Codeine Toxicity via Breast Milk

It was with great interest that we read the recent Perspectives piece by Drs. Anderson and Hannam on the issue of codeine toxicity via breast feeding [1]. This is an elegant modeling exercise which explores many of the complex issues underlying codeine metabolism in the mother-infant dyad in the context of breast feeding which illustrates that codeine toxicity via breast feeding is likely to be rare. We would however respectfully note that the conceptual basis of codeine biotransformation in the neonate is based on the assumption that codeine metabolism in the neonate is much reduced from adult values, an assumption that was also made by Willmann et al. in their modeling work exploring this issue [2]. While the vast majority of pathways of drug clearance function below expected adult values, in the case of CYP2D6 there is robust in vitro data demonstrating that CYP2D6 activity increases rapidly after birth and that its phenotype is consistent [3, 4]. As well, on the maternal side, CYP2D6 activity has been shown to increase over the course of pregnancy [5]. As noted by Drs. Anderson and Hannam, the alternate metabolic pathways do have reduced activity relative to adult values, and thus route switching would provide additional substrate for conversion to morphine. If this is factored into modeling—which in the case of Willmann et al. assumed 10% of adult values rather than the 80%–100% it is likely to be—the concentrations observed in the Lancet case of 2006 would be expected to be achieved notably given the combination of risk factors in this case [6, 7]. As well, the OCT1 transporter has a clear role in morphine clearance and genetic polymorphisms in transporter activity have been associated with significant alterations in codeine-morphine disposition—this association was made in 2013, unfortunately well after the Lancet case and thus was not explored at the time [8]. Given this and that the modeling was done with the assumptions being a 5 day old infant with short term maternal therapy versus an older infant in the context of prolonged maternal therapy—as noted by Koren et al.—or habitual use as per Heerfordt et al.—with consideration of maturation of CPY2D6 activity versus the immaturity of glucuronidation or CYP3A4 activity and hence the potential for both enhanced metabolism and route switching as well as altered OCT1 activity it would be of interest to consider conducting a modeling exercise taking these factors into account [6, 7, 9, 10]. Finally, while we agree with Willmann et al. and Drs. Anderson and Hannam that codeine toxicity via breast feeding is likely to be rare and primarily in the context of specific drug utilization and pharmacogenomic risk factors, it is worthy to note the work of Sistonen et al. who described central nervous system depression attributed to codeine in 26 infants in a cohort of 111 mother-infant pairs, this being related to a series of pharmacogenomic risk variants [10]. We agree with Drs. Anderson and Hannam that while short courses of maternal opioids are usually safe while breast feeding, caution is merited when multiple doses are used as under certain circumstances opioid toxicity is plausible, notably as Drs. Anderson and Hannam note that there are non-opioid therapeutic alternatives which transfer into breast milk in very small amounts [1, 7]. This article is linked to Anderson et al. papers. To view this article, visit https://doi.org/10.1002/pan.70227. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.