Abstract
Background and Objectives:
Minimally invasive approaches to benign hysterectomy are the current standard of care when feasible. Use of robotic-assisted laparoscopic hysterectomy (RA-LH) has been increasing; however, direct comparative data that accounts for uterine weight in conventional laparoscopic hysterectomy (CLH) and RA-LH is limited. We sought to examine the impact of uterine weight on immediate perioperative morbidity in CLH versus RA-LH. The primary outcome was a composite of complications including visceral injuries, conversions to abdominal procedures, and transfusions.
Methods:
A retrospective cohort study of patients who underwent a minimally invasive laparoscopic hysterectomy (CLH and RA-LH) in a single hospital system between January 1, 2014 and December 31, 2017 as identified by Current Procedural Terminology codes. The primary exposure was CLH or RA-LH. Uterine weight was categorized into four groups: <150 g, 150 to < 250 g, 250 to < 450 g, and ≥ 450 g.
Results:
A total of 1506 patients were included; 539 underwent CLH and 967 underwent RA-LH. Median uterine weight was higher in patients who underwent CLH (161.0 g) compared to RA-LH (147.0 g), P = .001. The odds of the composite of complications in CLH was 4.43 (2.84 – 6.92) higher than the odds of the composite in RA-LH. When stratified by the uterine weight, the odds of complications was significantly higher in CLH in the following categories: <150 g, 250 to < 450 g, and ≥ 450 g (OR: 4.41, 3.28, and 7.81, respectively).
Conclusion:
Surgical morbidity was lower in RA-LH across the spectrum of uterine weights compared to CLH. Patients may particularly benefit from RA-LH at higher uterine weights.
Keywords: Complications, Minimally invasive surgery, outcomes, Uterine size
INTRODUCTION
Hysterectomy is one the most frequently performed gynecologic surgeries in the United States with over 400,000 performed annually.1 Minimally invasive approaches have increased given evidence of superior outcomes compared to abdominal approaches.1–5 Current recommendations state that minimally invasive approaches should be performed for benign hysterectomy when feasible.6–7 Hysterectomy is frequently indicated for conditions that preclude use of a vaginal approach such as adnexal pathology, pelvic adhesive disease, or large uterine size; in such cases, laparoscopic or robotic hysterectomy is the preferred alternative that offers benefits of a minimally invasive approach.6,8,9
Since its introduction in 2001, the use of robotic-assisted laparoscopic hysterectomy has been increasing due to purported advantages of three-dimensional visualization, higher magnification, and improved ergonomics.10–15 Studies comparing operative outcomes of conventional laparoscopic hysterectomy (CLH) and robotic-assisted hysterectomy (RA-LH) have found mixed results;8,18,19 some demonstrating improved outcomes using a robotic approach while others finding no difference.8,15–17 Operative outcomes in both approaches are significantly impacted by uterine weight with increased complications, blood loss, and operative time at higher weights.18–19
Despite this association, direct comparative data on the impact of uterine weight on laparoscopic hysterectomy with or without robotic assistance is limited. Thus, we sought to examine the impact of uterine weight on immediate perioperative morbidity in CLH versus RA-LH.
METHODOLOGY
We performed a retrospective cohort study of patients who underwent a laparoscopic or robotic hysterectomy in a large-scale academic community hospital system between January 1, 2014 and December 31, 2017. This study was reviewed and approved by the Institutional Review Board. Patients were identified via Current Procedural Terminology codes for laparoscopic hysterectomy or robotic-assisted hysterectomy: 58570 – 58573, 58550 – 58554, and 58541 – 58544. Codes for both hysterectomy approaches included those with and without bilateral salpingo-oophorectomy. Procedures were confirmed after review of operative reports and were classified into two groups: CLH and RA-LH. Patients were excluded if they had a gynecologic malignancy diagnosis, were under the age of 18 years, had a missing uterine weight, or had a concomitant major procedure at the time of hysterectomy such as a procedure for pelvic organ prolapse, lymph node dissection, or a planned procedure in combination of another specialty. Minor concomitant procedures that included were cystoscopy, ureteral stent placement, bladder or vulvar biopsy, intraabdominal biopsy, lysis of adhesions, intrauterine device or hormonal implant removal, or vaginal laceration repair. Pathology reports were reviewed to obtain uterine weights. Gynecologic generalists, both hospital-employed and private practice, as well as surgeons from minimally invasive gynecologic surgery (MIGS), urogynecology, and gynecologic oncology performed all hysterectomies in this study. The hospital system is a large-scale teaching institution with residents in obstetrics and gynecology and fellows in MIGS and urogynecology regularly participating in nearly all hysterectomies.
