Objective
The purpose of this study was to describe national rates and trends of prophylactic bilateral oophorectomy or remaining oophorectomy (BO/RO) at hysterectomy in women without specific gynecologic disease.
Study Design
Data from the National Hospital Discharge Survey were analyzed for 1979-2004. Hysterectomies were divided into 2 groups: (1) hysterectomy with BO/RO and (2) hysterectomy alone (≥1 ovary remaining). Age-adjusted rates (AARs) were calculated with 2000 US census data.
Results
Approximately 3,686,000 hysterectomies with BO/RO were performed from 1979-2004. AARs of hysterectomy with BO/RO decreased during this period; the AARs in women ≥50 years old increased. The number of hysterectomies alone was 5,461,100, and AARs of hysterectomy alone decreased significantly from 2.9 per 1000 women in from 1979-1981 to 1.1 per 1000 women in 2001 ( P < .001). The proportion of women who underwent hysterectomy with BO/RO increased from 29% in 1979 to 45% in 2004.
Conclusion
Although AARs of prophylactic BO/RO decreased from 1979-2004, the actual proportion of BO/RO at hysterectomy increased.
Approximately 600,000 hysterectomies are performed annually in the United States, One-half of which involve salpingo-oophorectomy. Elective salpingo-oophorectomy is the removal of normal ovary(ies) at the time of another indicated procedure. Historically, bilateral salpingo-oophorectomy was performed frequently at time of hysterectomy to decrease future risk of ovarian cancer and reoperation for ovarian disease. It is estimated that a woman’s lifetime risk of ovarian cancer is 1 in 70. The American Cancer Society estimated that 21,650 new cases of ovarian cancer will be diagnosed in 2008 in the United States, with an estimated 15,520 deaths resulting from ovarian cancer. The utility of performing elective or prophylactic oophorectomy in women who undergo hysterectomy for benign indications is controversial. Many physicians believe that prophylactic bilateral oophorectomy or removal of remaining ovary is an effective means to prevent ovarian cancer. Of women diagnosed with ovarian cancer, 4.5-18.2% have undergone previous hysterectomy. Sightler et al estimated that approximately 1000 new cases of ovarian cancer could be prevented in the United States each year if bilateral oophorectomy were performed at time of hysterectomy for women ≥40 years old.
Others believe that prophylactic oophorectomy unnecessarily deprives women of years of beneficial ovarian function. Previous studies suggest that the rate of ovarian cancer after hysterectomy alone is low, with 2 women per 1000 experiencing a subsequent ovarian cancer after hysterectomy alone. Parrazini et al found that hysterectomy alone decreased subsequent development of ovarian cancer by 30%. Rocca et al found that women ≤45 years old who underwent bilateral oophorectomy at time of benign hysterectomy had a 2-fold increased risk of death during a 30-year follow-up period, when compared with women who were age-matched and who had not undergone oophorectomy. Recently, Parker et al, who reported long-term results from the Nurses’ Health study, showed that, although bilateral oophorectomy is associated with a decreased risk of breast and ovarian cancer compared with ovarian conservation, it is also associated with an increased risk of all-cause death, fatal and nonfatal coronary heart disease, and lung cancer.
The purpose of this study was to describe national rates and trends of prophylactic bilateral oophorectomy or remaining oophorectomy at the time of hysterectomy in women without ovarian or specific gynecologic disease with the use of data from the National Hospital Discharge Survey (NHDS) database, 1979–2004.
Materials and Methods
Data were abstracted from the NHDS, which is a computerized database that is maintained by the federal government that samples hospital inpatient discharge data from across the country. Medical records from 466 non-federal short-stay hospitals (8% of all hospitals) were selected by systematic random sample, and approximately 270,000 discharges were collected per year from January 1979 to December 2004. The survey recorded up to 7 discharge diagnoses and 4 procedures and patient information that included sex, age, race, marital status, length of hospital stay, hospital size (number of beds), hospital ownership, and insurance type or expected source of payment. Diagnoses and procedures were listed according to the International Classification of Diseases, 9th review, Clinical Modification (ICD-9-CM). Quality control programs have estimated the error rate at 4.3% for medical coding and data entry and 1.4% for demographic coding and data entry.
