Hysterectomy is the most common nonobstetric major surgical procedure performed in U.S. short-stay hospitals.1 The number of hysterectomies performed in the this country has increased in the last decade from approximately 540,000 in 1996 to 670,000 in 2002.1,2 Most hysterectomies are done to control or eliminate symptoms and therefore improve quality of life.3 Recent attention has focused on sexual functioning after hysterectomy, with numerous reports demonstrating significant improvements in this important domain.4–6 Most abdominal hysterectomies are “total” procedures that include removal of the cervix. With the advent of effective cervical screening and early anecdotal reports of the possible superiority of retaining the cervix for preserving sexual functioning,7,8 along with suggestions that reduced surgical time and postoperative morbidity may warrant use of this procedure,9 “supracervical” or “subtotal” procedures have garnered increased interest by some patients and physicians.10 Between 1988 and 1998 the rate of total abdominal hysterectomy (TAH) in Denmark fell by 38%, and the rate of supracervical hysterectomy (SCH) increased by 458%,11 and between 1990 and 1996, the rates of TAH and SCH in New York State showed a similar pattern, although TAH remained the more common procedure.12
Trials of women randomly assigned to undergo SCH or TAH in the United Kingdom and Denmark have found no difference in sexual functioning at 1-year follow-up,13,14 calling into question early suggestions that SCH might be the preferred procedure for women concerned with preserving or improving sexual function. However, observational studies continue to put forth conflicting reports of the effect of these 2 procedures.15,16 Interpreting results of these nonrandomized studies is difficult because without randomization, confounding variables such as expectation and preference might play a role in treatment effectiveness.
We conducted the Total or Supracervical Hysterectomy (TOSH) study, a multicenter randomized trial, comparing the effectiveness of TAH and SCH in women scheduled to undergo abdominal hysterectomy for symptomatic uterine leiomyomata or abnormal uterine bleeding. In prior articles, we reported that SCH and TAH are comparable in terms of clinical and economic outcomes.17,18 This report focuses on sexual functioning and health-related quality-of-life outcomes during 2 years of follow-up.
MATERIALS AND METHODS
Details of the TOSH study design, including a diagram of participant flow through the trial, have been reported.17 We recruited premenopausal women aged 30 or older who had decided with their gynecologists to undergo abdominal hysterectomy to treat their abnormal uterine bleeding or symptomatic uterine leiomyomata (bleeding, pressure, or pain). Patients entering the TOSH study with abnormal uterine bleeding were required to have undergone at least a 3-month trial of hormonal treatment and to have had a structural evaluation of the endometrial cavity within 6 months before randomization (by ultrasound, hysteroscopy, or hysterogram) to rule out lesions amenable to hysteroscopic treatment. We also included women aged older than 45 years whose follicle stimulating hormone level was 30 mIU/mL or less and who had undergone an endometrial biopsy within 6 months that was negative for hyperplasia or carcinoma. All participants were required to have a Pap test within 1 year before randomization showing no cellular changes suggestive of dysplasia.
Subjects were recruited from the gynecology clinics and practices of the University of Alabama, Birmingham; the University of Tennessee, Memphis; Wayne State University, Detroit; and clinics affiliated with the University of California, San Diego. Institutional review board approval was obtained at each site and at the University of California, San Francisco, which served as the coordinating center. All subjects signed informed consent.
Between January 1998 and April 2000 we screened 729 women and identified 163 who were eligible for the TOSH trial. Of these, 135 (83%) were willing to be randomly assigned to SCH or TAH and were enrolled in the study. Randomization was stratified by clinical center, using randomly permuted blocks of 4, 6, and 8 and a computer-generated random numbers sequence. Nurse coordinators opened a sealed, opaque envelope with the treatment assignment after consent was obtained. Study participants, research staff and health-care providers were not blinded to the treatment assignment.
Participating surgeons were free to perform total or supracervical hysterectomies using their customary techniques. However, bilateral oophorectomy could be performed only to treat abnormal findings or at the patient's specific request, and not electively according to the surgeon's practice style.
Each woman was followed for 2 years after randomization. We measured baseline sociodemographic, clinical, sexual functioning, and health-related quality-of-life variables at the randomization visit and conducted a telephone interview at 4 weeks postrandomization to assess short-term health-related quality-of-life outcomes. Other contacts included detailed clinic-based interviews every 6 months and brief telephone interviews at 3, 9, 15, and 21 months. Interim safety monitoring was carried out by an independent data and safety monitoring board. Follow-up data were available at 24 months for 64 of 67 subjects assigned to TAH (96%) and for 61 of 68 subjects assigned to SCH (90%). The difference in loss-to-follow-up was not statistically significant (P = .20), and no significant differences surfaced in the characteristics of women lost to follow-up compared with those who completed the study.
