Original Article
A Randomized Trial Comparing Vaginal and Laparoscopic Hysterectomy vs Robot-Assisted Hysterectomy Celine L€ onnerfors, MD, Petur Reynisson, MD, PhD, and Jan Persson, MD, PhD* From the Department of Obstetrics and Gynecology, Sk ane University Hospital and Lund University, Lund, Sweden (all authors).
ABSTRACT Study Objective: To investigate the hospital cost and short-term clinical outcome of traditional minimally invasive hysterectomy vs robot-assisted hysterectomy in women primarily not considered candidates for vaginal surgery. Design: Randomized controlled trial (Canadian Task Force classification I). Setting: University Hospital in Sweden. Patients: One hundred twenty-two women with uterine size %16 gestational weeks scheduled to undergo minimally invasive hysterectomy because of benign disease. Interventions: Robot-assisted hysterectomy or traditional vaginal or laparoscopic minimally invasive hysterectomy. Measurements and Main Results: All women underwent surgery as randomized. There were no demographic differences between the 2 groups. Vaginal hysterectomy was possible in 41% in the traditional minimally invasive group, at a mean hospital cost of $4579 compared with $7059 for traditional laparoscopic hysterectomy. This was reflected in a mean hospital cost of $993 more per robotic-assisted hysterectomy than for traditional minimally invasive hysterectomy when the robot was a preexisting investment. This hospital cost increased by $1607 when including investments and cost of maintenance. A perprotocol subanalysis comparing laparoscopy and robotics demonstrated similar hospital cost when the robot was a preexisting investment ($7059 vs $7016). Robotic-assisted hysterectomy was associated with less blood loss and fewer postoperative complications. Conclusion: A similar hospital cost can be attained for laparoscopy and robotics when the robot is a preexisting investment. From the perspective of hospital costs, robotic-assisted hysterectomy is not advantageous for treating benign conditions when a vaginal approach is feasible in a high proportion of patients. Journal of Minimally Invasive Gynecology (2015) 22, 78–86 Ó 2015 AAGL. All rights reserved. Keywords:
DISCUSS
Hysterectomy; Laparoscopic hysterectomy; Minimally invasive surgery; Robot-assisted laparoscopy; Vaginal hysterectomy
You can discuss this article with its authors and with other AAGL members at http://www.AAGL.org/jmig-22-1-JMIG-D-14-00303
Use your Smartphone to scan this QR code and connect to the discussion forum for this article now* * Download a free QR Code scanner by searching for ‘‘QR scanner’’ in your smartphone’s app store or app marketplace.
The potential benefit of robot-assisted laparoscopic surgery is enabling a higher proportion of minimally invasive surgical procedures. Despite guidelines supporting minimally invasive procedures, hysterectomy to treat benign gynecologic disease is still most commonly performed via laparotomy [1–9]. Vaginal hysterectomy is primarily performed in conjunction with surgery to treat prolapse, Disclosures: Dr. Persson is a proctor in robotic surgery. Corresponding author: Jan Persson, MD, PhD, Department of Obstetrics and Gynecology, Sk ane University Hospital, SE-22185 Lund, Sweden. E-mail: [email protected] Submitted June 18, 2014. Accepted for publication July 12, 2014. Available at www.sciencedirect.com and www.jmig.org 1553-4650/$ - see front matter Ó 2015 AAGL. All rights reserved. http://dx.doi.org/10.1016/j.jmig.2014.07.010
and the rate in the United States decreased from 22% in 2003 to 19% in 2009–2010, which coincides with the introduction of robotic-assisted surgery [1–4]. Robot-assisted laparoscopy has been widely adopted to treat benign gynecologic conditions, although no data have demonstrated a clinical or economic benefit over other operative approaches [3,10–17]. Recently, 2 large cohort studies found similar morbidity profiles as for laparoscopic hysterectomy, but a substantially increase in cost for robotic-assisted surgical procedures. However, factors that might influence the route of hysterectomy chosen, such as body mass index, uterine weight, and previous abdominal surgery, were not available [3,15]. A 2012 Cochrane review identified 2 randomized controlled trials of benign
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
gynecologic robotic-assisted hysterectomy and concluded that robotic surgery was not associated with improved effectiveness or safety [9]. However, both studies were potentially biased by inclusion of early robot adopters [9,18,19]. The Department of Obstetrics and Gynecology at Sk ane University Hospital is a tertiary referral unit for both gynecologic oncology and complex benign gynecologic surgery. Robotic surgery was introduced in October 2005, and to date .1600 women have undergone robotic-assisted surgery, with approximately 300 procedures performed each year [20–22]. Before the implementation of robotic-assisted surgery, traditional laparoscopy was routinely used for hysterectomy, and minimally invasive surgery was used to perform 78% of all hysterectomies to treat benign disease in 2012. The primary objective of the present study was to investigate the hospital cost of robotic-assisted hysterectomy compared with traditional minimally invasive hysterectomy (vaginal and laparoscopic) performed to treat benign gynecologic disorders in women with uterine size %16 gestational weeks after excluding women referred for vaginal hysterectomy. The secondary objective was to assess shortterm clinical outcome. Material and Methods Between January 2010 and June 2013, 125 women meeting the inclusion criteria were offered participation in the study (Fig. 1; Table 1). Preoperative evaluation in all women included medical and surgical history, and clinical examination including a gynecologic examination and vaginal ultrasonography. Each woman was assigned an individual clinical research file containing all study protocols. One hundred twenty-two opaque envelopes containing the assigned surgical method in the proportion of 1:1 were closed, shuffled, and then numbered. After inclusion, randomization occurred via telephone during which the envelopes were opened in consecutive order at the central randomization office. The date, clinical research file number, patient name and social security number, and the assigned surgical method were recorded in the central study log. Before randomization, surgical procedures were scheduled to be performed on a day when it was possible to include all 3 approaches, and after randomization the patient was assigned a specific surgeon. All patients were informed of their assignment. The route of traditional minimally invasive surgery was chosen by the designated surgeon, with vaginal hysterectomy as the first choice, followed by laparoscopic hysterectomy. The necessity of concomitant procedures, vaginal access, body mass index, and the presence of adhesions or endometriosis influenced the choice of surgical approach. Robot-assisted laparoscopic hysterectomy was performed using the da Vinci Si Surgical System (Intuitive Surgical, Inc., Sunnyvale CA). All 6 surgeons were consultants experienced in both vaginal and laparoscopic surgery, and 4 were gynecologic oncolo-
79
gists experienced in robotic-assisted surgery. The case load of hysterectomy to treat benign disease before commencing the study was approximately 110 annually, with 31% performed vaginally, 27% laparoscopically, and 15% robotically. However, most hysterectomies performed at our institution are to treat malignant disease, with the greatest proportion performed by the 4 gynecologic oncologists. The least experienced robotic surgeon had performed 49 robotic hysterectomies before the study, and a total of 231 robotic-assisted procedures by the end of the study, primarily in gynecologic oncology. All women received oral prophylactic antibiotic therapy including 200 mg doxycycline (Doxyferm; Nordic Group BV, Hoofddorp, Holland) and 800 mg metronidazole (Flagyl; Sanofi-Aventis, Paris, France). All procedures were performed with the patient under general anesthesia. For robotic-assisted hysterectomy, two or three 8-mm robot trocars and one assistant 5- or 12-mm trocar were used. Laparoscopic hysterectomy was performed with the use of 4 ports: a reusable umbilical port or a 12-mm port (Xcel; Ethicon EndoSurgery, Inc., Somerville, NJ) for the optics and 3 assistant ports, either three 5-mm or two 5-mm plus one 10-mm ports, in the lower quadrants. Vaginal hysterectomy was performed in the standard manner. For robot-assisted and total laparoscopic hysterectomy the peritoneum of the lateral sidewalls was opened, the ureters were visualized and lateralized, and the propria pelvic ligament or infundibulopelvic ligament and the round ligaments were divided, followed by electrocoagulation of the uterine arteries and cardinal ligaments. Then the vagina was incised, and the vaginal cuff was sutured using polyglactin 910 absorbable sutures (Vicryl; Ethicon) or V-Loc 180 sutures (Covidien; Mansfield, MA), at surgeon discretion either robotically, laparoscopically, or vaginally, with the latter in patients after vaginal coring of the uterus. During laparoscopic vaginal hysterectomy the procedure was similar except that the uterine arteries and cardinal ligaments were divided vaginally. Postoperative cystoscopy was not performed. Total operating room time (from patient entry to departure from the operating room, including administration of anesthesia), total operative time (skin to skin including placement of a catheter, application of a fornix presenter, docking, and dedocking), intraoperative blood loss, and complications were recorded. Hemoglobin and C-reactive protein concentrations and body temperature were noted on the first postoperative day. All women received similar pain medication and were given a daily dose of 4500 IE tinzaparin (Innohep; LEO Pharma AB, Copenhagen, Denmark) for 10 days postoperatively. Length of stay and immediate postoperative complications were recorded. All patients were advised to contact the department if necessary. A postoperative visit was planned for 4 months after surgery. To assess short-term clinical outcome, all adverse events within this period were included with the exception of uncomplicated lower urinary tract infections because this was thought to be unrelated to the surgical method.
