- Email: [email protected]
Mifepristone–misoprostol midtrimester abortion: impact of gestational age on the induction-to-abortion interval
Contraception 81 (2010) 97 – 101
Original research article
Mifepristone–misoprostol midtrimester abortion: impact of gestational age on the induction-to-abortion interval Olga Gómeza,⁎, Aina Borrása , Aintzane Rabanala , Montse Palacioa , Antonia Carcellerb , Oriol Colla , Eduard Gratacósa a
Department of Maternal–Fetal Medicine, ICGON, Hospital Clínic, University of Barcelona and Centre for Biomedical Research on Rare Diseases (CIBERER), 08028 Barcelona, Spain b Department of Pharmacy, ICGON, Hospital Clínic, University of Barcelona and Centre for Biomedical Research on Rare Diseases (CIBERER), 08028 Barcelona, Spain Received 8 September 2009; revised 19 September 2009; accepted 5 October 2009
Abstract Background: This study was conducted to explore the effect of gestational age (GA) on the induction-to-abortion interval of mifepristone– misoprostol midtrimester termination of pregnancy (TOP) regimen. Study Design: This study involved a consecutive series of 270 pregnancies between 12.0 and 22.6 weeks that have undergone legal TOP from April 2006 to June 2009. All women received a single oral dose of 200 mg mifepristone and, 36–48 h later, a course of misoprostol (an initial vaginal dose of 800 mcg plus four oral doses of 400 mcg at 3-hourly intervals). Treatment was considered to be a failure if abortion did not occur within 24 h. The impact of GA, parity and maternal age on the induction-to-abortion interval was assessed by means of Cox regression. Results: Overall, the mean GA at TOP was 18.0 weeks. The mean induction-to-abortion interval was 9.8 h (SD=8.2 h; range=1–50 h), and 246 women (91%) aborted successfully within 24 h. GA at TOP and parity were the only two variables independently associated with the induction-to-abortion interval. The mean induction-to-abortion interval was increased by about 50% in patients undergoing TOP between 20.0 and 22.6 weeks (12.9 h, SD=8.9), as compared with those at 16.0–19.6 weeks (7.8 h, SD=5.9) and 12.0–15.6 weeks (8.2 h, SD=8.3) (pb.001). The effect of parity on the induction-to-abortion interval was more modest, with a 20% increase in induction-to-abortion interval in nulliparous (10.1 h, SD=9.1), as compared with women with a previous live birth (8.1 h, SD=6.7). Conclusions: The mean induction-to-abortion interval increases by 4 h after 20 weeks GA. This information may be relevant for counseling and planning of the procedure. Published by Elsevier Inc. Keywords: Medical abortion; Mifepristone; Misoprostol
1. Introduction Recent advances in prenatal diagnosis have greatly improved the early detection of chromosomal abnormalities and major fetal congenital abnormalities during pregnancy. As a result, the number of indications for termination of pregnancy (TOP) is increasing, and at present, midtrimester TOP constitutes 10–15% of all induced abortions worldwide [1]. Medical second-trimester TOP regimens have been demonstrated to be a safe option and are the methods ⁎ Corresponding author. Department of Maternal–Fetal Medicine, Institut Clinic de Ginecologia, Obstetricia i Neonatologia, Hospital Clínic, 08028 Barcelona, Spain. Tel.: +34 678451555; fax: +34 93 2275605. E-mail address: [email protected] (O. Gómez). 0010-7824/$ – see front matter. Published by Elsevier Inc. doi:10.1016/j.contraception.2009.10.001
of choice to provide a fetal specimen suitable for pathological evaluation. The combination of mifepristone and misoprostol is now a recommended [1,2], safe and highly effective method [1–4]. It is generally accepted that the mean induction-toabortion interval with the mifepristone–misoprostol regimen ranges between 6 and 8 h, and 95–97% of women abort within 24 h after the first dose of misoprostol [3–5]. However, the results of previous clinical series suggest that the induction-to-abortion interval and the incidence of complications increase with gestational age (GA) [5,6]. Presently, the GA above which the induction-to-abortion interval increases has not been clearly established. This may be a relevant aspect for the planning of TOP and parents' counseling.
98
O. Gómez et al. / Contraception 81 (2010) 97–101
In this study, we analyzed the impact of GA at TOP, parity, maternal age, previous cesarean section and previous TOP on the outcome of a mifepristone–misoprostol regimen for induced abortion between 12.0 and 22.6 weeks.