Data were obtained from the charts of patients who met inclusion criteria. Basic demographic information including age, body mass index, race, and government insurance was obtained for analysis. Race was patient-reported and included: American Indian or Alaska Native, Asian, Black or African American, declined/not available, more than one race, Native Hawaiian or other Pacific Islander, or White. Eighty-five percent f individuals selected either “Black or African American” or “White,” thus we included an “Other” category for the remaining 15%. Pertinent medical history including hypertensive or diabetes diagnosis, current or former tobacco use, abnormal uterine bleeding or leiomyomata diagnosis, preoperative hemoglobin within 30 days of surgery, and use of chronic anticoagulation were variables extracted from our electronic health record database. Prior abdominal surgery, intraoperative lysis of adhesions, surgical indication, supracervical versus total hysterectomy, surgeon specialty performing hysterectomy, and surgeon volume were also analyzed. Surgeon specialty was divided into gynecologic generalists and gynecologic subspecialists, which included urogynecologists, minimally-invasive gynecologic surgeons, and gynecologic oncologists. Total number of hysterectomies performed per surgeon during the study period was reported. Annual surgeon volume was calculated based on total number of hysterectomies per year with low-volume defined as fewer than 12 per year and high-volume as 12 or greater; which is based on the definition used by Mowat et al. in a systemic review of surgical outcomes of low-volume vs high-volumes surgeons in gynecologic surgery.20
Surgical outcomes and complications examined included estimated blood loss (EBL), operative time, intraoperative visceral injury (bladder, ureter, bowel, or other organ), and the unanticipated need to convert to an open procedure. EBL was subjectively estimated by the surgeon in concert with anesthesia, operative time was calculated by incision to closure time, and intraoperative visceral injury was determined by review of operative reports. Transfusion within 30 days of surgery was an outcome also collected for analysis as most occur in the immediate postoperative period.
Our primary outcome was a composite of complications that contribute to surgical morbidity, which was defined as a one or more of the following: visceral injury, conversion to an abdominal procedure, or transfusion. A composite outcome was used due to the low frequency of the included variables. Important secondary outcomes included blood loss EBL ≥ 500 mL and operative time ≥ 240 minutes, both values cited in previously studies of hysterectomies including one by Sirota et al. who defined EBL ≥ 500 mL as a perioperative complication21 and Catanzarite et al. who found operative time ≥ 240 minutes to be associated with significantly increased surgical complications.22
Data were stratified according to surgical approach (CLH and RA-LH) to assess differences in demographic and clinical variables along with surgical outcomes. Descriptive statistics including mean, standard deviation and 95% confidence intervals (CI), or median and interquartile range were calculated for each continuous variable. Student t test or Mann-Whitney U test were performed to compare between two groups. Pearson’s χ2 test or Fisher exact tests were used to examine differences in categorical variables between groups where appropriate.
Uterine weights were categorized into four groups based on ascending weight in grams (g): < 150 g, 150 to < 250 g, 250 to < 450 g, and ≥ 450 g. These four groups were selected to allow for large, balanced categories for analysis while stratifying among the most common uterine weights given most uteri are < 150g. Uterine weight categories were compared within each type of hysterectomy to assess the odds of complications. Odds ratios (OR) and 95% confidence intervals (CI) comparing to the referent, < 150 g, were calculated accordingly. A P-value of < 0.05 was considered statistically significant.
Logistic regression was used to analyze the association between hysterectomy approach and the primary and secondary outcomes (composite of complications, prolonged operative time and blood loss > 500 mL, respectively) across each uterine weight category for each hysterectomy approach. All statistical analyses were performed using SAS 9.4 (Cary, NC).
RESULTS
Of the 3,787 patients who underwent a hysterectomy between January 1, 2014 – December 31, 2017, 1,506 patients met inclusion criteria for analysis; 539 patients underwent CLH (35.8%) and 967 underwent RA-LH (64.2%). Patient demographics between surgical approach differed in age, racial and ethnic distribution, and insurance type (Table 1). Patients who underwent a RA-LH were on average three years older, and they were more likely to be of Black or African American race and have government insurance (Table 1). Those with abnormal uterine bleeding and a slightly lower pre-operative hemoglobin (0.3 g/dL difference) more commonly underwent a CLH whereas those with a surgical indication of endometrial hyperplasia or an adnexal mass more often underwent a RA-LH (Table 1). Gynecologic generalists performed 52.5% of CLH and 49.0% of RA-LH and gynecologic subspecialists performed 47.5% of CLH and 51.0% of RA-LH with no difference between groups, P = 0.195. Median surgeon volume was lower in CLH at 98.0 (29.0 – 156.0) compared to RA-LH 177.0 (96.0 – 547.0), P < 0.001. High volume surgeons performed majority of the hysterectomies at 87.1%, which included 74.2% of CLH and 94.2% of RA-LH, P < 0.001 (Table 1). Overall, there was a low number of supracervical hysterectomies performed, 58/1506 (3.9%). However, there was a significantly higher amount in CLH (50/539 [9.3%]) compared to RA-LH (8/967 [0.8%]), P < 0.001 (Table 1).
