Surgical outcomes of robotic hysterectomy for large uterus weighing more than 1000 g: a retrospective study from a high-volume center.

Between January 2014 and December 2023, a total of 1,015 patients underwent RH at the Senior Department of Obstetrics and Gynecology in Chinese PLA General Hospital. After meticulous application of inclusion and exclusion criteria, data from 86 patients were included in the study. The average age of the population was 47.1 ± 4.5 years, with a mean BMI of 25.1 ± 4.0 kg/m 2 . The average number of gestations for the population was 2.6 ± 1.6. Among them, 13 (15.1%) patients had never given birth. The primary indication for surgery in 64 (74.4%) patients was large fibroids, while adenomyosis was the secondary indication in 17 (27.9%) patients, and a combination of both was present in 5 (5.8%) cases. Preoperative moderate anemia (hemoglobin < 9 g/dL) was present in 17 (19.8%) patients. Fifty-one (59.3%) patients had prior pelvic and abdominal surgery. The median uterine weight recorded was 1279 g with 65 (75.6%) patients having a uterine weight between 1000 and 1500 g, 15 (17.4%) patients between 1500 and 2000 g, and 6 (7.0%) patients with a uterine weight of ≥ 2000 g. In addition, 33 (38.4%) cases were classified as type A and 53 (61.6%) as type B. Furthermore, 77 (89.5%) patients had no or only mild adhesion, while 9 (10.5%) patients had moderate to severe adhesion. Demographic characteristics are detailed in Table  1 . Table 1 Baseline demographic characteristics of enrolled patients Total Uterine weight Lesion site Adhesion severity 1000–1500 g 1500–2000 g  > 2000 g P Type A Type B P None-mild Moderate to severe P Patients, n (%) 86 65 (75.6) 15 (17.4) 6 (7.0) 33 (38.4) 53 (61.6) 77 (89.5) 9 (10.5) Age in years, mean ± sd 47.1 ± 4.5 47.4 ± 4.7 45.7 ± 4.2 46.8 ± 4.5 0.409 47.5 ± 5.4 46.8 ± 3.9 0.524 47.1 ± 4.5 46.4 ± 4.6 0.669 Body mass index (kg/m 2 ), mean ± sd 25.1 ± 4.0 24.9 ± 3.9 26.2 ± 4.7 25.6 ± 2.4 0.467 26.0 ± 4.8 24.6 ± 3.4 0.132 25.1 ± 4.1 25.4 ± 2.6 0.874 Nullparity, n (%) 13 (15.1) 10 (15.4) 3 (20) 0 0.570 3 (9.1) 10 (18.9) 0.180 11 (14.3) 2 (22.2) 0.620 Indication for surgeries, n (%) 86 65 (75.6) 15 (17.4) 6 (7.0) 0.147 33 (38.4) 53 (61.6) 0.405 77 (89.5) 9 (10.5) 0.889 Fibroid 64 (74.4) 44 (67.7) 14 (93.3) 6 (100) 22 (66.7) 42 (79.2) 57 (74.0) 7 (77.8) Adenomyosis 17 (19.8) 16 (24.6) 1 (6.7) 0 8 (24.2) 9 (17.0) 15 (19.5) 2 (22.2) Fibroid and adenomyosis 5 (5.8) 5 (7.7) 0 0 3 (9.1) 2 (3.8) 5 (6.5) 0 Prior pelvic/abdominal surgery, n (%) 51 (59.3) 38 (58.5) 10 (66.7) 3 (50) 0.798 19 (57.6) 32 (60.4) 0.825 43 (55.8) 8 (88.9) 0.076 Prior cesarean delivery 24 (27.9) 16 (24.6) 5 (33.3) 3 (50) 0.368 11 (33.3) 13 (24.5) 0.460 20 (26.0) 4 (44.4) 0.434 Prior myomectomy 31 (60.8) 20 (30.8) 8 (53.3) 3 (50) 0.211 12 (36.4) 19 (35.8) 1.000 27 (35.0) 4 (44.4) 0.717 Prior excision of endometriosis 8 (9.3) 8 (12.3) 0 0 0.323 1 (3.0) 7 (13.2) 0.146 5 (6.5) 3 (33.3) 0.035 Preoperative anemia (hemoglobin < 9 g/dL), n (%) 17 (19.8) 12 (18.5) 3 (20) 2 (33.3) 0.682 7 (21.2) 10 (18.9) 1.000 14 (18.2) 3 (33.3) 0.280 Uterine Weight, median (range) 1274 (1010–2250) 1200 (1010–1496) 1800 (1500–1970) 2129 (2035–2250)  < 0.001 1266 (1010–2136) 1279 (1021–2250) 0.948 1266 (1010–2150) 1279 (1021–2250) 0.104 Data are expressed as mean ± standard deviation, absolute number (%), or median (range) Baseline demographic characteristics of enrolled patients Data are expressed as mean ± standard deviation, absolute number (%), or median (range) The median duration of the operation was 160 (75–390) minutes, correlating significantly with increasing uterine weight ( P  = 0.005) and adhesion severity ( P  = 0.028). The median estimated blood loss was 150 (20–2000) ml, with significant differences observed between the groups based on uterine type ( P  = 0.049). Only four (4/86, 4.7%) individuals required a transition to open surgery, with distinct differences identified in relation to the type of lesion ( P  = 0.019). The decision to convert to open surgery in all instances was prompted by excessive bleeding originating from the parametrium, particularly from the uterine vessels. Throughout the operations, a total of 4 (4/86, 4.7%) incidents of organ injuries were recorded, including one bladder injury and three vaginal lacerations, promptly identified and repaired during the surgery without subsequent complications. Additionally, six (6/86, 7.1%) cases necessitated intraoperative blood transfusions, and 11 (11/86, 12.8%) cases required postoperative blood transfusions. The postoperative transfusion rates varied significantly between Type A and Type B ( P  = 0.019). Among other post-operative complications, four (4/86, 4.7%) cases required medical intervention: one patient suffered a cuff infection and urinary tract infection requiring antibiotic therapy, one case of cuff dehiscence and bleeding leading to readmission and resolution after secondary suture, and one case of urinary retention requiring catheterisation for 2 weeks. There was also a significant difference ( P  = 0.006) between Type A and Type B in the rate of total surgical complications. The median length of postoperative hospital stay was 3 (2–14) days, with no significant differences observed between different groups. Notably, no cases of uterine sarcoma were identified in the final pathology. The results regarding surgical outcomes are shown in Table  2 . Table 2 Surgical data of enrolled patients Total Uterine weight Lesion site Adhesion severity 1000–1500 g 1500–2000 g  > 2000 g P Type A Type B P None-mild Moderate to severe P Patients, n (%) 86 65 (75.6) 15 (17.4) 6 (7.0) 33 (38.4) 53 (61.6) 77 (89.5) 9 (10.5) Operation time (min), median (range) 160 (75–390) 159 (75–354) 178.5 (133–269) 180 (125–390) 0.005 159 (70–354) 160 (80–390) 0.929 160 (75–390) 176.5 (125–355) 0.028 Estimated blood loss (ml), median (range) 150 (20–2000) 150 (20–2000) 150 (20–1500) 200 (100–1000) 0.053 150 (20–2000) 150 (20–1000) 0.049 150 (20–2000) 200 (100–600) 0.028 Intraoperation complications, n (%) 10 (11.6) 7 (10.8) 3 (20.0) 0 0.399 7 (18.2) 3 (5.7) 0.04 10 (13.0) 0 0.376 Conversion to laparotomy, n (%) 4 (4.7) 3 (4.6) 1 (6.7) 0 1.000 4 (12.1) 0 0.019 4 (5.2) 0 1.000 Transfusion in operation, n (%) 6 (7.1) 5 (7.7) 1 (6.7) 0 1.000 4 (12.1) 2 (3.8) 0.198 6 (7.8) 0 0.622 Bladder injury 1 (1.2) 0 1 (6.7) 0 0.244 1 (3.0) 0 0.384 1 (1.3) 0 1.000 Vaginal laceration 3 (3.5) 2 (3.1) 1 (6.7) 0 1.000 1 (3.5) 2 (3.8) 1.000 3 (3.9) 0 1.000 Postoperative complications, n (%) 14 (16.3) 10 (15.4) 2 (13.3) 2 (33.3) 0.598 8 (24.2) 6 (11.3) 0.139 14 (18.2) 0 0.344 Transfusion after operation, n (%) 11 (12.8) 7 (10.8) 2 (13.3) 2 (33.3) 0.334 8 (24.2) 3 (5.7) 0.019 11 (14.3) 0 0.273 Cuff infection 1 (1.2) 1 (1.5) 0 0 1.000 0 1 1.000 1 (1.3) 0 1.000 Urinary infection 1 (1.2) 1 (1.5) 0 0 1.000 0 1 1.000 1 (1.3) 0 1.000 Cuff dehiscence 1 (1.2) 1 (1.5) 0 0 1.000 0 1 1.000 1 (1.3) 0 1.000 Urinary retention 1 (1.2) 1 (1.5) 0 0 1.000 0 1 1.000 1 (1.3) 0 1.000 Total complications, n (%) 24 (27.9) 17 (26.2) 5 (33.3) 2 (33.3) 0.918 15 (45.5) 9 (17.0) 0.006 24 (31.2) 0 0.057 Total