Teplin, Vanessa MD1,2; Vittinghoff, Eric PhD3; Lin, Feng MS3; Learman, Lee A. MD, PhD2,3; Richter, Holly E. PhD, MD4; Kuppermann, Miriam PhD, MPH2,3,5
Most of the 600,000 hysterectomies performed annually in the United States are among premenopausal women with benign gynecologic diseases.1 These elective procedures aim to improve health-related quality of life by eliminating abnormal bleeding, chronic pelvic pain, or symptomatic leiomyomata.2 Oophorectomy is offered to women at the time of hysterectomy because it has been shown to decrease the risk of ovarian cancer and significantly improve premenstrual disorder, symptomatic endometriosis, and chronic pelvic pain.3–9 Approximately half of hysterectomies include elective oophorectomy,1 although little is known about how oophorectomy may affect the anticipated improvements in health-related quality of life after hysterectomy. Although a recent decision analysis addressed the risks and benefits of elective oophorectomy, health-related quality-of-life outcomes were not included in the analysis because of “insufficient data.”10
Oophorectomy in premenopausal women results in an acute decline in ovarian sex steroids.11 These hormones have been associated with maintenance of normal function in several health-related quality-of-life domains, including mental health, sexual functioning, and general well-being.12–14 Two cross-sectional studies have shown a decline in sexual functioning among women who underwent oophorectomy.15,16 However, a large cohort study found no negative effects of oophorectomy on sexual or psychological well-being.17 This investigation was limited because there was only one postsurgical evaluation and there were high rates of hormone use in the oophorectomy group, which may bias comparisons to women with ovarian conservation. Another recent cohort study reported improvements in depression scores and no decline in sexual activity after hysterectomy with a bilateral salpingo-oophorectomy (BSO), but this analysis did not include a direct comparison to women with ovarian conservation.18 A better understanding of how oophorectomy may affect health-related quality of life is critical to providing fully informed preoperative counseling to premenopausal women undergoing hysterectomy.
We investigated the relationship between concurrent oophorectomy at the time of hysterectomy and subsequent health-related quality-of-life outcomes among premenopausal women who participated in the Medicine or Surgery randomized trial or the Total or Supracervical Hysterectomy (TOSH) randomized trial.19,20 These trials demonstrated substantial improvements in health-related quality-of-life outcomes among symptomatic women after hysterectomy. The goal of this secondary analysis was to estimate whether women who underwent oophorectomy had less improvement in mental and physical health or sexual functioning compared with women who underwent hysterectomy with ovarian conservation.
MATERIALS AND METHODS
This is a secondary data analysis of premenopausal women who underwent hysterectomy between 1997 and 2000 as participants in either the Medicine or Surgery or TOSH randomized trials; details of the study designs have been previously reported.19–21 In brief, these trials recruited women simultaneously within four clinical sites using closely related inclusion and exclusion criteria. All participants were English-speaking premenopausal women, between the ages of 30 and 50 years, with abnormal uterine bleeding or symptomatic uterine leiomyomata (bleeding, pressure, or pain). Women aged 45 years or greater were included only if their follicle-stimulating hormone level was less than 30 milli-International Units/mL and they had a recent endometrial biopsy that was negative for carcinoma or hyperplasia. The trials were approved by the institutional review boards of the four clinical centers where patients were recruited and the coordinating center at the University of California, San Francisco.
For this analysis, we selected participants from the Medicine or Surgery or TOSH trials who underwent hysterectomy with or without oophorectomy and completed at least one interview after surgery. Participants were included in the BSO group if one or both ovaries were removed at the time of hysterectomy, leaving no ovaries present. The No BSO group consisted of women who had no ovaries removed or who underwent removal of one ovary, leaving one present, at the time of hysterectomy. Bilateral salpingo-oophorectomy was performed, according to the protocol of the Medicine or Surgery and TOSH trials, to treat abnormal intraoperative findings or at the patient's specific request, not electively according to the surgeon's practice style.
