Lithium in Pregnancy Tied to Infant Morbidity, Malformation Risk

Batya Swift Yasgur, MA, LSW

June 27, 2018

Lithium use during pregnancy is associated with increased risk for neonatal hospital readmission and increased risk for major malformations during the first trimester, new research suggests.

A collaborative meta-analysis encompassing more than 22,000 pregnancies found that for babies exposed to lithium in utero, the rate of readmission to the hospital within 28 days after birth was almost twice as high compared to babies whose mothers did not use lithium during pregnancy.

The prevalence of major malformations in babies exposed to lithium during the first trimester was also significantly higher compared to that of unexposed babies. However, the risk for major cardiac malformations was not found to be significant.

"It is of high importance for patients and clinicians to plan pregnancies and have balanced discussions about medication use in pregnancy before conception," lead author Trine Munk-Olsen, PhD, associate professor, National Center for Register-Based Research, Aarhus University, Denmark, told Medscape Medical News.

Additionally, "malformation risks should be considered jointly with the high relapse risk in pregnancy and post partum if, for example, bipolar disorder is untreated," she said.

The study was published online June 18 in Lancet Psychiatry.

Restricted Use

Lithium is an effective first-line pharmacotherapy for patients with bipolar disorder and other psychiatric conditions, but concerns about teratogenicity and maternal/offspring complications restrict its use, the authors write.

Most congenital abnormalities have been associated with lithium use during the first trimester, since this is the period when the embryo is especially vulnerable to teratogens.

Several studies have found lithium to be associated with risks for malformations, preterm birth, and other neonatal complications, but these findings have not been consistent across all studies.

Moreover, most previous studies have lacked the statistical power to detect significant effects or have had other flaws.

Meta-analyses can "improve the precision of estimates by increasing sample sizes," the authors point out.

"We did the study because of very limited solid evidence about lithium use in pregnancy and various maternal and infant outcomes," Munk-Olson recounted.

"Previous work on the topic was, until recently, based on highly selected case samples, suggesting that previous evidence for lithium use and malformations was overestimated," she continued.

She explained that she and her colleagues chose a meta-analytic approach that relied on data from six study sites in six countries "to collect as much information as possible about lithium-exposed pregnancies."

The six international cohorts consisted of three population-level, register-based cohorts in Denmark, Sweden, and Ontario, Canada, and three clinical cohorts in the Netherlands, the United Kingdom, and the United States.

The cohorts consisted of pregnant women whose pregnancies resulted in live-born singleton deliveries and who had either a mood disorder (bipolar disorder [BD] or major depressive disorder [MDD]) or who had taken lithium during pregnancy.

Patients were assigned either to the lithium-exposed group or the mood-disorder reference group.

The mood-disorder reference group consisted of pregnant women with a known history of BD or MDD who did not take lithium from 90 days before pregnancy until delivery.

Definitions of exposures, outcomes, potential confounders, and statistical analyses were harmonized across sites by use of a shared study protocol.

Outcomes were divided into four subcategories: pregnancy complications identified during pregnancy or within 42 days after delivery (preeclampsia, gestational diabetes, fetal distress, and postpartum hemorrhage); labor and delivery outcomes (cesarean delivery, preterm birth, low birth weight, and being small for gestational age); neonatal hospital readmission within 28 days of birth; and congenital malformations in the infant.

Major cardiac malformations were defined as atrial and atrioventricular septal defects and Ebstein's anomaly.

"Positively Surprised"

A total of 22,124 pregnancies were identified from all the cohorts, of which 727 were eligible for inclusion in the lithium-exposed group (557 [77%] from register-based cohorts and 170 [23%] from clinical cohorts).

Women in the lithium-exposed group tended to be older, nulliparous, and more likely to have filled a prescription for a psychotropic drug other than lithium during pregnancy, as compared to those in the reference group.

Lithium exposure during pregnancy was not found to be associated with preeclampsia, diabetes in pregnancy, fetal distress, or postpartum hemorrhage. No differences between the groups were found for cesarean delivery, preterm birth, low birth weight, or being small for gestational age.

On the other hand, in utero lithium exposure was associated with an increased risk for neonatal readmission to hospital within 28 days of birth (pooled prevalence, 27.5%; 95% confidence interval [CI], 15.8 - 39.1, vs pooled prevalence, 14.3%; 95% CI, 10.4 - 18.2; pooled adjusted odds ratio [aOR], 1.62, 95% CI, 1.12 - 2.33).

By age 1 year, diagnoses of major malformations were reported for 51 infants in the lithium-exposed group (pooled prevalence, 7.2%; 95% CI, 4.0 - 10.4), vs 856 in the reference group (pooled prevalence, 4.3%; 95% CI, 3.7 - 4.8).

Overall, lithium exposure was not significantly associated with an increased risk for major malformation nor with major cardiac malformations, but the authors note that statistical heterogeneity was high.

However, there was a significant difference between infants exposed to lithium during the first trimester (comprising 90% of infants who had been exposed to lithium during pregnancy) vs nonexposed infants.

In exposed infants vs nonexposed infants, lithium exposure was associated with an increased risk for major malformation (7.4% vs 4.3%; pooled aOR, 1.71; 95% CI, 1.07 - 2 .72).

There was no significant increase in major cardiac malformations in lithium-exposed babies, compared with the reference group (2.1% vs 1.6%, pooled aOR, 1.54; 95% CI, 0.64 - 3.70).

Notably, no cases of Ebstein's anomaly were found at any of the participating study sites.

The results remained "robust" after additional sensitivity analyses, including stratification by study design, the leave-one-out approach, and adjustment for additional variables in a subcohort from Danish and Swedish data.

"We were positively surprised that for the majority of our defined maternal and infant outcomes, we found no association with lithium use," Munk-Olsen commented.

Preconception Discussion

Commenting on the study for Medscape Medical News, Megan Galbally, MBBS, MPM, FRANZCP, PhD, professor, Murdoch University of Notre Dame, Australia, who was not involved with the study, said that it "adds to the available research a second larger study, where previously, the studies had been much smaller, which is important when examining rarer outcomes such as malformation risk."

Galbally, who is the author of an accompanying editorial, observed that the role that dose might have played was not examined in this study and that the study also did not account for potentially important confounders, such as obesity, alcohol, smoking, or substance misuse.

"These confounders are associated with increased risk of malformations, pregnancy complications, and poorer outcomes and have also been shown to be higher in women with bipolar and on lithium in pregnancy, making them highly relevant to include in research," she noted.

The study "continues to highlight the importance of preconception consultation and discussion about the potential for malformation risk when lithium is used in the first trimester," especially given the rate of unplanned pregnancies for all women.

Additionally, "any discussion about risk of lithium should also include discussions of risk of ceasing medication for maternal health and well-being."

The authors recommend that future research include "further pooling of evidence across countries, study sites, and presented results" to "quantify any magnitude of risk associated with lithium exposure during pregnancy."

The study received financial support from the Swedish Research Council through the Swedish Initiative for Research on Microdata in the Social and Medical Sciences Framework, the National Center for Advancing Translational Sciences, the National Institutes of Health, and ICES, which is funded by an annual grant from the Ontario Ministry of Health and Long-Term Care. Dr Olsen is supported by the National Institute of Mental Health, the Lundbeck Foundation Initiative for Integrative Psychiatric Research, and Aarhus University Research Foundation. The other authors' research funding is listed on the original article. Dr Galbally has disclosed no relevant financial relationships.

Lancet Psychiatry. Published online June 18, 2018. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: