Learning curve and perioperative outcomes of four-port assistant-independent robot-assisted total laparoscopic hysterectomy according to uterine size.

The number of cases by retrieved tissue weight, from the lightest to the heaviest, was 174 cases (SG), 162 cases (MG), 79 cases (LG), 30 cases (HG), and 40 cases (GG); the number of cases decreased as the weight increased. An analysis of the background characteristics of the five groups revealed no significant differences between the groups in terms of age, height, or BMI, except that the GG exhibited a significantly wider mean MAW than the SG and HG (Table  1 ). The mean weights of the retrieved uterus were 162 (55–249), 348 (250–499), 613 (501–747), 869 (752–990), and 1,457 (1,005–2,766) g in the SG, MG, LG, HG, and GG, respectively. The GG demonstrated significantly longer mean operative time (272 min), CT (174 min), closure time (18 min), and retrieval time (55 min), as well as a significantly higher blood loss (333 mL), than the SG, MG, LG, and HG; however, no significant differences in length of hospital stay were observed among the groups (Table  1 ). One case (2.5%) in the GG received allogeneic blood transfusion. Table 1 Comparison of background characteristics and peri-operative outcomes among groups classified by retrieved tissue weight Group small (S) medium (M) large (L) huge (H) giant (G) p -value Retrieved weight(g) <250 250≦ <500 500≦ <750 750≦ <1000 1000≦ <3000 No. of cases 174 162 79 30 40 Age* 47.2 ± 7.1 47.3 ± 5.1 48.0 ± 5.9 47.4 ± 3.9 49.1 ± 5.6 NS Height; cm* 158.7 ± 5.2 159.0 ± 5.2 159.1 ± 5.1 159.7 ± 5.0 159.0 ± 6.2 NS BMI; kg/㎡* 22.7 ± 4.1 23.1 ± 4.4 22.7 ± 3.9 22.9 ± 5.1 24.7 ± 4.3 NS MAW; cm* 278 ± 33 287 ± 29 285 ± 28 276 ± 28 300 ± 27 NS, except for S–G: <0.001; H–G: 0.016 Operation time; min* 153 ± 47 174 ± 53 194 ± 49 219 ± 87 272 ± 75 S; except for M–H: 0.071; L–H: 0.570; H–G: 0.071 Console time; min* 126 ± 43 138 ± 47 143 ± 44 154 ± 67 174 ± 56 NS, except for S–L: 0.041; S–G: <0.001; M–G: 0.004; L–G: 0.028 Preparation time; min* 15 ± 8 16 ± 8 16 ± 7 15 ± 7 16 ± 7 NS Closed time; min* 10 ± 6 11 ± 5 12 ± 7 12 ± 6 18 ± 11 NS, except for S–G: <0.001; M–G: 0.003; H–G: 0.028 Retrieval time; min* 5 ± 6 15 ± 13 25 ± 15 35 ± 20 55 ± 31 S, except for L–H: 0.146 Retrioeved weight; g* 162 ± 52 348 ± 70 613 ± 71 869 ± 74 1457 ± 454 S Blood loss; ml* 16 ± 29 23 ± 43 60 ± 114 61 ± 92 333 ± 341 NS, except for M–L: 0.046; S–G, M–G, L–G, and H–G: <0.001 Hospital stay; day* 6.2 ± 0.9 6.4 ± 1.5 6.4 ± 1.5 6.5 ± 0.9 6.3 ± 0.9 NS * Mean ± standard deviation MAW: Minimum abdominal width on abdominal X-ray, NS: not statistically significant, S: statistically significant Pairwise comparisons were performed using the appropriate post-hoc test Comparison of background characteristics and peri-operative outcomes among groups classified by retrieved tissue weight * Mean ± standard deviation MAW: Minimum abdominal width on abdominal X-ray, NS: not statistically significant, S: statistically significant Pairwise comparisons were performed using the appropriate post-hoc test In the analysis of tissue retrieval, the SG encompassed almost entirely of TV retrieval, with the proportion decreasing as the retrieval weight increased; the GG demonstrated no TV retrieval. The MG, LG, and HG most frequently employed WR (50%–60%); however, its use was lower in GG (30%), with 70% of cases involving SI retrieval (Supplementary Table 1). In the GG, the mean weight of tissue retrieved via WR was 1,264 (1,005–1,756) g, the retrieval time was 55 (26–97) min, and blood loss was 413 (2–535) mL. By contrast, the mean weight of tissue retrieved via SI was 1,539 (1,029–2,766) g, the retrieval time was 55 (10–161) min, and blood loss was 152 (10–1,247) mL. In the analysis of proficiency, the learning curve based on CT for the four surgeons revealed two distinct phases (phase 