Ovarian cancer causes more deaths than any other cancer of the female reproductive system.1 The only known effective method of preventing ovarian cancer is surgical removal of the ovaries, which may be performed electively in patients with no known increased risk for ovarian cancer or prophylactically in patients with increased genetic susceptibility for ovarian cancer. In patients at average risk for ovarian cancer, the potential benefit of cancer risk reduction with elective bilateral salpingo-oophorectomy must be balanced with the consequences of premature loss of estrogen production.2–5 Historically, elective bilateral salpingo-oophorectomy was considered appropriate in women 40 years or older.6–8 However, in 1999, the American College of Obstetricians and Gynecologists recommended that other factors in addition to age be considered in the decision to perform elective bilateral salpingo-oophorectomy9 and in 2008 revised its recommendation, suggesting that elective bilateral salpingo-oophorectomy should generally be avoided in premenopausal women.5 Additionally, in the last decade, several observational studies suggested that in the absence of estrogen replacement, the long-term adverse effects of elective bilateral salpingo-oophorectomy may outweigh the potential benefits for cancer risk reduction.10–15
In this context of uncertainty about the risks and benefits of elective bilateral salpingo-oophorectomy, little is known about the trends in U.S. elective bilateral salpingo-oophorectomy rates. Also, there is little information on complications with addition of elective bilateral salpingo-oophorectomy to hysterectomy.5 The objectives of this study were to examine recent U.S. trends in elective bilateral salpingo-oophorectomy by age group and surgical approach and to assess the association of in-hospital complications with elective bilateral salpingo-oophorectomy by surgical approach. We chose to examine trends during the periods from 1998 to 2001 and 2002 to 2006 to correspond with the release of the Women's Health Initiative study results in 2002,16 which is thought to have led to a widespread decrease in estrogen therapy use.17
MATERIALS AND METHODS
For this analysis, we used data from the Nationwide Inpatient Sample of the Healthcare Cost and Utilization Project. The Healthcare Cost and Utilization Project is a partnership between state data organizations and the national Agency for Healthcare Research and Quality. State partners contribute their statewide hospital discharge data to the Healthcare Cost and Utilization Project, which then translates and consolidates the data into uniformly formatted data sets. Using a stratified, probability design, the Nationwide Inpatient Sample is constructed to approximate a 20% sample of all U.S. community hospitals as defined by the American Hospital Association. The American Hospital Association defines community hospitals as all nonfederal short-term (average length of stay less than 30 days) general and specialty hospitals whose facilities are open to the public. The sampling frame consists of state-specific hospital discharge data provided to the Healthcare Cost and Utilization Project. Hospitals are selected based on five parameters for stratification: rural or urban location, bed size, geographic region, teaching status, and ownership. Within each stratum, a systematic random sample of hospitals equal in size to 20% of the universe for that stratum is drawn. The hospitals (clusters) are sorted by the first three digits of their zip code for the systematic sample. The Nationwide Inpatient Sample includes all discharges from the sampled hospitals and annually includes more than 900 hospitals and roughly seven million discharge records.
Details of the sampling strategy are described elsewhere.18 The number of states included and the hospitals sampled have differed each year. In 1998, the Nationwide Inpatient Sample included 22 states and covered 77% of the U.S. population; in 2006, it covered 37 states and 90% of U.S. population. The weighting process accounts for these differences and makes it possible for the data to be used to estimate the total number of inpatient hospital discharges in the United States.
The Nationwide Inpatient Sample data sets include patient information on age, race, health insurance status, county of residence, length of stay, and total charges; hospital characteristics (eg, ownership, size, teaching status) as well as admission and discharge status. Diagnoses and procedures are coded according to the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) criteria. Given that this study involved the use of deidentified data from a publicly available administrative data set, the Centers for Disease Control and Prevention determined that this project did not require human subject research review. Because there is no patient identifier in the Nationwide Inpatient Sample, the unit of analysis is typically considered to be the hospital discharge record because one patient can be admitted multiple times. However, because this analysis involves only hospitalizations when a hysterectomy is performed, we assume that each hospitalization represents one woman.
