In 2014, the Society of Gynecologic Oncology and the American College of Surgeons Commission on Cancer jointly approved use of minimally invasive surgery as a quality measure for the primary treatment of stage I–III endometrial cancer, the most common gynecologic cancer in the United States.1–3 Eight randomized controlled trials and a Cochrane database review demonstrate the superiority of minimally invasive surgery in this setting.4–6 Compared with laparotomy, laparoscopic surgery for the treatment of early-stage endometrial cancer is associated with almost identical survival as well as less blood loss, fewer severe postoperative adverse events, and shorter hospital stays.4,6,7 Additionally, laparoscopy is as effective as laparotomy in detection of advanced uterine cancer during surgical staging.5
Despite the overwhelming evidence to support minimally invasive surgery as a quality measure in uterine cancer care, there remain significant disparities in surgical care for this disease. These inequalities are based on multiple factors, including patient race, comorbidities and socioeconomic status, hospital size, region and cancer volumes, and surgeon variables.8 In the United States, surgical approach in the treatment of early-stage endometrial cancer largely remains a matter of physician preference and physician surgical skill set.8 In a temporal trends study of 32,560 U.S. women with nonmetastatic endometrial cancer who underwent hysterectomy between 2007 and 2011, we previously described that the percentage of patients treated with minimally invasive surgery increased from 36% in 2007 to 51% in 2011. Although a significant increase was observed over this 5-year period, the overall adoption rate of this surgical approach remained low during the study period, with utilization lowest in hospitals with few endometrial cancer cases, where only 23% of patients underwent hysterectomy surgery using minimally invasive surgery.8 Additionally, perioperative complication rates were approximately 2.3 times higher in the open compared with the minimally invasive cohorts.
Using the most contemporary data available from the National Inpatient Sample database, our primary objective was to conduct an updated analysis of the nationwide uptake of minimally invasive surgery to treat endometrial cancer. Secondary objectives included assessing complication rates and associated costs for women treated with minimally invasive compared with open surgery, identifying patient and hospital factors associated with choice of surgical approach, and projecting national cost savings possible with increased use of a minimally invasive approach. We hypothesized that minimally invasive hysterectomy rates in 2012–2013 would increase compared with rates observed during 2007–2011 and that fewer complications and lower overall procedural and inpatient costs would be observed with the minimally invasive approach.
MATERIALS AND METHODS
After the Johns Hopkins Medicine institutional review board (Baltimore, Maryland) approved this retrospective cohort study as institutional review board-exempt, access to the U.S. Healthcare Cost and Utilization Project-National (Nationwide) Inpatient Sample database was obtained. The study was deemed exempt because the database used in the analysis is publicly available. The National Inpatient Sample is the largest publicly available all-payer inpatient health care database in the United States and provides national estimates of inpatient hospital stays. It is comprised of a 20% stratified sample of hospital discharge data from 44 U.S. states. With a self-weighting design, the National Inpatient Sample estimates more than 36 million hospitalizations annually and represents more than 95% of the U.S. population.9 The National Inpatient Sample captures all inpatient surgical procedures with at least an overnight stay in the hospital.
All patients who underwent hysterectomy with or without lymphadenectomy in 2012–2013 for the treatment of endometrial cancer were identified by International Classification of Diseases, 9th Revision (ICD-9) codes. Patients with nonprimary endometrial cancer or metastatic cancer, as determined by diagnostic codes, were excluded from the study. Metastatic disease was ascertained by the presence of concomitant diagnostic codes for metastases; the positive predictive value and overall accuracy of administrative data sets to capture metastatic disease are 75–95%.10
Minimally invasive surgery was defined as vaginal, laparoscopic, or robot-assisted hysterectomy and as any of the first three listed ICD-9 codes if a fourth digit with a value of “1” was present (ie, 68.31, 68.41, 68.61), vaginal hysterectomy (ICD-9 codes 68.5, 68.7), or any robotically assisted hysterectomy (any hysterectomy code plus ICD-9 code 174). Open hysterectomy was defined as subtotal abdominal hysterectomy (ICD-9 code 68.3), total abdominal hysterectomy (68.4), radical abdominal hysterectomy (68.6), and other or unspecified hysterectomy (68.9). Lymphadenectomy was defined using the following ICD-9 codes: 40.10, 40.20, 40.50, 40.52, and 40.5.
