1 University of Aberdeen and 2 Grampian University Hospitals NHS Trust, Department of Obstetrics and Gynaecology, University of Aberdeen, Aberdeen Maternity Hospital, Foresterhill, Cornhill Road, Aberdeen AB25 2ZD, Scotland, UK.
![]() |
Abstract |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Key words: medical abortion/mifepristone/misoprostol/surgical vacuum aspiration
![]() |
Introduction |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
In the UK, mifepristone in combination with a prostaglandin analogue was licensed for the termination of pregnancy at up to 9 weeks amenorrhoea in 1991, and since 1995 has also been available for termination of pregnancy beyond 13 weeks. At present, surgical methods (vacuum aspiration) are used at 1013 weeks amenorrhoea, and there is little published work on the use of medical methods (Ashok et al., 1998a).
Vacuum aspiration is considered to be safe and effective but has been associated with major morbidity in up to 1% of women and minor morbidity in 10% (Joint Study of the Royal College of General Practitioners and the Royal College of Obstetricians and Gynaecologists, 1985), the major determinants of morbidity being gestational age and the procedure used to terminate the pregnancy. In the first trimester the complication rate is lowest at 78 weeks gestation (0.26 per 100 abortions) and increases progressively to 1.37 per 100 abortions at 13 weeks (Grimes and Cates, 1979
). Vacuum aspiration, usually performed under general anaesthesia, is currently the method of choice at 1013 weeks gestation.
The manufacturer's recommended regimen for early medical abortion comprises mifepristone 600 mg in combination with the prostaglandin analogue gemeprost. A randomized controlled trial comparing medical abortion with vacuum aspiration at gestations up to 9 weeks showed that although both methods were highly acceptable to women, medical abortion was more painful and less effective with advancing gestation (Henshaw et al., 1993,1994a
,b
).
Since then, a number of developments in the drug regimen have occurred based on randomized controlled studies (McKinley et al., 1993; World Health Organization Task Force, 1993
; El-Refaey et al., 1995
) and these led to the development in 1994 of a new medical regimen consisting of a reduced dose of mifepristone (200 mg) and the vaginal use of the prostaglandin analogue misoprostol (El-Refaey and Templeton, 1994
; Ashok et al., 1998b
,1999
). This prostaglandin has a number of advantages over gemeprost in that it is easily stored and transported and of very low cost.
This paper compares the efficacy, early medical sequelae (up to 8 weeks following termination) and acceptability of medical and surgical abortion at 1013 weeks gestation.
![]() |
Materials and methods |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Study participants
The study was approved by the Grampian Research Ethics Committee and was conducted at Aberdeen Royal Infirmary. Women presenting for termination of pregnancy at 1013 weeks amenorrhoea, gestation confirmed by transvaginal ultrasound, and fulfilling the requirements of the 1967 Abortion Act, were eligible for recruitment. They were also required to meet the following criteria: (i) eligible to undergo either surgical vacuum aspiration or medical abortion; (ii) singleton, viable intrauterine pregnancy; and (iii) ultrasonically estimated gestational age of 1013 weeks at the time of termination. The exclusion criteria were: (i) suspected ectopic pregnancy; (ii) chronic adrenal failure; (iii) long term corticosteroid treatment; (iv) haemorrhagic disorder and treatment with anticoagulants; (v) known allergy to mifepristone; (vi) smokers >35 years of age with electrocardiogram abnormalities; and (vii) breast feeding.
Eligible women were given a standardized information sheet describing medical and surgical methods of abortion and were asked if they were willing to be allocated to a method of abortion. Women were also informed of side effects and complications of each method. Women who agreed to be randomized were assigned to a method by opening consecutive sealed opaque envelopes containing a random number generated by computer. The randomization was prepared by the trial statistician using a randomized block design with blocks of 2, 4, 6 and 8, to ensure equal numbers in each group during the study.
The women who declined to be randomized were asked by the study nurse the reasons for their decision. Those not wishing to be randomized because they had a strong preference for a particular treatment were identified. The women willing to participate if they received their preferred treatment option constituted a non-randomized prospective cohort.
A total of 486 women were recruited (400 in the randomized arm) in order to achieve a 90% power of detecting at the 5% significance level a difference of 10% in the level of acceptability between the two termination methods. Eighty-six women were recruited to the preference arm of the trial. A further 93 women eligible to participate declined to take part in either the randomized or preference arm of the study. Analysis was by intention to treat, and women allocated to a method of treatment were attributed to that method for the purpose of analysis, whether or not they underwent their allocated procedure. Five women randomized to the medical group had surgical treatment by choice and three women who were randomized to surgery had medical treatment.