Our analysis included all discharge records for women aged 15 years or older with hysterectomy for benign gynecologic conditions. Hysterectomy was defined according to ICD-9-CM procedures 68.39, 68.4, 68.3, 68.49 (abdominal hysterectomy), 68.5, 68.59 (vaginal hysterectomy), and 68.31, 68.41, and 68.51 (total laparoscopic hysterectomy or laparoscopic assisted hysterectomy). Any type of hysterectomy with the presence of code 54.21 was coded as laparoscopic-assisted hysterectomy. To ensure that hysterectomy was performed only for benign gynecologic conditions, hospitalization records were excluded if they had an ICD-9-CM diagnosis code for cancer of the gynecologic or genitourinary tract (179–84, 188, 189, and 233.1–3), cancer of site unspecified (195, 199, 236, 239), ovarian neoplasm or cysts (220, 620.0–2), endometriosis (617.0–617.9), oophoritis, or salpingitis (614.0, 614.1, 614.2). Also excluded from our study population were patients with ICD-9-CM procedure codes for hysterectomy related to disorders of pregnancy (74.99, 74.0–74.4) or patients with ICD-9-CM diagnosis codes that indicate increased genetic susceptibility for ovarian cancer: V10.3, V10.00, V16.0, V16.3, V16.41, V50.42, V84.01, V84.02, 174.0–174.9, 153.0–153.9, and 233.0 (personal history of breast cancer, personal history of gastrointestinal tract cancer, family history of gastrointestinal tract cancer, family history of breast cancer, family history of ovarian cancer, genetic susceptibility to malignant neoplasm of breast, genetic susceptibility to malignant neoplasm of ovary, and malignant neoplasm of breast or colon). Women who underwent hysterectomy for benign gynecologic indications and who had no known risk factors for ovarian cancer were then classified into two groups based on whether or not they underwent concomitant bilateral salpingo-oophorectomy identified by the presence of ICD-9-CM procedure codes 65.51–65.54 or 65.61–65.64. Indications for hysterectomy were identified using the following ICD-9-CM diagnosis codes: leiomyomas (218), prolapse (618), abnormal uterine bleeding (626, 627.0, 627.1), endometrial hyperplasia (621.30–621.33), and cervical dysplasia (622.1). Each discharge record contains a maximum of 15 diagnoses codes; therefore, each hospitalization may be classified as having more than one surgical indication. Comorbidities included adhesive disease such as peritoneal adhesions, chronic cardiac, pulmonary, gastrointestinal and renal conditions as well as hypertension, diabetes, chronic anemia, and obesity.
Both surgical and medical complications were examined. Surgical complications included accidental operative laceration, gastrointestinal tract complications, including paralytic ileus, urinary tract complications, including injury to the ureters or bladder, wound complications, complications related to foreign bodies, hemorrhagic complications, including blood transfusion, infected seroma or abscess, and other complications (including other specified and unspecified complications of procedures not elsewhere classified). Medical complications included cardiac complications; pulmonary complications; vascular complications; central nervous system complications; renal–electrolyte complications, including acute renal failure, nonoperative site infections such as fever, septicemia, bacterial infection, urinary tract infection, and acute pyelonephritis; anesthesia complications; and other medical complications (including complications affecting other specified body systems not elsewhere classified).
Standard methods for analyzing weighted survey data were used using SUDAAN 9. Characteristics of women undergoing hysterectomy with and without elective bilateral salpingo-oophorectomy were compared. Comparisons were evaluated with the χ2 test for categorical variables and the two-sample t test for continuous variables. All calculated P values were two-sided; P<.05 was considered statistically significant. Programming and data results were confirmed by two independent researchers. Racial background was not available on 30% of the patients; race was therefore not included in our analyses.
