Trends and comparative study of robotic-assisted benign hysterectomy among morbidly obese patients in united States.

The Premier Healthcare Database (PHD) comprises U.S. Health Insurance Portability and Accountability Act (HIPAA) -compliant hospital-based, service-level, all-payer information on inpatient discharges and outpatient encounters. It represents approximately 25% of annual U.S. hospitals inpatient admissions. Since the data were de-identified, the Mayo Clinic Institutional Review Board deemed this study exempt from approval. The study included adult women (≥ 18 years old) who had elective benign hysterectomy during 2013–2020. We excluded patients with pelvic organ prolapse, pelvic evisceration, standalone resection of the ovary or the fallopian tube, other obstetric procedures related to pregnancy (e.g. Episiotomy, C-section) and their associated complications, missing or extreme costs and operative room (OR) times ( 99th percentile). Benign elective hysterectomy was defined using primary International Classification of Disease (ICD), Ninth or Tenth Revision, Clinical Modification (ICD-9-CM or ICD-10-CM) procedure codes for inpatient (IP) and Current Procedural Terminology (CPT) codes for outpatient (OP) settings. Patients who had any of the hysterectomy codes along with an ICD/S2900 modifier code for a robotic-assisted surgery or a documented charge code for robotic instrumentation were considered to have had a robotic procedure (Supplemental Appendix A). Patient socio-demographic variables include age, race, marital status and patient type (inpatient or outpatient). Patient clinical characteristics included primary payor, BMI, charlson comorbidity index (CCI) score, concomitant surgical procedures (e.g., removal of ovary or tube, repair of pelvic organ prolapse, lymphadenectomy), indications for surgery (e.g., endometriosis, fibroids, uterine bleeding, or chronic pelvic pain) and type of hysterectomy (total, sub-total). The CCI quantifies an individual’s burden of disease and has been shown to be an independent predictor of surgical mortality [ 10 , 11 ]. We also used complexity three indicators : (1) Uterine size > 250 g, (2) Presence of adhesions and (3) Personal history of gynecologic surgery. The facility and physician characteristics include geographic region, location, hospital type, hospital-bed size and surgeon specialty. Physician volume was defined as number of malignant or benign hysterectomy procedures by the operating surgeon during the 12 months prior to the index procedure as a proxy for the level of experience of the surgeon at the time when the patients received healthcare services. The physician volume was classified into equal quartiles categorized as low (0–12), medium-low (13–36), medium-high (37–122), and high (123–370). Perioperative 30-day complications were determined by reviewing ICD-9/10-CM diagnosis codes for morbidity not present on admission and were classified as intraoperative (during surgery) or 30-day postoperative. Postoperative 30-day complications were further categorized as following: cardiovascular, pulmonary, gastrointestinal, wound/infection, genitourinary, ureteral, and bladder injury. Metrics of resource utilization analyzed included OR time, blood transfusions, conversion to laparotomy, and 30-day postoperative readmission related to the primary surgery. OR time was calculated based on hospital charger bills for OR utilization and was defined as the time patient entered the operating room until they left the room (“wheels in” to “wheels out”), including the anesthesia induction, patient positioning, docking time, and extubation. The surgical utilization rates for open (OH), vaginal (VH), laparoscopic (LH), and robot-assisted (RH) surgeries between 2013 and 2020 were calculated. Comparing surgical modality trends employed the Cochran-Armitage trend test. Further, the overall perioperative 30-day complication rate and healthcare utilization cost (inflation adjusted to 2020 US dollars) were estimated by year between 2013 and 2020. The Mann-Kendal trend test was used to assess mean and median cost changes over time, and the Cochran-Armitage test was utilized to test the linear trend for the perioperative 30-day complication rate. Over the 8-year study period, the frequency of the hysterectomy method was also evaluated by year, taking into account BMI class in the overall population and OP. An analysis comparing RH versus LH outcomes among the morbidly obese (BMI > 40 kg/m 2 ) was conducted using data from 2017 to 2019. The reason for restricting the analysis to the most recent 3 years was to avoid ICD 9 to ICD 10 code transition issues that came into effect prior to 2016. In addition, 2020 data were excluded as this was a COVID-19 affected year. A 1:1 propensity score matched (PSM) analysis was used to adjust for differences between morbidly obese patients undergoing LH and RH to compare clinical outcomes. We used a multivariable logistic regression model to calculate the propensity (probability) of RH versus LH based on patient characteristics (race, marital status, CCI score, IP/OP, payor, year); hospital and physician characteristics (provider region, hospital-bed size, teaching/non-teaching hospital, urban/rural, physician volume, surgeon specialty); indications for hysterectomy (endometriosis, fibroids, uterine bleeding, chronic pelvic pain); hysterectomy type and patient complexity level. Each case received a 0–1 probability score. A greedy matching method paired robotic-assisted and laparoscopic patients one-to-one based on the log-odds transformation of the propensity score 9 (i.e., the algorithm selects a robotic case and matches it randomly to closest propensity score laparoscopic control without replacement). Matches between groups were assessed using standardized mean difference (SMD) scores < 0.1 for each covariate 10 . Within each matched pair cohort, chi-square tests were used to examine differences in complications (overall, intraoperative, post-discharge 30-days (cardiovascular, pulmonary, gastrointestinal, wound/infection, genitourinary, ureteral and bladder injury), perioperative 30-days (bleeding and blood transfusion)), conversion rate, post discharge 30-days readmission rate. Wilcoxon rank sum test was used to test differences in OR time.The statistical analysis was performed using R version 4.1.2.