Baseline demographic information, health-related quality of life, and sexual functioning were assessed before randomization in the Medicine or Surgery or TOSH trials. A brief interview to assess psychological well-being, physical health, and satisfaction with symptom level was conducted by telephone 4 weeks after randomization for women in the TOSH study assigned to supracervical or total hysterectomy and for women in the medicine group of the Medicine or Surgery study who subsequently crossed over to hysterectomy. The telephone interview occurred 4 weeks after surgery for women originally assigned to the hysterectomy group in the Medicine or Surgery study. Clinic-based interviews that included detailed measures of health-related quality-of-life and sexual functioning were performed at 6, 12, 18, and 24 months after randomization.
Participants in the Medicine or Surgery and TOSH studies underwent hysterectomy at varying time points after entering the trials. Therefore, to ensure that data for follow-up visits in this analysis accurately represented time since surgery, we organized follow-up data points relative to the date of hysterectomy. For the 4-week visit, the phone interview occurred 2–6 weeks after surgery. For subsequent follow-up visits, the interview was conducted at the following time points: 42 days or more but 6 months or less after hysterectomy for the 6-month visit, 1 year or more but 18 or 24 months or less after surgery for the 18- or 24-month visit. The average time from hysterectomy to the follow-up interview was the same between the BSO group and the No BSO group at all follow-up visits.
In the Medicine or Surgery and TOSH randomized trials, health-related quality of life and sexual functioning were assessed using several measures from the Medical Outcomes Study,22,23 including the Medical Outcomes Study SF-36.24–26 Detailed descriptions of these measures, as well as several unique scales developed specifically for the Medicine or Surgery and TOSH trials, have been previously reported.19,20,27 For this study, outcome measurements were selected from the Medicine or Surgery and TOSH trials that we hypothesized would be adversely affected by oophorectomy. We used the Medical Outcomes Study SF-36 Mental Component Summary as our primary outcome to determine the effect of BSO on the anticipated improvements in mental health after hysterectomy.
All health-related quality-of-life and sexual functioning measures were transformed to a 0-to-100 scale, with 100 representing the most favorable outcome of health or functioning. If fewer than 50% of items were answered on a given scale, the score was considered missing. Clinical symptoms were evaluated on a 4-point scale for frequency in the last 4 weeks, from never to every day. Hormone use during the last 3 months was ascertained by self-report at each visit. We defined hormone use as any use of estrogen (oral, transdermal, intradermal, or vaginal cream), progestins (oral or intradermal), combined oral contraceptives, or clonidine for hot flushes. Participants were provided the opportunity to indicate and describe use of “other” hormonal medications, including nonprescription preparations. We performed a test for interaction between BSO and study type (Medicine or Surgery or TOSH) to assess the effect of combining outcome data from two randomized trials.
Baseline differences between the BSO and No BSO groups were analyzed using t tests for continuous variables and χ2 or Fisher exact tests for categorical variables. To estimate the effects of BSO, we assessed changes from baseline to follow-up interviews for the full range of outcomes. To add precision to the comparison of the 2-year outcomes, we analyzed changes from baseline to 18 and 24 months as repeated measures, using a random effects model to account for within-participant correlation. All models controlled for clinic site to account for variation in patient populations across recruiting centers. To minimize confounding of associations between BSO and study outcomes, we adjusted for age and having a sexual partner at baseline, two covariates that differed between groups before surgery (P<.05). We considered hormone use a mediator in the relationship between BSO and study outcomes. Therefore, our primary models did not include hormone use as a covariate; the results estimate the combined effects of BSO and associated hormone use. However, because clinicians may be interested in the effect of BSO on posthysterectomy outcomes in the absence of hormone use, we adjusted for hormone use as a time-dependent covariate in supplementary analyses. In sensitivity analyses, we adjusted for baseline sleep problems, a covariate weakly associated with BSO (P=.09), and examined models from which covariates not associated with the outcome at P<.15 were omitted.
In the analysis of changes from baseline to 6 months, the sample of 48 women with BSO and 103 with no BSO provided 80% power in 2-sided tests with α of 5% to detect between-group differences of 46–57% of a standard deviation of the underlying outcome measure, depending on the within-subject correlation of the baseline and 6-month responses, which lay in the range between 0.4 and 0.6. These are moderate-to-large treatment effects. Minimum detectable between-group differences in the changes from baseline to 18 or 24 months were comparable, since the gains in efficiency provided by use of repeated measures were to some extent offset by lower levels of within-subject correlation over the longer time period. These estimates of minimum detectable effects incorporate a penalty for moderate loss of precision due to adjustment for confounding.
