Medications, Herbal
Preparations, and Natural Products in Breast Milk
As more women breastfeed and breastfeed longer in keeping with the World Health Organization (WHO) and the American Academy of Pediatrics (AAP) recommendations, questions about the safety of certain medications increase. A newer, excellent resource for information about drugs dur- ing lactation has been developed with the advice from an expert panel for the National Library of Medicine called LactMed. It is a drugs and lactation database, a peer-reviewed and fully referenced data- base of possible drugs used during lactation. The data include maternal and infant levels of drugs, possible effects on nurslings and on lactation itself, and a list of alternative drugs. The address ishttp://
toxnet.nlm.nih.gov/cgi-bin/sis/htmlgen?LACT.
With the plethora of resources about drugs, many of which are available to the lay person and mother herself, there is the risk for an untrained person mis- interpreting the data.16,22,64A major problem is even the professional untrained in lactation physiology who offers medical advice based on information gleaned from these resources. A professional needs to understand not just the plasma and milk levels but the pharmacology of the drug and physiology of lactation to give helpful instructions that will mitigate the effect of the drug on an infant and avoid discontinuing breastfeeding unnecessarily.
Despite the overwhelming advantages of human milk and the advantages of being breastfed, at times the risk of a maternal medication adversely affect- ing a nursing infant must be considered. Even when data about the medication, such as the milk/plasma
ratio, are available, a physician has to consider sev- eral factors related to each infant and each situation before deciding if breastfeeding can be initiated or continued.135 The more complicated a mother’s medical problems, the greater the possibility that the infant also has complications of prematurity or illness that will alter the ability to excrete the medication. This situation requires scientific infor- mation and experienced clinical judgment to appraise the problems and determine the therapeu- tic regimen. The clinician must determine the risk/
benefit ratio of continued breastfeeding. The data are meager and sometimes conflicting for some drugs, yet maternal medication is the single most common medical problem in managing breastfeed- ing patients reported to the Breastfeeding and Lactation Center.
There are a number of general reviews of drugs in breast milk and hundreds of articles about the effect of a specific medication in a particular infant.
The AAP Committee on Drugs22has published a list of drugs and other chemicals that transfer into human breast milk. The list, which is continually updated, is divided into those that are contraindi- cated, those that require temporary interruption of breastfeeding, and those that are compatible with breastfeeding. Concern about the issue of drugs in breast milk has spread. The U.S. Depart- ment of Health and Human Services and the Food and Drug Administration (FDA) have proposed a standard warning on all nonprescription drugs that are absorbed by the body: “As with any drug, if you 364
are pregnant or nursing a baby, seek professional advice before using this product.” Because studies of pregnant women have shown that they take five to eight medications on their own during preg- nancy and postpartum, clinicians’ education of these patients needs to continue.
A study of more than 14,000 pregnant women in 148 hospitals in 22 countries revealed that 79% of women received an average of 3.3 drugs.22 The drugs most often given were analgesics and anes- thetics. Of the 91% of women who initiated breast- feeding, 36% received methylergonovine and 5%
antibiotics. Another study of 885 women 3 to 5 months postpartum in Oslo showed that breast- feeding women took fewer medications (daily dose/
1000 women/day) than nonbreastfeeding women.92 The most common medication in the latter group was oral contraceptives. Colds, dyspepsia, hem- orrhoids, and breast infections were the disorders that precipitated the use of albuterol (salbutamol), clemastine fumarate (Tavist), dexchlorpheniramine maleate (cold preparations), phenylpropanolamine hydrochloride (Comtrex, Dimetane), cromolyn sodium, and methotrimeprazine hydrochloride (levo- mepromazine).92
No substitute exists for specific knowledge. It is equally inappropriate to discontinue breastfeeding when it is not medically necessary as it is to continue breastfeeding while taking contraindicated drugs.
Consideration of the pharmacokinetics contrib- utes to the understanding of the problems involved.
Some reported data have been extrapolated from experiments performed on cows, goats, and rodents.
Bovine experiments have been conducted using con- tinuous infusions, which provide data on the passage of a drug into milk under certain pH and plasma levels. In an effort to explain and clarify the issues involved, the literature has oversimplified the prob- lem so that individuals lacking a background in phar- macology or pediatrics have misused the published data to draw unwarranted conclusions.
Factors that influence the passage of a drug into the milk in humans include the size of the molecule, its solubility in lipids and water, whether it binds to protein, the drug’s pH, and diffusion rates. The fol- lowing outline summarizes these factors:
I. Drug
A. Route of administration 1. Oral
2. Intravenous (IV) 3. Intramuscular (IM)
4. Transdermal drug delivery system (TDDS) B. Absorption rate
C. Half-life or peak serum time D. Dissociation constant E. Volume of distribution
II. Size of molecule III. Degree of ionization IV. pH of substrate
A. Plasma: 7.4 B. Milk: 6.8 V. Solubility
A. In water B. In lipids
VI. Protein binding more to plasma than to milk protein
Passive diffusion is the principal factor in the passage of a drug from plasma into milk. The drug may appear in an active form or as an inactive metabolite.
