This retrospective study was reviewed and approved by the Human Ethics Committee of Kinan Hospital (Approval number: 283, Clinical outcomes of endoscopic surgery: retrospective analyses). The deidentified medical records of 548 patients who underwent TLH between 1 July 2010 and 31 December 2024 for uterine diseases were retrospectively reviewed, and data regarding TLH indications and histopathological diagnoses of uterine lesions were collected. In terms of major intraoperative complications, according to past reports [4], we focused mainly on organ injuries that were considered to have occurred as a result of surgical manipulation, particularly due to instrument handling within the pelvic cavity during TLH. Among the 25 cases reported by our hospital to the Japan Society of Gynaecologic and Obstetric Endoscopy and Minimally Invasive Therapy, we excluded the following 10 cases: 4 cases involving instrument breakage; 1 case involving an abdominal wall haematoma; 2 cases involving subcutaneous emphysema; 2 cases involving skin burns, which were presumably caused by the light source cable; and 1 case involving vaginal wall laceration, all of which were identified after the completion of TLH. Consequently, we included 15 cases in the analysis. With respect to the amount of blood loss, although we considered setting the threshold at 500 mL, which is generally regarded as a significant amount of blood loss in gynaecological surgery, TLH tends to result in less blood loss than conventional abdominal hysterectomies do. At that threshold (500 mL), only two patients met the criteria. Therefore, by considering the mean and standard deviation of the blood loss amount for our institution, we set the threshold at 300 mL ($\pm$ 2.5 SD), which corresponded to 20 cases, or approximately 3.6% of all cases. In combination with the aforementioned organ injuries as major intraoperative complications, the total number of cases was 27, representing just under 5% of all 548 cases. This proportion could be considered a reasonable figure in comparison with those of previous reports, especially in comparison with some reports in Japan [4, 8, 9] and was therefore used in our analysis [10]. For the 548 TLH patients, data on the following variables were extracted: (1) basic patient characteristics, primarily obtained from medical questionnaires documented in the medical records, such as age; body mass index (BMI), defined as body weight (kg) divided by the square of height (m); parity; history of gynaecological surgery; irregular menstruation; hypermenorrhoea; and dysmenorrhoea; (2) data related to preoperative conditions, such as the leiomyoma size (defined as the maximal diameter measured by magnetic resonance imaging (MRI)), the uterine size (defined as the average of the long and short diameters measured by MRI), the presence of ovarian endometriotic cysts or adenomyoma, and carbohydrate antigen 19-9 (CA19-9), carbohydrate antigen 125 (CA125), and lactate dehydrogenase (LDH) positivity; and (3) intraoperative/postoperative findings, such as the operation time, blood loss amount, intraoperative adhesion findings, pathological diagnoses of endometriosis, pathological diagnoses of adenomyosis, and resected uterine weight. Although this is a limitation of the study, owing to the retrospective nature of data collection, which focused on patients visiting the gynaecology department, information on comorbidities from other specialties, such as internal medicine and psychiatry, could not be obtained.

First, we compared basic information such as age, BMI, parity, leiomyoma diameter, uterine size, operation time, blood loss amount, and resected uterine weight between patients with and without major intraoperative complications. For each factor, the median, as well as the 25th percentile and 75th percentile values, are compared. The p values were obtained with the Wilcoxon rank-sum test, and p 0.05 was considered to indicate statistical significance. Next, from the available medical records, 22 representative factors were selected. The 548 patients were divided into two groups on the basis of the presence or absence of each factor. The proportions of patients with major intraoperative complications between these two groups were subsequently compared. Among these factors, advanced age was defined as 50 years or older as a substitute for menopausal status, and high BMI was defined as 30 kg/m² or greater. On the basis of previous reports [4, 8, 11], a large leiomyoma was defined as a lesion with a diameter of 8 cm or more, a large uterus was defined as a uterus with a diameter of 10 cm or more, and heavy uterine weight was defined as 500 g or more. Given the relatively small number of patients with major intraoperative complications (n = 27), and the possibility of low expected cell counts in the 2 $\times$ 2 contingency tables, Fisher’s exact test (two-tailed) was used as an appropriate statistical method for between-group comparisons.

Finally, candidates for multivariate analysis were selected on the basis of a p value of 0.30 or less in the aforementioned simple comparisons. We then extracted data on the following 9 representative factors: (1) high BMI (defined as 30 kg/m² or more), (2) hypermenorrhoea, (3) history of gynaecological surgery, (4) large leiomyoma (8 cm or more), (5) large uterus (10 cm or more), (6) coexistent ovarian endometriotic cyst, (7) intraoperative findings of adhesions, (8) pathological diagnosis of endometriosis, and (9) heavy uterine weight (500 g or more). The patients were then divided into two groups, and multivariate analysis was performed twice as follows: first, seven preoperative factors (1 through 6) were used; second, nine factors that included both preoperative and intraoperative/postoperative findings (1 through 5 and 7 through 9) were used.

Statistical analyses were performed with Microsoft Excel 365 (Microsoft Corporation, Redmond, WA, USA) and JMP version 18 for Windows (SAS Institute, Inc., Tokyo, Japan). The associations between patient characteristics and the likelihood of major intraoperative complications were assessed with the Wilcoxon rank-sum test, Fisher’s exact test (two-tailed) and multivariable logistic regression analysis. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to determine the strengths of the associations. A value of p 0.05 indicated statistical significance.

At our institution, TLH is performed primarily for uterine lesions, the distribution of which is shown in Table 1A. Similarly, the postoperative pathological findings around the uterine lesions were also recorded, as shown in Table 1B. As expected, approximately 85% of the patients in both Table 1A and Table 1B were diagnosed with leiomyoma or adenomyoma. Among the total cases, 73 involved adnexal tumour lesions, which were preoperatively identified and presumed to be benign. As shown in Table 2, 27 patients experienced major intraoperative complications, 8 of whom experienced multiple complications. Among these cases, 11 involved organ injury in addition to massive blood loss. However, in all of these cases, intraoperative repair was successfully performed, and no postoperative sequelae occurred.

The indications for TLH and the pathological diagnoses of the 548 patients were documented. The uterine diseases were classified accordingly. TLH, total laparoscopic hysterectomy.

In a simple comparison of patient characteristics between those with and without major intraoperative complications, significant differences were observed in uterine size, operation time, blood loss amount, and resected uterine weight (Table 3). Furthermore, as shown in Table 4, a simple analysis of the associations between individual factors and the likelihood of major intraoperative complications revealed 3 significant factors, namely, hypermenorrhoea, intraoperative findings of adhesions, and heavy uterine weight (500 g or more). In addition, the following 6 factors had p values of 0.30 or less when the proportions of patients with major intraoperative complications were compared between the groups: (1) coexistent ovarian endometriotic cyst, (2) large leiomyoma (8 cm or more), (3) high BMI (defined as 30 kg/m² or more), (4) history of gynaecological surgery, (5) large uterus (10 cm or more), and (6) pathological diagnosis of endometriosis. On the basis of these results, a total of 9 factors were included in the subsequent multivariate analysis.

Among the candidate factors that were selected on the basis of the results of the simple analyses, namely, p values of 0.30 or less according to Fisher’s exact test (two-tailed), significant differences were detected for (1) high BMI (defined as 30 kg/m² or greater), with an OR of 3.9 (95% CI: 1.0–11.9); (2) hypermenorrhoea, with an OR of 5.9 (95% CI: 2.0–25.9); and (3) coexistent ovarian endometriotic cysts, with an OR of 4.3 (95% CI: 1.4–11.5) among the preoperative factors; and (1) hypermenorrhoea, with an OR of 7.0 (95% CI: 2.2–30.9); (2) intraoperative findings of adhesions, with an OR of 3.2 (95% CI: 1.2–8.5); and (3) heavy uterine weight (500 g or greater), with an OR of 4.6 (95% CI: 1.3–15.9) among the intraoperative/postoperative factors (Table 5A and Table 5B). Additionally, a significant trend and a p value of less than 0.1 were detected for the pathological diagnosis of endometriosis.