Mr I. Manyonda, Department of Obstetrics and Gynaecology, St George's Hospital, Blackshaw Road, London SW17 0QT, UK.
Objective To conduct a prospective and concurrent evaluation of changes in health status and quality of life and psychological outcome measures over one year in women randomised to total or subtotal abdominal hysterectomy. The concurrent evaluation was the impact of total versus subtotal hysterectomy on bladder, bowel and sexual function.
Setting A large UK Teaching Hospital (St George's Hospital, London) and a large District General Hospital (Mayday University Hospital, Croydon).
Sample Two hundred and seventy-nine women undergoing hysterectomy for benign disease were randomly allocated to total hysterectomy (n= 146) or subtotal hysterectomy (n= 133).
Main outcome measures Quality of life assessment using the Short-Form-36 health survey (SF-36) and psychological outcome measures using the General Health Questionnaire-28 (GHQ-28) before the operation and 6 and 12 months after.
Results Quality of life and psychological symptoms were similar in the two groups at baseline. Following surgery, quality of life improved in six of the eight domains, with no significant difference between the groups, with the exception of emotions which showed a greater improvement in subtotal hysterectomy women between baseline and 12 months. When this difference was examined further by looking at change in the GHQ subscales, there were no significant differences between total and subtotal hysterectomy women in the amount of change in anxiety, depression, somatic symptoms or social dysfunction, between baseline and post-operative measurements. All women showed an improvement in psychological symptoms following both operations.
Conclusion Hysterectomy, whether total or subtotal, may improve quality of life and psychological outcome.
It is now generally recognised that the functional impact on patients' lives of clinical interventions is important in predicting demand for services, and that it is not sufficient to simply measure outcome of clinical intervention in terms of morbidity and mortality. Health status and quality of life outcomes measured prospectively and concurrently, complement mortality and morbidity measures. Quality of life is an important outcome variable in clinical research1 as medical interventions can affect it in both positive and negative ways. Psychiatric symptoms can arise as a result of physical illness, or might influence the manifestation and/or outcome of treatment of that illness. In this regard, arguably the example par excellence is the impact of hysterectomy. Historically, the uterus has been regarded as the regulator and controller of important physiological functions, a sexual organ, a source of energy and vitality and a maintainer of youth and attractiveness.2 Therefore, it is unsurprising that removal of the uterus can and does generate fierce debate and controversy. In any clinical study of the outcome of hysterectomy, the need to conduct prospective and concurrent evaluations of quality of life and psychological measures, as well as morbidity and mortality, cannot be over-emphasised.
Although hysterectomy has been performed for over 150 years, and is currently the most common major gynaecological operation both in the United Kingdom3 and the United States of America,4 an area of persistent controversy is whether the subtotal procedure, when the cervix is conserved, confers any benefits over total hysterectomy. Gynaecologists have been pilloried in the media for apparently destroying women's sex lives by performing total rather than subtotal hysterectomy, as it has been suggested that the cervix might be important in sexual function,5 while debate rages among gynaecologists themselves as to whether there are indeed any benefits of one procedure over the other. We have recently published findings from a large prospective, randomised, double-blind, two-centre study in which we found no benefits for one operation over the other in terms of bladder, bowel and sexual function.6 While early research suggested that hysterectomy might be associated with adverse psychological outcome,7 recent studies tend to suggest either no change8 or an improvement9 in psychological outcome. Recent research also suggests that hysterectomy might improve quality of life,10 but no previous studies have compared subtotal to total hysterectomy with regards to quality of life and psychological outcome. In the same group of women in whom we addressed the issue of pelvic organ function,6 we looked at quality of life and psychological outcome as a secondary outcome measure.
Between January 1996 and April 2000, women who had been offered abdominal hysterectomy for a benign indication at a large teaching hospital and a district general hospital were recruited. Exclusion criteria were age over 60, suspected malignancy, weight over 100 kg, previous pelvic surgery, known endometriosis, abnormal cervical smears, bothersome urinary incontinence and uterine prolapse. All subjects provided written informed consent. The Research Ethics Committees at each participating hospital approved the study.
This prospective and concurrent assessment of quality of life was undertaken as part of a comparative study of total versus subtotal hysterectomy in which the primary outcome measures were pelvic organ function (urinary, bowel and sexual functions).6 Therefore, the power calculations were based on the difference in urinary symptoms after hysterectomy. The alpha (type 1) value was set at 0.05 and the beta (type 2) value was set at 0.1. Based on a previous randomised controlled study in which 55% of women had stress incontinence after hysterectomy,11 it was determined that 138 women would be required in each arm to detect an 18% difference. Examination of previous studies of quality of life in patients undergoing hysterectomy12 showed that the sample size calculated on the basis of urinary symptoms was sufficient to address the issue of quality of life and psychological measures.
