human psychopharmacology Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/hup.2451
REVIEW ARTICLE
Serotonin reuptake inhibitors and breastfeeding: a systematic review Laura Orsolini* and Cesario Bellantuono Psychiatric Unit and DEGRA Center, United Hospital of Ancona and Academic Department of Experimental and Clinical Medicine, Polytechnic University of Marche, Ancona, Italy
Objective The postnatal period represents a critical phase for mothers because of physiological hormonal changes, the increase of emotional reactions and a greater susceptibility for the onset/recrudescence of psychiatric disorders. Despite the evidence of an increasing utilization of antidepressant drugs during breastfeeding, there is still few reliable information on the neonatal safety of the selective serotonin reuptake inhibitors (SSRIs) and selective noradrenergic reuptake inhibitors (SNRIs) [serotonin reuptake inhibitors (SRIs)] in nursing mothers. The aim of this study is to provide a systematic review on the neonatal safety profile of these drugs during breastfeeding, also assessing the limits of available tools. Methods MEDLINE and PubMed databases were searched without any language restrictions by using the following set of keywords: ((SSRIs OR selective serotonin inhibitor reuptake OR SNRIs OR selective serotonin noradrenaline inhibitor reuptake) AND (breastfeeding OR lactation OR breast milk)). A separate search was also performed for each SSRIs (paroxetine, fluvoxamine, fluoxetine, sertraline, citalopram and escitalopram) and SNRIs (venlafaxine and duloxetine). Results Sertraline and paroxetine show a better neonatal safety profile during breastfeeding as compared with other SRIs. Less data are available for fluvoxamine, escitalopram and duloxetine. Few studies followed up infants breastfeed for assessing the neurodevelopmental outcomes. Conclusions Literature review clearly indicates paroxetine and sertraline as the drugs that should be preferred as first line choice in nursing women who need an antidepressant treatment. Copyright © 2014 John Wiley & Sons, Ltd. key words—selective serotonin reuptake inhibitors (SSRIs); selective noradrenergic reuptake inhibitors (SNRIs); serotonin reuptake inhibitors (SRIs); breastfeeding; lactation; review
INTRODUCTION The postnatal period represents a relatively vulnerable phase for the mothers because of the physiological hormonal changes, the increase of emotional reactions and a greater susceptibility for the onset of psychiatric disorders (Gentile et al., 2007; Patil et al., 2011), especially in a context of lack of social support, isolation or unsatisfactory couple relationship (Razurel et al., 2013; Sidebottom et al., 2014), and a previous history of anxiety/mood disorders before and/or during the pregnancy (Howell et al., 2006; Dietz et al., 2007; Balestrieri et al., 2012). In particular, the first 3 postpartum months are those at greatest risk for the onset or relapse of a psychiatric disorder (Reck et al., 2009). A recent prospective *Correspondence to: L. Orsolini, United Hospital of Ancona and Academic Department of Experimental and Clinical Medicine, Polytechnic University of Marche, Via Conca, 71, 60020 Ancona, Italy. Tel: +39 071 5963313; Fax: +39 071 5963313 E-mail: [email protected]
study that included more than 1000 unselected pregnant women has, in fact, reported that 7.3% of sample had their first episode of depression in the perinatal period (Banti et al., 2011). The prevalence rate of a major depressive episode in the puerperium ranges from 12% to 16% (Viguera et al., 2011; Elisei et al., 2013); this prevalence rises to 26% among young unmarried mothers (Logsdon et al., 2009; Venkatesh et al., 2014). Approximately, from 5% to 20% of nursing women develop postpartum depression (PPD) within the first 6 months after delivery (O’Hara and McCabe, 2013; Wynter et al., 2013). Overall, it is estimated that the prevalence rate of PPD ranges from 0% to almost 60%, depending on the clinical and cultural characteristics of the sample (Halbreich and Karkun, 2006; Kozhimannil et al., 2011). It is also documented that at least 50% of mothers with PPD are treated with antidepressant drugs (ADs) (Georgiopoulos et al., 2001); considering that 60% of mothers initiate breastfeeding, a Received 21 May 2014 Accepted 30 October 2014
Copyright © 2014 John Wiley & Sons, Ltd.
