ARTICLE |
a Department of Pediatrics, Mt Scopus, Hadassah-Hebrew University Medical Center, Jerusalem, Israel
b BELTIS Teratology Information Center, Department of Neonatology, Rabin Medical Center, Petah Tikva, Israel; Department of
d Pediatric Cardiology
f Pediatrics, Schneider Children's Medical Center of Israel, Sackler School of Medicine, Tel Aviv University, Tel Aviv, Israel
c Department of Medicine, Pediatrics, and Health Policy, Management, and Evaluation
e MotheRisk Program, Divisions of Clinical Pharmacology and Toxicology
g Emergency Medicine, Department of Pediatrics, Hospital for Sick Children, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
h Clinical Pharmacology Unit, Division of Emergency Medicine, Children's Hospital, Boston, Boston, Massachusetts
| ABSTRACT |
|---|
|
|
|---|
METHODS. Between January 2000 and December 2005, we collected data on all of the newborns born at a single tertiary care hospital. Electrocardiograms of infants exposed to selective serotonin-reuptake inhibitor antidepressants in utero were compared with those of healthy control newborns matched on gestational age. The tracings were interpreted by a pediatric cardiologist who was unaware of the drug exposure.
RESULTS. We identified 52 newborns exposed to selective serotonin-reuptake inhibitor antidepressants in the immediate antepartum period and 52 matched control subjects. The mean QTc was significantly longer in the group of newborns exposed to antidepressants as compared with control subjects (409 ± 42 vs 392 ± 29 milliseconds). Five (10%) newborns exposed to selective serotonin-reuptake inhibitor antidepressants had a markedly prolonged QTc interval (>460 milliseconds) compared with none of the unexposed newborns. The longest QTc interval observed among exposed newborns was 543 milliseconds. All of the drug-associated repolarization abnormalities normalized in subsequent electrocardiographic tracings.
CONCLUSIONS. Antepartum use of selective serotonin-reuptake inhibitor antidepressants is associated with QTc interval prolongation in exposed neonates. Additional research using a standardized protocol is needed to determine whether exposure to selective serotonin-reuptake inhibitor antidepressants in late pregnancy is also associated with arrhythmias.
Key Words: pregnancy antidepressant adverse drug events electrocardiography long-QT syndrome
Abbreviations: SSRI—selective serotonin-reuptake inhibitor ECG—electrocardiography
Prolongation of the QT interval is associated with an increased risk of malignant arrhythmias and sudden death.1 Drug-induced QT prolongation is the most common reason for the withdrawal of medications from market or the imposition of restrictions on their use.2 Prominent recent examples of drugs removed from widespread use because of QT prolongation include terfenadine, astemizole, cisapride, and grepafloxacin.3
It has been estimated that
25% of women have symptoms of depression during pregnancy,4 most often in the second and third trimesters.5 Although some do not seek medical attention, a large proportion of depressed pregnant women receive antidepressant medication because of the well-documented negative impact of untreated depression on pregnancy outcomes.5 Since their introduction
2 decades ago, selective serotonin-reuptake inhibitor (SSRI) antidepressants have become the most widely used class of antidepressants by pregnant women. A recent report from the National Birth Defects Prevention Study6 revealed that SSRIs represent
80% of antidepressants used in pregnancy; given that treatment often continues into the second and third trimesters, every year
92000 newborns are exposed to these medications before delivery. Recent studies question the safety of these drugs during pregnancy, with the main focus being adverse neonatal outcomes.7–12
SSRI antidepressants alter cardiac repolarization by inhibiting the human ether-a-go-go potassium channel.13, 14 In adults, these drugs have been associated with prolongation of the QT interval and torsades de pointes during therapeutic use15–21 and after overdose.22, 23 However, little is known about the effects of SSRIs on cardiac repolarization in neonates. After the observation of a single case of QTc prolongation in a neonate born to a fluoxetine-treated mother,24 we explored the potential association between antenatal exposure to SSRI antidepressants and prolongation of the QTc interval in the first days of life.
| METHODS |
|---|
|
|
|---|
We identified all of the neonates with a gestational age of
35 weeks born to women receiving treatment with an SSRI antidepressant at the onset of labor and extracted their medical charts and ECG tracings. We excluded neonates born to women treated with any other chronic medication during pregnancy (whether known as QT prolonging or not), as well as those with gestational diabetes or hypothyroidism. We also excluded neonates with Apgar scores <7 (either 1 or 5 minutes), because asphyxia is a known risk factor for QT prolongation,2 and those with cardiac structural abnormalities identified by echocardiography.