Intervention
The two alternatives were medical or surgical abortion.
Medical termination
Women undergoing medical abortion were asked to attend the gynaecology ward and given 200 mg of mifepristone orally in hospital under nursing supervision. They were then admitted to the ward 3648 h later, when misoprostol 800 µg was administered vaginally. If products of conception were not passed, a further two doses (400 µg) of misoprostol were given either orally or vaginally at 3 h intervals depending on vaginal bleeding. Women who had vaginal bleeding heavier than a normal period were given misoprostol orally. After products of conception were passed, women were observed in the ward for a further 4 h. Following administration of prostaglandin, pulse, blood pressure, temperature and systemic symptoms were monitored hourly. Women were given oral (paracetamol 500 mg plus dihydrocodeine 10 mg) or parenteral (morphine 10 mg) analgesia every 46 h as required.
If products of conception were not passed 34 h after the third dose of prostaglandin, a speculum examination was undertaken and any products of conception in the vagina or cervix were removed. If products of conception were not identified, an ultrasound scan was performed and surgical evacuation undertaken if necessary prior to discharge.
Surgical termination
Standard techniques were used to perform vacuum aspiration, under general anaesthesia. Women were admitted to the day surgery unit and were given 800 µg of misoprostol for cervical priming 3 h pre-operatively.
In all other respects (e.g. counselling, infection screening) there were no differences between the two groups. All women were screened for genital tract infection including Chlamydia trachomatis and treated if necessary.
Women who underwent medical termination were invited to return to the hospital 1421 days after the termination of pregnancy for follow-up. At this visit, if vaginal bleeding had not ceased then an ultrasound scan was done and high vaginal and endocervical swabs taken if indicated. Patients were commenced on antibiotics or curettage performed if necessary. Those undergoing surgical termination of pregnancy were followed up by their referring doctor, usually at ~14 days post-procedure.
Outcome measures
These included efficacy, major medical complications and relatively minor short-term physical symptoms. Patient preferences prior to termination and acceptability following the procedure were also assessed. These were measured by self administered questionnaires prior to the procedure following randomization, prior to discharge and at 23 weeks following termination.
Efficacy and immediate medical complications prior to discharge
The efficacy of the procedure was defined as complete uterine evacuation without the need for a second procedure [subsequent surgical (re)curettage or medical regimen] within 8 weeks of abortion.
Women were asked to complete a modified menstrual distress questionnaire as well as a visual analogue scale for pain, after termination and prior to discharge. The former was used to assess symptoms such as nausea, vomiting, headache, hot flushes, dizziness, tiredness and diarrhoea on a five point scale (none to very severe) (Henshaw et al., 1994c). Analgesia use in hospital was also documented.
Medical sequelae at 2 weeks following the procedure
Complications recorded at the follow-up visit and unscheduled visits to hospital were documented.
A further questionnaire was completed and returned by post 23 weeks following the procedure to assess the following: (i) duration and severity of vaginal bleeding recorded on a menstrual calendar (each day on the menstrual calendar was subdivided into five to indicate the severity of bleeding, none to flooding). The values were added to give the total vaginal bleeding score (TVBS); (ii) a visual analogue scale to assess pain and analgesia use following discharge; (iii) a record of the time taken off work and to return to normal activity.
Medical sequelae at 8 weeks following the procedure
A third questionnaire was completed by the family doctor at 8 weeks following termination and all medical problems attributable to the abortion were documented. Women experiencing more than one problem were classified for analysis according to the primary complaint.
Major and minor complications attributable to the termination up to 8 weeks following the procedure were assessed. Major complications were classified according to categories previously defined (Joint Study of the Royal College of General Practitioners and the Royal College of Obstetricians and Gynaecologists, 1985). The minor complications are listed in Table V
.
|
Preference and acceptability of procedure
Preference for a particular method was included in the questionnaire completed following randomization and prior to the procedure. Acceptability was assessed with regards to preferred future method by means of questions included in the 23 weeks questionnaire, returned by post.
Statistical analysis
The data were entered into a personal computer-held database and analysed using the Statistics Package for Social Sciences program. The principal analysis compared the outcomes in the two arms of the randomized controlled trial. Independent and paired t-tests were used for continuous variables with a normal distribution, and MannWhitney's U-test for ordinal or non-parametric continuous variables. The 2 test or Fisher's Exact test, as appropriate, was used for independent nominal data and McNemar's test for paired data describing dichotomous variables. Confidence intervals (95% CI) were applied where appropriate. Comparisons were made between the randomized women and those entering the preference cohort with regard to their characteristics and outcomes.
