1. BREASTFEEDING MEDICINE
Volume 3, Number 1, 2008
© Mary Ann Liebert, Inc.
DOI: 10.1089/bfm.2007.9978
ABM Protocols
ABM Clinical Protocol #18: Use of Antidepressants
in Nursing Mothers
THE ACADEMY OF BREASTFEEDING MEDICINE PROTOCOL COMMITTEE
A central goal of The Academy of Breastfeeding Medicine is the development of clinical proto-
cols for managing common medical problems that may impact breastfeeding success. These pro-
tocols serve only as guidelines for the care of breastfeeding mothers and infants and do not de-
lineate an exclusive course of treatment or serve as standards of medical care. Variations in
treatment may be appropriate according to the needs of an individual patient.
BACKGROUND medication to the infant and mother, and the
benefits of treatment. This Protocol discusses
With estimates of between 5% and 25% of the importance of actively screening for and,
women experiencing depression in the post- when present, making the diagnosis of post-
partum year,1–3 it is critical that healthcare partum depression, how treatment can be de-
providers consider all treatment options, in- termined, and specifically addresses the med-
cluding the risks and benefits for nursing moth- ications for which there is sufficient evidence
ers. Many healthcare providers recognize the to make recommendations and provide data
short- and long-term negative effects that post- (selective serotonin reuptake inhibitors [SSRIs]
partum depression can have on mothers and and tricyclic antidepressants [TCAs]/hetero-
infants.4–6 Despite this, postpartum depression cyclics). We recognize that this is a complex is-
often goes undetected and untreated.2 Post- sue, and that there are many other factors that
partum depression is a treatable illness. Treat- impact the care of women with postpartum de-
ment options include psychotherapy (cognitive- pression, but which are beyond the scope of
behavioral, interpersonal psychotherapy),7–9 this protocol to discuss.
antidepressants,8,10,11 or a combination of med-
ication and therapy.8 The choice and approach
to treatment can be influenced by many factors, CLINICAL APPROACH TO IDENTIFYING
including the mother’s wish to breastfeed. POSTPARTUM DEPRESSION
Women may not receive medication, or receive
inadequate doses, because they are breastfeed- Postpartum depression is often missed by
ing, or may decide not to breastfeed because providers and mothers.2,12–14 The symptoms
they are concerned about medication use dur- of depression—depressed mood, sleep disrup-
ing lactation. Full consideration must be given tion, weight loss, fatigue, difficulty concentrat-
to the risks of untreated depression, risks of the ing, anxiety, loss of interest in usual activities—
44

2. ABM PROTOCOLS 45
can be difficult for mothers and providers to benefit of providing lasting changes in coping
distinguish from the normal experiences of skills and adaptation to the new role of moth-
new mothers. It is also important to differenti- erhood.
ate mothers suffering from postpartum de- If psychotherapy is unavailable or unaccept-
pression from those with postpartum blues as able to the mother, or the symptoms are severe,
misdiagnosis of such mothers can lead to un- antidepressants are an effective option. The ap-
necessary treatment. To distinguish symptoms proach to choosing an antidepressant is based
of depression from the “baby blues,” the tim- on a variety of factors. No antidepressant is
ing ( 2 weeks in duration, all day nearly every
day) and the severity (functional impairment) TABLE 1. RECOMMENDATIONS FOR IDENTIFYING WOMEN
must be evaluated.15 WITH POSTPARTUM DEPRESSION
For many women, acknowledgement of
feelings other than happiness following the • The preferred method for identifying women with
postpartum depression is the systematic use of a
birth of their infant can be devastating and validated screening tool such as the Edinburgh
embarrassing. If mothers have thoughts of Postnatal Depression Scale16 or the Postpartum
harming themselves or their infant, they are Depression Screening Scale17–19 at the obstetrical
postpartum visit and at well childcare visits in the
often afraid to bring these issues to their ob- postpartum year.
stetrician, family physician, pediatrician, mid- • Ask mothers if they feel down or anxious. Many
wife, child health nurse, or other healthcare women with postpartum depression report anxiety
as a primary symptom rather than depressed mood
professional for fear that they will be labeled or anhedonia. Excessive worrying about the baby’s
“crazy” or that their children will be taken or mother’s health should be explored.
away. Therefore, many women will not bring • Ask mothers if they are having trouble sleeping even
up their concerns or even identify them as a when they are exhausted and their child is
sleeping20 or if they are sleeping all the time and are
problem unless providers ask specific ques- unable to get out of bed.
tions or use a screening tool (see Table 1). De- • Ask mothers if they are losing or gaining weight.
pending upon the setting or the country, it is Many women with postpartum depression report a
poor appetite, but they eat because they need to
common for women to receive their peripar- keep their strength up or for nursing. Some mothers
tum and postpartum care from healthcare will gain weight.
providers other than physicians. In these cir- • Ask mothers directly but in an open, non-
threatening manner about thoughts or fears of
cumstances, communication between physi- harming their children. For example, “Many new
cians and these other healthcare providers mothers experience anxiety about their new infants.
may be crucial to making an accurate diagno- They may have thoughts that are unusual or frighten
sis and initiating timely treatment. them such as fears that they may harm their baby.
Does this ever happen to you?”21 Mothers who
experience intrusive thoughts do not wish to harm
their children and avoid the topics of their fears (i.e.,
CLINICAL APPROACH TO TREATING a mother is afraid her baby will drown therefore will
not bathe the baby and has her partner bathe the
POSTPARTUM DEPRESSION infant). It is important to distinguish the woman
with postpartum depression whose intrusive
Once a woman is identified as suffering from thoughts or fears of harming the infant are
postpartum depression, the choice of treatment incongruent with the mother’s wish to keep her
infant safe from the woman with postpartum
must be considered. While no treatment is an psychosis who is delusional and who may have
option, it is not the preferred approach. Post- thoughts of harming her infant to “save the infant
partum depression may last for months to from the devil or a life of torment.” Delusional
mothers are at great risk of harming their infants or
years and can have long-term effects for the themselves and must be immediately evaluated by a
health and well-being of mothers and in- psychiatrist.22
fants.4,5,9,23 In breastfeeding women with mild • Ask mothers if they have concerns or questions
about adapting to a new baby.
to moderate depression, the first-line treatment, • Consider the mother’s interactions with the infant,
if available, is psychotherapy. Psychotherapy including the responsiveness of mom and baby.
can be an effective treatment for women with • Difficulty in breastfeeding, or not enjoying
postpartum depression and carries no risks for breastfeeding, may be a warning sign that should be
further evaluated.
the infants. Psychotherapy may also have the

3. 46 ABM PROTOCOLS
proven safer or more effective than another in CLINICAL FACTORS AFFECTING
the postpartum period or during lactation. The ANTIDEPRESSANT CHOICE
majority of drugs including all antidepressants
are excreted in breastmilk. Data to inform clin- • There is no algorithm for antidepressant
ical decisions are derived primarily from case treatment choices in postpartum or lactating
reports or case series. Therefore, the initial women; however, articles by experts in the
treatment choice should be based on an in- field provide clinical guidance.28,29
formed clinical approach that takes into account • Obtain a history of previous antidepressant
the patient’s previous treatments for depression, treatment. In general, if a treatment was ef-
the targeted symptoms, family history of de- fective in the past and was tolerated, and
pression and their experiences with antide- there are no current contraindications, it is
pressants, current and past medical disorders, the likely first choice of treatment.
current medications, allergies, side effects of • Obtain a family history of treatment of de-
the medications, and maternal wishes. An in- pression. An immediate family member’s
dividualized risk-benefit analysis of the treat- history may be indicative of the mother’s
ments must be conducted (see Table 2).28 treatment response.
• Consider the primary symptoms that the
medication will be targeting and the poten-
tial side effect profile of the antidepressant.
TABLE 2. RISKS AND BENEFITS OF ANTIDEPRESSANT
TREATMENT IN LACTATING WOMEN For example, if the mother is particularly
anxious, a medication that might heighten
• The risks of untreated postpartum depression anxiety would not be the first choice. If the
include:
mother is experiencing hypersomnia, a med-
° persistence of symptoms
° possible increase in severity of symptoms, ication with sedation as a side effect would
including deterioration in functioning, and not be the first choice. If a mother has so-
thoughts (or even actions) of self-harm or harm to matic complaints such as nausea or diarrhea,
others
° relationship discord
a medication that may induce diarrhea
° impaired parenting would not be the first-line treatment.
° child neglect
° effects on the child’s development (including
behavior, social, and cognitive).