2. Patients and methods A retrospective study was performed on a consecutive series of 270 women requesting TOP between 12.0 and 22.6 weeks GA according to current Spanish legislation. The GA was determined based on ultrasound measurements in all cases (crown–rump length between 11 and 14 weeks [7] and biparietal diameter between 14 and 22 weeks [8]). Women with previous cesarean section were not excluded. The study protocol was approved by the Institutional Ethical Committee, and all cases signed a written consent form that included feticide with potassium chloride prior to mifepristone administration in the group of TOP between GA of 22 and 22.6 weeks [9]. According to our protocol, each woman received a single oral dose of 200 mg mifepristone. Women were admitted to the hospital 36–48 h later, and four tablets of misoprostol (a total dose of 800 mcg) were inserted into the posterior vaginal fornix by a staff member. Three hours after the first dose of misoprostol, 400-mcg doses were orally administered at 3-hourly intervals, to a maximum of four doses. The treatment was considered to be a failure if abortion did not occur within 24 h after the first dose of misoprostol. In these cases, a second oral dose of 200 mg mifepristone, plus a Dilapan® cervical dilator, was administered 3 h after the last dose of misoprostol and women were allowed to rest for 12 h before repeating a second course of vaginal misoprostol. If any patient failed to abort with this regimen, intra-amniotic F2α and/or cervical prostalglandine E2 was used as a second-line therapy. Parenteral analgesia (1 g paracetamol, 50 mg metamizol plus 5 mg methadone sc if necessary) was provided at 4- to 6-hourly intervals. Epidural was also offered if required. Surgical evacuation of the uterus was carried out in cases of clinically significant bleeding or suspicion of retained products of conception. Ultrasound of the uterine cavity was not routinely performed. The optimal interval to allow the spontaneous passage of the placenta after fetal expulsion was not uniformly standarized. Oxytocin (5 IU) was given intravenously in all cases after passage of the placenta or at the end of the surgical curettage. A database was set up to collect the clinical and outcome data of all the women during the study period (maternal age, parity, GA at TOP, indication for TOP, existence of a previous TOP, induction-to-abortion interval, rate of surgical curettage and maternal complications). The induction-toabortion interval was the period between the first dose of misoprostol and the fetal expulsion. Data were analyzed using the SPSS 13.1 for Windows Statistical Package. The dependent variable induction-to-abortion interval was analyzed by Cox regression. Covariates included in the model
were as follows: maternal age, GA at TOP, existence of previous live birth, previous TOP and previous cesarean section. Kaplan–Meier survival analysis was used to compare (using the Log rank test) the cumulative abortion rates in relation to the independent variables. A receiveroperating characteristic (ROC) curve was constructed to determine the best cutoff of GA in predicting successful vaginal delivery within 12 h. Quantitative variables were analyzed by Student's t test and one-way ANOVA with polynomial linear contrast test as appropriate. The χ2 test, linear-by-linear trend or Fisher's Exact Test was used to analyze the nominal data.
3. Results A total of 270 women underwent medical abortion between 12.0 and 22.6 weeks GA from April 2006 to June 2009. The demographic and clinical characteristics of the included population are shown in Table 1. Overall, the mean GA at TOP was 18.0 weeks (SD=3.43 weeks) and 53% of cases (144/270) corresponded to abortions of more than 17 weeks of gestation. One hundred twenty-one women (44.8%) had had a previous live birth, 24 women (8.9%) had had a previous TOP and 28 women (10.4%) had had a previous cesarean section. The most common indications for TOP in our center were major fetal malformations (122 cases, 45.2%) and chromosomal abnormalities (89 cases, 33%). The mean induction-to-abortion interval in our population was 9.8 h (SD=8.2 h; range=1–50 h). None of the women aborted or were admitted for bleeding or pain prior to misoprostol administration. Abortion was successful within 24 h after the first dose of misoprostol in 246 women (91.1%), and only 4 women (1.5%) failed to abort within 48 h and required a second-line therapy. Fig. 1 illustrates the survival function interpreted as the percentage of women that had not aborted at each time point. As illustrated in the chart, 25%, 50% and 75% of women aborted 4.3, 6.4 and 12.0 h following the first dose of misoprostol, respectively. GA at TOP (pb.001) and parity (pb.003) were the only two variables independently associated with the inductionTable 1 Demographic and clinical characteristics of the study population (N=270) Age (years) GA at TOP (weeks) 12.0–15.6 16.0–19.6 20.0–22.6 Previous live birth Nulliparous 12.0–15.6 16.0–19.6 20.0–22.6 Previous TOP Previous cesarean section Results are expressed as mean (SD) or n (%) as appropriate.
32.8 (5.44) 18.0 (3.34) 102 (38) 81 (30) 87 (32) 121 (44.8) 149 (55.2) 60 (40.3) 41 (27.5) 48 (32.2) 24 (8.9) 28 (10.4)
O. Gómez et al. / Contraception 81 (2010) 97–101
Fig. 1. Cumulative percentage of women who aborted in relation to the induction-to-abortion interval.
