Second‐trimester deliveries are frequently incomplete and associated with retained placenta and hemorrhage.1–3 If the placenta fails to deliver spontaneously, exploration of the uterus manually or with surgical instruments is necessary. These procedures may require general anesthesia along with the added risks of uterine perforation, hemorrhage, infection, and synechia.4,5 A noninvasive, inexpensive method to expel the placenta or reduce complications while waiting for spontaneous placental expulsion in second‐trimester deliveries would be beneficial.
An earlier study3 concluded that serial 15‐methyl prostaglandin (PG) F2α injections at doses of 250 μg every 20 minutes reduced the length of the third stage of labor in second‐trimester deliveries by half. This regimen, however, is expensive, requires increased nursing and pharmacy attention, is only available for parenteral administration, and cannot be used in women with a history of asthma.
Misoprostol is a PGE1 analogue that has uterotonic properties at all gestational ages, is inexpensive, and is easy to administer. Recent studies6,7 confirm misoprostol's efficacy to decrease blood loss in term third stage of labor patients, and rectal administration has been used to treat a retained placenta in a second‐trimester delivery.8
The purpose of this study was to determine whether serial oral misoprostol shortens the third stage of labor in second‐trimester pregnancy loss.
MATERIALS AND METHODS
Study subjects included women admitted for spontaneous or medically induced pregnancy terminations between 13 and 28 weeks' gestation from December 1, 1997, to January 1, 2002. Approval was obtained from the Orlando Regional Healthcare Institutional Review Committee; all subjects were at least 18 years old. Exclusion criteria included hypersensitivity to the drug. Previous low transverse segment cesarean deliveries and twin gestations were not excluded. The period from induction to delivery was defined as either the time from the start of an oxytocic to delivery of the fetus (in twins, the last fetus) or, in women with spontaneous labor, the time from the onset of contractions to fetal delivery. The length of the third stage of labor was defined as the time from delivery of the fetus until delivery of the placenta (in twins, from the last fetus to the last placenta), either by spontaneous passage, manual removal, or curettage. Manual placental removal included removal requiring ring forceps. Hemorrhage was defined as an estimate of blood loss exceeding 500 mL, decrease in hemoglobin concentration of at least 2 g/dL, or the need for transfusion. Post‐termination hemoglobin was measured after placental delivery and before discharge, but at no particular predesignated time. Fever was defined as a temperature of at least 100.4F occurring 24 hours or more after pregnancy termination. Antibiotics were not given routinely.
An informed consent was obtained on admission before the women received any medication. Subjects were randomized by means of opaque sealed envelopes containing computer‐generated random numbers to receive either look‐alike placebo or misoprostol. Randomization and drug preparation were performed by the hospital pharmacy, and the medication was on the floor at the time of fetal delivery. Both subjects and clinicians were blinded to group assignment. Pregnancies were terminated using intravenous oxytocin,9 or misoprostol (200 μg intravaginally increasing by 200 μg every 6 hours until 800 μg was reached, then 800 μg was continued every 6 hours).
After delivery of the fetus (the second fetus in twins), each woman was given intravenous oxytocin (40 U in 1000 mL of lactated Ringer solution at a rate of 125 mL per hour) and was observed for spontaneous passage of the placenta. Tension on the cord was limited to general traction. If the placenta had not delivered spontaneously by 10 minutes, the patient took the study drug. Women randomized to the control group received a look‐alike placebo, and those randomized to the study group received 200 μg of misoprostol. Both medications were administered orally. If the placenta was not delivered spontaneously within 1 hour of the first oral dose, the medication was repeated, and if it had not delivered within 1 hour of the second oral dose, the medication was repeated again. None of the subjects received more than three doses in total (including the original dose). Subjects were examined vaginally every hour for 6 hours, and the placenta was removed if possible. If significant vaginal bleeding was noted during the third stage, surgical intervention was performed at the managing physician's discretion. Abdominal ultrasound was used to assess the endometrial cavity when clinically indicated, ie, when there was suspicion of incomplete placental delivery or heavy bleeding. If the placenta was undelivered after 6 hours, curettage was scheduled. In this group, the time to placental removal was dependent primarily on operating room availability rather than clinical indications. Continuous data were expressed as mean ± standard deviation. χ2 test was used to compare the number of twins between groups, the indications for termination, methods of pregnancy termination, number of women not receiving medication, and to compare the number of patients who completed the third stage within 2 hours. Student t test was used to compare estimated gestational age. Sample size calculations were based on the assumption that 48%3 of the control group and 68% of the misoprostol group would deliver the placenta within 2 hours. Using a type I error of 0.05, a type II error of 0.2, and a 20% difference in successful completion of the third stage of labor within 2 hours, a sample of 104 subjects in each treatment group was required. We enrolled 248 patients to account for possible protocol deviations and withdrawals. The analysis was on intention to treat all randomized patients.
