REVIEW URRENT C OPINION

The beginnings of long QT syndrome Bettina F. Cuneo

Purpose of review The purpose of this study is to update the perinatal cardiologist and obstetrical care provider on the presentation and management of the fetus with long QT syndrome (LQTS). Recent findings LQTS is a known cause of sudden death in childhood, adolescence and young adulthood that presents during fetal life, but is often not recognized. Torsades de pointes (TdP) 28 atrioventricular block (AVB) are not always attributed to LQTS, although the most common LQTS rhythm, a fetal heart rate of less than third percentile for gestational age (GA), is not recognized as abnormal because it does not meet the standard obstetrical criteria for bradycardia. Early recognition and appropriate treatment can be life saving for the fetus and unsuspecting LQTS family members. Fetal rhythm phenotype and postnatal QTc can predict postnatal rhythm and suggest genotype: bradycardic fetuses usually have KCNQ1 mutation, while those with TdP and/or a postnatal QTc more than 500 ms have SCN5A, KCNH2 or uncharacterized mutations. Summary The fetus with repeated heart rates of less than third percentile of GA and those with TdP 28 AVB are likely to manifest the same rhythm after birth and have an LQTS mutation. Keywords arrhythmia, fetus, long QT syndrome, torsades de pointes

INTRODUCTION It has been recognized for over 50 years that the long QT syndrome (LQTS) is a cause of sudden death in childhood [1]. Although known to be ‘congenital’ and despite reports of stillbirths in LQTS families, it was not until 1995 that the ‘beginnings’ of LQTS were recognized in a healthy 38-week fetus with moderate bradycardia who delivered without incident at term [2]. This unassuming case report ushered in the era of fetal LQTS diagnosis, the opportunity to risk stratify and ultimately treat the fetus with LQTS.

RISK FACTORS AND CLINICAL PRESENTATION There are three groups of fetuses at an increased risk for LQTS. The first are those fetuses with a family history of LQTS. As LQTS is inherited as an autosomal dominant mutation, each fetus has a 50% chance of carrying the family mutation. There is phenotypic variability between family members with the same mutation; thus, the prenatal/postnatal course cannot always be predicted by the affected family member’s phenotype. The second group at risk for LQTS is those with no family history or an unrecognized family history www.co-cardiology.com

of LQTS who present with one of the LQTS rhythms. The signature LQTS rhythms – torsades de pointes (TdP) 28AV block – occur in about 25% of LQTS fetuses and often require in-utero treatment. The majority of LQTS fetuses will have sinus rhythm at a rate repeatedly less than third percentile for gestational age [3]. Preliminary data suggest that the TdP rhythm phenotype is more likely to occur with a de-novo LQTS mutation, germ cell mosaicism, a sporadic uncharacterized mutation or a family KCNH2 mutation in the pore region of the gene [4,5,6 ,7 ,8–10]. Another group of fetuses at risk for LQTS are those with a family history of intrauterine fetal demise (IUFD) or sudden infant death syndrome (SIDS) [11 ,12]. It is not yet known whether mutations reported in those with stillbirth or SIDS &

&

&

Department of Pediatrics and Obstetrics, University of Colorado School of Medicine, Children’s Hospital Colorado Heart Institute, The Colorado Institute for Maternal and Fetal Health, Aurora, Colorado, USA Correspondence to Bettina F. Cuneo, MD, Children’s Hospital Colorado, Box 100, 13123 E. 16th Ave, Aurora, CO 80045, USA. Tel: +1 720 777 2943; fax: +1 720 777 7290; e-mail: Bettina.Cuneo@childrenscolorado. org Curr Opin Cardiol 2015, 30:112–117 DOI:10.1097/HCO.0000000000000135 Volume 30  Number 1  January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

The beginnings of long QT syndrome Cuneo

KEY POINTS

(a) Atrium

 Long QT syndrome presents in fetal life with repeated heart rates of less than third percentile for gestational age, TdP and/or 28 AV block.  Treatment of in-utero TdP is possible with transplacental intravenous magnesium alone or with lidocaine.  The fetus with prenatal TdP will be very likely to have postnatal TdP and either a de-novo SCN5A or an uncharacterized mutation, or a family mutation in KCNH2.