80
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Fig. 1 Progression of patients through the study.
Enrollment
Eligible subjects offered participation (n =125)
Excluded (n =3) Declined participation (n =3)
Allocated to traditional minimally invasive surgery (n = 61)
Underwent robotic hysterectomy (n = 61)
Follow-up
Underwent allocated intervention (n = 61) Vaginal hysterectomy (n = 25 [41%]) Laparoscopic hysterectomy (n= 36[59%])
Allocated to robotic surgery (n = 61)
Lost to follow-up (n = 0) Discontinued intervention (n = 0)
Analysis
Allocation
Randomized (n = 122)
Analyzed (n = 61)
Analyzed (n = 61)
Excluded from analysis (n = 0)
Excluded from analysis (n = 0)
For estimation of cost-generating factors common to traditional, minimally invasive, and robotic-assisted hysterectomy we used admittance fee, operating room time, hospital length of stay, and cost of complications, readmission, and repeat interventions until 4 months after surgery. The hospital internal charge is based on the average real cost for the respective parameter. The per-minute charge for use of the operating room is based on the previous year’s
Lost to follow-up (n = 0) Discontinued intervention (n = 0)
mean institutional cost and includes the operating theater, 1 surgeon, 1 assistant, 1 scrub nurse, 1 circulating nurse, anesthetic staff, cleaning, and basic expendables such as gowns and gloves. The procedure cost for each approach was calculated, and included the basic cost and the specific cost, depending on the surgical approach. The traditional minimally invasive basic cost comprised the purchase price of the vaginal and/or laparoscopic equipment including
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
Table 1 Inclusion and exclusion criteria Inclusion criteria
Exclusion criteria
Need for total hysterectomy to treat benign indication Uterus size %16 gestational weeks Uterus/vagina size enabling vaginal retrieval, allowing for coring, when necessary No desire for additional pregnancies Informed consent
Suspicion of or known malignancy Known extensive intra-abdominal adhesions Contraindication for laparoscopic surgery Pacemaker or other electrosensitive implant Known bleeding disorder Immunosuppression therapy or other known increased risk of infection Women referred for vaginal hysterectomy Simultaneous need for prolapse surgery Allergy to metronidazole and/or doxycycline Inability to understand patient information
annual reinvestment, maintenance, and instrument wear and tear, with an estimated depreciation time of 7 years and an annual turnover of 300 procedures. In addition, the traditional minimally invasive specific cost included the cost of patient draping, sterilization of all instruments, and cost of disposable instruments when applicable, such as a 12-mm assistant port and a 5-mm suction-irrigation system (Stryker Endoscopy, San Jose, CA). A non-disposable fornix presenter was used. The robot basic cost, including investment and maintenance, was calculated from the purchase price of the da Vinci Si system, using a depreciation time of 7 years, the annual maintenance fee, and an annual turnover of 300 procedures. Because the robotic system was purchased for use within gynecologic oncology and for complex benign cases, other benign cases are performed only as excess capacity procedures, and thus in a setting such as ours the robot can be considered a preexisting investment insofar as less complicated procedures. When considering the robot as a preexisting investment, the robot basic cost was excluded from the robot procedure cost. The robot specific cost included the cost of patient draping, sterilization of all instruments including robotic instruments, ports and optics, robot draping (3- or 4-arm kit), cost of the actual number of robotic instruments, and disposable instruments used. A modified non-disposable trocar (Ternamian Endotip; Karl Storz GmbH & Co KG, Tuttlingen, Germany) hosted the optics. The electrocoagulation devices used were the bipolar forceps and monopolar scissors for traditional laparoscopy and their robotic counterparts for robotic-assisted surgery. Both
81
for robotic-assisted and traditional laparoscopy, standard reusable laparoscopic instruments were used, and the depreciation cost was estimated to be the same for robotic-assisted and traditional minimally invasive procedures. Because of their high cost, disposable sealing or cutting instruments or morcellators were not used. Inasmuch as all procedures were minimally invasive, no differences in societal cost was expected, and therefore was not included. For costs charged in Swedish Crowns (SEK) or Euros (V), we used the mean currency exchange rate between the US dollar and SEK or V for 2010 to 2012 as estimated by Swedish tax authorities (1$ 5 6.4294 SEK or 0.7411 V). The major factor that influences hospital cost at our institution is the operative time. Existing data on surgical times for robotic-assisted hysterectomy were analyzed (Analyseit; Analyse-it Software Ltd, Leeds, UK), found to have positive skewness, and therefore transformed into natural logarithms to perform the power analysis. A similar skewness was supposed for data from traditional minimally invasive surgery. We determined that 96 patients were needed to detect a difference of 30 minutes in operating room time between traditional minimally invasive and roboticassisted hysterectomy, with 80% power and a significance level of .05. To compensate for the uncertainty in the distribution of traditional minimally invasive data and the uncertainty of the number of hysterectomies available for the vaginal approach, we estimated a need for another 15%. Accounting for a drop out rate of 10%, a total of 122 women were included in the study. Institutional review board approval was obtained. The study was registered (www.clinicaltrials.gov NCT01865929). All included women gave written informed consent. Statistical analysis was performed using commercially available software (SPSS version 20; IBM Corp., Armonk, NY). For statistical analyses we used the c2 test, the Student t-test, or the Mann-Whitney U test, as appropriate. The gaussian distribution of the groups was tested using the Kolmogorov-Smirnov fitness test, and correlation was tested using the Pearson correlation. The power analysis was made using an unpaired t-test. All tests were 2-sided, and p ,.05 was considered significant. Results One hundred twenty-two women consented to participate in the study, were enrolled and randomized, and underwent surgery (Fig. 1). There were no demographic differences between groups (Table 2). Uterine myomas (44%) and abnormal bleeding (31%) were the most common indications for surgery. The mean operating room time was 145 minutes for traditional minimally invasive hysterectomy (103 minutes for vaginal hysterectomy and 174 minutes for laparoscopic hysterectomy), compared with 147 minutes for robotic-assisted
82
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 2 Demographic data Procedure Variable
Traditional minimally invasive hysterectomy
Robotic-assisted hysterectomy
p Value
Age, yr, median (range) Body mass index, median (range) Median parity, No. (range) No vaginal deliveries, % Previous abdominal surgery, % Previous cesarean delivery, %
46 (29–69) 24.9 (17.6–42) 2 (0–5) 21 56 18
47 (27–65) 24.9 (17–39.2) 2 (0–6) 23 54 23
.89 .86
hysterectomy. This being the major factor influencing hospital cost led to a mean hospital cost per patient of $993 less for traditional minimally invasive hysterectomy as a whole compared with robotic-assisted hysterectomy when considering the robot as a preexisting investment (Tables 3 and 4). This difference increased by $1607 if the robot basic cost was included. Intention-to-treat analysis was influenced by the surprisingly high number of women who were candidates for vaginal hysterectomy, and a per-protocol analysis of laparoscopic vs robotic-assisted hysterectomy was performed.
.85 .76 .44
When considering the robot as a preexisting investment, the hospital cost was similar for robotic-assisted and laparoscopic hysterectomy ($7016 vs $7059; p 5 .85), which was also found when comparing the robotic group in whom vaginal hysterectomy was not considered feasible by the assigned surgeon (n 5 38) with the 36 women who underwent a traditional laparoscopic approach ($7154 vs $7059; p 5 .72) (Table 4). Perioperative data are given in Table 5. Two conversions from laparoscopy to laparotomy occurred in the traditional minimally invasive group, one because of a large uterus
Table 3 Mean hospital cost (US$) per patient Procedure Traditional minimally invasive hysterectomy
Variable Patients, No. (%) Admittance feea Traditional minimally invasive basic cost Traditional minimally invasive specific cost Robot basic cost Robot specific costb Cost of conversion to laparotomy Mean operating room time, minc Postoperative stay, dayd Blood transfusion Vaginal procedure basic cost at repeat operation Vaginal procedure specific cost at repeat operation Total cost, mean (SD) p Value a b c d
Including readmission fee. Mean number of robotic instruments used, 2.8. Including operating room time at repeat operation. Including stay at readmission.
Cost per unit $671 $317 $377 $1607 $1876 $309 26 $549 $150 $104 $212
Units per patient 61 (50) 1.12 1 1
Mean cost per patient $752 $317 $377
0.03 145 1.42 0.03 0.07
$10 $3763 $778 $5 $7
0.07
$14 $6023 ($1881) ,.001
Robotic-assisted hysterectomy
Units per patient 61 (50) 1.05
Excluding Robot basic cost Mean cost per patient
Including Robot basic cost Mean cost per patient
$705
$705
0 (1) 1
0 $1876
$1607 $1876
147 1.12
$3815 $615
$3815 $615
0.02
$2
$2
0.02
$3
$3
$7016 ($1018) ,.001
$8623 ($1018) ,.001
L€ onnerfors et al.