5
ssris and snris in breastfeeding
substantial number of women with PPD who require ADs may wish to continue breastfeeding despite the drug treatment (American Academy of Paediatrics, 2012; WHO, 2013a, 2013b). Recent evidence, in fact, clearly shows that the use of second generation (nontricyclic) ADs during breastfeeding is steadily growing (Sachs and Committee on Drugs, 2013). Despite the evidence of an increasing utilization of such drugs during breastfeeding, there is still few reliable information on the neonatal safety of the ADs in nursing mothers. Selective serotonin reuptake inhibitors (SSRIs) and serotonin-noradrenaline reuptake inhibitors (SNRIs), usually defined serotonin reuptake inhibitors (SRIs), are nowadays considered a first line therapy in moderate/severe postpartum depressive/anxiety disorders, even though their safety during the breastfeeding is still a matter of investigation (Gentile et al., 2007; Fortinguerra et al., 2009; Davanzo et al., 2011; Patil et al., 2011). The benefits of breastfeeding becomes even more important if we take into account the recommendations by the World Health Organization (WHO, 2003), suggesting exclusive breastfeeding for approximately the first 6 months of life and the continuation of breastfeeding to at least 1 or 2 years thereafter (American Academy of Paediatrics, 2012). The mother milk is the ideal primary source of nutrients for the newborns because it contains immunological defence elements, growth-promoting factors and nucleotides, and promotes digestion and absorption enzymes (WHO, 2013a, 2013b). Moreover, several studies demonstrate that breastfed babies report a lower risk of respiratory infections, a lower mortality (Liu et al., 2012) and a minor rate of urinary and general bacterial infections compared with the bottle-fed babies (Duijts et al., 2009). The breastfeeding also assists cognitive development in the infant (Kramer et al., 2008). In addition, breastfeeding also provides advantages for the nursing mother, reducing the risk of ovarian and breast cancer (Butt et al., 2014) and enhancing psychological bonding between the mother and infant. The aim of this paper is to provide a systematic review on the neonatal safety profile of SRIs during breastfeeding and to analyse the limits of available tools for assessing the safety of such drugs in infants breastfed. We analysed all studies reporting data on the following: (i) neonatal drug exposure during breastfeeding; (ii) selected pharmacokinetic indices such as the relative infant dose (RID), milk drug concentration (MDC) and plasma drug concentration in infants (NPDC); (iii) milk-to-plasma ratio (M/P); (iv) data on short-term neonatal outcomes (i.e. potential disorders in sleep, activity, change in bowel Copyright © 2014 John Wiley & Sons, Ltd.
movements, weight, irritability and respiratory distress) until 6 months postpartum and long-term adverse events (i.e. neurodevelopment or behavioural disorders above 6 months postpartum). METHODOLOGY The MEDLINE and PubMed databases were initially searched on September 2013 by using the following set of keywords: ((SSRIs OR selective serotonin inhibitor reuptake OR SNRIs OR selective serotonin noradrenaline inhibitor reuptake) AND (breastfeeding OR lactation OR breast milk)). A separate search was also performed to complete the safety profile of the SSRIs (paroxetine, fluvoxamine, fluoxetine, sertraline, citalopram and escitalopram) and SNRIs (venlafaxine and duloxetine). Searches were updated on February 2014. To be included in the review, studies were required the following: (i) to be human studies; (ii) to investigate a sample of nursing women and infants exposed to SSRIs or SNRIs during breastfeeding; and (iii) to investigate short-term and/or long-term neonatal outcomes after exposure to SSRIs/SNRIs during breastfeeding. All studies without information on neonatal outcomes and/or without an exposure to SSRIs/SNRIs during the lactation were excluded. No language or time restrictions were placed on the electronic search. Secondary searches were performed using the reference list of included articles and relevant systematic reviews. We also tracked citations of included articles using Google Scholar to attempt to identify unpublished data. Eligible studies were case reports, prospective or retrospective cohort studies and clinical trials reporting information on neonatal exposure to SSRIs or SNRIs during the breastfeeding. Retrieved articles were checked independently by authors, data were compared and discrepancies were settled. The set of keywords initially identified 808 studies, of which 617 were excluded because of duplication and 87 because they did not meet the inclusion criteria. Finally, a total of 104 studies were considered in the analyses. Fourteen studies were selected for citalopram (CIT) (7 case reports and 7 prospective studies) for a total of 112 neonates exposed to CIT during breastfeeding; 8 for escitalopram (ESC) (5 case reports and 3 prospective studies) for a total of 37 infants exposed; 21 studies for fluoxetine (FLX) (14 prospective studies, 1 retrospective study, 5 case reports and 1 open trial) with 280 cases; 11 for fluvoxamine (FLV) (7 case reports and 4 prospective studies) for a total of 18 infants exposed; 17 studies were selected for paroxetine (PAR) (14 prospective studies and 3 case reports) for Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
6
l. orsolini and c. bellantuono
a total of 228 cases exposed; and, finally, 22 studies for sertraline (SER) (6 case reports, 13 prospective studies, 2 pilot studies and 1 random controlled trial) with 279 infants exposed. Eleven studies were selected for SNRIs: two case reports and one open label study for duloxetine (DLX) (eight infants exposed), while for venlafaxine (VEN)/ desvenlafaxine (ODV), four case reports and four prospective cohort studies for a total of 41 infants exposed during breastfeeding.