Ventricle

(b) Ventricle

 We do not understand why the same mutation causes a variety of phenotypes.

Atrium

are primarily de-novo mutations or whether they are found in other family members.

DIAGNOSIS OF LONG QT SYNDROME The diagnosis of LQTS rests on a high clinical suspicion. In some cases, the LQTS rhythms are fleeting or not recognized to be abnormal. The diagnosis is most challenging when the fetus presents with sinus rhythm, because only 17% of LQTS fetal heart rates meet the obstetrical definition of bradycardia: a fetal heart rate of 110 beats per minute or less [13]. However, almost 70% of LQTS fetal heart rates are less than third percentile for gestational age, which fall within the obstetrician’s range of normal heart rates [3]. Thus, an opportunity for prenatal recognition of LQTS is lost because of ascertainment bias. In addition to the fetal heart rate, a common feature of the LQTS fetus noted after 32 weeks is a nonreactive heart rate tracing. Because of a nonreactive tracing, LQTS fetuses can be delivered prematurely and emergently because of concerns for fetal distress [14]. The LQTS fetus can also present with incessant TdP causing heart failure and fetal hydrops. TdP can also be mistaken for supraventicular tachycardia and treated with QT-prolonging antiarrhythmic agents such as sotalol or amiodarone [7 ,15]. The diagnosis of ventricular tachycardia is based on the echocardiographic findings of atrioventricular (AV) dissociation with an atrial rate slower than the ventricular rate, whereas in supraventricular arrhythmia either the AV relation is 1 : 1 or the atrial rate is faster than the ventricular rate (atrial flutter) (Fig. 1). Periods of 28 block can also be seen in fetal LQTS; they can be isolated or occur with episodes of TdP [7 ,16]. The LQTS AV block occurs not because of conduction system disease per se, but &

&

FIGURE 1. M-mode echocardiograms of different types of fetal tachycardia. (a) Fetal ventricular tachycardia. The atrial rate is 110 and the ventricular rate is 260 bpm. (b) Atrial flutter. The atrial rate is 360 and the ventricular rate is 180 bpm.

because ventricular repolarization is so prolonged that the atrium is activated before the ventricle completely repolarizes. The findings of TdP with 28 AV block are helpful to confirm the diagnosis of LQTS. Interestingly, the timeline of LQTS rhythms seems related to gestational age. A slow sinus rate can be seen as early as 14 weeks, 28 AVB has been typically described at 18–40 weeks, but TdP generally does not occur until after 29 weeks [8–10]. Other echocardiographic features of LQTS include ‘18 AV block’ during sinus rhythm, defined as a prolonged mechanical PR interval [17] (Fig. 2). This finding may be primarily a prolonged isovolumic contraction time, because PR intervals measured during fetal magnetocardiography (fMCG, see next paragraph) have been normal (Fig. 3). Even during periods of LQTS 28 AV block, the prolonged isovolumic contraction time can be appreciated and differentiate true conduction system disease caused by maternal Sjogren’s antibodies from LQTS [18 ]. Biventricular hypertrophy and noncompaction can also be seen in fetal LQTS [15,19,20]. A diagnosis in utero of LQTS rests upon commercial genetic testing of fetal cells obtained by amniocentesis or percutaneous umbilical artery sampling (PUBS) [21]. Unfortunately, maternal cell-free DNA testing for fetal LQTS mutations is not yet available. A noninvasive diagnosis of LQTS is possible with findings of a prolonged QT interval during fMCG [22 ] (Fig. 3). fMCG gives precise electrophysiological data including beat-to-beat variability, QRS, QT

0268-4705 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

&&

&

www.co-cardiology.com

113

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Pediatrics

(a)

(a)

20 QRS: 86 ms

15 RR: 931 ms

PR: 109 ms

10

P: 46 ms

QT: 668 ms QTc: 692 ms

5

0

(b)

–5

0

0.5

1

1.5

2

2.5

Time (s)

(b) 8

6

4

2

0

FIGURE 2. Simultaneous mitral inflow/aortic outflow Doppler of fetuses with a slow sinus rate demonstrating the prolonged isovolumic contraction times in two fetuses with (a) SCN5A mutation and (b) CALM 2 mutation. The mechanical PR intervals are 170–180 ms, while the fMCG PR intervals were normal (see Fig. 3).