Table 1.
Demographic Characteristics
| Total | Route of Hysterectomy | P value | ||
|---|---|---|---|---|
| CLH | RA-LH | |||
| n (%) | 1506 (100) | 539 (35.8) | 967 (64.2) | |
| Age (yr), mean ± SD | 47.1 ± 0.2 | 45.3 ± 0.4 | 48.2 ± 0.3 | < .001 |
| BMI (kg/m2), mean ± SD | 31.1 ± 0.2 | 30.7 ± 0.3 | 31.3 ± 0.2 | .133 |
| *Race and ethnicity, n (%) | < .001 | |||
| Black or African American | 969 (68.5) | 298 (58.9) | 671 (73.9) | |
| White | 384 (27.2) | 189 (37.4) | 195 (21.5) | |
| None of the above | 61 (4.3) | 19 (3.7) | 42 (4.6) | |
| Tobacco use, n (%) | 557 (37.8) | 206 (39.6) | 351 (36.8) | .279 |
| Government insurance, n (%) | 127 (8.4) | 33 (6.1) | 94 (9.7) | .016 |
| Hypertension, n (%) | 408 (27.1) | 148 (27.5) | 260 (26.9) | .811 |
| Hgb (g/dL), mean ± SD | 12.4 ± 0.06 | 12.2 ± 0.09 | 12.5 ± 0.07 | .013 |
| Abnormal uterine bleeding, n (%) | 498 (33.1) | 209 (38.8) | 289 (29.9) | < .001 |
| Leiomyomata, n (%) | 914 (60.7) | 343 (63.6) | 571 (59.0) | .081 |
| Anticoagulation, n (%) | 5 (0.3) | 1 (0.2) | 4 (0.4) | .461 |
| Prior abdominal surgery, n (%) | 426 (48.5) | 184 (46.6) | 242 (50.1) | .299 |
| Lysis of adhesions, n (%) | 179 (11.9) | 62 (11.5) | 117 (12.1) | .732 |
| Gynecologic subspecialty | .195 | |||
| Generalist, n (%) | 757 (50.3) | 283 (52.5) | 474 (49.0) | |
| Subspecialist, n (%) | 749 (49.7) | 256 (47.5) | 493 (51.0) | |
| Surgeon volume, median (IQR) | 156.0 (75.0-547.0) | 98.0 (29.0-156.0) | 177.0 (96.0-547.0) | < .001 |
| Annual surgeon volume | < .001 | |||
| Low-volume, n (%) | 195 (12.9) | 139 (25.8) | 56 (5.8) | |
| High-volume, n (%) | 1311 (87.1) | 400 (74.2) | 911 (94.2) | |
| Indication for hysterectomy, n (%) | ||||
| Abnormal uterine bleeding | 967 (64.2) | 395 (73.3) | 572 (59.2) | < .001 |
| Prolapse | 8 (0.5) | 7 (1.3) | 1 (0.1) | .003 |
| Pelvic pain | 409 (27.2) | 152 (28.2) | 257 (26.6) | .497 |
| Cervical dysplasia | 98 (6.5) | 27 (5.0) | 71 (7.3) | .079 |
| Endometrial hyperplasia | 82 (5.4) | 14 (2.6) | 68 (7.0) | < .001 |
| Adnexal mass | 178 (11.8) | 39 (7.2) | 139 (14.4) | < .001 |
| Risk-reducing | 43 (2.9) | 9 (1.7) | 34 (3.5) | .039 |
| Other | 14 (0.9) | 4 (0.7) | 10 (1.0) | .571 |
| Supracervical hysterectomy | 58 (3.9) | 50 (9.3) | 8 (0.83) | < .001 |
CLH, conventional laparoscopic hysterectomy; RA-LH, robotic-assisted laparoscopic hysterectomy; SD, standard deviation; BMI, body mass index; Hgb, preoperative hemoglobin; IQR, interquartile range.
*Race was self-report.
Bold indicates statistical significance.
The median uterine weight was higher in patients who underwent CLH (161.0 g) versus RA-LH (147.0 g), P = 0.001 (Table 2). Our primary outcome, a composite of complications, was significantly lower in RA-LH at 3.1% (30/967) compared to CLH at 12.4% (67/539), P < 0.001. Unanticipated conversions accounted for the greatest proportion of complications (54/539 [10.0%] in CLH and 16/967 [3.1%] in RA-LH), P < 0.001. Transfusions were infrequent; however, were higher in CLH with 13 transfusions (2.4%) and 4 in RA-LH (0.4%), P < 0.001. There were no differences in intraoperative visceral injuries between hysterectomy approaches. 10/26 (38.5%) of visceral injuries affected the urinary tract with 6/10 in CLH and 4/10 in RA-LH; 7/10 (70.0%) occurred in uteri of 150 g to < 250 g. There were five vascular injuries with 3/5 in CLH and 2/5 in RA-LH; 60.0% occurred in uteri of 150 g to < 250 g. Small bowel injuries accounted for 5/26 (19.2%) visceral injuries and large bowel injuries occurred in 4/26 (15.4%) of hysterectomies with an intraoperative visceral injury. There was one unspecified bowel injury and one hepatic trocar injury. Of the ten total bowel injuries, 5/10 occurred in CLH and 5/10 occurred in RA-LH. 60% occurred in uteri < 250 g.
Table 2.
Operative Outcomes
| Total | Route of Hysterectomy | P value | ||
|---|---|---|---|---|
| CLH | RA-LH | |||
| Uterine weight (grams), median (IQR) | 150.3 (159) | 161.0 (199.2) | 147.0 (147) | .001 |
| EBL (mL), median (IQR) | 50 (125) | 100 (200) | 50 (75) | < .001 |
| Operative time (min), mean ± SD | 143.9 ± 1.6 | 172.1 ± 2.7 | 128.2 ± 1.7 | < .001 |
| Composite of complications | 97 (6.4) | 67 (12.4) | 30 (3.1) | < .0001 |
| Visceral injury, n (%) | 26 (1.7) | 14 (2.6) | 12 (1.2) | .15 |
| Conversion to open, n (%) | 70 (4.6) | 54 (10.0) | 16 (1.6) | < .001 |
| Transfusion, n (%) | 17 (1.1) | 13 (2.4) | 4 (0.4%) | < .001 |
CLH, conventional laparoscopic hysterectomy; RA-LH, robotic-assisted laparoscopic hysterectomy; IQR, interquartile range; EBL, estimated blood loss; SD, standard deviation.
Bold indicates statistical significance.
There were higher odds of the composite outcome of complications as uterine weight increased compared to a reference of < 150 g in combined analysis of hysterectomy approaches: OR of 2.04 (1.11 – 3.76) in uteri 150 - < 250 g, OR of 3.15 (1.71 – 5.80) in uteri 250 - < 450 g, and 9.17 (5.15 – 16.3) in uteri of ≥ 450 g (Table 3). In analysis of RA-LH alone, this trend was similar. When comparing approaches, CLH had a 4.43 (2.84 – 6.92) higher odds of the composite outcome than RA-LH overall. Evaluation of each weight category showed that CLH had a significantly higher odds of the composite in most categories compared to RA-LH: 4.41 (1.78 – 11.00) in uteri <150 g, 3.28 (1.32 – 8.15) in uteri 250 - < 450 g, and 7.81 (2.81 – 21.76) in uteri ≥ 450 g.
Table 3.
