in long‐term outcomes, including cancer‐free and overall survivals between patients operated on by male and female surgeons after elective colon cancer resections [28]. Interestingly, this study demonstrated the female surgeons' superiority in both short‐ and long‐term outcomes after emergent colonic resections [28]. Because the occurrence of postoperative complications is known to be a poor prognostic factor [29, 30], the authors speculate that the better short‐term outcomes for female surgeons may have led to superior long‐term outcomes [28].
The subgroup analysis compared the main outcomes for male and female surgeons according to years post‐registration and found that female surgeons had comparable surgical outcomes to their male counterparts for all 5‐year categories, except for the category representing 5 years' experience or less (i.e., incomplete general surgery training program phase). Only in this category did female surgeons have a statistically higher adjusted risk for surgical mortality than male surgeons (OR: 2.22, 95% CI: 1.04–4.74, p = 0.04). However, it was notable that no significant difference was observed in terms of severe complications in this category (OR: 0.96, 95% CI: 0.71–1.81, p = 0.80). Surgical skill, primarily technical proficiency, did not differ between female and male surgeons even in this early category, which is supported by previous studies reporting female medical students' and residents' superiority at learning surgical techniques over their male counterparts [31, 32]. However, determining the definitive cause of the higher surgical mortality among female surgeons in this younger generation is challenging. This difference may be related to a difference in complication rescue ability; however, it may also be related to the ability of the surgical team, including the supervising surgeons in these situations, and not only the young surgeon. This study was originally a confirmatory analysis based on a previous NCD study, hypothesizing that there was no difference between male and female surgeons. This was also the case in the subgroup analysis, which was divided into 5 years since registration of medical license. Therefore, we believe that it is not appropriate to conduct an exploratory analysis and the discussion after the initial analysis to investigate the factors behind the significant differences that emerged in the results.
In this study, the higher number of laparoscopic right hemicolectomies performed by female surgeons at institutions performing 19 or more such procedures per year was shown. Although this may be related to the number of female surgeons employed at the institution, it is more likely that the high number of surgeries performed at the institution itself reflects the high number of surgeries performed per female surgeon. However, we emphasize that our study is observational and not exploratory in nature, and this interpretation does not imply a causal relationship to given results.
Previous studies have found that female surgeons face several obstacles to choosing their careers and acquiring surgical proficiency, including an imbalance in the distribution of surgical cases assigned to female versus male surgeons, which occurs not only in Western countries but also in Japan [33, 34]. Kono et al. first demonstrated the gender discrepancy in surgical experience (i.e., fewer opportunities for surgical training for female surgeons) in six types of gastroenterological surgeries including right hemicolectomy in Japan using the NCD [34]. The gender gap tended to widen with increasing years of experience, especially for medium‐ and high‐difficulty surgeries, including right hemicolectomy (3.65 times higher in male surgeons in the category of 30–39 years) [34]. Furthermore, a gender gap in the assignment of surgical approach has also been reported. Foley et al. reported that male surgeons were more privileged than female surgeons with respect to the use of consoles and procedure completion in robotic colorectal surgery training programs [35]. Additionally, a tendency was observed for laparoscopic surgeries to be assigned more frequently to male surgeons than female surgeons in low anterior resection in Japan [8]. In Japan, patients typically do not have the option to choose their primary surgeon. Instead, primary surgeons, including the surgical approach, are assigned to surgeries either randomly or based on the judgment of the department head. Consequently, the role of supervisors in assigning cases to female surgeons is a crucial component of equalizing their training process to that of male surgeons. However, the average number of laparoscopic right hemicolectomies per surgeon in the female surgeon group was fewer, but not substantially different from that in the male surgeon group (i.e., the difference was less than one case during the study period). Therefore, it is unlikely that gender‐related differences in surgical opportunities had a significant impact on the study results.
This study has several additional limitations that should be considered. First, this is a retrospective observational study, which is inferior to prospective trials with respect to precision and validity, and it is impossible to adjust for unmeasured confounders. Second, because there were fewer female surgeons than male surgeons, it is possible that the results were disproportionately influenced by the outcomes of a single female surgeon, potentially introducing bias into the overall findings. Third, information about surgeon gender was self‐reported, which could result in the misclassification of this variable to biological sex.
In conclusion, this study found no significant differences between male and female surgeons in surgical mortality and severe complications in laparoscopic right hemicolectomy. Our findings suggest that providing well‐structured and efficient surgical training programs for all surgeons, regardless of gender, may help further improve surgical quality.

Author Contributions

Akihisa Matsuda: conceptualization, writing – original draft, investigation, project administration. Hideki Endo: investigation, data curation, methodology, writing – review and editing, formal analysis. Aya Yamagishi: conceptualization, writing – original draft. Hiroyuki Yamamoto: investigation, writing – review and editing, formal analysis, data curation. Shuji Takiguchi: writing – review and editing, project administration. Koki Otsuka: writing – review and editing, project administration. Takeshi Yamada: writing – review and editing, investigation. Hiroshi Yoshida: writing – review and editing. Hideki Ueno: supervision. Yuko Kitagawa: supervision. Ken Shirabe: supervision.

Ethics Statement
This study was approved by the Central Review Board of Nippon Medical School (M‐2023‐163).

Consent
The authors have nothing to report.

Conflicts of Interest
Hideki Endo and Hiroyuki Yamamoto are affiliated with the Department of Healthcare Quality Assessment at the University of Tokyo. The department is a social collaboration department supported by grants from the National Clinical Database, Intuitive Surgical Sarl, Johnson & Johnson KK, and Nipro Corporation. Yuko Kitagawa is the editor‐in‐chief, and Shuji Takiguchi, Hideki Ueno, and Ken Shirabe are members of the Annals of Gastroenterological Surgery. They were not involved in the editorial responsibilities or decision to accept this article for publication.

Supporting information

Figure S1: Flow diagram of included patients.

Table S1: The average number of laparoscopic right hemicolectomies per surgeon.