The comparison group consisted of newborns born to healthy mothers who took no medications before delivery. These infants were born during the same 6-year period in our center and underwent electrocardiography because of the detection of a systolic murmur during the first postnatal examination. Infants with any abnormality on echocardiography were excluded from the comparison group. Normal infants were stratified by gestational age, and then a random match for each exposed infant's gestational age was chosen from the potential control subjects within that gestational age stratum. Exposed and unexposed neonates were matched on gestational age, because advancing fetal age is associated with changes in the duration of ECG intervals.26
For the purpose of this study, ECGs were interpreted by a single experienced pediatric cardiologist (Dr Fogelman), who was blinded to drug exposure. The pediatric cardiologist measured the RR, PR, QRS, and QT intervals and defined whether the ECG was otherwise normal. The QT intervals were corrected for rate (QTc) using Bazett's correction27, 28 and subsequently replicated using the Fridericia formula.29 The mean of 2 consecutive ECG complexes in lead II30 was calculated, and a prolonged QTc was defined as an interval >460 milliseconds.31, 32 Although there is no consensus regarding the definition of prolonged QTc in neonates, 460 milliseconds is a widely used upper limit cited by authorities in both pediatric cardiology31 and neonatology32 and is 3 SDs higher than the mean of the general healthy newborn population.28 To test the robustness of our findings, we replicated our analysis using a less stringent cutoff (440 milliseconds), corresponding with the upper 2.5% of QTc values found in healthy newborns.28 This threshold was adopted recently by the European Society of Cardiology.33 The study was approved by the hospital's institutional research ethics board.
We compared continuous variables by using Student's t test or Mann-Whitney U test, as appropriate, and proportions by using Fisher's exact test. We used the Kruskal-Wallis test to compare the QTc intervals after exposure to different SSRI agents. Pearson's correlation (r) was used to assess the relationships between the QTc interval and various clinical and laboratory variables. For all of the analyses, P values were 2-sided and used a type I error rate of .05 as the cutoff for statistical significance. All of the statistical analyses were performed by using a standard statistical package (SPSS 15.0 [SPSS Inc, Chicago, IL]).
| RESULTS |
|---|
|
|
|---|
|
Five neonates (10%) born to SSRI antidepressant-treated mothers had a pathologically prolonged QTc interval (range: 462–543 milliseconds) as compared with none in the control group (P = .057; Table 2). Three of the infants with prolonged QTc were exposed to paroxetine and the other 2 to fluoxetine. Follow-up ECG showed normalization of the QTc intervals to
390 milliseconds within 48 hours in 3 of these infants, whereas the other 2 neonates had normal QTc intervals (390 and 412 milliseconds) on follow-up testing after hospital discharge. These 2 prolonged intervals were identified only during the chart review for this study. Families were contacted immediately; the children were
1 year old and were reported to be completely healthy. Repeat ECGs were performed by their pediatricians, and tracings were blindly interpreted by the pediatric cardiologist for the study (Dr Fogelman).
|
We found no association between QTc interval and birth weight (r = –0.13, P = .2 for all cohorts; r = –0.06, P = .6 for SSRI group; r = –0.13, P = .4 for control group), which was lower in exposed newborns and serves as a surrogate marker of long-term use of antidepressants by their mothers.12 Among the antidepressant-exposed group, no infant had abnormalities in serum sodium, potassium, calcium, or magnesium, and we found no correlation between QTc interval and any of these values. Finally, we found no significant differences in gender distribution, Apgar scores, QRS duration, or PR interval duration between the 2 groups.