Operative Outcomes and Uterine Weight
| < 150 g | 150 - < 250 g | 250 - < 450 g | ≥ 450 g | Total | |
|---|---|---|---|---|---|
| All, n = 1506 | 747 | 360 | 252 | 147 | |
| Total Composite, n (%) | 22 (3.0) | 21 (5.8) | 22 (8.7) | 32 (21.8) | |
| OR (95% CI) | ref | 2.04 (1.11 – 3.76) | 3.15 (1.71 – 5.80) | 9.17 (5.15 – 16.34) | |
| CLH, n = 539 | 252 | 119 | 94 | 74 | |
| Composite, n (%) | 15 (5.95) | 11 (9.24) | 14 (14.89) | 27 (36.49) | |
| OR (95% CI) | ref | 1.61 (0.72 –3.62) | 2.77 (1.28 – 5.98) | 9.08 (4.49 – 18.36) | |
| RA-LH n = 937 | 495 | 241 | 158 | 73 | |
| Composite, n (%) | 7 (1.41) | 10 (4.15) | 8 (5.06) | 5 (6.85) | |
| OR (95% CI) | ref | 3.02 (1.13 – 8.03) | 3.72 (1.33 – 10.42) | 5.13 (1.58 – 16.61) | |
| CLH vs RA-LH* | |||||
| OR (95% CI) | 4.41 (1.78 – 10.97) | 2.35 (0.97 – 5.71) | 3.28 (1.32 – 8.15) | 7.81 (2.81 – 21.76) | 4.43 (2.84 –6.92) |
| Total EBL > 500 mL, n (%) | 13 (1.7) | 18 (5.0) | 15 (6.0) | 23 (15.7) | |
| OR (95% CI) | ref | 2.97 (1.44 – 6.14) | 3.57 (1.68 – 7.62) | 10.47 (5.17 – 21.22) | |
| CLH, n = 539 | 252 | 119 | 94 | 74 | |
| EBL > 500 mL, n (%) | 8 (3.17) | 14 (11.76) | 10 (10.64) | 17 (22.97) | |
| OR (95% CI) | ref | 4.07 (1.66 – 9.99) | 3.63 (1.39 – 9.50) | 9.10 (3.74 – 22.12) | |
| RA-LH n = 937 | 495 | 241 | 158 | 73 | |
| EBL > 500 mL, n (%) | 5 (1.01) | 4 (1.66) | 5 (3.16) | 6 (8.22) | |
| OR (95% CI) | ref | 1.65 (0.44 – 6.22) | 3.20 (0.92 – 11.21) | 8.78 (2.61 – 29.55) | |
| CLH vs RA-LH* | |||||
| OR (95% CI) | 3.21 (1.04 - 9.93) | 7.90 (2.54 – 24.57) | 3.64 (1.21 – 11.01) | 3.33 (1.23 – 9.01) | 4.74 (2.78 – 8.06) |
| Total Op Time > 240 min, n (%) | 28 (3.8) | 2 (5.6) | 20 (7.9) | 38 (25.9) | |
| OR (95% CI) | ref | 1.51 (0.84 – 2.72) | 2.21 (1.22 – 4.00) | 8.95 (5.28 – 15.18) | |
| CLH, n = 539 | 252 | 119 | 94 | 74 | |
| Op time > 240 min, n (%) | 16 (6.35) | 17 (14.29) | 10 (10.64) | 23 (31.08) | |
| OR (95% CI) | ref | 2.46 (1.20 – 5.06) | 1.76 (0.77 – 4.02) | 6.65 (3.28 – 13.48) | |
| RA-LH, n = 937 | 495 | 241 | 158 | 73 | |
| Op time > 240 min, n (%) | 12 (2.42) | 3 (1.24) | 10 (6.33) | 15 (20.55) | |
| OR (95% CI) | ref | 0.51 (0.14 – 1.82) | 2.72 (1.15 – 6.42) | 10.41 (4.65 – 23.32) | |
| CLH vs RA-LH* | |||||
| OR (95% CI) | 2.73 (1.27 – 5.86) | 13.22 (3.79 – 46.10) | 1.76 (0.71 – 4.41) | 1.74 (0.82 – 3.70) | 3.23 (2.15 – 4.86) |
g, grams; CLH, conventional laparoscopic hysterectomy; RA-LH, robotic-assisted laparoscopic hysterectomy; Op time, operative time; min, minutes; EBL, estimated blood loss; mL, milliliters; OR, odds ratio; CI, confidence interval; ref, reference group.
Bold indicates statistical significance.
EBL and operative time were also higher in those who underwent CLH versus RA-LH (100 mL versus 50 mL, P < 0.001) and (172.1 ± 2.7 min versus 128.2 ± 1.7 min, P < 0.001), respectively.
Combined analysis of both hysterectomy approaches showed higher odds of blood loss > 500 mL as uterine weight increased beginning at weights > 150 g (Table 3). This trend was similar when analyzing CLH alone. However, in RA-LH the odds of blood loss > 500 mL only became significantly higher in uteri 450 g or greater. The odds of blood loss > 500 mL was higher across all uterine weight categories in CLH compared to RA-LH (< 150 g [3.21 (1.04 – 9.93)], 150 – < 250 g [7.90 (2.54 – 24.57)], 250 – < 450 g [3.64 (1.21 – 11.01)], and ≥ 450 g [3.33 (1.23 – 9.01)] (Table 3).