1, characterized by a positive slope, and phase 2, characterized by a negative slope). The turning points occurred at the 37th, 17th, 18th, and 15th case for surgeons 1, 2, 3, and 4, respectively (Fig.  2 ). Furthermore, surgeon 1 was inexperienced in laparoscopic surgery, whereas surgeon 4 was a certified laparoscopic surgeon. Regarding the number of cases in each group classified by the number of cases performed and arranged from the least to the most number of cases, Groups A, B, C, and D had 102, 80, 124, and 179 cases, respectively. In the comparison of CT among surgeons within each tissue recovery group, in Groups S and M (tissue recovery weight, < 500 g), a significant reduction in CT was observed between Groups A and B (classified by surgeon experience); however, Groups B and C exhibited no significant difference. Regarding tissue recovery weights of ≥ 500 g, Groups A, B, and C demonstrated no significant differences. Compared with Group D (skilled surgeons), Group A showed a significant difference in the weight range of ≥ 250 g but < 750 g; however, no significant differences were observed in any of the other weight groups. Groups B and C exhibited no significant differences from Group D in any of the weight groups (Table  2 ). On the other hand, the interaction between surgeon experience and uterine weight was not statistically significant (β = 0.0003, 95% CI − 0.078 to 0.079, P  = 0.994). In contrast, uterine weight itself was independently associated with console time (β = 0.119, 95% CI 0.053–0.184, P  < 0.001) (Supplementary Table 2). Fig. 2 Learning curves and turning points for four surgeons. The cumulative sum (CUSUM) graph was created based on console time for each case number Learning curves and turning points for four surgeons. The cumulative sum (CUSUM) graph was created based on console time for each case number Table 2 Comparative analysis of console time according to surgeon experience and retrieved tissue weight Group small (S) medium (M) large (L) huge (H) giant (G) Retrieved weight(g) <250 250≦ <500 500≦ <750 750≦ <1000 1000≦ <3000 No. of cases experienced (A) 1 to 20 146± 44* ( n =44)** 170 ± 46 ( n =33) 173 ± 43 ( n =13) 190 ± 50 ( n =5) 169 ± 48 ( n =7) (B) 21 to 40 121 ± 42 ( n =31) 126 ± 41 ( n =30) 148 ± 53 ( n =11) 152 ± 71 ( n =5) 130 ± 33 ( n =3) (C) 41 or more 108 ± 41 ( n =50) 132 ± 42 ( n =39) 138 ± 44 ( n =22) 112 ± 22 ( n =10) 169 ± 49 ( n =9) (D) skilled surgeons*** 126 ± 44 ( n =49) 131 ± 46( n =60) 133 ± 37 ( n =33) 178 ± 85 ( n =10) 183 ± 63 ( n =20) p-value**** A-B 0.047 # < 0.001 # 0.466 0.763 131 ± 46 ( n =60) A-C < 0.001 # 0.002 # 0.095 0.071 1.000 A-D 0.165 < 0.001 # 0.027 # 0.986 0.938 B-C 0.493 0.943 0.928 0.627 0.746 B-D 0.857 0.948 0.765 0.919 0.427 C-D 0.061 1.000 0.976 0.137 0.911 * Average console time (minutes) and standard deviation, ** Number of cases *** Skilled surgeons: >200 robotic surgery cases and >5 years of experience **** Pairwise comparisons were performed using the appropriate post-hoc test. # p-value < 0.05 Comparative analysis of console time according to surgeon experience and retrieved tissue weight * Average console time (minutes) and standard deviation, ** Number of cases *** Skilled surgeons: >200 robotic surgery cases and >5 years of experience **** Pairwise comparisons were performed using the appropriate post-hoc test. # p-value < 0.05 Conversely, regarding tissue recovery weights of < 250 g, the proportion of high-difficulty cases was the highest and lowest in Groups D (46.9%) and A (4.5%), respectively (Supplementary Table 3). Similarly, regarding blood loss, a significant difference was observed only in Group M (recovered weight of ≥ 250 g but < 500 g) compared with Group A, which had a small number of cases, and Group C, which comprised experienced surgeons; however, no significant differences were observed between the other