Postpartum Depression
Clinical Snapshot
Postpartum depression (PPD) is a major depressive episode with onset during pregnancy or in the weeks to months following childbirth, affecting an estimated 13% of new mothers — making it one of the most common complications of pregnancy and the postpartum period, more common than gestational diabetes or preeclampsia.[1] PPD carries genuine, well-documented consequences for both mother and infant when it goes unrecognized or untreated: impaired mother-infant bonding and attachment, disrupted infant cognitive and socioemotional development, family and relationship strain, and a real risk of self-harm and, in the most severe cases, maternal suicide — one of the leading contributors to maternal death in the first postpartum year in high-income countries.[1][2]
Two points anchor this entire chapter. First, the Edinburgh Postnatal Depression Scale (EPDS) is the validated screening standard, and current Canadian guidance recommends it at every perinatal encounter through 12 months postpartum — not a single check at the 6-week visit.[3] Second, and the single most safety-critical distinction in this chapter: postpartum psychosis is not a severe form of postpartum depression — it is a different, rare, and genuinely emergent condition that requires immediate psychiatric escalation, never routine follow-up (see Red Flags & Refer).[3][4] For a community or perinatal pharmacist, this chapter's centerpiece is breastfeeding-safety-informed antidepressant selection — a genuinely high-yield, practical counselling opportunity, given how often an unfounded fear that "no medication is safe while breastfeeding" leads a patient to either refuse necessary treatment or stop breastfeeding unnecessarily. Health Canada's December 2025 authorization of zuranolone, the first medication approved in Canada specifically for postpartum depression, also gives pharmacists a genuinely new — and genuinely different — treatment option to understand and counsel on correctly.[5]
Recognize the Condition
"Baby blues" versus postpartum depression — timing and severity are what separate them
Transient low mood, tearfulness, irritability, and mild anxiety — the "baby blues" — affect an estimated 50-80% of new mothers, typically beginning 2-3 days after delivery, peaking around day 5, and resolving on their own by about 2 weeks postpartum without any specific treatment.[2] This is expected, hormonally driven, and does not by itself indicate depression. Two features distinguish PPD from the baby blues: persistence beyond 2 weeks, and severity that meets full major depressive episode criteria — sustained low mood or anhedonia present most of the day, nearly every day, together with associated symptoms (sleep or appetite disturbance beyond what newborn care alone explains, guilt or worthlessness, concentration difficulty, psychomotor change, or thoughts of death) causing genuine functional impairment.[6][2] A patient whose low mood, tearfulness, or anxiety persists past the 2-week mark, or who meets full depressive-episode criteria at any point, has moved out of "normal adjustment" territory and warrants a formal EPDS screen and clinical assessment (see Assessment & Diagnosis) — this should never be waved off as "just the blues, it'll pass."
Onset window and DSM-5-TR terminology
DSM-5-TR does not carry a separate "postpartum depression" diagnosis — it is major depressive disorder with the peripartum onset specifier, applied when the most recent episode began during pregnancy or within 4 weeks of delivery.[6] In practice, both screening programs and most clinical guidance use a considerably broader window — up to 12 months postpartum — since a genuine, treatable postpartum-triggered depressive episode can present well beyond the narrow 4-week DSM-5-TR technical cut-off, and the underlying assessment and management does not meaningfully differ by the exact week of onset within that first year.[3]
Typical clinical presentation
Beyond the core mood and anhedonia symptoms, PPD very often presents with prominent anxiety, a pervasive sense of inadequacy or guilt specifically framed around mothering ("I'm a bad mother," "I'm failing my baby"), overwhelming fatigue that outpaces what sleep deprivation alone would explain, and difficulty bonding with the infant. Comorbid postpartum anxiety disorders and postpartum-onset obsessive-compulsive symptoms are common and are frequently under-recognized when a clinician screens for depression alone.[3]
Intrusive thoughts — common, distressing, and fundamentally different from psychosis
Unwanted, vivid, frightening intrusive thoughts of accidental or deliberate harm coming to the infant are reported by a substantial proportion of new mothers, with or without a depression or anxiety diagnosis, and are especially common in postpartum-onset OCD and PPD.[3] The clinically vital distinguishing feature is that these thoughts are ego-dystonic — the mother finds them horrifying, is not at any real risk of acting on them, often actively avoids situations that trigger them (a classic example: avoiding kitchen knives near the baby), and retains full insight that the thought is not something she wants or endorses. This is categorically different from the rare, ego-syntonic delusional beliefs of postpartum psychosis, where the person may believe that harming the infant is necessary, justified, or instructed by an outside force — often without the same horror or insight (see Red Flags & Refer for the full distinguishing picture). Asking directly and matter-of-factly about intrusive thoughts, and explaining that they are extremely common and not a sign the patient is a danger to her baby, is itself a genuinely therapeutic, destigmatizing intervention — not just a risk-screening exercise.[3]
Risk Factors
- Personal history of depression or anxiety — the single strongest predictor, including a prior episode of PPD itself, which carries a substantial recurrence risk in a subsequent pregnancy.