RESULTS Selective serotonin reuptake inhibitor Citalopram. Table 1 summarizes data on the RID (expressed as the rate of the weight-adjusted maternal dose), MDC, NPDC and M/P ratio for CIT and its metabolite desmethyl-citalopram (DCIT). In 6 out of 112 cases, the literature reports neonatal adverse events. Schmidt et al. (2000) reported a case of a nursing mother who started therapy with CIT (20–40 mg daily) 4 weeks after the birth. It was observed that an uneasy sleep in her infant starting 2 to 3 nights after CIT treatment was resolved after dose reduction (20 mg daily). In the prospective cohort study by Lee et al. (2004) was reported two cases of colic and decreased feeding and one case of irritability/restlessness. In the cases of colic, authors concluded that nonspecific and insignificant symptoms spontaneously resolved, while the case of irritability/restlessness, which was considered as a potential side effect of CIT, spontaneously subsided after breastfeeding was discontinued. A case report reported SSRI-associated symptoms (sleep disorders, hypotonia, hypertonia and irregular breathing) in an infant whose mother took CIT at a dose of 40 mg daily throughout pregnancy and breastfeeding. These symptoms spontaneously resolved after 3 weeks without any further problems (Franssen et al., 2006). Finally, there was a case of transient neurodevelopmental delay in a newborn breastfed and with a foetal exposure to CIT; after a follow-up of 1 year, the neurodevelopment was regular (Heikkinen et al., 2002). In 7 out of 20 infants, followed more than 6 months from delivery, long-term neonatal outcomes were assessed with the Denver Developmental Scale (Rampono et al., 2000). The RID for CIT ranged from 0.2% to 5.9%. Detectable NPDC was found in 15 infants (1.4–98.9 ng/mL); NPDC of its metabolite was detected in only two neonates (Weissman et al., 2004). Detected MDC of CIT (41–1190 ng/mL) was reported in only 27 infants. The M/P ratio ranged from 0.93 to 4.3 for CIT and from 0.9 to 6.3 for DCIT. Copyright © 2014 John Wiley & Sons, Ltd.
Escitalopram. Table 2 summarizes the results coming from eight studies including 37 neonates breastfed and exposed to ESC and its metabolite desmethylescitalopram (DESC). Only 1 out of 27 infants exposed to ESC during breastfeeding reported a necrotizing enterocolitis 5 days after delivery; the authors suggested this adverse event as related to the exposure to ESC throughout the pregnancy (Potts et al., 2007). Only three studies followed 11 infants for a long period showing no detrimental events (Castberg and Spigset, 2006; Rampono et al., 2006; Bellantuono et al., 2013), but only in Rampono et al. (2006), the infants were assessed with the Denver Developmental Scale (Rampono et al., 2006). Relative infant dose was measured in only 15 cases, ranging from 4.5% to 6.4%. The MDC of ESC was detected in nine infants (25–99 ng/mL). No detectable NPDC was identified for ESC and DESC. The M/P ratio ranged from 1.7 to 2.7. Fluoxetine. Table 3 summarizes studies concerning FLX neonatal exposure during lactation. Data reported short-term neonatal adverse events in 11 out of 280 cases: two cases of colic; one case of possible seizure; one case of irritability and restlessness (subsided after breastfeeding stopped); one case of somnolence, lethargy, fever and unresponsiveness; and one case of watery stool, uncontrollable crying, vomiting and poor sleeping. Kristensen et al. (1999) reported two cases of colic not likely FLX related. The infant who developed a possible seizure was also exposed to carbamazepine and buspirone during pregnancy and lactation period (Brent and Wisner, 1998). Hale et al. (2001) reported a case exposed to FLX in utero during the first 11 days of breastfeeding; the infant exhibited symptoms of extreme somnolence, and decreased feeding, hypotonia, moaning, grunting and fever. The authors concluded that prolonged exposure to FLX, for its long half-life, would produce higher plasma concentrations in the breastfed infant, and so, it was responsible of these effects. Chambers et al. (1999) reported decreased postnatal growth curves in 5 out of 26 neonates breastfed and also exposed in utero to FLX. Finally, Lester et al. (1993) reported a case of watery stool, incontrollable crying, vomiting and decreased sleeping in a neonate breastfed after 10 weeks and without a prenatal history of exposition to FLX, which resolved after discontinuation of breast milk. The average RID is slightly more elevated than the other SSRIs but always under the recommended safety limit of 10%, ranging from 0.54% to 6.81%. In the studies that evaluated M/P ratio, this value ranged Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 60
Case report (1)
Ziprasidone (1)
None
None
None
Alprazolam (1)
N.A.
N.A.
5.4
N.A.
0.7–5.9
5.9
4.8
N.A.
N.A.
N.A.
5.2
0.2–0.3
3.2–5.1
3.2–3.3
RID (%)
N.A.
273–381
205
N.A.
88–361 (N.Q.)
N.A.
104–432 (65)
N.A.
1.4–2.7
98.9
N.A.
N.A.
N.A.
7 (N.Q.)
N.A.
1.8–2.1 (1.4)
41–68 (25–26) N.A.
N.A.
1.9
N.D. (N.D.)
N.A.
N.D. (N.D.)
NPDC (ng/mL) CIT (DCIT)
N.A.
389
N.A.
340–1190
470 (150)
MDC (ng/mL) CIT (DCIT)
N.A.
2–2.8
2.07
N.A.
1.16–1.88
1.81–2.24 (2.05–2.91)
3.0 (3.0)
N.A.
0.93–1.79 (1.36–1.56)
N.A.
1.1–4.3 (0.9–6.3)
1.2–3.3 (1.3–4.1)
1.2–1.7
1.8 (1.8)
M/P ratio CIT (DCIT)
N.A.
0–8
4–6
0–8
0–16
N.A.