Because our main interest was in sexual functioning after SCH as compared with TAH, we used the Medical Outcomes Study (MOS) Sexual Problems Scale19 as our primary outcome measure. This scale includes 4 items asking how much of a problem in the past 4 weeks were lack of sexual interest, being unable to relax and enjoy sex, having difficulty becoming aroused, and having an orgasm. Response options include “not a problem,” a “little,” “somewhat,” and “very much” of a problem, as well as “did not have sexual activity.” A scale value contributed to the analysis only if the participant reported sexual activity.
To measure specific aspects of sexual functioning in more detail, we developed 4 new scales using a combination of items from the MOS Sexual Problems Scale and existing and adapted questions from other studies,4,20,21 along with a few items developed for this study. These scales included “pelvic problems interference with sex,” a 4-item measure of the frequency of pain during sexual activity, and the extent to which bleeding, pelvic pain, and pelvic problems overall interfered with sexual activity; “sexual desire,” a 4-item measure of the frequency of sexual desire and activity and the extent to which lack of sexual interest or an inability to relax and enjoy sex was a problem; “orgasm frequency and quality,” a 4-item measure asking about difficulty in having an orgasm, the frequency of having orgasm during sex and feeling satisfied after sex and the intensity of orgasm when experienced; and “satisfaction with sex,” a 2-item measure of the extent to which participants felt satisfied with their ability to have and enjoy sex. Items were adapted to include sexual activity with and without a partner.
We also developed a measure of body image using 2 of the constructs contained in the Body Attitudes Questionnaire,22 and adding 2 of our own items to generate a scale measuring the frequency of feeling feminine, good about one's body, physically attractive, and sexually attractive. To further describe the sexual functioning of these women we included several single-item measures of the frequency of sexual desire and activity alone or with a partner in the past four weeks,4 as well as a question asking how important sex was to the participant.23 Finally, we used several measures from the MOS24 to assess other health-related quality-of-life outcomes. Psychometric properties and more detailed descriptions of all study measures have been reported.6,25
We analyzed baseline differences between the SCH and TAH groups using t tests for continuous variables and Fisher exact or χ2 tests for categorical variables. In the primary analysis, linear regression models were used to assess differences in mean outcome values at 6 months and end of study between the groups assigned to SCH or TAH. In a secondary “as treated” analysis we compared the outcomes of all women who had undergone SCH with those who had undergone TAH. To add precision to the estimated between-group differences in end-of-study outcomes, we analyzed the18- and 24-month outcomes as repeated measures, using a random effect for each subject to account for within-subject correlation. Logistic models were used to assess binary outcomes (such as having a sexual partner) with generalized estimating equations used to estimate the model for the repeated 18- and 24-month measures. Finally, we used the bias-corrected bootstrap to obtain confidence intervals for between-group differences in the proportions at various levels of ordinal outcomes (such as categorized frequency of sexual activity).26 In this procedure, resampling of individuals rather than visits was used for account of within-individual correlation of the 18- and 24-month outcomes.
Although we observed significant between-group differences in the primary outcome at baseline, we neither adjusted the primary comparison of the outcomes at each follow-up time point for baseline values nor used change scores for the following reasons. First, in the presence of a between-group difference in the outcome at baseline, adjustment for the baseline values of the outcome introduces bias in estimates of the effect of treatment.27 Although analysis of change scores would have been unbiased for the comparison of mean changes in the presence of baseline differences, we considered them potentially misleading, because of ceiling effects. These effects were observed for both treatment groups (either by assignment or treatment received) in plots of the trajectories of the mean values of the sexual functioning and health-related quality-of-life scores at every time point. Ceiling effects also were evident in the proportions of women scoring in the low, middle, and upper ranges of the primary outcome scale at every visit, again by treatment assigned and undergone. In this context, larger improvements in the group with lower baseline scores are not interpretable as a greater treatment benefit, because the means in both groups seem to reach the same ceiling. Hence, our primary analysis focused on the levels achieved postrandomization, without accounting for the baseline differences.