experience groups in any of the recovered weight groups (Supplementary Table 4). Furthermore, when cases of RA-TLH were divided into three groups (Group A, ≤ 20 cases; Groups B and C, ≥ 20 cases; and Group D, skilled surgeons) and compared the proportion of cases in each weight group within each group, it was observed that in Group S (recovery weight, ≤ 250 g), Groups A (43.1%) and B + C (38.7%) had a higher proportion of cases than Group D (28.5%); conversely, in Group G (recovery weight, ≥ 1,000 g), Group D (11.7%) had the highest proportion of cases (Table  3 ). In Group G, skilled surgeons performed 21 of 40 cases (52.5%). Table 3 Distribution of cases by retrieved tissue weight according to surgeon experience No. of cases and their proportions (*) in Groups classified by retrieved tissue weight Group small (S) medium (M) large (L) huge (H) giant (G) Total Retrieved weight(g) <250 250≦ <500 500≦ <750 750≦ <1000 1000≦ <3000 Groups classofied by No. of cases experienced A 44 (43.1) 33 (32.4) 13 (12.2) 5 ( 4.9) 7 ( 6.9) 102 (100) B+C 79 (38.7) 67 (32.8) 31 (15.2) 15 ( 7.4) 12 ( 5.9) 204 (100) D 51 (28.5) 62 (34.6) 35 (19.6) 10 ( 5.6) 21 (11.7) 179 (100) Total 174 162 79 30 40 485 * Proportion of cases in each retrieved weight group within a cohort of patients who underwent robot-assisted total hysterectomy Distribution of cases by retrieved tissue weight according to surgeon experience * Proportion of cases in each retrieved weight group within a cohort of patients who underwent robot-assisted total hysterectomy Proportions are calculated within each surgeon-experience group. A: 1–20 robotic surgery cases; B: 21–40 cases; C: ≥41 cases; D: skilled surgeons with >200 robotic surgery cases and >5 years of experience. Regarding complications, the overall incidence was 2.9% ( n  = 14/485 cases). Group G exhibited the highest incidence rate by recovered weight at 7.5% ( n  = 3/40 cases; 1 of whom required blood transfusion), whereas the other groups demonstrated incidence rates ranging from 0% to 3.8%. However, most complications were classified as Clavien–Dindo Class II or lower, encompassing 10 and 3 cases of intra-pelvic infection and intra-abdominal hematoma, respectively. The only case classified as Class III or higher was in Group G (Class IIIb), which required robotic hemostasis on the third postoperative day owing to intra-abdominal hemorrhage (Table  4 ). Conversely, no cases required conversion to open surgery. Table 4 Complications according to retrieved tissue weight Group small (S) medium (M) large (L) huge (H) giant (G) Total Types of complications Treatment methods Retrieved weight(g) <250 250≦ <500 500≦ <750 750≦ <1000 1000≦ <3000 No. of cases 174 162 79 30 40 485 Clavien-Dindo classificastion Ⅰ 1 0 1 0 1 3 intra-abdominal hematoma follow-up observation Ⅱ 3 4 2 0 1 10 intra-pelvic infection antibiotic therpy Ⅲb 0 0 0 0 1 1 postoperative hemorrhage Robotic surgery No. of cases with complications 4 4 3 0 3 14 Complication rate (%) 2.3 2.5 3.8 0 7.5 2.9 Complication rate was calculated as the number of cases with complications divided by the total number of cases in each retrieved tissue-weight group Complications according to retrieved tissue weight Complication rate was calculated as the number of cases with complications divided by the total number of cases in each retrieved tissue-weight group