- Depression or anxiety symptoms already present during the current pregnancy — antenatal depression is the strongest specific predictor of PPD.
- Personal or family history of bipolar disorder — this specifically and disproportionately raises the risk of postpartum psychosis, not simply depression (see Red Flags & Refer and Special Populations).
- Personal history of postpartum psychosis in a prior pregnancy.
- Lack of social support, significant partner conflict, or intimate partner violence.
- An unplanned or unwanted pregnancy.
- A traumatic or medically complicated birth experience, including emergency cesarean delivery.
- Infant admitted to the NICU, or an infant with a significant health complication or congenital anomaly.
- Breastfeeding difficulties or early, unwanted breastfeeding cessation.
- Significant sleep deprivation.
- Socioeconomic stressors — poverty, housing instability, lack of paid parental leave, or immigration/refugee status with limited local support.
- Adolescent parenthood.
- Multiple gestation (twins or a higher-order pregnancy).
- Thyroid dysfunction, particularly postpartum thyroiditis — screen for this as a contributing or mimicking factor (see Assessment & Diagnosis).
Red Flags & Refer
Postpartum psychosis is a distinct, rare, and genuine psychiatric emergency — not a severe presentation of ordinary PPD. It affects roughly 1-2 per 1000 births, is strongly associated with bipolar disorder (as many as 1 in 6 postpartum relapses in a patient with known bipolar disorder are psychotic, manic, or otherwise severe enough to require hospitalization), and typically has an abrupt onset within the first 2 weeks postpartum — most often within the first several days.[3][4] It carries a real, elevated risk of both suicide and infanticide.[4] Any of the following requires the same-day activation of an emergency psychiatric pathway (emergency department, or a mobile/urgent perinatal psychiatry crisis service) — this is never a "routine follow-up in 1-2 weeks" situation:
- Delusions of any kind, especially ones involving the infant (for example, a belief the infant is defective, possessed, or needs to be "saved," or that someone or something intends harm to the infant or mother).
- Hallucinations — auditory, visual, or otherwise.
- Marked confusion, disorientation, or a rapidly fluctuating mental state — a hallmark feature that distinguishes psychosis from the more gradual, stable presentation of ordinary PPD.
- Severe insomnia that persists despite a genuine opportunity to sleep — a frequently reported early/prodromal feature.
- Bizarre, disorganized, or markedly out-of-character behaviour.
- Any thought of harming the infant that is ego-syntonic — the mother believes the act would be right, justified, or necessary — as opposed to the far more common, ego-dystonic intrusive thought that horrifies her (see Recognize the Condition). Do not assume every reported thought about infant harm is the common, benign kind; assess explicitly.
- Active suicidal ideation with a plan or intent, or a positive response to EPDS item 10 (thoughts of self-harm) combined with any other concerning feature (see Assessment & Diagnosis).
Also escalate urgently — though not necessarily to the same emergency pathway as suspected psychosis unless one of the features above is also present — for: any expressed inability to safely care for the infant, evidence of infant neglect, a first presentation of significant symptoms in a patient with no existing mental health care team, or new or worsening symptoms in a patient with a known bipolar disorder history (consider urgent psychiatric reassessment for an emerging mixed or manic episode, not simply "more depression").
