Drugs of Abuse in Human Milk Purchased via the Internet

Introduction

The Internet has enabled the exchange of human milk between women with an abundant supply and those who want to feed their child human milk but cannot meet their child's needs. Juxtaposed with the benefits of consuming human milk are several potential health risks to recipient infants from consuming milk from unknown sources. Because of the risks of infectious disease and exposure to illicit drugs, pharmaceuticals, and chemicals, the U.S. Food and Drug Administration cautions against feeding infants milk from unfamiliar sources. ¹ We have previously documented that 74% of milk we purchased via the Internet contained high counts of aerobic bacteria or detectable Gram-negative bacteria, which may pose a risk of infection. ²

Many drugs ingested by lactating women are detectable in milk.^3–5 The short- and long-term health effects of infant exposure to drugs via milk remain a subject of active research and are subject to evolving clinical guidance. ⁶ Healthcare providers and patients frequently must weigh the potential benefits to mothers of taking a prescription drug against the potential risk of exposure to the infant. However, illicit drugs provide no medical benefit to weigh against potential risk to the infant. For instance, cannabis remains the most popular recreational drug and concentrates potentially eightfold in human milk, but findings about the short- and long-term effects of infant exposure via human milk are very limited.^7–11 One study reported motor development deficits from exposure through breastmilk.^12–15 Narcotics (e.g., hydrocodone, oxycodone) and heroin have become more common substances of abuse among reproductive-age women, and narcotics have been associated with central nervous system depression and even death in infants exposed via milk.^7,16,17 Our previous research noted that approximately 30% of women looking to donate or sell their milk online state in their post or advertisement that they are “drug free” or otherwise indicate they do not use illicit drugs, but recipients of this milk are generally unable to know for certain that the milk they receive is free of drugs that are commonly abused. ¹⁸

As such, the objective of this study was to test milk purchased via the Internet for 13 categories of drugs of abuse so as to explore the extent of possible exposure to recipient infants.

Materials and Methods

Results

We obtained 102 milk samples for analysis and abstracted the corresponding advertisement for each. Most (71%) sellers included a statement in their advertisement about abstaining from some type(s) of drugs, although 29% did not indicate anything about drug use (whether prescribed or illicit) or abstinence (Table 1). No sellers admitted to using illicit drugs or abusing drugs in their advertisement. No specimen produced a result at or above the cutoff for any drug class, and therefore no specimen required any further confirmatory testing (all prevalence proportions = 0%; 95% confidence interval, 0.0, 2.9).

Discussion

In our analysis of 102 samples of milk purchased via the Internet, we found no evidence of contamination with drugs of abuse. The majority of sellers indicated in their advertisements they abstained from at least some kinds of drugs, although many statements (e.g., “drug free”) were too vague to determine whether they meant illicit drugs, pharmaceuticals, or both.

This study represents the first evaluation of potential exposure to drugs of abuse via human milk purchased via the Internet. We took many steps to remain anonymous (e.g., used a rented mailbox with no institution or investigator names attached, anonymous e-mail account) to obtain samples that were as representative as possible of those being sold online so as to avoid under- or over-representing drug users. We tested for a large number of the most commonly abused drugs, both illicit and prescription. These are the major strengths of the study.

Although our test results suggest that drug use is nonexistent or very rare among women selling human milk online, these findings should be considered exploratory because of several key limitations. First, our sample size was limited, and use of any individual drug is likely to be relatively uncommon among lactating women. For instance, the National Survey on Drug Use and Health (in 2012) reported that 5% of surveyed pregnant women reported using cannabis in the previous month, 0.3% used cocaine, and 0.2% used heroin, although use during lactation may differ. ⁷ Thus a much larger sample would be required to provide a narrower confidence interval around the prevalence estimate so as to provide greater assurance that exposure risk is closer to zero than 2.9%. Second, we tested for a limited list of drugs and potentially missed other drugs if present. Additionally, our milk samples may not be representative of all milk being sold online. Furthermore, the laboratory tests can detect drug use from a specific time window, which may not have been close enough to the time of ingestion. Thus, we may have missed women who used drugs intermittently or at low doses. Finally, drug metabolites may have degraded over time during the long storage period, to below the limits of detection leading them to be misclassified as negative.

We did not examine other prescription, over-the-counter, or illicit drugs of possible concern like the sleep aid zolpidem, alcohol, or galactogogues (e.g., metoclopramide) and herbs, which should be examined in future studies.

Conclusions

Based on the findings of this initial study, we conclude that exposure to drugs of abuse from human milk purchased via the Internet is likely uncommon. Larger future studies are needed to confirm our findings and to test for a greater range of drugs.