Table 1 presents baseline data on 161 women who underwent hysterectomy in the Medicine or Surgery and TOSH studies and completed at least one scheduled interview following surgery. The different subject numbers in Tables 2 and 3 represent participants selected from these 161 women with follow-up interviews that corresponded to time since hysterectomy. The majority of participants were African American and had a family income less than $25,000 (Table 1). The mean age was 45 years for women who had a BSO compared with a mean age of 40 years for women in the No BSO group (P<.001). Other sociodemographic and clinical characteristics were similar in both groups.
Health-related quality-of-life and sexual functioning were also similar between the groups before hysterectomy (Table 1). This was a highly symptomatic cohort in which the majority of the women had experienced several years of pelvic pain, pelvic pressure, and abnormal bleeding. The mean Mental Component Summary and Physical Component Summary scores for both groups were below the U.S. population norms (norms for women aged 35–44 years: Mental Component Summary 49, Physical Component Summary 51). Baseline clinical symptoms including hot flushes, pelvic pain, and urinary incontinence were similar between the groups. Women who underwent BSO were less likely to have a sexual partner than those in the No BSO group (67% versus 83%, P=.03).
Four weeks after surgery, the BSO group demonstrated less improvement in mental health outcomes than women with ovarian conservation. The BSO group had smaller improvements in psychological well-being (change score 3 versus 10, P=.04), and there was weak evidence for less improvement in the Mental Component Summary (change score 4 versus 9, P=.07). Concordant with results from the Medicine or Surgery and TOSH trials,19,20 both groups had a decline in the Physical Component Summary during this early postoperative period (change score –5 for both groups, P=.96).
Both groups had improvement in all health-related quality-of-life outcomes 6 months after surgery, but the BSO group continued to show less improvement in mental health measures (Table 2). The Mental Component Summary change score was 5 points lower in the BSO group (P=.03). There was also a trend toward less improvement in psychological well-being for the BSO group (P=.07). However, after 2 years of follow-up, improvements in both the Mental Component Summary and the psychological well-being scores were similar in the BSO and No BSO groups.
Figure 1 demonstrates different trends for the Mental Component Summary score over time between the groups. The mean Mental Component Summary score reached a peak in the No BSO group 4 weeks after surgery, followed by a slight decline over the next 2 years. The 2-year mean Mental Component Summary score remained higher than the baseline value. In contrast, the mean Mental Component Summary score in the BSO group gradually increased over time, eventually matching the level of the No BSO group after 1 year of follow-up.
Sleep problems constituted another outcome that showed less improvement in the BSO group at the 6-month visit (Table 2). The BSO group had less improvement in sleep problems compared with the No BSO group (P<.01). However, 2 years after surgery, this difference was no longer present. We found no statistically significant between-group differences in the frequency of hot flushes, headaches, urinary incontinence, or pelvic pain at any visit after surgery. The BSO group did have greater improvement in the symptom resolution scale 2 years after surgery (P=.02), but the satisfaction with symptom level scale was similar in both groups.
Table 3 presents sexual functioning outcomes by BSO status. Women in the BSO group experienced less improvement in body image at the 6-month visit (P=.01), but that difference did not persist after 2 years of follow-up (P=.13). There were no other statistically significant differences in measures of sexual functioning between the groups at any follow-up visit.
As expected, use of hormones was much higher in the BSO group at all follow-up visits. At the 6-month visit, 56% of the BSO group was using estrogen (oral, transdermal, or intradermal) compared with 8% of the No BSO group. One woman in the BSO group was using clonidine for control of hot flushes. There was no use of estrogen vaginal cream or progestins in either group. At the 24-month visit, 45% of the BSO group was using estrogen (oral, transdermal, or intradermal) compared with 14% of the No BSO group. Other reported hormone use at 24 months after hysterectomy included one woman in the BSO group on clonidine, one woman in the No BSO group taking progestin, and 2 women in the No BSO group using estrogen vaginal cream. No women reported use of nonprescription estrogens or estrogen-containing dietary supplements at any follow-up visit. When we examined study outcomes adjusting for hormone use in the supplementary models, there were no statistically significant differences in any outcome measure between groups at any follow-up visit. In these analyses, hormone use was associated with smaller improvements in study outcomes.
In our sensitivity analyses, results were essentially unchanged if we included baseline sleep problems as a covariate in the regression models. Results were also unaffected by omitting covariates not associated with the outcome at P<.15 using backward deletion. The effects of BSO on the primary outcome, Mental Component Summary score, appeared similar in both studies (Medicine or Surgery versus TOSH).
In our study, BSO attenuated the improvements in health-related quality of life experienced by symptomatic premenopausal women after hysterectomy. As previously reported in the Medicine or Surgery and TOSH trials,19,20 scores for all health-related quality-of-life outcomes improved after hysterectomy. However, in this secondary analysis, scores for the BSO group showed less improvement in mental health measures and body image compared to women with ovarian conservation for the first 6 months after surgery. After 2 years of follow-up, the BSO group had achieved the same level of improvement for these outcomes as the No BSO group. The BSO group did experience greater improvement in the symptom resolution scale at 2 years of follow-up. However, we suspect that this finding was due to lower baseline values in the BSO group combined with the ceiling effect of the extremely high scores achieved by both groups after surgery.
Studies of the Medical Outcomes Study scales among various adult populations have suggested that meaningful differences in health-related quality of life are found with 3- to 5-point changes.23 In addition, a 6-point increase on the Mental Component Summary has been used to define significant change in five tracer conditions from the Medical Outcomes Study.28 In our study, the change scores between groups for Mental Component Summary, psychological well-being, sleep problems, and body image at 6 months of follow-up ranged from 5 to 13 points. Therefore, the magnitude of change we observed suggests clinically relevant differences in health-related quality of life between women who underwent BSO compared with ovarian conservation.
To date, there have been few studies that evaluate the effect of BSO on mental health outcomes after hysterectomy among women with benign gynecologic disease (PubMed search terms “oophorectomy” and “psychology” or “mental health,” with the limits of English language and human subjects, all publication dates).15,17,18,29 Aziz and colleagues17 conducted a large prospective cohort study of perimenopausal women undergoing hysterectomy with or without oophorectomy. They reported no differences in an index for psychological well-being between women who had a BSO versus those with ovarian conservation. However, women were only evaluated at baseline and 1 year after hysterectomy. In our analysis, we were able to detect early effects of surgical menopause by assessing women 4 weeks and 6 months after surgery. In the Aziz study, women in the BSO group were three times more likely to have used estrogen than women with ovarian conservation. However, the authors did not examine the association between BSO and study outcomes accounting for the effect of hormone use. Our study improves on this design by creating multivariate models that control for hormone use as time-dependent covariates.
Unlike most prior investigations, our study did not demonstrate poorer sexual functioning in the BSO group compared to women with ovarian conservation after hysterectomy.15,30–32 Most of the previous studies have been cross-sectional designs in which preoperative sexual functioning was unknown. Baseline sexual functioning may be a confounding factor in the relationship between BSO and postsurgical sexual functioning. Our study has the advantage of measuring sexual functioning before surgery, which was used as the baseline value for all postoperative comparisons.
There are several limitations to our study which must be considered in counseling patients regarding the effect of BSO at the time of hysterectomy. This is not a randomized trial of BSO, so residual confounding may have affected our results. Specifically, because participants were not randomized, we cannot rule out the possibility that preoperative counseling by the physician or the participants' personal beliefs regarding BSO may have biased reporting of postoperative outcomes. However, the only sociodemographic difference between the groups was age at surgery, which likely reflects the primary criteria by which participants requested, or physicians suggested, a BSO. We controlled for age in the analysis to account for the possible influence of this difference between groups in determining outcome measures. In addition, this is an exploratory analysis with multiple outcomes for the various domains within health-related quality of life. Given the possibility of chance findings in this setting, our conclusions must be interpreted with some caution and merit replication.