Postpartum psychosis and ordinary PPD can be genuinely difficult to tell apart from any single symptom viewed in isolation; it is the overall pattern — abrupt (days, not weeks) onset, a fluctuating and confused mental state, and any delusional or hallucinatory content — that should trigger emergency-level concern rather than routine reassurance.[3]
Assessment & Diagnosis
Edinburgh Postnatal Depression Scale (EPDS)
The EPDS is a 10-item, patient-completed self-report screening tool, validated specifically for the perinatal period and deliberately built without heavy reliance on somatic items (sleep, appetite, fatigue) that overlap with normal postpartum physiology.[7] Each item is scored 0-3, for a maximum total score of 30. Current Canadian guidance recommends a cut-off score of 11 or higher to maximize sensitivity for detecting depression — a lower, more sensitive threshold than the scale's original 1987 validation cut-off of 12-13; use 11 as the current, evidence-based screening threshold rather than the older figure some clinicians still reference.[3] The Patient Health Questionnaire-9 (PHQ-9), at a cut-off of 10 or higher, has comparable test characteristics and is an acceptable alternative where the EPDS is not in local use; the PHQ-2 and Whooley questions are less optimal screening tools, with only moderate specificity.[3]
EPDS item 10 asks specifically about thoughts of self-harm and must be reviewed on every completed scale regardless of the total score — a nonzero response on item 10 warrants same-visit, direct follow-up questioning about intent and safety, even when the overall score falls below the general screening cut-off.[7][2]
A positive EPDS screen is not itself a diagnosis — it identifies patients who need a full clinical assessment against major depressive disorder criteria (see below). Current Canadian guidance recommends screening once per trimester during pregnancy, at the postpartum obstetrical follow-up visit, and through routine family practice and pediatric well-child visits up to 12 months postpartum — not a single point-in-time check.[3]
Confirming the diagnosis
Diagnosis requires meeting full DSM-5-TR major depressive episode criteria, with the peripartum onset specifier applied when onset falls during pregnancy or within 4 weeks of delivery (see Recognize the Condition for the broader clinical window most guidance actually uses).[6] Distinguish PPD explicitly from: the self-limited baby blues (see Recognize the Condition); a postpartum anxiety disorder or postpartum-onset OCD, which can occur with or without comorbid depression and needs its own targeted assessment; a hypomanic, manic, or mixed episode in a patient with unrecognized bipolar disorder — screening for a personal or family history of bipolar disorder before starting an antidepressant is genuinely important, since antidepressant monotherapy can precipitate or worsen mania in undiagnosed bipolar disorder; and postpartum psychosis (see Red Flags & Refer), which is a different condition, not simply a severity tier of PPD.[3]
Basic laboratory work-up
A reasonable baseline work-up includes thyroid-stimulating hormone (TSH) — postpartum thyroiditis is common and can mimic or contribute to depressive and fatigue symptoms — and a complete blood count, given that postpartum anemia from delivery blood loss can compound fatigue and is easily correctable. Neither test is required to make the diagnosis, but both are reasonable, low-burden checks to rule out a contributing or confounding medical cause, particularly for an atypical presentation or one that does not respond as expected to initial treatment.[2]
Treatment Targets
- Achieve full symptomatic remission, not just partial symptom reduction — a partial response leaves genuine residual risk to both mother and infant.
- Keep the mother, and by extension the infant, safe — reassess for suicidality and psychosis risk at every relevant encounter, not only at the point of initial diagnosis.
- Support and preserve the breastfeeding relationship wherever the patient wishes to continue it and doing so is clinically reasonable — treatment choice should very rarely be the reason breastfeeding stops (see Special Populations).
- Restore day-to-day functioning and actively support mother-infant bonding and attachment.
- Correctly distinguish and separately manage comorbid postpartum anxiety, postpartum-onset OCD, and — where present — an underlying bipolar disorder, rather than defaulting to treating every postpartum mood symptom as unipolar PPD.
- Engage the partner and broader support system as an active part of the treatment plan, not an afterthought (see Non-Pharmacologic Treatment).
- Prevent progression to a chronic or recurrent course, and reduce the risk of recurrence in a future pregnancy.
Choose Initial Treatment
Current Canadian guidance uses illness severity, not a single fixed algorithm, to select initial treatment.[3]
Mild-to-moderate PPD can often be successfully managed with non-pharmacologic treatment alone — structured psychotherapy (CBT or interpersonal therapy), psychosocial support, and lifestyle measures — without necessarily starting medication (see Non-Pharmacologic Treatment).[3]
Moderate-to-severe PPD typically needs pharmacotherapy, generally combined with psychotherapy rather than either alone.[3]
Severe PPD, PPD with psychotic features, or a rapidly deteriorating presentation needs pharmacotherapy plus urgent psychiatric involvement, with neuromodulation (electroconvulsive therapy) a genuine option for the most severe or treatment-refractory presentations.[3]
1st line, starting fresh sertraline is generally the preferred first SSRI for a patient with no prior antidepressant trial, given its combination of standard antidepressant efficacy and the most reassuring, most extensively studied breastfeeding-safety data of any antidepressant (see Special Populations for the full breastfeeding-safety discussion).[8][9] This is a preference for a first, untried patient — not a reason to switch a patient who is already doing well on a different antidepressant.
