Objective
The purpose of this study was to compare operative time and intra- and postoperative complications between total laparoscopic hysterectomy and robotic-assisted total laparoscopic hysterectomy.
Study Design
This study was a blinded, prospective randomized controlled trial conducted at 2 institutions. Subjects consisted of women who planned laparoscopic hysterectomy for benign indications. Preoperative randomization to total laparoscopic hysterectomy or robotic-assisted total laparoscopic hysterectomy was stratified by surgeon and uterine size (> or ≤12 weeks). Validated questionnaires, activity assessment scales, and visual analogue scales were administered at baseline and during follow-up evaluation.
Results
Sixty-two women gave consent and were enrolled and randomly assigned; 53 women underwent surgery (laparoscopic, 27 women; robot-assisted, 26 women). There were no demographic differences between groups. Compared with laparoscopic hysterectomy, total case time (skin incision to skin closure) was significantly longer in the robot-assisted group (mean difference, +77 minutes; 95% confidence interval, 33–121; P < .001] as was total operating room time (entry into operating room to exit; mean difference, +72 minutes; 95% confidence interval, 14–130; P = .016). Mean docking time was 6 ± 4 minutes. There were no significant differences between groups in estimated blood loss, pre- and postoperative hematocrit change, and length of stay. There were very few complications, with no difference in individual complication types or total complications between groups. Postoperative pain and return to daily activities were no different between groups.
Conclusion
Although laparoscopic and robotic-assisted hysterectomies are safe approaches to hysterectomy, robotic-assisted hysterectomy requires a significantly longer operative time.
Hysterectomy is the most common nonobstetric procedure performed for reproductive-aged women in the United States. Each year, approximately 615,000 hysterectomies are performed; approximately 20 million women have had a hysterectomy.
Compared with abdominal hysterectomy, the advantages of laparoscopic hysterectomy are decreased postoperative intravenous analgesia requirements, shorter length of hospital stay, and quicker return to work and daily activities. Longer operating times have been shown to be offset by shorter hospital stays, with similar hospital costs overall.
Compared with abdominal hysterectomy, laparoscopic hysterectomy is associated with less blood loss and fewer abdominal wall infections and febrile episodes.
Nevertheless, the laparoscopic approach requires a higher level of technical skills, especially with total laparoscopic hysterectomy for which the entire procedure, including suturing of the vaginal cuff, is performed by laparoscopic route. Abdominal hysterectomy remains the most common surgical approach, likely because of the more difficult learning curve that is associated with minimally invasive techniques and the relative lack of advanced laparoscopic surgeons who are skilled in laparoscopic tissue dissection and suturing.
Robotic systems were developed to facilitate laparoscopy by allowing more ergonomic movements that are easier to perform and are more precise. The da Vinci Surgical System (Intuitive Surgical Inc, Sunnyvale, CA) has EndoWrist instruments that mimic the human wrist. It can also filter out tremor and other unintentional hand motions. The surgeon who is seated at a console commands the laparoscope and 2-3 laparoscopic surgical instruments. Robotics has been adopted widely, despite the lack of abundant data that has demonstrated a benefit over other operative approaches.
To address this deficiency, we designed a prospective randomized trial that compared robotic-assisted vs laparoscopic hysterectomy in the treatment of patients who undergo hysterectomy for benign indications. The primary outcome was total case time from incision to closure. Secondary outcomes were intraoperative and postoperative complications, the impact of surgery on the activities of daily living, and narcotic use for 6 weeks based on patient responses to daily postoperative diaries.
Materials and Methods
Cleveland Clinic (IRB #07-150) was first approved on Feb. 21, 2007; Brigham and Women’s Hospital (IRB #2008P000220) was first approved March 14, 2008. After institutional review board approvals for subject consent were obtained, subjects were recruited from the Departments of Gynecology at the Cleveland Clinic and at Brigham and Women’s Hospital and underwent surgery between June 2007 and March 2011. Inclusion criteria included women ≥18 years old who were to undergo laparoscopic hysterectomy for benign indications. Other concomitant laparoscopic or antiincontinence procedures (eg, excision of endometriosis or midurethral sling procedures) were performed at the primary surgeon’s discretion. Exclusion criteria were suspected malignancy, medical illness that precludes laparoscopy, inability to give informed consent, morbid obesity (body mass index [BMI], >44 kg/m 2 ), or need for concomitant bowel resection. Patients who chose not to participate in this study were offered the standard evaluation and treatment that includes a discussion regarding the risks and benefits of different approaches to hysterectomy as deemed appropriate by the primary surgeon.