• The risks of treatment with antidepressants include: CHOOSING AN ANTIDEPRESSANT
° Maternal: side effects of the medication, potential DURING BREASTFEEDING
drug interactions
° Infant: exposure through breastmilk transmission;
limited data on the long-term effects on child When considering the use of any medication
development in a lactating woman, providers must consider
• The benefits of treatment include the resolution of the factors that influence infant serum levels,
depressive symptoms that, in turn, will potentially
improve maternal self-esteem, parenting, maternal- the most accurate measure of infant exposure.
infant interaction, and child outcomes.24 Factors affecting the passage of medication into
• The medical and psychological benefits of breastmilk must be considered (route of ad-
breastfeeding to infants and mothers are well
established.25–27 Depressed mothers may benefit ministration, absorption rate, half-life and peak
additionally from breastfeeding because of the sense serum time, dissociation constant, volume of
of accomplishment and active, positive participation distribution, molecular size, degree of ioniza-
in their infant’s care and development.
• The risks of breastfeeding for depressed mothers
tion, pH of plasma [7.4] and milk [6.8], solu-
should be considered and may include: bility of the drug in water and in lipids, greater
° Sleep deprivation due to total dependence on the binding to plasma protein than to milk pro-
mother may exacerbate or precipitate depressive tein), factors affecting the amount of drug re-
symptoms.
° If mothers are nursing and taking medications, the ceived by the infant (milk yield, colostrum vs.
feelings of guilt and anxiety associated with mature milk, concentration of the drug in the
exposing their infant to medication may exacerbate milk, how well the breast was emptied during
their depressive symptoms.
the previous feeding), and an infant’s ability to

4. ABM PROTOCOLS 47
absorb, detoxify, and excrete the drug. Up-to- testinal distress, headaches, sexual dysfunc-
date information about medication use during tion, nervousness, or sedation. Except for flu-
lactation is available on TOXNET lactmed at oxetine, which has a half-life of 4–6 days, most
http://toxnet.nlm.nih.gov. SSRIs have a half-life of 24–48 hours. A newer,
Most antidepressant studies provide milk related class of antidepressants, selective sero-
levels, or milk to mother’s plasma ratio, that tonin and norepinepherine reuptake inhibitors
are not constant and depend on factors such as (SSNRIs or SNRIs), are becoming more widely
dose, frequency, duration of dosing, maternal used because of what appears to be better effi-
variation in drug disposition, drug interac- cacy with fewer side effects, especially for neu-
tions, and genetic background. Few studies ropathic pain. Since the SSRIs have been in use
provide infant serum levels, although they are longer and there are more data concerning lac-
the best measure of infant exposure. Most stud- tation, this discussion will focus on the SSRIs.
ies suggest that infant daily dosages (calculated All SSRIs have been detected in breastmilk,
based on maternal dose and milk levels) are although paroxetine31–33 and sertraline33–38
safest if the level is 10% or less of the “thera- usually produced undetectable infant serum
peutic dose for infants (or the adult dose stan- levels.30 Neither of these medications has been
dardized by weight).” found to exceed the recommended 10% mater-
nal level. In contrast, fluoxetine,39–44 in 22% of
cases, and citalopram,33,45,46 in 17% of cases,
SPECIFIC ANTIDEPRESSANTS have exceeded the 10% maternal level.30 There
are virtually no case reports of escitalopram
Data from a recent meta-analysis indicated and few case reports of fluvoxamine47–52 in
that all antidepressants were detected in breast- nursing mothers, most likely because escitalo-
milk but not all were found in infant serum.30 pram is only recently available and fluvoxam-
Infant serum levels of nortriptyline, paroxetine, ine was indicated for obsessive compulsive dis-
and sertraline were undetectable in most cases. order, not depression, and therefore is not used
Infant serum levels of citalopram and fluoxe- as frequently. In most studies, no infant ad-
tine exceeded the recommended 10% maternal verse events are reported for any of these
level in 17% and 22% of cases, respectively. Few medications. The few infant adverse events
adverse outcomes are reported for any of the reported include uneasy sleep, colic, irritabil-
antidepressants. There were an insufficient ity, poor feeding, and drowsiness.53–55 In one
number of cases for all other antidepressants to case, an infant seizure was reported during the
make conclusions. time that the mother was taking fluoxetine.56
However, the relationship of fluoxetine to the
reported seizure was confounded by other
SSRIs medication exposures, and infant serum con-
centration was not obtained.