For a patient who responded well to a specific antidepressant during the current pregnancy or a prior depressive episode, restarting or continuing that same agent postpartum is generally preferable to switching to sertraline by default — continuity avoids the relapse risk and lost time of a fresh medication trial. The specific choice can still be reviewed against the breastfeeding-safety data in Special Populations, but should not be changed reflexively.
Health Canada-authorized December 2025 — first medication approved in Canada specifically for PPD zuranolone is a rapid-acting, orally administered neuroactive steroid (a positive allosteric modulator of GABA-A receptors) given as a single 14-day course, with symptom improvement reported as early as day 3 and sustained through day 45 in trial data.[5][10] It is mechanistically and practically different from a standard antidepressant: it is not intended as ongoing maintenance therapy, retreatment with a further course has not been studied, and a plan for what happens after the 14-day course — including whether a standard antidepressant is still needed — should be made before starting it, not worked out afterward (see Treatment Failure/Next Step).[10] A 2025 Cochrane review of this drug class as a whole — covering both zuranolone and its earlier intravenous predecessor brexanolone — found these neurosteroid agents probably improve depressive response and remission compared with placebo, alongside a real increase in maternal adverse events; brexanolone itself was never marketed in Canada and has since been withdrawn from the US market entirely in favour of oral zuranolone (see Drug Tables).[11] It carries real practical counselling points: significant sedation/dizziness risk, with impairment that can last beyond 12 hours after a dose; a genuine misuse/dependence and abrupt-discontinuation withdrawal profile similar to other GABAergic CNS depressants; and it is contraindicated in pregnancy — confirm pregnancy has been ruled out and discuss effective contraception before starting.
Do not advise stopping breastfeeding as a default response to starting antidepressant therapy — for the great majority of patients this is unnecessary, and can itself worsen the loss-of-control feelings PPD already causes (see Special Populations).
Do not manage postpartum psychosis symptoms through this routine pathway — return to Red Flags & Refer and escalate immediately instead.
Do not start antidepressant monotherapy without first asking about a personal or family history of bipolar disorder or mania — unrecognized bipolar disorder is a genuine, clinically important explanation for an apparent "treatment failure" or a new manic/mixed episode after starting an antidepressant (see Treatment Failure/Next Step).
Non-Pharmacologic Treatment
Structured psychotherapy — first-line for mild-to-moderate PPD, and a valuable adjunct at any severity
Cognitive behavioural therapy (CBT) and interpersonal therapy (IPT) both have a solid evidence base specifically in the postpartum population.[3] IPT is particularly well suited to PPD because it explicitly targets the major interpersonal role transition of becoming a parent, relationship strain with a partner, and social isolation — themes that are often central to a new mother's specific distress in a way generic supportive counselling may not fully address.
Partner and family involvement
Actively encourage the patient to discuss treatment options with her partner or closest support person, and to bring that person to an appointment where treatment is being discussed — current Canadian guidance specifically recommends this collaborative approach rather than treating perinatal mental health as the mother's issue alone.[3] Practical, concrete family involvement — a partner or family member taking overnight infant-care shifts to protect a consolidated block of the mother's sleep, sharing household and newborn-care tasks, and simply checking in regularly about mood and coping — is a genuine, evidence-consistent intervention, not just a supportive gesture. Screen partners too, where possible: paternal or partner postpartum depression is real, affects a meaningful minority of partners, and is more common when the birthing parent also has PPD — untreated partner depression undermines the whole family's ability to support the new mother and infant.
Peer support and group-based interventions
Structured postpartum support groups, delivered in person or through a facilitated peer-support program, reduce isolation and normalize the experience of struggling — many patients find genuine relief simply from learning how common these symptoms are.
