Prescription Medicine While Breastfeeding: A Complete Lactation Guide

How Drugs Move into Human Milk

Drug entry into breast milk is not random. The physical and chemical properties of a molecule determine how much of a maternal dose reaches a nursing infant.

The Four Key Pharmacokinetic Factors

Molecular weight. Drugs with molecular weight above 500 Da cross into milk poorly. Most biologics, including insulin and enoxaparin, have molecular weights in the thousands and transfer at clinically negligible levels. Heparin (MW ~15,000 Da), for example, is undetectable in milk.

Protein binding. Highly protein-bound drugs (above 85% bound) stay largely in the maternal circulation. Ibuprofen (99% protein bound) produces milk concentrations well below 1% of the maternal plasma level.

Oral bioavailability. Even if a drug reaches milk, the infant's gut absorption matters. Vancomycin crosses into milk at measurable concentrations, but its oral bioavailability is <5%, so systemic infant exposure remains minimal. [1]

Lipid solubility and ionization. Lipophilic, non-ionized compounds concentrate in the fat-rich hindmilk. This is why sedating antihistamines and some psychotropics warrant more caution than their plasma levels alone would suggest.

The Infant Relative Dose Threshold

The infant relative dose (RID) standardizes exposure across drugs. The formula is:

RID (%) = (infant dose via milk in mg/kg/day) divided by (maternal dose in mg/kg/day) multiplied by 100.

An RID below 10% is the accepted clinical threshold for most drugs, first described by Hale and later endorsed by the Academy of Breastfeeding Medicine (ABM). [2] Many commonly prescribed drugs fall well below this mark.

The Hale Lactation Risk Categories

Thomas Hale's five-tier system, published in Medications and Mothers' Milk (now in its 19th edition), remains the most widely used clinical classification in North America. [3]

| Category | Label | Meaning | |---|---|---| | L1 | Safest | Controlled studies show no infant risk; RID typically <1% | | L2 | Safer | Limited studies, no increase in adverse effects | | L3 | Moderately safe | No controlled studies; risk possible but remote | | L4 | Possibly hazardous | Positive evidence of risk; benefit may outweigh risk | | L5 | Contraindicated | Risk clearly outweighs benefit |

The NIH LactMed database provides an open-access, regularly updated alternative and is the resource most cited by the ABM. [4] Both tools should be consulted together because each may contain more recent data on a specific drug.

Pain Management During Breastfeeding

First-Line Options

Ibuprofen (L1) is the preferred oral analgesic during lactation. Its high protein binding (99%) and short half-life (~2 hours) produce an RID of approximately 0.1 to 0.6%. A 2020 analysis in Breastfeeding Medicine confirmed that ibuprofen 400 mg every six hours produces undetectable levels in most milk samples beyond two hours post-dose. [5]

Acetaminophen (L1) is also well established. Standard adult doses (325 to 1,000 mg every four to six hours) yield an infant RID of roughly 8.8%, just under the 10% threshold, and no hepatotoxicity has been observed in nursing infants at maternal therapeutic doses. [6]

Opioids: A More Careful Calculation

Codeine is no longer recommended during lactation. The FDA issued a black box warning in 2017 after neonatal deaths were linked to ultra-rapid CYP2D6 metabolism converting codeine to morphine at unexpectedly high rates. [7] Oxycodone (L3) is sometimes used short-term with close infant monitoring, particularly post-cesarean, though its RID of 3.4 to 8% warrants caution. Morphine (L3) may be preferred over oxycodone because of its lower oral bioavailability in the infant. Hydrocodone (L3) carries a similar caution; the FDA updated labeling in 2015 to warn of neonatal respiratory depression. [8]

Key principle: for any opioid, use the lowest effective dose, the shortest duration possible, and watch the infant for sedation, poor feeding, or breathing changes.

Antibiotics During Breastfeeding

Routinely Compatible Choices

The majority of antibiotics commonly prescribed in primary care are compatible with breastfeeding.

• Amoxicillin (L1): RID 0.9 to 1.8%. The most-prescribed antibiotic in outpatient medicine; decades of safety data exist.
• Amoxicillin-clavulanate (L1): Similar RID; monitor infant for loose stools or candidiasis.
• Azithromycin (L2): RID approximately 5.9%; widely used for respiratory and pelvic infections during lactation.
• Cephalexin (L1): RID <1%; first choice for mastitis and skin infections. [9]
• Nitrofurantoin (L2): Appropriate for uncomplicated urinary tract infections except in infants under one month or those with G6PD deficiency, where hemolysis risk is elevated. [10]

Antibiotics That Require Avoidance or Modification

Tetracyclines (doxycycline, tetracycline) are generally avoided for courses beyond 3 weeks because of theoretical dental staining, though short courses (7 to 10 days) carry low actual risk given poor oral absorption in the presence of milk calcium.