| DISCUSSION |
|---|
|
|
|---|
Several observations in the present study support a causal relationship between antenatal exposure to SSRI antidepressants and neonatal QTc prolongation. First, SSRI medications and their active metabolites cross the placenta to the fetus and appear in the amniotic fluid, allowing for possible absorption through fetal gastrointestinal and respiratory tracts.34–36 Second, reversible QT prolongation has been described previously in adults and children receiving SSRI antidepressants, particularly after large doses. Third, the high incidence of long QTc observed among neonates exposed to antidepressants (10%) far exceeds that anticipated by chance, because only 1% of healthy newborns have QTc intervals >460 milliseconds.28 Finally, repolarization abnormalities in 3 exposed newborns were shown to normalize shortly after delivery, coincident with the cessation of drug exposure. This reversibility supports a causal relationship between antenatal drug exposure and QTc prolongation in neonates and also suggests that congenital "long QT syndrome" (which affects
1 in 10000 individuals,37 a 1000-fold lower figure compared with our study incidence) is not a competing explanation for our findings. Additional QT-prolonging drugs, such as amiodarone and methadone, have also been reported to cause neonatal QT prolongation in the same mechanism after maternal use during pregnancy.38, 39
It should be noted that we identified no alternative explanation for the reversible QTc prolongation. The differential diagnosis for QTc prolongation in a neonate includes familial long QT syndrome, hypoxia, brain injury and hemorrhage, medications, endocrine causes, and electrolyte disturbances.32, 40 All of the infants in this study were healthy, born to healthy mothers, with no family history suggestive of long QT syndrome, no evidence of hypoxia, and with normal electrolyte levels. Neonates born to mothers with other known drug and nondrug causes of QT prolongation were a priori excluded from enrollment to further minimize any other potential confounding mechanism of repolarization abnormalities.
Although SSRI-induced QT prolongation in adults is relatively uncommon, a large proportion of neonates in our study had abnormally prolonged QTc intervals. It should be noted that this drug effect has probably occurred in utero as well and was only identified after birth. Prenatal magnetocardiograms could be used to identify such intrauterine QT changes in future studies.41 Although the mechanisms for enhanced effects of SSRIs on the newborn's heart are not yet fully elucidated, several compelling explanations exist. These involve ontogeny of the CYP2D6 (the main metabolic pathway of SSRIs) and CYP2C9 (involved in the metabolism of fluoxetine), as well as altered protein binding and CYP2D6 polymorphism.
Limited CYP2D6 protein amount and activity at 30% have been documented in the fetal liver.42 In the first month of life, CYP2D6 activity increases to reach approximately two thirds of adult values between 1 month and 5 years of age. On an individual basis, altered pharmacodynamic impact because of CYP2D6 polymorphism may also play an important role in reduced SSRI clearance.43 CYP2C9 catalytic activity expression is at 1% to 2% of mature values during the first trimester and progressively increases during pregnancy to reach
30% of mature values at birth.44 Lower CYP2D6 and CYP2C9 activities in neonates can result in higher circulating SSRI levels at this age, thereby partially explaining the increased rate of QTc prolongation observed in this study.
Another mechanism that contributes to enhanced systemic exposure in newborns is their lower serum levels of binding proteins and, as a result, higher free drug levels compared with adults.45 For example, fluoxetine is highly bound (94%–95%) to serum
-1-acid glycoprotein,46 which is present at lower levels in neonates. A third mechanism for the high rate of QTc prolongation is that the neonatal heart might be more sensitive than that of adults to drugs that interfere with the delayed rectifier current.47
| CONCLUSIONS |
|---|
|
|
|---|
| FOOTNOTES |
|---|
Address correspondence to Yaron Finkelstein, MD, Hospital for Sick Children, MotheRisk Program, Divisions of Clinical Pharmacology and Toxicology and Emergency Medicine, Department of Pediatrics, 555 University Avenue Toronto, Ontario, Canada M5N2N5. E-mail: yfinkel{at}yahoo.com
The authors have indicated they have no financial relationships relevant to this article to disclose.
| What's Known on This Subject SSRIs are the most widely used class of antidepressants by pregnant women. Several reports of SSRI-induced QT prolongation in adults exist, and we previously reported such a case of prolonged QT after maternal fluoxetine use. No studies have examined the effects of maternal SSRI use in pregnancy and neonatal QT prolongation.
|
| What This Study Adds The mean QTc was significantly longer in the group of newborns exposed to SSRI antidepressants as compared with control subjects. Ten percent of SSRI-exposed newborns had a markedly prolonged QTc interval (>460 milliseconds) compared with none of the unexposed newborns. The longest QTc interval observed was 543 milliseconds.
|
| REFERENCES |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||