The SSRIs are the most widely prescribed Although the association between fluoxetine
antidepressant class and include citalopram and observed effects is uncertain, the long-term
(20–60 mg), escitalopram (10–20 mg), fluoxe- effects on neurobehavior and development
tine (20–80 mg), fluvoxamine (50–300 mg), from exposure to this potent serotonin reup-
paroxetine (20–60 mg), and sertraline (50–200 take blocker, or any of the SSRIs, during a pe-
mg). SSRIs improve depression and anxiety by riod of rapid central nervous system develop-
blocking the serotonin transporter and thereby ment have not been adequately studied.57 In
increasing serotonin availability in the synapse. addition, the reduced weight gain identified in
The medications are usually prescribed for de- one study may have clinical significance in
pressive or anxiety disorders but may be pre- some situations, and should be monitored care-
scribed for fibromyalgia, neuropathic pain, and fully in any breastfeeding baby whose mother
premenstrual symptoms and disorders. Com- is on fluoxetine.58 The U.S. Food and Drug Ad-
mon maternal side effects include gastroin- ministration specifically advised the manufac-

5. 48 ABM PROTOCOLS
turer to revise the labeling of fluoxetine to con- Other antidepressants
tain a recommendation against its use by nurs-
Other common antidepressants include mir-
ing mothers.59 The current labeling contains
tazapine, an antidepressant that works by block-
this revision.
ing the presynaptic noradrenergic receptors that
Bearing all this information in mind, sertra-
control norepinephrine and serotonin release,
line and paroxetine are often the most likely to
venlafaxine33,64 and duloxetine, which are
be prescribed, because of their low to zero con-
SNRIs, and bupropion, which is a norepineph-
centrations in breastmilk. This is based on a
rine and dopamine reuptake inhibitor. Sporadic
presumption that there will be lower central
case reports were found for these medica-
nervous system effects compared to some of
tions,65,66 and there are an insufficient number to
the other SSRIs with higher breastmilk con-
report significant outcomes for nursing infants.
centrations.
One case report of a seizure in an infant exposed
to bupropion through breastmilk is published,
TCAs/heterocyclics but attributing causation is cautioned.67
The TCAs (amitriptyline, amoxapine,
Herbal/natural
clomipramine, desipramine, doxepin, mapro-
tiline, nortriptyline, protriptyline, and trim- St. John’s Wort, an herbal medication, has
ipramine) are one of the older classes of anti- been used for the treatment of mild to moder-
depressants. They are effective for the treatment ate depression for many years, especially in Eu-
of depressive and anxiety disorders and are of- rope. Its use as a treatment for depression is
ten used in low doses for sleep and chronic pain. controversial in the United States. Only one
The therapeutic mechanisms are most likely re- study of sufficient numbers was available for
lated to the blockade of the norepinephrine review.68 In this study there were increased
transporter, which thereby increases norepi- rates of colic, drowsiness, and lethargy in the
nephrine availability in the synapses. These St. John’s Wort group compared to controls,
medications also block the dopamine and sero- but this was confounded by concomitant anti-
tonin pumps, which may contribute to their depressant treatment in the study group. No
therapeutic mechanisms. Unfortunately, they long-term effects were noted, and no effect on
also block muscarinic cholinergic receptors, H1 milk production.
histamine receptors, and alpha-1-adrenergic re- Omega-3 fatty acids are currently being stud-
ceptors, which most likely account for their ied as a treatment for depression during preg-
wide array of unpleasant side effects. Despite nancy and the postpartum period.69 Omega-3
being effective and inexpensive they are not fatty acids appear to be of little risk to mothers
used as frequently as SSRIs because of their side and infants as they are natural essential ele-
effects, which can include hypotension, seda- ments of one’s diet and are often depleted dur-
tion, dry mouth, urinary retention, weight gain, ing pregnancy and breastfeeding. The primary
sexual dysfunction, and constipation. In addi- negative side effect is the “fishy smell” and the
tion, in an overdose, these medications can lack of sufficient evidence at this time to con-
cause cardiac arrhythmias and death. Among sider it a treatment for depression.
this class of medications, only nortriptyline has There is little or no evidence that ethnic or
a sufficient number of reported cases to com- regional “medicines” are safe or effective; thus
ment on its use during lactation. In most cases, their use by healthcare providers is strongly
nortriptyline is undetectable in infant serum. Its cautioned.
metabolite has been detected, but no adverse
events have been reported.60–62 Insufficient
numbers of cases have been reported on the RECOMMENDATIONS FOR
other medications; however, use of doxepin is ANTIDEPRESSANT TREATMENT
often cautioned because of a case report of hy- IN LACTATING WOMEN
potonia, poor feeding, emesis, and sedation in
a breastfeeding infant that resolved after dis- • Current evidence suggests that the risks of
continuation of nursing.63 untreated maternal depression can have se-

6. ABM PROTOCOLS 49
rious and long-term effects on mothers and and benefits of the treatment to make an in-
infants and that treatment may improve out- formed decision.