Finally, a most important factor that has received relatively little attention is the infant. Will the infant absorb the chemical from the intestinal tract?
If the infant absorbs the chemical, can the infant detoxify and excrete it, or will minimal amounts in the milk build in the infant’s system? Is the infant premature, small for gestational age, or high risk because of complications of the pregnancy or deliv- ery? Is the drug a material that could be safely given to an infant directly, and at what risk? What dos- ages and blood levels are safe? These latter two questions are more critical than the pharmacoki- netic theory. The ultimate question faced by the physician is, “Can this infant be safely exposed to this chemical as it appears in breast milk without a risk that exceeds the benefits of being breastfed?”
Almost any drug present in a mother’s blood will appear to some degree in her milk.
Characteristics of Drugs
PROTEIN BINDING
Drugs entering the circulation become protein bound or remain free in the circulation. The protein-bound component of the drug serves as an inactive reservoir for the drug that is in equilibrium with the free drug. Most drugs enter the mammary alveolar cells in the unbound form (Figure 12-1).
At term, plasma proteins may be reduced and the fatty acid and hypoprotein fraction slightly increased in the mother, which results in the dis- placement of some drugs from plasma proteins.
During the early postpartum period, for 5 to 7 weeks the free fraction of some drugs increases and there- fore more readily crosses into milk (e.g., salicylate, phenytoin, diazepam).
For most drugs, a higher concentration will be found in the plasma than in the milk. Only the small free fraction of drug can cross the biologic mem- brane. The total concentration in milk is only min- imally influenced by binding of drugs in milk
proteins (milk protein concentration is 0.9% in mature milk). Only those drug molecules that are free in solution can pass through the endothelial pores, either by diffusion or by reverse pinocytosis.
Pinocytosisis the process whereby drug molecules dissolved in the interstitial fluid attach to receptors located at the surface of the cell membrane. The cell membrane invaginates at the site of the drug attachment, bringing the drug into the cell. The membrane is pinched off, and the drug, surrounded by membrane, remains in the cell. Then the mem- brane is dissolved, leaving the drug molecule free in the cell.
Reverse pinocytosis is the process by which the apical membrane evaginates after fusion of the intracellular membrane-bound secretion granules with the plasma membrane. The granules include lipids, proteins, lactose, drug molecules, and other cellular constituents. The evagination of the plasma membrane is pinched off and released into the alveolar lumen. Within the extravascular space, the drug may be bound to proteins in the intersti- tial fluid. Some agents in free solution can pass into the alveolar milk directly by way of the spaces between the mammary alveolar cells. These paracellular areas account for a major portion of the fluid changes across the epithelium. These spaces between adjacent alveolar cells serve to carry water-soluble drugs from the tissue into the milk.
The intercellular junctions are “open” at delivery as lactation is being established and gradually
“tighten” over the next few days. The amount of drug passed into milk on day 1 is greater than on day 3 or later. The composition of the milk changes from colostrum to mature milk, altering the amount of pro- tein and fat, which could also influence drug levels in the milk. It is always important to know when plasma and milk samples were measured in relationship to the onset of lactation. Furthermore, some studies have been done on nonlactating women by pumping
enough milk to measure the drug. These “weaning samples” provide only misinformation.
Ionization
Drugs that are nonionized are excreted in the milk in greater amounts than are ionized compounds.
Depending on the pH of the solvent and the drug dissociation constant (pKa), many weak electrolytes are more or less ionized in solution. Blood plasma and interstitial fluid are slightly alkaline (pH 7.4).
Drugs that are weak acids are ionized to a greater extent in alkaline solution and are more extensively bound to protein. The amount of drug excreted from plasma (pH 7.4) to milk (pH 6.8 to 7.3, aver- age 7.0) depends on the pH of the compound. Thus a weakly acidic compound has a higher concentra- tion in plasma than in milk. Conversely, weakly alkaline compounds are in equal or higher levels in the milk than in the plasma.
The degree of drug ionization changes with the pH of the plasma and milk. Weak bases become more ionized with decreasing pH; thus the ionized component will increase in milk. The concentration in plasma and milk for the nonionized fraction will be the same, but the total amount of drug in the milk will be greater than in plasma. The sulfonamides demon- strate the effect of the pKaon the concentration of drug that reaches the milk. Sulfacetamide, with a low pKa (5.4), has a low milk/plasma (M/P) ratio (0.08), whereas sulfanilamide has a pKaof 10.4 and an M/P ratio of 1.00 (Table 12-1).