As reported in the Medicine or Surgery and TOSH studies,19,20 over half of the study participants were low-income African-American women living in the southeastern United States, which may limit the generalizability of our findings to other clinic populations. In addition, because of inclusion of participants from the TOSH randomized trial, the rate of supracervical hysterectomies (SCHs) in this analysis was higher than in most general gynecology settings in the United States.33,34 Although this may somewhat limit the generalizability of our findings, we do not believe the high rate of SCH significantly biased our results. First, there was no statistically significant difference in the rate of SCH between the BSO groups. Second, the effect of SCH on study outcomes is likely limited, given that the TOSH study reported no difference in sexual functioning at 2 years of follow-up between the SCH and total hysterectomy groups.19
Hormone use was common among the BSO group at 6 months and 2 years of follow-up. The Medicine or Surgery and TOSH studies were not designed to adjudicate the self-report of hormone use through review of medical records. However, women were queried about a short recall time of the last 3 months, which likely limited the potential for error in reporting hormone use. For the first 6 months after hysterectomy, the BSO group had less improvement in the Mental Component Summary, despite much higher rates of hormone use, as demonstrated in the primary analyses which did not adjust for hormone use. Our attempt to estimate the effect of BSO in the absence of hormone use should be interpreted with caution because hormone use was unexpectedly associated with poorer outcomes in this analysis. This effect is most likely due to confounding by indication. Had we been able to remove the beneficial effects of hormone use in the BSO group, the difference in average improvements in Mental Component Summary scores at 6 months in favor of the No BSO group would likely have been larger.
Improvement in health-related quality of life is the primary goal for women undergoing hysterectomy for benign gynecologic disease. In our study, women who underwent BSO at the time of hysterectomy had smaller improvements in some health-related quality-of-life outcomes for the first 6 months after surgery. However, by 2 years of follow-up, virtually every measure of health-related quality of life and sexual functioning were similar between the groups, irrespective of estrogen use. These findings may assist physicians in counseling premenopausal women regarding the decision to perform a BSO at the time of hysterectomy for benign gynecologic disease.
1.Keshavarz H, Hillis SD, Kieke BA, Marchbanks PA. Hysterectomy surveillance: United States, 1994–1999. MMWR CDC Surveill Summ 2002;51 (SS-5):1–8.
2.Carlson KJ, Miller BA, Fowler FJ Jr. The Maine Women's Health Study. I. Outcomes of hysterectomy. Obstet Gynecol 1994;83:556–65.
3.Averette HE, Nguyen HN. The role of prophylactic oophorectomy in cancer prevention. Gynecol Oncol 1994;55:S38–41.
4.Beard RW, Kennedy RG, Gangar KF, Stones RW, Rogers V, Reginald PW, et al. Bilateral oophorectomy and hysterectomy in the treatment of intractable pelvic pain associated with pelvic congestion. Br J Obstet Gynaecol 1991;98:988–92.
5.Cronje WH, Vashisht A, Studd JW. Hysterectomy and bilateral oophorectomy for severe premenstrual syndrome. Hum Reprod 2004;19:2152–5.
6.Namnoum AB, Hickman TN, Goodman SB, Gehlbach DL, Rock JA. Incidence of symptom recurrence after hysterectomy for endometriosis. Fertil Steril 1995;64:898–902.
7.Prophylactic oophorectomy. ACOG Practice Bulletin No. 7. Clinical management guidelines for obstetrician-gynecologists. American College of Obstetricians and Gynecologists. Int J Gynaecol Obstet 1999;67:193–9.
8.Kauppila A. Prophylactic oophorectomy. Ann Chir Gynaecol 1982;71:305–7.
9.Piver MS. Prophylactic oophorectomy: reducing the U.S. death rate from epithelial ovarian cancer. A Continuing Debate. Oncologist 1996;1:326–30.
10.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.
11.Shifren JL. Androgen deficiency in the oophorectomized woman. Fertil Steril 2002;77 suppl:S60–2.
12.Davison SL, Davis SR. Androgens in women. J Steroid Biochem Mol Biol 2003;85:363–6.
13.Dennerstein L, Randolph J, Taffe J, Dudley E, Burger H. Hormones, mood, sexuality, and the menopausal transition. Fertil Steril 2002;77 suppl:S42–8.
