Adverse Events Following Misoprostol Induction of Labor

Editor’s note: This article first appeared in Midwifery Today, Issue 71, Autumn 2004.
Subscribe to Midwifery Today Magazine

Introduction

“Off-label” use of misoprostol (Cytotec) for labor induction has been steadily increasing for 10 years, even though this use is approved neither by the U.S. Food and Drug Administration, other national drug regulatory agencies, the pharmaceutical industry, the Cochrane Library nor a number of national obstetric organizations, including the British Royal College of Obstetricians and Gynaecologists. Some obstetricians, particularly in the U.S., continue to promote induction with misoprostol, even though the available evidence suggests possible serious risks including uterine rupture, maternal mortality and perinatal mortality.

In January 2004, The Cochrane Library commented on the paucity of data on the serious risks of induction with misoprostol and asked for help: “The studies reviewed were not large enough to exclude the possibility of rare but serious adverse events, particularly uterine rupture, which has been reported anecdotally following misoprostol use in women with and without previous caesarean section. The authors request information on cases of uterine rupture known to readers.”(1) This paper addresses this appeal.

Methods

In the past eight years, I have consulted on sixteen medico-legal cases involving adverse events following misoprostol induction of labor in the U.S. In every case, I have carefully reviewed all prenatal, intrapartum, postpartum and neonatal medical records, including all electronic fetal monitoring strips; all laboratory results, including maternal ultrasound examinations; non-stress tests; cord blood gases; and the infant EEGs, MRIs and neurological examinations. All cases of uterine hyperstimulation and hypoxic fetal heart rate patterns have been confirmed by thorough review of the electronic fetal monitoring strips. All cases of uterine rupture are not just dehiscence but clear, significant ruptures of 6-18 cm, confirmed by the operative report on the emergency caesarean section done in each case of uterine rupture. All cases of amniotic fluid embolism have been confirmed by the pathologists’ autopsy reports. Finally, all cases of hypoxic ischemic encephalopathy of the infants have been confirmed by EEG, MRI and pediatric neurological examination. As some claim the adverse effects of misoprostol for labor induction are limited to women with previous caesarean section or to improper dosage, this information is included in each case described below.

Results (see Table 1)