Between July 2020 and October 2025, our hospital performed 493 total hysterectomies for benign uterine tumors after obtaining adequate informed consent. As we generally did not impose size restrictions, 486 cases (98.6%) were robot-assisted total laparoscopic hysterectomies (RA-TLHs), and 7 cases (1.4%) were total abdominal hysterectomies (TAH). However, this study excluded TAH cases as they did not align with the study objectives. With the exception of one case (7,141 g), the weight of the retrieved tissue in the RA-TLH group was ≤ 3,000 g in all other cases; therefore, this analysis focused on 485 cases weighing 3,000 g or less and classified them into five groups. The study population comprised 174, 162, 79, 30, and 40 cases with retrieved tissue weights of < 250 g (small group [SG]), ≥ 250 g but < 500 g (medium group [MG]), ≥ 500 g but < 750 g (large group [LG]), ≥ 750 g but < 1,000 g (huge group [HG]), and ≥ 1,000 g but < 3,000 g (giant group [GG]), respectively (Fig.  1 ). Fig. 1 Flowchart for patients and surgical procedures in this study. RA-TLH: Robot-assisted total laparoscopic hysterectomy, TAH: Total abdominal hysterectomy, * This case was excluded because it was the only one in which the retrieved weight exceeded 3,000 g (7,141 g) Flowchart for patients and surgical procedures in this study. RA-TLH: Robot-assisted total laparoscopic hysterectomy, TAH: Total abdominal hysterectomy, * This case was excluded because it was the only one in which the retrieved weight exceeded 3,000 g (7,141 g) We compared patient characteristics, including age, body mass index (BMI), and minimum abdominal width (MAW; defined as the narrowest abdominal width on routine preoperative anteroposterior plain abdominal X-rays), as well as perioperative surgical outcomes, including operative time, console time (CT), preparation time, closure time, time to retrieve resected tissue, weight of retrieved tissue, blood loss, conversion to open surgery, length of hospital stay, and complications, across the five groups. Regarding the SG, we focused on high-difficulty cases identified through preoperative evaluation, including physical examination, imaging studies, and medical history. Surgery was performed under general anesthesia with the patient in the lithotomy position at a head-down angle of 25°–30° using a levitator. A Belles needle was inserted through the umbilicus; after securing the surgical field at a 10-mmHg pneumoperitoneum pressure, ports were created. The surgical robot employed was the Da Vinci X or Xi (Intuitive Surgical, Sunnyvale, CA, USA), with a side-docking approach rolling in from the patient’s left side. RA-TLH was performed as a four-port assistant-independent surgery without using an assist port in all cases. In this study, “assistant-independent” refers specifically to the absence of an assistant port during RA-TLH. The surgeon independently performed uterine manipulation, tissue retraction, and surgical field exposure using the robotic instruments, including the third robotic arm. A bedside assistant was present when necessary for standard perioperative support and tasks that could not be performed through the robotic instruments; however, no assistant port was used for surgical manipulation or tissue retraction. Briefly, an 8-mm Da Vinci endoscopic port was first created using the optical method, horizontally 4 cm to the right of the point 4 cm directly above the navel. The remaining three 8-mm Da Vinci ports were arranged in a straight line, spaced 8 cm apart from each other [ 7 ]. Considering forceps manipulation, we employed a proprietary “port-hopping” technique [ 7 ]. The forceps used encompassed ProGrasp™ Forceps, Monopolar Curved Scissors, Maryland Bipolar Forceps, and Large SutureCut™ Needle Driver. Two highly experienced robotic surgeons (skilled robotic surgeons) with over five years of robotic surgery experience and more than 200 cases of RA-TLH and six surgeons (novice robotic surgeons) with no prior robotic surgery experience performed these surgeries. Among these six surgeons, the learning curves for the console time were evaluated using cumulative sum (CUSUM) analysis [ 9 , 10 ] for the four surgeons (surgeons 1–4) who had experience with ≥ 60 RA-TLH cases.CUSUM analysis was performed separately for each novice surgeon using cases in chronological order. For each surgeon, the CUSUM