complication according to Clavien-Dindo Grade 1–2, n (%) 16 (18.6) 12 (18.5) 2 (13.3) 2 (33.3) 0.640 10 (30.3) 6 (11.3) 0.044 16 (20.8) 0 0.199 Total complication according to Clavien-Dindo Grade 3–4, n (%) 8 (9.3) 5 (7.7) 3 (20.0) 0 0.323 5 (15.2) 3 (5.7) 0.251 8 (10.4) 0 0.592 Hospital stay (d), median (range) 3 (2–14) 3 (2–14) 3 (3–7) 4 (3–7) 0.852 3 (3–12) 3 (2–14) 0.431 3 (2–14) 4 (3–7) 0.309 Data are expressed as mean ± standard deviation, absolute number (%), or median (range) Surgical data of enrolled patients Data are expressed as mean ± standard deviation, absolute number (%), or median (range) Logistic regression analysis revealed that only uterine lesion type was an independent predictor of overall surgical complications ( P  = 0.006). Lesions classified as Type A were associated with an increased risk compared to Type B (OR = 4.07, 95% CI 1.51–10.99). No significant influence of age, BMI, previous abdominal surgery, preoperative anaemia, adhesion severity or uterine weight was found as a predictor of total complications.

The medical records of patients who underwent RH with large uteri at the Senior Department of Obstetrics and Gynecology of Chinese PLA General Hospital were reviewed from January 2014 to December 2023. The study imposed no age restrictions. Eligibility criteria required (1) availability of complete uterine weight records and preoperative imaging (transvaginal ultrasound and/or pelvic MRI), and (2) pathologically confirmed benign uterine pathology. Exclusion criteria comprised (1) contraindications to minimally invasive surgery, (2) advanced endometriosis exceeding uncomplicated Enzian FA classification [ 15 ], and (3) prior pelvic radiotherapy or concurrent pelvic malignancies. Uterine weight, excluding the adnexa weight, was measured postoperatively and patients were categorized into three groups based on uterine weight: 1000 to < 1500 g, 1500 to < 2000 g, and ≥ 2000 g. Different types of uterine structures were categorized based on the placement of the prominent lesions: either in the lower segment and/or broad ligament myoma, and/or cervical region with an inverted “T” or “L” uterine shape classified as Type A, or in the fundus or upper segment without lesions in Type A location classified as Type B (Fig.  1 ). Adhesion severity were re-assessed using the modified American Fertility Society score [ 16 ] and classified into two groups: the none-mild group and the moderate to the severe group. Fig. 1 Schematic drawing and MRI images show different types of uterus categorized based on the placement of the prominent lesions. ① and ②: Lesions mailnly in the lower segment and/or broad ligament myoma, and/or cervical region with an inverted “T” or “L” uterine shape classified as Type A; ③: Lesions mainly in the uterine fundus or upper segment without lesions in Type A location classified as Type B Schematic drawing and MRI images show different types of uterus categorized based on the placement of the prominent lesions. ① and ②: Lesions mailnly in the lower segment and/or broad ligament myoma, and/or cervical region with an inverted “T” or “L” uterine shape classified as Type A; ③: Lesions mainly in the uterine fundus or upper segment without lesions in Type A location classified as Type B The study documented various demographic characteristics including age, body mass index (BMI), parity, history of abdominal surgery, operative time, uterine weight and pathology. Intraoperative data included operative time, estimated blood loss, conversion to laparotomy, intraoperative