Recurrent Third-Trimester Fetal Loss and Maternal Mosaicism for Long-QT Syndrome

1. Recurrent Third-Trimester Fetal Loss and Maternal
Mosaicism for Long-QT Syndrome
Todd E. Miller, PhD; Elicia Estrella, MS; Robert J. Myerburg, MD; Jocelyn Garcia de Viera, MD;
Niberto Moreno, BS; Paolo Rusconi, MD; Mary Ellen Ahearn, MS; Lisa Baumbach, PhD;
Paul Kurlansky, MD; Grace Wolff, MD; Nanette H. Bishopric, MD
Background—The importance of germ-line mosaicism in genetic disease is probably underestimated, even though recent
studies indicate that it may be involved in 10% to 20% of apparently de novo cases of several dominantly inherited
genetic diseases.
Methods and Results—We describe here a case of repeated germ-line transmission of a severe form of long-QT syndrome
(LQTS) from an asymptomatic mother with mosaicism for a mutation in the cardiac sodium channel, SCN5A. A male
infant was diagnosed with ventricular arrhythmias and cardiac decompensation in utero at 28 weeks and with LQTS after
birth, ultimately requiring cardiac transplantation for control of ventricular tachycardia. The mother had no ECG
abnormalities, but her only previous pregnancy had ended in stillbirth with evidence of cardiac decompensation at 7
months’ gestation. A third pregnancy also ended in stillbirth at 7 months, again with nonimmune fetal hydrops. The
surviving infant was found to have a heterozygous mutation in SCN5A (R1623Q), previously reported as a de novo
mutation causing neonatal ventricular arrhythmia and LQTS. Initial studies of the mother detected no genetic
abnormality, but a sensitive restriction enzyme–based assay identified a small (8% to 10%) percentage of cells harboring
the mutation in her blood, skin, and buccal mucosa. Cord blood from the third fetus also harbored the mutant allele,
suggesting that all 3 cases of late-term fetal distress resulted from germ-line transfer of the LQTS-associated mutation.
Conclusions—Recurrent late-term fetal loss or sudden infant death can result from unsuspected parental mosaicism for
LQTS-associated mutations, with important implications for genetic counseling. (Circulation. 2004;109:3029-3034.)
Key Words: long-QT syndrome Ⅲ genetics Ⅲ genes
Inherited forms of the long-QT syndrome (LQTS) have
been associated with Ͼ200 different mutations in 7 genes
encoding cardiac ion channels, their regulatory subunits, and
a membrane anchoring protein.1,2 The vast majority of these
mutations are dominant point mutations (Romano-Ward syn-
drome) that result in amino acid substitutions or non-sense
codons. Current molecular genetic screening protocols can
identify potentially causative mutations in approximately two
thirds to three fourths of familial cases.3,4
See p 2930
Recently, case reports and an epidemiological study have
implicated LQTS in sudden unexplained infant death (sudden
infant death syndrome [SIDS]),5–7 but genetic testing for
LQTS is not widely available and is infrequently performed
in cases of perinatal death. Hence, the extent of the contri-
bution of LQTS to sudden infant death remains unknown.
When parents are clinically normal, a genetic basis for
sudden death may not be suspected but could arise either
from de novo mutation in the fetus or through parental
germ-line transmission of a mutation. This latter phenomenon
is surprisingly common and may underlie up to 30% of what
otherwise appear to be de novo cases of dominantly inherited
genetic disorders.8–11 Here, we describe the case of a clini-
cally unaffected mother with somatic mosaicism for a known
pathogenic mutation in the cardiac sodium channel, SCN5A,
who transmitted a severe form of LQTS to multiple succes-
sive pregnancies. These findings have significant implica-
tions for the evaluation of causes of late-term fetal distress
and demise. and for genetic counseling of apparently de novo
cases of LQTS.
