Pediatric Cardiology

, Volume 30, Issue 6, pp 747–751

Late Ventricular Potentials and QT Dispersion in Familial Dysautonomia


  • Udi Nussinovitch
    • Department of Internal Medicine BChaim Sheba Medical Center
    • Sackler Faculty of MedicineTel Aviv University
  • Uriel Katz
    • Department of Pediatric CardiologyChaim Sheba Medical Center
    • Sackler Faculty of MedicineTel Aviv University
  • Moshe Nussinovitch
    • Department of Pediatrics CSchneider Children’s Medical Center of Israel
    • Sackler Faculty of MedicineTel Aviv University
    • Hypertension Unit and Department of Internal Medicine DChaim Sheba Medical Center
    • Sackler Faculty of MedicineTel Aviv University
Original Article

DOI: 10.1007/s00246-009-9419-0

Cite this article as:
Nussinovitch, U., Katz, U., Nussinovitch, M. et al. Pediatr Cardiol (2009) 30: 747. doi:10.1007/s00246-009-9419-0


Familial dysautonomia is a worldwide disorder characterized by maldevelopment and dysfunction of the autonomic and sensory systems. Despite major improvements in disease management in recent years, sudden death remains the cause of death in up to 43% of patients. The aim of this study was to evaluate electrocardiographic markers of sudden death in familial dysautonomia. A comparative case series design was used. Electrocardiographic measurements were performed in 13 patients with familial dysautonomia, 7 male and 6 female, aged 9–46 years. QT was measured from all leads and corrected QT (QTc) was calculated with the Bazett formula. QT dispersion (QTd), a marker of arrhythmogenicity, was calculated and corrected for heart rate. Late ventricular potential parameters, predictive of arrhythmias, were calculated as well. Findings were compared to a matched control group using the Mann–Whitney–Wilcoxon test. A prolonged QT interval was noted in 30.7% of patients. Several QT dispersion parameters were significantly abnormal in the study group compared to the controls. All late potential parameters were within normal range in both groups. In conclusion, patients with familial dysautonomia commonly have electrocardiographic abnormalities and may be at a higher risk for adverse cardiac events.


Familial dysautonomiaQT dispersionLate ventricular potentials

Familial dysautonomia (FD) is an autosomal recessive disease characterized by maldevelopment and dysfunction of the sensory and autonomic nervous systems [1, 3]. It occurs worldwide and affects Ashkenazi Jews almost exclusively. A mutation in the IKBKAP gene on chromosome 9q31 is responsible for more than 99% of cases [19]. The carrier rate may be as high as 1 in 18 in high-risk populations [19].

The clinical characteristics of FD may include hypotonia, feeding difficulties, alacrimation, frequent changes in skin color, body temperature dysregulation, dysphagia, gastrointestinal dysmotility, recurrent aspiration pneumonia, and spinal deformity [12]. Recurrent dysautonomic crises occur in 40% of patients. The frequency is individual and may be daily, weekly, or less. The crises are triggered by emotional and physical stress, sleep deprivation, and visceral pain. Manifestations include vomiting, dysphagia, abdominal pain, hypertension, and tachycardia [1].

Substantial progress in the management of FD has been made in the last five decades. Today, gastrointestinal disorders are being treated earlier both surgically and medically, and infections are being treated more aggressively. As a result, the average life span of patients with FD has increased from <5 years to adulthood [2]. Nevertheless, there has been no change in the rate of sudden unexplained death, which remains the most common cause of death in FD, accounting for up to 43% of all cases [4].

Although some studies have suggested that sudden death may be due to severe bradyarrhythmias and unopposed vagal stimulation [3, 12, 13], placement of pacemakers had a limited effect on mortality rates in small-scale studies [6, 13]. Others suggested that a proportion of the cases of sudden death might be explained by electrocardiogram (ECG) findings of prolonged corrected QT (QTc) and JT intervals [10, 11, 16].

Abnormal depolarization and repolarization may be evaluated noninvasively by late ventricular potentials (LPs) and QT interval dispersion (QTd), respectively. LPs are low-amplitude, high-frequency signals observed at the end of the QRS complex. They may represent a conduction disturbance during depolarization [7] and serve as a marker of increased risk of arrhythmias. LPs can be detected using signal-averaged ECG (SAECG). Although SAECG has been recorded in patients with FD [5], only one of its parameters was published, which is insufficient for LP evaluation and proarrhythmogenicity risk stratification. The QTd is defined as the maximal minus the minimal QT interval. An interlead variation in QT interval reflects dispersion of ventricular repolarization [9, 18] and has been associated with a higher risk of induction of ventricular tachycardia on electrophysiological studies [24]. The QTd has been evaluated in patients with FD [10], but it was measured manually, and not all dispersion parameters were included.