Two hundred forty‐eight women were enrolled on admission, but 11 were not randomized because of technical error, leaving 237 for randomization (Figure 1). One hundred eighteen women were assigned to misoprostol and 119 to placebo. Twin gestations represented five (4%) and three (3%) in the misoprostol and placebo groups, respectively (P = .46). The two groups were comparable in all measured demographic and obstetric characteristics including mean maternal age, gravidity, race, mean estimated gestational age (19.7 ± 2.9 versus 19.0 ± 2.8 weeks in the misoprostol and placebo groups, respectively), previous cesarean deliveries, and mean maternal weight at admission. There were no significant differences between the misoprostol and control groups, respectively, in the indications for delivery: fetal death (36 versus 36), preterm premature ruptured membranes (33 versus 34), fetal anomaly (27 versus 26), advanced cervical dilatation (18 versus 19), and maternal indication (four versus four). In addition, there were no significant differences between the misoprostol and placebo groups in the methods of pregnancy termination: misoprostol (111 versus 113), spontaneous (six versus five), and oxytocin (one versus one).
Fifty‐eight (49%) and 55 (46%) of the misoprostol and placebo groups, respectively, did not receive their medication (P = .65, χ2) (Table 1). The major induction and outcome variables were similar between the misoprostol and control groups (Table 2). The placenta was out within 2 hours of fetal delivery in 67 (57%) and 62 (52%) of the misoprostol and placebo groups, respectively (P = .47). There was no difference in the measured outcomes in patients who received medication (Table 3). Complications are listed in Table 4. There was no difference in the frequency of hemorrhage, need for transfusion, or curettage between the groups when tested hour by hour during the 6‐hour observation period. Figure 2 shows the cumulative percentage of patients who received medication who delivered over the 6‐hour study period (P = not significant for all tested measurements, χ2). Figure 3 shows the cumulative percentage of placentas delivered in all subjects (P = not significant for all tested times, χ2).
Retained placenta occurs more commonly in second‐trimester deliveries and in women whose pregnancies are terminated with PG.1,10 The frequency of complications and morbidity appears to increase as the third stage of labor lengthens.11 Thus, any method that would shorten the third stage of labor or reduce complications while waiting for spontaneous placental expulsion would be helpful.
Although many publications have addressed the different methods to induce labor in a second‐trimester pregnancy, very little prospective data are available on management options of the third stage of labor in second‐trimester deliveries. Indeed, neither the definition of prolonged third stage of labor nor the incidence of morbidity from prolonged third stage of labor is consistent among publications. Prolonged third stage of labor and retained placenta in second‐trimester deliveries has been defined as the time from fetus to spontaneous placental expulsion exceeding 30 minutes,11 1 hour,12 2 hours,13,14 and 3 hours.15 Using standard intravenous oxytocin doses, the retained placenta rates have varied from 8%13 to 83%15 and hemorrhage rates from 0%15 to 6%.14 However, the studies reporting on these complications were not designed to prospectively evaluate management of the third stage of labor in second‐trimester deliveries.