Methods
Clinical Subjects
A 1340-g black male infant was delivered by emergency cesarean
section at 32 weeks’ gestation because of fetal arrhythmias and
hydrops. Ventricular arrhythmias were initially diagnosed in utero at
28 weeks of gestation. An initial obstetrical ultrasound at 28 weeks
Received December 1, 2003; revision received March 2, 2004; accepted March 15, 2004.
From the Departments of Medicine (Division of Cardiology) (T.E.M., R.J.M., N.M., N.H.B.), Pediatrics (E.E., J.G.d.V., P.R., M.E.A., L.B., G.W.,
N.H.B.), and Molecular and Cellular Pharmacology (N.H.B.), University of Miami School of Medicine, and Miami Heart Research Institute (P.K.), Miami
Fla.
Correspondence to Nanette H. Bishopric, MD, FACC, FAHA, Professor of Pharmacology, Medicine and Pediatrics, Department of Molecular and
Cellular Pharmacology (R-189), PO Box 016189, Miami, FL 33101. E-mail n.bishopric@miami.edu
© 2004 American Heart Association, Inc.
Circulation is available at http://www.circulationaha.org DOI: 10.1161/01.CIR.0000130666.81539.9E
3029at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

2. revealed no gross abnormalities, but a subsequent fetal echocardio-
gram identified frequent bursts of atrial and ventricular tachycardia
and episodes of 2:1 AV block. ␤-Blockers were ineffective, and a
trial of amiodarone resulted in marked worsening of arrhythmias,
with development of myocardial dysfunction and fetal hydrops
necessitating emergency caesarian section. An ECG demonstrated
AV conduction abnormalities and a QTc interval of 550 milliseconds
(Figure 1A). After delivery, the newborn continued to have drug-re-
sistant ventricular tachyarrhythmias, frequent episodes of ventricular
flutter, and torsade de pointes (Figure 1B). A chest x-ray was
consistent with hyaline membrane disease and mild cardiomegaly.
Tachyarrhythmias were partially responsive to lidocaine infusion but
resisted a variety of other antiarrhythmic interventions, including
epicardial pacing, esmolol infusion, bicarbonate, magnesium, cal-
cium, mexiletine, flecainide, and left sympathetic ganglionectomy.
Ultimately, the infant received orthotopic heart transplantation at 5.5
months of age. He is now almost 3 years of age and has had no
further arrhythmias.
The infant’s mother was a 20-year-old black woman with no
clinical evidence or family history of heart disease, cardiac arrhyth-
mias, prematurity, syncope, cardiac arrest or sudden cardiac death, or
autoimmune disorders. Her baseline ECG was normal (Figure 1C),
Figure 1. Recurrent fetal demise and
fetal arrhythmias. A–C, ECG findings in
index case and mother. Leads I, II, and
VI are shown for each tracing. A, Index
case, 2 weeks of age: 3-lead ECG show-
ing prolonged QTc (550 ms) and 2:1 AV
block (arrows indicate p waves). B, Infant
ECG, showing torsade de pointes arising
on a background of ventricular paced
rhythm at 150 bpm and preceded by
T-wave alternans, a typical feature of
LQTS. C, Normal maternal ECG. D, Pedi-
gree of proband. Filled symbols indicate
intrauterine heart failure; IUFD, intrauter-
ine fetal death.
3030 Circulation June 22, 2004
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

3. and her QTc interval did not lengthen abnormally during amiodarone
infusion given for treatment of fetal arrhythmias (not shown). Her 1
previous pregnancy had ended in stillbirth at 28 weeks (Figure 1D);
a report of the postmortem evaluation of the male fetus described
marked cardiomegaly and a supernumerary digit but no other gross
abnormalities. Prenatal care was started 10 weeks into her second
pregnancy, and all prenatal laboratory studies were normal. The
infant’s father was 19 years of age and in excellent health, but he
declined to undergo genetic evaluation. The mother’s father and
mother were 23 and 18, respectively, at the time of her birth.