value was calculated sequentially as the cumulative sum of the difference between the console time of each case and the mean console time across all cases performed by that surgeon during the study period. For example, for the second case, the CUSUM value from the first case was added to the difference between the CT of the second case and the surgeon-specific mean CT. Calculating this process through the final case yielded a continuous graph. Furthermore, to evaluate the surgeon’s proficiency, cases of RA-TLH were classified into four groups on the basis of the surgeon’s experience: 1–20 cases (group A), 21–40 cases (group B), ≥ 41 cases (group C), and skilled robotic surgeons (group D). Subsequently, surgical outcomes were compared across the tissue retrieval groups within each of these groups. Furthermore, regarding retrieval methods, this study investigated three approaches: conventional transvaginal (TV) retrieval, transvaginal retrieval using a wound retractor (WR), and retrieval via a small incision (a 3-cm horizontal incision 1 cm above the pubic bone [SI]) [ 11 ]. Regarding statistical analyses , continuous variables were summarized as means with standard deviations or medians with interquartile ranges, as appropriate. The distribution of continuous variables and the homogeneity of variance were assessed before parametric analyses. Because console time showed a right-skewed distribution, log-transformed console time was used for the multivariable regression analysis. For comparisons of console time across the five retrieved tissue weight categories, Welch’s one-way analysis of variance (ANOVA) was used because of unequal sample sizes and the potential for heterogeneity of variance. When the overall test was significant, Games–Howell post-hoc tests were performed for pairwise comparisons. This method was selected because it does not require equal variances and accounts for multiple pairwise comparisons. To evaluate whether the association between retrieved tissue weight and console time differed according to surgeon experience, a multivariable linear regression model was constructed using log-transformed console time as the dependent variable. Surgeon experience was classified as experienced surgeons (the two experienced surgeons) or novice surgeons (the six novice surgeons). The model included surgeon experience, log-transformed retrieved tissue weight, the interaction between surgeon experience and log-transformed retrieved tissue weight, age, BMI. Heteroscedasticity-robust (HC3) standard errors were used. For experience by retrieved tissue weight subgroup analyses, sample sizes and 95% confidence intervals were reported, and estimates from small subgroups were interpreted cautiously. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant. All statistical analyses were performed using the Statistical Package for the Social Sciences (version 29; SPSS 29). The Institutional Ethics Committee of Tokyo International Ohori Hospital approved this study (approval number: 2025-0003).

Four-port assistant-independent RA-TLH was feasible and was associated with favorable perioperative outcomes in this single-center retrospective cohort. Among cases with retrieved tissue weights of < 500 g, console time decreased significantly after approximately 20 cases, and the learning-curve inflection points for four novice surgeons occurred between 15 and 37 cases. However, these findings should be interpreted as observations of the learning curve rather than validated thresholds for surgical proficiency. For very large uteri, including those with retrieved tissue weights ≥ 1,000 g but > 3,000 g, favorable outcomes were observed in appropriately selected patients, predominantly managed by skilled robotic surgeons. Further multicenter prospective studies are warranted to determine the generalizability of these findings.

