New data continues to be published almost monthly which supports that human milk is the finest source of nutrition for human infants. During the first week postnatally, human milk is filled with immunoglobulins (IgG, IgA, and IgM), interferons, and other antibacterial substances which results in colonization of the GI tract of the newborn with benign bacterial flora. The presence of numerous growth factors(IgF-1), cytokines and gastric hormones enhance development of protective barriers in the GI tract and stimulate elimination of meconium present during gestation. Other key benefits to the infant include perfect nutrition, enhanced neurocognitive development, stronger immune function, and at least a three fold reduction in infectious disease such as upper respiratory infections, otitis media, and necrotizing enterocolitis.

Although recent studies have suggested that the number of women who choose to breastfeed their infant is presently 72% and rising in the USA, the number of women who discontinue breastfeeding in order to take a medication is still significant. Surveys in many countries indicate that 90-99% of women who breastfeed will receive at least some medication during the first week postpartum. Other studies suggest that the use of medications is one of the major reasons women discontinue breastfeeding prematurely. The most frequently used drugs include analgesics, hypnotics, and methylergometrine.

Because so many women ingest medications during the early neonatal period, it is not surprising that one of the most common questions encountered in pediatrics is concerning the use of various drugs during lactation. Unfortunately, most physicians simply review the package insert, or advise the mother to not breastfeed without having done a thorough study of the literature to find the true answer. Discontinuing breastfeeding is often the wrong decision and most mothers could easily continue to breastfeed their infants and take the medication without risk to the infant.

In the past 20 years a large number of studies have been published providing a better understanding of the kinetics of drug entry into human milk. Most of the physiocochemical properties that facilitate transfer (molecular weight, pKa, lipophilicity) are well known, but ultimately the degree of transfer of the medication must be determined in humans. Rodent studies are virtually worthless as the albumin content in rodent milk is many times higher than in humans. This may account for the fact that most of the rodent studies show milk levels that are many times too high and simply do not correlate with human studies. Benefits to breastfeeding women are significant and include a major reduction in bone loss in later life, and a reduction in the incidence of breast and ovarian cancer. Other benefits include enhanced weight loss, reduced blood loss postpartum, and a more desirable inter-pregnancy interval.

The publication several years ago by the American Academy of Pediatrics recommending at least 1 year of breastfeeding was a major step in acceptance by a major medical body that breastfeeding is immunologically and nutritionally the best way to feed infants. The Academy, provided with evidence by numerous publications that breastfed infants are not only healthier but may also gain in developmental skills as well, finally felt that the evidence was overwhelming and justified a stronger stance by this large body of pediatricians. Due to enormous support for breastfeeding in the medical community many mothers are now resistant to discontinuing breastfeeding just to take a medication, particularly if the interruption is based solely on the recommendation of their physician. Although interrupting breastfeeding may appear safest to the physician, it is simply not necessary in most cases as the amount of drug transferred to milk is normally quite small.

We all know that most medications penetrate milk to some degree. However, the amount transferred into human milk is generally low. With only a few exceptions, the dose of the medication transferred to the infant via milk is generally far subclinical. It is ultimately the clinicians responsibility to review the documentation available before coming to a decision as to breastfeeding. There are almost always other medications that could be substituted that are safe for both mother and infant.

The pharmaceutical manufacturer’s resistance to using medications in breastfeeding mothers is almost always based on legal reasons, not clinical reasons.

Because the PDR basically lists only the pharmaceutical’s package insert, the standard recommendation is to not take the medication while breastfeeding. The PDR is the poorest source for obtaining accurate breastfeeding information. The amount of drug excreted into milk depends on the following factors:

Drugs transfer into human milk if they:

  • Are highly lipid soluble.
  • Attain high concentrations in maternal plasma.
  • Are low in molecular weight (< 500 ).
  • Are low in protein binding.
  • Pass into the brain easily.

However, once medications are transferred into human milk then other kinetic factors are involved. One of the most important is the oral bioavailability of the medication to the infant. Numerous medications are either destroyed in the infant’s gut, fail to be absorbed through the gut wall, or are rapidly picked up by the liver. Once in the liver they are either metabolized or stored.

Drugs enter milk primarily by diffusion driven by equilibrium forces between the maternal plasma compartment and the maternal milk compartment. They pass from the maternal plasma through capillary walls into the alveolar cell lining the alveolus. Medications must generally pass through both lipid membranes of the alveolar cell to penetrate milk; although early on they may pass between the alveolar cells (first 72 hours postpartum). During the first three days of life large gaps between alveolar cells exist. These gaps permit enhanced access for most drugs, many immunoglobulins, maternal living cells (lymphocytes, leukocytes, macrophages) and other maternal proteins to the milk. By the end of the first week, the alveolar cells swell under the influence of prolactin; subsequently closing the intracellular spaces and reducing the intracellular entry of most maternal drugs, proteins, and other substances, to the milk compartment. It is generally agreed that medications penetrate into milk more during the colostral period than in mature milk. However, the absolute dose transferred during the colostral period is still low as the total volume of milk is generally less than 30-100 mL total volume/day.

