ARCH GEN PSYCHIATRY/VOL 68 (NO. 11), NOV 2011

PK B17 조 201076436 박성진

Introduction The prevalence of Autism Specrum Disorders (ASDs)

has increased over recent years.◦ First epidemiologic survey (1966) : 4 ~ 5 cases /

10,000 population◦ Recent studies : 100 cases / 10000 population (1%)

There is scientific concern about environmental factors that may contribute to autism risk.

Use of antidepressant medications during pregnancy also shows a secular increase in recent decades.◦ SSRI have become the first line treatment for depression and other

psychiatric condition, including anxiety, obsessive-compulsive dis-orders

◦ SSRI can cross the placenta and are secreted in breast milk.

Introduction Systematic study of ASD and prenatal exposure to

SSRIs and other antidepressant medications is needed.◦ No prior studies have addressed this question◦ There are many reports about ASD patients have

atypical serotonin patterns in their blood speci-men Some reports indicate possible subtle effects in reaching

milestone achievement of children by fetal SSRI exposure Some experimental studies using rodent models indicate

that SSRI exposure during early brain development has consequences for some behaviors in life

Distinguishing the effects of medication exposure

from the effects of the underlying condition that led to treatment

is very important

Methods#Study design and setting a population based, Case Control Study Study sample was drawn from the medical record of

‘Childhood Autism Perinatal study’◦ A large case-control study of prenatal, perinatal,

and neonatal risk factors for ASDs set within the Kaiser Permanente Medical Care Program in North-ern California (KPNC)

◦ Represntative of the total population in the region(California)

◦ It can be derived prospectively recorded information on mothers’ use of antidepressant medications, mental health history of mothers, and demographic and medical covariates.

Methods#Case/Control selection

Case children (n=420) Children with at least 1 diagnosis of autism,

Asperger syndrome, pervasive developmen-tal disorder not otherwise specified (Jan1995 ~ Nov 2002)

Control children (n=2100) Children without an ASD diagnosis were

randomly sampled from the remaining co-hort

Ratio of 5 control children / 1 case child. Frequency matched to case children by sex,

birth year, and hospital of birth.

Methods#Parameterization of antidepressant administration of mother

Classification of antidepressnats (1) SSRIs

: citalopram hydrobromide, fluoxetine, fluvoxaminemaleate, paroxetine hydrochlo-ride, and sertraline hydrochloride

(2) dual-action antidepressants (DAA): nefazodone hydrochloride, trazodone hydrochloride, and venlafaxine hydrochloride: including serotonin-noradrenergic-reuptake inhibitors, noradrenergic and specific serotoninergic antidepressants, and noradrenaline-reuptake inhibitors

(3) tricyclic antidepressants (TCA): amitriptyline hydrochloride, desipramine hydrochloride, doxepin hydrochloride, imipramine hydrochloride, nortriptyline hydrochloride, and protriptyline hydrochloride

4 exposure time were defined (1) preconception (3months prior to the estimated LMP) (2) first trimester (first 90 days after the LMP) (3) second trimester (91-180 days after the LMP) (4) third trimester (181 days after the LMP to date of delivery)

Definition of exposure : presciption by doctor + record of dispensed prescription

Methods#Mothers’ psychiatric conditions

Analytic concern about treatment vs underly-ing condition

Categorization of mother’s psychiatric conditions Major depressive disorder Anxiety or phobia Bipolar disorder Obsessive-compulsive disorder Adjustment disorder Schizophrenic disorders

Mental health history of mother was included as a variable in

selected logistic models

Statistical Analysis#Other relevant medical and utilization history

data Data on the mothers’ demographic characteristics

Self-reported age at delivery, race/ethnicity, educational at-tainment, and parity

Data on infants’ characteristics Birth year, sex, birth weight, and gestational age

#Statistical tests The Chi square test (for categorical variables) The Fisher exact test (for proportions) The 2-sample t test (for continuous variables)

Unconditional logistic regression analysis to estimate unadjusted and adjusted relative risks of ASD

associated with antidepressant use by the mother during the year before delivery of the study child.

ResultsThe Characteristics of the final study population

The final study population in-cluded 298 children with ASD and 1507 control children. ◦ This is 73% of initial cases and 72%

of initial controls◦ Some population data was dis-

carded by statistical restriction The characteristics of the final

study population◦ Mothers

Age Race ethnicity Education

◦ Children Birth weight < 2500g Gestational Age < 37 wks

◦ Similar characteristics Proportion of male Distribution of birth facilities Parity

Results Twenty case mothers (6.7%) and 50

control mothers(3.3%) ◦ had at least 1 prescription for an an-

tidepressant in the year prior to the birth of the study child

◦ The majority of these case and control mothers were prescribed SSRIs

#20 Case mothers 13 (65%) were prescribed SSRIs only, 2 (10%) were prescribed an SSRI in combi-

nation with another antidepressant, 5 (25%) were prescribed 1 or more non-

SSRI antidepressants only.

#50 Control mothers 25 (50%) were prescribed SSRIs only 9 (18%) were prescribed an SSRI in combi-

nation with another antidepressant 16 (32%) were prescribed 1 or more non-

SSRI antidepressants only.

Results

To reduce bias, variables adjustment was done◦ for maternal age, race/ethnicity, education, and child’s birth

weight, sex, birth year, and facility of birthMothers of children subsequently diagnosed with ASD were,

◦ twice as likely to have at least 1 antidepressant prescription in the year prior to delivery of the study child

In adjusted models with women not exposed to SSRIs as the reference group (Table 4, model 1), risk of ASD was significantly associated with an SSRI pre-scription during the preconception period, during the first trimester of preg-nancy, and anytime during the year.

For the second and third trimesters, during which fewer numbers of women were prescribed an SSRI, point estimates were also elevated but did not reach statistical significance

ConclusionAlthough the number of children exposed

prenatally to selective serotonin reuptake inhibitors in this population was low, results suggest that exposure, especially during the first trimester, may modestly increase the risk of ASD.

The potential risk associated with exposure must be balanced with the risk to the mother or fetus of untreated mental health disorders.

Further studies are needed to replicate and extend these findings.