Rates of elective bilateral salpingo-oophorectomy per 10,000 women aged 15 years and older were calculated using denominators from U.S. Census Bureau population estimates of female civilian residents. Tests for linear trends in overall and age-specific rates were performed using weighted least squares regression.19 The SUDAAN analytical procedure PROC Descript was used to test for linear trends in the distribution of surgical approach. Linear trends were assessed from 1998 to 2001 and from 2002 to 2006. Multivariable regression models were used to assess the association between elective bilateral salpingo-oophorectomy and 1) length of stay (linear); or 2) the presence of a complication (logistic) adjusting for age group, number of comorbidities, surgical indication, and hospital teaching status. Models were fit for all hysterectomies and stratified by each surgical approach.
From 1998 to 2006, 2,250,041 patients with no known increased risk for ovarian cancer underwent hysterectomy for benign gynecologic indications. A total of 873,770 (39%) of these patients had elective bilateral salpingo-oophorectomy. Among women who underwent elective bilateral salpingo-oophorectomy, 72% had abdominal hysterectomy; 15% and 13% had vaginal and laparoscopic hysterectomies, respectively. Compared with women undergoing hysterectomy only, women undergoing elective bilateral salpingo-oophorectomy were older (mean age, 50.4 compared with 44.1 years; P<.05) and were more likely to have two or more comorbid conditions (16% compared with 11%, P<.05) (Table 1). Patients who underwent elective bilateral salpingo-oophorectomy were more likely to have uterine leiomyomas as a diagnosis compared with patients who had hysterectomy only (63% compared with 55%, P<.05). The overall elective bilateral salpingo-oophorectomy rate during the 9-year study period was 8.3 per 10,000 women per year (Table 2). The elective bilateral salpingo-oophorectomy rate increased from 7.8 per 10,000 in 1998 to 9.0 per 10,000 in 2001 (P trend <.05) and then declined from 9.0 per 10,000 in 2002 to 7.4 per 10,000 in 2006 (P trend <.05). Similarly, the proportion of hysterectomies accompanied by elective bilateral salpingo-oophorectomy increased from 38% in 1998 to 41% in 2001 (P trend <.05) and then decreased from 40% in 2002 to 36% in 2006 (P trend <.05) (data not shown). From 2002 to 2006, the proportion of elective bilateral salpingo-oophorectomy performed laparoscopically increased from 11.6% to 17.7% (P<.05), whereas the proportion performed abdominally decreased from 73.1% to 69.3% (P<.05). The proportion of elective bilateral salpingo-oophorectomy performed vaginally did not change significantly over the time period (Table 2).
The rates of elective bilateral salpingo-oophorectomy varied with age (Fig. 1). The overall rate was highest among women aged 45 to 49 years (26.5 per 10,000) and those aged 50 to 54 years (20.3 per 10,000) and lowest among women aged 34 years or less (0.9 per 10,000) and those aged 35 to 39 years (5.1 per 10,000). From 1998 to 2001, elective bilateral salpingo-oophorectomy rates increased among women aged 35 to 39, 40 to 44, and 45 to 49 years (P for trend <.05). From 2002 to 2006, elective bilateral salpingo-oophorectomy rates decreased among all age groups (P trend <.05); the decline was greatest among those aged 45 to 49 years (20.3%). In 2006, 23% of women aged 40 to 44 years and 45% of women aged 45 to 49 years undergoing hysterectomy for benign gynecologic indications had concomitant elective bilateral salpingo-oophorectomy (data not shown).