Demographic data were collected on patient factors including age at diagnosis, obesity status, Elixhauser comorbidity measure, race, and payer type. The Elixhauser comorbidity measure (Appendix 1, available online at http://links.lww.com/AOG/A838) was used to analyze how comorbidities affect clinical outcomes in the inpatient setting.11,12 The Elixhauser comorbidity measure is based on 30 acute and chronic conditions significantly associated with in-hospital mortality. The summary score—a weighted combination of the Elixhauser comorbidities—was categorized as zero (no comorbidity), one to two comorbidities, three to five comorbidities, and greater than five comorbidities.
Data were also collected on the following hospital factors: type, geographic region, teaching status, size, median patient household income, and annual hospital case volume of patients with endometrial cancer. High-volume hospitals were defined as having 30 or more endometrial cancer hysterectomy operations per year, medium volume as 10–29 operations, and low volume as less than 10 cases per year. Hospital size (by number of beds) was reported as small, medium, or large as defined by the Healthcare Utilization Project and combined teaching–rural status of the hospital.13 A government hospital was defined as a facility operated by the Veterans’ Administration.
The adverse events analyzed in this study included overall inpatient perioperative complications, intraoperative surgical complications, vascular injury (laceration), major blood loss (acute posthemorrhagic anemia), hospital stay greater than 2 days, required aftercare, surgical site infection, pneumonia, venothrombosis, sepsis, wound or fascial dehiscence, and death.
Descriptive analyses and the evaluation of factors associated with surgical approach were conducted on all patients with endometrial cancer in our study cohort. Demographic and hospital characteristics for patients who underwent open surgery compared with minimally invasive surgery were compared using χ2 tests (categorical variables) and independent t tests (continuous variables). Hierarchical logistic regression was used to identify patient- and hospital-level factors associated with receiving open surgery compared with minimally invasive surgery.
To assess the association between surgical approach and the risk of preoperative complications (Table 1), and to estimate costs associated with surgical approach (Table 2), we created a propensity-matched cohort to minimize the effects of observed confounding. Open and minimally invasive surgery cohorts were matched using one-to-one propensity score matching. Propensity score was estimated from the logistic regression that modeled the probability of undergoing open compared with minimally invasive surgery as a function of the baseline patient and hospital characteristics. Standardized difference was used to assess balance between the two surgical groups prematching and postmatching (Table 3). Hierarchical logistic regression modeling was used to estimate the association between surgical approach and the risk of perioperative complications. To account for the highly skewed nature of health care cost distribution, generalized linear modeling with log link and γ distribution was used to estimate the adjusted cost (http://www.ncbi.nlm.nih.gov/pubmed/11469231). Descriptive and regression analyses were performed using SAS 9.4.
To analyze cost, billed charges were multiplied by hospital cost-to-charge ratios.14 Cost data were adjusted for inflation using the Gross Domestic Product Implicit Price Deflator.15 Multiple linear regression and propensity score matching were used to calculate costs for open surgery and minimally invasive surgery±surgical complications (Table 2). To account for the highly skewed nature of distributions of health care costs, the generalized linear model incorporated a log link and γ distribution. We also estimated health care cost in a hypothetical scenario with 80% use of minimally invasive surgery, assuming the same weighted overall count and complication proportions for each group. This scenario implicitly assumes there is a causal link between minimally invasive surgery and reduced complications as well as a causal link between complications and increased costs.
After excluding patients with metastatic endometrial cancer and nonprimary cancers of the endometrium, we identified 9,799 patients in the database who underwent hysterectomy for the treatment of endometrial cancer in 2012–2013. In sum, 4,661 patients underwent minimally invasive surgery and 5,138 patients underwent open surgery (Table 3). During the study period, 47.6% of patients underwent minimally invasive surgery and 52.4% of patients underwent open surgery.
