Choline is a nutrient that is a precursor of phosphatidylcholine and the plasma concentrations of both nutrients are controlled by endogenous synthesis and dietary intake [1]. Both are important for the efficient turnover of lipids from the liver and blood. Choline is also important for the control of plasma homocysteine concentration and is the precursor of the neurotransmitter acetylcholine, which is important for the proper functioning of cholinergic neurons peripherally and in the brain.

Patients with bipolar affective disorder are effectively managed with pharmacological intervention, such as lithium, haloperidol and benztropine, but studies on the influence of these drugs on plasma choline and phosphatidylcholine concentrations are limited. There is evidence that lithium can decrease the plasma availability of these important cell components [2, 3], but whether this translates into depletion in the brain supply remains questionable [4].

When women with bipolar affective disorder become pregnant, pharmacological management is complicated because of possible risks to the foetus from the use of medications. Lithium, for example is classified as a category D drug [5], that is, having the potential to cause foetal malformations, including foetal cardiac malformations [6, 7]. However in the case of pregnant patients with affective disorder, the benefits of therapy can outweigh the risks. We report on the pregnancy outcomes of a patient with bipolar affective disorder treated with mood stabilizers in the antenatal clinic of the University Hospital of the West Indies.

She was admitted to the antenatal ward at 38 weeks + 5 days and labour was induced, however due to failure to progress, a caesarean section was performed with the birth of a male infant at 39 weeks. The infant's birth Apgar scores were good: 9 at one minute and 10 at 5 minutes. The infant's birth weight was 2500 g, which is below the mean for a term baby in the Jamaican population [8]. Both infant and mother were discharged after three days and no follow-up data of either was collected.

We found that in the case of our patient, there was an unusually low placental weight and a low infant birth weight when compared with data recorded from three control patients and from previous studies of our population [8]. These previous studies also recorded an association of low birth weight infants with low haemoglobin concentrations, especially during the first trimester. However, this was not a factor in this case, as the patient maintained normal plasma concentrations of haemoglobin throughout the pregnancy.

Low weight gain during pregnancy is another risk factor that contributes to low infant birth weight [9]; however, the control patients that experienced similar weight gain did not give birth to low birth weight infants. Although our comparisons are limited by the lack of dietary intake information, previous reports have confirmed that mood stabilizers can contribute to low birth weight outcome [10].

On further comparison of this patient with controls, it appeared that there was a decrease in plasma choline and phosphatidylcholine concentrations in this patient. Both nutrients are especially important during pregnancy and are actively transported to the foetus [1, 11]. The decreases in plasma concentrations of these nutrients in our patient were unexpected, as plasma concentrations of both are increased during pregnancy [12, 13], possibly to ensure adequate supply to the foetus. Phosphatidylcholine, for example, supplies important long chain polyunsaturated fatty acids, and deficiency of polyunsaturated fatty acids to the foetus is a known risk factor for negative foetal outcomes such as low birth weight [14]. Furthermore, animal studies have demonstrated that inadequate maternal supply of these nutrients impairs cognitive and memory functions of pups and that dietary supplementation with these nutrients during pregnancy can prevent these effects [1, 15].

Our data analysis was limited by the lack of information on the actual amounts of choline and phosphatidylcholine that were consumed by this patient during pregnancy and therefore whether inadequate dietary intake contributed to the unexpected depletions. However, previously documented evidence supports negative influences of at least one of the drugs involved (lithium) on these nutrients. We therefore conclude that there is need for further studies to clarify the causal associations between drug therapy, maternal outcomes, foetal outcomes and the availability of these nutrients in patients being treated for bipolar affective disorder. Whether benefits could be derived from dietary supplementation with choline and phosphatidylcholine should also be considered.

Signed written informed consent was received from all patients reported in this paper allowing for publication of the data. A copy of the written consent is available for review by the editor-in-chief of this journal.