We here investigated the safety and proficiency of four-port RA-TLH without an assistant, focusing on retrieved tissue weight and surgeon experience. Specifically, we clarified two points: (1) the safety of this procedure for giant uteri and (2) the learning curve considering the surgeon’s experience and retrieved tissue weight. In RA-TLHs, operating time, blood loss, and the rate of conversion to open surgery increase when uterine weight exceeds 750 g [ 12 ]. Conversely, recently, an increasing number of studies have revealed that RA-TLH can be safely performed even on giant uteri weighing > 1,000 g [ 13 – 15 ]. The present study suggested that in Group G (retrieved weight, ≥ 1,000 g but < 3,000 g), CT and blood loss significantly increased, indicating that giant uteri present a higher surgical difficulty. In addition, retrieved tissue weight was significantly associated with CT, whereas the interaction between surgeon experience and retrieved weight was not statistically significant. This analysis demonstrated that retrieved weight is an independent factor. However, as no increase in hospital stay or severe complications was observed and that no cases of conversion to open surgery were noted, it was believed that with appropriate case selection and surgeon assignment, the procedure can be performed on giant uteri while maintaining its minimally invasive nature. On the other hand,　although no conversion to open surgery occurred in this cohort, the absence of conversion events should not be interpreted as evidence of a zero conversion risk. The relatively small number of cases with retrieved tissue weights ≥ 1,000 g limits the precision with which rare adverse outcomes can be estimated. Several studies on the learning curve for robot-assisted surgery have been conducted; Seamon et al. and Lim et al. have reported that the learning curve flattens after approximately 20–40 cases [ 16 – 18 ]. Similarly, in this study, the “turning point,” a proficiency indicator, occurred between 15 and 37 cases, aligning with previous studies. Conversely, a distinctive feature of this study is that it centered on solo surgery, which is not assistant-dependent. As this surgical technique reduces the influence of the assistant’s role in exposing the surgical field and retracting tissues, we believe that we evaluated a learning curve that more directly reflects the surgeon’s skills. Furthermore, when proficiency was evaluated by retrieved tissue weight, for uteri weighing  20 cases); therefore, for uteri of this size, approximately 20 cases were associated with a learning-curve inflection point in cases with retrieved tissue weights of < 500 g. Conversely, for uteri weighing ≥ 500 g, the experience groups demonstrated no significant difference in CT. Importantly, the absence of significant differences in console time among surgeon-experience groups for cases with retrieved tissue weights ≥ 500 g should not be interpreted as evidence that surgeon experience has little influence on the management of large uteri. Rather, these findings are likely influenced by selective case allocation, because more technically challenging cases were preferentially assigned to experienced surgeons. In particular, skilled surgeons performed 21 of the 40 cases (52.5%) with retrieved tissue weights ≥ 1,000 g but < 3,000 g. Therefore, the present data cannot determine whether novice surgeons would achieve comparable outcomes in very large uteri. Furthermore, the presence of groups that demonstrated no difference in CT between novice and experienced surgeons for small uteri may be due to the assignment of high-difficulty cases, including those with severe adhesions, to experienced surgeons. Bleeding volume was low, averaging ≤ 100 mL in all groups except for Group G ( ≥  1,000g but < 3,000 g); no significant differences were observed based on surgeon experience. This finding is believed to be because the expanded field of view and superior forceps maneuverability provided by robotic surgery enabled even novice surgeons to perform stable hemostasis. The results of this study aligned with previous reports [ 19 , 20 ], suggesting that robot-assisted surgery contributes to decreased blood loss and accelerated postoperative recovery compared with laparoscopic surgery. The incidence of perioperative complications in conventional RA-TLH has been reported to be 5%–10%, which is deemed comparable to that of laparoscopic surgery [ 21 , 22 ]. In this study, the complication rate was low a 2.9%; even in the group with giant uteri weighing ≥ 1,000 g, the rate was lower than that reported in previous studies [ 23 , 24 ]. Furthermore, only one case (0.21%) was classified as Clavien–Dindo grade III or higher; no cases required conversion to open surgery. These