complications and postoperative details such as hospital stay, haemoglobin fall and postoperative complications. Operative complications were graded according to the Clavien-Dindo complication scale. Patients were followed up 1 month and 3 months after surgery. The length of hospital stay was calculated from the day after surgery until discharge, excluding patients with missing follow-up data. There were five high-volume surgeons [ 17 ] (LL, WF, WY, YM and CG) individual caseload > 200 hysterectomies/year) with da Vinci certification during the study period, but the main steps were similar to the conventional laparoscopic hysterectomy. The robot was positioned at the patient's foot side, and the surgery was conducted with three arms. Before starting the hysterectomy, we inserted a uterine manipulator (Atom Medical, Tokyo, Japan) and then docked the da Vinci Si robotic system with a 12 mm trocar for the camera positioned more than 3 cm above the uterine fundus. The intersection point between the horizontal line of the umbilical cord and the vertical line of the left transiliac anterior superior spine was the 2-arm puncture point with an 8 mm trocar, the symmetrical point on the right side was the 3-arm puncture point, and the midpoint between the lens hole and the 2-arm puncture point was the assistant puncture point with a 10 mm trocar. The point between the camera and the 3-arm puncture point was the 1-arm puncture point (Fig.  2 ). For Type A, the retroperitoneum was accessed through the pararectal and paravesical spaces to locate the ureters and parametrial tunnel leading to the bladder. Additionally, the uterine artery was tried to be ligated inside the internal iliac artery, with the ureter kept in sight, to achieve devascularization of the uterus at the outset of the procedure and reduce the risk of bleeding. The uterus was removed transvaginally using a cold scalpel “paper roll” coring technique [ 18 ]. Fig. 2 The image of robotic port sites in hysterectomy. “0” represent camera port; “1” represents the 1-arm puncture point; “2” represents the 2-arm puncture point; “3” represents the 3-arm puncture point; “4” represents the assistant puncture point The image of robotic port sites in hysterectomy. “0” represent camera port; “1” represents the 1-arm puncture point; “2” represents the 2-arm puncture point; “3” represents the 3-arm puncture point; “4” represents the assistant puncture point The SPSS 20.0 (IBM, Chicago, USA) program was used for statistical analysis. Kolmogorov–Smirnov test was performed to determine whether data were sampled from a Gaussian distribution. Variables with normal distribution were expressed as mean ± standard deviation, whereas non-Gaussian variables were expressed as median (range). The correlations between intraoperative data and various factors were analyzed by the chi-square test, Student’s t test or the nonparametric Kruskal–Wallis test. Considering our comparisons are descriptive and not hypothesis-driven, consistent with observational studies, adjustment of P -value was not applied. Logistic regression for binary outcomes was performed to investigate the association between perioperative complications and demographic characteristics. Variables were initially selected based on clinical relevance that prior literature reported (e.g., age, BMI, uterine weight, surgical history). In addition, we also specify that variables with P  < 0.10 in univariable analysis were retained for logistic regression. A two-sided P  < 0.05 was considered to indicate statistical significance.