Table 4 details patient and hospital factors associated with choice of surgical approach with odds ratios and 95% confidence intervals (CIs) detailed in the table. Demographic characteristics most frequently associated with open hysterectomy were patients of black race, “other” race, or Hispanic ethnicity compared with white patients (adjusted odds ratio [OR] 1.46, 95% CI 1.30–1.65; adjusted OR 2.39, 95% CI 1.99–2.87; and adjusted OR 1.18, 95% CI 1.03–1.36). Conversely, patients with multiple comorbidities (Elixhauser score greater than 5) were less likely to receive open surgery (adjusted OR 0.50, 95% CI 0.40–0.62). Obese patients and patients undergoing lymphadenectomy were also less likely to receive open surgery (adjusted OR 0.89, 95% CI 0.83–0.95 and adjusted OR 0.78, 95% CI 0.73–0.84, respectively). Patient median household income and payer status were not associated with surgical approach with the exception that patients on Medicaid were less likely to receive open surgery than patients with private insurance (adjusted OR 0.81, 95% CI 0.70–0.94).
Patients at rural hospitals were much more likely to undergo open surgery than patients treated at urban teaching hospitals (adjusted OR 14.34, 95% CI 9.66–21.27). Patients at medium or high endometrial cancer volume hospitals were less likely to receive open surgery than patients at low-volume hospitals (adjusted OR 0.46, 95% CI 0.34–0.62 and adjusted OR 0.35, 95% CI 0.13–0.94). Notably, 43.4% of patients with endometrial cancer overall had surgery at low-volume hospitals (with fewer than 10 endometrial cancer cases performed annually), and 58.2% of patients underwent open surgery at these hospitals. Conversely, patients treated at large hospitals were less likely to receive open surgery (adjusted OR 0.74, 95% CI 0.57–0.96). Hospital size was correlated with hospital cancer volume (Spearman's ρ: 0.21; P<.000). Additionally, 87.8% of rural hospitals were also low endometrial cancer volume facilities (compared with 75.0% of urban nonteaching and 33.1% of urban teaching hospitals). Rural hospital was correlated with lower cancer volume (Spearman's ρ: 0.38; P<.000).
Furthermore, patients treated at government hospitals were more likely to receive open surgery than patients at private hospitals (adjusted OR 1.66, 95% CI 1.15–2.39); hospitals in the West had lower use of open surgery than hospitals in the northeast United States (adjusted OR 0.47, 95% CI 0.32–0.68).
The propensity-matched cohort included 7,388 patients (3,694 matched pairs) (Table 3). In total, 1,589 patients (21.5%) experienced a major perioperative complication during their respective hospitalizations (Table 3). After propensity score matching, open abdominal surgery was associated with a 2.80 times greater risk of complications than minimally invasive surgery (95% CI 2.48–3.17). Open surgery was correlated with higher rates of intraoperative surgical complications (adjusted OR 2.33, 95% CI 1.99–2.74) and major blood loss (adjusted OR 2.68, 95% CI 2.24–3.20). Open surgery was also associated with a 21-fold increased risk of hospitalization greater than 2 days (adjusted OR 21.01, 95% CI 18.45–23.92).
The propensity score matched analysis showed mean open surgery costs to be $1,243 greater than minimally invasive surgery ($14,712 compared with $13,469, 95% CI $875–1,610) (Table 2). There was no significant difference in cost between uncomplicated open and minimally invasive surgery cases ($12,269 compared with $12,301, 95% CI of difference −$358 to $294). Additionally, there were no statistical differences in mean costs between open cases with complications and minimally invasive cases with complications ($20,617 compared with $20,795, 95% CI of difference −$743 to $387). Assuming the same weighted overall count and complication proportion for each group, providing minimally invasive surgery for 80% of patients in this study would have averted 2,733 complications and saved $19 million in 2012–2013.
