The main insights gained from this study are as follows: 1) Misoprostol as a single agent for pregnancy termination in the second trimester is highly effective, with a low failure rate of 6.3%. 2) Misoprostol was significantly higher effective in pregnancies with a dead fetus in utero than that those with a live fetus. The higher success rate was more obvious in the first 12 hours of induction. 3) Side effects of misoprostol were comparable in the two groups but the group of pregnancies with a live fetus had higher rate of oxytocin requirement. 4) Fetal viability, fetal weight or gestational age and an initial Bishop score were significantly related with total amount of misoprostol dosage used for induction. While fetal viability and fetal weight / gestational age were still an independent factor of total dosage use on multiple regression analysis, Bishop score was not significantly, possibly due to too low frequency of the cases with score of 4. The new insights gained from this study are probably helpful in counseling and predicting the successful outcomes. Also, we provide evidence-based insights to support a revisit on current practice in second trimester termination of pregnancy.
Our finding of higher effectiveness in the group of dead fetuses was consistent with that in the preliminary study by Srisomboon et al [7] who compare the induction-to-delivery time between the groups receiving intracervical misoprostol 200 mcg every 12 hours, though different in route, dosage and drug administration interval. Likewise, side effects found in the two groups were not significant different, similar to the findings in the study comparing complications of second- trimester medical termination of pregnancy for fetal anomalies with pregnancy with intrauterine fetal demise [8].
In evaluation of independent impact factors on the effectiveness of misoprostol, we preferred to assess total amount of misoprostol dosage required for successful termination rather than induction to fetal delivery time. This is due to the fact that, induction-to-delivery time is associated with multiple factors, including higher misoprostol dosage itself which certainly shortens the active phase of labor, oxytocin use and variations of labor curve. Actually, the main purpose of misoprostol is to prime the cervix to become favorable rather than stimulate uterine contraction, though spontaneous contractions commonly occur after misoprostol use.
Theoretically, the effectiveness of misoprostol is expected to be higher in pregnancies with fetal death than those with viable fetuses and also in pregnancies in late gestation than those in early gestation. However, the data supporting such a belief is very limited, leading to different recommendations among various guidelines. For examples, WHO guideline [5] recommends buccal, intravaginal or sublingual 400 mcg every 3 hours for pregnancy termination at 12–28 weeks of gestation, while FIGO guideline [9] recommends 400 mcg every 3 hours for 13–24 weeks and 200 mcg every 4 hours for 25–28 weeks. Additionally, several guidelines recommend the same regimens regardless of fetal viability. Nevertheless, our findings suggest that for misoprostol administration, fetal viability and gestational age should be taken into account. As noted above, even in our practice in the past, we used the same regimen for second trimester termination from 14–28 weeks of gestation, regardless of fetal viability and Bishop score (0–4). Note that, in our practice, vaginal misoprostol 400 mcg every 6 hours, different from 3-hour interval as recommended by WHO guideline, also is highly effective in both groups. Importantly, our findings support the FIGO guideline [9], which is based on both fetal viability and gestational age. For example, the FIGO guideline recommends a lower dosage to be 200 mcg for fetal dead during 13–26 weeks and 100 mcg during 27–28 weeks, and higher dosage for pregnancies with a live fetus. Actually, the recommended dosages by the FIGO have been decided on the basis of limited evidence and expert opinion, new evidence is regularly emerging and thus there is a need to review and revise the recommendations in the future [9]. In other words, pregnancy termination, based on our findings, the same regimen for both pregnancies with dead fetuses and live fetuses and for all gestational age (14–28 weeks), as recommended by several guidelines including that of our own practice, is no longer justified.
Note that all pregnancies with viable fetuses were sacrificed, though gestational age of greater than 24 weeks of which we usually make attempts to resuscitate for the highest chance of survival, because they were associated with severe perinatal morbidity. Accordingly, in termination of pregnancy after 24 weeks with expectation of live birth, the misoprostol regimen should be considered differently, lower dosage or longer interval of administration, and intrapartum surveillance must be incorporated to avoid non-reassuring fetal heart rate associated with misoprostol.
The limitations of this study include: 1) retrospective nature, though based on prospective database with high reliability; 2) Though the sample size was adequate to address the primary objective, it might be too low for multivariate analysis of some potentials confounders to show their significant effect, for example low frequency of cases with Bishop score of 4. 3) The results represent only the regimen of 400 mcg every 6 hour, which may not reproducible for other regimens. The strengths of this study are as follows: 1) high homogeneity of drug administration in both groups, the same regimen in terms of route, dosage, and interval; 2) comprehensive analysis to compare the effectiveness based on time-event analysis and multiple regression analysis; 3) The findings could represent the effectiveness in actual practice because of being conducted on service setting, not research setting.
Future studies concerning termination of pregnancy should focus on incorporation of all of the potential predictors to develop models, probably derived from machine learning, in predicting the chance of successful termination, including misoprostol dosage, route, interval, fetal viability, initial Bishop score, gestational age / estimated fetal weight, etc., to guide the best regimen for an individual rather than one regimen for all.
In conclusion, the regimen of vaginal misoprostol 400 mcg every 6 hours is highly effective for second trimester termination with the overall success rate of more than 90%. The regimen is significantly higher effective in pregnancies with a dead fetus. Also, the effectiveness is significantly associated with birth weight / gestational age, fetal viability and initial Bishop score. Our findings suggest that the regimen of misoprostol as a single agent for second trimester termination should be modified, depending on fetal viability and gestational age.