Women who underwent supracervical, total abdominal, laparoscopic-assisted, and vaginal hysterectomy were identified with the ICD-9-CM codes 68.3, 68.4, 68.51, and 68.59. The diagnosis codes for hysterectomy included menorrhagia/metrorrhagia/menometrorrhagia (ICD-9-CM code 626.2), irregular menstrual bleeding (ICD-9-CM code 626.4), dysfunctional uterine bleeding (ICD-9-CM code 626.8), uterine myoma (ICD-9-CM codes 218.0, 218.1, 218.2, 218.9), postmenopausal bleeding (ICD-9-CM code 627.1), genital pain (ICD-9-CM codes 625.0-9), dysplasia of cervix (ICD-9-CM code 622.1), genital prolapse (ICD-9-CM codes 618.0, 618.1, 618.2, 618.3, 618.4, 618.5, 618.6, 618.9), cervicitis/endocervicitis (ICD-9-CM code 616.0), and disorders of the uterus not elsewhere classified (ICD-9-CM codes 621.0-9).
Women who underwent hysterectomy with concomitant bilateral oophorectomy or remaining oophorectomy, which was defined as removal of both ovaries or removal of the remaining ovary, were identified by the following procedures: removal of both ovaries at the same operative episode (ICD-9-CM codes 65.51 and 65.63), removal of remaining ovary (ICD-9-CM codes 65.52 and 65.54), or removal of remaining ovary and tube (ICD-9-CM codes 65.62 and 65.64). After we excluded women with diagnoses of ovarian disease (benign and malignant) and gynecologic diagnoses that warrant ovarian removal (eg, endometriosis, upper genital tract cancer), 2 groups of women were created: group 1, women who underwent hysterectomy with bilateral oophorectomy or remaining oophorectomy; and group 2, women who underwent hysterectomy alone (with at least 1 ovary remaining).
The numbers of tabulated surgical procedures were inflated to national averages according to designated hospital weights. Age-adjusted rates (AARs) of procedures were calculated by the direct method of rate adjustment, with the use of the 2000 projected United States population as the standard population. Rates were expressed as the number of women who underwent procedures per 1000 women in the population. Numbers were rounded to the nearest hundred because this database reflects an estimated number of procedures. When the estimated number of cases per year was based on <60 records in the database, the estimate was considered unreliable.
Postoperative complications during the index hospitalization were tabulated. Complications were categorized broadly into cardiovascular, respiratory, urinary tract, circulatory/bleeding, any organ injury, and postoperative infection. The number of deaths and patients who received a blood transfusion were also assessed.
Analyses were performed with Stata statistical software (release 9.0; StataCorp, College Station, TX). Weighted methods were used for statistical testing and estimates to account for the sample design of the survey. All statistical tests were 2-sided and evaluated at the .05 significance level. Spearman’s correlation coefficient was used to compare AARs. Univariable associations were analyzed with the Student t test for continuous data and with the χ 2 test for categoric data. Multivariable logistic regression was used to identify independent determinants of hysterectomy with bilateral oophorectomy or remaining oophorectomy. Variables with probability values < .1 were considered for inclusion in the models. Models were developed with forward stepwise regression based on the likelihood ratio statistic. Variables were retained in the model if the Wald χ 2 test statistic had a probability value of ≤ .05. Separate logistic regression models that controlled for age and surgical route were developed to evaluate the association between hysterectomy with bilateral oophorectomy or remaining oophorectomy and postoperative complications. In addition, separate logistic regression models that controlled for age, race, and surgical route were used to evaluate the association between common diagnoses and hysterectomy with bilateral oophorectomy or remaining oophorectomy.