The most frequent surgical approach for patients receiving benign hysterectomy between 2013 and 2020 was LH (37.4%), followed by RH (32.7%), OH (23.8%), and VH (6.1%). During the time period studied, LH rose from 34.2% to 41.6%, RH from 31.8% to 36.5%, while OH declined by 10%. VH stayed stable ranging from 5 to 6%. In 2020, 84% of benign hysterectomy surgeries were MIS (LH (42%), RH (37%), and VH (5%)). In morbidly obese (BMI > 40 kg/m 2 ) individuals RH was the procedure of choice (36.5%), followed closely by LH (32.6%) (Fig.  1 ). During this time interval RH increased from 33.7% to 40.8%, LH increased from 27.4% to 37.6%, OH fell from 33.3% to 17.9%, and VH fell from 5.5% to 3.7%. By 2020, 83% of morbidly obese women’s benign hysterectomy surgeries were MIS [RH (41%), LH (38%), and VH (4%) procedures.] The 30-day postoperative complication rate for benign hysterectomy patients decreased from 11.5% to 9.0% between 2013 and 2020, whereas mean and median expenses increased ($602 and $589, respectively). This increase in cost represents about a 6–7% increase relative to 2013. Similarly, the 30-day postoperative complication rate decreased 5.7% in morbidly obese patients (BMI > 40 kg/m 2 ). However, in the morbidly obese patients, a statistically significant increase in mean and median expenditures($89 and $194, respectively) was not observed (Fig.  2 ). In this cohort, 67% of benign hysterectomies were done OP. Figure  3 shows breakdown of surgical approach by overall (IP and OP) and OP setting. The proportion of patients undergoing OP decreases slightly with increasing BMI (64% in BMI  50). The proportion of women undergoing RH rises with increasing BMI. RH represented 32% of hysterectomies in overall settings and 46% in OP settings amongst women with BMI  50 kg/m 2 ) women. Fig. 1 Trends in surgical routes for patients with morbid obesity (BMI > 40 kg/m2) undergoing benign hysterectomy Trends in surgical routes for patients with morbid obesity (BMI > 40 kg/m2) undergoing benign hysterectomy Fig. 2 Trends in 30 post-operative complication rates and healthcare costs among patients with morbid obesity (BMI > 40 kg/m2) undergoing for benign hysterectomy Trends in 30 post-operative complication rates and healthcare costs among patients with morbid obesity (BMI > 40 kg/m2) undergoing for benign hysterectomy Fig. 3 Surgical approach utilization by patient’s BMI status in US from 2013–2020 Surgical approach utilization by patient’s BMI status in US from 2013–2020 Between 2017 and 2019, we identified 20,418 morbidly obese (BMI > 40 kg/m 2 ) women who underwent benign hysterectomy. Figure  4 shows the study population flowchart for comparative analysis. Of these patients, 7879 (38.6%) had RH, 6944 (34%) LH, 4767 (23.3%) OH, and 828 (4.1%) VH. Table  1 displays the demographic, clinical, and hospital characteristics of individuals undergoing RH and LH. Gynecologic oncologists (29% RH, 11% LH), high-volume doctors (18% RH, 6.8% LH), and institutions with more than 300 beds (71% RH, 55% LH) favoured RH. Most RHs were carried out in urban as opposed to rural hospitals when compared with LH (95% vs. 85%). Fig. 4 Patient study population for the comparative analysis of different minimally invasive surgical approaches Patient study population for the comparative analysis of different minimally invasive surgical approaches Table 1 Demographic and clinical differences between patients undergoing laparoscopic and robot-assisted surgery Characteristic Laparoscopic, N  = 6944 (34%) Robotic, N  = 7879 (39%) OR 1 95% CI 1 p -value Age 18–34 Years 817 (12%) 722 (9.2%) — — 35–44 Years 2,987 (43%) 3,013 (38%) 0.99 0.87, 1.13 0.89 45–64 Years 2,852 (41%) 3,635 (46%) 1.04 