During these investigations, the mother became pregnant for a
third time by a second male partner. Regular prenatal examinations
and monthly ultrasound studies revealed no abnormalities until the
seventh month. At 28 weeks’ gestation, echocardiography revealed
no fetal heartbeat, and a stillborn female infant was delivered by
induction. Autopsy of this fetus revealed a severely enlarged but
structurally normal heart and nonimmune hydrops fetalis, consistent
with low-output cardiac failure; other organ systems were unremark-
able. Cord blood was obtained for genetic analysis.
All subjects were recruited and evaluated in accordance with
regulations set forth by the University of Miami Committee for the
Protection of Human Subjects and under a human subjects research
protocol approved by the University of Miami Institutional Review
Board. Autopsy results on the stillborn fetuses and cord blood from
the third fetus were provided with written consent of the mother.
DNA Isolation
Genomic DNA was obtained from lymphocytes, cultured skin
fibroblasts, paraffin-embedded tissue, and buccal swab with a
Puregene DNA Isolation Kit (Gentra Systems). DNA yields were
determined spectrophotometrically.
Single-Stranded Conformational
Polymorphism Analysis
The first 300 base pairs of exon 28 of SCN5A were amplified with
polymerase chain reaction (PCR).3 PCR was also performed on most
of the exons from LQT1, LQT2, KCNE1, and KCNE2 with the use of
published primers and conditions.3,12 Single-stranded conformational
polymorphism (SSCP) analysis was performed on a Bio-Rad Protean
II xi Cell electrophoresis apparatus. Gels were made with or without
glycerol using MDE Gel Solution (Cambrex BioScience) according
to the manufacturer’s recommendations. Gels were developed with a
silver stain kit from Bio-Rad according to their instructions.
DNA Sequence Analysis
PCR products to be sequenced were purified with a QIAquick PCR
Purification Kit (Qiagen). Cycle sequencing reactions were per-
formed by use of reagents from Applied Biosystems according to
their recommendation. The products were resolved on an ABI Prism
3100 Genetic Analyzer and analyzed with ABI Sequence Analysis
software. HinfI (10 U/␮L, New England Biolabs) digestions were
done for 3 hours at 37°C in 20-␮L reactions with up to 400 ng of
PCR product and 10 U of enzyme. Digestion products were analyzed
by electrophoresis in 2% agarose gels and staining with ethidium
bromide. The intensity of bands was quantified with Sigmascan
software (SPSS Science).
Results
Mutation Analysis
To identify the molecular abnormality in the index case, we
conducted mutation screening of 5 cardiac ion channel genes,
including KCNQ1, HERG, SCN5A, minK, and MiRP-1, cov-
ering Ϸ85% of the mutations known to be associated with
LQTS.3 Analysis of the proband’s lymphocyte DNA revealed
an unusual banding pattern of the first half of exon 28 of
SCN5A on SSCP analysis. DNA sequencing of this fragment
revealed that the infant was heterozygous for a previously
reported mutation at nucleotide 48683,13,14 (Genbank
NM_198056), with both a normal guanine residue and a
mutant adenine residue (Figure 2A). This introduces a muta-
tion in the second position of codon 1623, which changes a
positively charged arginine to a neutral glutamine (R1623Q).
No mutations were found in any of the other genes
investigated.
Detection of Maternal Mosaicism
Initial SSCP analysis of the mother’s lymphocyte DNA gave
an ambiguous result (not shown), but DNA sequencing
suggested that she was homozygous for the wild-type allele
(Figure 2B and 2E). To resolve this discrepancy, we took
advantage of the fact that the R1623Q mutation destroys an
Figure 2. Mutation detection in proband, mother, and third
fetus. A, B, D, and E, DNA sequence of sense strand of indi-
cated source. Arrows show position of mutation at nucleotide
4865. C, PCR products of DNA spanning mutation, after diges-
tion with HinfI, amplified from maternal lymphocytes (ML), pro-
band lymphocytes (PL), cord blood from third fetus (CB), mater-
nal fibroblasts (MF), maternal buccal epithelium (MB), and
lymphocytes from 3 unrelated individuals (NC). A, Proband lym-
phocyte DNA; B, maternal lymphocyte DNA. D, Maternal DNA
resistant to HinfI digestion; E, DNA from unaffected, unrelated
individual. Note that intensity of mutant band in stillborn cord
blood is intermediate between that of heterozygous brother (PL)
and the mosaic mother (ML), possibly reflecting incursion of
maternal blood after fetal demise.