comes for mothers and infants. Therefore • Mothers should be monitored carefully in
treatment is strongly preferred. the initial stages of treatment for changes in
• However, it is important not to label moth- symptoms, including worsening of symp-
ers who are only suffering from mild cases toms. Specifically, women with histories of
of “baby blues” as “depressed.” We must bipolar disorder, which may be undiag-
make a distinction. If symptoms are mild, nosed, are at increased risk of developing a
there is no reason to initiate antidepressant mood episode of depression, mania, or psy-
medication treatment in the first 2 weeks chosis in the postpartum period. While this
postpartum. is rare, mothers and partners should be
• When available and when symptoms are in made aware of the symptoms to watch for
the mild–moderate range, psychotherapy is such as increased insomnia, delusions, hal-
the first line of treatment for lactating lucinations, racing thoughts, and talking/
women as it carries no known risk for the in- moving fast and contact their mental health
fant. Mothers must be monitored and re- provider immediately.
evaluated. If they are not improving or their • Infants should be evaluated prior to the ini-
symptoms are worsening, antidepressant tiation of a new medication during breast-
drug treatment must be considered. feeding and monitored carefully by the pe-
• Psychotherapy in addition to antidepressant diatrician, including carefully following
medication is recommended for women with growth. Serum levels are not indicated on a
severe symptoms. regular basis without a clinical indication or
• Women with moderate to severe symptoms concern.
may request only antidepressant drug treat- • Strategies that may be used to decrease in-
ment, and this must be considered as the fant exposure, but for which there is little ev-
benefits of treatment likely outweigh the idence, include medication administration
risks of the medication to the mother or in- immediately after feedings and pumping
fant. and discarding the breastmilk obtained dur-
• There is no widely accepted algorithm for ing the peak serum levels.
antidepressant medication treatment of de-
pression in lactating women. An individual-
ized risk-benefit analysis must be conducted CONCLUSIONS AND SUGGESTIONS
in each situation and take into account the FOR FUTURE RESEARCH
mother’s clinical history and response to
treatment, the risks of untreated depression, Despite many publications of antidepres-
the risks and benefits of breastfeeding, the sants and breastfeeding, the scientific literature
benefits of treatment, the known and un- lacks both the breadth and depth for clinicians
known risks of the medication to the infant, and mothers to make confident decisions about
and the mother’s wishes. individual medications. Multiple reviews of
• If a mother has no history of antidepressant the literature broadly suggest TCAs and sero-
treatments, an antidepressant, such as parox- tonin reuptake inhibitors are relatively safe,
etine or sertraline, that has evidence of lower and all recommend individual risk-benefit as-
levels in breastmilk and infant serum and sessments.
few side effects is an appropriate first choice. The literature suffers from a lack of any ran-
• If mothers have been successfully treated domized clinical trials in lactating women for
with a particular SSRI, TCA, or SNRI in the any class of antidepressant. The majority of
past, the data regarding this particular anti- studies are case reports or case series, and most
depressant should be reviewed, and it have small samples sizes. Those studies that re-
should be considered as a first-line treatment port larger samples (n 25) primarily report
if there are no contraindications. a variety of medications. Only six controlled
• Mothers should be provided the information studies (one retrospective,70 five prospec-
regarding the known and unknown risks tive42,45,46,54,71) were found that used a variety

7. 50 ABM PROTOCOLS
of controls—some control for depression, while 7. O’Hara MW, Stuart S, Gorman LL, et al. Efficacy of
others do not. None of the studies sufficiently interpersonal psychotherapy for postpartum depres-
sion. Arch Gen Psychiatry 2000;57:1039–1045.
controlled for level of depression. In addition,
8. Appleby L, Warner R, Whitton A, et al. A controlled
the case reports are limited by confounding study of fluoxetine and cognitive-behavioral counsel-
with in utero exposure, the range of infant ages, ing in the treatment of postnatal depression. BMJ
inconsistencies in the timing of when samples 1997;314:932–936.
were obtained, lack of information about the 9. Murray L, Cooper PJ, Wilson A, et al. Controlled trial
amount of medication in foremilk versus of the short- and long-term effect of psychological
treatment of post-partum depression: 2. Impact on the
hindmilk, and no information about infant con- mother-child relationship and child outcome. Br J Psy-
sumption as average breastmilk volumes are chiatry 2003;182:420–427.
not provided. The majority of studies provide 10. Cohen LS, Viguera AC, Bouffard SM, et al. Venlafax-
information about the amount of medication ine in the treatment of postpartum depression. J Clin
detected in breastmilk and maternal serum. Psychiatry 2001;62:592–596.