Bivins et al1 published one of the first prospective studies designed exclusively to evaluate the management of the third stage of labor in second‐trimester deliveries and concluded that an injection of high‐dose oxytocin into the umbilical vein did not shorten the third stage of labor or decrease the need for surgical intervention. Retained placenta was defined as exceeding 1 hour, and the incidence of operative removal was not significant at 33% and 29% in the oxytocin and placebo groups, respectively.
A more recent study, also designed specifically to address the third stage of labor in second‐trimester deliveries, established 15‐methyl PG F2α injections every 20 minutes as a method of shortening the third stage of labor.3 Of all patients enrolled, approximately 70% had completed the third stage of labor by 2 hours, and the median time to deliver the placenta was significantly shorter in the 15‐methyl PG F2α group at 40 minutes versus 92 minutes in the control group. The incidence of hemorrhage and curettage in the 15‐methyl PG F2α group was 8% and 24%, respectively, and was not statistically different from the control group. This study concluded that, although the third stage of labor was shortened, there was no obvious improvement in complication rate or need for surgical intervention when 15‐methyl PG F2α was used.
The current study suggests that serial oral misoprostol in doses of 200 μg every hour is not effective in shortening the third stage of labor. Why oral misoprostol is not as equally effective as 15‐methyl PG F2α is not clear. The mechanism involved in physiologic placental separation and expulsion includes uterine contractions, and both agents are potent uterotonic PGs that have been shown to be effective in the third stage of labor.16,17 In addition, the time to peak plasma levels in both oral misoprostol and parenteral 15‐methyl PG F2α is 30 minutes.18 It may be that the method of pregnancy termination is more operative in retained placenta than the type of third stage of labor management. In the 15‐methyl PG F2α study,3 less than 50% of the study subjects received misoprostol for pregnancy termination, whereas over 90% of the subjects in the current study received misoprostol for pregnancy termination. Although misoprostol is an effective aborting agent, it may be that women who receive the drug in large doses for pregnancy termination are less sensitive to the effects of the rather small doses used in this study. This is consistent with recent evidence that women whose labors are induced with misoprostol may be at increased risk for postpartum hemorrhage.19
In addition to the lack of shortening of the third stage of labor, the misoprostol group showed no improvement in the frequency or severity of complications. Twelve percent of our misoprostol group had a hemorrhage, and 72% of those occurred after the first hour from fetal delivery. Twenty‐six percent of our misoprostol group underwent a dilation and curettage, and 84% of these occurred after the first hour. Our lower curettage rates compared with previous studies1,12,14 probably reflect our 6‐hour observation period, which allowed time for spontaneous placental passage. However, the higher hemorrhage rate suggests that even though three in four placentas will eventually deliver without surgical intervention over a 6‐hour observation period, waiting is associated with a high chance of hemorrhage. In fact, the lowest curettage rates over the 6‐hour observation period were in the first 60 minutes after fetal delivery.
Our choice of the misoprostol dosing schedule was based on our observation that a significant complaint of shivering occurred as higher doses were given. The oral route was chosen over the vaginal route because of concern that vaginal bleeding would wash out the tablet. Because misoprostol appears to be effectively absorbed across mucous membranes including rectal8 and buccal,20 other methods of administering the agent may have resulted in different outcomes. Because there were no reported side effects believed to be specifically related to misoprostol, increasing the oral dose may also have improved outcomes.
Although the statistical design was intention to treat, we acknowledge the potential for a type 2 statistical error because half of our enrolled patients did not receive their assigned trial medication. However, to achieve our objective of demonstrating a 20% difference in successful completion of the third stage of labor within 2 hours, and assuming 50% of the patients would not receive the trial medication, would require enrolling 416 patients or 10 years at the current rate of enrollment. Assuming a 30% difference in successful completion of the third stage of labor within 2 hours using the same statistical assumptions and parameters, 46 subjects in each group would be required.
Prolonged third stage of labor in second‐trimester deliveries is undefined. It appears from this study that a 6‐hour observation period is excessive, and earlier intervention to prevent complications may be prudent.11
The prophylactic use of misoprostol in the third stage of labor in second‐trimester pregnancy terminations does not appear to be effective in reducing either the time to placental delivery or complications.