![]() |
Results |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
For descriptive purposes, the data in certain sections have been combined for women who underwent a particular method either by choice or chance.
Medical abortion
Of the 203 women allocated to the medical group, 45 (22.2%) experienced some vaginal bleeding following mifepristone administration and prior to the administration of prostaglandin; nine (4.5%) women rated this to be heavier than a normal period. One (0.5%) woman aborted on mifepristone alone, prior to the administration of prostaglandin. The median time interval from administration of mifepristone to prostaglandin administration was 44.00 h (range 33.8348.58).
The median number of doses of prostaglandin required was 2 (range 03). Products of conception were identified by inspection in 192 (94.6%) women. Twenty women (9.9%) also had abortion confirmed on ultrasound scan prior to discharge. Of the 203 women who underwent medical abortion 139 (68.5%) aborted within 6 h of prostaglandin administration. The median induction abortion interval was 5.00 h (range 2.0027.58). Following the third dose of prostaglandin, 57 women (28.1%) did not pass products of conception within 34 h and required a speculum examination.
Surgical abortion
All 242 surgical terminations were carried out under general anaesthesia on a dedicated theatre list. The consultant responsible for patients carried out 159 (65.7%) of the vacuum aspirations. The median time interval between cervical priming and surgical evacuation was 2.42 h (range 0.177.00).
Efficacy and immediate medical complications prior to discharge
There were no significant differences in overall efficacy between medical abortion and vacuum aspiration. Gestation did not influence efficacy in either group (Table II). Of the 11 (5.4%) failures in the medical group who subsequently had surgery, three (1.5%) had a continuing pregnancy, one (0.5%) a missed abortion and seven (3.4%) had incomplete abortions. Surgical evacuation following failed medical abortion was undertaken prior to discharge in 10 (90.9%) women, and within 2 weeks of the procedure in one woman. Products of conception from five women were sent for histological examination and chorionic villi were identified in four of these. Of the five women (2.1%) who required a second procedure in the surgical group, medical treatment was undertaken following a failed attempt to dilate the cervix at vacuum aspiration in one woman (0.4%). Surgical (re)curettage was undertaken in the other four women (1.7%) for incomplete abortion within 4 weeks of the first procedure. Products of conception were identified in all three women where histological examination was requested.
|
Of the 203 women who underwent medical abortion, 61 (30.0%) required no analgesia, 120 (59.1%) requested oral analgesia only, and 22 (10.9%) requested parenteral opiate analgesia. All women who underwent surgical termination were given diclofenac per rectally 100 mg or paracetamol per rectally 1000 mg in theatre following the procedure. However, a further 96 (39.7%) women required additional analgesia with 91 (37.6%) women requesting oral analgesia and 5 (2.1%) requesting parenteral opiates.
Side effects experienced by women on the ward are shown in Table III. Four (2.0%) women who had medical abortion required syntometrine to control bleeding. Two (0.8%) women in the surgical group had a blood loss >500 ml, one (0.4%) woman requiring a blood transfusion.
|
Short-term sequelae 23 weeks following the two methods of termination are shown in Table IV. The duration and total vaginal bleeding score was greater in women randomized to medical abortion. The median (range) time taken to return to normal activity was 2.5 days (014) in women who underwent medical abortion and 2.5 days (020) in those who underwent vacuum aspiration with no significant difference between the two groups. There were no differences in median (range) time taken to return to work, 2.0 days (021) in the medical group and 2.0 days (021) in the surgical group.
|
All (both randomized and preference) women who underwent medical abortion were invited to attend follow-up 2 weeks later in hospital. Of these, 113 (55.7%) agreed to attend for follow-up, of whom 72 (63.7%) attended. The remaining 41 (36.3%) women failed to keep their hospital appointment. The remaining 90 (44.3%) women who declined the initial invitation to attend hospital were followed up by their family doctor. Women undergoing surgical abortion were followed up by the family doctor, and only seven (2.9%) women followed up in hospital for medical reasons.
Of the 72 women who underwent medical abortion and were followed up in hospital, six required oral antibiotics for suspected pelvic infection. All seven women who were followed up in hospital following vacuum aspiration required oral antibiotics for suspected infection.