to-abortion interval in our population assessed by Cox regression. GA at TOP was the strongest predictive variable. An ROC curve analysis determined that a cutoff of 20 weeks GA optimized the prediction of abortion within 12 h. In pregnancies at or beyond 20 weeks at TOP, induction-toabortion interval was increased by about 50% when compared with earlier gestations, and the proportion of cases delivering before 12 h was reduced by 20% (Table 2). In order to explore the impact of GA at earlier pregnancy stages, cases below 20 weeks were arbitrarily divided into two further subgroups (12–15.6 and 16–19.6 weeks), but no significant differences were observed (Table 2). The impact of parity on the evaluated parameters was more modest. Thus, patients with at least one previous live birth had a 20% of reduction in the induction-to-abortion interval and an 8% increase in the proportion of cases delivering within 12 h. Uterine curettage was required in 63 cases (23.3%). As shown in Table 2, the proportion of patients requiring a
99
surgical curettage progressively decreased throughout gestation duration (pb.001). No cases of uterine rupture occurred in the group of patients with a previous cesarean section. Serious complications were reported in three women: a case of uterine atony, which was resolved with surgical curettage plus oxytocin infusion, a uterine perforation occurring during a surgical curettage and an episode of transient coronary artery spasm in a primiparous woman at 17 weeks GA. This later complication was reported in a 32year-old woman with no previous medical antecedents. Two hours after the first dose of misoprostol, spontaneous amniorrhexis was evidenced and the patient referred uterine contractions. Parenteral analgesia (50 mg metamizol plus 5 mg methadone sc) was provided according to our protocol. At this moment, she developed an acute thoracic pain associated with vomiting and hypotension, as well as a decreased level of consciousness. Electrocardiogram and blood laboratory serial test confirmed the diagnosis of acute coronary artery spasm possibly in relation to misoprostol administration. Treatment with ephedrine and nitroglycerin was administered, and the patient spontaneously recovered consciousness; the thoracic pain progressively disappeared. Curettage was then performed for uterine evacuation. A lowdose perfusion of nitroglycerin was maintained the following 12 h. Troponine I showed no significant increase when determined at 6-h intervals. Angio-CT showed no signs of pulmonary tromboembolism. The patient was discharged 2 days later.
4. Discussion Our study confirms previous studies suggesting that mifepristone–misoprostol is a safe and highly effective regimen for second-trimester TOP. A high proportion of women (91%) aborted successfully within 24 h and 75% aborted 12 h after the first dose of misoprostol. The study provides evidence that both GA at TOP and parity were
Table 2 Induction-to-abortion interval [in hours, mean (SD)], proportion of cases requiring surgical curettage and proportion of women who aborted within 12 and 24 h in relation to GA at TOP and parity
GA (weeks) 12–15.6 16–19.6 20–22.6 Parity Nulliparous Previous live birth
Abortion b12 h
Abortion b24 h
Induction-to-abortion interval
Uterine curettage
8.16 (8.3)⁎ 7.8 (5.9)⁎ 12.9 (8.9⁎)
38/102 (37.3%)⁎⁎ 15/81 (18.5%)⁎⁎ 10/87 (11.5%)⁎⁎
83/102 (81.4%)⁎⁎ 66/81 (81.5%)⁎⁎ 51/87 (59%)⁎⁎
94/102 (92.2%)⁎⁎⁎ 78/81 (96.3%)⁎⁎⁎ 74/87 (85.1%)⁎⁎⁎
10.1 (9.1)† 8.1 (6.7)†
36/149 (24.2%)†† 27/121 (22.3%)††
105/149 (70.5%)††† 95/121 (78.5%)†††
131/149 (87.9%)†††† 115/121 (95%)††††
⁎ pb.001, ANOVA with polynomial linear contrast. ⁎⁎ pb.001, linear-by-linear trend. ⁎⁎⁎ p=.105, linear-by-linear trend. † pb.008, Student's t test. †† p=.721, χ2 test. ††† p=.134, χ2 test. †††† pb.04, χ2 test.
100
O. Gómez et al. / Contraception 81 (2010) 97–101
independently associated with the induction-to-abortion interval, with GA having the strongest effect. The mean induction-to-abortion interval increased by about 4 h after 20 weeks GA, and the proportion of patients delivering within 12 h decreased by 20%. Parity had a lesser impact on the clinical outcome. The mean induction-to-abortion interval was about 2 h longer in nulliparous women, with a very modest reduction in the number of patients delivering within 12 h. The present study is one of the few published large series of TOP between GA 17.0 and 22.6 weeks managed with a mifepristone–misoprostol regimen [6,10–12]. The mean GA at TOP and the percentage of pregnancies of more than 17 weeks were higher in our study compared with those of others. Ashok et al. [6] reported the largest series of 999 cases to date. The mean induction-to-abortion interval was shorter in their study (6.25 h vs. 9.8 h in the present series). However, the mean GA at TOP and the proportion of gestations of more than 17 weeks were lower than ours (15 vs. 18 weeks and 25% vs. 32%, respectively). Again, the mean induction-toabortion interval reported by Ashok in the group of gestations of more than 17 weeks was shorter than ours (8.7 h vs. 11.7 h). Nevertheless, the proportion of gestations of more than 20 weeks, which corresponded to the group with the highest interval in our study, was not reported. The main, novel finding of this study is that GA has a strong influence in the duration of the induction-to-abortion interval. To our knowledge, the GA above which the induction-to-abortion interval increases has not been previously reported. An ROC curve analysis determined that a cutoff of 20 weeks best discriminated cases with a remarkable difference in the duration of the interval. Although timing of TOP is largely determined by the GA at diagnosis, this information may be relevant at the time of planning the procedure. As an additional finding, the present study confirms that the lower the GA at TOP, the higher the chance of incomplete abortion requiring curettage, as previously reported in other clinical series [6,12–14]. In the present study, a GA at TOP of less than 16 weeks was associated with a 2.5-fold increase in the risk of surgical curettage. The reason for this association has not been investigated. It should be noted that the rate of surgical evacuation was high in our study. It was performed in 22% of women, whereas previous studies have reported rates between 5% and 11% [3,4]. This high rate may reflect our local practice and experience. Before April 2006, we used ultrasound criteria to define complete abortion and surgical curettage was performed in 47% of cases. At present, the diagnosis of RPC is based on clinical criteria, and consequently, the rate of surgical abortion decreased by 50% [15]. However, we have not yet standardized the optimal interval time to allow the spontaneous passage of the placenta after fetal expulsion, and probably, many of the women who underwent surgical curettage would have benefited from a more expectant management.
