1 Department of Obstetrics and Gynaecology, University of Wales College of Medicine and the University Hospital of Wales, Cardiff CF14 4XN, UK, 2 H. A. Béclère, Clamart, France, 3 St Josephs Health Centre, London, Canada, 4 Clin Méd. Chateauguay, Canada, 5 Belfast City Hospital, Belfast, UK, 6 University Clinic Obstetrics and Gynaecology, Graz, Austria, 7 Twenteborg H., Almelo, The Netherlands, 8 St Josephs Hospital, Veldhoven, The Netherlands, 9 University of British Columbia, Vancouver, Canada and 10 Hôpital Conception, Marseilles, France
11 To whom correspondence should be addressed. e-mail: amsonn{at}cardiff.ac.uk
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: endometrial thermal balloon therapy/hysterectomy/long-term follow-up/menorrhagia/UBT
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the UK, menorrhagia affects about 22% of premenopausal women over the age of 35 years, and 5% of women aged 3049 years consult their general practitioners (GPs) with heavy bleeding each year. First-line therapy is usually medication prescribed by GPs, but should these be ineffective, further treatment is necessary. Consideration of surgical interventions for menorrhagia accounts for some 12% of NHS gynaecology referrals (Sculpher et al., 1996). Unlike hysterectomy, a substantial proportion of women who undergo endometrial ablation will continue to menstruate, some with persistent or recurrent menorrhagia necessitating further surgery. Recent reports assessed the rate of further surgery in women who were treated with UBT in comparison with rollerball ablation or TCRE, but the follow-up period was relatively short, 2 and 3 years respectively (Bongers et al., 2000
; Loffer, 2001
).
We report the results of a study in which a cohort of women complaining of menorrhagia and who underwent UBT were followed up for 46 years. The primary outcome measure was avoidance of hysterectomy. Secondary outcome measures included: (i) need for other surgical interventions e.g. TCRE, rollerball or repeat balloon ablation; (ii) impact of patient- and treatment-dependant variables on avoidance of hysterectomy; and (iii) current bleeding pattern in women who had not undergone hysterectomy.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Ten centres participated in the current study and, where necessary, ethical approval was obtained. Patients were eligible for inclusion in the study if: (i) they were treated by UBT between June 1994 and August 1996; (ii) the starting balloon pressure gradient was 150 mmHg; (iii) they were not taking hormone replacement therapy (HRT) in the first year as part of another study; (iv) the 12-month follow-up (short-term) data for the original study were complete; and (v) they agreed to sign an informed consent form for the study. Patients were excluded from this study if: (i) they were treated by UBT with a starting gradient pressure of <150 mmHg; (ii) they were treated outside the clinical study protocol outlined above; (iii) the data recording of the 12-month follow-up was not complete; and (iv) they did not agree to take part or sign the patient informed consent form. As a result of the above criteria, 260 women from 10 participating centres were eligible for inclusion in the study. Two centres (41 eligible patients) did not take part in this study.
Surgery
Details of treatment protocol, technique and typical post-operative course have been described previously (Amso et al., 1998). In summary, the balloon was checked before insertion into the uterus and was filled with a variable volume of 5% dextrose water until a mean starting pressure of 167 ± 8 mmHg was achieved. The treatment cycle commenced when the fluid temperature reached 85 ± 5°C and continued for 8 min. The controller unit continuously monitors, displays and controls pre-set balloon pressure, temperature and duration of treatment. At the conclusion of the treatment, the balloon was emptied, removed and checked again for any leaks.
Follow-up
The 10 investigators participating in this study were provided with a list of their patients from the 260 eligible women who met the protocols selection criteria. The patients were mailed a study questionnaire asking them about their current menstrual history (days and severity of spotting or bleedinglight, equivalent to normal loss, or heavy) and, where applicable the date of cessation of their periods and any adverse events or disease since their first balloon ablation. They were also asked about current gynaecological problems, medication including HRT and whether they continued to bleed for some time after the ablation and then stopped (natural amenorrhoea) or their period stopped soon after the ablation (non-natural amenorrhoea). If further surgery (hysterectomy or repeat ablation) had taken place, the women were asked to give the date of the operation. In one centre, the ethics committee requested that the womens own GPs were contacted first for confirmation of the contemporaneous addresses of the women and whether they were still registered with their GPs before mailing the questionnaires. If a woman did not respond within the first 34 weeks, another questionnaire was mailed with a stamped, addressed envelope.