Sleep, exercise, and lifestyle measures
Protecting even one consolidated 4-5 hour sleep block per 24 hours (typically by having a partner or family member cover an overnight infant-care shift) is a concrete, achievable, and genuinely evidence-relevant target, given how strongly sleep fragmentation compounds mood symptoms. Regular physical activity, once medically cleared postpartum, has a modest but real antidepressant benefit and is a reasonable adjunct at any severity. Bright light therapy is a further non-pharmacologic option with evidence specifically in mild-to-moderate perinatal depression and a favourable side-effect profile.[3]
Address breastfeeding difficulties directly, when present
Breastfeeding difficulty and pain are themselves a recognized contributor to postpartum distress. A prompt referral to a lactation consultant or breastfeeding medicine clinic to resolve a correctable latch, supply, or pain problem can meaningfully improve mood symptoms in a patient for whom breastfeeding struggles are a major, identifiable stressor — independent of, and in addition to, any pharmacologic treatment for the depression itself.
Special Populations
Breastfeeding and lactation safety — the central pharmacist counselling opportunity
The most common, and most consequential, misconception a pharmacist will encounter in PPD is the belief that no antidepressant is safe during breastfeeding. For the great majority of patients this is false, and correcting it — with real, specific data rather than a blanket reassurance — is genuinely high-value counselling. The standard way to compare antidepressant transfer into breast milk is the relative infant dose (RID): the infant's dose received via milk, expressed as a percentage of the mother's own weight-adjusted dose. An RID under 10% is the conventional threshold generally considered acceptable for continued breastfeeding.[8]
| SSRI | Relative infant dose (RID) | Infant serum detectability | Practical positioning |
|---|---|---|---|
| Sertraline | Approximately 1% (reported range roughly 0.5-1.2%) | Undetectable in essentially every infant studied — 0 of 15 in the largest recent paired maternal/infant sampling study | Preferred first choice to start fresh while breastfeeding; the most reassuring, most extensively studied data of any antidepressant. |
| Paroxetine | Approximately 2% | Undetectable in the large majority — 8 of 9 in the same study | Comparably low transfer to sertraline; reasonable to continue if already effective, though its overall tolerability/discontinuation profile makes it a less preferred agent to start fresh. |
| Escitalopram | Generally low, under the conventional 10% threshold | Generally low, though less extensively characterized than sertraline or paroxetine | A reasonable second preferred first choice; less breastfeeding-specific data available than for sertraline. |
| Citalopram | Approximately 6% (reported range roughly 0.7-9%) | Detectable in a meaningful minority — 4 of 13 in the same study; a separate pooled analysis found 1 of 5 infants exceeded 10% of the mother's own plasma level, the threshold defined as "elevated" in that analysis | Reasonable to continue if already effective; less preferred than sertraline or escitalopram to start fresh while breastfeeding. |
| Fluoxetine | The highest of the commonly used SSRIs — up to approximately 6.8% | The active metabolite norfluoxetine is detectable in the large majority of studied infants (roughly 85%), reflecting its long half-life and genuine accumulation over time | Reasonable to continue if already effective and tolerated; use particular caution starting it fresh with a newborn or preterm infant, whose drug clearance is still immature. |
| Zuranolone | Reported at under 1% in the product's own lactation data | Not well characterized beyond the RID figure itself — a newly marketed drug with limited postmarketing lactation experience | A low RID number alone is not the same thing as an established safety record. Current product labelling recommends discontinuing breastfeeding during the 14-day course unless the benefit is judged to outweigh the risk, with infant monitoring for excess sleepiness or poor feeding if breastfeeding continues. |
Practical counselling points: for the great majority of breastfeeding patients on a well-tolerated SSRI, continue it — do not switch a working agent purely because of breastfeeding without a specific, identified reason. Routine infant drug-level testing is not needed; instead, counsel the family to watch for and report excessive infant sleepiness or difficulty rousing for feeds, poor feeding or weight gain, irritability, or unusual fussiness, and to bring these up rather than assuming they must be unrelated to the medication. A patient who was taking a specific antidepressant safely through pregnancy generally has the strongest reason to simply continue it postpartum rather than switch to a "preferred lactation" agent she has never tried.
Postpartum psychosis prevention in known bipolar disorder
For a patient with known bipolar disorder, or a prior episode of postpartum psychosis, prophylactic mood-stabilizer therapy — most commonly lithium — reinitiated immediately after delivery is a recognized, evidence-supported strategy to reduce recurrence risk, and should be arranged proactively with the patient's psychiatric team well before delivery rather than reactively once symptoms appear.[3] This decision, the specific agent chosen, and its own breastfeeding-safety profile (lithium requires close infant monitoring and specialist involvement if breastfeeding continues) belongs with the patient's treating psychiatrist or a reproductive mental health specialist — a community pharmacist's role is recognizing this history and supporting, not independently directing, that plan.