Eligible women were randomised by computer generated numbers. The sealed opaque envelope was only opened after the surgical incision. The women and the investigator (RT) were blinded to the type of operation for one year. While women could always self-examine to establish which operation they had undergone, we strongly discouraged this, and our view is that those women who participated were highly motivated and compliant. All operations were carried out or supervised by an experienced surgeon using the clamp–cut–ligate method with polyglycolic sutures and antibiotic prophylaxis.13 During subtotal hysterectomy, the endocervical canal was electrocoagulated. When indicated, oophorectomy was the only concomitant procedure.
To assess health status and quality of life, the SF-36 questionnaire was used. The SF-36 is a 36-item self-administered questionnaire which measures health in eight multi-item dimensions: health perception, mental health, energy, physical function, role limitations due to physical factors, role limitations due to emotional factors, social function and pain.14 For each dimension, item scores are coded, summed and transformed on a scale from 0 (worst health) to 100 (best health). The SF-36 questionnaire is a widely used and robust measure of quality of life in patients.15 Data from population studies have shown that the SF-36 is reliable and able to detect differences between groups defined by age, sex, socio-economic status, geographical region and clinical conditions.
To assess psychological symptoms the GHQ-28 was utilised. It is a screening instrument designed to detect current psychiatric disorders.16 It is designed to cover four identifiable elements of distress: depression, anxiety, social dysfunction and somatic symptoms, and is intended for studies in which an investigator requires more information than is provided by a single severity score.17 The GHQ scoring system was used in which a two-point score rates each problem as present or absent and are scored as 0-0-1-1,18 therefore each subscale ranges from 0 (no symptoms) to 7 (severe symptoms).
Screening of the data indicated that a number of the variables were skewed. Therefore, change scores were calculated (by subtracting post-operative scores from baseline scores) and analyses carried out looking at differences between total and subtotal hysterectomy in change scores, using t tests or Mann–Whitney as appropriate. Friedman tests were used to assess change in quality of life in the total sample.
Figure 1 is a CONSORT diagram of the sample at recruitment and follow up. We recruited 279 women, of whom 146 underwent total hysterectomy (with bilateral salpingo-oophorectomy in 81), and 133 had subtotal hysterectomy (with bilateral salpingo-oophorectomy in 61). Twenty-six women failed to attend for their 6-month follow up, while 32 did not return for their 12-month appointment. This represents a 91% return rate at 6 months and 88% at 12 months. Two hundred and sixty-three women (94%) answered the GHQ-28 questionnaire pre-operatively, 232 (83%) at 6 months (21 attended but did not complete questionnaire) and 230 (82%) at 12 months (16 attended but did not complete questionnaire). Women in the total abdominal hysterectomy and subtotal abdominal hysterectomy groups were broadly comparable at baseline.Table 1 shows the demographics of the study population and indications for hysterectomy. While the patients in the two groups were matched for demographic characteristics, menorrhagia was more frequent in the subtotal abdominal hysterectomy group (P= 0.03) despite randomisation. This difference is likely to be due to chance, and therefore unlikely to affect important aspects of quality of life or psychological symptoms following hysterectomy. There were no significant differences between total and subtotal abdominal hysterectomy women on baseline measures of quality of life (see Table 2) or mental health.
Table 1. Demographic data. Values are presented as mean [SD], median (range) or n (%).
Table 2. Differences in quality of life for women following subtotal or total abdominal hysterectomy. Values are presented as mean (SD). Higher change scores indicate greater improvement. Scores were rounded up for clarity.
Differences at baseline
Change between baseline and 6 months post-operative
Change between pre-operative and 12 months post-operative
With the exception of role limitations due to emotions, which showed greater improvement in women who had subtotal hysterectomy between baseline and 12 months, we found no significant differences between total and subtotal hysterectomy on quality of life (SF-36) or psychological measures (GHQ) before and after hysterectomy (Table 2). The issue of role limitation due to emotions was examined further by looking at change in the GHQ subscales, but there were no significant differences between women who had total and subtotal hysterectomy in the amount of change in anxiety, depression, somatic symptoms or social dysfunction, between baseline and post-operative measurements.
In view of the lack of difference between the randomised groups, analysis was carried out to look at the effect of hysterectomy on quality of life for all the women in the study.Table 3 shows that all aspects of quality of life improved following hysterectomy, with the exception of reported energy/vitality which remained unchanged, and pain which was worse. However,Table 3 also shows that mean scores for quality of life were lower in this sample than norms from community samples of women aged 40–44 years.
Table 3. Quality of life and psychological symptoms before and after hysterectomy (in the whole sample). Data are mean (SD). Higher scores indicate better functioning for the quality of life subscales (SF-36) and worse mental health for the GHQ subscales.
Norms taken from Oxford Healthy Life Survey 1991/1992, HSRU, Oxford.
Analysis of the GHQ-28 showed that all aspects of mental health also significantly improved after hysterectomy (see Table 3).