0–8
2–33
12–26
0–8
3–42
0–8
N.A.
0–14
Lactation exposure (wks)
Irregular breathing, sleep disorders and hypo/ hypertonia spontaneously resolved Not reported in short term (MO, PO)
Not reported in short/ long term (MO, PO, OA) Not reported in short term (MO, PO) 1 case of transient neurodevelopmental delay spontaneously resolved Not reported in short term (MO, PO) 2 cases of colic and decreased feeding 1 case of irritability/ restlessness Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short/ long term (MO, PO) Not reported in short term (MO, PO) Uneasy sleep
Neonatal adverse events
CIT, citalopram; DCIT, desmethyl-citalopram; wks, weeks; N.A., not available; N.D., not detected; N.Q., not quantified; MO, maternal observation; PO, paediatrician observation; OA, objective assessment with several scales; RID, relative infant dose; MDC, milk drug concentration; NPDC, plasma drug concentration; M/P, milk-to-plasma ratio.
Werremeyer (2009)
40
20–80
Case report (7)
Case report (1)
20–40
Case report (2)
20–40
20–60
Case report (2)
Case report (1)
None
20 None
None
None
10–60
20
Weissman et al. (2004) Gorman et al. (2012) Jensen et al. (1997) Öhman et al. (1997) Spigset et al. (1997) Lepola et al. (2000) Schmidt et al. (2000) Franssen et al. (2006)
None
None
20–40
20–50
None
20–40
None
Prospective cohort study (10) Prospective cohort study (31)
Berle et al. (2004) Lee et al. (2004)
None
36
Other drugs (n. of infants exposed)
20
Prospective cohort study (7) Prospective cohort study (3) Prospective cohort study (11)
Rampono et al. (2000) Nordeng et al. (2001) Heikkinen et al. (2002)
Maternal daily dose (mg/day)
Prospective cohort study (2) Prospective cohort study (33) Case report (1)
Type of study (sample size)
Citalopram during the breastfeeding
Authors
Table 1.
ssris and snris in breastfeeding 7
Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 20
Case report (1) 20 20 10
Case report (1)
Case report (1) Case report (2)
None None
Reboxetine
None
Valproic acid
None
10–15 5–10
None
10–20
None
Other drugs
N.A. N.A.
4.6
N.A.
6.4
N.A.
5.3
4.5
RID (%)
N.A. N.A.
N.A.
N.A.
25–76
N.A.
27–99 (10–49)
N.A.
MDC (ng/mL) ESC (DESC)
N.A. N.A.
N.A.
N.A.
N.A.
N.A.
N.D. (N.D.)
N.D. (N.D.)
NPDC (ng/mL) ESC (DESC)
N.A. N.A.
N.A.
N.A.
2.03–2.49
1.7–2.7 (1.8–3.1) N.A.
N.A.
M/P ratio ESC (DESC)
5 days 4–20
0–6
0–7 (in monotherapy) 2–12
2–33
3–32
4–8
Lactation exposure (wks)
Not reported in short term (MO, PO) Not reported in short/long term (MO, PO, OA) Not reported in short term (MO, PO) Not reported in short/long term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Necrotizing enterocolitis Not reported in short/long term (MO, PO)
Neonatal adverse events
ESC, citalopram; DESC, desmethyl-escitalopram; wks, weeks; N.A., not available; N.D., not detected; N.Q., not quantified; MO, maternal observation; PO, paediatrician observation; OA, objective assessment with several scales; RID, relative infant dose; MDC, milk drug concentration; NPDC, plasma drug concentration; M/P, milk-to-plasma ratio.
Hackett et al. (2006) Potts et al. (2007) Bellantuono et al. (2013)
Rampono et al. (2006) Gorman et al. (2012) Castberg and Spigset (2006) Gentile (2006a)
10–20
Maternal daily dose (mg/day)
Prospective cohort study (5) Prospective cohort study (8) Prospective cohort study (18) Case report (1)
Type of study (sample size)
Escitalopram during the breastfeeding
Ilett et al. (2005)
Authors
Table 2.
8 l. orsolini and c. bellantuono
Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 20 20
Brent and Wisner (1998) Case report (1)
Hale et al. (2001)
Terbutamine
N.A.
3–242 (18–314)
6–235 (2–222)
121 (124)
N.A.
N.A.
39–177 (N.A.)
N.A.
N.A.
3.4
3.3
N.A.
0.54
N.A.
N.A.
3.3
N.A.
114 (124)
N.A.
69 (90)
67 (52)
28.8 (41.6)
N.A.
77.2 (81.3)
64 (104)
578
68.55
2.4–3.8 12–80 (8–74)
N.A.
Carbamazepine N.A. Buspirone
None
None
None
None
None
None
None
None
None
None
N.A.
N.A.
FLX (NFLX)
NPDC (ng/mL)
0.68 (0.56)
N.A.
N.A.
0.52–1.45 (0.08–1.1)
N.A.
N.A.
N.A.
FLX (NFLX)
M/P ratio
N.A. N.A.