After adjusting for age, diagnoses, comorbid conditions, and hospital teaching status, mean lengths of stay over the entire study period were 2.9 days for elective bilateral salpingo-oophorectomy and 2.5 days for hysterectomy only (P<.01) (Table 3). After stratifying by route of hysterectomy, elective bilateral salpingo-oophorectomy was associated with a significantly longer mean length of stay for laparoscopic and vaginal procedures but not abdominal procedures. Approximately 20% of women undergoing elective bilateral salpingo-oophorectomy had at least one in-hospital complication compared with 17% of women undergoing hysterectomy only (Table 3). After adjusting for potential confounders, women who had elective bilateral salpingo-oophorectomy were found to be at a slightly increased risk of developing at least one in-hospital complication compared with women undergoing hysterectomy only (odds ratio [OR], 1.07; 95% confidence interval [CI], 1.05–1.10). After stratifying by route of hysterectomy, elective bilateral salpingo-oophorectomy was associated with an increased risk of any complication only when the procedure was done vaginally (OR 1.12; 95% CI 1.08–1.17). There was a decreased risk of any complication associated with elective bilateral salpingo-oophorectomy when the procedure was done abdominally (OR 0.91; 95% CI 0.89–0.94) or laparoscopically (OR 0.89; 95% CI 0.83–0.94). Patients who had elective bilateral salpingo-oophorectomy done vaginally were at increased risk of developing a surgical complication compared with patients who had hysterectomy only (OR 1.16; 95% CI 1.12–1.21). Specifically, hemorrhagic complications (5.6% compared with 4.9%; P<.05), urinary tract complications (5.2% compared with 4.2%; P<.05), and gastrointestinal complications (3.0% compared with 2.4%; P<.05) were more common among those undergoing vaginal hysterectomy with elective bilateral salpingo-oophorectomy compared with those undergoing vaginal hysterectomy only.
We found that rates of elective bilateral salpingo-oophorectomy in the United States increased from 1998 to 2001 and decreased from 2002 to 2006. Between 2002 and 2006, the decline in elective bilateral salpingo-oophorectomy rates was greatest among women aged 45 to 49 years. These observations could be attributed to several factors. It is interesting to note that the elective bilateral salpingo-oophorectomy rate decreased significantly since 2002, the same year that the Women's Health Initiative early-termination results were published.16 Although the Women's Health Initiative did not address estrogen replacement therapy in younger women undergoing bilateral salpingo-oophorectomy, it may have resulted in gynecologists and patients becoming increasingly reluctant toward elective bilateral salpingo-oophorectomy as a result of concerns about the safety of estrogen therapy. Buist et al17 showed that 5 months after release of the Women's Health Initiative results, there was a 46% decline in hormone therapy prevalence and 28% decrease in estrogen therapy use. The decrease in the elective bilateral salpingo-oophorectomy rates also coincides with observational studies in the early 2000s linking elective bilateral salpingo-oophorectomy in premenopausal women with increased long-term morbidity and mortality, especially in patients not taking estrogen replacement therapy.10–15 Another factor that may also have influenced the downward trend in the elective bilateral salpingo-oophorectomy rate is the recent overall decrease in national hysterectomy rates in the United States.20 However, the fact that the proportion of hysterectomies accompanied by elective bilateral salpingo-oophorectomy has also decreased over the same time period suggests that the decrease in rates of elective bilateral salpingo-oophorectomy is not wholly the result of the decrease in hysterectomy rates.
We found an overall lower frequency (39%) of hysterectomies to be accompanied by elective bilateral salpingo-oophorectomy than has been described previously in the literature.21–23 We believe that this is the result of the relatively restrictive definition for elective bilateral salpingo-oophorectomy that we used in our study such as the exclusion of patients at increased genetic risk of ovarian cancer. Another observation in our study was that the introduction of laparoscopy into general gynecology seems to have influenced hysterectomy and elective bilateral salpingo-oophorectomy practice. In recent years, the proportion of elective bilateral salpingo-oophorectomy performed laparoscopically has increased markedly, whereas the proportions performed abdominally have decreased. The proportions performed vaginally have not changed significantly. These observations might indicate that the decision to or not to have elective bilateral salpingo-oophorectomy influences the surgical approach for the hysterectomy and that physicians are choosing laparoscopy compared with abdominal approach if they are going to perform elective bilateral salpingo-oophorectomy.