Fluoroquinolones (ciprofloxacin, levofloxacin) are classified L3; animal data raised concerns about cartilage toxicity. Brief courses are sometimes used when no alternative exists, but they are not first-line. [11]

Chloramphenicol (L4) is avoided because of the risk of "gray baby" syndrome even at low infant doses.

Mental Health Medications

Antidepressants

Depression affects 10 to 15% of postpartum people, yet treatment rates remain low partly because of medication concerns. [12] The evidence supporting several antidepressants during lactation is substantial.

Sertraline (L2) is the antidepressant of choice during lactation according to ABM Protocol 18 (2015, reaffirmed 2021). [13] Its RID ranges from 0.4 to 2.2%, and published case series covering hundreds of mother-infant dyads have found no measurable infant serum levels or adverse outcomes in full-term infants.

Paroxetine (L2) has an RID below 2% and is similarly well studied. It is avoided near delivery, however, because of neonatal adaptation syndrome risk.

Escitalopram (L2) and fluoxetine occupy the next tier. Fluoxetine (L2, though sometimes listed L3 in premature infants) has an active metabolite (norfluoxetine) with a half-life of 4 to 16 days; RID ranges up to 14.6% in some studies, which is above the 10% threshold. For new prescriptions during lactation, sertraline or paroxetine are generally preferred. [14]

SNRIs: Venlafaxine (L3) produces an active metabolite (O-desmethylvenlafaxine) with significant milk transfer; use requires monitoring. Duloxetine (L3) data are more limited but suggest an RID of approximately 0.1%, which is low.

Anxiolytics and Sleep Aids

Short-acting benzodiazepines such as lorazepam (L3) may be used cautiously at the lowest effective dose for acute anxiety, but accumulation with daily use raises sedation risk in the infant. Zolpidem (L3) has an RID of approximately 1.5%, but sedation in the nursing parent itself creates safety concerns during nighttime infant care.

Cardiovascular and Thyroid Medications

Hypertension

Postpartum hypertension is common and under-recognized. The preferred agents during lactation are:

• Nifedipine (L2): RID <3%; the calcium channel blocker most studied in lactating people.
• Labetalol (L2): RID approximately 0.6%; widely used postpartum.
• Enalapril (L2): The ACE inhibitor with the most lactation data; RID roughly 0.2%.

ACE inhibitors are generally avoided in the first month of life because of the theoretical risk of neonatal renal impairment, though data in older infants are reassuring. [15]

Avoid: Atenolol (L3, higher RID ~6.8% with case reports of bradycardia and cyanosis) and ACE inhibitors in mothers of premature infants.

Thyroid Disease

Levothyroxine (L1) is fully compatible. Endogenous T4 is already present in milk; supplemental levothyroxine at standard replacement doses does not meaningfully alter infant thyroid function. [16]

Propylthiouracil (PTU, L2) is preferred over methimazole for treating maternal hyperthyroidism during lactation. PTU is more highly protein-bound and produces lower milk concentrations; however, doses should be kept at or below 300 mg/day and infant thyroid function should be monitored at 4 and 8 weeks. Methimazole (L3) is an alternative when PTU cannot be used; at doses <20 mg/day, infant thyroid function has generally remained normal in published case series. [17]

Galactagogues: Medications to Increase Milk Supply

Not every low-supply situation requires a drug. The ABM Protocol 9 (2018) emphasizes that frequent, effective milk removal, ideally 8 to 12 times per 24 hours, is the foundation of supply management and that galactagogues should only be considered after mechanical causes of low supply have been addressed. [18]

Domperidone

Domperidone is a dopamine antagonist that raises prolactin by blocking D2 receptors in the pituitary. It does not cross the blood-brain barrier readily, so CNS side effects are limited. An RID of approximately 0.01 to 0.1% means direct infant exposure is minimal. Typical doses in published trials range from 10 mg three times daily to 20 mg three times daily. A 2012 Cochrane review found domperidone significantly increased milk volume in mothers of preterm infants compared with placebo. [19]

Caution: Domperidone prolongs the cardiac QT interval. Health Canada issued a warning in 2012, and the FDA has not approved domperidone for any indication in the United States, meaning U.S. Patients must access it through compounding pharmacies or from Canada. The ABM recommends a baseline ECG for patients with cardiac risk factors before initiating therapy. [18]

Metoclopramide

Metoclopramide (Reglan) at 10 mg three times daily raises serum prolactin and has modest short-term evidence for increasing milk output in some populations. Unlike domperidone, it crosses the blood-brain barrier and carries a black box warning for tardive dyskinesia with prolonged use. The FDA label specifically warns against use for more than 12 weeks. [20] Given this profile, the ABM considers metoclopramide a second-line option.