Finally, the findings of the present study support the concept that misoprostol–mifepristone is a safe regimen for TOP during the second trimester of pregnancy. No cases of uterine rupture occurred in the group of 28 patients with a previous cesarean section, and serious complications were reported in only three women. We report a case in which misoprostol appeared to be the cause of a coronary artery spasm in a young woman without previous cardiovascular disease. Although misoprostol has been reported to be a safe drug for medical TOP in women with cardiac disorders [16], further data are required to determine whether this case represents an isolated incident or reflects an infrequent but potentially adverse effect associated with misoprostol. In conclusion, a regimen of mifepristone–misoprostol was a safe and highly effective method for second-trimester TOP in our population. The study confirmed the hypothesis that increased GA at TOP was associated with a longer induction interval and a lower proportion of patients delivering within 12 h. In this study, 20 weeks GA was the best cutoff to discriminate this risk.
References [1] Medical methods for termination of pregnancy. WHO Technical Report Series 871. Geneva 1997. [2] Gynaecology. RCoOa. The care of women requesting induced abortion. Evidence-based Clinical Guideline No. 7. The Royal College of Obstetricians and Gynaecology. September 2004. [3] Gemzell-Danielsson K, Lalitkumar S. Second trimester medical abortion with mifepristone–misoprostol and misoprostol alone: a review of methods and management. Reprod Health Matters 2008;16:162–72. [4] Lalitkumar S, Bygdeman M, Gemzell-Danielsson K. Mid-trimester induced abortion: a review. Hum Reprod Update 2007;13:37–52. [5] Ho PC, Blumenthal BP, Gemzell-Danielsson K, Gómez Ponce de León R, Mittal S, Tang O. Misoprostol for the termination of pregnancy with a live fetus at 13 to 26 weeks. Intern J Gynaecol Obstet 2007;99: S178–81. [6] Ashok PW, Templeton A, Wagaarachchi PT, Flett GM. Midtrimester medical termination of pregnancy: a review of 1002 consecutive cases. Contraception 2004;69:51–8. [7] Robinson HP, Fleming JE. A critical evaluation of sonar “crown– rump length” measurements. Br J Obstet Gynaecol 1975;82: 702–10. [8] Mul T, Mongelli M, Gardosi J. A comparative analysis of secondtrimester ultrasound dating formulae in pregnancies conceived with artificial reproductive techniques. Ultrasound Obstet Gynecol 1996;8:397–402. [9] Silva LV, Cecatti JG, Pinto e Silva JL, Amaral E, Barini R. Feticide does not modify duration of labor induction in cases of medical termination of pregnancy. Fetal Diagn Ther 2008;23:192–7. [10] Ashok PW, Templeton A. Nonsurgical mid-trimester termination of pregnancy: a review of 500 consecutive cases. Br J Obstet Gynaecol 1999;106:706–10. [11] Goh SE, Thong KJ. Induction of second trimester abortion (12–20 weeks) with mifepristone and misoprostol: a review of 386 consecutive cases. Contraception 2006;73:516–9. [12] Kapp N, Borgatta L, Stubblefield P, Vragovic O, Moreno N. Mifepristone in second-trimester medical abortion: a randomized controlled trial. Obstet Gynecol 2007;110:1304–10.
O. Gómez et al. / Contraception 81 (2010) 97–101 [13] Lo TK, Lau WL, Lai FK, et al. The effect of gestational age on the outcome of second-trimester termination of pregnancies for foetal abnormalities. Prenat Diagn 2008;28:508–11. [14] Nilas L, Glavind-Kristensen M, Vejborg T, Knudsen UB. One or two day mifepristone–misoprostol interval for second trimester abortion. Acta Obstet Gynecol Scand 2007;86:1117–21.
101
[15] Guix C, Palacio M, Figueras F, et al. Efficacy of two regimens of misoprostol for early second-trimester pregnancy termination. Fetal Diagn Ther 2005;20:544–8. [16] Bagga R, Choudhary N, Suri V, et al. First and second trimester induced abortions in women with cardiac disorders: a 12-year analysis from a developing country. J Obstet Gynaecol 2008;28:732–7.