–2 P: 33 ms PR: 60 ms

–4

QT: 409 ms QTc: 628 ms

RR: 424 ms

–6 QRS: 73 ms

–8 0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

1.4

1.6

1.8

2

Time (s)

and P wave duration and repolarization characteristics by recording the primary (0.5–2 pT) magnetic fields generated by the electric currents in the fetal and maternal heart. Unfortunately fMCG is still a research tool, limited to only a few centres worldwide. In a recent study, LQTS was correctly identified or excluded by direct measurement of the corrected QT interval during fMCG in 29/30 fetuses at risk for LQTS [22 ].

(c)

3 QRS: 44 ms

2.5 2

RR: 524 ms PR: 81 ms

1.5

P: 36 ms

QT: 392 ms QTc: 541 ms

1 0.5 0

&

0.5 –1

BEYOND DIAGNOSIS: THE ELECTROPHYSIOLOGY OF FETAL LONG QT SYNDROME In addition to direct measurement of the QT interval and confirmation of TdP morphology, fMCG provides insight into the electrophysiology of LQTS before birth [22 ]. We studied 21 genotype and/or phenotype-positive individuals by fMCG; four of the 21 had episodes of TdP in utero. Episodes of TdP lasted 1–459 s, and mean ventricular rate varied 170–250 bpm (Fig. 4). Half of the fetuses were hydropic; none died suddenly in utero. Patterns of initiation and termination of TdP varied between individuals and between episodes in the same individuals, and included patterns not seen after birth. For example, initiation of TdP was not pause dependent but followed an aberrantly &

114

www.co-cardiology.com

1.5 –2 0

0.2

0.4

0.6

0.8

1

1.2

Time (s)

FIGURE 3. Magnetocardiographic signal averaged fetal heart rate tracings of three fetuses with long QT syndrome. Note that the PR intervals of (a) and (c) are normal, while the mechanical PR intervals shown of the same patients in Fig. 2 are prolonged. The varieties of T-wave morphology are demonstrated. (a) SCN5A R1623Q mutation; (b) SCN5A L409P; (c) CALM 2 mutation.

conducted beat with little or no change in cycle length; TdP spontaneously terminated into multiple rhythms but did not degenerate into ventricular fibrillation despite prolonged and frequently recurrent episodes. Repolarization abnormalities, Volume 30  Number 1  January 2015

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

The beginnings of long QT syndrome Cuneo SCN5A R1623Q 1–20 (10)s 6

SCN5A L409P 2–194 (17)s 5

KCNH2 T613K 3–459 (86)s 4

KCNH2 G628S 10–69 (31)s 3

Top

Sinus

Heart block

2400

Time/sec

FIGURE 4. Chronology of long QT syndrome rhythms during continuous magnetocardiographic monitoring of four long QT syndrome fetuses. Numbers represent range and mean times in TdP.

including T-wave and QRS alternans, were also present in seven out of 21 individuals, five of whom had pre or postnatal TdP.