Miller et al Recurrent Fetal Loss and Maternal LQTS Mosaicism 3031
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

4. HinfI restriction enzyme site (GANTC3AANTC). As ex-
pected, the wild-type exon 28 PCR product was completely
digested by HinfI, but more than half of the DNA from the
index case remained undigested, corresponding to the pres-
ence of homoduplex mutant and heteroduplex strands lacking
the restriction site (Figure 2C, lanes NC and PL). Analysis of
the mother’s lymphocyte DNA also revealed a small amount
of material resistant to complete digestion (Figure 2C, lanes
ML, MF, and MB). When purified, amplified, and sequenced,
this DNA was found to consist almost exclusively of the
mutant allele (Figure 2D), consistent with maternal somatic
mosaicism for the R1623Q mutation.
We next used the HinfI assay to compare levels of
mosaicism in several different maternal tissues (Figure 2C).
Using a standard curve developed by serial dilution of
proband DNA into normal DNA, we determined that the
HinfI assay was sufficiently sensitive to permit detection of
mutant DNA corresponding to Ϸ4% heterozygous cells.
Comparison of DNA from 3 different tissues suggested that
mosaicism was greatest in circulating lymphocytes (Ϸ10%),
less in buccal epithelium (8%), and least in skin fibroblasts
(4%) (Figure 2C). These differences may reflect genuine but
small variations in the proportion of mutation-bearing cells or
alternatively variability in PCR amplification. A comparably
low level of mosaicism in the heart would be consistent with
the mother’s lack of symptoms or ECG findings, although
this could not be directly confirmed.
The mother’s third pregnancy ended in stillbirth at 28
weeks. Analysis of fetal cord blood by HinfI digestion and
sequencing revealed the presence of the mutant allele (Figure
2C, lane CB).
Discussion
In this case report, we describe the late-term loss or near loss
of 3 successive pregnancies in a woman with somatic
mosaicism for a lethal LQTS mutation. Although material for
genetic studies was available only for 2 of the 3 infants, we
provide strong evidence for germ-line transfer of the mutation
in at least 2 and probably 3 pregnancies. Clinical and
postmortem evidence suggests that all 3 pregnancies were
complicated by fetal cardiac decompensation, and in the
second infant, heart failure was observed to develop in
connection with incessant ventricular tachyarrhythmias.
Heart failure, bradycardia, cardiac hypertrophy, and
tachyarrhythmias are frequent features of the antenatal pre-
sentation of LQTS.15
This case provides evidence that late-term fetal loss, like
SIDS, can stem from unsuspected LQTS. Fetal demise after
the sixth month of gestation is rare, occurring in Ͻ4 per 1000
pregnancies (National Center for Health Statistics, http://
www.cdc.gov/nchs/about/major/fetaldth/abfetal.htm), and
known causes (Rh incompatibility, maternal autoimmune
disease, thrombophilia, infection with TORCH pathogens,
major congenital malformations) were excluded in each of
the 3 pregnancies reported here. As with SIDS, fetal LQTS
and neonatal LQTS are rarely familial, and death may occur
in the absence of a documented ventricular tachyarrhythmia
or suggestive family history.16,17 Since Schwartz18 originally
proposed that some SIDS cases may reflect congenital LQTS,
several lines of supportive evidence have emerged, including
an ECG screening study of newborns,19 a case of a near-miss
SIDS infant who carried an apparently de novo SCN5A
mutation,5 and a case of a SIDS infant in whom postmortem
molecular screening identified an apparently de novo muta-
tion on KCNQ1.7 More recently, 2 of 93 sequential cases of
SIDS were found to have missense mutations in SCN5A.6 The
latter study likely underestimates the true contribution of
LQTS to unexplained infant death because even in clinically
well-defined cases, mutations are not always identified.3,4
Our findings demonstrate that LQTS mutations, particularly
those affecting the sodium channel encoded by SCN5A,
confer a risk for late-term fetal loss and for infant sudden
death.