Original research article
Mifepristone–misoprostol midtrimester abortion: impact of gestational age on the induction-to-abortion interval Olga Gómeza,⁎, Aina Borrása , Aintzane Rabanala , Montse Palacioa , Antonia Carcellerb , Oriol Colla , Eduard Gratacósa a
Department of Maternal–Fetal Medicine, ICGON, Hospital Clínic, University of Barcelona and Centre for Biomedical Research on Rare Diseases (CIBERER), 08028 Barcelona, Spain b Department of Pharmacy, ICGON, Hospital Clínic, University of Barcelona and Centre for Biomedical Research on Rare Diseases (CIBERER), 08028 Barcelona, Spain Received 8 September 2009; revised 19 September 2009; accepted 5 October 2009
Abstract Background: This study was conducted to explore the effect of gestational age (GA) on the induction-to-abortion interval of mifepristone– misoprostol midtrimester termination of pregnancy (TOP) regimen. Study Design: This study involved a consecutive series of 270 pregnancies between 12.0 and 22.6 weeks that have undergone legal TOP from April 2006 to June 2009. All women received a single oral dose of 200 mg mifepristone and, 36–48 h later, a course of misoprostol (an initial vaginal dose of 800 mcg plus four oral doses of 400 mcg at 3-hourly intervals). Treatment was considered to be a failure if abortion did not occur within 24 h. The impact of GA, parity and maternal age on the induction-to-abortion interval was assessed by means of Cox regression. Results: Overall, the mean GA at TOP was 18.0 weeks. The mean induction-to-abortion interval was 9.8 h (SD=8.2 h; range=1–50 h), and 246 women (91%) aborted successfully within 24 h. GA at TOP and parity were the only two variables independently associated with the induction-to-abortion interval. The mean induction-to-abortion interval was increased by about 50% in patients undergoing TOP between 20.0 and 22.6 weeks (12.9 h, SD=8.9), as compared with those at 16.0–19.6 weeks (7.8 h, SD=5.9) and 12.0–15.6 weeks (8.2 h, SD=8.3) (pb.001). The effect of parity on the induction-to-abortion interval was more modest, with a 20% increase in induction-to-abortion interval in nulliparous (10.1 h, SD=9.1), as compared with women with a previous live birth (8.1 h, SD=6.7). Conclusions: The mean induction-to-abortion interval increases by 4 h after 20 weeks GA. This information may be relevant for counseling and planning of the procedure. Published by Elsevier Inc. Keywords: Medical abortion; Mifepristone; Misoprostol
1. Introduction Recent advances in prenatal diagnosis have greatly improved the early detection of chromosomal abnormalities and major fetal congenital abnormalities during pregnancy. As a result, the number of indications for termination of pregnancy (TOP) is increasing, and at present, midtrimester TOP constitutes 10–15% of all induced abortions worldwide [1]. Medical second-trimester TOP regimens have been demonstrated to be a safe option and are the methods ⁎ Corresponding author. Department of Maternal–Fetal Medicine, Institut Clinic de Ginecologia, Obstetricia i Neonatologia, Hospital Clínic, 08028 Barcelona, Spain. Tel.: +34 678451555; fax: +34 93 2275605. E-mail address: [email protected] (O. Gómez). 0010-7824/$ – see front matter. Published by Elsevier Inc. doi:10.1016/j.contraception.2009.10.001
of choice to provide a fetal specimen suitable for pathological evaluation. The combination of mifepristone and misoprostol is now a recommended [1,2], safe and highly effective method [1–4]. It is generally accepted that the mean induction-toabortion interval with the mifepristone–misoprostol regimen ranges between 6 and 8 h, and 95–97% of women abort within 24 h after the first dose of misoprostol [3–5]. However, the results of previous clinical series suggest that the induction-to-abortion interval and the incidence of complications increase with gestational age (GA) [5,6]. Presently, the GA above which the induction-to-abortion interval increases has not been clearly established. This may be a relevant aspect for the planning of TOP and parents' counseling.
98
O. Gómez et al. / Contraception 81 (2010) 97–101
In this study, we analyzed the impact of GA at TOP, parity, maternal age, previous cesarean section and previous TOP on the outcome of a mifepristone–misoprostol regimen for induced abortion between 12.0 and 22.6 weeks.
