Under the Norwegian law on termination of pregnancy, women have the right to termination of pregnancy on request within the first 12 weeks of gestation, dating from the last menstrual period.1 According to the Norwegian Ministry of Health and Social Affairs 13,500–15,000 terminations are performed each year in Norway.2,3 They are all performed in a hospital setting completely free of charge. In 2005, 23.2% of all legally induced terminations were performed at 9 to 12 weeks of gestation.3 Until 1998, all first-trimester abortions in Norway were achieved surgically by vacuum aspiration. The combination of mifepristone and misoprostol has been available in Norway since 1998. Over the last decade, medical abortion with the combination of mifepristone and misoprostol has become a widely used method for pregnancy termination both for early first-trimester4 abortions and second-trimester abortions.5 In 2006, 46.5% of all abortions up to 63 days of gestation were achieved medically.6 For late first-trimester abortions (performed between day 63 and 90 of gestation) the method used in most countries has been exclusively surgically, although Ashok et al7 in 1998 reported medical abortion with mifepristone and misoprostol as an effective, safe, and acceptable treatment alternative also for this group of patients. Except for studies from the Aberdeen group,7–9 only two other groups have presented results with the treatment method for this gestational age.10,11 All have a success rate of approximately 90–95%. There are other studies that use misoprostol alone for abortions with the gestational length 9–13 weeks.12,13 With a success rate of approximately 85% for misoprostol alone, it is lower than for the combination regime with mifepriston and misoprostol.
Medical abortion at 63 to 90 days of gestation was introduced at the Department of Obstetrics and Gynecology, Haukeland University Hospital, Bergen, Norway, in October 2005. The objective of this study was to evaluate medical abortion as a treatment alternative for late first-trimester abortions and to evaluate the decrease in β-hCG after abortion at 63–90 days of gestation.
MATERIALS AND METHODS
Participants for this case–control study (N=254) were selected on the basis of the woman requesting and being offered a medical abortion (under the terms of the Norwegian Termination of Pregnancy Law) with a documented pregnancy length between 63 and 90 days at the Department of Obstetrics and Gynecology, Haukeland University Hospital clinic between October 2005 and April 2007. Women were excluded if they had adrenal and respiratory diseases, in situ intrauterine devices, or a nonviable pregnancy. Oral and written information about the methods were given, and all the patients chose by informed consent to undergo a medical termination. The Regional Committee for Medical and Health Research Ethics, Western Norway approved the study (number 2009/738).
Four clinical visits were scheduled. At the first visit, gestational age was determined and choice of method was decided. Gestational age was defined by vaginal ultrasound measurement of fetal crown-rump length (CRL). Norwegian law defines the upper limit for termination on request to CRL 66 mm.1 Patients with a CRL length of 24–66 mm and a gestational length from 63–90 days were included in the study.14 At the second consultation, mifepristone 200 mg was administered orally, and β-hCG (referred to as 1hCG), and serum-hemoglobin (s-Hb) tests were taken. Rhesus negative–women received Rhesus Immune globulin. The women were hospitalized as outpatients 48 hours later. They self-administered 800 micrograms misoprostol vaginally. Paracetalmol 800 mg with 60 mg codeinsulphate and 50 mg diclofenac were administrated rectally as analgesia and was repeated every 4 hours if required. The opioid ketobemidone 5 mg was administrated intramuscularly as an alternative.
All women were observed for pain, vaginal bleeding, and expulsion of the fetus, pregnancy products, or both. Pain and bleeding were recorded as none/low, moderate, or strong. If products of conception were not passed within 4 hours, further doses of 400 micrograms of misoprostol were given orally at 3-hour intervals, up to a maximum of five doses. The termination was confirmed by visual inspection of products of conception by the nursing staff. Vaginal ultrasonography was performed in cases of severe bleeding, ongoing pregnancy, infection, or missed abortion. The induction-to-abortion interval was defined as the time interval in hours from administration of misoprostol to passing the products of conception and was routinely recorded. The women were observed in the ward for 2 more hours after completion.
To assess treatment, a fourth consultation between the 8th and the 14th day after intake of mifepristone was scheduled. Clinical examination and ultrasonography as well as s-Hb and β-hCG (referred to as 2hCG) tests were performed. Complications, bleeding, and pain from discharge to follow-up were assessed. Acceptability of the procedure was measured by a standard questionnaire that the women filled out at the follow-up visit: I. Are you satisfied with the method? II. Would you choose the method again? III. Would you recommend the method to others? Answering alternatives for all questions were yes, no, or unsure.
β-hCG was measured with a commercial two-site chemiluminescent immunometric assay kit (Immulite 2000 HCG; Siemens Medical Solutions Diagnostics, Los Angeles, CA). The assay has a detection limit of 3 international units/L.
Main results are presented in medians and range. Multivariable logistic regression was carried out with the use of opiate as the dependent variable and former parity and gestational age as the independent variable. Linear regression was carried out with the induction-to-abortion time as the dependent variable and with gestational age at termination as the independent variable. Statistical significance was defined using a probability value of <.05. Statistical analysis was carried out using the Statistical Package for Social Sciences (SPSS, Chicago, IL).
During the implementation period, 254 women requesting termination of pregnancy in late first trimester underwent a medical procedure with mifepristone and misoprostol in our department and volunteered to participate in this study. One woman did not have a blood sample drawn for the 1hCG determination. Her pregnancy was confirmed by clinical and ultrasound examination at the first visit. The other 253 pregnancies were identified both with the use of ultrasound examination and β-hCG determinations. Two hundred eight participants (81.9%, 95% confidence interval [CI] 77.9–85.9%) returned for the planned follow-up consultation or had the 2hCG sample drawn or both. Through hospital medical files or phone consultations, follow-up information on a further 29 patients (11.4%, 95% CI 8.1–14.7%) was recorded. Despite repetitive attempts to contact them, the remaining 17 women (6.7%, 95% CI 4.1–9.3%) were lost to follow-up.
The 254 women represent approximately 55% of the total number of late first-trimester terminations at our clinic during the study period. Patient characteristics are shown in Table 1.
Overall complete abortion rate with no complications was 91.7% (95% CI 88.9–94.6%). Ultrasound examination was performed on the ward because of insecurity regarding completeness of the procedure in 15 women (7.6%, CI 3.9–11.3%), and one of them required surgical aspiration. In total 21 women (8.3%, 95% CI 4.9–11.7%) required surgical intervention (Table 2). Two of them were because of incomplete procedure and six were performed later than 3 days after discharge. Two women had an ongoing pregnancy, and one of those women had not completed the procedure.
Induction-to-abortion time for different gestational lengths are shown in Table 3 and Figure 1. The median number of misoprostol applications required was two, equal to 1,200 micrograms of misoprostol. Two women aborted after mifepristone alone while they were at home but were given the first dose of misoprostol to complete the procedure. A total of 85 women (33.5%, 95% CI 27.7–39.3%) aborted after only one dose of misoprostol while 228 (89.8%, 95% CI 86.0–93.5%) had aborted after three doses of misoprostol. The median induction-to-abortion time was 4.5 hours, and 100 women (43.9%, 95% CI 37.4–50.3%) had aborted within 4 hours and 214 (93.9%, 95% CI 90.7–97.0%) within 8 hours. The linear regression analysis showed no significant association between the induction-to-abortion interval and gestational length (P=.5).
A follow-up consultation was scheduled 8–14 days after the actual termination; however, the consultations took place 7–69 days (median 17) after termination because of various practical and personal reasons, and 187 women (73.6%, 95% CI 68.2–79.0%) came for clinical follow-up. An ultrasound examination of the uterine content was made. No ongoing pregnancies or gestational sacks were detected. The median thickness of the uterine content was 6 (range 1–22) mm. Only one woman had an intrauterine content above 20 mm.
Before discharge from the clinic, pain was rated in 214 women (84.3%, 95% CI 79.8–88.7) by a nurse asking the woman to rate her level of pain. Paracetalmol/codeinsulphate and diclofenac were administrated rectally as analgesia to all the women; 116 (54.2%, 95% CI 47.5–60.9%) reported that they had moderate pain, and 85 (39.7%, 95% CI 33.2–46.3%) had strong pain. Eighty-two women (38.9%, 95% CI 32.2–46.3%) required opiates. More nulliparous than multiparous women required opiates also when gestational length was added as a variable, and the multivariable logistic regression analysis showed there was a negative association between number of injections of opioids required with increasing parity (P=.005, odds ratio 0.66, 95% CI 0.49–0.88).
Level of vaginal bleeding during the procedure was recorded for 209 women (82.3%, 95% CI 72.6–87.0%) and after discharge for 184 (72.4%, 95% CI 66.9–77.9) by the women themselves at visit four (Table 4). Seven women needed surgical aspiration because of hemorrhage (Table 2). The level of s-Hb was measured before treatment and at visit four. One woman had a s-Hb below 9 g/dL at follow-up and came for an extra control, but she did not require surgical intervention. The median length of bleeding after the termination was 14 (range 1–40) days.
The initial intention of this study was to measure β-hCG both at the time of administration of mifepristone and 8–14 days after the abortion. The first β-hCG sample was taken in 253 women at the predetermined time. The 2hCG was quantified in 180 women (70.9%, 95% CI 66.2–75.6%) 9–71 days after the administration of mifepristone. The 1hCG levels ranged from 15,468 to 247,640 international units/L, with a median value of 81,125 international units/L. The values of 2hCG varied from less than 3 international units/L to 3,049 international units/L, with a median value of 136.5 international units/L (Fig. 2). The decline in β-hCG-levels from 1hCG to 2hCG was more than 97.5%. The 2hCG values were above 1,000 international units/L in 10 women. These samples were drawn 9–19 days after mifepristone administration. An additional consultation and/or another determination of serum hCG was scheduled for these women. All these abortions were complete. The women's β-hCG values have been analyzed separately and show that their β-hCG declines were more than 97.5%.
Judging by the women's own evaluation, the procedure seemed to be highly acceptable (Table 5). A total of 91.0% women (95% CI 87.5–94.5%) said they were content with the method, 76.1% (95% CI 70.9–81.4%) would opt for the method of treatment if they ever needed a termination of pregnancy again, and 81.9% (95% CI 77.2–86.6%) would recommend the treatment to a friend.
This study of 254 women represents the first report of a method for late first-trimester medical abortion undertaken in Norway. The mifepristone–misoprostol method used was found to be an effective method for termination of pregnancy, had acceptable side effects, and was generally well tolerated. The medical procedure has become the treatment method of choice for late first-trimester abortion at the Department of Obstetrics and Gynecology, Haukeland University Hospital clinic and is included in the Norwegian national guidelines.15
The combined complete abortion rate of 91.7% for the treatment in this study is similar to those reported in other studies in which comparable protocols were used.7,9,10 The minor differences in efficacy may be attributed to the differences in number of misoprostol administrated7,9,10 and the institutions experience with the method. The decision to intervene surgically was related to the experience of the gynecologist on call, as there were no strict criteria regarding the indication for surgical intervention. Two (0.8%) ongoing pregnancies were detected, but in only one of the cases was the treatment procedure fully completed. The continuing pregnancy rate is similar to that of 1.5% reported by Hamoda et al,9 where more than 1,000 patients were treated. The absolute numbers of continuing pregnancies in this study was small, and a comparison of failure rate in relation to gestational age could therefore not be made.
The results for induction-to-abortion time in this study are similar to previous reports.7–9 The complete expulsion rate before administration of prostaglandins, is also in line with former studies.8 Earlier reports have shown the efficacy of late first-trimester medical abortion regimens to be negatively correlated with the gestational age and that induction-to-abortion time was gestational-age dependent.8,9 In contrast, no such associations could be found here. The reason for the latter is presently unknown, but it is probable that our number of patients is not high enough to detect minor differences.
In line with the data presented by Hamoda et al8,9 a positive correlation between gestational age and the risk of surgical evacuation was shown. Earlier studies have reported that the likelihood of requiring surgery significantly increases when receiving more than three doses.9,16 In this study, 72.2% of the women who required the maximum number of misoprostol applications had a vacuum aspiration.
Painful contractions similar to labor or menstrual cramps were reported from almost all the women in the ward. All received prophylactic analgesia as described, and the majority indicated an acceptable level of pain. In this study, the use of analgesics is considerably higher than in other studies.9,10,16 Lalitkumar et al5 state that the analgesic regimens for medical abortion vary widely, for instance, for second-trimester medical abortion. The rationale for the differences is unknown. Possible reasons may be both at the patient or medical personnel levels, in tradition, and/or in information given. Other studies have found that women with a previous live birth were less likely to require opiates.9,10 At the follow-up, some women reported that they had not been prepared for the level of pain. Our experience indicates that the procedure is relatively painful and that it is of great importance to give adequate pain relief. It is important to provide high-quality information to ensure realistic expectation about the pain as this will enhance the acceptance and tolerability to the majority of women.
Bleeding is an expected consequence of the treatment, and seven of the women required a surgical evacuation to control bleeding. None of the patients required a blood transfusion as reported in other studies.4,9,10 The median duration of bleeding after termination was similar to what has been reported for early first-trimester abortions.17
The validation of completion by observing the expelled products of conception seems to have been successful as no ongoing pregnancies were detected after discharge. This study found that β-hCG levels decrease sharply after late first-trimester abortion when the abortion has been successful. This is in concordance with early first-trimester pregnancy termination abortion studies,18,19 but to our knowledge this has never before been published for late first-trimester termination. We chose to determine β-hCG 8–14 days after termination as this was the point of time used in the adapted protocol for combined clinical and ultrasonographic control examination. Additionally it is essential to detect a failed termination as early as possible.
