Should Lithium Dose Be Reduced Prior to Delivery?

The dosing of lithium presents certain challenges during pregnancy.  During pregnancy, there is a physiologic increase in the glomerular filtration rate (GFR), which is a measure of how quickly the kidney clears renally excreted compounds, including lithium.  Thus, lithium blood levels tend to fall gradually during the first and second trimesters of pregnancy and start to increase toward pre-pregnancy levels during the third trimester (Wesseloo et al, 2017).

How clinicians manage lithium dosing during pregnancy varies.  (We have reviewed this topic previously HERE.) One approach is to monitor patients clinically and to increase the lithium dose if there are any breakthrough symptoms.  The primary advantage of this approach is that lithium dose is kept to the lowest effective dosage; however, in this setting, it is possible that lithium levels may drop to subtherapeutic levels at some point during the pregnancy.  In theory, this may increase risk for relapse.  Thus, this approach may not be ideal in patients with extremely brittle bipolar disorder, where mood stabilization is clearly associated with a specific lithium blood level, or in women with very severe bipolar disorder, where there is a risk of self-harm, suicidality or psychosis.  

Another approach is to frequently monitor lithium blood levels during pregnancy and to adjust the dosage so that it remains within the therapeutic window (0.5 to 1.0 mmol/L). The primary advantage of this approach is that lithium levels are kept within the therapeutic range, which theoretically decreases risk for relapse.  The potential disadvantages are that higher levels may increase risk for lithium toxicity in the mother and in the infant after delivery.  In addition, there is preliminary evidence suggesting that higher doses of lithium during the first trimester may be associated with an increased risk for cardiovascular malformation.  For a daily dose of 600 mg or less, the relative risk was 1.11 and increased to 1.60 for 601 to 900 mg and 3.22 for doses greater than 900 mg.  In addition, another study suggested that higher lithium levels at the time of delivery may be associated with higher risk of neonatal complications (Newport et al, 2005).  

Lithium Dosing Around the Time of Delivery

Various papers have provided guidelines on lithium dosing during pregnancy and the immediate postpartum period.  Ultimately, the goal is to reduce the risk of complications and toxicity in the mother and the newborn, while at the same time minimizing the mother’s risk for recurrent illness.  Some have suggested a dose reduction of 30%?50% at the onset of labor, and some have recommended holding lithium prior to delivery.  The rationale for this approach is that, after delivery, there may be an abrupt increase in lithium levels due to changes in renal clearance and fluid volume after delivery.  

While these recommendations make sense, they have not been tested in a real world clinical cohort.  To test the clinical validity of these recommendations, Molenaar and colleagues conducted a retrospective observational cohort study, following maternal lithium blood level changes following delivery and examining the association between neonatal lithium blood levels at delivery and neonatal outcomes.

This study focused on a total of 78 women with a total of 100 pregnancies who were referred for psychiatric treatment during pregnancy and included women with at least one lithium blood level measurement during the last  week of pregnancy and the first postpartum week. There were a total of 233 maternal lithium blood level measurements; 55 (23.6%) during the week before delivery and 178 (76.4%) in the week after.  In contrast to the predicted rise in lithium levels occurring after delivery, this study observed no fluctuations in maternal lithium blood levels surrounding delivery (up to the first week postpartum).

In addition, they looked at outcomes in a subset of 29 newborns who had neonatal lithium blood levels drawn (either from the umbilical cord or venipuncture within 24 hours of delivery).  Neonatal lithium levels ranged from 0.05 to 1.16 mmol/L.

About half of the newborns in this cohort (14/29) had complications.  The most commonly reported complications were hyperbilirubinemia (n = 6), respiratory complications (n = 5), and hypotonia (n = 3). While they observed a strong correlation between maternal and neonatal lithium blood levels, there was no association between neonatal lithium blood levels at delivery and neonatal outcomes.  

The finding of no association between neonatal lithium levels and risk for neonatal complications differs from the study published by Newport and colleagues.  However, the Molehaar and colleagues note that in the Newport study, neonates were divided into two groups: those with low lithium levels (under 0.64 mmol/L) and those with higher lithium levels.   The high lithium level group in Newport’s study often had blood levels higher than 0.7 with some in this group having toxic levels (>1.2 mmol/L). The low lithium level group were most often born to mothers who decreased or held their lithium prior to delivery and typically had subtherapeutic levels (below 0.5 mmol/L).

It is worth noting that, while risk for adverse neonatal outcomes in this study was not correlated with lithium blood levels, about half of the neonates had complications, suggesting that exposure to lithium (at any level) near the time of delivery may be associated with worse outcomes.  

Some Final Thoughts

While there are diverging opinions regarding the management of lithium dosing during pregnancy and at the time of delivery, it is important to keep some guiding principles in mind.  Our overarching goal is to reduce risk for relapse in the mother while at the same time minimizing risk of complications in the mother and her child.  While relapse rates are clearly elevated during pregnancy, it seems that women are the most vulnerable to relapse during the postpartum period. Thus, we must consider the possibility that decreasing the dose of lithium prior to delivery — which may result in subtherapeutic levels in the mother — may increase risk for postpartum illness.

When there is limited information to inform our decisions, it is indeed challenging to determine what is the “best” option.  In the absence of clear guidelines in an extremely vulnerable population, the bedrock of our care should be close monitoring during pregnancy and into the postpartum period.  Maintenance of a mood stabilizer decreases the risk for relapse; however, monitoring and adjusting lithium blood levels is but one component of the care we provide.  Other interventions can help to decrease risk for relapse, including early detection and treatment of recurrent symptoms, careful management of sleep, stress reduction, and bolstering social supports.  

Based on the information we have now, there is not evidence to indicate that we should decrease the lithium dose proximate to delivery.  The data from the Molenaar study indicates that there is not a significant increase in lithium levels immediately after delivery, or during the first week, in the mother.  If the mother has increased her lithium dosage during pregnancy, she will need to adjust her dosage to pre-pregnancy levels, but this change can be made more gradually during the first few weeks after delivery.

We must acknowledge, however, that there is a risk for adverse outcomes in children exposed to lithium near the time of delivery.  Although the Molenaar study did not show a correlation between lithium levels and risk for adverse outcomes at delivery, it was a very small study and about half of the children had complications.  Lithium toxicity is more likely to occur at higher blood levels, and it is likely that neonates, because they have a less mature nervous system, may be more vulnerable to the effects of lithium.    Rather than decreasing lithium dosage proximate to delivery to minimize risk to the newborn, our approach would be to use the lowest effective dosage during pregnancy.

Ruta Nonacs, MD PhD

Molenaar NM, Poels EMP, Robakis T, Wesseloo R, Bergink V.  Management of lithium dosing around delivery: An observational study.  Bipolar Disord. 2021 Feb; 23(1):49-54.  Free article.

Newport DJ, Viguera AC, Beach AJ, Ritchie JC, Cohen LS, Stowe ZN. Lithium placental passage and obstetrical outcome: implications for clinical management during late pregnancy. Am J Psychiatry. 2005;162(11):2162?2170. 

Wesseloo R, Wierdsma AI, van Kamp IL, et al. Lithium dosing strategies during pregnancy and the postpartum period. Br J Psychiatry. 2017;211(1):31?36.