  • Case 1
    In 1995, a 34-year-old gravida 2, para 1, with previous caesarean section was given misoprostol vaginally, 25 micrograms (mcg) every four hours X 2, for a total 50 mcg, followed by oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns was followed by uterine rupture. Apgar was 2, 4, 4. The baby died at three days of age from hypoxic ischemic encephalopathy (HIE).
  • Case 2
    In 1997, a 35-year-old gravida 3, para 2, with no previous c-section was given misoprostol vaginally, 50 mcg every four to six hours X 3, for a total 150 mcg, with no oxytocin. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred. Apgar was 1, 2, 5. The child has HIE with cerebral palsy and mental retardation.
  • Case 3
    In 1998, a 32-year-old gravida 3, para 1, ab 1, with one previous c-section was given 25 mcg misoprostol vaginally, then after four hours 25 mcg, then six hours later 50 mcg, for a total 100 mcg, with no oxytocin. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns was followed by uterine rupture. The child has HIE.
  • Case 4
    In 1998, a 24-year-old primip was given misoprostol, 25 mcg orally, then after four hours 50 mcg orally every four hours X 7, for a total 375 mcg, followed by oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred. The child has severe HIE with cerebral palsy and mental retardation.
  • Case 5
    In 1998, a 35-year-old gravida 9, para 5, with no previous c-section was given misoprostol vaginally, 50 mcg every four to six hours X 3, for a total 150 mcg, with no oxytocin. Uterine hyperstimulation was followed by uterine rupture. The mother developed disseminated intravascular coagulopathy requiring massive blood transfusion.
  • Case 6
    In 1999, a 33-year-old gravida 3, para 1, ab 1, with no previous c-section was given misoprostol vaginally, 50 mcg every four hours X 2, for a total 100 mcg, followed by oxytocin drip. Uterine hyperstimulation accompanied by severe hypoxic fetal heart rate patterns with meconium followed. Apgar was 1, 2, 5. The child had HIE and died after seven days.
  • Case 7
    In 1999, a 26-year-old gravida 2, para 1, with no previous c-section was given misoprostol vaginally, 50 mcg, one dose, followed by oxytocin drip. Severe uterine hyperstimulation and meconium ensued. Mother died just after giving birth. The diagnosis was amniotic fluid embolism.
  • Case 8
    In 1999, a 32-year-old gravida 2, para 1, with a previous c-section was given misoprostol vaginally, 50 mcg every four hours X 2, for a total 100 mcg. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns ended in uterine rupture. Apgar was 1, 3, 3. The child has HIE with cerebral palsy and mental retardation.
  • Case 9
    In 1999, a 34-year-old gravida 4, para 3, with no previous c-section was given misoprostol vaginally, 50 mcg, and four hours later 25 mcg, for a total 75 mcg, followed by an oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred with “pea soup” meconium. Apgar was 1, 1, 2. The child has severe HIE. The mother died just after giving birth. The diagnosis was amniotic fluid embolism.
  • Case 10
    In 2000, an 18-year-old gravida 2, ab 1, with no previous c-section was given misoprostol orally, 50 mcg, then after four hours misoprostol vaginally, 50 mcg, and after four hours, 75 mcg vaginally, for a total 175 mcg, followed by oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred. The mother died shortly after giving birth. The diagnosis was amniotic fluid embolism. The baby died shortly after birth from severe hypoxia resulting from uterine hyperstimulation and maternal shock.
  • Case 11
    In 2000, a 28-year-old gravida 2, para 1, with a previous c-section was given misoprostol vaginally, 50 mcg, one dose, followed by oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns led to uterine rupture. Apgar was 1, 5, 6. The child died at one month of age from severe HIE.
  • Case 12
    In 2000, a 25-year-old primip was given misoprostol vaginally, 100 mcg, a single dose, followed by oxytocin drip. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred with meconium. Apgar was 4, 4, 5. The child has HIE with cerebral palsy and mental retardation.
  • Case 13
    In 2000, a 30-year-old gravida 2, para 1, with no previous c-section was given misoprostol vaginally, 25 mcg every four hours X 4, for a total 100 mcg. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred. Apgar was 1, 3, 5. The mother suffered uterine rupture, and the child has HIE with cerebral palsy and mental retardation.
  • Case 14
    In 2001, a 32-year-old primip was given misoprostol vaginally, 25 mcg every four hours X 2, for a total 50 mcg. Uterine hyperstimulation with severe hypoxic fetal heart rate patterns occurred. The mother died during childbirth. The diagnosis was amniotic fluid embolism. The baby died at birth after severe hypoxia due to uterine hyperstimulation and maternal shock.
  • Case 15
    In 2001, a 33-year-old gravida 2, para 1, with previous c-section was given misoprostol vaginally, 50 mcg, one dose, then one dose prostaglandin E 2 vaginal insert (Cervadil), then oxytocin drip. Uterine hyperstimulation, accompanied by severe hypoxic fetal heart rate patterns and meconium, was followed by uterine rupture. Apgar was 1, 3, 3. The child has severe HIE.
  • Case 16
    In 2002, a 31-year-old primip was given misoprostol vaginally, 25 mcg, and then after four hours 50 mcg every four hours X 4, for a total 225 mcg misoprostol, followed by oxytocin drip. Uterine hyperstimulation, accompanied by severe hypoxic fetal heart rate patterns, occurred. An emergency caesarean section was done for severe fetal hypoxia. The child has HIE with cerebral palsy and mental retardation.

Table 1: Summary of 16 cases of misoprostol induction followed by adverse events

Previous caesarean section
7
Total dose of misoprostol
 
50 mcg
7
75 or 100 mcg
5
125 or 150 mcg
3
375 mcg
1
Uterine hyperstimulation
16
Meconium
5
Uterine rupture
7
Previous c-section
5
No previous c-section
2
Hypoxic ischemic encephalopathy
14
Neonatal deaths
4
Maternal mortality
4
Amniotic fluid embolism
4

Summary

All 16 cases of misoprostol induction of labor followed by adverse events involved uterine hyperstimulation. Five of the 16 cases had meconium. Seven of the cases involved uterine rupture. Five of the 16 women had had a previous caesarean section, and all five of these experienced uterine rupture. Of the two women in whom uterine rupture occurred without a previous caesarean section, one had three vaginal doses of 50 mcg misoprostol with no oxytocin drip, and the other had four vaginal doses of 25 mcg misoprostol with no oxytocin. Fourteen of the sixteen newborns were diagnosed as having hypoxic ischemic encephalopathy with later permanent neurological sequellae. All fourteen of these mothers experienced uterine hyperstimulation accompanied by severe hypoxic fetal heart rate patterns. All four neonatal deaths were attributed to severe hypoxic ischemic encephalopathy. All four maternal deaths were attributed to amniotic fluid embolism.

Discussion

The literature contains a large number of experimental trials on misoprostol induction of labor but, as pointed out by the editor of the British Journal of Obstetrics and Gynaecology (BJOG) in January 2004, these trials “have been too small to provide clear evidence on rare but important outcomes.”(2) In the same issue of the BJOG, a trial is reported with 641 participants, but the editor points out that this sample size, while the largest single trial to date, is still too small.(3) Efforts to overcome small sample size with meta-analysis often fail because of serious disparities among the treatment protocols used in the trials.(4)

The authors of this BJOG paper, like the authors of a number of other small trials, are willing to say misoprostol induction of labor is “safe.” Even an oft-quoted review paper states: “Because there were so few serious adverse effects, the relative risk of rare adverse outcomes with the use of misoprostol for labor induction remains unknown.” Having said this, the authors of the review then turn around and conclude, “There is strong and consistent evidence to support the use of misoprostol for induction in the third trimester.”(5)

No one disagrees that induction with misoprostol is effective and cheap. The central issue is whether it is safe. Safety in medical treatment is a value judgment: does the possibility that the procedure will improve the situation (efficacy) outweigh the chance that it will make things worse (risk)? The final judgment can only be made by the person taking the chance—the patient. It is inappropriate for a doctor, midwife or nurse to tell a patient that a treatment is “safe,” only telling her what these two chances are and leaving it to the patient to decide whether it is “safe.”

It is similarly inappropriate for the authors of research papers to write that misoprostol induction is “safe.” Furthermore, such an assessment is meaningless. What do the authors mean? That it is risk-free? Certainly not. But by quoting these papers, some providers of maternity services have tried to justify their failure to wait for adequate evidence.

Uterine hyperstimulation is a common denominator in all 16 of the cases described above. A review of the hospital records of these cases reveals that American maternity care staff understand that induction with misoprostol carries the risk of uterine hyperstimulation and subsequent uterine rupture. However, they also appear to have little or no appreciation for or ability to recognize a far more common risk—induction with misoprostol leading to uterine hyperstimulation which, by narrowing the window between contractions, reduces the time available for oxygen to reach the fetus. This can lead to fetal hypoxia and subsequent hypoxic ischemic encephalopathy. Doctors, midwives and nurses urgently need improved training in the identification and management of uterine hyperstimulation. Correct early identification of uterine hyperstimulation could prevent further doses of misoprostol and prevent addition of oxytocin to the labor management.

During the 1990s induction with misoprostol spread rapidly in the U.S. prior to sufficient evidence of its safety. It was only in 1999 that data accumulated showing the markedly increased incidence of uterine rupture when inducing with misoprostol in women with previous caesarean section. After a decade of what must be a considerable number of uterine ruptures accompanied by maternal and perinatal mortality, the American College of Obstetricians and Gynecologists recommended at the end of 1999 against misoprostol induction with a previous caesarean section. The five cases cited here of uterine rupture following misoprostol induction with VBAC serve as a reminder that obstetric practice must wait for adequate scientific research.

It is now assumed by many obstetricians that if a woman has had no previous caesarean section and the dosage of misoprostol is kept low enough, uterine rupture is not a risk. But the anecdotal data, both here and in other cases reported to The Cochrane Library, the FDA and Searle Pharmaceutical, suggest that the risk of uterine rupture following induction with misoprostol in women without a previous caesarean section remains higher than with no pharmacological induction. The anecdotal data also suggest that keeping the dose of misoprostol low does not eliminate the risk of uterine rupture and amniotic fluid embolism. Knowledge of the true risks of uterine hyperstimulation, fetal hypoxia, hypoxic ischemic encephalopathy, uterine rupture and amniotic fluid embolism when inducing with misoprostol awaits further research evidence using far larger sample sizes.

That misoprostol is a precipitating factor in amniotic fluid embolism is less well known. In the U.S. the package insert for Cytotec (misoprostol) states that amniotic fluid embolism resulting in maternal and fetal death has been reported with the use of misoprostol during pregnancy. The FDA and Searle Pharmaceutical, which manufacturers Cytotec, have had cases of amniotic fluid embolism after misoprostol induction reported to their adverse drug reaction programs. As a result, and with the active collaboration of the FDA(6), in August 2000 Searle sent a letter to all physicians in the U.S. stating: “Serious adverse effects reported following off-label use of Cytotec in pregnant women include maternal and fetal death, uterine hyperstimulation, amniotic fluid embolism, severe vaginal bleeding.”(7)

The genesis of amniotic fluid embolism is described in Williams Obstetrics: “Vigorous uterine contractions combined with a long, firm cervix and a birth canal that resists stretch may lead to uterine rupture or extensive lacerations of the cervix, vagina, vulva or perineum. It is in these latter circumstances that the rare condition of amniotic fluid embolism most likely develops.”(8) Misoprostol is known to cause unusually vigorous contractions of the uterus, and in the autopsy reports in the four cases of amniotic fluid embolism in this paper, microscopic examination showed hemorrhages and lacerations in the cervix and myometrium.

In the four cases of amniotic fluid embolism in this series, the pathophysiology was almost certainly: misoprostol induction leading to uterine hyperstimulation (proven by electronic fetal monitoring), leading to small or even microscopic lacerations of the genital tract, including the uterine wall (proven on microscopic examination at autopsy). This broke down the normal barrier between the amniotic fluid and the blood stream, causing amniotic fluid containing significant amounts of larger debris, such as fetal skin cells, to leak into the maternal blood stream, leading to embolism.

Amniotic fluid embolism was once so rare that the estimated frequency was 1:50,000–80,000 births. But now, perhaps associated with the increasing use of misoprostol for labor induction, there is some evidence of increased incidence of amniotic fluid embolism. In the U.S. in the year 2000, the Massachusetts Department of Public Health Maternal Mortality and Morbidity Review Committee reported amniotic fluid embolism as increasing to become the second most common cause of maternal mortality in that state.(9) Amniotic fluid embolism leading to maternal death (about half of women with amniotic fluid embolism die) is increasing in the U.K.—17 women died from 1994 to 1996, compared to 10 from 1991 to 1993.(10)

Conclusions

The data in this report are anecdotal. They therefore cannot be used to estimate levels of risk but are limited to suggesting possibilities for later confirmation by research. However, the anecdotal evidence in this report is consistent with previous anecdotal reports, lending strong support to the following position:

  • Other labor induction drugs, such as oxytocin, exist, for which we have sufficient data on risks. The administration of these drugs can be immediately and effectively stopped if adverse reactions occur, and their use is approved by drug regulatory agencies.
  • To date no experimental trial of misoprostol induction of labor has been large enough to exclude the possibility of rare but serious adverse events. Since the anecdotal data here and elsewhere suggest serious adverse events, it is prudent to refrain from using misoprostol for induction of labor until approved research trials with much larger sample sizes have been conducted.

Another conclusion is to eschew pharmacological induction of labor without medical indications or with merely marginal or social indications, as the serious risks will not be counterbalanced by any significant benefits.

Obstetric drug regulation is failing with regard to the non-FDA approved use of drugs prior to adequate scientific evaluation. Drug regulation by litigation is not an elegant solution but may be the only recourse at the moment to protect the public. Drug regulation by public education is another solution, but most legal settlements have included a gag rule, so the public never hears about them. Obstetricians and hospital administrators concerned with the high cost of obstetric malpractice insurance would be wise to discourage practices that carry a high risk of litigation, such as induction of labor with misoprostol. And as all medico-legal cases known to me that involve adverse events following misoprostol induction of labor (except for cases still in process) have had large out-of-court settlements in favor of the family, insurance carriers might want to reconsider their coverage of misoprostol induction of labor.

References:

  • (2004). The Cochrane Library, 1: 2.
  • Thornton, J. (2004). Editors Choice. Br J Obstet Gynaecol 111(1): 10.
  • van Gemund, N., et al. (2004). A randomized trial comparing low dose vaginal misoprostol and dinoprostone for labour induction. Br J Obstet Gynaecol 111(1): 42–49.
  • Sanches-Ramos, L., et al. (1997). Misoprostol for cervical ripening and labor induction: a meta-analysis. Obstet Gynecol 98: 633–42.
  • Goldberg, A., et al. (2001). Drug therapy: misoprostol and pregnancy. N Engl J Med 344(1): 38–47.
  • Freidman, M. (Searle). (2001) Manufacturer’s Warning Regarding Unapproved Uses of Misoprostol. N Engl J Med 344(1): 61.
  • Searle Pharmaceutical Company. (2000, Aug). Letter to practicing physicians in the US regarding use of Cytotec for labor induction.
  • William’s Obstetrics, 21st Edition. (2001). New York: McGraw Hill, p. 447.
  • Maternal Mortality and Morbidity Review in Massachusetts. (2000). “A Bulletin for Health Care Professionals,” number 1.
  • United Kingdom Department of Health. “Why Mothers Die.” (1993–1996).

About Author: Marsden Wagner

Marsden Wagner, MD, was a perinatologist and perinatal epidemiologist from California. He was Regional Officer for Maternal and Child Health in the European Regional office of the World Health Organization for 15 years. He raised four children as a single father. Marsden was an outspoken supporter of midwifery, speaking at Midwifery Today conferences and contributing numerous articles to Midwifery Today magazine (see below). Marsden passed away in April 2014 at the age of 84. His books, Born in the USA, Creating Your Birth Plan, and Pursuing the Birth Machine, are invaluable for anyone involved in birth. View all posts by

Skip to content