0.91, 1.19 0.56 65 + Years 288 (4.1%) 509 (6.5%) 0.94 0.74, 1.19 0.61 Race Black 1,385 (20%) 1,557 (20%) — — Hispanic 426 (6.2%) 741 (9.6%) 1.3 1.11, 1.52 0.001 Other 233 (3.4%) 355 (4.6%) 1.35 1.10, 1.66 0.004 White 4,803 (70%) 5,098 (66%) 0.92 0.83, 1.01 0.088 Unknown 97 128 Marital Status Married 3,593 (52%) 4,330 (55%) — — Other 436 (6.3%) 161 (2.1%) 0.17 0.14, 0.22 < 0.001 Single 2,884 (42%) 3,338 (43%) 0.9 0.84, 0.98 0.01 Unknown 31 50 Charlson Comorbitity Score CCI = 0 5,096 (73%) 5,647 (72%) — — CCI = 1 1,431 (21%) 1,631 (21%) 1 0.92, 1.10 0.94 CCI > = 2 417 (6.0%) 601 (7.6%) 1.03 0.88, 1.19 0.73 Payor Commercial 4,390 (63%) 5,093 (65%) — — Medicaid 1,437 (21%) 1,377 (17%) 0.92 0.84, 1.02 0.12 Medicare 642 (9.2%) 990 (13%) 1.08 0.93, 1.25 0.3 Other 475 (6.8%) 419 (5.3%) 0.66 0.57, 0.77 < 0.001 Provider Census Region Midwest 1,937 (28%) 2,318 (29%) — — Northeast 465 (6.7%) 642 (8.1%) 1.1 0.95, 1.28 0.19 South 3,387 (49%) 3,722 (47%) 0.81 0.74, 0.88 < 0.001 West 1 , 155 (17%) 1,197 (15%) 0.6 0.53, 0.68 < 0.001 Provider Number of Beds 0 TO 99 Beds 732 (11%) 199 (2.5%) — — 100 TO 199 Beds 1,345 (19%) 1,153 (15%) 2.79 2.31, 3.37 < 0.001 200 TO 299 Beds 1,117 (16%) 973 (12%) 3.12 2.58, 3.79 < 0.001 300 TO 399 Beds 1,007 (15%) 1,432 (18%) 3.93 3.25, 4.76 < 0.001 400 TO 499 Beds 815 (12%) 1,065 (14%) 4.65 3.81, 5.69 < 0.001 500 + Beds 1,928 (28%) 3,057 (39%) 4.58 3.80, 5.55 < 0.001 Provider Teaching Status No 4,107 (59%) 4,283 (55%) — — Yes 2,837 (41%) 3,546 (45%) 0.75 0.68, 0.82 < 0.001 Urban or Rural Rural 1,021 (15%) 383 (4.9%) — — Urban 5,923 (85%) 7,496 (95%) 2.37 2.07, 2.72 < 0.001 Inpatient or Outpatient Inpatient 886 (13%) 831 (11%) — — Outpatient 6,058 (87%) 7,048 (89%) 1.15 1.01, 1.31 0.038 Physician Volume Low (0–12) 2,321 (33%) 1,641 (21%) — — Medium low (13–36) 2,521 (36%) 2,400 (30%) 1.28 1.17, 1.40 < 0.001 Medium High (37–122) 1,627 (23%) 2,420 (31%) 1.59 1.37, 1.84 < 0.001 High (123–370) 475 (6.8%) 1,418 (18%) 1.68 1.14, 2.48 0.009 Physician specialty Gynecological Oncology (GO) 760 (11%) 2,259 (29%) — — Obstetrics/Gynecology (OBG) 5,873 (85%) 5,398 (69%) 0.52 0.46, 0.59 < 0.001 Others or unknown 311 (4.5%) 222 (2.8%) 0.39 0.31 , 0.49 < 0.001 Hysterectomy Type Sub Total Hysterectomy 279 (4.0%) 218 (2.8%) — — Total Hysterectomy 5,227 (75%) 7,088 (90%) 1.24 0.99, 1.54 0.058 Other Hysterectomy 1,438 (21%) 573 (7.3%) 0.42 0.33, 0.53 < 0.001 Endometriosis 2,724 (39%) 3,289 (42%) 1.02 0.95, 1.10 0.54 Fibroids 3,758 (54%) 4,551 (58%) 1.15 1.06, 1.24 < 0.001 Inflammation or Infection 1,283 (18%) 1,725 (22%) 1.24 1.09, 1.41 0.001 Uterine Bleeding 4,730 (68%) 4,512 (57%) 0.82 0.76, 0.88 < 0.001 Chronic Pelvic Pain 2,255 (32%) 2,016 (26%) 0.94 0.87, 1.03 0.18 Endometrial Hyperplasia 734 (11%) 1,168 (15%) 0.93 0.83, 1.05 0.25 Removal Of Ovary Or Tube 6,487 (93%) 7,550 (96%) 1.15 0.98, 1.36 0.083 Repair Of Pelvic Organ Prolapse 76 (1.1%) 174 (2.2%) 2.08 1.54, 2.84 < 0.001 Lymphadenectomy 9 (0.1%) 73 (0.9%) 6.27 3.05, 14.7 < 0.001 BMI Category BMI 40–50 5,862 (84%) 6,340 (80%) — — BMI 50+ 1,082 (16%) 1,539 (20%) 1.1 1.00, 1.22 0.043 Large_Uterus 1,036 (15%) 1,376 (17%) 0.94 0.49, 1.86 0.85 Adhesions 1,707 (25%) 2,149 (27%) 0.83 0.43, 1.65 0.59 Morbid Obese (BMI > 40) 6,944 (100%) 7,879 (100%) Personal History of Surgery 401 (5.8%) 655 (8.3%) 1.15 0.60, 2.26 0.69 complexity Indicator BMI > 40 Only 4,207 (61%) 4,345 (55%) — — 1 Complex Condition 2,347 (34%) 2,919 (37%) 1.16 0.58, 2.24 0.67 2 + Plus Complex Conditions 390 (5.6%) 615 (7.8%) 1.36 0.34, 5.22 0.66 1 OR = Odds Ratio, CI Confidence Interval Demographic and clinical differences between patients undergoing laparoscopic and robot-assisted surgery 1 OR = Odds Ratio, CI Confidence Interval A