Future pregnancy planning and recurrence risk
Discuss recurrence risk directly with every patient who has had PPD before conceiving again — a prior episode is the single strongest predictor of a future one. Proactive planning (arranging earlier antenatal screening, identifying a treatment plan in advance, and considering prophylactic treatment for a patient with a bipolar or postpartum psychosis history) is preferable to waiting for symptoms to reappear.
NICU admission and preterm infants
A NICU admission is itself both a risk factor for PPD and a setting that complicates lactation-safety decisions, given a preterm infant's genuinely immature drug clearance — this is a specific reason to favour sertraline or another low-transfer agent, and to involve the NICU care team directly in any medication decision affecting a breastfeeding preterm infant.
Adolescent parents
Adolescent mothers face a compounded risk profile — developmental, social, and often socioeconomic — and screening should not be skipped or softened simply because the patient is young; involve a parent, guardian, or other trusted adult support person in the treatment conversation with the patient's consent wherever appropriate.
Socioeconomic and cultural context
Screening and follow-up plans should account for real barriers to care — lack of paid leave, transportation, childcare for other children, language barriers, and cultural expectations around postpartum confinement or family involvement. A plan that assumes unlimited time, transportation, or privacy will fail for many patients; ask directly what is realistic for this specific patient rather than defaulting to a one-size-fits-all follow-up schedule.
Monitoring & Follow-Up
- Reassess symptom severity every 1-3 weeks while the patient remains symptomatic, extending to every 6-8 weeks once symptoms have fully remitted — current Canadian guidance's explicit monitoring cadence.[3]
- Repeat the EPDS (or PHQ-9) at follow-up visits to track objective change, not only subjective impression, and continue routine screening through 12 months postpartum even after an initial episode has resolved.[3]
- Reassess for emerging mania, mixed features, or psychosis at every follow-up visit, not only at the initial diagnosis — new activation, decreased need for sleep without fatigue, or any of the Red Flags & Refer features after starting an antidepressant warrants prompt reassessment rather than being assumed to be simple treatment response.
- For any patient on a psychotropic while breastfeeding, ask specifically and at every visit about infant feeding, alertness/sedation, weight gain, and irritability — this is clinical monitoring, not routine infant drug-level testing (see Special Populations).
- For a patient completing a 14-day zuranolone course, schedule a specific follow-up at or shortly after course completion to reassess symptoms and confirm the plan for what comes next — whether that is starting or continuing a standard antidepressant, or, for a genuinely full responder, close monitoring alone.
- Reassess safety (suicidality, capacity to care for the infant, support system adequacy) at every encounter for the duration of active treatment, not only when directly prompted by the patient.
Common Errors to Avoid
- Blurring postpartum psychosis with routine PPD — triaging delusions, hallucinations, marked confusion, or an ego-syntonic thought of infant harm as "severe PPD" and scheduling routine follow-up instead of same-day emergency escalation. This is the single most consequential error possible in this chapter.[4]
- Dismissing genuine, persistent symptoms past 2 weeks postpartum as "just the baby blues" and delaying a formal EPDS screen and assessment.
- Screening only once, at the traditional 6-week postpartum visit, rather than through pregnancy and up to 12 months postpartum as current guidance actually recommends.[3]
- Skipping or not specifically reviewing EPDS item 10 (self-harm ideation) simply because the total score is below the general cut-off.
- Assuming every reported thought about infant harm is the common, benign, ego-dystonic kind without directly assessing whether it is instead an ego-syntonic delusional belief.
- Advising a patient to stop breastfeeding by default when starting antidepressant therapy, rather than reviewing the specific, genuinely reassuring lactation-safety data for most agents.
- Defaulting to fluoxetine or citalopram for a fresh antidepressant trial in a breastfeeding patient without considering their comparatively higher relative infant dose and infant-serum detectability versus sertraline or paroxetine.
- Switching a patient who is already doing well on a specific antidepressant to sertraline purely because sertraline is "the lactation-preferred SSRI" — continuity of an already-effective agent is generally the better choice.
- Starting antidepressant monotherapy without first asking about a personal or family history of bipolar disorder, risking an unrecognized manic or mixed episode.