In this case, the mother was found to harbor a small
population of cells with the same SCN5A (R1623Q) mutation
found in her children, indicating that she is a somatic mosaic
for LQTS. Genetic mosaicism results when a mutation occurs
in a single cell of the developing embryo and is propagated
only to its descendant cells in the adult. Mosaicism can affect
either somatic or germ-line lineages or both, depending on the
location and embryological stage at which the mutation
occurs. It has been estimated that 10% to 20% of ostensibly
de novo cases of retinoblastoma,9 hemophilia B,10 Duchenne
muscular dystrophy,8 and hemophilia A20 are the result of
parental mosaicism for these genetic disorders. Examples of
germ-line mosaicism in asymptomatic parents have been
documented in several autosomal dominant genetic diseases,
including neurofibromatosis,21 osteogenesis imperfecta,22 and
hypertrophic cardiomyopathy.23
The R1623Q mutation was first reported in an Asian
female infant who, like our proband, presented perinatally
with 2:1 AV block, lengthened QT interval, cardiac failure,
and life-threatening ventricular arrhythmias. The child is now
at least 8 years old and is free of arrhythmias on 600 mg/d
mexiletine.24 Our patient is now almost 3 years old and is also
free of arrhythmias after receiving an orthotopic heart trans-
plant. A third, sporadic instance of this mutation has also
been reported in an infant.3,25 The severe phenotype mani-
fested by these infants is typical of a subset of LQTS patients
who present in fetal and early neonatal life.5,13,16,17,26 A
review of 22 case studies from the English language litera-
ture16 described a subset of LQTS patients with a severe
neonatal-onset phenotype characterized by very long QTc
intervals (mean, 665 milliseconds), 2:1 AV block, T-wave
alternans, functional bradycardia, and a poor prognosis, with
9 of 22 dead in the first year of life and 50% mortality by 4
years of age. In comparison, a study of LQTS in a large
pediatric collaborative study (mean age at diagnosis, 7
years)27 found a mean QTc of 520 milliseconds and a low rate
of AV block (4%). A further striking feature of the perinatal
LQTS phenotype was the infrequency of a family history of
the disease (1 of 22, or 4.5%, compared with 60% in the
collaborative study of pediatric LQTS16). The important role
of de novo mutations in perinatal LQTS is supported by the
fact that disease-causing mutations at the same SCN5A
residue (1623) have arisen independently at least 3
times,3,14,25 and a number of other spontaneous mutations in
LQTS-associated genes have been linked to lethal arrhyth-
3032 Circulation June 22, 2004
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

5. mias in the fetus and neonate.5,28,29 The poor prognosis of
infants with neonatal LQTS undoubtedly also contributes to
low heritability.
The evidence for SCN5A mutations as a cause of perinatal
LQTS seems to conflict with earlier studies showing that
patients with LQTS generally have clinical manifestations
later in life than patients with mutations at other loci and that
SCN5A mutation penetrance is strongly age dependent.30
However, the findings presented here demonstrate that certain
SCN5A mutations can have an early, aggressive phenotype
characterized by arrhythmias and death before birth or in
early infancy. These mutations do not appear to overlap with
those previously reported to cause disease later in life.30
Additional work is required to explain the mechanisms
underlying these distinct clinical presentations.