subset of 4175 morbidly obese patients who underwent RH or LH were then matched using PSM. After matching, all covariates had an SMD score < 0.1 (Supplemental Fig.  1 ). RH was associated with a significantly lower 30-day perioperative bleeding rate (2.8% vs. 4.0%, p  = 0.003), a lower conversion rate (3.5% vs. 7.3%, p   50 kg/m 2 ) patients after matching indicated comparable clinical outcomes (Supplemental Fig.  2 ). RH was associated with a significantly lower rate of conversion to laparotomy (4.4% vs. 8.9%, p  < 0.001), with comparable post-perioperative 30-days complication rates, readmission rates and OR time (196 min vs. 198 min, p  = 0.6) to LH (Supplemental Table 1 ). Table 2 Matched clinical outcomes comparing laparoscopic vs. robotic surgery in women with morbid obesity (BMI > 40 kg/m2) Outcomes Laparoscopic, N  = 4157 (50%) 1 Robotic, N  = 4157 (50%) 1 p -value 2 Overall Complications (Post Discharge 30 Days) 240 (5.8%) 217 (5.2%) 0.3 Intraoperative Complications 64 (1.5%) 70 (1.7%) 0.6 Cardiovascular (Post Discharge 30 Days) 38 (0.9%) 32 (0.8%) 0.5 Pulmonary (Post Discharge 30 Days) 24 (0.6%) 32 (0.8%) 0.3 Gastrointestinal (Post Discharge 30 Days) 21 (0.5%) 23 (0.6%) 0.8 Wound and Infection (Post Discharge 30 Days) 65 (1.6%) 49 (1.2%) 0.13 Genitourinary (Post Discharge 30 Days) 39 (0.9%) 29 (0.7%) 0.2 Ureteral & Bladder Injury (Post Discharge 30 Days) 13 (0.3%) 8 (0.2%) 0.3 Bleeding Related (Perioperative 30 Days) 168 (4.0%) 118 (2.8%) 0.003 Blood Transfusion (Perioperative 30 Days) 74 (1.8%) 55 (1.3%) 0.092 Conversion 305 (7.3%) 144 (3.5%) < 0.001 Readmission (Post Discharge 30 Days) 105 (2.5%) 109 (2.6%) 0.8 OR Time in minutes 0.003 Median (IQR) 172 (135, 221) 165 (133, 210) Mean (SD) 186 (77) 179 (74) 1:1 Matching based on: Patient Characteristics: Race, Marital Status, CCI Score, Inpatient/Outpatient, Payor Type, admission year; Hospital and Physician Characteristics: Provider region, Hospital Bed Size, Teaching/Non-Teaching Hospital, Urban/Rural, Physician Volume, Surgeon specialty; Indication for Hysterectomy: Endometriosis, Fibroids, Uterine Bleeding, Chronic Pelvic Pain; Hysterectomy Type and patient complexity 1 n (%) 2 Pearson’s Chi-squared test; Fisher’s exact test; Wilcoxon rank sum test Matched clinical outcomes comparing laparoscopic vs. robotic surgery in women with morbid obesity (BMI > 40 kg/m2) 1:1 Matching based on: Patient Characteristics: Race, Marital Status, CCI Score, Inpatient/Outpatient, Payor Type, admission year; Hospital and Physician Characteristics: Provider region, Hospital Bed Size, Teaching/Non-Teaching Hospital, Urban/Rural, Physician Volume, Surgeon specialty; Indication for Hysterectomy: Endometriosis, Fibroids, Uterine Bleeding, Chronic Pelvic Pain; Hysterectomy Type and patient complexity 1 n (%) 2 Pearson’s Chi-squared test; Fisher’s exact test; Wilcoxon rank sum test

Our data showed that RH facilitated the widespread diffusion of MIS nationally for elective benign hysterectomies. Based on our findings, RH is associated with some improved outcomes compared to LH for morbidly obese patients undergoing benign hysterectomies. RH had a lower conversion rate and a shorter operating time. These results may suggest that the subset of morbidly obese patients may specifically benefit from the robotic-assisted hysterectomy.