Medical sequelae at 8 weeks following the procedure
Of the 445 women in the study, 385 (86.5%) agreed to the researchers contacting the family doctor at 8 weeks following the procedure to complete a questionnaire. Information was obtained from family doctors in 365 (94.8%) of these. Overall, 73 (20.0%) women consulted the family doctor with problems related to termination. There was no significant difference in the consultation rates between the medical and surgical group. Seven (4.4%) women who had medical and 17 (8.2%) who had surgical treatment were given oral antibiotic treatment by the family doctor for presumed pelvic infection (Table V).
There were no significant differences in the rates of major complications within 8 weeks of abortion. Two (0.8%) women in the surgical group and one (0.5%) woman in the medical group required i.v. antibiotics for presumed pelvic infection.
In total, 15 of the 445 women in the study had an unscheduled visit to hospital, six (3.0%) in the medical and nine (3.7%) in the surgical group. Of these, one (0.5%) women in the medical group and seven (2.9%) who had vacuum aspiration were re-admitted (P = 0.04).
Of the 41 women who underwent medical abortion but failed to keep their hospital appointment, information was obtained from the family doctor in 33. Thus, in only eight women was no follow-up information obtained.
Preferences and acceptability
Prior to termination and following randomization, 351 (80%) of 441 women who returned the questionnaire had a definite preference for a particular method. Of these, 253 women (72%) preferred the medical method and 98 (28%) preferred vacuum aspiration (P < 0.0001).
Following termination, 118 women who underwent the medical method and 111 women who had vacuum aspiration returned questionnaires, but only 67 (35.6%) women who had the medical method and 96 (53.3%) who underwent surgery answered the question regarding future preferred method. Forty-seven women (70%) who underwent medical termination would opt for the same method in future, and 76 women (79%) who underwent vacuum aspiration would opt for the same method in futurea statistically significant difference between the two groups (P < 0.0001).
![]() |
Discussion |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Although the drop-out rate was higher than expected (short-term follow-up rate with regards to acceptability), which would directly affect the power calculation, the results achieved are believed to be a true reflection of differences between the groups. However, data regarding problems related to termination were obtained from family doctors and in only eight women who underwent medical abortion was no information obtained. All these women had passed products of conception prior to discharge.
Overall, 11.8% of women who underwent medical abortion and 8.3% who underwent surgery were managed as in-patients. The majority of women who required overnight admission had geographical reasons.
Following randomization but prior to termination, nearly 80% of women had a definite preference for a particular method. Of these, the majority preferred the medical method. Of those who underwent medical abortion, 70% would opt for the same method in future while 79% who underwent vacuum aspiration would opt for the same in futurea statistically significant difference. Based on these results, it appears that medical abortion is less acceptable than surgery, although the response was from a small number of women who underwent either procedure (one-third of all women who underwent medical abortion and half of who underwent surgery returned the questionnaire), which highlights difficulties in post-termination surveys.
Medical abortion has proven to be a safe and effective alternative to vacuum aspiration in the early first trimester and for midtrimester termination of pregnancy (Ashok and Templeton, 1999). The efficacy is up to 97.5% overall and at gestations of 79 weeks ~97% (Ashok et al., 1998b
). Even when the efficacy falls to 94% at 9 weeks gestation, acceptability among women remains high, as demonstrated in a previous randomized study (Henshaw et al., 1993
). To demonstrate a significant difference of 23% in efficacy, as found in this study, would have required 1750 women, but the clinical relevance would have been doubtful given the level of acceptability.
There is now increasing experience of medical abortion in the late first and early second trimester of pregnancy in China (Cheng, 1999). Mifepristone in an oral 200 mg dose in combination with vaginal misoprostol of up to three doses has an efficacy of 94.0%. The vaginal administration of misoprostol was more effective than the oral route. In our study we used a combination of vaginal and oral misoprostol with success rates similar to the above study. It has been shown that in the second trimester, provided the first dose of misoprostol is administered vaginally, there is no advantage in the vaginal administration of subsequent doses (El-Refaey and Templeton, 1995
).
Re-admission rates to hospital were lower than in previous studies with only 1.8% of women requiring re-admission (Henshaw et al., 1994c). This may be associated with our current policy of pre-operative screening for pathogenic organisms. However, significantly more women in the surgical group were re-admitted following termination, although there were no significant differences between the two groups in family doctor consultation rates or in the requirement for antibiotics for presumed pelvic infection.
The current care provided to women undergoing medical abortion in the UK includes in-patient hospitalization on a day case basis. Medical abortion using mifepristone and home administration of misoprostol has been shown to be feasible in the USA at gestations up to 63 days (Schaff et al., 2000). Schaff and colleagues showed no difference in efficacy in relation to gestation, with medical abortion being highly acceptable (91%). The feasibility of carrying out the procedure at home needs to be evaluated in different settings and in the context of further studies.