REVIEW ARTICLE
Serotonin reuptake inhibitors and breastfeeding: a systematic review Laura Orsolini* and Cesario Bellantuono Psychiatric Unit and DEGRA Center, United Hospital of Ancona and Academic Department of Experimental and Clinical Medicine, Polytechnic University of Marche, Ancona, Italy
Objective The postnatal period represents a critical phase for mothers because of physiological hormonal changes, the increase of emotional reactions and a greater susceptibility for the onset/recrudescence of psychiatric disorders. Despite the evidence of an increasing utilization of antidepressant drugs during breastfeeding, there is still few reliable information on the neonatal safety of the selective serotonin reuptake inhibitors (SSRIs) and selective noradrenergic reuptake inhibitors (SNRIs) [serotonin reuptake inhibitors (SRIs)] in nursing mothers. The aim of this study is to provide a systematic review on the neonatal safety profile of these drugs during breastfeeding, also assessing the limits of available tools. Methods MEDLINE and PubMed databases were searched without any language restrictions by using the following set of keywords: ((SSRIs OR selective serotonin inhibitor reuptake OR SNRIs OR selective serotonin noradrenaline inhibitor reuptake) AND (breastfeeding OR lactation OR breast milk)). A separate search was also performed for each SSRIs (paroxetine, fluvoxamine, fluoxetine, sertraline, citalopram and escitalopram) and SNRIs (venlafaxine and duloxetine). Results Sertraline and paroxetine show a better neonatal safety profile during breastfeeding as compared with other SRIs. Less data are available for fluvoxamine, escitalopram and duloxetine. Few studies followed up infants breastfeed for assessing the neurodevelopmental outcomes. Conclusions Literature review clearly indicates paroxetine and sertraline as the drugs that should be preferred as first line choice in nursing women who need an antidepressant treatment. Copyright © 2014 John Wiley & Sons, Ltd. key words—selective serotonin reuptake inhibitors (SSRIs); selective noradrenergic reuptake inhibitors (SNRIs); serotonin reuptake inhibitors (SRIs); breastfeeding; lactation; review
INTRODUCTION The postnatal period represents a relatively vulnerable phase for the mothers because of the physiological hormonal changes, the increase of emotional reactions and a greater susceptibility for the onset of psychiatric disorders (Gentile et al., 2007; Patil et al., 2011), especially in a context of lack of social support, isolation or unsatisfactory couple relationship (Razurel et al., 2013; Sidebottom et al., 2014), and a previous history of anxiety/mood disorders before and/or during the pregnancy (Howell et al., 2006; Dietz et al., 2007; Balestrieri et al., 2012). In particular, the first 3 postpartum months are those at greatest risk for the onset or relapse of a psychiatric disorder (Reck et al., 2009). A recent prospective *Correspondence to: L. Orsolini, United Hospital of Ancona and Academic Department of Experimental and Clinical Medicine, Polytechnic University of Marche, Via Conca, 71, 60020 Ancona, Italy. Tel: +39 071 5963313; Fax: +39 071 5963313 E-mail: [email protected]
study that included more than 1000 unselected pregnant women has, in fact, reported that 7.3% of sample had their first episode of depression in the perinatal period (Banti et al., 2011). The prevalence rate of a major depressive episode in the puerperium ranges from 12% to 16% (Viguera et al., 2011; Elisei et al., 2013); this prevalence rises to 26% among young unmarried mothers (Logsdon et al., 2009; Venkatesh et al., 2014). Approximately, from 5% to 20% of nursing women develop postpartum depression (PPD) within the first 6 months after delivery (O’Hara and McCabe, 2013; Wynter et al., 2013). Overall, it is estimated that the prevalence rate of PPD ranges from 0% to almost 60%, depending on the clinical and cultural characteristics of the sample (Halbreich and Karkun, 2006; Kozhimannil et al., 2011). It is also documented that at least 50% of mothers with PPD are treated with antidepressant drugs (ADs) (Georgiopoulos et al., 2001); considering that 60% of mothers initiate breastfeeding, a Received 21 May 2014 Accepted 30 October 2014
Copyright © 2014 John Wiley & Sons, Ltd.