Mean hospital cost (US$) per patient according to surgical approach Procedure
Variable Patients, No. (%) Admittance feeb Vaginal approach basic cost Vaginal approach specific cost Laparoscopic basic cost Laparoscopic specific cost Robot specific costc Cost of conversion to laparotomy Mean operating room time, mind Postoperative stay, daye Blood transfusion Total cost, mean (SD) p Value, traditional vs robotic a b c d e
$/Unit $671 $104 $212 $375 $309 $1876 $309 $26 $549 $150
Traditional Minally invasive hysterectomy
Robotic-assisted hysterectomya
Vaginal hysterectomy
Laparoscopic hysterectomy
All robotic patients
Robotic patients not suitable candidates for vaginal surgery
Units per patient
Units per patient
Units per patient
Units per patient
25 (20) 1.12 1.08 1.08 0.04 0.04
103 1.4 0.08
Mean cost per patient $752 $112 $229 $15 $12
$2678 $769 $12 $4579 ($1654) ,.001
Mean cost per patient
36 (30) 1.11 0.92 0.92 1 1
$746 $96 $195 $375 $309
0.06 174 1.4
$17 $4537 $784 $7059 ($1260) .85
Mean cost per patient
61 (50) 1.05
$705
1
$1876
1
$1876
147 1.12
$3815 $615
152 1.13
$3952 $621
$7016 ($1018)
38 1.05
Mean cost per patient
Traditional vs Robotic-Assisted Hysterectomy
Table 4
$705
$7154 ($1015)
Excluding robot basic cost. Including readmission fee. Mean number of robotic instruments used, was 2.8. Including operating room time at repeat operation. Including stay at readmission.
83
84
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 5 Perioperative dataa Procedure Variable
Vaginal
Laparoscopic
Traditional minimally invasive
Robotic
p Valueb
Patients Conversion to laparotomy Intraoperative bleeding, mL Intraoperative complications Operative time, min Operating room time, min Uuterine weight, g Concomitant procedures Inpatient time, day Postoperative complications Vaginal cuff hematoma Vaginal cuff dehiscence Port infection Repeat operation Readmission Change in hemoglobin concentration, g/L Postoperative temperature, C Postoperative C-reactive protein concentration, mg/L
25 0 50 (0–350) 0 59 (29–118) 91 (59–154) 152 (30–433) 4 (19) 1.4 (0.87) 5 (20) 5 (20) 0 0 2 (8) 3 (12) 15 (1–27) 37 (36–37.5) 15 (4.6–69)
36 2 (5.6) 100 (10–600) 1 (2.8) 104 (54–223) 163 (116–286) 163 (31–694) 27 (75) 1.4 (0.6) 7 (19.4) 6 (16.7) 1 (2.8) 0 2 (5.6) 4 (11) 18 (3–34) 36.7 (35.6–37.7) 12 (0.6–34)
61 (50) 2 (3.3) 100 (0–600) 1 (1.6) 86 (29–223) 148 (59–286) 154 (30–694) 31 (51) 1.4 (0.81) 12 (19.7) 11 (18) 1 (1.6) 0 4 (6.6) 7 (11.5) 16 (1–34) 36.7 (35.6–37.7) 13 (0.6–69)
61 (50) 0 50 (0–400) 1 (1.6) 76 (43–210) 140 (98–280) 180 (54–1114) 36 (59) 1.1 (0.52) 4 (6.6) 2 (3.3) 1 (1.6) 1 (1.6) 1 (1.6) 3 (4.9) 8 (0–24) 36.9 (36.0–38.0) 12 (1.4–90)
.50 .001 1.0 .54 .59 .53 .47 .09 .01 .02 1.0 1.0 .21 .32 .004 .23 .72
a b
Unless otherwise indicated, values are given as No. (%), median (SD), or median (range). Traditional minimally invasive vs robotic approach.
and the other because of a last-minute change to an inexperienced assistant. There was no difference in uterine weight (p 5 .53), median operating room time (p 5 .59), and median operative time (p 5 .54) between the traditional minimally invasive and robotic-assisted groups. Operative time was significantly related to uterine weight in both groups (p ,.01). The patients operated on per surgeon ranged from 3 to 50, with 3 surgeons (JP, PR, CB) performing 85% of the procedures; all 3 surgeons performed hysterectomies via all routes. Intraoperative complications were rare, with 1 accidental cystotomy occurring in each group; contributing factors were adhesions after previous cesarean section deliveries. Both lesions were repaired intraoperatively with no need for conversion. Median blood loss was lower in the robotic-assisted group, 50 mL vs 100 mL; p ,.05). No patient received an intraoperative blood transfusion. One patient in the traditional minimally invasive group received a postoperative blood transfusion because of occult bleeding. This patient had an extended hospital stay; however, no further intervention was needed. There were fewer complications in the robotic-assisted group (p 5 .01) (Table 5). Postoperatively, no patients experienced thrombosis or a gastrointestinal, cardiac, or pulmonary complication. A vaginal cuff hematoma developed in more women in the traditional minimally invasive group (11 of 61 [18%]); 3 of these patients (27%) underwent secondary surgery. The remaining 7 patients and the 2 in the
robotic-assisted group with vaginal cuff hematomas were managed conservatively. Vaginal cuff dehiscence developed in 1 woman in each group, and a secondary cuff-closure procedure was performed vaginally. In addition, 3 women in the traditional minimally invasive group underwent a repeat operation vaginally to evacuate a vaginal cuff hematoma. An additional 3 women in the traditional minimally invasive group and 2 in the robot-assisted group were readmitted because of symptoms associated with vaginal cuff hematomas or infection. Uncomplicated lower tract urinary infection was noted in 4 women in each group but was not included as a postoperative complication because it was not thought to be related to the surgical method. No further complications were observed at the 4-month follow-up. Discussion A paradigm shift from traditional minimally invasive hysterectomy to robotic-assisted hysterectomy in women with benign disease and uterine size %16 gestational weeks will lead to substantially increased hospital costs, primarily because of the lower cost of vaginal hysterectomy. In the present study, vaginal hysterectomy was possible in 41% patients, even after excluding women referred to undergo the vaginal approach. A better short-term clinical outcome was observed after robotic-assisted hysterectomy; however,
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
although the complication rate was low in all subgroups, it is debatable whether this can motivate the higher cost. However, per-protocol analysis indicates that laparoscopic and robotic-assisted hysterectomy can be performed at similar hospital cost because of higher robot capacity that entails excluding the cost of investment and maintenance, i.e., the basic cost of the robot. This cost differs among institutions, depending on the number of procedures performed; however, the difference becomes less pronounced when 300 to 400 procedures or more are performed annually and the cost for instruments and disposables accounts for most of the cost of the procedure. The power analysis of this randomized trial was based on an expected operating room time of 30 minutes less in the robotic group. Even though women referred for vaginal hysterectomy were excluded before randomization, a surprisingly high rate of women in the traditional minimally invasive group were suitable candidates for vaginal hysterectomy. The mean operating room time for vaginal hysterectomy was 71 minutes less than for laparoscopic hysterectomy, and aggregating the 2 approaches led to no difference in operating room time between robotic-assisted and traditional minimally invasive surgery. Consequently, this unanticipated difference in expected and observed mean operating room time in the traditional minimally invasive group was reflected in higher hospital cost for robotic surgery. However, a per-protocol analysis comparing laparoscopic and robotic-assisted hysterectomy resulted in a similar mean hospital cost. This was primarily due to the substantially shorter operating room time observed with roboticassisted procedures, which counteracted the increased cost of robotic draping, instruments, and sterilization. Despite the high rate of vaginal hysterectomy, our findings differ from those of 2 previous randomized studies that compared hysterectomy via laparoscopy vs roboticassisted surgery, which found similar clinical outcome but longer operative time with robotic surgery [18,19]. We believe this discrepancy reflects the fact that the robot, contrary to the above-mentioned studies, was well implemented at our institution before onset of our study. This is supported by recent studies that suggest a longer learning curve for robotics than initially reported [23–25]. All 6 participating surgeons were highly experienced, and no difference in operative time was observed between surgeons, hence enabling evaluation of the potential of the techniques rather than comparing individual surgeon skills. The low overall rate of intraoperative complications, 1.6%, was probably influenced by surgeon experience, as suggested by other authors [26,27]. This is further supported by Boggess et al [28], who reported a rate of 2.1% in 152 patients undergoing robotic-assisted hysterectomy, primarily to treat complex benign indications, in which most were performed by 1 highly experienced surgeon. Rosero et al [3] recently reported a rate of 4.67% for robotic-assisted hysterectomy and 5.33% for laparoscopic hysterectomy when investigating all hysterectomies per-
85