2. Patients and methods A retrospective study was performed on a consecutive series of 270 women requesting TOP between 12.0 and 22.6 weeks GA according to current Spanish legislation. The GA was determined based on ultrasound measurements in all cases (crown–rump length between 11 and 14 weeks [7] and biparietal diameter between 14 and 22 weeks [8]). Women with previous cesarean section were not excluded. The study protocol was approved by the Institutional Ethical Committee, and all cases signed a written consent form that included feticide with potassium chloride prior to mifepristone administration in the group of TOP between GA of 22 and 22.6 weeks [9]. According to our protocol, each woman received a single oral dose of 200 mg mifepristone. Women were admitted to the hospital 36–48 h later, and four tablets of misoprostol (a total dose of 800 mcg) were inserted into the posterior vaginal fornix by a staff member. Three hours after the first dose of misoprostol, 400-mcg doses were orally administered at 3-hourly intervals, to a maximum of four doses. The treatment was considered to be a failure if abortion did not occur within 24 h after the first dose of misoprostol. In these cases, a second oral dose of 200 mg mifepristone, plus a Dilapan® cervical dilator, was administered 3 h after the last dose of misoprostol and women were allowed to rest for 12 h before repeating a second course of vaginal misoprostol. If any patient failed to abort with this regimen, intra-amniotic F2α and/or cervical prostalglandine E2 was used as a second-line therapy. Parenteral analgesia (1 g paracetamol, 50 mg metamizol plus 5 mg methadone sc if necessary) was provided at 4- to 6-hourly intervals. Epidural was also offered if required. Surgical evacuation of the uterus was carried out in cases of clinically significant bleeding or suspicion of retained products of conception. Ultrasound of the uterine cavity was not routinely performed. The optimal interval to allow the spontaneous passage of the placenta after fetal expulsion was not uniformly standarized. Oxytocin (5 IU) was given intravenously in all cases after passage of the placenta or at the end of the surgical curettage. A database was set up to collect the clinical and outcome data of all the women during the study period (maternal age, parity, GA at TOP, indication for TOP, existence of a previous TOP, induction-to-abortion interval, rate of surgical curettage and maternal complications). The induction-toabortion interval was the period between the first dose of misoprostol and the fetal expulsion. Data were analyzed using the SPSS 13.1 for Windows Statistical Package. The dependent variable induction-to-abortion interval was analyzed by Cox regression. Covariates included in the model
were as follows: maternal age, GA at TOP, existence of previous live birth, previous TOP and previous cesarean section. Kaplan–Meier survival analysis was used to compare (using the Log rank test) the cumulative abortion rates in relation to the independent variables. A receiveroperating characteristic (ROC) curve was constructed to determine the best cutoff of GA in predicting successful vaginal delivery within 12 h. Quantitative variables were analyzed by Student's t test and one-way ANOVA with polynomial linear contrast test as appropriate. The χ2 test, linear-by-linear trend or Fisher's Exact Test was used to analyze the nominal data.
3. Results A total of 270 women underwent medical abortion between 12.0 and 22.6 weeks GA from April 2006 to June 2009. The demographic and clinical characteristics of the included population are shown in Table 1. Overall, the mean GA at TOP was 18.0 weeks (SD=3.43 weeks) and 53% of cases (144/270) corresponded to abortions of more than 17 weeks of gestation. One hundred twenty-one women (44.8%) had had a previous live birth, 24 women (8.9%) had had a previous TOP and 28 women (10.4%) had had a previous cesarean section. The most common indications for TOP in our center were major fetal malformations (122 cases, 45.2%) and chromosomal abnormalities (89 cases, 33%). The mean induction-to-abortion interval in our population was 9.8 h (SD=8.2 h; range=1–50 h). None of the women aborted or were admitted for bleeding or pain prior to misoprostol administration. Abortion was successful within 24 h after the first dose of misoprostol in 246 women (91.1%), and only 4 women (1.5%) failed to abort within 48 h and required a second-line therapy. Fig. 1 illustrates the survival function interpreted as the percentage of women that had not aborted at each time point. As illustrated in the chart, 25%, 50% and 75% of women aborted 4.3, 6.4 and 12.0 h following the first dose of misoprostol, respectively. GA at TOP (pb.001) and parity (pb.003) were the only two variables independently associated with the inductionTable 1 Demographic and clinical characteristics of the study population (N=270) Age (years) GA at TOP (weeks) 12.0–15.6 16.0–19.6 20.0–22.6 Previous live birth Nulliparous 12.0–15.6 16.0–19.6 20.0–22.6 Previous TOP Previous cesarean section Results are expressed as mean (SD) or n (%) as appropriate.
32.8 (5.44) 18.0 (3.34) 102 (38) 81 (30) 87 (32) 121 (44.8) 149 (55.2) 60 (40.3) 41 (27.5) 48 (32.2) 24 (8.9) 28 (10.4)
O. Gómez et al. / Contraception 81 (2010) 97–101
Fig. 1. Cumulative percentage of women who aborted in relation to the induction-to-abortion interval.