5
ssris and snris in breastfeeding
substantial number of women with PPD who require ADs may wish to continue breastfeeding despite the drug treatment (American Academy of Paediatrics, 2012; WHO, 2013a, 2013b). Recent evidence, in fact, clearly shows that the use of second generation (nontricyclic) ADs during breastfeeding is steadily growing (Sachs and Committee on Drugs, 2013). Despite the evidence of an increasing utilization of such drugs during breastfeeding, there is still few reliable information on the neonatal safety of the ADs in nursing mothers. Selective serotonin reuptake inhibitors (SSRIs) and serotonin-noradrenaline reuptake inhibitors (SNRIs), usually defined serotonin reuptake inhibitors (SRIs), are nowadays considered a first line therapy in moderate/severe postpartum depressive/anxiety disorders, even though their safety during the breastfeeding is still a matter of investigation (Gentile et al., 2007; Fortinguerra et al., 2009; Davanzo et al., 2011; Patil et al., 2011). The benefits of breastfeeding becomes even more important if we take into account the recommendations by the World Health Organization (WHO, 2003), suggesting exclusive breastfeeding for approximately the first 6 months of life and the continuation of breastfeeding to at least 1 or 2 years thereafter (American Academy of Paediatrics, 2012). The mother milk is the ideal primary source of nutrients for the newborns because it contains immunological defence elements, growth-promoting factors and nucleotides, and promotes digestion and absorption enzymes (WHO, 2013a, 2013b). Moreover, several studies demonstrate that breastfed babies report a lower risk of respiratory infections, a lower mortality (Liu et al., 2012) and a minor rate of urinary and general bacterial infections compared with the bottle-fed babies (Duijts et al., 2009). The breastfeeding also assists cognitive development in the infant (Kramer et al., 2008). In addition, breastfeeding also provides advantages for the nursing mother, reducing the risk of ovarian and breast cancer (Butt et al., 2014) and enhancing psychological bonding between the mother and infant. The aim of this paper is to provide a systematic review on the neonatal safety profile of SRIs during breastfeeding and to analyse the limits of available tools for assessing the safety of such drugs in infants breastfed. We analysed all studies reporting data on the following: (i) neonatal drug exposure during breastfeeding; (ii) selected pharmacokinetic indices such as the relative infant dose (RID), milk drug concentration (MDC) and plasma drug concentration in infants (NPDC); (iii) milk-to-plasma ratio (M/P); (iv) data on short-term neonatal outcomes (i.e. potential disorders in sleep, activity, change in bowel Copyright © 2014 John Wiley & Sons, Ltd.
movements, weight, irritability and respiratory distress) until 6 months postpartum and long-term adverse events (i.e. neurodevelopment or behavioural disorders above 6 months postpartum). METHODOLOGY The MEDLINE and PubMed databases were initially searched on September 2013 by using the following set of keywords: ((SSRIs OR selective serotonin inhibitor reuptake OR SNRIs OR selective serotonin noradrenaline inhibitor reuptake) AND (breastfeeding OR lactation OR breast milk)). A separate search was also performed to complete the safety profile of the SSRIs (paroxetine, fluvoxamine, fluoxetine, sertraline, citalopram and escitalopram) and SNRIs (venlafaxine and duloxetine). Searches were updated on February 2014. To be included in the review, studies were required the following: (i) to be human studies; (ii) to investigate a sample of nursing women and infants exposed to SSRIs or SNRIs during breastfeeding; and (iii) to investigate short-term and/or long-term neonatal outcomes after exposure to SSRIs/SNRIs during breastfeeding. All studies without information on neonatal outcomes and/or without an exposure to SSRIs/SNRIs during the lactation were excluded. No language or time restrictions were placed on the electronic search. Secondary searches were performed using the reference list of included articles and relevant systematic reviews. We also tracked citations of included articles using Google Scholar to attempt to identify unpublished data. Eligible studies were case reports, prospective or retrospective cohort studies and clinical trials reporting information on neonatal exposure to SSRIs or SNRIs during the breastfeeding. Retrieved articles were checked independently by authors, data were compared and discrepancies were settled. The set of keywords initially identified 808 studies, of which 617 were excluded because of duplication and 87 because they did not meet the inclusion criteria. Finally, a total of 104 studies were considered in the analyses. Fourteen studies were selected for citalopram (CIT) (7 case reports and 7 prospective studies) for a total of 112 neonates exposed to CIT during breastfeeding; 8 for escitalopram (ESC) (5 case reports and 3 prospective studies) for a total of 37 infants exposed; 21 studies for fluoxetine (FLX) (14 prospective studies, 1 retrospective study, 5 case reports and 1 open trial) with 280 cases; 11 for fluvoxamine (FLV) (7 case reports and 4 prospective studies) for a total of 18 infants exposed; 17 studies were selected for paroxetine (PAR) (14 prospective studies and 3 case reports) for Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
6
l. orsolini and c. bellantuono
a total of 228 cases exposed; and, finally, 22 studies for sertraline (SER) (6 case reports, 13 prospective studies, 2 pilot studies and 1 random controlled trial) with 279 infants exposed. Eleven studies were selected for SNRIs: two case reports and one open label study for duloxetine (DLX) (eight infants exposed), while for venlafaxine (VEN)/ desvenlafaxine (ODV), four case reports and four prospective cohort studies for a total of 41 infants exposed during breastfeeding.