Seventeen women were lost to follow-up. Although the Department of Obstetrics and Gynecology, Haukeland University Hospital is the only delivery unit and abortion clinic in the area, and none of these women have given birth within 9 months after treatment, it cannot be completely determined that they have not had problems that were dealt with elsewhere.
Overall, the acceptability of the treatment was high. These findings are similar to the American study for late first-trimester abortion and other studies on acceptability for early first- and late-trimester terminations.10,20,21 The acceptability in this study is slightly lower than in the American study. We don't have an obvious explanation, but cultural and national differences need to be raised as a possibility.
This trial has confirmed that the combination of mifepristone and misoprostol in repeated doses is effective for late first-trimester termination. The side effects are tolerable, and the majority of women were satisfied with the method and would recommend it to others. Medical abortion is an effective and acceptable method for termination of pregnancy in late first-trimester, it offers a valuable alternative to surgical abortion, and should be offered routinely, thus increasing women's choice.
4. Kulier R, Gulmezoglu AM, Hofmeyr GJ, Cheng LN, Campana A. Medical methods for first trimester abortion. Cochrane Database of Systematic Reviews, 2004. Issue 2. Art. No.: CD002855.
5. Lalitkumar S, Bygdeman M, Gemzell-Danielsson K. Mid-trimester induced abortion: a review. Hum Reprod Update 2007;13:37–52.
6. Health NIoP. Utførte svangerskapsavbrudd til og med 9. uke med bruk av abortfremkallende legemidler 2000–2006. Available at: http://www.fhi.no/dav/e0f5a9543b.doc
. Retrieved May 5, 2008.
7. Ashok PW, Flett GM, Templeton A. Termination of pregnancy at 9–13 weeks' amenorrhoea with mifepristone and misoprostol. Lancet 1998;352:542–3.
8. Hamoda H, Ashok PW, Flett GM, Templeton A. Medical abortion at 64 to 91 days of gestation: a review of 483 consecutive cases. Am J Obstet Gynecol 2003;188:1315–9.
9. Hamoda H, Ashok PW, Flett GM, Templeton A. Medical abortion at 9–13 weeks' gestation: a review of 1076 consecutive cases. Contraception 2005;71:327–32.
10. Bracken H, Ngoc NT, Schaff E, Coyaji K, Ambardekar S, Westheimer E, et al. Mifepristone followed in 24 hours to 48 hours by misoprostol for late first-trimester abortion. Obstet Gynecol 2007;109:895–901.
11. Largeaud M, El Guindi W, Perotti F, Montoya Y, Carles G, Seve B. [Medical termination of pregnancy at 9–14 weeks gestation. Prospective study of 105 cases in Saint-Laurent-du-Maroni (French Guyana)]. J Gynecol Obstet Biol Reprod (Paris) 2004;33:119–24.
12. Carbonell Esteve JL, Varela L, Velazco A, Cabezas E, Tanda R, Sánchez C. Vaginal misoprostol for late first trimester abortion. Contraception 1998;57:329–33.
13. Carbonell JL, Varela L, Velazco A, Tanda R, Sánchez C. Vaginal misoprostol for abortion at 10–13 weeks' gestation. Eur J Contracept Reprod Health Care 1999;4:35–40.
14. Robinson HP, Fleming JE. A critical evaluation of sonar “crown-rump length” measurements. Br J Obstet Gynaecol 1975;82:702–10.
15. Bjørge L, Løkeland M, Oppegaard K. Provosert abort Norsk gynekologisk veileder: Norsk Gynekologisk forening; 2009. Available at: http://www.legeforeningen.no/id/157791.0
. Retrieved January 23, 2010.
16. Ashok PW, Templeton A, Wagaarachchi PT, Flett GM. Midtrimester medical termination of pregnancy: a review of 1002 consecutive cases. Contraception 2004;69:51–8.
17. Christin-Maitre S, Bouchard P, Spitz IM. Medical termination of pregnancy. N Engl J Med 2000;342:946–56.
18. Fiala C, Safar P, Bygdeman M, Gemzell-Danielsson K. Verifying the effectiveness of medical abortion; ultrasound versus hCG testing. Eur J Obstet Gynecol Reprod Biol 2003;109:190–5.
19. Parashar P, Iversen OE, Midboe G, Myking O, Bjorge L. Medical abortion in the first trimester: The use of serum hCG and endometrial thickness as markers of completeness. Eur J Contracept Reprod Health Care 2007:1–6.
20. Berer M. Medical abortion: issues of choice and acceptability. Reprod Health Matters 2005;13:25–34.
© 2010 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.
21. Henshaw RC, Naji SA, Russell IT, Templeton AA. Comparison of medical abortion with surgical vacuum aspiration: women's preferences and acceptability of treatment. BMJ 1993 18;307:714–7.