formed to treat benign disease in the United States in 2009–2010. The overall low rates of intraoperative complications and conversions and an acceptable rate of postoperative complications support the overall feasibility of minimally invasive surgery. This is further illustrated by recent reports of decreasing rates of abdominal hysterectomy and an increased use of minimally invasive surgery overall after the introduction of robotic-assisted surgery [3,15,16]. In the present study, because of the associated cost of disposable instruments, their use was kept at a minimum, which may not reflect the situation at other institutions. Expensive single-use sealing and cutting instruments are frequently used in traditional laparoscopy, as reported by Paraiso et al [18], whereas these instruments are rarely needed in robotic-assisted surgery. Hence, we may have underestimated the instrument cost for traditional laparoscopy in general. Whether this increased cost would be somewhat counteracted by shorter operative time, better hemostasis, and decreased incidence of vaginal cuff hematomas remains unknown. Major strengths of the present study are that it was performed at an institution with well-implemented programs within both traditional minimally invasive and roboticassisted surgery, and procedures were performed only by highly experienced surgeons. Additional strengths are good methods of allocation concealment, preventing an imbalance in the baseline prognostic value by excluding patients with factors believed to be strongly related to outcome, and using an intention-to-treat analysis. In addition, no protocol violations occurred, and no patients were lost to follow-up. The surprisingly high rate of vaginal hysterectomies led to the study being underpowered for comparing traditional laparoscopy with robotic-assisted hysterectomy, which is an unexpected weakness. An additional weakness is that although a trend was observed toward fewer conversions, repeat operations, and readmissions with robotic surgery, the study was not powered to detect differences in infrequent adverse advents. Ascertainment bias due to lack of blinding is unavoidable in a setting such as ours. However, blinding the administering clinician to the treatment allocation is impossible for all surgical trials, and at our hospital the patient must specifically either consent to the suggested surgical approach or be able to decline the procedure. The primary factor determining hospital cost is operative time, which is expected to be independent of whether the study was blinded. When used for gynecologic cancer surgery, given a high annual caseload and an experienced team, the robot can be economically viable [14,29,30]. To date, no other studies have found a clinical or economical benefit for robotics in comparison with traditional laparoscopic procedures for benign indications [3,15,16,18,19]. A per-protocol analysis of our results showed a similar hospital cost for laparoscopic and robotic-assisted hysterectomy to treat benign disease
86
when the latter is performed because of excess robot availability. This might indicate that robotic surgery can be economically feasible for treatment of complex benign cases such as severe endometriosis, intra-abdominal adhesions, high body mass index, or a large uterus. A randomized study addressing these issues is ongoing at our institution. Acknowledgments We would like to thank Christer Borgfeldt, MD, PhD, Pia Teleman, MD, PhD, and Thomas Bossmar, MD, PhD, Department of Obstetrics and Gynecology, Sk ane University Hospital and Lund University, Lund, Sweden for their contribution as participating surgeons. References 1. American College of Obstetricians and Gynecologists. ACOG Committee Opinion 444: choosing the route of hysterectomy for benign disease. Obstet Gynecol. 2009;114:1156–1158. 2. American Association of Gynecologic Laparoscopists. AAGL Position Statement: route of hysterectomy to treat benign uterine disease. J Minim Invasive Gynecol. 2011;18:1–3. 3. Rosero EB, Kho KA, Joshi GP, Giesecke M, Schaffer JI. Comparison of robotic and laparoscopic hysterectomy for benign gynecologic disease. Obstet Gynecol. 2013;122:778–786. 4. Wu JM, Wechter ME, Geller EJ, Nguyen TV, Visco AG. Hysterectomy rates in the United States, 2003. Obstet Gynecol. 2007;110:1091–1095. 5. Nieboer TE, Johnson N, Lethaby A, et al. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2009;(8):CD003677. 6. Stjerndahl JH, L€ ofgren M, Renstr€om B. Yearly report on hysterectomies performed for benign indications in 2011. Reconfirmation from the register on gynecologic surgeries. 2012. Available at: http://www.gynop. org/rapportering/rapporter/GynopHystrapport2012.pdf. Accessed December 19, 2013. 7. Wright KN, Jonsdottir GM, Jorgensen S, Shah N, Einarsson JI. Costs and outcomes of abdominal, vaginal. laparoscopic and robotic hysterectomies. JSLS. 2012;16:519–524. 8. Einarsson JI, Matteson KA, Schulkin J, Chavan NR, SangiHaghpeykar H. Minimally invasive hysterectomies: a survey on attitudes and barriers among practicing gynecologists. J Minim Invasive Gynecol. 2010;17:167–175. 9. Liu H, Lu D, Wang L, Shi G, Song H, Clarke J. Robotic surgery for benign gynaecological disease. Cochrane Database Syst Rev. 2012;(2):1–26. 10. Pasic RP, Rizzo JA, Fang H, Ross S, Moore M, Gunnarsson C. Comparing robot-assisted with conventional laparoscopic hysterectomy: impact on cost and clinical outcomes. J Minim Invasive Gynecol. 2010;17:730–738. 11. Reza M, Maeso S, Blasco JA, Andradas E. Meta-analysis of observational studies on the safety and effectiveness of robotic gynaecological surgery. Br J Surg. 2010;97:1772–1783. 12. Sarlos D, Kots LA. Robotic versus laparoscopic hysterectomy: a review of recent comparative studies. Curr Opin Obstet Gynecol. 2011;4:283–288.
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015 13. Barbash GI, Glied SA. New technology and health care costs: the case of robot-assisted surgery. N Engl J Med. 2010;363:701–704. 14. Barnett JC, Judd JP, Wu JM, Scales CD Jr, Myers ER, Havrilesky LJ. Cost comparison among robotic, laparoscopic, and open hysterectomy for endometrial cancer. Obstet Gynecol. 2010;116:685–693. 15. Wright JD, Ananth CV, Lewin SN, et al. Robotically assisted vs laparoscopic hysterectomy among women with benign gynecologic disease. JAMA. 2013;309:689–698. 16. Jonsdottir GM, Jorgensen S, Cohen SL, et al. Increasing minimally invasive hysterectomy: effect on cost and complications. Obstet Gynecol. 2011;117:1142–1149. 17. O’Neill M, Moran PS, Teljeur C, et al. Robot-assisted hysterectomy compared to open and laparoscopic approaches: systematic review and meta-analysis. Arch Gynecol Obstet. 2013;287:907–918. 18. Paraiso MF, Ridgeway B, Park AJ, et al. A randomized trial comparing conventional and robotically assisted total laparoscopic hysterectomy. Am J Obstet Gynecol. 2013;208:368.e1–368.e7. 19. Sarlos D, Kots L, Stevanovic N, von Felten S, Sch€ar G. Robotic compared with conventional laparoscopic hysterectomy: a randomized controlled trial. Obstet Gynecol. 2012;120:604–611. 20. L€onnerfors C, Persson J. Implementation and applications of robotic surgery within gynecologic oncology and gynecology: analysis of the first thousand cases. Ceska Gynekol. 2013;78:12–19. 21. Persson J, Reynisson P, Borgfeldt C, Kannisto P, Lindahl B, Bossmar T. Robot assisted laparoscopic radical hysterectomy and pelvic lymphadenectomy with short and long term morbidity data. Gynecol Oncol. 2009; 113:185–190. 22. Persson J, Imboden S, Reynisson P, Andersson B, Borgfeldt C, Bossmar T. Reproducibility and accuracy of robot-assisted laparoscopic fertility sparing radical trachelectomy. Gynecol Oncol. 2012; 127:484–488. 23. Woelk JL, Casiano ER, Weaver AL, Gostout BS, Trabuco EC, Gebhart JB. The learning curve of robotic hysterectomy. Obstet Gynecol. 2013;121:87–95. 24. Lenihan JP, Kovanda C, Seshadri-Kreaden U. What is the learning curve for robotic assisted gynecologic surgery. J Minim Invasive Gynecol. 2008;15:589–594. 25. Bell MC, Torgerson JL, Kreaden U. The first 100 da Vinci hysterectomies: an analysis of the learning curve. S D Med. 2009;62:91. 93–95. 26. Garry R, Fountain J, Mason S, et al. The eVALuate study: two parallel randomised trials, one comparing laparoscopic with abdominal hysterectomy, the other comparing laparoscopic with vaginal hysterectomy. BMJ. 2004;328:129. 27. Tian YF, Lin YS, Lu CL, et al. Major complications of operative gynecologic laparoscopy in Southern Taiwan: a follow-up study. J Minim Invasive Gynecol. 2007;14:284–292. 28. Boggess JF, Gehrig PA, Cantrell L, et al. Perioperative outcomes of robotically assisted hysterectomy for benign cases with complex pathology. Obstet Gynecol. 2009;114:585–593. 29. Reynisson P, Persson J. Hospital costs for robot-assisted laparoscopic radical hysterectomy and pelvic lymphadenectomy. Gynecol Oncol. 2013;130:95–99. 30. Lau S, Vaknin Z, Ramana-Kumar AV, Halliday D, Franco EL, Gotlieb WH. Outcomes and cost comparison after introducing robotics program for endometrial cancer surgery. Obstet Gynecol. 2012;119: 717–724.