RESULTS Selective serotonin reuptake inhibitor Citalopram. Table 1 summarizes data on the RID (expressed as the rate of the weight-adjusted maternal dose), MDC, NPDC and M/P ratio for CIT and its metabolite desmethyl-citalopram (DCIT). In 6 out of 112 cases, the literature reports neonatal adverse events. Schmidt et al. (2000) reported a case of a nursing mother who started therapy with CIT (20–40 mg daily) 4 weeks after the birth. It was observed that an uneasy sleep in her infant starting 2 to 3 nights after CIT treatment was resolved after dose reduction (20 mg daily). In the prospective cohort study by Lee et al. (2004) was reported two cases of colic and decreased feeding and one case of irritability/restlessness. In the cases of colic, authors concluded that nonspecific and insignificant symptoms spontaneously resolved, while the case of irritability/restlessness, which was considered as a potential side effect of CIT, spontaneously subsided after breastfeeding was discontinued. A case report reported SSRI-associated symptoms (sleep disorders, hypotonia, hypertonia and irregular breathing) in an infant whose mother took CIT at a dose of 40 mg daily throughout pregnancy and breastfeeding. These symptoms spontaneously resolved after 3 weeks without any further problems (Franssen et al., 2006). Finally, there was a case of transient neurodevelopmental delay in a newborn breastfed and with a foetal exposure to CIT; after a follow-up of 1 year, the neurodevelopment was regular (Heikkinen et al., 2002). In 7 out of 20 infants, followed more than 6 months from delivery, long-term neonatal outcomes were assessed with the Denver Developmental Scale (Rampono et al., 2000). The RID for CIT ranged from 0.2% to 5.9%. Detectable NPDC was found in 15 infants (1.4–98.9 ng/mL); NPDC of its metabolite was detected in only two neonates (Weissman et al., 2004). Detected MDC of CIT (41–1190 ng/mL) was reported in only 27 infants. The M/P ratio ranged from 0.93 to 4.3 for CIT and from 0.9 to 6.3 for DCIT. Copyright © 2014 John Wiley & Sons, Ltd.
Escitalopram. Table 2 summarizes the results coming from eight studies including 37 neonates breastfed and exposed to ESC and its metabolite desmethylescitalopram (DESC). Only 1 out of 27 infants exposed to ESC during breastfeeding reported a necrotizing enterocolitis 5 days after delivery; the authors suggested this adverse event as related to the exposure to ESC throughout the pregnancy (Potts et al., 2007). Only three studies followed 11 infants for a long period showing no detrimental events (Castberg and Spigset, 2006; Rampono et al., 2006; Bellantuono et al., 2013), but only in Rampono et al. (2006), the infants were assessed with the Denver Developmental Scale (Rampono et al., 2006). Relative infant dose was measured in only 15 cases, ranging from 4.5% to 6.4%. The MDC of ESC was detected in nine infants (25–99 ng/mL). No detectable NPDC was identified for ESC and DESC. The M/P ratio ranged from 1.7 to 2.7. Fluoxetine. Table 3 summarizes studies concerning FLX neonatal exposure during lactation. Data reported short-term neonatal adverse events in 11 out of 280 cases: two cases of colic; one case of possible seizure; one case of irritability and restlessness (subsided after breastfeeding stopped); one case of somnolence, lethargy, fever and unresponsiveness; and one case of watery stool, uncontrollable crying, vomiting and poor sleeping. Kristensen et al. (1999) reported two cases of colic not likely FLX related. The infant who developed a possible seizure was also exposed to carbamazepine and buspirone during pregnancy and lactation period (Brent and Wisner, 1998). Hale et al. (2001) reported a case exposed to FLX in utero during the first 11 days of breastfeeding; the infant exhibited symptoms of extreme somnolence, and decreased feeding, hypotonia, moaning, grunting and fever. The authors concluded that prolonged exposure to FLX, for its long half-life, would produce higher plasma concentrations in the breastfed infant, and so, it was responsible of these effects. Chambers et al. (1999) reported decreased postnatal growth curves in 5 out of 26 neonates breastfed and also exposed in utero to FLX. Finally, Lester et al. (1993) reported a case of watery stool, incontrollable crying, vomiting and decreased sleeping in a neonate breastfed after 10 weeks and without a prenatal history of exposition to FLX, which resolved after discontinuation of breast milk. The average RID is slightly more elevated than the other SSRIs but always under the recommended safety limit of 10%, ranging from 0.54% to 6.81%. In the studies that evaluated M/P ratio, this value ranged Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 60
Case report (1)
Ziprasidone (1)
None
None
None
Alprazolam (1)
N.A.
N.A.
5.4
N.A.
0.7–5.9
5.9
4.8
N.A.
N.A.
N.A.
5.2
0.2–0.3
3.2–5.1
3.2–3.3
RID (%)
N.A.
273–381
205
N.A.
88–361 (N.Q.)
N.A.
104–432 (65)
N.A.
1.4–2.7
98.9
N.A.
N.A.
N.A.
7 (N.Q.)
N.A.
1.8–2.1 (1.4)
41–68 (25–26) N.A.
N.A.
1.9
N.D. (N.D.)
N.A.
N.D. (N.D.)
NPDC (ng/mL) CIT (DCIT)
N.A.
389
N.A.
340–1190
470 (150)
MDC (ng/mL) CIT (DCIT)
N.A.
2–2.8
2.07
N.A.
1.16–1.88
1.81–2.24 (2.05–2.91)
3.0 (3.0)
N.A.
0.93–1.79 (1.36–1.56)
N.A.
1.1–4.3 (0.9–6.3)
1.2–3.3 (1.3–4.1)
1.2–1.7
1.8 (1.8)
M/P ratio CIT (DCIT)
N.A.
0–8
4–6
0–8
0–16
N.A.
0–8
2–33
12–26
0–8
3–42
0–8
N.A.