The mutation detected in our patient lies in the S4 region of
SCN5A domain IV, which participates in coupling of channel
activation and inactivation.31 DNA sequence alignment of the
human isoforms of this sodium channel shows that the 3
brain-type (1A, 2A, and 3A), the skeletal muscle (4A), and
the cardiac (5A) isoforms are Ͼ90% identical in this region,
suggesting that this part of the protein serves a general and
essential function of the channel. The severity of disease
associated with mutations in this portion of the channel
further underscores its functional importance. In addition to 3
apparently de novo R1623Q mutations being associated with
a severe, early-onset phenotype of LQTS (this report and
others14,25), the same residue, when mutated to leucine instead
of glutamine, also causes the Romano-Ward form of LQTS,3
and another mutation 3 residues away has been associated
with the Brugada syndrome.32 Furthermore, mutation of the
homologous residue (R1448) in the skeletal SCN4A isoform
has been reported to cause paramyotonia congenita,33 a
disease characterized by cold-induced myotonia.34 In vitro
functional studies on the R1623Q mutant protein have found
unique inactivation gating defects that lead to both prolonged
spontaneous opening and early reopening, resulting in pro-
longed decay of INa.35 These gating defects were sensitive to
lidocaine, which restored more normal rapid inactivation
kinetics; possibly consistent with this, lidocaine was the only
antiarrhythmic agent demonstrating partial clinical benefit in
the surviving infant reported here.
The role of genetics in fetal demise is likely to be
overlooked, particularly when parents have no evidence of
disease. Dominant mutations such as those leading to LQTS,
hypertrophic cardiomyopathy, and other disorders associated
with sudden death are often assumed to have arisen de novo
in the affected individual (see elsewhere36 for review). The
possibility of genetic mosaicism might not be explored, partly
because of technical difficulties in detection and partly
because of a low level of suspicion when both parents are
clinically normal. Nonetheless, the distinction between de
novo mutation and parental germ-line mosaicism is critical
because mosaicism confers a considerably higher risk of
disease recurrence in subsequent pregnancies. The frequency
of de novo or transmitted mosaic LQTS mutations in fetal
demise or SIDS remains to be established. Our observation of
repeated germ-line transmission of LQTS from a mother with
unsuspected mosaicism implies that perinatal cardiac moni-
toring, in addition to genetic counseling, should be consid-
ered after recurrent late-term fetal loss or infant death.
Genetic disorders linked to cardiac sudden death may be an
important focus for future genetic epidemiological studies of
unexplained third-trimester pregnancy loss.
Acknowledgments
This work was supported by a collaborative grant from the Miami
Heart Research Institute (N.H.B. and R.J.M.), a Grant-in-Aid from
the Florida-Puerto Rico Affiliate of the American Heart Association
(L.B.), and a generous contribution from the Dr John T. Macdonald
Foundation for Medical Genetics. We are grateful to Drs Mark
Keating and Igor Splawski for providing control DNA samples,
unpublished data, and technical suggestions throughout these exper-
iments. We thank Dr Louis Elsas for his review and comments on the
manuscript.
References
1. Keating MT, Sanguinetti MC. Molecular and cellular mechanisms of
cardiac arrhythmias. Cell. 2001;104:569–580.
2. Mohler PJ, Schott JJ, Gramolini AO, et al. Ankyrin-B mutation causes
type 4 long-QT cardiac arrhythmia and sudden cardiac death. Nature.
2003;421:634–639.
3. Splawski I, Shen J, Timothy KW, et al. Spectrum of mutations in long-QT
syndrome genes. KVLQT1, HERG, SCN5A, KCNE1, and KCNE2. Circu-
lation. 2000;102:1178–1185.
4. van Langen IM, Birnie E, Alders M, et al. The use of genotype-phenotype
correlations in mutation analysis for the long QT syndrome. J Med Genet.
2003;40:141–145.
5. Schwartz PJ, Priori SG, Dumaine R, et al. A molecular link between the
sudden infant death syndrome and the long-QT syndrome. N Engl J Med.
2000;343:262–267.