There were higher odds of prolonged operative time in uteri 250 g or greater in combined analysis of both hysterectomy approaches (Table 3). This finding was similar in RA-LH. Comparing CLH to RA-LH, there was 3.23 (2.15 – 4.86) higher odds of prolonged operative time in CLH compared to RA-LH overall. Stratification by uterine weight showed higher odds of prolonged operative time in CLH in uterine weight categories < 250 g compared to RA-LH but not over 250 g (Table 3).
DISCUSSION
This single institution study, which examined the impact of uterine weight on immediate perioperative morbidity in laparoscopic compared to robotic hysterectomy, found that CLH had four times higher odds of complication than RA-LH overall. Moreover, we found the odds of surgical complications was higher in larger uteri (>250 g) regardless of approach which was more pronounced in uteri > 450 g.
Upon stratification by ascending uterine weight categories of <150 g, 150 to < 250 g, 250 to < 450 g, and ≥ 450 g, we found a significantly higher odds of complications in CLH in all categories except 150 to < 250 g compared to RA-LH. We noted a greater proportion of CLH performed in the > 450 g group at 13.7% versus 7.8% in RA-LH which may impact the higher complication rate, however the proportion of RA-LH in the 250 to < 450 g group was higher than CLH and, similarly, demonstrated higher complication rates in CLH. As uterine weight increased, the odds of blood loss > 500 mL was also higher across all CLH categories, and the odds of prolonged operative time was only significantly higher in CLH when uteri were less than 250 g.
Our results demonstrate that RA-LH has a lower odds of a composite outcome of complications (visceral injury, conversion, transfusion), blood loss > 500 mL, and prolonged operative time compared to CLH. These findings are consistent with a number of prior studies; however, results have been mixed due to sample size limitations, learning curves during implementation of robotics, and inclusion of only select uterine weights. Furthermore, previous studies have not evaluated the impact of uterine weight when comparing CLH versus RA-LH. Prior studies of laparoscopic hysterectomies alone that have evaluated the impact of uterine weight have found increased complications, operative time, and blood loss with increasing uterine weight; however, robotic hysterectomies were not directly compared to the conventional laparoscopic approach. They may have been included in analysis; however, many large national databases code both hysterectomy approaches as laparoscopic.18–19
In the limited number of comparative studies between CLH and RA-LH, results have also been inconsistent. Advantages to RA-LH, including lower blood loss, have been documented though without differences in intraoperative complications or conversions.8,15–17,23 Disadvantages such as longer operative time and higher costs have also been cited.8,9,17,24 Most studies, many of which are limited by sample size, have not demonstrated differences in complications,11,25 and a recent systemic review by Alshowaikh et al. that included over 50,000 patients similarly concluded that there was no difference in surgical or patient outcomes.24 Most these studies were performed during implementation of robotics which likely impacted robotic outcomes.8,9,16 Payne et al. found the opposite effect on operative time with surgeon/team experience and demonstrated cost savings with RA-LH at higher uterine weights.8,26 Similarly, Alshowaikh et al. concluded that RA-LH may be more profitable in complex cases.24 Both single institution studies and systemic reviews have noted advantages to RA-LH to be most apparent at higher uterine weights,24,26 which is similar to our findings. Although prior studies lack stratification by uterine weight; we demonstrate that complications, blood loss > 500 mL, and prolonged operative time increase at a higher rate in CLH compared to RA-LH as uterine weight increases with a notable seven-fold higher odds of complications in CLH when uterine weight was 450 g or greater.
In addition to evaluating the impact of uterine weight, this studies’ strengths include a large sample size, examination of outcomes after the early robotic implementation period with experienced teams and surgeons, and 53 total surgeons including benign gynecology and all gynecologic surgical subspecialties with no difference between hysterectomy approach in generalists versus subspecialists. Surgeon experience and surgeon preference may impact selection of operative approach and impact outcomes independent of uterine weight and hysterectomy approach. While total hysterectomies per surgeon during the study period were higher in RA-LH, high-volume surgeons performed nearly 90% of hysterectomies included in this study with multiple generalists and subspecialists regularly performing both CLH and RA-LH. Further stratification by surgeon volume was limited by small sample size of hysterectomies performed by low-volume surgeons. Although surgeon experience and operative outcomes is also impacted by involvement of trainees, residents and fellows are involved in nearly every hysterectomy regardless of approach or uterine weight thus is unlikely to alter study outcomes.