A Randomized Trial Comparing Vaginal and Laparoscopic Hysterectomy vs Robot-Assisted Hysterectomy Celine L€ onnerfors, MD, Petur Reynisson, MD, PhD, and Jan Persson, MD, PhD* From the Department of Obstetrics and Gynecology, Sk ane University Hospital and Lund University, Lund, Sweden (all authors).
ABSTRACT Study Objective: To investigate the hospital cost and short-term clinical outcome of traditional minimally invasive hysterectomy vs robot-assisted hysterectomy in women primarily not considered candidates for vaginal surgery. Design: Randomized controlled trial (Canadian Task Force classification I). Setting: University Hospital in Sweden. Patients: One hundred twenty-two women with uterine size %16 gestational weeks scheduled to undergo minimally invasive hysterectomy because of benign disease. Interventions: Robot-assisted hysterectomy or traditional vaginal or laparoscopic minimally invasive hysterectomy. Measurements and Main Results: All women underwent surgery as randomized. There were no demographic differences between the 2 groups. Vaginal hysterectomy was possible in 41% in the traditional minimally invasive group, at a mean hospital cost of $4579 compared with $7059 for traditional laparoscopic hysterectomy. This was reflected in a mean hospital cost of $993 more per robotic-assisted hysterectomy than for traditional minimally invasive hysterectomy when the robot was a preexisting investment. This hospital cost increased by $1607 when including investments and cost of maintenance. A perprotocol subanalysis comparing laparoscopy and robotics demonstrated similar hospital cost when the robot was a preexisting investment ($7059 vs $7016). Robotic-assisted hysterectomy was associated with less blood loss and fewer postoperative complications. Conclusion: A similar hospital cost can be attained for laparoscopy and robotics when the robot is a preexisting investment. From the perspective of hospital costs, robotic-assisted hysterectomy is not advantageous for treating benign conditions when a vaginal approach is feasible in a high proportion of patients. Journal of Minimally Invasive Gynecology (2015) 22, 78–86 Ó 2015 AAGL. All rights reserved. Keywords:
DISCUSS
Hysterectomy; Laparoscopic hysterectomy; Minimally invasive surgery; Robot-assisted laparoscopy; Vaginal hysterectomy
You can discuss this article with its authors and with other AAGL members at http://www.AAGL.org/jmig-22-1-JMIG-D-14-00303
Use your Smartphone to scan this QR code and connect to the discussion forum for this article now* * Download a free QR Code scanner by searching for ‘‘QR scanner’’ in your smartphone’s app store or app marketplace.
The potential benefit of robot-assisted laparoscopic surgery is enabling a higher proportion of minimally invasive surgical procedures. Despite guidelines supporting minimally invasive procedures, hysterectomy to treat benign gynecologic disease is still most commonly performed via laparotomy [1–9]. Vaginal hysterectomy is primarily performed in conjunction with surgery to treat prolapse, Disclosures: Dr. Persson is a proctor in robotic surgery. Corresponding author: Jan Persson, MD, PhD, Department of Obstetrics and Gynecology, Sk ane University Hospital, SE-22185 Lund, Sweden. E-mail: [email protected] Submitted June 18, 2014. Accepted for publication July 12, 2014. Available at www.sciencedirect.com and www.jmig.org 1553-4650/$ - see front matter Ó 2015 AAGL. All rights reserved. http://dx.doi.org/10.1016/j.jmig.2014.07.010
and the rate in the United States decreased from 22% in 2003 to 19% in 2009–2010, which coincides with the introduction of robotic-assisted surgery [1–4]. Robot-assisted laparoscopy has been widely adopted to treat benign gynecologic conditions, although no data have demonstrated a clinical or economic benefit over other operative approaches [3,10–17]. Recently, 2 large cohort studies found similar morbidity profiles as for laparoscopic hysterectomy, but a substantially increase in cost for robotic-assisted surgical procedures. However, factors that might influence the route of hysterectomy chosen, such as body mass index, uterine weight, and previous abdominal surgery, were not available [3,15]. A 2012 Cochrane review identified 2 randomized controlled trials of benign
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
gynecologic robotic-assisted hysterectomy and concluded that robotic surgery was not associated with improved effectiveness or safety [9]. However, both studies were potentially biased by inclusion of early robot adopters [9,18,19]. The Department of Obstetrics and Gynecology at Sk ane University Hospital is a tertiary referral unit for both gynecologic oncology and complex benign gynecologic surgery. Robotic surgery was introduced in October 2005, and to date .1600 women have undergone robotic-assisted surgery, with approximately 300 procedures performed each year [20–22]. Before the implementation of robotic-assisted surgery, traditional laparoscopy was routinely used for hysterectomy, and minimally invasive surgery was used to perform 78% of all hysterectomies to treat benign disease in 2012. The primary objective of the present study was to investigate the hospital cost of robotic-assisted hysterectomy compared with traditional minimally invasive hysterectomy (vaginal and laparoscopic) performed to treat benign gynecologic disorders in women with uterine size %16 gestational weeks after excluding women referred for vaginal hysterectomy. The secondary objective was to assess shortterm clinical outcome. Material and Methods Between January 2010 and June 2013, 125 women meeting the inclusion criteria were offered participation in the study (Fig. 1; Table 1). Preoperative evaluation in all women included medical and surgical history, and clinical examination including a gynecologic examination and vaginal ultrasonography. Each woman was assigned an individual clinical research file containing all study protocols. One hundred twenty-two opaque envelopes containing the assigned surgical method in the proportion of 1:1 were closed, shuffled, and then numbered. After inclusion, randomization occurred via telephone during which the envelopes were opened in consecutive order at the central randomization office. The date, clinical research file number, patient name and social security number, and the assigned surgical method were recorded in the central study log. Before randomization, surgical procedures were scheduled to be performed on a day when it was possible to include all 3 approaches, and after randomization the patient was assigned a specific surgeon. All patients were informed of their assignment. The route of traditional minimally invasive surgery was chosen by the designated surgeon, with vaginal hysterectomy as the first choice, followed by laparoscopic hysterectomy. The necessity of concomitant procedures, vaginal access, body mass index, and the presence of adhesions or endometriosis influenced the choice of surgical approach. Robot-assisted laparoscopic hysterectomy was performed using the da Vinci Si Surgical System (Intuitive Surgical, Inc., Sunnyvale CA). All 6 surgeons were consultants experienced in both vaginal and laparoscopic surgery, and 4 were gynecologic oncolo-
79
gists experienced in robotic-assisted surgery. The case load of hysterectomy to treat benign disease before commencing the study was approximately 110 annually, with 31% performed vaginally, 27% laparoscopically, and 15% robotically. However, most hysterectomies performed at our institution are to treat malignant disease, with the greatest proportion performed by the 4 gynecologic oncologists. The least experienced robotic surgeon had performed 49 robotic hysterectomies before the study, and a total of 231 robotic-assisted procedures by the end of the study, primarily in gynecologic oncology. All women received oral prophylactic antibiotic therapy including 200 mg doxycycline (Doxyferm; Nordic Group BV, Hoofddorp, Holland) and 800 mg metronidazole (Flagyl; Sanofi-Aventis, Paris, France). All procedures were performed with the patient under general anesthesia. For robotic-assisted hysterectomy, two or three 8-mm robot trocars and one assistant 5- or 12-mm trocar were used. Laparoscopic hysterectomy was performed with the use of 4 ports: a reusable umbilical port or a 12-mm port (Xcel; Ethicon EndoSurgery, Inc., Somerville, NJ) for the optics and 3 assistant ports, either three 5-mm or two 5-mm plus one 10-mm ports, in the lower quadrants. Vaginal hysterectomy was performed in the standard manner. For robot-assisted and total laparoscopic hysterectomy the peritoneum of the lateral sidewalls was opened, the ureters were visualized and lateralized, and the propria pelvic ligament or infundibulopelvic ligament and the round ligaments were divided, followed by electrocoagulation of the uterine arteries and cardinal ligaments. Then the vagina was incised, and the vaginal cuff was sutured using polyglactin 910 absorbable sutures (Vicryl; Ethicon) or V-Loc 180 sutures (Covidien; Mansfield, MA), at surgeon discretion either robotically, laparoscopically, or vaginally, with the latter in patients after vaginal coring of the uterus. During laparoscopic vaginal hysterectomy the procedure was similar except that the uterine arteries and cardinal ligaments were divided vaginally. Postoperative cystoscopy was not performed. Total operating room time (from patient entry to departure from the operating room, including administration of anesthesia), total operative time (skin to skin including placement of a catheter, application of a fornix presenter, docking, and dedocking), intraoperative blood loss, and complications were recorded. Hemoglobin and C-reactive protein concentrations and body temperature were noted on the first postoperative day. All women received similar pain medication and were given a daily dose of 4500 IE tinzaparin (Innohep; LEO Pharma AB, Copenhagen, Denmark) for 10 days postoperatively. Length of stay and immediate postoperative complications were recorded. All patients were advised to contact the department if necessary. A postoperative visit was planned for 4 months after surgery. To assess short-term clinical outcome, all adverse events within this period were included with the exception of uncomplicated lower urinary tract infections because this was thought to be unrelated to the surgical method.
