Many women of childbearing age are prescribed opioids for continual or intermittent pain relief. Approximately half of all pregnancies in the United States are unplanned, and as many as 86% of pregnancies are unintended in opioid-abusing women. Since many pregnancies are unintended, many medication exposures occur before women realize that they are pregnant. Virtually all drugs or their metabolites cross the placenta easily because of their lipid solubility, low molecular weight, and not being highly protein bound. Exposures that occur from implantation through the eighth week can produce fetal anomalies.

Counseling about opioids after early fetal exposure is often limited to case reports, small series, and retrospective cohort studies. Interpretation of these reports is limited by the underlining condition or comorbidities, animal studies that are usually unreliable in predicting human response, retrospective and uncontrolled methodologies, patient recall often later during gestation or afterward, and selective acceptance by medical journals of reports with positive findings. Advising about absolute risk is difficult, if not impossible, and large populations are required to determine even relative teratogenic risks.

Two reports in this issue of the American Journal of Obstetrics and Gynecology relate to the risk associated with exposure to opioids in early gestation and to a strategy to warn patients about potentially harmful pregnancy effects. The first, reported by Broussard et al from the Centers for Disease Control and Prevention, examined whether maternal opioid treatment between 1 month before pregnancy and the first trimester was associated with birth defects that contribute to infant morbidity and mortality. The other article in this issue is a secondary analysis by You et al from Northwestern University about an improved prescription drug warning strategy intended for women of childbearing age.

Oral analgesics, either alone or compounded with other additives or with other medications, represent one of the most common groups of medications taken during the preconceptional period. Prescribing of opioid analgesics has grown, and in 2002 constituted 10% of all drug abuse cases. Opioid-containing medication in early pregnancy was observed in 2-2.6% of all patients evaluated in the report by Broussard et al. Effects from maternal opioid therapy other than codeine have not been thoroughly studied, but previous reports have been inconsistent or shown no increase in risks for birth defects following prenatal exposures to oxycodone, propoxyphene, or meperidine. Most previous studies considered congenital heart defects (CHD) of any type as a single group, since sample sizes were insufficient to examine individual heart defects.

The study by Broussard et al utilized data gathered from the National Birth Defects Prevention Study (1997 through 2005), which consisted of an ongoing multisite, population-based case-control study of >30 types of major structural birth defects. In their report, fetal exposure to therapeutic opioids was associated with a greater risk of conoventricular septal defects (odds ratio [OR], 2.7; 95% confidence interval [CI], 1.1–6.3), atrioventricular septal defects (OR, 2.0; 95% CI, 1.2–3.6), hypoplastic left heart syndrome (OR, 2.4; 95% CI, 1.4–4.1), spina bifida (OR, 2.0; 95% CI, 1.3–3.2), and gastroschisis (OR, 1.8; 95% CI, 1.1–2.9). Although the data set is the largest to date to study opioid exposure and CHD, sample sizes for some individual CHD categories were borderline for what is required to observe these effects. Codeine and hydrocodone accounted for most of the statistically significant findings and were the most frequently prescribed drugs, representing 69% of all reported exposures. A stronger relationship with treatment confined to the critical window for embryologic development (3-8 weeks postfertilization) was consistent with what would be expected if the associations were causal.

The investigators also found statistically significant associations between early pregnancy opioid treatment and infants with spina bifida. Further investigation is necessary, since this association conflicts with conclusions from epidemiologic studies of human pregnancy. Associations observed in the present investigation between maternal opioid treatment and infants with hydrocephaly, glaucoma, or gastroschisis were not reported previously. Given the probability that some findings may result by chance, their results should be treated cautiously for now.

Limitations in conducting any case-control observational study are worthy of mention. Exposure information was gathered from maternal recall, which might result in self-report bias and exposure misclassification, particularly given the variable time to interview. The average time from delivery until interview was 9-11 months, and up to 2 years in certain cases. The study did not collect information on medication dose, so dose-response relationships could not be sought. Many oral analgesics were multiple-component products, with ingredients that may influence outcomes.

Also in this issue is a report by You and colleagues dealing with enhancing prescription drug warning strategies. Medication labels are often the only way that many women are informed about the potential danger of a medication in the context of pregnancy. Warning symbols can be misinterpreted or misleading, however. Instructions and precautions can be complex, and consumer feedback is seldom tested for comprehension. This secondary analysis evaluated the effectiveness of an enhanced pregnancy labeling strategy. Feedback was sought from 132 women of childbearing age who were assigned prescription containers with: the current teratogen warning; a label with a simplified text; or a label with a simplified text and icons. Their findings offer further evidence for setting forth best practices in the design and development of patient-centered warning labels on prescription drug containers. A teratogen warning label that had easy-to-read messages with icons significantly improved patient comprehension. This evidence is particularly relevant for reproductive-age women, since many become pregnant without a full understanding of the teratogenic potential of their treatment.

Whether this preferred labeling strategy will translate into fewer medication errors or undesired exposures remains to be seen. Also, it is possible that such warning labels could cause a patient not to fill her prescription or to discontinue taking the medication out of fear for any potential harm despite risks being theoretical or low. A teratogen warning label with an explicit, understandable message in conjunction with icons should hopefully encourage more prompt discussion with a patient’s obstetrician. No mention is made in this report about nonprescription medications. Containers for over-the-counter drugs often have a different (and often diverse) appearance, and additional study would be helpful.

It is important to place into perspective that an increased relative risk for any rare birth defect with any medication exposure usually translates into only a modest absolute increase above the baseline risk. Therefore, determining the clinical significance of an increased relative risk of a rare birth defect is important. The background risk of major structural malformations at birth is 3% and another 3% by age 5 years, with 8-10% of persons having ≥1 functional abnormalities by age 18 years. An early fetal ultrasound is important to confirm viability and gestational dating. Maternal serum testing, chorionic villous sampling, amniocentesis, and fetal blood sampling have no selective role in determining early fetal effects from opioid or any other drug exposure. Fetal ultrasound examination by midgestation is valuable to visualize such sites of common anomalies as the face, intracranium, heart, abdominal wall, spine, and urine-collecting system.