As with any retrospective study, ours is limited by the available data, lack of randomization, and potential coding errors in electronic health record databases. While over 1,500 patients were included, patient differences between approaches were notable for a higher incidence of leiomyomata, abnormal uterine bleeding, lower hemoglobin, and higher uterine weights in patients who underwent CLH, which could account for the increased conversions, blood loss, and transfusions in this study. However, transfusions were rare, and the statistically significant difference in hemoglobin may not be clinically significant given both were > 12.0 g/dL. Comparisons between blood loss is limited by the subjective estimation by individual surgeons; however, it is estimated in consultation with anesthesia at this institution and backed up by transfusion data as routine postoperative hemoglobin values were not obtained. While surgeons may differ in threshold for transfusion, standard practice is for transfusion for hemoglobin values < 7.0 g/dL or < 8.0 g/dL if signs or symptoms of anemia are present, which is reviewed by the blood bank prior to issuing blood products. On manual review of transfusions administered in this study, all patients had hemoglobin values < 8.0 g/dL except two whose hysterectomies were complicated by significant blood loss.
While we categorized all patients with uterine weights ≥ 450 g together to focus on the most common uterine weights and allow adequate sample size distribution with stratification, findings may differ at extremely high uterine weights. During statistical analysis, we did not adjust for the multiple comparisons during the regression modeling, which may increase the possibility of a Type I error. Manual Bonferroni correction could avoid this issue; however, it also has been determined to be over conservative thus was not performed.
CONCLUSIONS
This study highlights the impact of uterine weight on adverse surgical outcomes that contribute to patient morbidity in CLH compared to RA-LH. Patients may benefit from robotic-assisted hysterectomy due to the lower odds of intraoperative conversions, blood loss > 500 mL, and prolonged operative time. Patients with uteri ≥ 450 g may benefit the most given the seven-fold reduced odds of conversion in RA-LH compared to CLH. These findings may help inform surgeons of operative outcomes and impact surgical counseling and operative planning.
Footnotes
Acknowledgements: The authors would like to thank Christina Johnson, MD for case acquisition and data collection.
Disclosure: none
Conflict of interests: none
Funding sources: none
IRB approval: This study was reviewed and approved by the Christiana Care Hospital Institutional Review Board (IRB00000480) per exempt category 45 CFR 46.104(d)(4)(iii) with waiver of HIPAA Authorization 45 CFR 164.512(i)(2)(ii).
References:
- 1.Wright JD, Herzog TJ, Tsui J, et al. Nationwide trends in the performance of inpatient hysterectomy in the United States. Obstet Gynecol. 2013;122(2 Pt 1):233–241. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Aarts JWM, Nieboer TE, Johnson N, et al. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2015;2015(8):CD003677. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nieboer TE, Hendriks JC, Bongers MY, Vierhout ME, Kluivers KB. Quality of life after laparoscopic and abdominal hysterectomy: a randomized controlled trial. Obstet Gynecol. 2012;119(1):85–91. [DOI] [PubMed] [Google Scholar]
- 4.Nieboer TE, Johnson N, Lethaby A, et al. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2009(3):CD003677. [DOI] [PubMed] [Google Scholar]
- 5.Wiser A, Holcroft CA, Tulandi T, Abenhaim HA. Abdominal versus laparoscopic hysterectomies for benign diseases: evaluation of morbidity and mortality among 465,798 cases. Gynecol Surg. 2013;10(2):117–122. 10.1007/s10397-013-0781-9. [DOI] [Google Scholar]
- 6.AAGL Advancing Minimally Invasive Gynecology Worldwide. AAGL Position Statement: route of hysterectomy to treat benign uterine disease. J Minim Invasive Gynecol. 2011;18(1):1–3. [DOI] [PubMed] [Google Scholar]
- 7.ACOG Committee Opinion No 701: Choosing the route of hysterectomy for benign disease. Obstet Gynecol. 2017;129(6):e155–e159. [DOI] [PubMed] [Google Scholar]
- 8.Payne TN, Dauterive FR. A comparison of total laparoscopic hysterectomy to robotically assisted hysterectomy: surgical outcomes in a community practice. J Minim Invasive Gynecol. 2008;15(3):286–291. [DOI] [PubMed] [Google Scholar]