Statistical analysis
As we had anticipated that a proportion of women are lost to follow-up or would not respond to our questionnaire (non-participants group), it was essential to establish whether there was a significant risk of bias in the group who responded and hence for whom long-term follow-up data were available (participants group). To that effect, we compared the baseline and short-term follow-up characteristics, available from the original database, of the participants group and the non-participants group to ensure that they were comparable. Nine variables were selected for this comparison. Four were patient dependent (age at the time of the initial ablation, dysmenorrhoea score before and after treatment, uterine position, and depth of uterine cavity); three were intervention dependent (volume of fluid in balloon, starting pressure and treatment duration); and two were short-term outcome measurements (initial result and hysterectomy after balloon ablation). The level of significance adopted was defined using the Bonferroni method. This approach can be approximated by dividing the level of significance ( = 5%) by the number of variables tested (corrected
= 5/
, i.e. 5/9 = 0.555%). The ShapiroWilk test was used to test the hypothesis of normality of the two population groups (participants and non-participants). This was followed by a comparison of the means of these groups using Students t-test, if the sample variances were homogenous. All statistical tests were two-tailed. A non-parametric test such as MannWhitney U-test was applied whenever parametric tests where not applicable. For descriptive variables in a contingency table, Pearson
2- and M-L
2-tests were employed to test the homogeneity of distribution.
Relevant variables from the respondents to the questionnaires were analysed in accordance with appropriate statistical methods, and type 1 error () at 5% was accepted taking into account multiple testing.
Life-table analysis using the actuarial method (Greenwood, 1926; Cox and Oaks, 1984
) with 3 monthly intervals (quarters) was used to determine the risk of treatment failure after surgery, the primary end-point being the time when hysterectomy was performed. The short-term outcome data for the non-participants group were included in the calculation of the life-table analysis up to the point of their last contact. The same approach was adopted for repeat ablation or when calculating the probability of becoming amenorrhoeic.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Of the 187 participants who were entered in the life-table analyses, the time elapsed between the initial UBT and their response to the questionnaire was between 4 and 5 years in 38 patients, between 5 and 6 years in 122 patients, and between 6 and 7 years in 27 patients.
Table I shows comparison of the participants and non-participants groups for four patient-dependent variables, three intervention-dependent variables and two short-term outcome variables. Significant difference was reported for the depth of the uterine cavity and the dysmenorrhoea score before, but not after, the treatment. The mean depth of the uterine cavity in the participants group exceeded by 0.56 cm that of the non-participants group. This variable is of doubtful clinical relevance and when tested by logistic regression model was found to have little or no clinical impact on the risk of hysterectomy. More non-participants had dysmenorrhoea of any severity before treatment (P < 0.0009). However, there was no significant difference between the two groups in the post-treatment dysmenorrhoea scores at the short-term follow-up contact (P = 0.463). According to this information, analysis of the long-term UBT outcome based only on the participants group was appropriate and hence, any systematic selection bias was unlikely.
|
|
Further surgery
In the participants group, 25 women (13.3%) underwent hysterectomy and 21 (11.2%) had a second ablation during the follow-up period. The most common indication for hysterectomy was continued menorrhagia (n = 20). Other indications included fibroids (n = 2) and bleeding with unclear reason (n = 1). None had cancer as an indication. In two women the indications for hysterectomy were not known. The largest proportion of women requiring hysterectomy [18.8% (6/32)], was those with a retroverted uterus, followed by those with an axial uterus [14% (6/42)] and the least affected [11.7% (13/111)] were women with an anteverted uterus. None of the women who had repeat ablation underwent hysterectomy.
Of the 21 repeat ablations among the participants, seven were for women who had spotting only, 11 were carried out in women with light periods and three in women who had a normal menstrual pattern after their initial procedure. Seven repeat ablations were carried out in each of the uterine position group, hence the proportion of repeat ablations was highest [27% (7/26)] in the retroverted position, compared with 19% (7/36) in the axial uterine group and 7% (7/97) in the anteverted group. Of those who had a second ablation, 13 women were amenorrhoeic at the time of the follow-up and seven continued to bleed, four of whom had light bleeding, one was bleeding normally, two continued to have heavy loss, and in one the severity of bleeding was unknown. The impact of uterine position on the probability of not having re-ablation in non-hysterectomized women was 77% for the retroverted uterus group, 83% for axial position and 94% for women with anteverted uterus.