Despite level I evidence supporting the value of minimally invasive surgery for the treatment of early-stage endometrial cancer, in 2013, the rate of open abdominal hysterectomy in the United States remains high. After propensity score matching, our data demonstrate a 2.80 times greater risk of major perioperative complications and significantly higher health care costs associated with open compared with minimally invasive hysterectomy surgery in this setting. In this contemporary analysis of the U.S. National Inpatient Sample database, patients who underwent minimally invasive hysterectomy surgery for nonmetastatic endometrial cancer were significantly older, more likely to have undergone a lymphadenectomy procedure, and were just as likely to be obese compared with their counterparts who underwent open surgery, yet the minimally invasive cohort still experienced fewer perioperative complications, demonstrating that age, obesity, and extent of surgical staging are not relative contraindications to the less invasive approach. Additionally, as observed in our 2007–2011 National Inpatient Sample analysis,8 patients treated at high-volume endometrial cancer centers were more likely to undergo a minimally invasive hysterectomy than lower and medium-volume hospitals. Taken together, these data may be evidence that in high-volume centers, minimally invasive operations can be far more widely offered, irrespective of patient factors or surgical complexity.
The indisputably increased risk of perioperative complications associated with open hysterectomy takes on particular significance given the high rates of obesity in the endometrial cancer patient population. Currently, 80% of patients with endometrial cancer are obese and the incidence of endometrial cancer is expected to climb sharply as the obesity epidemic worsens.1,16 Consequently, this is a population that will especially benefit from less invasive procedures. The safety of minimally invasive surgery in obese women with endometrial cancer is demonstrated in several reports. Wysham et al17 found no differences in complication rates between obese and nonobese patients with endometrial cancer treated with robotic surgery. Likewise, a Korean study reported no significant differences in perioperative or long-term complications or overall survival between obese and nonobese patients treated with laparoscopic surgery.18
In the current study, women of black race, “other” race, or Hispanic ethnicity were more likely than white women to receive open surgery. Our findings are consistent with the results of previous studies and emphasize a disturbing trend in which women of minority race receive substandard cancer care. In our previous National Inpatient Sample study, we showed that black women were more than twice as likely as white women to undergo open surgery.8 Similarly, a National Surgical Quality Improvement Program study found that black patients with endometrial cancer were more likely to receive open surgery and more likely to experience postoperative complications than white patients.19 However, this study found no significant difference in complications between black and white patients who underwent laparoscopic surgery. Such data suggest that the poorer surgical and perioperative outcomes observed in black patients are more likely to be the result of an open surgical approach than biologic factors.
One of our most provocative study findings was the pronounced use of open surgery by rural and low endometrial cancer volume hospitals compared with medium- to high-volume cancer care facilities in urban areas. Patients treated at rural hospitals were 14.3 times more likely to undergo open abdominal hysterectomy than patients in urban settings. This is especially relevant because 43% of patients treated for endometrial cancer in 2012–2013 had surgery at low-volume hospitals, defined as fewer than 10 endometrial cancer cases performed annually. Our previous study reported that the rate of minimally invasive surgery for inpatients with low-stage endometrial cancer at low-volume hospitals was 24% between 2007 and 2011.8 The current study, showing a utilization rate of 41.8% in the same patient population, demonstrates progress, although use remains well below the national average of 52% at these low-volume facilities. This disparity may reflect a lack of gynecologic oncologist availability or a surgeon trained in minimally invasive surgery at these low-volume facilities. Although it may be unfeasible for all patients with endometrial cancer living in rural settings to receive care at larger, urban centers, primary care physicians and gynecologists should consider referring these patients to a gynecologic oncologist or high-volume gynecology surgeon, when possible.20 Nevertheless, even after referral to a specialist, some patients will choose to make the trip to an urban center whereas others will assume the disadvantages of an open surgical approach in exchange for being cared for closer to home. Further study is needed regarding patient, health care provider, and geographic barriers to receiving cancer care at higher volume, urban centers.