In the repeated-measures analysis with adjustment for baseline scores specified in the protocol, the planned sample of 160 participants was expected to provide 90% power to detect a between-group difference in mean sexual problem scores of approximately 0.42 standard deviations. A data and safety monitoring board monitored the study every 6–12 months for adverse trends in either of the treatment groups. At each meeting, the board deemed the adverse study outcomes to be consistent with those published for hysterectomies in general and allowed the study to continue without modification.
Baseline sociodemographic, clinical, sexual functioning and health-related quality-of-life characteristics are presented in Table 1. Women who were randomly assigned to SCH had significantly higher scores on the Sexual Problems Scale (indicating fewer problems; 69 compared with 55, P = .03), and seemed to have better functioning on all of the other sexual functioning measures, although statistically significant differences were not observed. The treatment groups were comparable on all other measured variables at baseline.
Mean scores and proportions at 6 months and 2 years are found in Table 2. Although women assigned to SCH reported a higher mean score on the orgasm frequency and quality scale at 6 months (73 compared with 64, P = .05), most likely due to their better sexual functioning at baseline, by 1 year these differences had dissipated, and at 2 years no significant differences in the Sexual Problems Scale or any of the secondary outcomes were apparent. “As treated” analyses yielded similar findings.
Trajectories of the mean values in the Sexual Problems Scale are depicted in Figure 1. The significantly lower mean baseline Sexual Problems Scale score of women allocated to TAH reached a plateau at 1 year similar to the one that women assigned to SCH reached by 6 months. Moreover, despite baseline differences in sexual problem scale scores, by 12 months the majority of the sexually active women in both groups had scores above 90 (Fig. 2). Figure 3 presents the trajectories in MCS and PCS scores over time. The trajectories paralleled each other very closely, demonstrating an early improvement on the mental component summary that was sustained over time, and a dip in the physical component summary at 4 weeks after surgery, which rose dramatically by 6 months, and again was sustained over time.
This randomized trial of total or supracervical hysterectomy found no significant difference in sexual functioning or quality-of-life outcomes. Both groups demonstrated substantial improvement in most sexual functioning and health-related quality-of-life measures, and by 2 years the outcomes of both randomization groups were similar.
Our findings provide further validation of those reported by investigators in the United Kingdom13 and Denmark.14,28 Thakar and colleagues13 found no differences in sexual function at 6 or 12 months after randomization to subtotal compared with total abdominal hysterectomy among 279 women seeking care at 2 London hospitals. More recently, Gimbel, Zobbe and others11,14,28 randomly assigned 319 Danish women with noncancerous uterine conditions to TAH or SCH, also finding no difference in sexual functioning or health-related quality of life at 12 months of follow-up. Our study population differed substantially from those of these studies. The British study excluded women with a body weight exceeding 100 kg, previous pelvic surgery, and known endometriosis. Our trial did not make these exclusions, and the mean weight of the women in our study (83 kg) was far greater than the 70 kg in the British trial. Moreover, the British trial included postmenopausal women who were aged 60 years or younger, whereas menopausal patients were excluded from our study. Two-thirds of the women in the British study were white, compared with 23% of our study population. The Danish study included women with a broader range of conditions and weights; however, almost all participants were white (V. Zobbe, personal communication). The similarity of findings among the 3 trials despite differences in the characteristics of their populations supports the validity of concluding that SCH and TAH have similar effects on sexual functioning and health-related quality of life. The longer follow-up in our study indicates that these findings persist beyond the first year after surgery.
Our study had several limitations. First, the one statistically significant imbalance between randomized groups in baseline variables was in our primary outcome measure, the MOS Sexual Problems Scale. We do not know why this imbalance occurred, whereas all of the measures of health-related quality of life, along with body image and sexual activity, were balanced. We believe this was due to chance and that the differences in orgasm frequency and quality at 6 months may be due in part to differences in sexual functioning at baseline. It is also possible that SCH patients resumed sexual activity sooner than TAH patients, although clinical outcomes suggest similar complication rates and days of restricted activity.17 By 1 year these differences had dissipated, however, leading to our conclusion that the long-term effects of these 2 approaches to treatment were similar.
In addition, only patients reporting sexual activity could contribute values for the MOS Sexual Problem Scale. Patients who did not have sexual activity because of sexual dysfunction issues would therefore not contribute a value. However, similar proportions of the 2 treatment groups reported sexual activity at the 18- or 24-month visit either with (79% in SCH compared with 75% in TAH) or without a partner (19% in SCH compared with 23% in TAH). This is consistent with the similarity of other measures of sexual functioning among the subset of sexually active women, making it unlikely that strong bias was introduced by not including data from sexually inactive women.