1. Bivins HA, Cope DA, Newman RB, Eller DP. Randomized trial of umbilical vein oxytocin in mid-trimester pregnancy losses. Am J Obstet Gynecol 1993;169:1070–3.
2. Romero R, Hsu YC, Athanassiadis AP, Hagay Z, Avila C, Nores J, et al. Preterm delivery: A risk factor for retained placenta. Am J Obstet Gynecol 1990;163:823–5.
3. Carlan SJ, Gushwa JP, O'Brien WF, Vu Thao. Effect of intramuscular 15-methyl prostaglandin F2
alpha after second trimester delivery. Obstet Gynecol 1997;89:5–9.
4. Schenker JR, Margalioth EJ. Intrauterine adhesions: An updated appraisal. Fertil Steril 1982;37:593–610.
5. Castadot RG. Pregnancy termination: Techniques, risks and complications and their management. Fertil Steril 1986;45:5–17.
6. el-Rafaey H, O'Brien P, Morafa W, Walder J, Rodeck C. Use of misoprostol in the prevention of postpartum hemorrhage. Br J Obstet Gynecol 1997;104:336–9.
7. Hofmeyr GJ, Nikodem VC, de Jager M, Gelbart BR. A randomized placebo controlled trial of oral misoprostol in the third stage of labor. Br J Obstet Gynecol 1998;105:971–5.
8. Li YT, Yin CS. Delivery of retained placenta by misoprostol in second trimester abortion. Int J Gynecol Obstet 2001;74:215–6.
9. Winkler CL, Gray SE, Hauth JC, Owen J, Tucker JM. Mid-second trimester labor induction: Concentrated oxytocin compared with prostaglandin E2 vaginal suppositories. Obstet Gynecol 1991;77:297–300.
10. Blumenthal PD, Castleman LD, Jain JK. Abortion by labor induction. In: Paul M, Lichtenberg ES, Borgatta L, Grimes DA, Stubblefield PG, eds. A clinician's guide to medical and surgical abortion. 1st ed. New York: Churchill Livingston, 1999:139–54.
11. Kirz DS, Haag MK. Management of the third stage of labor in pregnancies terminated by prostaglandin E2. Am J Obstet Gynecol 1989;160:412–4.
12. Ben-Arie A, Hazan Y, Goldchmit R, Hagay Z. Safety of extraovular catheter insertion for second trimester abortion. Obstet Gynecol 2000;96:529–32.
13. Perry KG, Rinehart BK, Terrone DA, Martin RW, May WL, Roberts WE. Second trimester uterine evacuation: A comparison of intra-amniotic (15-S)-15-methyl-prostaglandin F2α
and intravaginal misoprostol. Am J Obstet Gynecol 1999;181:1057–61.
14. Lurie S, Appelman Z, Katz Z. Curettage after mid-trimester termination of pregnancy, is it necessary? J Reprod Med 1991;36:786–8.
15. Hogg BB, Owen J. Laminaria versus extraamniotic saline solution infusion for cervical ripening in second trimester labor inductions. Am J Obstet Gynecol 2001;184:1145–8.
16. Bamigboye AA, Hofmeyr GJ, Merrell DA. Rectal misoprostol in the prevention of postpartum hemorrhage: A placebo controlled trial. Am J Obstet Gynecol 1998;179:1043–6.
17. Oleen MA, Mariano JP. Controlling refractory atonic postpartum hemorrhage with Hemabate sterile solution. Am J Obstet Gynecol 1990;162:205–8.
18. Zieman M, Fong SK, Benowitz NL, Banskter D, Darney PD. Absorption kinetics of misoprostol with oral or vaginal administration. Obstet Gynecol 1990;90:88–92.
19. Hofmeyr GJ. Induction of labour with misoprostol. Curr Opin Obstet Gynecol 2001;13:577–81.
20. Carlan SJ, Blust D, O'Brien WF. Buccal misoprostol for cervical ripening. Am J Obstet Gynecol 2002;186:229–33.