to-abortion interval in our population assessed by Cox regression. GA at TOP was the strongest predictive variable. An ROC curve analysis determined that a cutoff of 20 weeks GA optimized the prediction of abortion within 12 h. In pregnancies at or beyond 20 weeks at TOP, induction-toabortion interval was increased by about 50% when compared with earlier gestations, and the proportion of cases delivering before 12 h was reduced by 20% (Table 2). In order to explore the impact of GA at earlier pregnancy stages, cases below 20 weeks were arbitrarily divided into two further subgroups (12–15.6 and 16–19.6 weeks), but no significant differences were observed (Table 2). The impact of parity on the evaluated parameters was more modest. Thus, patients with at least one previous live birth had a 20% of reduction in the induction-to-abortion interval and an 8% increase in the proportion of cases delivering within 12 h. Uterine curettage was required in 63 cases (23.3%). As shown in Table 2, the proportion of patients requiring a
99
surgical curettage progressively decreased throughout gestation duration (pb.001). No cases of uterine rupture occurred in the group of patients with a previous cesarean section. Serious complications were reported in three women: a case of uterine atony, which was resolved with surgical curettage plus oxytocin infusion, a uterine perforation occurring during a surgical curettage and an episode of transient coronary artery spasm in a primiparous woman at 17 weeks GA. This later complication was reported in a 32year-old woman with no previous medical antecedents. Two hours after the first dose of misoprostol, spontaneous amniorrhexis was evidenced and the patient referred uterine contractions. Parenteral analgesia (50 mg metamizol plus 5 mg methadone sc) was provided according to our protocol. At this moment, she developed an acute thoracic pain associated with vomiting and hypotension, as well as a decreased level of consciousness. Electrocardiogram and blood laboratory serial test confirmed the diagnosis of acute coronary artery spasm possibly in relation to misoprostol administration. Treatment with ephedrine and nitroglycerin was administered, and the patient spontaneously recovered consciousness; the thoracic pain progressively disappeared. Curettage was then performed for uterine evacuation. A lowdose perfusion of nitroglycerin was maintained the following 12 h. Troponine I showed no significant increase when determined at 6-h intervals. Angio-CT showed no signs of pulmonary tromboembolism. The patient was discharged 2 days later.
4. Discussion Our study confirms previous studies suggesting that mifepristone–misoprostol is a safe and highly effective regimen for second-trimester TOP. A high proportion of women (91%) aborted successfully within 24 h and 75% aborted 12 h after the first dose of misoprostol. The study provides evidence that both GA at TOP and parity were
Table 2 Induction-to-abortion interval [in hours, mean (SD)], proportion of cases requiring surgical curettage and proportion of women who aborted within 12 and 24 h in relation to GA at TOP and parity
GA (weeks) 12–15.6 16–19.6 20–22.6 Parity Nulliparous Previous live birth
Abortion b12 h
Abortion b24 h
Induction-to-abortion interval
Uterine curettage
8.16 (8.3)⁎ 7.8 (5.9)⁎ 12.9 (8.9⁎)
38/102 (37.3%)⁎⁎ 15/81 (18.5%)⁎⁎ 10/87 (11.5%)⁎⁎
83/102 (81.4%)⁎⁎ 66/81 (81.5%)⁎⁎ 51/87 (59%)⁎⁎
94/102 (92.2%)⁎⁎⁎ 78/81 (96.3%)⁎⁎⁎ 74/87 (85.1%)⁎⁎⁎
10.1 (9.1)† 8.1 (6.7)†
36/149 (24.2%)†† 27/121 (22.3%)††
105/149 (70.5%)††† 95/121 (78.5%)†††
131/149 (87.9%)†††† 115/121 (95%)††††
⁎ pb.001, ANOVA with polynomial linear contrast. ⁎⁎ pb.001, linear-by-linear trend. ⁎⁎⁎ p=.105, linear-by-linear trend. † pb.008, Student's t test. †† p=.721, χ2 test. ††† p=.134, χ2 test. †††† pb.04, χ2 test.
100
O. Gómez et al. / Contraception 81 (2010) 97–101
independently associated with the induction-to-abortion interval, with GA having the strongest effect. The mean induction-to-abortion interval increased by about 4 h after 20 weeks GA, and the proportion of patients delivering within 12 h decreased by 20%. Parity had a lesser impact on the clinical outcome. The mean induction-to-abortion interval was about 2 h longer in nulliparous women, with a very modest reduction in the number of patients delivering within 12 h. The present study is one of the few published large series of TOP between GA 17.0 and 22.6 weeks managed with a mifepristone–misoprostol regimen [6,10–12]. The mean GA at TOP and the percentage of pregnancies of more than 17 weeks were higher in our study compared with those of others. Ashok et al. [6] reported the largest series of 999 cases to date. The mean induction-to-abortion interval was shorter in their study (6.25 h vs. 9.8 h in the present series). However, the mean GA at TOP and the proportion of gestations of more than 17 weeks were lower than ours (15 vs. 18 weeks and 25% vs. 32%, respectively). Again, the mean induction-toabortion interval reported by Ashok in the group of gestations of more than 17 weeks was shorter than ours (8.7 h vs. 11.7 h). Nevertheless, the proportion of gestations of more than 20 weeks, which corresponded to the group with the highest interval in our study, was not reported. The main, novel finding of this study is that GA has a strong influence in the duration of the induction-to-abortion interval. To our knowledge, the GA above which the induction-to-abortion interval increases has not been previously reported. An ROC curve analysis determined that a cutoff of 20 weeks best discriminated cases with a remarkable difference in the duration of the interval. Although timing of TOP is largely determined by the GA at diagnosis, this information may be relevant at the time of planning the procedure. As an additional finding, the present study confirms that the lower the GA at TOP, the higher the chance of incomplete abortion requiring curettage, as previously reported in other clinical series [6,12–14]. In the present study, a GA at TOP of less than 16 weeks was associated with a 2.5-fold increase in the risk of surgical curettage. The reason for this association has not been investigated. It should be noted that the rate of surgical evacuation was high in our study. It was performed in 22% of women, whereas previous studies have reported rates between 5% and 11% [3,4]. This high rate may reflect our local practice and experience. Before April 2006, we used ultrasound criteria to define complete abortion and surgical curettage was performed in 47% of cases. At present, the diagnosis of RPC is based on clinical criteria, and consequently, the rate of surgical abortion decreased by 50% [15]. However, we have not yet standardized the optimal interval time to allow the spontaneous passage of the placenta after fetal expulsion, and probably, many of the women who underwent surgical curettage would have benefited from a more expectant management.