80
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Fig. 1 Progression of patients through the study.
Enrollment
Eligible subjects offered participation (n =125)
Excluded (n =3) Declined participation (n =3)
Allocated to traditional minimally invasive surgery (n = 61)
Underwent robotic hysterectomy (n = 61)
Follow-up
Underwent allocated intervention (n = 61) Vaginal hysterectomy (n = 25 [41%]) Laparoscopic hysterectomy (n= 36[59%])
Allocated to robotic surgery (n = 61)
Lost to follow-up (n = 0) Discontinued intervention (n = 0)
Analysis
Allocation
Randomized (n = 122)
Analyzed (n = 61)
Analyzed (n = 61)
Excluded from analysis (n = 0)
Excluded from analysis (n = 0)
For estimation of cost-generating factors common to traditional, minimally invasive, and robotic-assisted hysterectomy we used admittance fee, operating room time, hospital length of stay, and cost of complications, readmission, and repeat interventions until 4 months after surgery. The hospital internal charge is based on the average real cost for the respective parameter. The per-minute charge for use of the operating room is based on the previous year’s
Lost to follow-up (n = 0) Discontinued intervention (n = 0)
mean institutional cost and includes the operating theater, 1 surgeon, 1 assistant, 1 scrub nurse, 1 circulating nurse, anesthetic staff, cleaning, and basic expendables such as gowns and gloves. The procedure cost for each approach was calculated, and included the basic cost and the specific cost, depending on the surgical approach. The traditional minimally invasive basic cost comprised the purchase price of the vaginal and/or laparoscopic equipment including
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
Table 1 Inclusion and exclusion criteria Inclusion criteria
Exclusion criteria
Need for total hysterectomy to treat benign indication Uterus size %16 gestational weeks Uterus/vagina size enabling vaginal retrieval, allowing for coring, when necessary No desire for additional pregnancies Informed consent
Suspicion of or known malignancy Known extensive intra-abdominal adhesions Contraindication for laparoscopic surgery Pacemaker or other electrosensitive implant Known bleeding disorder Immunosuppression therapy or other known increased risk of infection Women referred for vaginal hysterectomy Simultaneous need for prolapse surgery Allergy to metronidazole and/or doxycycline Inability to understand patient information
annual reinvestment, maintenance, and instrument wear and tear, with an estimated depreciation time of 7 years and an annual turnover of 300 procedures. In addition, the traditional minimally invasive specific cost included the cost of patient draping, sterilization of all instruments, and cost of disposable instruments when applicable, such as a 12-mm assistant port and a 5-mm suction-irrigation system (Stryker Endoscopy, San Jose, CA). A non-disposable fornix presenter was used. The robot basic cost, including investment and maintenance, was calculated from the purchase price of the da Vinci Si system, using a depreciation time of 7 years, the annual maintenance fee, and an annual turnover of 300 procedures. Because the robotic system was purchased for use within gynecologic oncology and for complex benign cases, other benign cases are performed only as excess capacity procedures, and thus in a setting such as ours the robot can be considered a preexisting investment insofar as less complicated procedures. When considering the robot as a preexisting investment, the robot basic cost was excluded from the robot procedure cost. The robot specific cost included the cost of patient draping, sterilization of all instruments including robotic instruments, ports and optics, robot draping (3- or 4-arm kit), cost of the actual number of robotic instruments, and disposable instruments used. A modified non-disposable trocar (Ternamian Endotip; Karl Storz GmbH & Co KG, Tuttlingen, Germany) hosted the optics. The electrocoagulation devices used were the bipolar forceps and monopolar scissors for traditional laparoscopy and their robotic counterparts for robotic-assisted surgery. Both
81
for robotic-assisted and traditional laparoscopy, standard reusable laparoscopic instruments were used, and the depreciation cost was estimated to be the same for robotic-assisted and traditional minimally invasive procedures. Because of their high cost, disposable sealing or cutting instruments or morcellators were not used. Inasmuch as all procedures were minimally invasive, no differences in societal cost was expected, and therefore was not included. For costs charged in Swedish Crowns (SEK) or Euros (V), we used the mean currency exchange rate between the US dollar and SEK or V for 2010 to 2012 as estimated by Swedish tax authorities (1$ 5 6.4294 SEK or 0.7411 V). The major factor that influences hospital cost at our institution is the operative time. Existing data on surgical times for robotic-assisted hysterectomy were analyzed (Analyseit; Analyse-it Software Ltd, Leeds, UK), found to have positive skewness, and therefore transformed into natural logarithms to perform the power analysis. A similar skewness was supposed for data from traditional minimally invasive surgery. We determined that 96 patients were needed to detect a difference of 30 minutes in operating room time between traditional minimally invasive and roboticassisted hysterectomy, with 80% power and a significance level of .05. To compensate for the uncertainty in the distribution of traditional minimally invasive data and the uncertainty of the number of hysterectomies available for the vaginal approach, we estimated a need for another 15%. Accounting for a drop out rate of 10%, a total of 122 women were included in the study. Institutional review board approval was obtained. The study was registered (www.clinicaltrials.gov NCT01865929). All included women gave written informed consent. Statistical analysis was performed using commercially available software (SPSS version 20; IBM Corp., Armonk, NY). For statistical analyses we used the c2 test, the Student t-test, or the Mann-Whitney U test, as appropriate. The gaussian distribution of the groups was tested using the Kolmogorov-Smirnov fitness test, and correlation was tested using the Pearson correlation. The power analysis was made using an unpaired t-test. All tests were 2-sided, and p ,.05 was considered significant. Results One hundred twenty-two women consented to participate in the study, were enrolled and randomized, and underwent surgery (Fig. 1). There were no demographic differences between groups (Table 2). Uterine myomas (44%) and abnormal bleeding (31%) were the most common indications for surgery. The mean operating room time was 145 minutes for traditional minimally invasive hysterectomy (103 minutes for vaginal hysterectomy and 174 minutes for laparoscopic hysterectomy), compared with 147 minutes for robotic-assisted
82
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 2 Demographic data Procedure Variable
Traditional minimally invasive hysterectomy
Robotic-assisted hysterectomy
p Value
Age, yr, median (range) Body mass index, median (range) Median parity, No. (range) No vaginal deliveries, % Previous abdominal surgery, % Previous cesarean delivery, %
46 (29–69) 24.9 (17.6–42) 2 (0–5) 21 56 18
47 (27–65) 24.9 (17–39.2) 2 (0–6) 23 54 23
.89 .86
hysterectomy. This being the major factor influencing hospital cost led to a mean hospital cost per patient of $993 less for traditional minimally invasive hysterectomy as a whole compared with robotic-assisted hysterectomy when considering the robot as a preexisting investment (Tables 3 and 4). This difference increased by $1607 if the robot basic cost was included. Intention-to-treat analysis was influenced by the surprisingly high number of women who were candidates for vaginal hysterectomy, and a per-protocol analysis of laparoscopic vs robotic-assisted hysterectomy was performed.
.85 .76 .44
When considering the robot as a preexisting investment, the hospital cost was similar for robotic-assisted and laparoscopic hysterectomy ($7016 vs $7059; p 5 .85), which was also found when comparing the robotic group in whom vaginal hysterectomy was not considered feasible by the assigned surgeon (n 5 38) with the 36 women who underwent a traditional laparoscopic approach ($7154 vs $7059; p 5 .72) (Table 4). Perioperative data are given in Table 5. Two conversions from laparoscopy to laparotomy occurred in the traditional minimally invasive group, one because of a large uterus
Table 3 Mean hospital cost (US$) per patient Procedure Traditional minimally invasive hysterectomy
Variable Patients, No. (%) Admittance feea Traditional minimally invasive basic cost Traditional minimally invasive specific cost Robot basic cost Robot specific costb Cost of conversion to laparotomy Mean operating room time, minc Postoperative stay, dayd Blood transfusion Vaginal procedure basic cost at repeat operation Vaginal procedure specific cost at repeat operation Total cost, mean (SD) p Value a b c d
Including readmission fee. Mean number of robotic instruments used, 2.8. Including operating room time at repeat operation. Including stay at readmission.
Cost per unit $671 $317 $377 $1607 $1876 $309 26 $549 $150 $104 $212
Units per patient 61 (50) 1.12 1 1
Mean cost per patient $752 $317 $377
0.03 145 1.42 0.03 0.07
$10 $3763 $778 $5 $7
0.07
$14 $6023 ($1881) ,.001
Robotic-assisted hysterectomy
Units per patient 61 (50) 1.05
Excluding Robot basic cost Mean cost per patient
Including Robot basic cost Mean cost per patient
$705
$705
0 (1) 1
0 $1876
$1607 $1876
147 1.12
$3815 $615
$3815 $615
0.02
$2
$2
0.02
$3
$3
$7016 ($1018) ,.001
$8623 ($1018) ,.001
L€ onnerfors et al.
Mean hospital cost (US$) per patient according to surgical approach Procedure
Variable Patients, No. (%) Admittance feeb Vaginal approach basic cost Vaginal approach specific cost Laparoscopic basic cost Laparoscopic specific cost Robot specific costc Cost of conversion to laparotomy Mean operating room time, mind Postoperative stay, daye Blood transfusion Total cost, mean (SD) p Value, traditional vs robotic a b c d e
$/Unit $671 $104 $212 $375 $309 $1876 $309 $26 $549 $150
Traditional Minally invasive hysterectomy
Robotic-assisted hysterectomya
Vaginal hysterectomy
Laparoscopic hysterectomy
All robotic patients
Robotic patients not suitable candidates for vaginal surgery
Units per patient
Units per patient
Units per patient
Units per patient
25 (20) 1.12 1.08 1.08 0.04 0.04
103 1.4 0.08
Mean cost per patient $752 $112 $229 $15 $12
$2678 $769 $12 $4579 ($1654) ,.001
Mean cost per patient
36 (30) 1.11 0.92 0.92 1 1
$746 $96 $195 $375 $309
0.06 174 1.4
$17 $4537 $784 $7059 ($1260) .85
Mean cost per patient
61 (50) 1.05
$705
1
$1876
1
$1876
147 1.12
$3815 $615
152 1.13
$3952 $621
$7016 ($1018)
38 1.05
Mean cost per patient
Traditional vs Robotic-Assisted Hysterectomy
Table 4
$705
$7154 ($1015)
Excluding robot basic cost. Including readmission fee. Mean number of robotic instruments used, was 2.8. Including operating room time at repeat operation. Including stay at readmission.
83
84
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015
Table 5 Perioperative dataa Procedure Variable
Vaginal
Laparoscopic
Traditional minimally invasive
Robotic
p Valueb
Patients Conversion to laparotomy Intraoperative bleeding, mL Intraoperative complications Operative time, min Operating room time, min Uuterine weight, g Concomitant procedures Inpatient time, day Postoperative complications Vaginal cuff hematoma Vaginal cuff dehiscence Port infection Repeat operation Readmission Change in hemoglobin concentration, g/L Postoperative temperature, C Postoperative C-reactive protein concentration, mg/L
25 0 50 (0–350) 0 59 (29–118) 91 (59–154) 152 (30–433) 4 (19) 1.4 (0.87) 5 (20) 5 (20) 0 0 2 (8) 3 (12) 15 (1–27) 37 (36–37.5) 15 (4.6–69)
36 2 (5.6) 100 (10–600) 1 (2.8) 104 (54–223) 163 (116–286) 163 (31–694) 27 (75) 1.4 (0.6) 7 (19.4) 6 (16.7) 1 (2.8) 0 2 (5.6) 4 (11) 18 (3–34) 36.7 (35.6–37.7) 12 (0.6–34)
61 (50) 2 (3.3) 100 (0–600) 1 (1.6) 86 (29–223) 148 (59–286) 154 (30–694) 31 (51) 1.4 (0.81) 12 (19.7) 11 (18) 1 (1.6) 0 4 (6.6) 7 (11.5) 16 (1–34) 36.7 (35.6–37.7) 13 (0.6–69)
61 (50) 0 50 (0–400) 1 (1.6) 76 (43–210) 140 (98–280) 180 (54–1114) 36 (59) 1.1 (0.52) 4 (6.6) 2 (3.3) 1 (1.6) 1 (1.6) 1 (1.6) 3 (4.9) 8 (0–24) 36.9 (36.0–38.0) 12 (1.4–90)
.50 .001 1.0 .54 .59 .53 .47 .09 .01 .02 1.0 1.0 .21 .32 .004 .23 .72
a b
Unless otherwise indicated, values are given as No. (%), median (SD), or median (range). Traditional minimally invasive vs robotic approach.
and the other because of a last-minute change to an inexperienced assistant. There was no difference in uterine weight (p 5 .53), median operating room time (p 5 .59), and median operative time (p 5 .54) between the traditional minimally invasive and robotic-assisted groups. Operative time was significantly related to uterine weight in both groups (p ,.01). The patients operated on per surgeon ranged from 3 to 50, with 3 surgeons (JP, PR, CB) performing 85% of the procedures; all 3 surgeons performed hysterectomies via all routes. Intraoperative complications were rare, with 1 accidental cystotomy occurring in each group; contributing factors were adhesions after previous cesarean section deliveries. Both lesions were repaired intraoperatively with no need for conversion. Median blood loss was lower in the robotic-assisted group, 50 mL vs 100 mL; p ,.05). No patient received an intraoperative blood transfusion. One patient in the traditional minimally invasive group received a postoperative blood transfusion because of occult bleeding. This patient had an extended hospital stay; however, no further intervention was needed. There were fewer complications in the robotic-assisted group (p 5 .01) (Table 5). Postoperatively, no patients experienced thrombosis or a gastrointestinal, cardiac, or pulmonary complication. A vaginal cuff hematoma developed in more women in the traditional minimally invasive group (11 of 61 [18%]); 3 of these patients (27%) underwent secondary surgery. The remaining 7 patients and the 2 in the
robotic-assisted group with vaginal cuff hematomas were managed conservatively. Vaginal cuff dehiscence developed in 1 woman in each group, and a secondary cuff-closure procedure was performed vaginally. In addition, 3 women in the traditional minimally invasive group underwent a repeat operation vaginally to evacuate a vaginal cuff hematoma. An additional 3 women in the traditional minimally invasive group and 2 in the robot-assisted group were readmitted because of symptoms associated with vaginal cuff hematomas or infection. Uncomplicated lower tract urinary infection was noted in 4 women in each group but was not included as a postoperative complication because it was not thought to be related to the surgical method. No further complications were observed at the 4-month follow-up. Discussion A paradigm shift from traditional minimally invasive hysterectomy to robotic-assisted hysterectomy in women with benign disease and uterine size %16 gestational weeks will lead to substantially increased hospital costs, primarily because of the lower cost of vaginal hysterectomy. In the present study, vaginal hysterectomy was possible in 41% patients, even after excluding women referred to undergo the vaginal approach. A better short-term clinical outcome was observed after robotic-assisted hysterectomy; however,
L€ onnerfors et al.
Traditional vs Robotic-Assisted Hysterectomy
although the complication rate was low in all subgroups, it is debatable whether this can motivate the higher cost. However, per-protocol analysis indicates that laparoscopic and robotic-assisted hysterectomy can be performed at similar hospital cost because of higher robot capacity that entails excluding the cost of investment and maintenance, i.e., the basic cost of the robot. This cost differs among institutions, depending on the number of procedures performed; however, the difference becomes less pronounced when 300 to 400 procedures or more are performed annually and the cost for instruments and disposables accounts for most of the cost of the procedure. The power analysis of this randomized trial was based on an expected operating room time of 30 minutes less in the robotic group. Even though women referred for vaginal hysterectomy were excluded before randomization, a surprisingly high rate of women in the traditional minimally invasive group were suitable candidates for vaginal hysterectomy. The mean operating room time for vaginal hysterectomy was 71 minutes less than for laparoscopic hysterectomy, and aggregating the 2 approaches led to no difference in operating room time between robotic-assisted and traditional minimally invasive surgery. Consequently, this unanticipated difference in expected and observed mean operating room time in the traditional minimally invasive group was reflected in higher hospital cost for robotic surgery. However, a per-protocol analysis comparing laparoscopic and robotic-assisted hysterectomy resulted in a similar mean hospital cost. This was primarily due to the substantially shorter operating room time observed with roboticassisted procedures, which counteracted the increased cost of robotic draping, instruments, and sterilization. Despite the high rate of vaginal hysterectomy, our findings differ from those of 2 previous randomized studies that compared hysterectomy via laparoscopy vs roboticassisted surgery, which found similar clinical outcome but longer operative time with robotic surgery [18,19]. We believe this discrepancy reflects the fact that the robot, contrary to the above-mentioned studies, was well implemented at our institution before onset of our study. This is supported by recent studies that suggest a longer learning curve for robotics than initially reported [23–25]. All 6 participating surgeons were highly experienced, and no difference in operative time was observed between surgeons, hence enabling evaluation of the potential of the techniques rather than comparing individual surgeon skills. The low overall rate of intraoperative complications, 1.6%, was probably influenced by surgeon experience, as suggested by other authors [26,27]. This is further supported by Boggess et al [28], who reported a rate of 2.1% in 152 patients undergoing robotic-assisted hysterectomy, primarily to treat complex benign indications, in which most were performed by 1 highly experienced surgeon. Rosero et al [3] recently reported a rate of 4.67% for robotic-assisted hysterectomy and 5.33% for laparoscopic hysterectomy when investigating all hysterectomies per-
85