To our knowledge, this is the first trial to compare the impact of total versus subtotal abdominal hysterectomy on quality of life and psychological symptoms. The study was prospective, randomised and double-blind (the person administering the outcome measures and the patients were blinded to the treatment for a period of one year). It was a two-centre study, with a variety of surgeons performing the operation, and thus introducing a measure of external validity. We conducted a minimum 12-month post-operative follow up, as long term complications may evolve over time. A resolution to the controversy of total versus subtotal hysterectomy has long been overdue. We reported recently that subtotal hysterectomy confers no added benefits over total hysterectomy with respect to bladder, bowel and sexual functions.6 That study was designed from the outset to concurrently measure quality of life and psychological outcome, because it has been argued that the functional impact of a clinical intervention should be the definitive arbiter of treatment success, and measurement of quality of life has been recommended for outcome assessment of treatments for menorrhagia.19
We hypothesised that subtotal hysterectomy would confer advantages over total hysterectomy. With respect to pelvic organ function, not only did we find no such advantage, in fact, we found that both operations were associated with significant improvements in some aspects of urinary function, and had no deleterious effects on other parameters of pelvic organ function.6 That we should find an improvement in quality of life following hysterectomy is therefore consistent. Our findings concur with those of Crosignani et al.,12 who reported improvement in quality of life using SF-36 in women after vaginal hysterectomy compared with endometrial resection, and Schulter et al.,20 who compared abdominal hysterectomy and transcervical resection of the endometrium. Similarly, others have reported symptom relief following hysterectomy associated with marked improvement in quality of life.21 Presumably, women feel better because they no longer suffer the symptoms that were the indication for a hysterectomy in the first place. Kjerulffe et al.22 reported that 12 and 24 months after hysterectomy symptoms were resolved completely or mostly in 95.8% and 96% women, respectively, while 85.3% and 81.6%, respectively, reported an improvement in health. The reduced SF-36 scores observed at baseline are consistent with the scores reported for other women with menorrhagia23 and concur with other findings.10,24
We found that although quality of life improved, the scores for all domains were lower than normal at 6 and 12 months post-operatively, reported energy/vitality remained unchanged and pain scores were worse. This is difficult to explain, but suggests that full recovery from hysterectomy may take longer than is currently appreciated. Physical recovery and relief of symptoms may explain the overall improvement in quality of life, but nevertheless, these women have undergone major surgery that has involved the removal of an organ central to the concept of womanhood. There may be ambivalences about this, which might take much longer to resolve, and therefore affect quality of life. The worsening in pain scores is even more difficult to explain, and we intend to address the issue of quality of life in a 5- to 10-year follow up of the same group of women, especially as other studies have not concurred with ours. For example, Rannestad et al.10 showed that compared with controls the health and functioning domains were lower than before hysterectomy and reached the same levels after 6 and 12 months.
Hysterectomy has traditionally been considered to be associated with adverse psychiatric sequelae.7 However, earlier studies often lacked conceptual clarity and had methodological flaws, many being retrospective analyses with inadequate measures of outcome. In recent years, more robust instruments have been employed to study psychological outcome following clinical interventions. Thus, more recent studies on the psychological sequelae of hysterectomy, using the GHQ or equivalent tools, have arrived at entirely different conclusion.8,25 In general, recent research indicates that women with pre-operative depression are at increased risk for depression after surgery.9,26 Our own findings that hysterectomy does not lead to a significant increase in symptoms of psychiatric illness are consistent with other prospective studies.8,25 Indeed, we found that psychiatric symptoms decreased after hysterectomy for all women, regardless of type of hysterectomy. This is consistent with the Maryland study, the largest prospective study to date of 1299 women, which reported a substantial decrease in depression and anxiety levels after hysterectomy.21
This study has a number of limitations. Although a variety of surgeons performed the operations, nevertheless only two centres were involved, and this may well limit the external validity of the study. It is possible that, as we did not have a control group of women who did not undergo surgery, women may have improved regardless of surgery. Thus, we acknowledge that our results should be replicated on a wider scale before our recommendations can be generalised.
In symptomatic women with menstrual or related disorders where conservative treatment has failed, total or subtotal hysterectomy may improve quality of life and reduce psychiatric symptoms, with both procedures having an equally beneficial effect. Perhaps the key to the optimal counselling of women when deciding which procedure to offer may lie in providing the woman with as much information as possible, and then allowing her to choose the type of operation. Such empowerment of women may well greatly improve satisfaction rates from hysterectomy.27
We would like to thank all our colleagues who generously allowed us to recruit their patients into the study. We owe a huge debt of gratitude to all the women who graciously consented to participate.
Source of funding
This study was funded by the NHS Executive London (NHSE-LRO) Research & Development, Responsive Funding Programme (Grant SPGS 202, awarded to ITM). The views expressed in the publication are those of the authors and not necessarily those of the NHS Executive or the Department of Health.
Conflict of interest statement
None of the authors has any conflict of interest with regards to this article.