Our present study shows that RH is feasible and safe even in cases of large uteri weighing more than 1000 g. Overall, the incidence of conversion to laparotomy and serious adverse events associated with RH was acceptable. Increasing uterine weight and adhesion severity showed a trend towards longer operative times. But they were not associated with the incidence of conversion to laparotomy, perioperative adverse events, or postoperative hospital stay. RH in large uteri, characterized by Type A, was associated with a relatively higher rate of conversion to laparotomy, more blood loss and higher postoperative blood transfusion rates. In other words, within the robotic surgery cohort, RH offers enhanced safety and reliability for Type B patients in comparison with Type A patients. Removing a large uterus represents a challenge, irrespective of the surgical approach. With advancements in surgical technique, the upper weight limit of minimally invasive resection of large uterus is constantly increasing. Robotic surgery using EndoWrist instruments and 3D cameras is essential for procedures in restricted surgical areas [ 19 ], and the mechanical elevation of the robotic arm provides more space compared to conventional laparoscopic surgery [ 20 ]. Currently, there is no standardized threshold value to define a large uterus, with varying criteria utilized across different sources [ 21 – 23 ]. Uccela et al. observed that approximately 1 in 17 patients who underwent hysterectomy presented with uteri weighing over 1000 g, indicating a considerable likelihood of encountering significantly large myomas [ 24 ], but data on robotic removal of giant uteri greater than 1000 g are still limited. Therefore, we used 1000 g as the threshold in our retrospective cohort. The median weight of the uterine specimens taken from our patient group was 1274 g, suggesting that the procedures performed by our surgical team may have been more complex than those in previous studies. Conversion to laparotomy during minimally invasive surgery is a significant complication. Laparoscopic and robotic conversion rates for uteri over 1000 g vary from 4.4 to 17.4% [ 21 , 25 , 26 ]. Previously reported independent risk factors include larger uterus size, BMI and a history of adhesions from previous surgery [ 27 ]. Our result revealed a relatively low conversion rate of 4.7%, primarily due to significant hemorrhage from the parametrium. The subgroup analysis showed that uterine weight is not the only factor in the decision to convert, while uterine type classified by lesion location may play a role in the decision. These results appear to be easily clarified in clinical practice. If the uterus is large enough to fill the entire abdominal cavity, its weight seems less important. However, lesions in the lower segment of uterine can impact adjacent blood vessels, leading to vessel stretching, irritation, and displacement observed during surgery. We have categorized the different types of uterus according to their position, in other words, whether they displace the vessel. A preliminary assessment of the uterine type can be made by preoperative imaging to help choose the surgical approach. It is also advised to utilize a vascular clamp or a suture technique when dealing with inflamed and enlarged blood vessels. The vessels should be blocked without cutting them at first, and then gradually removed to minimize bleeding from surrounding tissues. Although exposure of the internal iliac artery can sometimes be challenging, it is best to occlude the uterine artery directly from the onset of the internal iliac artery. Our results also suggest that the length of surgery tends to increase with increasing uterine weight and adhesion severity, regardless of the location of the lesion. This trend may be due to the complexity associated with a larger uterus, which makes it difficult to achieve a clear surgical field and limits instrument manoeuvrability. In addition, more time is often required for ureterolysis and adhesiolysis when treating severe adhesions. Although not investigated in this study, it is widely accepted that removing a bulky uterus would also result in a prolonged operative time. Typically, the “paper roll” technique is employed to systematically remove sections of the uterus from the outer layer to the inner layer. In instances where a portion of the tumor proves challenging to extract, a midline incision can be performed followed by segmental resection through mobilizing the incisional edge. Interestingly, existing literature indicates a correlation between an increased monthly volume of minimally invasive hysterectomy cases and a heightened occurrence of intraoperative and postoperative complications [ 17 ]. The rates of postoperative adverse events in our study by five high-volume surgeons were comparable to prior literature reporting on complication rates regardless of size, which suggests that RH for this size uteri is safe. However, it can be postulated that since higher-volume surgeons are less likely to convert to open, operative times are in turn prolonged and blood loss is increased, and for this reason, they incur increased perioperative morbidity. Regarding the amount of blood loss, several previous studies have shown