Results
Approximately 3,686,000 women underwent hysterectomy with bilateral oophorectomy or remaining oophorectomy from 1979-2004; the numbers ranged from a low of 111,500 in 1989 to a high of 177,100 in 2002. The demographics of this population are listed in Table 1 . Eighty-four percent of hysterectomies with bilateral oophorectomy or remaining oophorectomy were performed abdominally, and 16% were performed vaginally or as a laparoscopically assisted vaginal hysterectomy. The overall AAR of hysterectomy with bilateral oophorectomy or remaining oophorectomy varied during the study period of 1979-2004 (range, 1.0–1.5 procedures per 1000 women; P < .001; Figure 1 ). For women <50 years old, the absolute number of hysterectomies with bilateral oophorectomy or remaining oophorectomy increased from 86,700 in 1979 to 111,200 in 2001 and then trended downward to 85,600 in 2004. However, as seen in Figure 2 , the AAR decreased from a high of 1.6 in 1980 to a low of 0.8 per 1000 women in 2004 ( P < .001). The proportion of women <50 years old who underwent hysterectomy with bilateral oophorectomy or remaining oophorectomy increased from 26% in 1979 to 37% in 2004. For women ≥50 years old, the absolute number of hysterectomies with bilateral oophorectomy or remaining oophorectomy increased from 36,100 in 1979 to a high of 69,100 in 2000, after which it trended downward to 60,300 in 2004. The AARs for this group varied from 1.1-2.1 per 1000 women during this time period ( P = .02; Figure 2 ). The overall proportion of women ≥50 years old who underwent hysterectomy with bilateral oophorectomy or remaining oophorectomy increased from 44% in 1979 to a high of 70% in 2000 and remained stable at 64% in 2004. Linear trends of AARs from 1979-2004 for hysterectomy with bilateral oophorectomy or remaining oophorectomy in women <50 years old and women ≥50 years old are shown in Figure 2 , with hysterectomy alone trend shown for reference.
| Characteristic | Hysterectomy only, n | Percentage | Bilateral oophorectomy, n | Percentage | P value |
|---|---|---|---|---|---|
| Race | .009 | ||||
| White | 3,715,700 | 68 | 2,4991700 | 68 | |
| Black | 756,000 | 14 | 476,600 | 13 | |
| Other | 989,400 | 18 | 710,300 | 19 | |
| Marital status | < .001 | ||||
| Single | 491,300 | 9 | 306,500 | 8 | |
| Married | 2,897,900 | 53 | 1,852,600 | 50 | |
| Other | 2,071,900 | 38 | 1,527,000 | 41 | |
| Region | .588 | ||||
| Northeast | 811,900 | 15 | 562,200 | 15 | |
| Midwest | 1,304,500 | 24 | 858,600 | 23 | |
| South | 2,313,200 | 42 | 1,560,000 | 42 | |
| West | 1,031,500 | 19 | 705,200 | 19 | |
| Insurance | < .001 | ||||
| Private | 3,924,000 | 72 | 2,724,100 | 74 | |
| Public assistance | 1,127,100 | 21 | 646,600 | 18 | |
| Other | 410,000 | 8 | 315,300 | 9 | |
| Hospital size: no. of beds | < .001 | ||||
| 6-99 | 887,100 | 16 | 660,100 | 18 | |
| 100-199 | 1,344,700 | 25 | 907,400 | 25 | |
| 200-299 | 1,019,100 | 19 | 693,400 | 19 | |
| 300-499 | 1,313,700 | 24 | 891,400 | 24 | |
| ≥500 | 896,500 | 16 | 533,800 | 14 | |
| Hospital ownership | < .001 | ||||
| Proprietary | 596,900 | 11 | 473,100 | 13 | |
| Government | 853,600 | 16 | 541,100 | 15 | |
| Not for profit | 4,010,500 | 73 | 2,671,800 | 72 | |
| Age, y | < .001 | ||||
| <25 | 147,900 | 3 | 28,400 | 1 | |
| 25-29 | 543,500 | 10 | 113,300 | 3 | |
| 30-34 | 948,800 | 17 | 200,200 | 5 | |
| 35-39 | 1,258,900 | 23 | 353,300 | 10 | |
| 40-44 | 1,046,000 | 19 | 757,300 | 21 | |
| 45-49 | 508,600 | 9 | 1,009,100 | 27 | |
| 50-54 | 206,500 | 4 | 536,100 | 15 | |
| 55-59 | 142,400 | 3 | 217,500 | 6 | |
| 60-64 | 146,200 | 3 | 172,100 | 5 | |
| 65-69 | 167,900 | 3 | 130,000 | 4 | |
| 70-74 | 169,300 | 3 | 80,400 | 2 | |
| 75-79 | 101000 | 2 | 57,200 | 2 | |
| 80-84 | 52,900 | 1 | 21,400 | 1 | |
| ≥85 | 21,300 | <1 | 9,700 | <1 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