Preoperatively, subjects underwent a standardized evaluation that included medical history and physical examination. Additionally, they filled out valid and reliable questionnaires that included the 36-Item Short Form Health Survey (SF-36) quality-of-life assessment, pain scales, and functional activities of daily life questionnaires. Subjects were assigned randomly according to a computer-generated randomization schedule with random block sizes with the use of the SAS statistical software package (SAS Institute Inc, Cary, NC) and were stratified by surgeon and by uterine size into those who had a uterus >12-week size either by examination or ultrasound scan and those who had a uterus <12-week size either by examination or ultrasound scan. Surgeries were scheduled based on the randomization; all patients were blinded to their assignment.
There were 5 enrolling surgeons included in this study (M.F.R.P., B.R., J.E.J., T.F., and J.I.E.). All 5 surgeons were experienced laparoscopists who had performed 75-400 total laparoscopic hysterectomies before enrollment into the trial and had also completed robotic training provided by Intuitive Surgical Inc. All enrolling surgeons had performed at least 20 cases of operative robot-assisted surgery before enrolling patients into this trial. Assisting surgeons included Female Pelvic Medicine and Reconstructive Surgery fellows, Obstetrics and Gynecology residents and medical students. Laparoscopic-assisted hysterectomy was performed with the use of 4 ports: an umbilical port for the laparoscopic camera, 2 ports (either 5 or 10/12 mm) in the bilateral lower quadrants, and one 5-mm port placed at the level of the umbilicus, lateral to the rectus muscle on either side for retraction. The robotic-assisted hysterectomy was performed with the da Vinci Surgical System with the use of the following ports: an umbilical port for the laparoscopic camera, one 10/12-mm port placed in the right or left subcostal area lateral to the rectus for suture introduction, two 8-mm robotic ports placed in bilateral lower quadrants, and one 5-mm port approximately 8 cm inferior to the right or left subcostal margin laterally for retraction as previously reported. The technique to perform the hysterectomy was performed in a standard fashion, with the entirety of the hysterectomy performed laparoscopically (with or without the assistance of the robot), which included the electrocoagulation of the uterine arteries and cardinal ligaments, incision of the vagina, and suturing of the vaginal cuff. Electrocoagulation for vascular pedicles during the laparoscopic hysterectomies was performed with either the Ligasure (Covidien Corporation, Boulder, CO), Gyrus PK (Gyrus ACMI, Southborough, MA), or Harmonic Scalpel (Ethicon Endo-Surgery, Cincinnati, OH). The electrocoagulation for vascular pedicles during the robotic-assisted hysterectomies was performed with bipolar forceps, Gyrus PK, or Harmonic Scalpel. Routine cystoscopy was also performed to assess for lower urinary tract injury. There were no changes in practice over the time period of performance of the surgeries such as robots with a teaching station or advances in electrosurgical instruments in the robotic arm.
Data points that were recorded by operating room personnel during the procedure included total operating room time (entry into room and exit from room), total case time (initial incision to skin closure), and intraoperative complications. Daily pain medication use included in-patient patient-controlled analgesia intravenous narcotics, and outpatient oral analgesics. Quality-of-life questionnaires and diaries that evaluated pain, activity level, and return to daily activities were filled out by subjects. Weekly assessments of return to normal activity and assessments at 2, 4, and 6 weeks of the patients’ surgical pain scale and ability to perform activities of daily living were collected. Patients returned to the clinic for a 4- to 6-week postoperative visit, at which time a physician or a research nurse who was blinded to the patient’s surgical group assignments also performed an assessment. At 6 months, research nurses administered these questionnaires over the phone (SF-36, pain scales, and functional activities of daily life questionnaires).
We determined that 23 subjects in each arm were needed to detect a difference of ≥30 minutes in operating time between conventional vs robotic-assisted laparoscopic hysterectomy with 90% power and a significance level of .05 (NCSS and PASS; Number Cruncher Statistical Systems, Kaysville, UT). We assumed a standard deviation of 30 minutes for surgery time for both groups as observed from previous studies. We then accounted for a 10% dropout rate at 6 months by adding 3 patients to each group so that each arm would require 26 patients, for a total of 52 subjects. Those patients who withdrew after consenting to participate in this study were included as an intention-to-treat analysis.