The studies done in cows and goats with con- stant infusions demonstrate this principle more dra- matically because the pH of bovine plasma is 7.4 to 7.5 and the pH of bovine milk is 6.5. Under normal circumstances, however, concentrations of drugs are rarely constant, and there is a delay in achieving a new equilibrium. During periods of rapidly decreasing blood levels, some back diffusion occurs into the plasma.
Metabolism
Liver Brain Kidney Breast Bone Fat
Distribution to tissues
Inter action
Excretion Excretion Stor
age
Stor Storage age
Excretion
Membrane
Gut Lungs Skin IM IV
Absorption Bloodstream
(bound and unbound)
Figure 12-1. Distribution pathways for drugs once absorbed during lactation. (Modified from Rivera- Calimlim L: The significance of drugs in breast milk,Clin Perinatol 14:51, 1976.)
Molecular Weight
The passage of molecules into the milk also depends on the size of the molecule, or the molec- ular weight (mol wt, in daltons). Water-filled mem- branous pores permit the movement of molecules of less than 100 mol wt. Because of action similar to the limitation of transport of certain large- molecular-weight chemicals across the placenta, insulin and heparin are not found in human milk either, presumably because of the molecule’s size.
Solubility
The alveolar epithelium of the breast is a lipid bar- rier that is most permeable in the first few days of lactation, when colostrum is being produced. The solubility of a compound in water and in lipid is a determining factor in its transfer. Nonionized drugs, which are lipid soluble, usually dissolve and descend in the lipid phase of the membrane.
The solubility is closely linked to the manner in which the drug crosses the membranes (Table 12-2). The membrane of the alveolar epithe- lial cells is composed of lipoprotein, glycolipid, phospholipid, and free lipids, as described in Chapter 4. The transfer of water-soluble drugs
and ions is inhibited by this hydrophobic barrier.
Water-soluble materials pass through pores in the basement membrane and paracellular spaces. Low lipid solubility of a nonionized compound will diminish its excretion into milk.
Lipid solubility affects the profile of the drug in the milk and plasma. A drug with high lipid solubil- ity will have parallel elimination curves in the plasma and the milk. A drug with low lipid solubil- ity will clear the plasma at a constant rate, but the clearance curve for the milk will peak lower and later, and the drug will linger in the milk. A pro- longed terminal elimination phase may exist when time between feedings is long.
Mechanisms of Transport
Drugs pass into milk by simple diffusion, carrier- mediated diffusion, or active transport, as follows:
Simple diffusion: Concentration gradient decreases Carrier-mediated diffusion: Concentration gradient
decreases
Active transport: Concentration gradient increases
Pinocytosis
Reverse Pinocytosis. Pharmacokinetic principles relate to the specific variation with time of the drug concentration in the blood or plasma as a result of its absorption, distribution, and elimination. Ulti- mately, by extrapolation of these factors, one deter- mines the effect of the drug. The most elementary kinetic model is based on the body as a single com- partment. Distribution of the drug in the compart- ment is assumed to be uniform and rapidly equilibrated. In the single-compartment model, the volume of distribution of a drug is considered to be the same as that of the plasma, assuming a rapid uniform distribution.43The volume of distri- bution (Vd) is calculated as follows:
Vd ẳ Total amount drug in body Concentration of drug in plasma The absorption and elimination are considered to be exponential or first-order kinetics. A two- compartment model of drug kinetics takes into account the phase of decreasing drug concentration as the drug distributes into the tissues. Initially, concentrations fall rapidly as the drug distributes, then first-order elimination follows. When consid- ering the pharmacokinetics of drugs in breast milk, one must also consider that elimination in the breast is by two potential routes: excreted with the milk to the infant and back diffusion into the plasma to reequilibrate with the falling level in the plasma.
TABLE 12-1 Association Between Milk/Plasma (M/P) Ratios and Dissociation Constants (pKa) of Sulfonamides Sulfonamide Milk/Plasma Ratio pKa
Sulfacetamide 0.08 5.4
Sulfadiazine 0.21 6.5
Sulfathiazole 0.43 7.1
Sulfamethazine 0.51 7.4
Sulfapyridine 0.85 8.4
Sulfanilamide 1.00 10.4
Modified from Gaginella TS: Drugs and the nursing mother- infant,US Pharm3:39, 1978.