findings suggest that this surgical technique is applicable to giant uteri without compromising safety, even as a solo surgery that does not necessitate an assistant. In this procedure, no assistant port was used; we ensured sufficient distance between ports and adequate surgical field exposure even for giant uteri by combining the four-port configuration with the port-hopping technique. Furthermore, the use of the third arm facilitated the surgeon to independently perform stable traction and retraction, minimizing assistant-dependence. Moreover, as this procedure can be performed with one assistant, it can potentially contribute to savings in human resources and costs. Conversely, although some limitations regarding suction maneuvers and needle insertion/removal were noted, these did not pose significant clinical difficulties due to the introduction of barbed sutures insertable through 8-mm ports and the use of the third arm. The limitations of this study encompassed its retrospective single-center design and the nonrandomized case selection. In particular, exceedingly challenging cases, including those involving a giant uterus or severe adhesions, were performed by highly skilled surgeons, introducing selection bias into the inter-surgeon comparison. Furthermore, this study did not examine difficulty factors other than uterine size (e.g., fibroid location, uterine mobility, and the severity of endometriosis); future prospective studies that take these factors into account are warranted. The most significant feature of this study is that it evaluated the learning curve in solo surgery, which does not rely on an assistant, while taking uterine size (classified in this study based on the weight of the resected tissue), a measure of surgical difficulty, into account. Consequently, we believe that we have provided clinically useful indicators regarding the uterine size that novice surgeons should start with and the number of cases required for achieving proficiency.

Total hysterectomy represents the standard treatment for benign uterine conditions. Laparoscopic total hysterectomy has recently gained widespread adoption as the standard minimally invasive procedure as it causes less postoperative pain, a shorter hospital stay, and an earlier return to daily activities than open surgery [ 1 , 2 ]. However, in cases involving a giant uterus, the operative space within the pelvis is substantially constrained, and the difficulty of laparoscopic surgery increases owing to limited visual access, reduced uterine mobility, and restricted forceps manipulation [ 3 , 4 ]. Robot-assisted surgery provides benefits not achieved in conventional laparoscopic surgery, including a three-dimensional field of view, high degrees of freedom for forceps manipulation, and superior ergonomics [ 5 , 6 ]. However, in cases involving a giant uterus, the close proximity between the camera and uterus or collisions between robotic arms, which can restrict surgical field exposure, may limit even robot-assisted surgery. One solution is to move the camera port position higher, as is performed in laparoscopic surgery. However, in cases of a giant uterus, limitations exist regarding how high the camera can be positioned. However, using the previously reported four-port technique [ 7 ], adequate distance can be maintained between the ports, and viewing the fibroids from a slightly lateral angle becomes possible as the camera is not positioned in the midline. Furthermore, the lateral aspects of the uterus can be more effectively managed by adopting port hopping and utilizing the strong force of the robotic forceps to sufficiently retract the enlarged uterus laterally, thereby facilitating surgery with greater freedom of movement. Regarding four-port robot-assisted hysterectomy, we have previously reported that combining the four-port technique, which does not use an assistant port, with port hopping facilitates “solo surgery,” a characteristic of robotic surgery, while maintaining safety, thereby reducing manpower and healthcare costs [ 8 ]. Therefore, this study aimed to evaluate the true surgical skill of the surgeon using the four-port technique through a learning curve. In laparoscopic and robotic surgery, the skill level of the assistant frequently impacts surgical outcomes. However, as the four-port technique does not necessitate an assistant, it enables a more direct assessment of the surgeon’s skill. Furthermore, uterine size was considered an important marker of operative complexity, although it does not fully capture surgical difficulty. Therefore, this study also evaluated surgeons on the basis of the weight of retrieved tissue and investigated indicators for safely performing surgery.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1