11. Suri R, Burt VK, Altshuler LL, et al. Fluvoxamine for
Some studies also provide information about
postpartum depression. Am J Psychiatry 2001;158:
infant serum levels of medication. Few studies 1739–1740.
report infant behavioral outcomes. 12. Bagedahl-Strindlund M, Monsen BK. Postnatal de-
pression: A hidden illness. Acta Psychiatr Scand
1998;98:272–275.
13. Morris-Rush JK, Freda MC, Bernstein PS. Screening
ACKNOWLEDGMENTS
for postpartum depression in an inner-city popula-
tion. Am J Obstet Gynecol 2003;188:1217–1219.
This work was supported in part by a grant 14. Heneghan AM, Silver EJ, Bauman LJ, et al. Do pedi-
from the Maternal and Child Health Bureau, atricians recognize mothers with depressive symp-
U.S. Department of Health and Human Ser- toms? Pediatrics 2000;106:1367–1373.
vices. 15. American Psychiatric Association. Diagnostic and Sta-
tistical Manual of Mental Disorders-IV-TR, 4th edition.
American Psychiatric Publishing, Washington, DC,
2000.
REFERENCES 16. Cox JL, Holden JM, Sagovsky R. Detection of postna-
tal depression. Development of the 10-item Edin-
1. O’Hara MW, Neunaber DJ, Zekoski EM. Prospective burgh Postnatal Depression Scale. Br J Psychiatry 1987;
study of postpartum depression: Prevalence, course, 150:782–786.
and predictive factors. J Abnorm Psychol 1984;93: 17. Beck CT, Gable RK. Postpartum Depression Screen-
158–171. ing Scale: development and psychometric testing.
2. Chaudron LH, Szilagyi PG, Kitzman HJ, et al. De- Nurs Res 2000;49:272–282.
tection of postpartum depressive symptoms by 18. Beck CT, Gable RK. Comparative analysis of the per-
screening at well-child visits. Pediatrics 2004;113: formance of the Postpartum Depression Screening
551–558. Scale with two other depression instruments. Nurs Res
3. Gaynes BN, Gavin N, Meltzer-Brody S, et al. Perina- 2001;50:242–250.
tal Depression: Prevalence, Screening Accuracy, and Screen- 19. Beck CT, Gable RK. Further validation of the Post-
ing Outcomes. Evidence Report/Technology Assess- partum Depression Screening Scale. Nurs Res 2001;50:
ment No. 119. Prepared by the RTI–University of 155–164.
North Carolina Evidence-Based Practice Center, un- 20. Chaudron LH, Klein MH, Remington P, et al. Predic-
der Contract Number 290-02-0016. AHRQ Publication tors, prodromes and incidence of postpartum de-
Number 05-E006-2. Agency for Healthcare Research pression. J Psychosom Obstet Gynecol 2001;22:103–112.
and Quality, Rockville, MD, 2005. 21. Wisner KL, Peindl KS, Gigliotti T, et al. Obsessions
4. Murray L, Hipwell A, Hooper R, et al. The cognitive and compulsions in women with postpartum de-
development of 5-year-old children of postnatally de- pression. J Clin Psychiatry 1999;60:176–180.
pressed mothers. J Child Psychol Psychiatry 1996;37: 22. Spinelli MG. A systematic investigation of 16 cases of
927–935. neonaticide [see comment]. Am J Psychiatry 2001;158:
5. Murray L, Sinclair D, Cooper P, et al. The socioemo- 811–813.
tional development of 5-year-old children of postna- 23. Sinclair D, Murray L. Effects of postnatal depression
tally depressed mothers. J Child Psychol Psychiatry on children’s adjustment to school. Teacher’s reports.
1999;40:1259–1271. Br J Psychiatry 1998;172:58–63.
6. Kahn RS, Zuckerman B, Bauchner H, et al. Women’s 24. Weissman MM, Pilowsky DJ, Wickramaratne PJ, et al.
health after pregnancy and child outcomes at age 3 Remissions in maternal depression and child psy-
years: a prospective cohort study. Am J Public Health chopathology: a STAR*D-child report. JAMA 2006;
2002;92:1312–1318. 295:1389–1398.

8. ABM PROTOCOLS 51
25. Scariati PD, Grummer-Strawn LM, Fein SB. A lon- 43. Hendrick V, Stowe ZN, Altshuler LL, et al. Fluoxe-
gitudinal analysis of infant morbidity and the extent tine and norfluoxetine concentrations in nursing in-
of breastfeeding in the United States. Pediatrics fants and breast milk. Biol Psychiatry 2001;50:775–782.