- Treating zuranolone as a substitute for ongoing antidepressant maintenance therapy rather than what it actually is — a single 14-day course requiring its own follow-up plan.
- Assuming a low relative infant dose figure alone is equivalent to an established safety record, particularly for a newly marketed, sedating agent like zuranolone.
- Failing to involve a partner or support person in the treatment conversation, or overlooking a partner's own postpartum depression.
Clinical Decision Algorithm
Drug Tables
How to use this table: doses are adult PPD-specific doses and require patient-specific confirmation. See Special Populations for the full breastfeeding relative-infant-dose comparison. Drug pricing is intentionally omitted. Select any drug name to open its complete Pharmasist Drug Database page.
SSRIs — Preferred While Breastfeeding
The reasonable first choices for a patient with no prior antidepressant trial, given the most reassuring lactation-safety data of any antidepressant class.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Sertraline | Initial: 50 mg PO once daily Usual maintenance: 50-200 mg/day | GI upset, insomnia, sexual dysfunction, sweating | Serotonergic interactions with other serotonergic drugs/MAOIs; caution with NSAIDs/anticoagulants (bleeding risk) | Relative infant dose approximately 1%; undetectable in infant serum in essentially every recent study — the preferred default first choice. |
| Escitalopram | Initial: 10 mg PO once daily Usual maintenance: 10-20 mg/day | GI upset, insomnia, sexual dysfunction | Serotonergic interactions with other serotonergic drugs/MAOIs; dose-related QTc prolongation at higher doses | A reasonable second preferred first choice; less breastfeeding-specific data than sertraline but a generally low relative infant dose. |
SSRIs — Reasonable to Continue, Less Preferred to Start Fresh While Breastfeeding
Each has a genuinely higher relative infant dose and/or more frequent infant-serum detectability than sertraline or escitalopram — reasonable to continue if a patient is already doing well on one, less preferred as a first choice while nursing.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Paroxetine | Initial: 20 mg PO once daily Usual maintenance: 20-50 mg/day | Sedation, weight gain, anticholinergic effects, pronounced discontinuation symptoms if stopped abruptly | Potent CYP2D6 inhibitor — numerous drug interactions; contraindicated with an MAOI | Relative infant dose approximately 2%, comparably low to sertraline, but short half-life and discontinuation-symptom burden make it a less preferred agent to start fresh. |
| Citalopram | Initial: 20 mg PO once daily Usual maintenance: 20-40 mg/day (do not exceed 40 mg/day) | Dose-related QTc prolongation, GI upset, sedation | Avoid combining with other QTc-prolonging drugs; contraindicated with an MAOI | Relative infant dose approximately 6%; a meaningful minority of studied infants have detectable serum levels — reasonable to continue if already effective. |
| Fluoxetine | Initial: 10-20 mg PO once daily Usual maintenance: 20-60 mg/day | GI upset, insomnia, activation/anxiety on initiation | Potent CYP2D6 inhibitor; contraindicated with an MAOI | Long half-life and the active metabolite norfluoxetine give the highest relative infant dose of the commonly used SSRIs; use particular caution starting fresh with a newborn or preterm infant. |
Other Antidepressants — Second-Line / Alternative
Reasonable options after an inadequate SSRI trial, or for a specific tolerability problem.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Venlafaxine (extended-release) | Initial: 37.5-75 mg PO once daily Usual maintenance: 75-225 mg/day | Hypertension (dose-related), GI upset, sweating, discontinuation symptoms | Serotonergic interactions with other serotonergic drugs/MAOIs | Reasonable second-line SNRI after an inadequate SSRI response; confirm current lactation data for the specific formulation. |
| Bupropion (sustained-release) | Initial: 150 mg PO once daily Usual maintenance: 150-300 mg/day | Insomnia, dry mouth, seizure risk at higher doses | Lowers seizure threshold — avoid combining with other agents that do the same | Limited but generally reassuring lactation literature; a reasonable alternative when sexual dysfunction or sedation from an SSRI/SNRI is not tolerated; avoid in a seizure disorder or eating disorder history. |
Rapid-Acting Neurosteroid Therapy
Mechanistically distinct from standard antidepressants — GABA-A receptor positive allosteric modulators, not serotonergic agents; a single time-limited course, not maintenance therapy.