The impact of patient- and treatment-dependent variables on achieving a successful outcome (i.e. avoidance of hysterectomy) in the participants group is depicted in Table III. There was no statistical relationship between the age at initial procedure, pretreatment dysmenorrhea score and balloon pressure drop during the operation between those who had a hysterectomy and those who remained hysterectomy free. However, there was a significant difference in the post-treatment dysmenorrhoea score, with more frequent dysmenorrhoea among those who underwent hysterectomy (P < 0.00001). There was no significant relationship between uterine position and post treatment dysmenorrhoea score (P = 0.73). Statistical analysis using a 2-test indicated that there was most likely a statistical relationship between uterine position and outcome. However, it is essential to remember that the design of this study did not enable us to conclude that there is a relationship between cause and effect.
|
Life-table analysis
At 46 years after the first endometrial ablation, the probability of avoiding hysterectomy was 86% of all women [95% confidence interval (CI) 8190.5%]. Similarly, in women who had not had hysterectomy, the probability of avoiding re-ablation was 88% (95% CI 83.492.8%). The overall probability of avoiding any form of surgery (hysterectomy or re-ablation) was 75% of all women (95% CI 69.481.2%) (Figure 1). The probability of avoiding further surgery (hysterectomy or ablation) varied according to uterine position and was the lowest in the retroverted uterus group (61%), increasing to 69% for the axial position, and was highest (80%) in women with an anteverted uterus (Figure 1). At 46 years follow-up, the probability of amenorrhoea in women who had not had hysterectomy was 52.1% (95% CI 43.152.7%) (Figure 2).
|
|
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Earlier reservations concerning success rates of the second-generation devices stipulated that techniques performed under direct vision are more likely to treat all endometrial areas (OConnor and Magos, 1996). The probability of not having hysterectomy 5 years after initial ablation was 91% with conventional endometrial resection (OConnor and Magos, 1996
) and 79% following endometrial laser ablation (Phillips et al., 1998
). Similarly, the probability of not having any surgery in these two studies was 80 and 75% respectively. A randomized trial of endometrial ablation versus hysterectomy for the treatment of dysfunctional uterine bleeding reported the probability of further surgical treatment by any method after 4 years at 36% and by hysterectomy at 24% in women randomized to ablation (Aberdeen Endometrial Ablation Trials Group, 1999
). A randomized study between another second generation device, microwave endometrial ablation (MEA), and TCRE reported that both techniques achieved high satisfaction and acceptability rates and both improved quality of life after 1 year (Cooper et al., 1999
). The same authors reported that at 2 years follow-up, the two modalities continued to show highly significant reduction in bleeding and pain scores, but significantly more women were satisfied after MEA, and with its impact on menstrual loss, than TCRE (Bain et al., 2002
). Hysterectomy rates were equally similar between the two groups (11.6 and 12.7%, respectively).
In a previous report a number of factors were found to be associated with improved menstrual loss after treatment (Amso et al., 1998). In that study, with the exception of uterine position and post ablation dysmenorrhoea score at short-term follow-up, no other patient or treatment-dependant variable was found to influence the outcome. The statistical independence of uterine position and post-treatment dysmenorrhoea score indicate that each of them may independently influence the long-term outcome. It is likely that the position of the heater element within the uterine cavity, especially in women with retroverted uteri, has a significant impact on outcome as judged by menstrual flow or the need for further surgery (Watermeyer and Amso, 2002
).
The choice between hysterectomy or repeat ablation may be determined by the perceived severity of symptoms and their impact on womens life-style and expectations balanced by their desire to avoid major surgery.
In women who became amenorrhoeic it was difficult to exclude natural menopause without endocrine studies. At the time of follow-up, 30 women were receiving HRT for peri- or post-menopausal symptoms, yet only seven women were menstruating. The nature of the HRT preparation was not known in all women, hence it was not possible to determine its relative contribution to the menstrual status.
This study was not designed to evaluate the economic differences between UBT and hysterectomy. Studies that compared the costs of TCRE with hysterectomy reported that for patients with abnormal uterine bleeding treated with TCRE, there was a significantly shorter hospital stay and quicker return to normal daily activity (Sculpher et al., 1996; Hidlebaugh, 2000
). Data from a randomized clinical trial indicated that at 2 years follow-up, TCRE costs (including cost of reintervention) were only 71% of the costs of abdominal hysterectomy (Sculpher, 1998
).
The role of any surgical procedure must balance its benefits and risks. The surgical nature of thermal balloon ablation, even though it is associated with minimal complications, means that medical therapy will most likely remain the first choice. However, a recent randomized study suggested that immediate TCRE is more effective than medical treatment and does not result in an increase in hysterectomies (Cooper et al., 2001). Additionally, clinicians must be aware that there is an as yet undetermined risk of this surgery on the early warning signs of endometrial hyperplasia or cancer in menopausal women who have previously undergone ablation. However, the high rate of avoidance of hysterectomy over a long period of time means that UBT offers women who are reluctant to undergo hysterectomy the dual advantages of relatively conservative surgery and freedom from prolonged medical therapy.
![]() |
Acknowledgements |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Amso, N.N., Stabinsky, S.A., McFaul, P., Blanc, B., Pendley, L. and Neuwirth, R. (1998) Uterine thermal balloon therapy for the treatment of menorrhagia: the first 300 patients from a multi-centre study. International collaborative uterine thermal balloon working group. Br. J. Obstet. Gynaecol., 105, 517523.[ISI][Medline]
Bain, C., Cooper, K.G. and Parkin, D.E. (2002) Microwave endometrial ablation versus endometrial resection: A randomised controlled trial. Obstet. Gynecol., 99, 983987.