1997;99:e5. 44. Suri R, Stowe ZN, Hendrick V, et al. Estimates of nurs-
26. Dell S, To T. Breastfeeding and asthma in young chil- ing infant daily dose of fluoxetine through breast
dren: findings from a population-based study. Arch milk. Biol Psychiatry 2002;52:446–451.
Pediatr Adolesc Med 2001;155:1261–1265. 45. Heikkinen T, Ekblad U, Kero P, et al. Citalopram in
27. Dignam DM. Understanding intimacy as experienced pregnancy and lactation. Clin Pharmacol Ther 2002;2:
by breastfeeding women. Health Care Women Int 184–191.
1995;16:477–485. 46. Lee A, Woo J, Ito S. Frequency of infant adverse events
28. Burt VK, Suri R, Altshuler L, et al. The use of psy- that are associated with citalopram use during breast-
chotropic medications during breast-feeding. Am J feeding. Am J Obstet Gynecol 2004;190:218–221.
Psychiatry 2001;158:1001–1009. 47. Arnold LM, Suckow RF, Lichtenstein PK. Fluvoxam-
29. Hendrick V, Burt VK, Altshuler LL. Psychotropic ine concentrations in breast milk and in maternal and
guidelines for breast-feeding mothers. Am J Psychia- infant sera. J Clin Psychopharmacol 2000;20:491–493.
try 1996;153:1236–1237. 48. Hagg S, Granberg K, Carleborg L. Excretion of flu-
30. Weissman AM, Levy BT, Hartz AJ, et al. Pooled anal- voxamine into breast milk. Br J Clin Pharmacol 2000;49:
ysis of antidepressant levels in lactating mothers, 286–288.
breast milk, and nursing infants. Am J Psychiatry 49. Hendrick V, Fukuchi A, Altshuler L, et al. Use of ser-
2004;161:1066–1078. traline, paroxetine and fluvoxamine by nursing
31. Misri S, Kim J, Riggs KW, et al. Paroxetine levels in women. Br J Psychiatry 2001;179:163–166.
postpartum depressed women, breast milk, and in- 50. Piontek CM, Wisner KL, Perel JM, et al. Serum flu-
fant serum. J Clin Psychiatry 2000;61:828–832. voxamine levels in breastfed infants. J Clin Psychiatry
32. Stowe ZN, Cohen LS, Hostetter A, et al. Paroxetine in 2001;62:111–113.
human breast milk and nursing infants. Am J Psychi- 51. Yoshida K, Smith B, Kumar RC. Fluvoxamine in
atry 2000;157:185–189. breast-milk and infant development. Br J Clin Phar-
33. Berle JO, Steen VM, Aamo TO, et al. Breastfeeding macol 1997;44:210–211.
during maternal antidepressant treatment with sero- 52. Wright S, Dawling S, Ashford JJ. Excretion of fluvox-
tonin reuptake inhibitors: Infant exposure, clinical amine in breast milk. Br J Clin Pharmacol 1991;31:209.
symptoms, and cytochrome p450 genotypes. J Clin 53. Lester BM, Cucca J, Andreozzi L, et al. Possible asso-
Psychiatry 2004;65:1228–1234. ciation between fluoxetine hydrochloride and colic in
34. Epperson CN, Anderson GM, McDougle CJ. Sertra- an infant. J Am Acad Child Psychiatry 1993;32:1253–1255.
line and breast-feeding. N Engl J Med 1997;336:1189– 54. Merlob P, Stahl B, Sulkes J. Paroxetine during breast-
1190. feeding: Infant weight gain and maternal adherence
35. Stowe ZN, Owens MJ, Landry JC, et al. Sertraline and to counsel. Eur J Pediatr 2004;163:135–139.
desmethylsertraline in human breast milk and nurs- 55. Schmidt K, Olesen OV, Jensen PN. Citalopram and
ing infants. Am J Psychiatry 1997;154:1255–1260. breast-feeding: Serum concentration and side effects
36. Epperson N, Czarkowski KA, Ward-O’Brien D, et al. in the infant. Biol Psychiatry 2000;47:164–165.
Maternal sertraline treatment and serotonin transport 56. Brent NB, Wisner KL. Fluoxetine and carbamazepine
in breast-feeding mother-infant pairs. Am J Psychiatry concentrations in a nursing mother/infant pair. Clin
2001;158:1631–1637. Pediatr 1998;37:41–44.