Herbal Galactagogues and Prescription Alternatives

Fenugreek, blessed thistle, and shatavari are frequently self-prescribed but lack controlled trial data demonstrating efficacy in term-infant dyads. No herbal preparation appears in the ABM's evidence-graded galactagogue protocol as a recommended first or second line. [18] Sulpiride and chlorpromazine have been studied in small trials but are not used in standard U.S. Practice.

Lactation Suppression

Some clinical situations, including stillbirth, neonatal death, maternal serious illness, or patient preference, require pharmacologic suppression of lactation.

Cabergoline (L5 for lactation) at a single oral dose of 1 mg taken within 24 hours of delivery is the preferred pharmacologic approach where it is available. A 1991 randomized controlled trial (N=272) published in Obstetrics and Gynecology found cabergoline superior to bromocriptine for suppression success and tolerability. [21] Cabergoline inhibits prolactin secretion by acting on pituitary dopamine D2 receptors. Its long half-life (63 to 68 hours) means a single dose is sufficient in most cases.

Bromocriptine was the standard agent until case reports of puerperal stroke and myocardial infarction prompted the FDA to withdraw its lactation-suppression indication in 1994. It is no longer recommended for this purpose. [22]

Non-pharmacologic methods (tight breast binding, cabbage leaves, ice packs) are less effective than cabergoline but appropriate when drug use is contraindicated.

Drugs That Are Contraindicated During Breastfeeding (L5)

The following drug classes carry enough evidence of infant harm that breastfeeding should be stopped or pumped milk discarded if their use is necessary:

| Drug or Class | Primary Concern | |---|---| | Amiodarone | Iodine load; theoretical thyroid suppression in infant | | Radioactive iodine (I-131) | Direct thyroid ablation risk; ABM recommends cessation for 2 to 8 weeks minimum [23] | | Cyclophosphamide / methotrexate | Immunosuppression, neutropenia in infant | | Lithium | RID 12 to 30%; neonatal toxicity, cardiac defects on fetal exposure | | Ergotamine | Vasospasm, ergotism in infant; suppresses prolactin | | Cocaine / PCP | Direct CNS toxicity in infant |

The American Academy of Pediatrics (AAP) 2013 policy statement, "The Transfer of Drugs and Therapeutics Into Human Breast Milk," remains a foundational reference document for this list. [24]

When to Consult a Lactation-Trained Clinician or Pharmacist

Standard primary care visits often lack the time to reconcile a full medication list against lactation safety data. A referral to a board-certified lactation consultant (IBCLC) or a clinical pharmacist with lactation training is appropriate in the following situations:

• More than two concurrent chronic medications requiring review
• Psychiatric polypharmacy (antipsychotic plus mood stabilizer)
• Maternal chronic illness (epilepsy, rheumatoid arthritis, HIV, cancer)
• Preterm infant under 32 weeks gestational age (higher GI permeability and reduced renal clearance)
• Any maternal medication with an RID above 10%

The Academy of Breastfeeding Medicine's clinical protocols are freely accessible at bfmed.org and provide decision frameworks for specific drug classes. The online LactMed database at https://www.ncbi.nlm.nih.gov/books/NBK501922/ is updated monthly and should be the first lookup tool for any unfamiliar drug. [4]

Monitoring the Nursing Infant

The nursing infant is the second patient in this clinical equation. General signs of adverse drug exposure in a nursing infant include:

• Unusual sedation or difficulty waking for feeds
• Poor latch or weak suck
• Decreased wet diapers (fewer than 6 per 24 hours in an infant over 5 days old)
• Irritability, jitteriness, or high-pitched cry
• Skin rash, diarrhea, or blood in stool

Parents should be given a specific list of symptoms to watch for based on the drug's known pharmacology, not a generic "call if worried" instruction. For sedating drugs (opioids, benzodiazepines, sedating antihistamines), explicit written guidance about infant arousal and feeding intervals is standard of care per ABM Protocol 28. [25]

Practical Timing Strategies to Reduce Infant Exposure

For drugs with short half-lives and high oral bioavailability, feeding timing relative to dose can reduce infant exposure by 30 to 50%:

1. Feed the infant or pump a full session immediately before taking the medication.
2. The next feed occurs when maternal plasma levels are on the decline, typically after one to two half-lives.
3. For drugs with half-lives below 4 hours (ibuprofen, most beta-lactam antibiotics), timing can be clinically meaningful.
4. For drugs with half-lives above 12 hours (fluoxetine, sertraline, methadone), timing feeds has minimal effect on total infant exposure over 24 hours.

"Pump and dump" is rarely necessary. It is indicated only when a drug with known infant toxicity must be used for a finite period, such as with radioactive contrast (wait time approximately 12 to 24 hours depending on agent) or after general anesthesia with volatile agents. [26]