Objective
We sought to determine the differential rates and complications of hysterectomy type in the year prior to and following the introduction of robotic technology.
Study Design
This was a retrospective chart review of 461 hysterectomies performed from July 2007 through June 2008 (period 1) and July 2008 through June 2009 (period 2) at Virginia Commonwealth University Medical Center.
Results
In all, 199 vs 262 hysterectomies were performed in periods 1 and 2: open, 52.3% vs 43.1%; laparoscopic, 18.1% vs 8.0%; robotic, 2.5% vs 24.8%; and vaginal, 27.4% vs 24.1%, respectively. The increase in robotic hysterectomies in period 2 was associated only with a decline in laparoscopic hysterectomy ( P < .0001). Major morbidity by route was 23.04% open, 11/1% vaginal, 7.02% laparoscopic, and 4.29% robotic ( P < .0001).
Conclusion
Route of hysterectomy changed significantly after the introduction of robotic technology primarily due to a change in management of pelvic organ prolapse. Open hysterectomy was associated with significantly higher complication rates.
Gynecologists perform approximately 600,000 hysterectomies in the United States each year making it one of the most common major nonobstetrical surgeries in women. Optimizing outcomes and minimizing complications of this procedure, therefore, could have a profound impact in women’s health.
Despite numerous studies that demonstrate lower morbidity associated with vaginal and laparoscopic hysterectomies, the majority is still performed via laparotomy. A review of all hysterectomies performed in the United States in 2003 found that 66% were approached via laparotomy. A recent retrospective analysis of hysterectomies performed in Illinois between 2000 and 2005 reported that 82% of all cases in teaching hospitals vs 77% in nonteaching hospitals were performed abdominally. Laparoscopic hysterectomy, even after adjusting for confounders, was associated with fewer complications than either the vaginal or open group. In addition to a reduction in surgical morbidity, studies have clearly shown that vaginal and laparoscopic surgery allows faster recovery time, reduced inpatient hospitalization, improved cosmesis, decreased blood loss, and less postoperative pain.
Success with nonabdominal hysterectomy is highly dependent upon the experience and skill of the surgeon and the persistently high rates of abdominal hysterectomy may be related to surgical exposure during residency. Current statistics show that, of the average 120 hysterectomies that residents complete during their training, as few as 12 are performed vaginally. Exposure to advanced laparoscopic procedures such as total laparoscopic hysterectomy may be even less and given the steep learning curve for traditional laparoscopy, many surgeons may be hesitant to choose this approach in patients with challenging pathology such as large uteri, pelvic adhesive disease, or a history of pelvic surgeries.
In 2005, the da Vinci Surgical system (Intuitive Surgical, Sunnyvale, CA) was approved for use in gynecologic surgery by the US Food and Drug Administration. This technology provides the advantages of a 3-dimensional image of the operative field and instruments with a wristlike range of motion that are not counterintuitive, thereby potentially shortening the laparoscopic learning curve and facilitating the completion of more challenging cases through a minimally invasive approach.
The purpose of this study was to assess the impact on route of hysterectomy for the year prior to and following the introduction of the da Vinci surgical system in an academic teaching hospital and to determine which gynecologic diagnoses were associated with choice of surgical approach. Surgical morbidities and lengths of hospital stay for the different routes of hysterectomy were compared.
Materials and Methods
The da Vinci surgical system was available for use at Virginia Commonwealth University (VCU) Medical Center in April 2008. One benign gynecologist and 1 urogynecologist had completed training by May 2008 and each performed 4 proctored cases in the ensuing 6 weeks. Two additional faculty members, a gynecologist and a second urogynecologist, completed their robotic training by November 2008 and had full access to the system at that time. The remainder of the benign gynecology division, comprising 6 members, did not complete robotic training and continued to follow their traditional surgical paradigm. We chose to analyze cases from July 2007 through June 2008 (period 1), and July 2008 through June 2009 (period 2) because June 2008 represented the first month that cases were successfully completed independently by members of the gynecology division at VCU. Access to the robotic system was readily available after it was purchased, which permitted scheduling all desired robotic cases as such.
The VCU Institutional Review Board approved this study. We performed a retrospective chart and surgical case log review of all hysterectomies performed for benign indications during the 2 specified time periods. Cases defined as “open” included total abdominal and supracervical hysterectomy; “vaginal” included total vaginal hysterectomy with or without additional reconstructive procedures; “laparoscopic” included laparoscopic-assisted vaginal hysterectomy, laparoscopic supracervical hysterectomy, and total laparoscopic hysterectomy; and “robotic” included robotic total laparoscopic hysterectomy and robotic supracervical hysterectomy, with or without additional reconstructive procedures. Surgical case logs, maintained by the division of gynecology, were reviewed and the following data were extracted and recorded on de-identified spreadsheets: age, race, insurance status, indication for procedure, route of hysterectomy, estimated blood loss, and intraoperative complications such as genitourinary tract injury, bowel injury, and transfusion. Surgical pathology reports for each case were reviewed to determine the uterine weight in grams and final pathological diagnosis. Electronic medical records were then reviewed to determine the overall length of stay as well as early postoperative complications and any postoperative readmissions to the hospital. Complications including urinary tract infection, pneumonia, wound infection/cuff cellulitis, febrile morbidity, need for blood transfusion, delayed return of bowel function, deep vein thrombosis, and need for reoperation were recorded. Data analyses were conducted using software (SAS 9.1.3; SAS Institute Inc, Cary, NC). Rates of complications and routes of hysterectomy by time period were compared using Fisher’s exact test and χ 2 statistic, respectively. Analysis of variance was used to test the hypothesis that the mean length of stay was the same for all hysterectomy types. All analyses were conducted by intention to treat. A secondary analysis was performed after excluding cases of pelvic organ prolapse because in these cases, the hysterectomy was performed as part of a reconstructive procedure as opposed to being the primary surgical indication. A P value < .05 was considered statistically significant.
Results
A total of 461 hysterectomies were performed throughout the study period, 199 in the year prior to, and 262 in the year following the introduction of the robotic system. Demographic data, by route of hysterectomy for time periods 1 and 2, are presented in Table 1 . The respective rates of route of hysterectomy for each time period were as follows: open, 52.3% vs 43.1%; laparoscopic, 18.1% vs 8.0%; robotic, 2.5% vs 24.8%; and vaginal, 27.4% vs 24.1%, respectively. An overall comparison of route of hysterectomy across the 2 time periods revealed a significant difference ( P < .001). The rise in robotic hysterectomies in period 2 was associated with a decline in laparoscopic ( P < .0001) but not abdominal or vaginal hysterectomy. There was, however, a trend toward a significant decline in open vs nonlaparotomy cases from periods 1-2 ( P = .052) ( Table 2 ). When excluding the hysterectomy cases that were performed for the primary indication of pelvic organ prolapse, the rates were: open, 59.8% vs 51.1%; laparoscopic, 19.5% vs 9.1%; robotic, 2.4% vs 19.9%; and vaginal, 18.3% vs 19.9%, respectively ( P < .0001).