In most instances, the most important determinant of drug penetration into milk is the mother’s plasma level. Almost without exception, as the level of the medication in the mother’s plasma rises, the concentration in milk increases as well. Drugs enter milk and exit milk as a function of the mother’s plasma level. As soon as the maternal plasma level of a medication begins to fall equilibrium forces drive the medication out of the milk compartment back into the maternal plasma for elimination. In some instances, drugs are trapped in milk (ion trapped). This means that due to the lower pH of human milk (7.0-7.2), the ionic state of the drug changes and prevents its’ exit back into the maternal circulation. This is important in weakly basic drugs such as the barbiturates. With the iodides, such as 131I or any ionic form of iodine, the drug may concentrate in milk due to a pumping system on the alveolar cell wall. Thus iodides, particularly radioactive ones, should be avoided as their milk concentrations are exceedingly high.

Two other physicochemical factors are important in evaluating drugs in breastfeeding mothers. These are the degree of protein binding, and lipid solubility. Drugs that are very lipid soluble penetrate into milk in higher concentrations almost without exception.

Of particular interest are the drugs that are active in the central nervous system (CNS). CNS-active drugs invariably have the unique characteristic requisite to enter milk. Therefore, if a drug is active in the central nervous system higher levels in milk can be expected; although, they still are often subclinical. Many of the neuroactive drugs produce Relative Infant Doses of >5%. Protein binding also plays an important role. Drugs circulate in the maternal plasma either bound to albumin or freely soluble in the plasma. It is the free component (unbound) that transfers into milk while the bound fraction stays in the maternal circulation. Therefore, drugs that have high maternal protein binding (warfarin) have reduced milk levels simply because they are excluded from the milk compartment. Once a drug has entered the mother’s milk and has been ingested by the infant it must traverse through the infant’s GI tract prior to absorption. Some drugs are poorly stable in this environment due to the proteolytic enzymes and acids. In general, the infant’s stomach is quite acidic and the acidity can denature many drugs. This includes the aminoglycoside family, Omeprazole, and large peptide drugs such as Heparin or Insulin. Other drugs are simply poorly absorbed by the gastrointestinal tract, and are poorly absorbed into the infant’s blood stream. Oral bioavailability is a useful tool to estimate just how much of the drug will be absorbed by the infant. In addition, many drugs are sequestered in the liver (first pass) and may never actually reach the plasma compartment where they are active. Such absorption characteristics tend ultimately to reduce the overall effect of many drugs. There are certainly exceptions to this rule and one must always be aware that the action of a drug in the GI tract can also be profound, producing diarrhea, constipation, and occasionally syndromes such as pseudomembranous colitis.

One of the more popular methods for estimating risk is to determine the Relative Infant Dose (RID). The RID is calculated by dividing the infant’s dose via milk (mg/kg/day) by the mothers dose in mg/kg/day. The RID gives the clinician a feeling for just how much medication the infant is exposed to on a weight-normalized basis. However, many authors calculate the infant dose without normalizing for maternal and infant weight so be cautious.

Key Points About Breastfeeding and Medications

  • Avoid using medications where possible. Herbal drugs, high dose vitamins, unusual supplements, etc. are simply not necessary; recommend avoiding the risk.
  • If the Relative Infant Dose is less than 10%, most medications are quite safe to use. The RID of the vast majority of drugs is < 1%.
  • Choose drugs for which we have published data, rather than those recently introduced.
  • Evaluate the infant for risks. Be slightly more cautious with premature infants or neonates. Be less concerned about older infants.
  • Observe breastfeeding mothers for depression, as the incidence is high.
  • Recommend antidepressants in depressed breastfeeding mothers, the risks of a depressed mother to the infant are simply too high.
  • Most drugs are quite safe in breastfeeding mothers, while the risks of not breastfeeding and of using infant formulas are much higher for the infant.
  • Discontinuing breastfeeding for some hours/days may be required, particularly with radioactive compounds. Choose drugs with short half-lives, high protein binding, low oral bioavailability or high molecular weight. Adapted from Hale TW, Ilett KF. Drug Therapy and Breastfeeding: From Theory to Clinical Practice, First Edition ed. London: Parthenon Publishing; 2002.

Lastly, it is extremely important to always evaluate the infant’s ability to handle small amounts of medications. Some infants, such as premature or ill infants, may not be suitable candidates for certain medications. But remember that early postpartum, the amount of milk produced is so low that the clinical dose of drug transferred is often low, so even in premature neonates, the dose is still probably low because of the limited amount of milk used in these infants.