- 9.Shashoua AR, Gill D, Locher SR. Robotic-assisted total laparoscopic hysterectomy versus conventional total laparoscopic hysterectomy. JSLS. 2009;13(3):364–369. [PMC free article] [PubMed] [Google Scholar]
- 10.Abitbol J, Munir A, How J, et al. The shifting trends towards a robotically-assisted surgical interface: clinical and financial implications. Health Policy Technol. 2020;9(2):157–165. [Google Scholar]
- 11.Jeong SY, Kim K, Ryu JW, Cha J, Park ST, Park SH. Comparison of surgical outcomes of robotic versus conventional laparoscopic hysterectomy of large uterus with gynecologic benign disease. J Pers Med. 2022;12(12):2042. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Liu H, Lu D, Wang L, Shi G, Song H, Clarke J. Robotic surgery for benign gynaecological disease. Cochrane Database Syst Rev. 2012(2):CD008978. [DOI] [PubMed] [Google Scholar]
- 13.Shah NT, Wright KN, Jonsdottir GM, Jorgensen S, Einarsson JI, Muto MG. The feasibility of societal cost equivalence between robotic hysterectomy and alternate hysterectomy methods for endometrial cancer. Obstet Gynecol Int. 2011(;2011):570464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Truong MD, Tholemeier LN. Role of robotic surgery in benign gynecology. Obstet Gynecol Clin North Am. 2022;49(2):273–286. [DOI] [PubMed] [Google Scholar]
- 15.Wright JD, Ananth CV, Lewin SN, et al. Robotically assisted vs laparoscopic hysterectomy among women with benign gynecologic disease. JAMA. 2013;309(7):689–698. [DOI] [PubMed] [Google Scholar]
- 16.Kilic GS, Moore G, Elbatanony A, Radecki C, Phelps JY, Borahay MA. Comparison of perioperative outcomes of total laparoscopic and robotically assisted hysterectomy for benign pathology during introduction of a robotic program. Obstet Gynecol Int. 2011(;2011:683703. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Martino MA, Berger EA, McFetridge JT, et al. A comparison of quality outcome measures in patients having a hysterectomy for benign disease: robotic vs. non-robotic approaches. J Minim Invasive Gynecol. 2014;21(3):389–393. [DOI] [PubMed] [Google Scholar]
- 18.Louie M, Strassle PD, Moulder JK, Dizon AM, Schiff LD, Carey ET. Uterine weight and complications after abdominal, laparoscopic, and vaginal hysterectomy. Am J Obstet Gynecol. 2018;219(5):480.e1-480–e8. [DOI] [PubMed] [Google Scholar]
- 19.Uccella S, Cromi A, Bogani G, Casarin J, Formenti G, Ghezzi F. Systematic implementation of laparoscopic hysterectomy independent of uterus size: clinical effect. J Minim Invasive Gynecol. 2013;20(4):505–516. [DOI] [PubMed] [Google Scholar]
- 20.Mowat A, Maher C, Ballard E. Surgical outcomes for low-volume vs high-volume surgeons in gynecology surgery: a systematic review and meta-analysis. Am J Obstet Gynecol. 2016;215(1):21–33. [DOI] [PubMed] [Google Scholar]
- 21.Sirota I, Tomita SA, Dabney L, Weinberg A, Chuang L. Overcoming barriers to vaginal hysterectomy: An analysis of perioperative outcomes. J Turk Ger Gynecol Assoc. 2019;20(1):8–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Catanzarite T, Saha S, Pilecki MA, Kim JY, Milad MP. Longer operative time during benign laparoscopic and robotic hysterectomy is associated with increased 30-day perioperative complications. J Minim Invasive Gynecol. 2015;22(6):1049–1058. [DOI] [PubMed] [Google Scholar]
- 23.Okumura A, Kondo E, Nii M, Kubo-Kaneda M, Yoshida K, Ikeda T. Comparison of surgical outcomes between robot-assisted laparoscopic hysterectomy and conventional total laparoscopic hysterectomy in gynecologic benign disease: a single-center cohort study. J Robot Surg. 2023;17(5):2221–2228. [DOI] [PubMed] [Google Scholar]
- 24.Alshowaikh K, Karpinska-Leydier K, Amirthalingam J, et al. Surgical and patient outcomes of robotic versus conventional laparoscopic hysterectomy: a systematic review. Cureus. 2021;13(8):e16828. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Paraiso MF, Ridgeway B, Park AJ, et al. A randomized trial comparing conventional and robotically assisted total laparoscopic hysterectomy. Am J Obstet Gynecol. 2013;208(5):368.e1-368–e3687. [DOI] [PubMed] [Google Scholar]
- 26.Moawad GN, Abi Khalil ED, Tyan P, et al. Comparison of cost and operative outcomes of robotic hysterectomy compared to laparoscopic hysterectomy across different uterine weights. J Robot Surg. 2017;11(4):433–439. [DOI] [PubMed] [Google Scholar]