The aim of the present study was to further evaluate ECG markers for sudden death in FD.


Study Design

A comparative case series design was used. The research protocol was approved by the Institutional Helsinki Committee. All participants gave written informed consent.

Study Subjects

The study group included 13 patients with FD, who account for about one-ninth of all known patients with FD currently living in Israel. Nine patients reported having occasional autonomic crises. None of the patients was a smoker, and all were diabetes-free. Findings were compared with an age- and sex-matched group of healthy subjects. None of the patients or control subjects was using drugs known to cause QT prolongation.


A complete physical examination was conducted. ECGs were recorded in the resting supine position with a 12-lead MAC5500 device (GE Medical System, Milwaukee, WI). Electrodes were placed in anatomical positions according to standard procedure.

ECGs made with technical errors or missing leads and ECGs of inadequate quality were excluded from the analysis. Printouts were scanned at a resolution of 3200 dpi and digitalized with the Canoscan 4200F (Canon Inc., Tokyo). QT was measured from all leads with custom-written software validated for accuracy and consistency. Uniform criteria were used throughout [14]. QTc was calculated with the Bazett formula. A long QT was defined as a QTc >440 ms for a child younger than 15 years, >430 ms for an adult male, and >450 ms for an adult female [14]. QTd was computed from one randomly selected beat in steady state by subtracting the minimal QT interval from the maximal QT interval. QTc dispersion (QTcd) was calculated in a similar manner. Dispersion was evaluated separately for 12 leads and for the precordial leads (V1–V6). The QTd ratio (QTdR) was calculated by correcting QTd for the corresponding RR interval.

LP measurements were conducted using the MAC5500 device and LP software (GE Medical System). The subject’s skin was cleaned with alcohol prior to electrode placement to decrease the noise level to <0.7 μV. Leads were positioned according to the Frank corrected orthogonal lead system, representing the X, Y, and Z bipolar axes. At least 200 consecutive beats were digitally recorded and averaged. A computer-based algorithm was used to calculate the following parameters: (1) duration of filtered QRS complex (fQRS); (2) root-mean-square voltage of the terminal 40 ms (RMS40); and (3) time during which the low-amplitude QRS signal (LAS) remained below 40 μV. LP measurements were considered abnormal when two of the following three criteria were met: (1) fQRS duration >114 ms, (2) RMS40 <20 μV, and (3) LAS duration >38 ms [7].

Statistical Analysis

Data were analyzed with Microsoft Excel version 2003 (Microsoft Corp., Seattle, WA, USA) and JMP version 7.0 (SAS Institute, Cary, NC, USA). Results are presented as medians, means, and standard deviations. Abnormal results were defined as more than two standard deviations from the normal range. Findings were compared between the groups by nonparametric Mann–Whitney–Wilcoxon test (normal approximation). A P-value <0.05 was considered statistically significant.


The study group consisted of seven male and six female patients aged 9–46 years (mean, 24.0 ± 10.9 years). Average age at diagnosis was 2.5 years.

Mean QTc in the FD group was 428.66 ± 40.16 ms, similar to that in the control group. Mean QT interval in both the FD and the control group were within normal range. Despite the overall normal average QT and QTc intervals, a prolonged QTc was noted in four patients with FD (30.7%; 451, 477, 493, and 494 ms) compared to zero subjects in the control group; three of the patients had an excessively prolonged QTc (>470 ms).

There was a significant increase in several QTd parameters (Table 1). QTcd measured from precordial leads was 109.93 ± 37.86 ms in the FD group compared to 76.6 ± 29.02 ms in the control group (P = 0.023). Precordial QTdR was abnormally high in the FD group: 0.126 ± 0.046% compared to 0.078 ± 0.033% in the control group (P = 0.016). All other dispersion parameters were higher in the FD group, although the difference did not reach statistical significance.
Table 1

QT dispersion in 13 patients with familial dysautonomia compared to aged-matched healthy volunteers





QT, ms

371.5 ± 33.1 (362)

406.4 ± 38.28 (418)


QTc, ms

428.66 ± 40.16 (410)

419.7 ± 21.98 (427)