14.Sarrel PM. Sexuality and menopause. Obstet Gynecol 1990;75(4 Suppl):26S–30S; discussion 1S–5S.
15.Nathorst-Boos J, von Schoultz B. Psychological reactions and sexual life after hysterectomy with and without oophorectomy. Gynecol Obstet Invest 1992;34:97–101.
16.Nathorst-Boos J, von Schoultz B, Carlstrom K. Elective ovarian removal and estrogen replacement therapy: effects on sexual life, psychological well-being and androgen status. J Psychosom Obstet Gynaecol 1993;14:283–93.
17.Aziz A, Brannstrom M, Bergquist C, Silfverstolpe G. Perimenopausal androgen decline after oophorectomy does not influence sexuality or psychological well-being. Fertil Steril 2005;83:1021–8.
18.Farquhar CM, Harvey SA, Yu Y, Sadler L, Stewart AW. A prospective study of 3 years of outcomes after hysterectomy with and without oophorectomy. Am J Obstet Gynecol 2006;194:711–7.
19.Kuppermann M, Summitt RL Jr, Varner RE, McNeeley SG, Goodman-Gruen D, Learman LA, et al. Sexual functioning after total compared with supracervical hysterectomy: a randomized trial. Obstet Gynecol 2005;105:1309–18.
20.Kuppermann M, Varner RE, Summitt RL Jr, Learman LA, Ireland C, Vittinghoff E, et al Effect of hysterectomy vs medical treatment on health-related quality of life and sexual functioning: the Medicine or Surgery (Ms) randomized trial. JAMA 2004;291:1447–55.
21.Learman LA, Summitt RL Jr, Varner RE, Richter HE, Lin F, Ireland CC, et al. Hysterectomy versus expanded medical treatment for abnormal uterine bleeding: clinical outcomes in the medicine or surgery trial. Obstet Gynecol 2004;103:824–33.
22.Stewart AL, Ware JE. Measuring functioning and well-being: the Medical Outcomes Study approach. Durham (NC): Duke University Press; 1999.
23.Stewart AL, Greenfield S, Hays RD, Wells K, Rogers WH, Berry SD, et al. Functional status and well-being of patients with chronic conditions: results from the Medical Outcomes Study [published erratum appears in JAMA 1989;262:2542]. JAMA 1989;262:907–13.
24.McHorney CA, Ware JE Jr, Lu JF, Sherbourne CD. The MOS 36-item Short-Form Health Survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994;32:40–66.
25.McHorney CA, Ware JE Jr, Raczek AE. The MOS 36-Item Short-Form Health Survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:247–63.
26.Ware JE, Snow KK, Kosinkski M, Gandek B. SF-36 Health Survey: manual and interpretation guide. Boston (MA): The Health Institute, New England Medical Center; 1993.
27.Varner RE, Ireland CC, Summitt RL Jr, Richter HE, Learman LA, Vittinghoff E, et al. Medicine or Surgery (Ms): a randomized clinical trial comparing hysterectomy and medical treatment in premenopausal women with abnormal uterine bleeding. Control Clin Trials 2004;25:104–18.
28.Ware JE, Kosinski M. SF-36 Physical & Mental Health Summary Scales: a manual for users of version 1. Lincoln (RI): Quality Metric; 2001.
29.Taylor M. Psychological consequences of surgical menopause. J Reprod Med 2001;46 suppl:317–24.
30.Bellerose SB, Binik YM. Body image and sexuality in oophorectomized women. Arch Sex Behav 1993;22:435–59.
31.Dennerstein L, Koochaki P, Barton I, Graziottin A. Hypoactive sexual desire disorder in menopausal women: a survey of Western European women. J Sex Med 2006;3:212–22.
32.Rhodes JC, Kjerulff KH, Langenberg PW, Guzinski GM. Hysterectomy and sexual functioning. JAMA 1999;282:1934–41.
33.Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990–1997. Obstet Gynecol 2002;99:229–34.
34.Jacobson GF, Shaber RE, Armstrong MA, Hung YY. Hysterectomy rates for benign indications. Obstet Gynecol 2006;107:1278–83.