| Drug | Dosage | Adverse effects | Drug interactions | Comments |
|---|---|---|---|---|
| Zuranolone | Standard dose: 50 mg PO once daily at bedtime with fat-containing food, for 14 days If not tolerated: Reduce to 40 mg once daily for the remainder of the 14-day course Renal/hepatic impairment or strong CYP3A4 inhibitor: Reduce to 30 mg once daily for the 14-day course | Somnolence, dizziness, sedation; impairment can last beyond 12 hours post-dose | Additive CNS depression with alcohol, benzodiazepines, opioids, or other sedatives; dose reduction needed with a strong CYP3A4 inhibitor | First medication authorized in Canada specifically for PPD (December 2025); a single 14-day course, not retreated or continued indefinitely; contraindicated in pregnancy — confirm effective contraception. |
Not Accessible or Not Recommended
Included so neither is mistakenly assumed to be a current, routine option. Brexanolone's own phase 3 registration trials established the neurosteroid mechanism's efficacy for PPD before zuranolone existed as an oral alternative.[12]
| Drug | Dosage | Reason avoided | Drug interactions | Comments |
|---|---|---|---|---|
| Brexanolone Avoid | N/A | Not accessible in Canada — never marketed here and withdrawn from the US market entirely as of 2025 | — | Historical interest only; zuranolone is the current oral alternative with the same mechanism of action. |
| Doxepin Avoid | N/A | Case reports of infant sedation and respiratory depression — avoid while breastfeeding | — | Additive CNS/respiratory depression with other sedating agents |
References
- Maternal depression and child development — Paediatrics & Child Health, Canadian Paediatric Society, Mental Health and Developmental Disabilities Committee. Paediatr Child Health 2004;9(8):575-583 (reaffirmed periodically)
- Guideline No. 454: Identification and Treatment of Perinatal Mood and Anxiety Disorders — Journal of Obstetrics and Gynaecology Canada, Society of Obstetricians and Gynaecologists of Canada (Shea A, Jumah NA, Forte M, Cantin C, Bayrampour H, Butler K, Francoeur D, Green C, Cook J). J Obstet Gynaecol Can 2024;46(10):102696
- Canadian Network for Mood and Anxiety Treatments (CANMAT) 2024 Clinical Practice Guideline for the Management of Perinatal Mood, Anxiety, and Related Disorders — Canadian Journal of Psychiatry (Vigod SN, Frey BN, Clark CT, et al.). Can J Psychiatry 2025;70(6):429-489
- The global prevalence of postpartum psychosis: a systematic review — BMC Psychiatry (VanderKruik R, Barreix M, Chou D, Allen T, Say L, Cohen LS; Maternal Morbidity Working Group). BMC Psychiatry 2017;17:272
- Biogen Receives Health Canada Authorization for ZURZUVAE (zuranolone), the First and Only Treatment Indicated for Adults with Postpartum Depression in Canada — Biogen Canada Inc.. News release, 9 December 2025
- Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR) — American Psychiatric Association. Washington (DC): American Psychiatric Publishing; 2022
- Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale — British Journal of Psychiatry (Cox JL, Holden JM, Sagovsky R). Br J Psychiatry 1987;150:782-786
- Pooled analysis of antidepressant levels in lactating mothers, breast milk, and nursing infants — American Journal of Psychiatry (Weissman AM, Levy BT, Hartz AJ, Bentler S, Donohue M, Ellingrod VL, Wisner KL). Am J Psychiatry 2004;161(6):1066-1078
- Sertraline, citalopram and paroxetine in lactation: passage into breastmilk and infant exposure — Frontiers in Pharmacology (Den Besten-Bertholee D, Touw DJ, Damer EA, Mian P, Ter Horst PGJ). Front Pharmacol 2024;15:1414677
- Zuranolone for the Treatment of Postpartum Depression — American Journal of Psychiatry (Deligiannidis KM, Meltzer-Brody S, Maximos B, et al.). Am J Psychiatry 2023;180(9):668-675
- Brexanolone, zuranolone and related neurosteroid GABA-A receptor positive allosteric modulators for postnatal depression — Cochrane Database of Systematic Reviews (Wilson CA, et al.). Cochrane Database Syst Rev 2025
- Brexanolone injection in post-partum depression: two multicentre, double-blind, randomised, placebo-controlled, phase 3 trials — The Lancet (Meltzer-Brody S, Colquhoun H, Riesenberg R, et al.). Lancet 2018;392(10152):1058-1070
Professional reference only. Apply clinical judgment and verify patient-specific decisions against current authoritative guidance and product information.