that the amount of blood loss increases with greater uterine weight [ 12 , 14 , 25 ]. However, Carbonnel et al. found no significant difference in uterine weight and increased blood loss, suggesting a subtle difference [ 28 ]. Notably, a substantial enlargement of the uterus volumes frequently presents with abnormal uterine bleeding, potentially leading to anemia. In our study cohort, 19.8% of individuals presented with moderate anemia prior to surgery, while the same percentage of patients needed blood transfusions both during and after the operation, a rate exceeding those reported in comparable studies. Nevertheless, assessing surgical safety solely on transfusion rates presents challenges due to discrepancies in transfusion protocols among various medical facilities. The study has certain limitations that should be noted. First, the single-institution retrospective nature of this study inherently carries a risk of selection bias and may compromise the generalizability of our findings to diverse clinical populations. While our preliminary positive associations between uterine-type classification and surgical outcomes are encouraging, we acknowledge that the current classification system's reliance on retrospective anatomical assessments lacks standardized quantitative criteria. Second, the subgroup analysis of large uteri (≥ 2000 g) included only six patients, which limits the statistical power and precludes definitive conclusions for this high-risk population. Third, intraoperative uterine extraction time was not systematically documented, potentially affecting the positive findings. Additionally, the absence of patient-reported outcomes (e.g., quality-of-life metrics) and cost-effectiveness analyses impedes a comprehensive assessment of robotic surgery. Furthermore, although all procedures were performed by surgeons with expertise in robotic techniques, potential inter-surgeon variability in operative decision-making was not quantified or controlled. Lastly, the lack of direct comparisons with laparoscopic or vaginal approaches hinders a thorough evaluation of the relative advantages across minimally invasive modalities. Future multicentre prospective studies with standardized protocols, larger sample sizes, and detailed cost-effectiveness evaluations are necessary to address these limitations. Our retrospective study suggests that RH is feasible and safe even in cases of large uteri weighing more than 1000 g. Within the robotic surgery cohort, Type B uteri demonstrated enhanced safety and reliability compared to Type A uteri. These findings should be interpreted as preliminary evidence of differential outcomes specific to robotic surgery rather than absolute advantages of RH over other modalities. Surgeons are advised to meticulously strategize the surgical approach before the operation and exercise precision during the procedure, especially when considering the location of the lesion.

Hysterectomy is a widely practiced surgical procedure worldwide, with about 90% of cases performed to address benign conditions such as fibroids and adenomyosis [ 1 ]. Despite advances in minimally invasive techniques, 22–25% of hysterectomies for benign indications in the United States and France are still performed via laparotomy [ 2 , 3 ]. Prior to the advent of robotic surgery, laparoscopy served as the primary minimally invasive approach, offering advantages such as shorter hospital stays, faster recovery times and less post-operative pain [ 4 ]. Robotic surgery has advanced to provide supplementary advantages including enhanced dexterity, a three-dimensional perspective with improved depth perception, and ergonomic benefits for the operating surgeon [ 5 ]. These advancements demonstrate particular utility in complex scenarios involving obesity or large uterine size [ 6 ]. Hysterectomy is clinically indicated for patients presenting with refractory symptoms (e.g., severe anemia unresponsive to medical therapy, debilitating pain, or compressive sequelae) who demonstrate persistent symptomatology despite conservative management and have no desire for future fertility preservation [ 7 ]. Despite the potential advantages, performing a robotic hysterectomy (RH) for large uteri presents significant challenges. Previous studies have shown that as the uterine weight increases, the complexity of the procedure escalates, affecting perioperative outcomes including operative time, blood loss, and the likelihood of conversion to laparotomy [ 8 – 14 ]. In clinical practice, reliance on uterine weight alone for surgical guidance and outcome prediction appears inadequate. Other factors like pelvic adhesion and lesion sites can affect exposure of anatomical structures and uterine mobilization, impacting perioperative results. There is a lack of comprehensive data on robotic surgery for uteri weighing over 1000 g. Given the need for extensive experience to assess the true value of robotic surgery, this study intends to analyze cases of large uteri undergoing benign hysterectomies via robotic surgery at our high-volume medical institution over the past decade. The study aims to examine the influence of potential risk factors on surgical outcomes and offer recommendations on the optimal surgical technique for such cases.