The data on the effect of concomitant bilateral salpingo-oophorectomy on complication rates at the time of hysterectomy for benign gynecologic conditions are not consistent.21,22,24 After stratifying by surgical approach, our study suggests that elective bilateral salpingo-oophorectomy increased the risk of in-hospital complications only when the procedure was done vaginally. For hysterectomies done laparoscopically and abdominally, adding elective bilateral salpingo-oophorectomy was associated with a decreased risk of in-hospital complications. One possible explanation for this apparent protective effect of elective bilateral salpingo-oophorectomy is that generally the cases that are anticipated to be difficult are the ones that are done abdominally or laparoscopically, and elective bilateral salpingo-oophorectomies are not done in such cases. Surgeons would be unlikely to take the time to perform an elective bilateral salpingo-oophorectomy for a particularly difficult case or one in which a complication had already occurred. It is also possible that the vaginal approach is less safe for performing bilateral salpingo-oophorectomy and another approach should be considered when patients desire elective bilateral salpingo-oophorectomy.
Our analysis is not without limitations, including those that are common when using large administrative data sets. The Nationwide Inpatient Sample data set does not provide all the patient characteristics that may influence the decision to undergo elective bilateral salpingo-oophorectomy (eg, parity, a history of osteoporosis, or education) or physician characteristics (eg, age, years of surgical experience, or subspecialty training) that may influence not only patient counseling on elective bilateral salpingo-oophorectomy, but also surgical complication rates. Another limitation is that inaccuracies in the diagnoses and procedures listed on the hospital discharge summary are possible, and we were unable to perform record review. Also by using the Nationwide Inpatient Sample data set, we relied solely on inpatient hospital records in an era when an increasing proportion of surgical procedures is being performed in the outpatient setting.25 Finally, we were unable to capture complications that occurred after discharge and were either managed as outpatients or readmitted.
Despite these limitations, this study has significant strengths. With a total of more than 2,200,000 patients over a study period of 9 years, this is a large nationally representative database that allows for the results to be generalized to the entire U.S. population. It also allows for examination of more narrowly defined age groups and determination of age-specific rates. An important strength of the study was the ability to quantify perioperative morbidity as it applies to women undergoing hysterectomy with and without bilateral salpingo-oophorectomy. This study will facilitate counseling and aid women and their providers in the choice between ovarian conservation and oophorectomy when hysterectomy is performed for benign gynecologic indications. Our study showed that the rate of elective bilateral salpingo-oophorectomy in the United States has been decreasing since 2002. This change may be partially attributable to emerging data, largely from observational studies, on long-term adverse effects of elective bilateral salpingo-oophorectomy. However, it is important to note that ovarian cancer causes more deaths than any other cancer of the reproductive system and the only effective means of preventing it, in average risk patients, is elective bilateral salpingo-oophorectomy BSO. Prospective studies that will provide valid data on the short- and long-term health consequences of elective bilateral salpingo-oophorectomy and the effect of estrogen therapy are urgently needed to guide clinical practice and patient counseling.
2. Kauff ND, Satagopan JM, Robson ME, Scheuer L, Hensley M, Hudis CA, et al. Risk-reducing salpingo-oophorectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2002;346:1609–15.
3. Averette HE, Nguyen HN. The role of prophylactic oophorectomy in cancer prevention. Gynecol Oncol 1994;55:S38–41.
4. Grann VR, Panageas KS, Whang W, Antman KH, Neugut AI. Decision analysis of prophylactic mastectomy and oophorectomy in BRCA1-positive or BRCA2-positive patients. J Clin Oncol 1998;16:979–85.
5. Elective and risk reducing salpingo-oophorectomy. ACOG Practice Bulletin No. 89. American College of Obstetricians and Gynecologists. Obstet Gynecol 2008;111:231–41.
6. American College of Obstetricians and Gynecologists. Prophylactic oophorectomy. ACOG Technical Bulletin No. 111. Washington, DC: ACOG; 1987.