0–14
Lactation exposure (wks)
Irregular breathing, sleep disorders and hypo/ hypertonia spontaneously resolved Not reported in short term (MO, PO)
Not reported in short/ long term (MO, PO, OA) Not reported in short term (MO, PO) 1 case of transient neurodevelopmental delay spontaneously resolved Not reported in short term (MO, PO) 2 cases of colic and decreased feeding 1 case of irritability/ restlessness Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Not reported in short/ long term (MO, PO) Not reported in short term (MO, PO) Uneasy sleep
Neonatal adverse events
CIT, citalopram; DCIT, desmethyl-citalopram; wks, weeks; N.A., not available; N.D., not detected; N.Q., not quantified; MO, maternal observation; PO, paediatrician observation; OA, objective assessment with several scales; RID, relative infant dose; MDC, milk drug concentration; NPDC, plasma drug concentration; M/P, milk-to-plasma ratio.
Werremeyer (2009)
40
20–80
Case report (7)
Case report (1)
20–40
Case report (2)
20–40
20–60
Case report (2)
Case report (1)
None
20 None
None
None
10–60
20
Weissman et al. (2004) Gorman et al. (2012) Jensen et al. (1997) Öhman et al. (1997) Spigset et al. (1997) Lepola et al. (2000) Schmidt et al. (2000) Franssen et al. (2006)
None
None
20–40
20–50
None
20–40
None
Prospective cohort study (10) Prospective cohort study (31)
Berle et al. (2004) Lee et al. (2004)
None
36
Other drugs (n. of infants exposed)
20
Prospective cohort study (7) Prospective cohort study (3) Prospective cohort study (11)
Rampono et al. (2000) Nordeng et al. (2001) Heikkinen et al. (2002)
Maternal daily dose (mg/day)
Prospective cohort study (2) Prospective cohort study (33) Case report (1)
Type of study (sample size)
Citalopram during the breastfeeding
Authors
Table 1.
ssris and snris in breastfeeding 7
Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 20
Case report (1) 20 20 10
Case report (1)
Case report (1) Case report (2)
None None
Reboxetine
None
Valproic acid
None
10–15 5–10
None
10–20
None
Other drugs
N.A. N.A.
4.6
N.A.
6.4
N.A.
5.3
4.5
RID (%)
N.A. N.A.
N.A.
N.A.
25–76
N.A.
27–99 (10–49)
N.A.
MDC (ng/mL) ESC (DESC)
N.A. N.A.
N.A.
N.A.
N.A.
N.A.
N.D. (N.D.)
N.D. (N.D.)
NPDC (ng/mL) ESC (DESC)
N.A. N.A.
N.A.
N.A.
2.03–2.49
1.7–2.7 (1.8–3.1) N.A.
N.A.
M/P ratio ESC (DESC)
5 days 4–20
0–6
0–7 (in monotherapy) 2–12
2–33
3–32
4–8
Lactation exposure (wks)
Not reported in short term (MO, PO) Not reported in short/long term (MO, PO, OA) Not reported in short term (MO, PO) Not reported in short/long term (MO, PO) Not reported in short term (MO, PO) Not reported in short term (MO, PO) Necrotizing enterocolitis Not reported in short/long term (MO, PO)
Neonatal adverse events
ESC, citalopram; DESC, desmethyl-escitalopram; wks, weeks; N.A., not available; N.D., not detected; N.Q., not quantified; MO, maternal observation; PO, paediatrician observation; OA, objective assessment with several scales; RID, relative infant dose; MDC, milk drug concentration; NPDC, plasma drug concentration; M/P, milk-to-plasma ratio.
Hackett et al. (2006) Potts et al. (2007) Bellantuono et al. (2013)
Rampono et al. (2006) Gorman et al. (2012) Castberg and Spigset (2006) Gentile (2006a)
10–20
Maternal daily dose (mg/day)
Prospective cohort study (5) Prospective cohort study (8) Prospective cohort study (18) Case report (1)
Type of study (sample size)
Escitalopram during the breastfeeding
Ilett et al. (2005)
Authors
Table 2.
8 l. orsolini and c. bellantuono
Hum. Psychopharmacol Clin Exp 2015; 30: 4–20 DOI: 10.1002/hup
Copyright © 2014 John Wiley & Sons, Ltd. 20 20
Brent and Wisner (1998) Case report (1)
Hale et al. (2001)
Terbutamine
N.A.
3–242 (18–314)
6–235 (2–222)
121 (124)
N.A.
N.A.
39–177 (N.A.)
N.A.
N.A.
3.4
3.3
N.A.
0.54
N.A.
N.A.
3.3
N.A.
114 (124)
N.A.
69 (90)
67 (52)
28.8 (41.6)
N.A.
77.2 (81.3)
64 (104)
578
68.55
2.4–3.8 12–80 (8–74)
N.A.
Carbamazepine N.A. Buspirone
None
None
None
None
None
None
None
None
None
None
N.A.
N.A.
FLX (NFLX)
NPDC (ng/mL)
0.68 (0.56)
N.A.
N.A.
0.52–1.45 (0.08–1.1)
N.A.
N.A.
N.A.
FLX (NFLX)
M/P ratio
N.A. N.A.