37. Wisner KL, Perel JM, Blumer J. Serum sertraline 57. Drug Safety Site.com. Fluoxitine. http:/ /drugsafe-
and N-desmethylsertraline levels in breast-feeding tysite.com/fluoxetine. Last accessed November 27,
mother-infant pairs. Am J Psychiatry 1998;155:690– 2007.
692. 58. Chambers CD, Anderson PO, Thomas RG, et al.
38. Dodd S, Stocky A, Buist A, et al. Sertraline analysis Weight gain in infants whose mothers take fluoxetine.
in the plasma of breast-fed infants. Aust N Z J Psy- Pediatrics 1999;104:e61.
chiatry 2001;35:545–546. 59. Nightingale SL. Fluoxetine labeling revised to iden-
39. Ito S, Koren G. Antidepressants and breast-feeding tify phenytoin interaction and to recommend against
[see comment]. Am J Psychiatry 1997;154:1174. use in nursing mothers. JAMA 1994;271:1067.
40. Kristensen JH, Ilett KF, Hackett LP, et al. Distribution 60. Wisner KL, Perel JM. Nortriptyline treatment of
and excretion of fluoxetine and norfluoxetine in hu- breast-feeding women. Am J Psychiatry 1996;153:295.
man milk. Br J Clin Pharmacol 1999;48:521–527. 61. Wisner KL, Perel JM. Serum nortriptyline levels in
41. Epperson CN, Jatlow PI, Czarkowski K, et al. Mater- nursing mothers and their infants. Am J Psychiatry
nal fluoxetine treatment in the postpartum period: 1991;148:1234–1236.
Effects on platelet serotonin and plasma drug levels 62. Wisner KL, Perel JM, Findling RL, et al. Nortriptyline
in breastfeeding mother-infant pairs. Pediatrics 2003; and its hydroxymetabolites in breastfeeding mothers
112:e425. and newborns. Psychopharmacol Bull 1997;33:249–251.
42. Heikkinen T, Ekblad U, Palo P, et al. Pharmacokinet- 63. Frey OR, Scheidt P, von Brenndorff AI. Adverse ef-
ics of fluoxetine and norfluoxetine in pregnancy and fects in a newborn infant breast-fed by a mother treated
lactation. Clin Pharmacol Ther 2003;73:330–337. with doxepin. Ann Pharmacother 1999;33:690–693.

9. 52 ABM PROTOCOLS
64. Ilett KF, Kristensen JH, Hackett LP, et al. Distribution ABM protocols expire five years from the
of venlafaxine and its O-desmethyl metabolite in hu- date of publication. Evidenced-based revi-
man milk and their effects in breastfed infants. Br J
sions are made within five years, or sooner
Clin Pharmacol 2002;53:17–22.
65. Baab SW, Peindl KS, Piontek CM, et al. Serum bupro- if there are significant changes in the evi-
pion levels in 2 breastfeeding mother-infant pairs. J dence.
Clin Psychiatry 2002;63:910–911.
66. Aichhorn WMD, Whitworth ABM, Weiss UMD, et al.
Mirtazapine and breast-feeding [letter]. Am J Psychi- Contributors
atry 2004;161:2325. *Linda H. Chaudron, M.D., M.S.
67. Chaudron LH, Schoenecker CJ. Bupropion and *Stephanie A.M. Giannandrea, B.A.
breastfeeding: A case of a possible infant seizure. J
Clin Psychiatry 2004;65:881–882.
68. Lee A, Minhas R, Matsuda N, et al. The safety of St. Protocol Committee
John’s wort (Hypericum perforatum) during breast- Caroline J. Chantry, M.D., FABM
feeding. J Clin Psychiatry 2003;64:966–968. Co-Chairperson
69. Freeman MP, Hibbeln JR, Wisner KL, et al. Random- Cynthia R. Howard, M.D., MPH, FABM,
ized dose-ranging pilot trial of omega-3 fatty acids for Co-Chairperson
postpartum depression. Acta Psychiatr Scand 2006;113:
31–35.
Ruth A. Lawrence, M.D., FABM
70. Chambers CD, Anderson PO, Thomas RG, et al. Kathleen A. Marinelli, M.D., FABM,
Weight gain in infants breastfed by mothers who take Co-Chairperson
fluoxetine. Pediatrics 1999;104:e61. Nancy G. Powers, M.D., FABM
71. Yoshida K, Smith B, Craggs M, et al. Investigation of
*Lead author
pharmacokinetics and of possible adverse effects in
infants exposed to tricyclic antidepressants in breast-
milk. J Affect Disord 1997;43:225–237. For reprint requests: abm@bfmed.org