Our analysis suggests the use of MIS is the preferred approach for morbidly obese patients in the US who are undergoing benign hysterectomy. Additionally, RH facilitated diffusion of MIS amongst patients with superobesity (BMI > 50 kg/m 2 ) with a safe complication profile. RH was associated with a reduction of complications over LH across the BMI spectrum. The main benefit of RH over LH was a significantly lower conversion rate to laparotomy and a slight decrease in OR time. Robotic gynecologic surgery became popular after the US Food and Drug Administration approval in 2005 [ 3 ]. In addition, recent studies in EC suggest that RH reduces perioperative morbidity without raising treatment-related 30-day costs [ 12 – 14 ]. Despite several studies on the trends and outcomes of different surgical approaches for benign hysterectomy [ 15 – 17 ], few have focused on morbidly obese patients [ 18 ]. The observed increase in RH implementation is consistent with prior benign gynecologic surgery studies [ 19 – 22 ]. Wright et al. [ 21 ] showed a substantial increase in the use of RH for benign indications across the US since the early-adoption phase. RH increased from 0.5% in 2007 to 9.5% of all hysterectomies in 2010. Furthermore, the rate of all RH between 2005 and 2013 rapidly increased, reaching 31.07% in 2013, the same proportion (31.55%) as LH [ 22 ]. Our study confirmed that RH has been steadily increasing over the years since its introduction. Our study adds significant data regarding the trends of MIS, focusing on the morbidly obese population undergoing benign hysterectomies. RH was more common in higher BMI classes. This may indicate surgical preference and technical advances of the robotic platform over laparoscopy for severely obese patients. Improved surgeon ergonomics during the robotic procedures, better exposure to the surgical field, and the use of a fourth arm are potential advantages which may explain the increased adoption of the robotic approach [ 23 , 24 ]. In the US, the majority of benign hysterectomies are completed in the OP [ 19 ]. A significant shift from the IP to OP setting was previously observed from 2008 to 2014 (+ 44.2% shift, P  < 0.0001). Our study echoed the similar migration, suggesting that RH may facilitate the adoption of OP benign hysterectomies even in complex surgical candidates such as patients with morbid obesity (BMI > 40 kg/m 2 ) [ 25 ]. In fact, RH may enable more OP surgery allowing patients to access gynecologic surgical care at more sites of service [ 12 ]. To our knowledge, the current analyses represent the first US evaluation of RH adoption and its impact on clinical outcomes for benign hysterectomies focusing on women with morbid obesity. Previous comparative analyses of RH vs. LH in unstratified BMI patients undergoing benign hysterectomy reported similar perioperative morbidity, with RH associated with longer surgery and higher costs [ 21 , 26 ]. Swenson et al. [ 27 ] showed similar major postoperative complications, readmissions, and reoperations between RH and other MIS approaches, while the RH had a longer operative time. Conversely, our study discovered decreased 30-day perioperative bleeding rates, lower conversion rates, and shorter OR times in morbidly obese patients. Moreover, our findings confirmed a lower conversion rate to laparotomy, even among superobese women undergoing RH hysterectomy. These benefits of RH are essential when operating on complicated patients. However, although these comparative data may be influenced by surgeon skill and expertise, especially in laparoscopic hysterectomy, our PSM also accounted for “physician volume” to homogenize the two groups. Moreover, Lim et al. [ 22 ] has shown that robotic-assisted surgery provides improved outcomes compared to laparoscopy even when both approaches are used by gynecologic surgeons with relevant high-volume experience. In addition, the surgery for EC revealed a relationship between BMI and conversion to laparotomy, with the likelihood of conversion rising as BMI rose [ 28 ]. Cusimano et al. [ 29 ] showed that