Finally, the findings of the present study support the concept that misoprostol–mifepristone is a safe regimen for TOP during the second trimester of pregnancy. No cases of uterine rupture occurred in the group of 28 patients with a previous cesarean section, and serious complications were reported in only three women. We report a case in which misoprostol appeared to be the cause of a coronary artery spasm in a young woman without previous cardiovascular disease. Although misoprostol has been reported to be a safe drug for medical TOP in women with cardiac disorders [16], further data are required to determine whether this case represents an isolated incident or reflects an infrequent but potentially adverse effect associated with misoprostol. In conclusion, a regimen of mifepristone–misoprostol was a safe and highly effective method for second-trimester TOP in our population. The study confirmed the hypothesis that increased GA at TOP was associated with a longer induction interval and a lower proportion of patients delivering within 12 h. In this study, 20 weeks GA was the best cutoff to discriminate this risk.
References [1] Medical methods for termination of pregnancy. WHO Technical Report Series 871. Geneva 1997. [2] Gynaecology. RCoOa. The care of women requesting induced abortion. Evidence-based Clinical Guideline No. 7. The Royal College of Obstetricians and Gynaecology. September 2004. [3] Gemzell-Danielsson K, Lalitkumar S. Second trimester medical abortion with mifepristone–misoprostol and misoprostol alone: a review of methods and management. Reprod Health Matters 2008;16:162–72. [4] Lalitkumar S, Bygdeman M, Gemzell-Danielsson K. Mid-trimester induced abortion: a review. Hum Reprod Update 2007;13:37–52. [5] Ho PC, Blumenthal BP, Gemzell-Danielsson K, Gómez Ponce de León R, Mittal S, Tang O. Misoprostol for the termination of pregnancy with a live fetus at 13 to 26 weeks. Intern J Gynaecol Obstet 2007;99: S178–81. [6] Ashok PW, Templeton A, Wagaarachchi PT, Flett GM. Midtrimester medical termination of pregnancy: a review of 1002 consecutive cases. Contraception 2004;69:51–8. [7] Robinson HP, Fleming JE. A critical evaluation of sonar “crown– rump length” measurements. Br J Obstet Gynaecol 1975;82: 702–10. [8] Mul T, Mongelli M, Gardosi J. A comparative analysis of secondtrimester ultrasound dating formulae in pregnancies conceived with artificial reproductive techniques. Ultrasound Obstet Gynecol 1996;8:397–402. [9] Silva LV, Cecatti JG, Pinto e Silva JL, Amaral E, Barini R. Feticide does not modify duration of labor induction in cases of medical termination of pregnancy. Fetal Diagn Ther 2008;23:192–7. [10] Ashok PW, Templeton A. Nonsurgical mid-trimester termination of pregnancy: a review of 500 consecutive cases. Br J Obstet Gynaecol 1999;106:706–10. [11] Goh SE, Thong KJ. Induction of second trimester abortion (12–20 weeks) with mifepristone and misoprostol: a review of 386 consecutive cases. Contraception 2006;73:516–9. [12] Kapp N, Borgatta L, Stubblefield P, Vragovic O, Moreno N. Mifepristone in second-trimester medical abortion: a randomized controlled trial. Obstet Gynecol 2007;110:1304–10.
O. Gómez et al. / Contraception 81 (2010) 97–101 [13] Lo TK, Lau WL, Lai FK, et al. The effect of gestational age on the outcome of second-trimester termination of pregnancies for foetal abnormalities. Prenat Diagn 2008;28:508–11. [14] Nilas L, Glavind-Kristensen M, Vejborg T, Knudsen UB. One or two day mifepristone–misoprostol interval for second trimester abortion. Acta Obstet Gynecol Scand 2007;86:1117–21.
101
[15] Guix C, Palacio M, Figueras F, et al. Efficacy of two regimens of misoprostol for early second-trimester pregnancy termination. Fetal Diagn Ther 2005;20:544–8. [16] Bagga R, Choudhary N, Suri V, et al. First and second trimester induced abortions in women with cardiac disorders: a 12-year analysis from a developing country. J Obstet Gynaecol 2008;28:732–7.