6. Ackerman MJ, Siu BL, Sturner WQ, et al. Postmortem molecular analysis
of SCN5A defects in sudden infant death syndrome. JAMA. 2001;286:
2264–2269.
7. Schwartz PJ, Priori SG, Bloise R, et al. Molecular diagnosis in a child
with sudden infant death syndrome. Lancet. 2001;358:1342–1343.
8. van Essen AJ, Abbs S, Baiget M, et al. Parental origin and germline
mosaicism of deletions and duplications of the dystrophin gene: a
European study. Hum Genet. 1992;88:249–257.
9. Sippel KC, Fraioli RE, Smith GD, et al. Frequency of somatic and
germ-line mosaicism in retinoblastoma: implications for genetic coun-
seling. Am J Hum Genet. 1998;62:610–619.
10. Ketterling RP, Vielhaber E, Li X, et al. Germline origins in the human F9
gene: frequent G:C3A:T mosaicism and increased mutations with
advanced maternal age. Hum Genet. 1999;105:629–640.
11. Moyhuddin A, Baser ME, Watson C, et al. Somatic mosaicism in neu-
rofibromatosis 2: prevalence and risk of disease transmission to offspring.
J Med Genet. 2003;40:459–463.
12. Abbott GW, Sesti F, Splawski I, et al. MiRP1 forms IKr potassium
channels with HERG and is associated with cardiac arrhythmia. Cell.
1999;97:175–187.
13. Makita N, Shirai N, Nagashima M, et al. A de novo missense mutation of
human cardiac Naϩ
channel exhibiting novel molecular mechanisms of
long QT syndrome. FEBS Lett. 1998;423:5–9.
14. Yamagishi H, Furutani M, Kamisago M, et al. A de novo missense
mutation (R1623Q) of the SCN5A gene in a Japanese girl with sporadic
long QT syndrome: mutations in brief no. 140: online. Hum Mutat.
1998;11:481.
15. Cuneo BF, Ovadia M, Strasburger JF, et al. Prenatal diagnosis and in
utero treatment of torsades de pointes associated with congenital long QT
syndrome. Am J Cardiol. 2003;91:1395–1398.
16. Trippel DL, Parsons MK, Gillette PC. Infants with long-QT syndrome
and 2:1 atrioventricular block. Am Heart J. 1995;130:1130–1134.
17. Gorgels AP, Al Fadley F, Zaman L, et al. The long QT syndrome with
impaired atrioventricular conduction: a malignant variant in infants.
J Cardiovasc Electrophysiol. 1998;9:1225–1232.
18. Schwartz PJ. Cardiac sympathetic innervation and the sudden infant death
syndrome: a possible pathogenetic link. Am J Med. 1976;60:167–172.
19. Schwartz PJ, Stramba-Badiale M, Segantini A, et al. Prolongation of the
QT interval and the sudden infant death syndrome. N Engl J Med.
1998;338:1709–1714.
Miller et al Recurrent Fetal Loss and Maternal LQTS Mosaicism 3033
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

6. 20. Leuer M, Oldenburg J, Lavergne JM, et al. Somatic mosaicism in hemo-
philia A: a fairly common event. Am J Hum Genet. 2001;69:75–87.
21. Lazaro C, Gaona A, Lynch M, et al. Molecular characterization of the
breakpoints of a 12-kb deletion in the NF1 gene in a family showing
germ-line mosaicism. Am J Hum Genet. 1995;57:1044–1049.
22. Lund AM, Schwartz M, Raghunath M, et al. Gly802Asp substitution
in the pro alpha 2(I) collagen chain in a family with recurrent osteo-
genesis imperfecta due to paternal mosaicism. Eur J Hum Genet.
1996;4:39–45.
23. Forissier JF, Richard P, Briault S, et al. First description of germline
mosaicism in familial hypertrophic cardiomyopathy. J Med Genet. 2000;
37:132–134.