RISK STRATIFICATION It has been reported that prenatal rhythm phenotype suggests genotype, facilitating the risk stratification of fetal/neonatal LQTS [7 ]. In 43 LQTS individuals identified during fetal life, 32 out of 43 had sinus rhythm at a rate of less than third percentile for gestational age (GA), the shortest neonatal mean QTc interval (compared with those with other fetal LQTS rhythms), and the highest proportion of KCNQ1 mutations. No fetus with an isolated sinus bradycardia had neonatal TdP. On the contrary, individuals with in-utero TdP all had postnatal TdP, a significantly longer neonatal mean QTc interval, and either SCN5A or KCNH2 mutations. Two individuals with SCN5A (R1623 Q) mutations died in utero or suddenly at 4 months of age [4,15]. It was also noted that the fetus with a family history of LQTS was more likely to present with isolated sinus rhythm than with TdP or 28 AV block; the exception, as shown in this report and others, would be a KCNH2 family mutation. The four remaining individuals in this study presented with isolated 28 AV block. Only one of the four individuals had a neonatal QTc of more than 600 ms and this fetus developed neonatal TdP. This infant also had a mutation in CALM 2 [6 ]. These results suggest that rhythm phenotype and neonatal QTc are helpful in predicting both the genotype and the risk for neonatal TdP. Another feature of the fetal TdP is the very consistent phenotype of the de-novo SCN5A R1523 mutations and the phenotypic variability &

&

of the familial KCNH2 mutations. Thus, a fetus with a family KCNH2 mutation may present with TdP, heart failure and a very prolonged postnatal QTc, although the proband is asymptomatic with a QTc in the near normal range [23]. This phenotypic variability, along with the number of family members with no recognized symptoms, underscores the importance of cascade testing, or identifying family members who are carriers of the same mutation in the event the fetus is identified as proband [3].

IN-UTERO MANAGEMENT AND TREATMENT Recommendations for care of the fetus with a family history of LQTS include regular surveillance of fetal heart rate and rhythm, withholding QTc-prolonging medications during the pregnancy, delivering at a cardiac centre with experienced paediatric electrophysiologists and paediatric cardiothoracic surgeons, and postnatal genetic testing [24 ]. The asymptomatic LQTS fetus with a heart rate of less than third percentile or 28 AV block needs only to be observed. The premature LQTS fetus with TdP and hydrops need not be delivered emergently for several reasons. First, transplacental intravenous magnesium has restored sinus rhythm and resolved hydrops in premature fetuses with either SCN5A or KCNH2 mutations [8,15,16]. Second, in the event that monotherapy with intravenous magnesium is unsuccessful in rhythm control, a continuous infusion of lidocaine can be added [15]. Third, as prenatal TdP predicts postnatal TdP, the likelihood that these infants will undergo placement of a dual chamber pacemaker or internal cardiac defibrillator is high [7 ,10]. Placing either of these devices can be technically more challenging in a premature infant.

0268-4705 ß 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins

&&

&

www.co-cardiology.com

115

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Pediatrics

Yet, despite the success of in-utero therapy, over 80% of fetuses with TdP were delivered prematurely and without treatment, putting them at risk for an unfavourable outcome at worst, or many months in the hospital at best. Transplacental beta-adrenergic blocking agents such as propranolol have not successfully restored sinus rhythm in TdP and hydrops, and have been only partially effective in maintaining normal rhythm, most likely because the transplacental transfer rate is only 25–30% [25]. They can be considered in the fetus with intermittent, very short-lived runs of TdP and no heart failure. Other ways to optimize the outcome of fetal LQTS include withholding QTc-prolonging agents commonly used during pregnancy such as pitocin, used to augment labour, ondansetron, an antiemetic, and erythromycin, and optimizing maternal magnesium, vitamin D and calcium levels in the hope of maintaining sinus rhythm. If it is known that the fetus has LQTS, premature delivery can be postponed if there is 28 AV block or a nonreactive fetal heart rate tracing. Finally, anticipatory postnatal care can improve pre and postnatal outcome [10].

CONCLUSION The next steps in improving the outcome of LQTS individuals of all ages include improved ascertainment during fetal life and the recognition that in-utero treatment is available and effective. Ascertainment of fetal LQTS is incomplete for several reasons. First, it is a rare disease, occurring in one out of 2000 individuals; thus, most obstetrics practices could potentially see one patient every 4 years. Second, the history obtained in the obstetrics office does not include leading questions relating to LQTS, such as a family history of LQTS, recurrent syncope, sudden unexplained death at a young (

The beginnings of long QT syndrome.

The purpose of this study is to update the perinatal cardiologist and obstetrical care provider on the presentation and management of the fetus with l...
309KB Sizes 1 Downloads 12 Views