7. Boike G, Averette HE, Hoskins W, Chmiel J, Zuber K, Karnell L, et al. National survey of ovarian carcinoma: 111. Women with prior hysterectomy: a failure of prevention? Gynecol Oncol 1993;49:112.
8. Sightler S, Boike G, Estape R, Averette H. Ovarian cancer in women with prior hysterectomy: a 14-year experience at the University of Miami. Obstet Gynecol 1991;78:681–4.
9. American College of Obstetricians and Gynecologists. Prophylactic oophorectomy. ACOG Practice Bulletin 7. American College of Obstetrics and Gynecologists. Washington, DC: ACOG; 1999.
10. Falkeborn M, Schairer C, Naessen T, Persson I. Risk of myocardial infarction after oophorectomy and hysterectomy. J Clin Epidemiol 2000;53:832–7.
11. Taylor M. Psychological consequences of surgical menopause. J Reprod Med 2001;46(suppl):317–24.
12. de Kleijn MJ, van der Schouw YT, Verbeek AL, Peeters PH, Banga JD, van der Graaf Y. Endogenous estrogen exposure and cardiovascular mortality risk in postmenopausal women. Am J Epidemiol 2002;155:339–45.
13. Melton LJ 3rd, Khosla S, Malkasian GD, Achenbach SJ, Oberg AL, Riggs BL. Fracture risk after bilateral oophorectomy in elderly women. J Bone Miner Res 2003;18:900–5.
14. Howard BV, Kuller L, Langer R, Manson JE, Allen C, Assaf A, et al. Risk of cardiovascular disease by hysterectomy status, with and without oophorectomy: the Women's Health Initiative Observational Study. Circulation 2005;111:1462–70.
15. Parker WH, Broder MS, Liu Z, Shoupe D, Farquhar C, Berek JS. Ovarian conservation at the time of hysterectomy for benign disease. Obstet Gynecol 2005;106:219–26.
16. Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. JAMA 2002;288:321–33.
17. Buist DS, Newton KM, Miglioretti DL, Beverly K, Connelly MT, Andrade S, et al. Hormone therapy prescribing patterns in the United States. Obstet Gynecol 2004;104:1042–50.
18. HCUP NIS. Nationwide Inpatient Sample (NIS), Healthcare Cost and Utilization Project (HCUP). Rockville (MD): Agency for Healthcare Research and Quality; 2005.
19. Gillum BS, Graves EJ, Jean L. Trends in hospital utilization: United States, 1988–92. Vital Heath Stat 13 1996:1–71.
20. Whiteman MK, Hillis SD, Jamieson DJ, Morrow B, Podgornik MN, Brett KM, et al. Inpatient hysterectomy surveillance in the United States, 2000–2004. Am J Obset Gynecol 2008;198:34.e1–7.
21. Lowder JL, Oliphant SS, Ghetti C, Burrows LJ, Meyn LA, Balk J. Prophylactic bilateral oophorectomy or removal of remaining ovary at the time of hysterectomy in the United States, 1979–2004. Am J Obstet Gynecol 2010;202:538.e1–9.
22. Gross CP, Nicholson W, Powe NR. Factors affecting prophylactic oophorectomy in postmenopausal women. Obstet Gynecol 1999;94:962–8.
23. Keshavarz H, Hillis S, Kieke B, Marchbanks P. Hysterectomy surveillance—United States, 1994–1999. MMWR CDC Surveill Summ 2002;51:1–8.
24. Elit L, Rosen B, Goel V, McLaughlin J, Fung MK, Shime J, Narod S. Prophylactic oophorectomy in Ontario. Fam Cancer 2001;1:143–8.
© 2010 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
25. Morrison JE Jr, Jacobs VR. Outpatient laparoscopic hysterectomy in a rural ambulatory surgery center. J Am Assoc Gynecol Laparosc 2004;11:359–64.