TABLE 12-2 Predicted Distribution Ratios of Drug Concentrations in Milk and Plasma
General Drug Type Milk/Plasma (M/P) Ratio Highly lipid-soluble drugs 1
Highly protein-bound drugs
in maternal serum <1
Small (mol wt<200) water-
soluble drugs 1
Weak acids 1
Weak bases 1
Actively transported drugs >1
Modified from Gaginella TS: Drugs and the nursing mother- infant,US Pharm3:39, 1978.
With access to the volume of distribution of the drug in question, the amount of the dose, and the weight of the mother, the concentration of drug in breast milk could be theoretically calculated as follows:
Concentration in breast milk
ẳ Dose
Volume of distribution
Other models have been developed for measur- ing the amount of drug that reaches the infant when the M/P ratio is not known. Using a stepwise linear regression for acidic and basic drugs, based on the drug’s pKa, the plasma protein binding value, and the octanol/water partition coefficient, an M/P ratio can be calculated. In a study of several proposed equations, the error is lowest for the drugs with the highest M/P ratio, that protein binding is the most important single predictor, and that the M/P ratios for basic drugs are more accurately predictable.
The concentration of the drug in the circulation of the mother depends on the mode of administra- tion: oral, IV, IM, or TDDS. Absorption through the skin, the lungs (inhalants), or vaginally may also need to be considered.
The levels in the blood depend on the route of administration. The curves produced by bolus IV medication peak high and early and taper sharply, thus making avoiding peak plasma levels more fea- sible. Absorption from IM dosing is less rapid but follows a similar but less sharp curve. Oral dosing depends on other factors, such as whether the med- ication is taken between or during meals. Depend- ing on the curve of uptake and removal of drug from the plasma, the area under the curve varies. Single doses are simple area-under-the-curve calculations, but calculations for multiple doses or chronic use vary with the steady state of the drug in the body.
TDDS patches deliver the medication at a constant rate continuously.
Nonelectrolytes such as ethanol, urea, and anti- pyrine enter the milk by diffusion through the lipid membrane barrier and may reach the same concen- trations in the milk as in the plasma, regardless of the pH. The main entrance site of molecules is at the basement laminal membrane, where water- soluble materials pass through the alveolar pores.
Nonionized drugs cross the membrane more easily than ionized ones because of the structure of the membrane. The nonionized drugs pass through the membrane by diffusion. When simple diffusion takes place, the M/P ratio is 1.0. Passive diffusion provides the same ratio regardless of the plasma concentrations of the drug or the volume of milk secreted. Different M/P ratios depend on the bind- ing to protein and are a measure of the protein-free fraction. The dissimilar ratios for the sulfa drugs
(see Table 12-1) partly result from the difference in protein binding and partly from ionization.
Large molecules depend on their lipid solubility and ionization to cross the membrane, because they pass in a lipid-soluble nonionized form. The M/P ratio is determined when equilibrium exists in the amount of nonionized drug in the aqueous phase on both sides of the membrane. When drugs are only partially ionized, the nonionized fraction determines the concentration that crosses the membrane. The drugs for which the nonionized fraction is not very lipid soluble will pass only in limited degree into breast milk.
Passive drug transport may occur in the form of facilitated diffusion. The active compound is trans- ported across the cell membrane by a carrier enzyme or protein. The gradient is toward a lesser or equal concentration in both simple diffusion and facilitated diffusion and is controlled by chemical activity gradients. Facilitated diffusion usually involves a water-soluble substance too large to pass through the membrane pores.
Active transport mechanisms provide a process whereby the gradient is “uphill,” or higher, in the milk. The process is similar to facilitated diffusion except that metabolic energy is required to over- come the gradient. Examples of substances actively transported include glucose, amino acids, calcium, magnesium, and sodium. Pinocytosis and reverse pinocytosis, as described previously, are involved in the transport of very large molecules and pro- teins. Chloride ions are secreted into milk via an active apical membrane pump, whereas sodium and potassium are diffused by electrical gradient.
Because the level of sodium is kept low, an active return of sodium may occur into the plasma, referred to as a reverse pump. The TDDS depends on absorption of the drug through the skin at a steady rate; it has become a significant route of administration for certain medications. The deliv- ery rate is determined by diffusion of drug from the reservoir matrix through the epidermis. This method offers some advantages, including conve- nience of dosing, reduced dosing frequency, ease of reaching a steady state, increased patient compli- ance, avoidance of first-pass hepatic biotransforma- tion, avoidance of peaks and valleys in blood levels, and reduction of side effects through heightened selectivity of drug action.71The level in the plasma remains constant during the drug’s anticipated life span while the patch is in place. The technology is limited to drugs with low molecular weight that are hydrophilic and can diffuse through the stra- tum corneum. The top molecular weight is 500 dal- tons. For patient compliance and economics the patch size is limited to 50 cm in diameter. Occa- sional patients experience skin irritation. Currently patches are limited to drugs that are potent in small