Bongers, M.Y., Mol, B.W.J., Dijkhuizen, F.P.H. and Brölmann, H.A.M. (2000) Is balloon ablation as effective as endometrial electroresection in the the treatment of menorrhagia? J. Laparoendosc. Adv. Surg. Tech., 10, 8592.[ISI]
Cooper, K.G., Bain, C. and Parkin, D.E. (1999) Comparison of microwave endometrial ablation and transcervical resection of the endometrium for treatment of heavy menstrual loss: a randomised trial. Lancet, 354, 18591863.[CrossRef][ISI][Medline]
Cooper, K.G., Jack, S.A., Parkin, D.E. and Grant, A.M. (2001) Five-year follow-up of women randomised to medical management or transcervical resection of the endometrium for heavy menstrual loss: clinical and quality of life outcomes. Br. J. Obstet. Gynaecol., 108, 12221228.[CrossRef][ISI]
Cox, R.D. and Oaks, D. (1984) Analysis of survival data. Chapman and Hall, London, UK.
Gervaise, A., Fernandez, H., Capella-Allouc, S., Taylor, S., La Vieille, S., Hamou, J. and Gomel, V. (1999) Thermal balloon ablation versus endometrial resection for the treatment of abnormal uterine bleeding. Hum. Reprod., 14, 27432747.
Grainger, D.A., Tjaden, B.L., Rowland, C. and Meyer, W.R. (2000) Thermal balloon and rollerball ablation to treat menorrhagia: two-year results of a multicentre, prospective, randomized, clinical trial. J. Am. Assoc. Gynecol. Laparosc., 7, 175179.[ISI][Medline]
Greenwood, M.A. (1926) Report on the natural duration of cancer, appendix 1: the errors of sampling of the survivorship tables. Reports on public health and medical subjects. Number 33, App. 1. Stationery Office, London, UK.
Hidlebaugh, D.A. (2000) Cost and quality-of-life issues with different surgical therapies for the treatment of abnormal uterine bleeding. Obstet. Gynecol. Clin. North. Am., 27, 451465.[ISI][Medline]
Loffer, F.D. (2001) Three-year comparison of thermal balloon and rollerball ablation in treatment of menorrhagia. J. Am. Assoc. Gynecol. Laparosc., 8, 4854.[ISI][Medline]
Meyer, W.R., Walsh, B.W., Grainger, D.A., Peacock, L.M., Loffer, F.D. and Steege, J.F. (1998) Thermal balloon and rollerball ablation to treat menorrhagia: a multicenter comparison. Obstet. Gynecol., 92, 98103.
Neuwirth, R.S., Duran, A.A., Singer, A., MacDonald, R. and Bolduc, L. (1994) The endometrial ablator: A new instrument. Obstet. Gynecol., 83, 792796.[Abstract]
OConnor, H. and Magos, A. (1996) Endometrial resection for the treatment of menorrhagia. N. Engl. J. Med., 335, 151156.
Phillips, G., Chien P.F.W. and Garry. R. (1998) Risk of hysterectomy after 1000 consecutive endometrial laser ablations. Br. J. Obstet. Gynecol., 105, 897903.[ISI][Medline]
Sculpher, M. (1998) The cost-effectiveness of preference-based treatment allocation: the case of hysterectomy versus endometrial resection in the treatment of menorrhagia. Health Econ., 7, 129142.[CrossRef][ISI][Medline]
Sculpher, M.J., Dwyer, N., Byford, S. and Stirrat, G.M. (1996) Randomized trial comparing hysterectomy and transcervical endometrial resection: effect on health related quality of life and costs two years after surgery. Br. J. Obstet. Gynaecol., 103, 142149.[ISI][Medline]
Singer, A., Almanza, R., Gutierrez, A., Haber, G., Bolduc, C.R. and Neuwirth, R., (1994) Preliminary clinical experience with a thermal balloon endometrial ablation method to treat menorrhagia. Obstet. Gynecol., 84, 732734.
Vilos, G.A., Fortin, C.A., Sanders, B.A., Pendley, L. and Stabinsky, S.A. (1997) Clinical trial of the Uterine Thermal Balloon for treatment of menorrhagia. J. Am. Assoc. Gynecol. Laparosc., 4, 559565.[Medline]
Watermeyer, S.R. and Amso, N.N. (2002) Observational case reports of intra and post operative ultrasound monitoring of uterine balloon ablation. J. Obstet. Gynecol., 22, 432435.[CrossRef]
Submitted on March 8, 2002; resubmitted on October 28, 2002; accepted on January 16, 2003.