In summary, medical abortion is as safe and effective as vacuum aspiration at 1013 weeks gestation. The introduction of a medical method of abortion at 1013 weeks could have a considerable impact on the provision of medical services, as well as increasing women's choice of methods, particularly for those women who wish to avoid surgery and anaesthesia.
![]() |
Acknowledgements |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
![]() |
Notes |
---|
![]() |
References |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
Ashok, P.W., Flett, G.M. and Templeton, A. (1998a) Termination of pregnancy at 913 weeks amenorrhoea with mifepristone and misoprostol. Lancet, 352, 542543.[ISI][Medline]
Ashok, P.W., Penney, G.C., Flett, G.M. et al. (1998b) An effective regimen for early medical abortion: a report of 2000 consecutive cases. Hum. Reprod., 13, 29622965.
Ashok, P.W. Flett, G.M. and Templeton, A (1999) Effective, low cost regimens for medical termination of pregnancy at all gestations. Contemp. Rev. Obstet. Gynaecol., 207212.
Brewin, C.R. and Bradley, C. (1989) Patient preferences and randomised clinical trials. Br. Med. J., 299, 313315.[ISI][Medline]
Cheng, L. (1999) The Shanghai multicentre randomized comparative trial on termination of 1016 weeks gestation with mifepristone plus misoprostol. Chin. J. Obstet. Gynecol., 5, 268271.
El-Refaey, H. and Templeton, A. (1994) Early induction of abortion by a combination of mifepristone and misoprostol administered by the vaginal route. Contraception, 49, 111114.[ISI][Medline]
El-Refaey, H. and Templeton, A (1995) Induction of abortion in the 2nd trimester by a combination of misoprostol and mifepristonea randomized comparison between two misoprostol regimens. Hum. Reprod., 10, 475478.[Abstract]
El-Refaey, H., Rajasekar, D., Abdalla. M. et al. (1995) Induction of abortion with mifepristone (RU486) and oral or vaginal misoprostol. N. Engl. J. Med., 332, 983987.
Grimes, D.A. and Cates, W. (1979) Complications from legally-induced abortion: a review. Obstet. Gynaecol. Surv., 34, 177191.[Medline]
Henshaw, R.C., Naji, S.A., Russell, I.T. et al. (1993) Comparison of medical abortion with vacuum aspiration; women's preferences and acceptability of treatment. Br. Med. J., 307, 714717.[ISI][Medline]
Henshaw, R.C., Naji, S.A., Russell, I.T. et al. (1994a) A prospective economic evaluation comparing medical abortion (using mifepristone and gemeprost) with surgical vacuum aspiration. Br. J. Fam. Plan., 20, 6468.[ISI]
Henshaw, R.C., Naji, S.A., Russell, I.T. et al. (1994b) Psychological responses following medical abortion (using mifepristone and gemeprost) with surgical vacuum aspiration. Acta Obstet. Gynecol. Scand., 73, 812818.[ISI][Medline]
Henshaw, R.C., Naji, S.A., Russell, I.T. et al. (1994c) A comparison of medical abortion (using mifepristone and gemeprost) with surgical vacuum aspiration: efficacy and early medical sequelae. Hum. Reprod., 9, 21672172.[Abstract]
Information and Statistics Division (2000) Abortion Statistics 1998. In Health Briefing No. 00/01. ISD Publications, Edinburgh.
Joint Study of the Royal College of General Practitioners and the Royal College of Obstetricians and Gynaecologists (1985) Induced abortion operations and their early sequelae. J. R. Coll. Gen. Pract., 35, 175180.[ISI][Medline]
McKinley, C., Thong, K.J. and Baird, D.T (1993) The effect of dose of mifepristone and gestation on the efficacy of medical abortion with mifepristone and misoprostol. Hum. Reprod., 8, 15021505.[Abstract]
Office for National Statistics (1999) Abortion Statistics 1998Series AB no 26. Office for National Statistics Legal Abortion in England and Wales, 1999, HMSO, London 2000.
Schaff, E.A., Fielding, S.L., Eisinger, S.H. et al. (2000) Low-dose mifepristone followed by vaginal misoprostol at 48 hours for abortion up to 63 days.Contraception, 61, 4146.[ISI][Medline]
World Health Organization Task Force (1993) WHO Task Force on Post-ovulatory Methods of Fertility Regulation. Termination of pregnancy with reduced doses of mifepristone. Br. Med. J., 307, 532537.[ISI][Medline]
Submitted on April 27, 2001; accepted on September 18, 2001.