RH and LH conversion rates were equal in patients with BMI > 30 kg/m², whereas LH was higher in those with BMI > 40 kg/m². Given that one of the primary causes of conversion is obesity-related anesthetic indications, such as intolerance of the Trendelenburg position, the fixed robotic arms supporting the weight of the abdominal wall may lower intra-abdominal pressure, therefore minimizing pulmonary complications [ 30 , 31 ]. Our results revealed that morbidly obese people can safely undergo RH. Additionally, the low complication rates in our analysis may drive the increased adoption of the RH, optimizing the treatment for obese patients with benign gynecologic disorders. Robotic surgery is criticized for taking longer and costing more. Our finding of shorter OR times with RH might be explained by the fact that we restricted our analysis to a more contemporary group of patients. Robotic surgeons, as a community, have gained experience. Given that each additional minute in the OR increases costs [ 32 ], the shorter or comparable surgery time of RH compared to LH may represent another important advantage of this MIS approach. We expect RH OR times to reduce relative to LH as more surgeons mature and more systematic robotic training programs [ 20 , 33 ] are implemented. Our study, observes an increasing uptake of MIS over OH for benign indications from 2013 to 2020. There was a simultaneous decline in the overall complication rates and no additional healthcare utilization costs associated with this trend over the 8-year study period in both the overall population and morbidly obese (BMI > 40 kg/m 2 ) patients undergoing benign hysterectomy. The potential reason for the cost offset could be the decrease in the complication rates and OP migration in recent years. Future cost-effectiveness studies are needed to assess the economic value of RH in benign gynecologic surgery. Of note, emerging data highlight the potential role of transvaginal natural orifice transluminal endoscopic surgery (vNOTES) as an alternative surgical approach for benign hysterectomy. This technique, which can incorporate conventional laparoscopic or robotic assistance, has gained popularity in gynecology, primarily due to its association with a short operative time, lack of visible scars, and superior cosmetic results achieved through the use of natural orifices [ 34 ]. Furthermore, recent comparative reports with “classic” laparoscopic hysterectomy indicate that vNOTES is a feasible, safe, and effective option, even in obese women requiring a hysterectomy [ 35 ]. While it offers a viable alternative to traditional techniques in select patient populations, additional research is warranted, particularly in the subset of morbidly obese patients.The main strengths of this study include the analysis of a large sample of patients with morbid obesity undergoing benign hysterectomy. Moreover, a comprehensive comparative analysis between the main MIS approaches was performed. There are several limitations. First, the PHD is a hospital administrative database with inherent limitations with respect to the completeness of the data. This study relied on ICD and CPT coding provided by the hospitals to Premier. Although coding variations exist within and across hospitals, the degree to which coding variation may affect this analysis is unknown. Hospitals that submit data to Premier may differ from non reporting hospitals due to the fact that Premier hospitals submit data to drive quality efforts, thus affecting the ability to generalize results to all U.S. hospitals. Yet, since PHD is U.S. hospital discharge database, outpatient prescription and the history of medical treatment is not available in