24. Miura M, Yamagishi H, Morikawa Y, et al. Congenital long QT
syndrome and 2:1 atrioventricular block with a mutation of the SCN5A
gene. Pediatr Cardiol. 2003;24:70–2.
25. Westenskow P, Splawski I, Timothy KW, et al. Compound mutations: a
common cause of severe long-QT syndrome. Circulation. 2004;109:
1834–1841.
26. Chang IK, Shyu MK, Lee CN, et al. Prenatal diagnosis and treatment of
fetal long QT syndrome: a case report. Prenat Diagn. 2002;22:
1209–1212.
27. Garson A Jr, Dick M 2nd, Fournier A, et al. The long QT syndrome in
children: an international study of 287 patients. Circulation. 1993;87:
1866–1872.
28. Valdivia CR, Ackerman MJ, Tester DJ, et al. A novel SCN5A arrhythmia
mutation, M1766L, with expression defect rescued by mexiletine. Car-
diovasc Res. 2002;55:279–289.
29. Wedekind H, Smits JP, Schulze-Bahr E, et al. De novo mutation in the
SCN5A gene associated with early onset of sudden infant death. Circu-
lation. 2001;104:1158–1164.
30. Schwartz PJ, Priori SG, Spazzolini C, et al. Genotype-phenotype corre-
lation in the long-QT syndrome: gene-specific triggers for life-threatening
arrhythmias. Circulation. 2001;103:89–95.
31. Stuhmer W, Conti F, Suzuki H, et al. Structural parts involved in acti-
vation and inactivation of the sodium channel. Nature. 1989;339:
597–603.
32. Chen Q, Kirsch GE, Zhang D, et al. Genetic basis and molecular
mechanism for idiopathic ventricular fibrillation. Nature. 1998;392:
293–296.
33. Ptacek LJ, George AL Jr, Barchi RL, et al. Mutations in an S4 segment
of the adult skeletal muscle sodium channel cause paramyotonia con-
genita. Neuron. 1992;8:891–897.
34. McClatchey AI, Van den Bergh P, Pericak-Vance MA, et al.
Temperature-sensitive mutations in the III-IV cytoplasmic loop region of
the skeletal muscle sodium channel gene in paramyotonia congenita. Cell.
1992;68:769–774.
35. Kambouris NG, Nuss HB, Johns DC, et al. Phenotypic characterization of
a novel long-QT syndrome mutation (R1623Q) in the cardiac sodium
channel. Circulation. 1998;97:640–644.
36. Zlotogora J. Germ line mosaicism. Hum Genet. 1998;102:381–386.
3034 Circulation June 22, 2004
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from

7. H. Bishopric
Paolo Rusconi, Mary Ellen Ahearn, Lisa Baumbach, Paul Kurlansky, Grace Wolff and Nanette
Todd E. Miller, Elicia Estrella, Robert J. Myerburg, Jocelyn Garcia de Viera, Niberto Moreno,
Recurrent Third-Trimester Fetal Loss and Maternal Mosaicism for Long-QT Syndrome
Print ISSN: 0009-7322. Online ISSN: 1524-4539
Copyright © 2004 American Heart Association, Inc. All rights reserved.
is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation
doi: 10.1161/01.CIR.0000130666.81539.9E
2004;109:3029-3034; originally published online June 7, 2004;Circulation.
http://circ.ahajournals.org/content/109/24/3029
World Wide Web at:
The online version of this article, along with updated information and services, is located on the
http://circ.ahajournals.org//subscriptions/
is online at:CirculationInformation about subscribing toSubscriptions:
http://www.lww.com/reprints
Information about reprints can be found online at:Reprints:
document.Permissions and Rights Question and Answerthis process is available in the
click Request Permissions in the middle column of the Web page under Services. Further information about
Office. Once the online version of the published article for which permission is being requested is located,
can be obtained via RightsLink, a service of the Copyright Clearance Center, not the EditorialCirculationin
Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:
at VA MED CTR BOISE on February 10, 2016http://circ.ahajournals.org/Downloaded from