QTd, ms

    All leads

99.69 ± 31.75 (96)

92.88 ± 33.78 (86)



96.46 ± 33.2 (96)

75.11 ± 26.13 (80)


QTdR, %

    All leads

0.13 ± 0.045 (0.136)

0.0977 ± 0.0467 (0.088)



0.126 ± 0.046 (0.132)

0.078 ± 0.033 (0.088)


QTcd, ms

    All leads

113.72 ± 36.48 (110.83)

95.02 ± 39.54 (88.02)



109.93 ± 37.86 (106.4)

76.6 ± 29.02 (88.02)


FD familial dysautonomia, QTd QT interval dispersion, QTdR QTd ratio corrected for RR interval, QTcd QT corrected dispersion

Values are mean ± SD. Median values are given in parentheses. Measurements were calculated separately for 12 leads and for precordial leads (V1–V6)

None of the ECGs in the FD group were excluded because of high noise level. LP test parameters were normal in all patients with FD (Table 2).
Table 2

Signal-averaged ECG in 13 patients with familial dysautomonia

Patient no.

Age (year)


fQRS duration (ms)

RMS40 (μV)

LAS duration <40 μV (ms)















































































fQRS filtered QRS complex, RMS40 root-mean-square voltage of the terminal 40 ms, LAS low amplitude signal

The late potential test was considered abnormal when two of three criteria were met: (1) Fqrs >114 ms; (2) RMS40 <20 μV; and (3) LAS >38 ms. None of the patients with FD had abnormal late potential results


Patients with FD have a shorter-than-normal life expectancy and a substantial proportion of cases of sudden death remain unexplained [4]. We recently reported a high rate of left ventricular hypertrophy and remodeling in patients with FD (manuscript submitted for publication). Left ventricular hypertrophy has long been recognized as a strong independent risk factor for ventricular arrhythmias and sudden death [2022]. These findings prompted the present evaluation of the arrhythmogenic substrate to ventricular tachyarrhythmias using two noninvasive ECG-based techniques.

The finding of a prolonged QTc interval in 30.7% of the patients with FD was consistent with previous studies in the medical literature [10, 11, 16]. A prolonged QTc interval in some patients can be explained in part by a decrease in autonomic activity, similar to the effect of pharmacological blockade in normal subjects [8]. During the follow-up period of this study, no lethal arrhythmia was documented. Therefore, no obvious link between prolonged QT intervals and sudden death was noted in this patient group. Further investigation into the potential role of prolonged QT parameters in sudden death is warranted. In addition, in the present study, several QTd parameters were significantly abnormal in the FD group. Our results are consistent with those of the only previous study of QTd and QTcd in patients with FD [10]. However, the previous study did not investigate QTdR and did not calculate precordial dispersion. The QTdR is known to be a more significant predictive factor for arrhythmogenicity than the QTd in some patients [15]. Thus, our finding of an abnormally high precordial QTdR supports the notion that patients with FD are at higher risk of ventricular arrhythmias than healthy age-matched subjects. Moreover, since QT dispersion of the precordial leads is believed to be a better index than QTd measured from all 12 leads [23], the finding of abnormal precordial repolarization parameters further supports on arrhythmogenic risk in FD.

Findings for the LP test in our patients with FD were unremarkable. The LP test has been widely used in survivors of myocardial infarction [7], and it is believed to have a high negative predictive value for future sudden death [17]. Nevertheless, a negative LP test may suggest that arrhythmias are induced by a mechanism other than delayed depolarization or re-entry [7]. Given the present findings, we speculate that it is abnormal repolarization rather than abnormal depolarization that increases the risk of sudden death in patients with FD. Therefore the SAECG is of limited use as a screening test for sudden death risk quantification in FD.

Study Limitations

First, the resting ECGs and SAECGs were performed in the daytime, although the majority of sudden deaths (68%) occur at night [4]. Using the present design, we could not predict whether the results would have been different in other settings. Second, the study group was small because of the low prevalence of FD. Future prospective studies with larger study groups and prolonged follow-up may shed further light on the significance of proarrhythmogenic markers in FD.


We thank Dr. Hillary Voet of the Hebrew University of Jerusalem, Faculty of Agricultural Food and Environmental Quality Sciences, for her assistance with the statistical analysis. Also, we thank Gloria Ginzach and Phyllis Curchack Kornspan for their editorial assistance. This study is dedicated to the memory of Haim Giron.

Copyright information

© Springer Science+Business Media, LLC 2009