the database. Moreover, this retrospective study is limited by potential selection bias in the allocation of patients in each surgical group. This limitation was minimized by PSM analyses. Second, the ICD 9 to 10 code transition issue (2013–2016) and the COVID impacted year (2020) prevented us from including more patients, the comparative analysis was limited to 3 years (2017–2019). Third, PHD does not capture visits outside the primary hospital IP/OP settings, and 30-day readmission rates were calculated based on readmission to the same hospital. Finally, PHD does not collect data on the cost of each surgical procedure (i.e., the cost of robotic instruments). In fact, no specific direct cost analysis was performed, while our cost findings relate to perioperative (30-day) costs.

The surgical techniques for benign gynecologic diseases continue to evolve as new technologies are adopted. According to an analysis of surgical data gathered in US between 1998 and 2010, open hysterectomies declined from 65% to 54% over time in favor of minimally invasive surgeries (MIS). By 2010, 8.2% of all hysterectomies were performed through robotic-assisted surgery, a substantial increase from 0.9% in 2008. This increase in robotic hysterectomies has been incorporated as evidence into professional society recommendations [ 1 ]. In 2015, a Cochrane systematic review of published clinical trials concluded, “When vaginal hysterectomy is not feasible, laparoscopic hysterectomy may avoid the need for open hysterectomy [ 2 ]. There is no evidence that robotic-assisted hysterectomy is of benefit in this population.” Later the 2017 American College of Obstetrician and Gynecologists (ACOG) committee report No. 701 echoed these conclusions but mentioned the increasing use of robotic-assisted laparoscopy and called for the generation of well-designed studies to compare the outcomes of these alternative techniques [ 1 ]. In 2020, ACOG issued a committee opinion (No. 810) acknowledging that use of robotic in the patients with noncancerous conditions (e.g., hysterectomy, myomectomy, endometriosis) was increasing although the evidence for robot-assisted hysterectomy was scarce [ 3 ]. They concluded that robotic-assisted gynecologic surgery can be performed safely in centers with experienced surgeons (ACOG, 2020). So far, experimental evidence comparing robotic-assisted to laparoscopy among gynecology patients with benign disease is sparse. Studies were underpowered with small numbers of randomized patients which precludes meaningful interpretation of the results [ 4 ]. Accumulating evidence highlighted that robotic platforms are particularly beneficial for morbidly obese (BMI > 40 kg/m 2 ) endometrial cancer (EC) patients [ 5 ]. Robotic-assisted surgery achieved comparable or slightly better outcomes than laparoscopy (e.g., fewer wound complications, less blood loss, lower conversion rates, shorter hospital stay) [ 6 – 8 ]. Chan and colleagues examined data from the National Inpatient Sample and reported that over 80% of MIS on morbidly obese EC patients utilized robotic approach [ 9 ]. Whether these findings are generalizable to a population of women with benign gynecological disease is unknown. Our primary objective is to compare perioperative outcomes and complications of robotic-assisted and laparoscopic benign hysterectomies among patients with morbid (BMI > 40 kg/m 2 ) and supermorbid (BMI > 50 kg/m 2 ) obesity. As a secondary aim, we analyzed the trends in the frequency of different surgical routes over the study period.

Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1 Supplementary Material 2 Supplementary Material 2 Supplementary Material 3 Supplementary Material 3