In the 2012–2013 National Inpatient Sample database, most patients (52.4%) treated for nonmetastatic endometrial cancer underwent open hysterectomies. This represents a plateauing from 2011, when 49% of patients with endometrial cancer underwent open surgery.8 Level I and II data, including several randomized controlled trials and a Cochrane database review supporting minimally invasive surgery as the superior surgical approach to treat early-stage endometrial cancer, has been publically available since 2009.4–7 Additionally, the Society of Gynecology Oncology and the American College of Surgeons Commission on Cancer have jointly approved minimally invasive surgery as a quality measure for the treatment of stage I-–III endometrial cancer.3 Although a steady decline in open hysterectomies was observed from 2007 to 2011 in our previous National Inpatient Sample study,8 the more contemporary data demonstrate that the uptake of minimally invasive surgery appears to have leveled off in the last 2 years for those patients who require at least one night of hospitalization or inpatient care. In that article, we defined a minimally invasive hysterectomy rate of 80% as the benchmark for women with early-stage endometrial cancer. Our modeling analysis demonstrated that, if 80% instead of 47.6% of women in the current analysis had undergone minimally invasive surgery, more than 2,700 complications potentially would have been averted and millions of dollars in health care costs saved.
Our findings also indicate that open hysterectomy surgery is more costly than less invasive approaches—largely because of the costs associated with surgical complications—and that more widespread performance of minimally invasive surgery could potentially result in substantial health care cost savings. Even in cases with no complications, there was no significant difference in cost between uncomplicated open and minimally invasive cases. Because 22.3% of all study patients experienced a serious complication, the total cost of complications resulting from open surgery is enormous and, therefore, suggests that minimally invasive surgery is an excellent value. Scalici et al2 project that using less invasive procedures for 90% of endometrial cancer cases nationwide would save the U.S. health care system $534 million annually.
A study limitation is that the National Inpatient Sample database does not record events occurring after hospital discharge, so the incidence and cost of surgical complications may be even higher than we calculated. Additionally, patients who undergo same-day discharge after hysterectomy are not represented in this data set. The National Inpatient Sample provides no surgeon information, so we were unable to confirm the effect of subspecialty surgeons compared with generalist gynecologists in this regard. However, our data demonstrate that patients experience the best surgical outcomes at large, urban, high-volume cancer centers, where they may be more likely to be operated on by a gynecologic oncologist or minimally invasive surgeon. Other limitations associated with an administrative claims database include time lag and incomplete coding of claims data and the possibility that the study missed patients with metastatic disease not acknowledged at the time of discharge. Nevertheless, the database represents the largest, most substantive U.S. surgical database available to study contemporary national trends in cancer outcomes. Additionally, our propensity-matched cohorts provide a robust comparison of outcomes for patients treated with either open or minimally invasive approaches.
In conclusion, more than half of all U.S. women diagnosed with early-stage endometrial cancer and treated as inpatients still undergo open abdominal hysterectomy surgery. This approach is consistently associated with a higher rate of potentially preventable patient harm and health care costs, confirmed in our propensity-matched analysis in a large U.S. patient sample. The observed trends suggest a need to increase minimally invasive training opportunities for obstetrics and gynecology residents, fellows, and practicing gynecologic surgeons. Given the improved surgical outcomes noted at large centers with high cancer volumes, referral of patients with endometrial cancer to these facilities should be encouraged by the American College of Obstetricians and Gynecologists, by the Centers for Medicaid and Medicare Services, and other regulatory agencies. Additionally, it may be warranted to require hospitals to publish their minimally invasive surgery rates and to offer government incentives to hospitals with rates above the national average. Such changes may help to reduce the enormous burden of surgical complications, preventable patient harm, and unnecessary health care spending while standardizing the quality of endometrial cancer care on a national level.
1. Koh W, Greer BE, Abu-Rustum NR, Apte SM, Campos SM, Chan J, et al. Uterine neoplasms, version 1.2014. J Natl Compr Canc Netw 2014;12:248–80.
2. Scalici J, Laughlin BB, Finan MA, Wang B, Rocconi RP. The trend towards minimally invasive surgery (MIS) for endometrial cancer: an ACS-NSQIP evaluation of surgical outcomes. Gynecol Oncol 2015;136:512–5.
3. Society of Gynecologic Oncology. Available at: https://www.sgo.org/newsroom/sgo-issues/sgo-issues-may-15-2014/
. Retrieved December 28, 2015.