Another limitation is that participants were informed which procedure they were going to undergo, in part to be sure that those with a cervix knew that they would need to continue cervical cancer screening, and there was no way to blind the participating gynecologists. Patients and physicians with a strong bias for or against cervical removal were unlikely to participate in the trial (which had had the benefit of reducing the likelihood of biased reporting or treatments during follow-up). Also, the specifics of the surgical procedures were left to attending physicians, enhancing generalizability but preventing us from knowing whether a specific aspect of the interventions we studied had a particular effect on our outcomes. In addition, although we used the MOS Sexual Problems Scale as our primary outcome measure, we used newly created measures for some secondary outcomes, limiting our ability to make direct comparisons with other studies in these domains. Finally, most of the women in our study lived in the southeastern United States, where hysterectomy rates are higher than other parts of the country,29 and their average body mass index was higher than that of similarly aged U.S. women.30
Earlier reports from the TOSH study showed that there were no significant differences in clinical and economic outcomes of SCH compared with TAH over 2 years of follow-up. This analysis confirms and extends findings from studies conducted in other populations that there also seems to be no difference in sexual functioning or health-related quality-of-life after SCH compared with TAH. These results should lead to more informed decision making by patients and providers as they consider alternative techniques of hysterectomy.
1.DeFrances CJ, Hall MJ. 2002 National Hospital Discharge Survey. Adv Data 2004;1–29.
2.Keshavarz H, Hillis SD, Kieke BA, Marchbanks PA. Hysterectomy surveillance—United States, 1994–1999. MMWR CDC Surveill Summ 2002;51(SS05):1–8.
3.Farquhar CM, Steiner CA. Hysterectomy rates in the United States 1990-1997. Obstet Gynecol 2002;99:229–34.
4.Rhodes JC, Kjerulff KH, Langenberg PW, Guzinski GM. Hysterectomy and sexual functioning. JAMA 1999;282:1934–41.
5.Dragisic KG, Milad MP. Sexual functioning and patient expectations of sexual functioning after hysterectomy. Am J Obstet Gynecol 2004;190:1416–8.
6.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.
7.Kilkku P, Gronroos M, Hirvonen T, Rauramo L. Supravaginal uterine amputation vs. hysterectomy: effects on libido and orgasm. Acta Obstet Gynecol Scand 1983;62:147–52.
8.Zussman L, Zussman S, Sunley R, Bjornson E. Sexual response after hysterectomy-oophorectomy: recent studies and reconsideration of psychogenesis. Am J Obstet Gynecol 1981;140:725–9.
9.Hasson HM. Cervical removal at hysterectomy for benign disease: risks and benefits. J Reprod Med 1993;38:781–90.
10.Munro MG. Supracervical hysterectomy: a time for reappraisal. Obstet Gynecol 1997;89:133–9.
11.Gimbel H, Settnes A, Tabor A. Hysterectomy on benign indication in Denmark 1988-1998: a register based trend analysis. Acta Obstet Gynecol Scand 2001;80:267–72.
12.Sills ES, Saini J, Applegate MS, McGee M, Gretz HF 3rd. Supracervical and total abdominal hysterectomy trends in New York State: 1990-1996. J Urban Health 1998;75:903–10.
13.Thakar R, Ayers S, Clarkson P, Stanton S, Manyonda I. Outcomes after total versus subtotal abdominal hysterectomy. N Engl J Med 2002;347:1318–25.
14.Zobbe V, Gimbel H, Andersen BM, Filtenborg T, Jakobsen K, Sorensen HC, et al. Sexuality after total vs. subtotal hysterectomy. Acta Obstet Gynecol Scand 2004;83:191–6.
15.Saini J, Kuczynski E, Gretz HF 3rd, Sills ES. Supracervical hysterectomy versus total abdominal hysterectomy: perceived effects on sexual function. BMC Womens Health 2002;2:1.
16.Kim DH, Lee YS, Lee ES. Alteration of sexual function after classic intrafascial supracervical hysterectomy and total hysterectomy. J Am Assoc Gynecol Laparosc 2003;10:60–4.
17.Learman LA, Summitt RL Jr, Varner RE, McNeeley SG, Goodman-Gruen D, Richter HE, et al. A randomized comparison of total or supracervical hysterectomy: surgical complications and clinical outcomes. Obstet Gynecol 2003;102:453–62.