Evaluation of the Infant

  • Inquire about the infant - Always inquire as to the infant’s age, size, and stability. It’s an important criteria before using medications in breastfeeding mothers.
  • Infant age - premature and newborn infants are at somewhat greater risk.
  • Infant stability - unstable infants with poor GI stability may increase the risk of using medications.
  • Pediatric Approved Drugs - generally are less hazardous if long-term history of safety is recognized.
  • Dose - in a premature infant various doses may be more risky than in a 1 year old healthy infant.
  • Drugs that alter milk production - may be much more risky during the neonatal period than after the milk supply is well established.

General Suggestions for the Clinician

1. Determine if the drug is absorbed from the GI tract. Many drugs such as the aminoglycosides, vancomycin, cephalosporin antibiotics (third generation), morphine, magnesium salts, and large protein drugs (heparin) are so poorly absorbed that it is unlikely the infant will absorb significant quantities. At the same time observe for GI side effects from the medication trapped in the GI compartment of the infant (e.g. diarrhea).

2. Review the monograph provided herein on the drug. Review the theoretical infant dose and the relative infant dose (RID) and compare that to the pediatric dose if known. Remember, most of the Theoretic Infant Doses were derived from the Cmax (highest milk concentration of the drug) that is published. Unfortunately, the milk/plasma ratio is virtually worthless information unless you know the maternal plasma level. It does not provide the user with information as to the absolute amount of drug transferred to the infant via milk. Even if the drug has a high milk/plasma ratio, if the maternal plasma level of the medication is very small (such as with propranolol), then the absolute amount (dose) of a drug entering milk will still be quite small and often subclinical.

3. Try to choose shorter half-life drugs, as they generally peak rapidly and then are eliminated from the maternal plasma thus exposing the milk compartment (and the infant) to reduced levels of medication. Urge the mother not to breastfeed while the drug is at it’s peak level in the maternal plasma. Determine the time-to-peak interval as this will indicate how long the mother must wait before feeding. However, you must first determine the dosage form administered. If the tablet formulation is a prolonged release form, then all prior assumptions about half-life are pointless and you must assume that the drug has a long half-life (12-24 hours). Unfortunately, avoiding peaks early postpartum is simply not very practical and drugs with low milk levels should be chosen.

4. Be cautious of drugs (or their active metabolites) that have long pediatric half-lives as they can continually build up in the infant's plasma over time. The barbiturates, benzodiazepines, meperidine, and fluoxetine are classic examples where higher levels in the infant can and do occasionally occur.

5. If you are provided a choice, choose drugs that have higher protein binding because they are more often sequestered in the maternal circulation and do not transfer as readily to the milk or the infant. Remember, it’s the free drug that transfers into the milk compartment. Without doubt, the most important parameter that determines drug penetration into milk is plasma protein binding. Choose drugs with high protein binding.

6. Although not always true, I have generally found neuroleptic drugs frequently penetrate milk in higher levels simply due to their physicochemistry. If the drug in question produces sedation, depression, or other neuroleptic effects in the mother, it is likely to penetrate the milk and may produce similar effects in the infant. Sedative drugs (particularly phenothiazines) may contribute to an elevated risk of SIDS, although this is poorly documented.

Be cautious of herbal drugs as many contain chemical substances that may be dangerous to the infant. Numerous poisonings have been reported. Prior to using, advise the mother to contact a lactation consultant or herbalist who is knowledgeable about their use in breastfeeding mothers. Do not exceed standard recommended doses. Try to use pure forms, not large mixtures of unknown herbals. Do not overdose, use only minimal amounts.

8. With radioactive compounds check the NRC table in the index of this volume. The NRC recommendations are quite good. They can be copied and provided to your radiologist. They are available from the Nuclear Regulatory Commission’s web page address in the appendix. Also, I’ve included a number of new tables in the appendix on radioisotopes and radioactive procedures.

9. Use the Theoretic Infant Dose to estimate the maximum dose the infant would receive per kilogram per day. This data is derived from the literature and is a close estimate of the peak amount an infant would receive. Another very useful evaluation of the dose is to determine the Relative Infant Dose. The box below shows the calculation. In general, a Relative infant dose of < 10% is considered safe and its use is becoming increasingly popular by numerous investigators. But you will still find many useful drugs may approach 20% RID( metronidazole, fluconazole, etc.). Because they are incredibly safe medications, they are still quite useful in breastfeeding mothers. Remember the RID is only a guideline, certain safer medications simply don’t fall within this rule.

10. Hence, it is important that the clinician evaluate all the risks of the drug in relationship to the absolute dose received by the infant and the ability of the infant to handle even that dose. It is no longer acceptable for the clinician to interrupt lactation merely because of heightened anxiety on their part. The risks of formula feeding are significant and should not be trivialized. There are few drugs that have well documented and significant toxicity in infants as a result of breastfeeding and we know most of these. At the same time, no medication should be used lightly in a breastfeeding mother.

The following review of drugs is a compilation of what has been published concerning the use of various medications in breastfeeding women. It is unfortunate that manufacturers are not required by law to provide breastfeeding data on all their products. This would make evaluation of using new products in breastfeeding women much easier.


Thomas W. Hale