4. Galaal K, Bryant A, Fisher AD, Al-Khaduri M, Kew F, Lopes AD. Laparoscopy versus laparotomy for the management of early stage endometrial cancer. The Cochrane Database of Systematic Reviews 2012, Issue 9. Art. No.: CD006655. doi: 10.1002/14651858.CD006655.pub2.
5. Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Schlaerth JB, Mannel RS, et al. Laparoscopy compared with laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group Study LAP2. J Clin Oncol 2009;27:5331–6.
6. Walker JL, Piedmonte MR, Spirtos NM, Eisenkop SM, Schlaerth JB, Mannel RS, et al. Recurrence and survival after random assignment to laparoscopy versus laparotomy for comprehensive surgical staging of uterine cancer: Gynecologic Oncology Group LAP2 Study. J Clin Oncol 2012;30:695–700.
7. Palomba S, Falbo A, Mocciaro R, Russo T, Zullo F. Laparoscopic treatment for endometrial cancer: a meta-analysis of randomized controlled trials (RCTs). Gynecol Oncol 2009;112:415–21.
8. Fader AN, Weise RM, Sinno AK, Tanner EJ 3rd, Borah BJ, Moriarty JP, et al. Utilization of Minimally Invasive Surgery in Endometrial Cancer Care: A Quality and Cost Disparity. Obstet Gynecol 2016;127:91–100.
9. Healthcare Cost and Utilization Project (HCUP). Overview of the National (Nationwide) Inpatient Sample (NIS) 2015. Available at: https://www.hcup-us.ahrq.gov/nisoverview.jsp
. Retrieved December 28, 2015.
10. Whyte JL, Engel-Nitz NM, Teitelbaum A, Gomez Rey G, Kallich JD. An evaluation of algorithms for identifying metastatic breast, lung, or colorectal cancer in administrative claims data. Med Care 2015;53:e49–57.
11. Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care 1998;36:8–27.
12. Agency for Healthcare Research and Quality. Overview of the Agency for Healthcare Research and Quality patient safety indicators. 2013. Available at: www.qualityindicators.ahrq.gov/Modules/psi_resources.aspx
. Retrieved April 15, 2016.
13. Healthcare Cost and Utilization Project (HCUP). Bedsize of hospital. Available at: http://www.hcup-us.ahrq.gov/db/vars/hosp_bedsize/nisnote.jsp
. Retrieved December 28, 2015.
14. Cost-to-charge ratio files: 2011 Nationwide Inpatient Sample (NIS) user guide. Rockville (MD): Agency for Healthcare Research and Quality. Available at: https://www.hcup-us.ahrq.gov/db/state/CCR2011NISUserGuide.pdf
. Retrieved November 10, 2015.
15. Medical Expenditure Panel Survey: using appropriate price indices for analyses of health care expenditures or income across multiple years. Rockville (MD): Agency for Healthcare Research and Quality; 2013. Available at: http://meps.ahrq.gov/mepsweb/about_meps/Price_Index.shtml
. Retrieved November 10, 2015.
16. Fader AN, Arriba LN, Frasure HE, von Gruenigen VE. Endometrial cancer and obesity: epidemiology, biomarkers, prevention, and survivorship. Gynecol Oncol 2009;114:121–7.
17. Wysham WZ, Kim KH, Roberts JM, Sullivan SA, Campbell SB, Roque DR, et al. Obesity and perioperative pulmonary complications in robotic gynecologic surgery. Am J Obstet Gynecol 2015;213:33.e1–7.
18. Baek MH, Lee SW, Park JY, Kim D, Kim JH, Kim YM, et al. Feasibility and safety of laparoscopic surgery for obese Korean women with endometrial cancer: long-term results at a single institution. J Korean Med Sci 2014;29:1536–43.
19. Mahdi H, Lockhart D, Moslemi-Kebria M, Rose PG. Racial disparity in the 30-day morbidity and mortality after surgery for endometrial cancer. Gynecol Oncol 2014;134:510–5.
20. Melamed A, Katz Eriksen JL, Hinchcliff EM, Worley MJ Jr, Berkowitz RS, Horowitz NS, et al. Same-day discharge after laparoscopic hysterectomy for endometrial cancer. Ann Surg Oncol 2016;23:178–85.