Groups were compared on dichotomous variables with the use of Fisher exact tests and on continuous variables with the 2-sided Student t tests. Wilcoxon rank sum test was used to compare the groups in terms of the quality-of-life variables from the SF-36: physical component summary score, mental component summary score, overall health, health vs 1 year ago, vigorous activities, moderate activities, lifting groceries, climbing several stairs, climbing 1 flight, bending, walking >1 mile, walking several hundred yards, walking 100 yards, and bathing or dressing. Mixed models analysis was used for the following variables measures longitudinally: activity, pain at rest, pain during normal activities, pain during exercise, and worst pain today.
Results
Sixty-two subjects were consented, enrolled, and randomly assigned ( Figure ). Nine subjects withdrew from the study before surgery (laparoscopy group, 5; robot-assisted group, 4), and 53 subjects underwent surgery (laparoscopy group, 27; robot-assisted group, 26). One subject in the laparoscopic group had missing data for the primary outcome and was excluded from analysis, which left 26 in the laparoscopic group and 26 in the robot-assisted group. There were no differences in age, BMI, race, and parity among all subjects and those who ultimately underwent surgery. Because all subjects who withdrew from the study did so before the intervention, the remainder of the analysis reflects all subjects who underwent hysterectomy.
Demographic data are listed in Table 1 . Subjects in the surgical group had a mean (±SD) age of 44.7 ± 6.1 years, a mean BMI of 30.6 ± 7.4 kg/m 2 , and median parity of 2 (range, 0–6) with no differences between groups. Seventy-seven percent of the subjects were white; 20% of them were smokers; 9.2% of them had at least 1 cesarean delivery; 32% of them had a history of previous laparoscopy, and 46% of them had a previous laparotomy (with no differences between groups). Indications for hysterectomy did not differ between groups: fibroid tumors (59%), pelvic pain (67%), endometriosis (10%), abnormal bleeding (60%), and ovarian cyst (18%); most subjects (80%) indicated >1 reason for hysterectomy.
| Variable | Procedure | P value | |
|---|---|---|---|
| Conventional | Robot-assisted | ||
| Average age, y | 45.6 | 43.8 | .29 |
| Body mass index, kg/m 2 | 31.4 | 29.9 | .54 |
| Smoking history, % | 25 | 8.0 | .27 |
| White race, % | 73.1 | 68 | .48 |
| Median parity, n | 2 | 2 | .30 |
| Previous laparoscopy, % | 36.0 | 28.0 | .76 |
| Previous laparotomy, % | 40.0 | 52.0 | .57 |
Ninety-two percent of surgical cases were completed per the randomization assignment. One subject who was assigned randomly to laparoscopy required a laparotomy because of bleeding and inability to maintain pneumoperitoneum. Two subjects who were assigned randomly to the robot-assisted group were converted to laparoscopy, one because of robot malfunction and the other because of an inability to ventilate the subject. One subject who was assigned randomly to the laparoscopy group underwent a robotic-assisted hysterectomy because of error.
The total operating room time (time from entry into to exit from the operating room) was significantly shorter in the laparoscopic group compared with the robot-assisted group (171.6 ± 75.8 minutes vs 245.8 ± 117.1 minutes; P = .01), as was the total case time from incision to closure (102.7 ± 63.7 minutes vs 172.8 ± 89.0 minutes; P = .002). There were no differences between groups in estimated blood loss, hematocrit change, and uterine weight (293.9 ± 299.9 g vs 282.9 ± 214.7 g; range, 35–1242 g). There were no intraoperative transfusions, bladder, ureteral, rectal, or small-bowel injuries in either group. The mean docking time for the robot-assisted group was 6 ± 3.9 minutes. Concomitant procedures are listed in Table 2 and show no difference between groups.