formed to treat benign disease in the United States in 2009–2010. The overall low rates of intraoperative complications and conversions and an acceptable rate of postoperative complications support the overall feasibility of minimally invasive surgery. This is further illustrated by recent reports of decreasing rates of abdominal hysterectomy and an increased use of minimally invasive surgery overall after the introduction of robotic-assisted surgery [3,15,16]. In the present study, because of the associated cost of disposable instruments, their use was kept at a minimum, which may not reflect the situation at other institutions. Expensive single-use sealing and cutting instruments are frequently used in traditional laparoscopy, as reported by Paraiso et al [18], whereas these instruments are rarely needed in robotic-assisted surgery. Hence, we may have underestimated the instrument cost for traditional laparoscopy in general. Whether this increased cost would be somewhat counteracted by shorter operative time, better hemostasis, and decreased incidence of vaginal cuff hematomas remains unknown. Major strengths of the present study are that it was performed at an institution with well-implemented programs within both traditional minimally invasive and roboticassisted surgery, and procedures were performed only by highly experienced surgeons. Additional strengths are good methods of allocation concealment, preventing an imbalance in the baseline prognostic value by excluding patients with factors believed to be strongly related to outcome, and using an intention-to-treat analysis. In addition, no protocol violations occurred, and no patients were lost to follow-up. The surprisingly high rate of vaginal hysterectomies led to the study being underpowered for comparing traditional laparoscopy with robotic-assisted hysterectomy, which is an unexpected weakness. An additional weakness is that although a trend was observed toward fewer conversions, repeat operations, and readmissions with robotic surgery, the study was not powered to detect differences in infrequent adverse advents. Ascertainment bias due to lack of blinding is unavoidable in a setting such as ours. However, blinding the administering clinician to the treatment allocation is impossible for all surgical trials, and at our hospital the patient must specifically either consent to the suggested surgical approach or be able to decline the procedure. The primary factor determining hospital cost is operative time, which is expected to be independent of whether the study was blinded. When used for gynecologic cancer surgery, given a high annual caseload and an experienced team, the robot can be economically viable [14,29,30]. To date, no other studies have found a clinical or economical benefit for robotics in comparison with traditional laparoscopic procedures for benign indications [3,15,16,18,19]. A per-protocol analysis of our results showed a similar hospital cost for laparoscopic and robotic-assisted hysterectomy to treat benign disease
86
when the latter is performed because of excess robot availability. This might indicate that robotic surgery can be economically feasible for treatment of complex benign cases such as severe endometriosis, intra-abdominal adhesions, high body mass index, or a large uterus. A randomized study addressing these issues is ongoing at our institution. Acknowledgments We would like to thank Christer Borgfeldt, MD, PhD, Pia Teleman, MD, PhD, and Thomas Bossmar, MD, PhD, Department of Obstetrics and Gynecology, Sk ane University Hospital and Lund University, Lund, Sweden for their contribution as participating surgeons. References 1. American College of Obstetricians and Gynecologists. ACOG Committee Opinion 444: choosing the route of hysterectomy for benign disease. Obstet Gynecol. 2009;114:1156–1158. 2. American Association of Gynecologic Laparoscopists. AAGL Position Statement: route of hysterectomy to treat benign uterine disease. J Minim Invasive Gynecol. 2011;18:1–3. 3. Rosero EB, Kho KA, Joshi GP, Giesecke M, Schaffer JI. Comparison of robotic and laparoscopic hysterectomy for benign gynecologic disease. Obstet Gynecol. 2013;122:778–786. 4. Wu JM, Wechter ME, Geller EJ, Nguyen TV, Visco AG. Hysterectomy rates in the United States, 2003. Obstet Gynecol. 2007;110:1091–1095. 5. Nieboer TE, Johnson N, Lethaby A, et al. Surgical approach to hysterectomy for benign gynaecological disease. Cochrane Database Syst Rev. 2009;(8):CD003677. 6. Stjerndahl JH, L€ ofgren M, Renstr€om B. Yearly report on hysterectomies performed for benign indications in 2011. Reconfirmation from the register on gynecologic surgeries. 2012. Available at: http://www.gynop. org/rapportering/rapporter/GynopHystrapport2012.pdf. Accessed December 19, 2013. 7. Wright KN, Jonsdottir GM, Jorgensen S, Shah N, Einarsson JI. Costs and outcomes of abdominal, vaginal. laparoscopic and robotic hysterectomies. JSLS. 2012;16:519–524. 8. Einarsson JI, Matteson KA, Schulkin J, Chavan NR, SangiHaghpeykar H. Minimally invasive hysterectomies: a survey on attitudes and barriers among practicing gynecologists. J Minim Invasive Gynecol. 2010;17:167–175. 9. Liu H, Lu D, Wang L, Shi G, Song H, Clarke J. Robotic surgery for benign gynaecological disease. Cochrane Database Syst Rev. 2012;(2):1–26. 10. Pasic RP, Rizzo JA, Fang H, Ross S, Moore M, Gunnarsson C. Comparing robot-assisted with conventional laparoscopic hysterectomy: impact on cost and clinical outcomes. J Minim Invasive Gynecol. 2010;17:730–738. 11. Reza M, Maeso S, Blasco JA, Andradas E. Meta-analysis of observational studies on the safety and effectiveness of robotic gynaecological surgery. Br J Surg. 2010;97:1772–1783. 12. Sarlos D, Kots LA. Robotic versus laparoscopic hysterectomy: a review of recent comparative studies. Curr Opin Obstet Gynecol. 2011;4:283–288.
Journal of Minimally Invasive Gynecology, Vol 22, No 1, January 2015 13. Barbash GI, Glied SA. New technology and health care costs: the case of robot-assisted surgery. N Engl J Med. 2010;363:701–704. 14. Barnett JC, Judd JP, Wu JM, Scales CD Jr, Myers ER, Havrilesky LJ. Cost comparison among robotic, laparoscopic, and open hysterectomy for endometrial cancer. Obstet Gynecol. 2010;116:685–693. 15. Wright JD, Ananth CV, Lewin SN, et al. Robotically assisted vs laparoscopic hysterectomy among women with benign gynecologic disease. JAMA. 2013;309:689–698. 16. Jonsdottir GM, Jorgensen S, Cohen SL, et al. Increasing minimally invasive hysterectomy: effect on cost and complications. Obstet Gynecol. 2011;117:1142–1149. 17. O’Neill M, Moran PS, Teljeur C, et al. Robot-assisted hysterectomy compared to open and laparoscopic approaches: systematic review and meta-analysis. Arch Gynecol Obstet. 2013;287:907–918. 18. Paraiso MF, Ridgeway B, Park AJ, et al. A randomized trial comparing conventional and robotically assisted total laparoscopic hysterectomy. Am J Obstet Gynecol. 2013;208:368.e1–368.e7. 19. Sarlos D, Kots L, Stevanovic N, von Felten S, Sch€ar G. Robotic compared with conventional laparoscopic hysterectomy: a randomized controlled trial. Obstet Gynecol. 2012;120:604–611. 20. L€onnerfors C, Persson J. Implementation and applications of robotic surgery within gynecologic oncology and gynecology: analysis of the first thousand cases. Ceska Gynekol. 2013;78:12–19. 21. Persson J, Reynisson P, Borgfeldt C, Kannisto P, Lindahl B, Bossmar T. Robot assisted laparoscopic radical hysterectomy and pelvic lymphadenectomy with short and long term morbidity data. Gynecol Oncol. 2009; 113:185–190. 22. Persson J, Imboden S, Reynisson P, Andersson B, Borgfeldt C, Bossmar T. Reproducibility and accuracy of robot-assisted laparoscopic fertility sparing radical trachelectomy. Gynecol Oncol. 2012; 127:484–488. 23. Woelk JL, Casiano ER, Weaver AL, Gostout BS, Trabuco EC, Gebhart JB. The learning curve of robotic hysterectomy. Obstet Gynecol. 2013;121:87–95. 24. Lenihan JP, Kovanda C, Seshadri-Kreaden U. What is the learning curve for robotic assisted gynecologic surgery. J Minim Invasive Gynecol. 2008;15:589–594. 25. Bell MC, Torgerson JL, Kreaden U. The first 100 da Vinci hysterectomies: an analysis of the learning curve. S D Med. 2009;62:91. 93–95. 26. Garry R, Fountain J, Mason S, et al. The eVALuate study: two parallel randomised trials, one comparing laparoscopic with abdominal hysterectomy, the other comparing laparoscopic with vaginal hysterectomy. BMJ. 2004;328:129. 27. Tian YF, Lin YS, Lu CL, et al. Major complications of operative gynecologic laparoscopy in Southern Taiwan: a follow-up study. J Minim Invasive Gynecol. 2007;14:284–292. 28. Boggess JF, Gehrig PA, Cantrell L, et al. Perioperative outcomes of robotically assisted hysterectomy for benign cases with complex pathology. Obstet Gynecol. 2009;114:585–593. 29. Reynisson P, Persson J. Hospital costs for robot-assisted laparoscopic radical hysterectomy and pelvic lymphadenectomy. Gynecol Oncol. 2013;130:95–99. 30. Lau S, Vaknin Z, Ramana-Kumar AV, Halliday D, Franco EL, Gotlieb WH. Outcomes and cost comparison after introducing robotics program for endometrial cancer surgery. Obstet Gynecol. 2012;119: 717–724.