18.Showstack J, Kuppermann M, Lin F, Vittinghoff E, Varner RE, Summitt RL Jr, et al. Resource use for total and supracervical hysterectomies: results of a randomized trial. Obstet Gynecol 2004;103:834–41.
19.Sherbourne CD. Social functioning: sexual problems measure. In: Stewart AL, Ware JE, editors. Measuring functioning and well-being: The Medical Outcomes Study approach. Durham (NC): Duke University Press;1992. p. 194–204.
20.Taylor JF, Rosen RC, Leiblum SR. Self-report assessment of female sexual function: psychometric evaluation of the Brief Index of Sexual Functioning for Women. Arch Sex Behav 1994;23:627–43.
21.Greendale GA, Petersen L, Zibecchi L, Ganz PA. Factors related to sexual function in postmenopausal women with a history of breast cancer. Menopause 2001;8:111–9.
22.Ben-Tovim DI, Walker MK. The influence of age and weight on women′s body attitudes as measured by the Body Attitudes Questionnaire (BAQ). J Psychosom Res 1994;38:477–81.
23.Greendale GA, Hogan P, Schumaker S. Sexual functioning in postmenopausal women: the postmenopausal estrogen/progestin interventions (PEPI) trial. J Womens Health 1996;5:445–58.
24.Stewart A, Ware J. Measuring functioning and well-being: The Medical Outcomes Study approach. Durham (NC): Duke University Press; 1992.
25.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.
26.Efron B, Tibshirani R. Bootstrap measures for standard errors, confidence intervals, and other measures of statistical accuracy. Stat Sci 1986;1:54–77.
27.Vittinghoff E, Glidden D, Shiboski S, McCulloch C. Longitudinal data. Regression methods in biostatistics: linear, logistic, survival, and repeated measures models. New York (NY): Springer; 2005. p. 262–6.
28.Gimbel H, Zobbe V, Andersen BM, Filtenborg T, Gluud C, Tabor A. Randomised controlled trial of total compared with subtotal hysterectomy with one-year follow up results. BJOG 2003;110:1088–98.
29.Lepine L, Hillis S, Marchbanks P, Koonin L, Morrow B, Kieke B, et al. Hysterectomy surveillance–United States, 1980-1993. MMWR CDC Surveill Summ 1997;46:1–15.
30.McTigue KM, Garrett JM, Popkin BM. The natural history of the development of obesity in a cohort of young U.S. adults between 1981 and 1998. Ann Intern Med 2002;136:857–64.
The TOSH Research Group
Clinical centers: University of Alabama, Birmingham: R. E. Varner, MD (principal investigator), H. Richter, MD, PhD (co-investigator), Robert L. Holley, MD (co-investigator), Seine Chiang, MD (co-investigator), V. Bannister, RN (clinic coordinator), L. Abdo, RN, CCRC (clinic coordinator), Alison Myrich (research technician). University of California, San Diego: E. Barrett-Connor, MD (principal investigator), D. Goodman-Gruen, MD, PhD (co- investigator), A. Fleming, RN (clinic coordinator). University of Tennessee, Memphis: R. L. Summitt Jr, MD (principal investigator), F. W. Ling, MD (co-investigator), Gary H. Lipscomb, MD (co-investigator), Marie Woodruff, RN, CCRC (clinic coordinator), D. Bran, RN, CCRC (clinic coordinator), M. F. Braslow (senior research technician). Wayne State University, Detroit: S. G. McNeeley, MD (principal investigator), S. Hendrix, DO (co-investigator), D. Kmak, MD (co-investigator), Paul R. Makela, MD (co-investigator), J. Mowery, RN (clinic coordinator), M. McNamee, RN (research nurse). Coordinating Center: University of California, San Francisco: E. Washington, MD, MSc (study chair), S. B. Hulley, MD, MPH (principal investigator), L. Learman, MD, PhD (co-investigator, gynecology), M. Kuppermann, PhD, MPH (co-investigator, quality of life), J. Showstack, PhD (co-investigator, health economics), E. Vittinghoff, PhD (statistician), F. Lin, MS (statistical programmer), C. Ireland, MPH (project director), C. Gehrman (programmer).
Funding was provided by Agency for Health Care Research and Quality, H Hubbard RN, MPH (project officer).
Data and Safety Monitoring Board, C. Westhoff, MD, MSc (chair), J. Wittes, PhD, M. Hlatky, MD (2000–2003), and W. Applegate, MD (1997–2000).