| Concomitant procedure | Procedure | P value | |
|---|---|---|---|
| Conventional, n (%) | Robot-assisted, n (%) | ||
| Culdoplasty or vault suspension | 6 (23) | 6 (23) | 1.0 |
| Adnexal surgery | 4 (15) | 3 (12) | .69 |
| Lysis of adhesions | 5 (19) | 7 (27) | .51 |
| Sling | 2 (8) | 0 | .49 |
| Excision of endometriosis | 3 (12) | 3 (12) | 1.0 |
| >1 concomitant procedure | 3 (12) | 4 (15) | .69 |
Longer operating room time was associated significantly with increased uterine weight ( P = .04) but with no other variables that included age, BMI, parity, history of laparoscopy, history of laparotomy, and number of cesarean deliveries. We dichotomized operating room time into 2 groups: >185 minutes (greater than the 75th percentile) or ≤185 minutes. Randomization group, BMI, history of laparotomy, history of laparoscopy, number of cesarean deliveries, and uterine weight did not have significant associations with this outcome (all P > .05). However, when all variables were entered into a multivariable model, all variables were associated significantly with operating room time of >185 minutes ( P < .0001), with the exception of uterine weight and number of cesarean deliveries. Operating room time of >185 minutes was not associated with any specific complication or number of complications but was associated with a longer length of stay (1.22 ± 0.53 days vs 2 ± 1.56 days; P = .013).
The patients enrolled per surgeon ranged between 3 and 32. All surgeons performed hysterectomies by both routes. Among the 3 surgeons with individual data that could be compared, each surgeon had significantly longer total case times in the robot-assisted group compared with the laparoscopic group (range, 46–117 minutes). Even the surgeon with the fastest, most consistent total case times took a mean 46 minutes longer per case in the robot-assisted group.
Postoperatively, there were no deep vein thromboses, pulmonary emboli, repeat operations or infectious, gastrointestinal, cardiac, or pulmonary complications. Three subjects required blood transfusions in the postoperative period (2 after robot-assisted hysterectomy; 1 after conventional laparoscopic hysterectomy). The average length of stay was 1.4 ± 0.9 days, with no difference between groups.
Both groups did not differ significantly in terms of the quality-of-life variables at baseline and 6 months after surgery (all P > .30; Table 3 ). Subjects in both groups noted that they were somewhat limited during vigorous activity before surgery; 6 months after surgery most women in both groups responded that they were not limited at all. In the longitudinal analyses, there were significant overall improvements over time for all variables ( P < .0001); however, the arms did not differ significantly in the levels (all P > .20) or in the trends over time (all P > .15). Table 4 demonstrates that postoperative pain and return to normal activity were not different between groups.
| Variable | Baseline | At 6 mo | ||||
|---|---|---|---|---|---|---|
| Conventional (n = 22) | Robot-assisted (n = 23) | P value a | Conventional (n = 19) | Robot-assisted (n = 19) | P value a | |
| 36-Item Short Form Health Survey scores (mean ± SD) | ||||||
| Physical component summary score | 44 ± 8 | 47 ± 10 | .42 | 53.7 ± 8 | 53 ± 8 | .80 |
| Mental component summary score | 47 ± 11 | 52 ± 6 | .05 | 45 ± 14 | 50 ± 11 | .35 |
| 36-Item Short Form Health Survey questions (median) | ||||||
| 1. Health status b | 4 | 4 | .39 | 4 | 4 | .90 |
| 2. Health compared with 1 year ago c | 3 | 3 | .16 | 5 | 4 | .28 |
| 3a. Vigorous activities d | 2 | 2 | .72 | 3 | 3 | .87 |
| 3b. Moderate activities d | 3 | 3 | .99 | 3 | 3 | .86 |
| 3c. Lifting/carrying groceries d | 3 | 3 | .76 | 3 | 3 | .10 |
| 3d. Climbing several flights of stairs d | 3 | 3 | .91 | 3 | 3 | .50 |
| 3e. Climbing 1 flight of stairs d | 3 | 3 | .37 | 3 | 3 | .17 |
| 3f. Bending, kneeling or stooping d | 3 | 3 | .66 | 3 | 3 | .50 |
| 3g. Walking >1 mile d | 3 | 3 | .78 | 3 | 3 | .90 |
| 3h. Walking several blocks d | 3 | 3 | .37 | 3 | 3 | .18 |
| 3i. Walking 1 block d | 3 | 3 | .30 | 3 | 3 | .19 |
| 3j. Bathing or dressing yourself d | 3 | 3 | .40 | 3 | 3 | .97 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
