Main

Hereditary arrhythmia syndromes in children include FHC and congenital LQTS. Recent molecular biologic advances have led to identification of multiple specific causative genes in both of these heterogenous syndromes. Genetically engineered mouse models can be used to study the electrophysiologic phenotype of specific mutations. FHC is an inherited disease caused by specific mutations in the genes encoding the sarcomeric structural components, such as the β-myosin heavy-chain and tropomyosin proteins. Patients affected by FHC are at risk for ventricular arrhythmias and sudden death. This may occur as early as childhood or adolescence. Occasionally, sudden cardiac death is the first clinical manifestation of the disease. Some affected families have a more malignant clinical course than others, suggesting that specific genetic mutations have phenotypic significance(1,2). The discovery of a reliable marker for patients at particularly high risk for life-threatening arrhythmias may be of clinical importance by allowing early medical intervention and preventive therapy(3).

Regional variability in the QT interval (QTd) on surface ECG has been shown to be significantly increased in humans with hypertrophic cardiomyopathy who experience ventricular arrhythmias(4,5). Gender-related differences in QT intervals have also been demonstrated in mice(6). In this study, a genetically engineered murine model of FHC was used to study the electrophysiologic characteristics of the disease. The defect studied is a missense mutation Arg403Gln in the cardiac α-myosin heavy-chain gene. This is highly homologous to the β-isoform expressed in the myocardium of some affected families and has been shown in humans to be a malignant mutation with early mortality compared with other FHC mutations(1,7). We have previously shown that these mice have inducible ventricular arrhythmias during electrophysiologic testing. The purpose of this study was to evaluate the feasibility and significance of QTd measurements in assessing the risk of ventricular arrhythmias in FHC mice.

METHODS

Surface resting ECGs and studies were performed on 32 FHC mice (19 males, 13 females) and 39 wild-type littermate controls (22 males, 17 females). The protocol for the in vivo mouse electrophysiology study has been previously described(6,8,9).

Animal care. In compliance with the American Association for the Accreditation of Laboratory Animal Care and after approval of the Institutional Animal Care and Use Committee, the animals were housed four per cage in a facility with 12-h light/dark cycles. They were allowed free access to food and water.

Preprocedural preparation. Each animal was anesthetized with a combination of ketamine and pentobarbital administered intraperitoneally (0.033 mg/kg). Surface ECGs were obtained using 27-gauge s.c. needle electrodes in each limb. Heart rate and rhythm were monitored continuously throughout the procedure.

Operative procedure. The operative procedure has been fully described elsewhere(6,8,9). Briefly, for intracardiac studies, a cutdown of the internal jugular vein was performed under an operating microscope. A pacing eletrode was placed in the right atrium and ventricle using the electrogram tracing to guide placement. For epicardial studies, pacing electrodes were attached to the right atrium and right and left ventricles after thoracotomy and incision of the pericardial sac(6,8,9).

Electrophysiology study. Standard electrophysiologic protocols for pacing and extrastimulus testing were followed to assess baseline conduction parameters and arrhythmia inducibility(6,8,9). As in humans, if programmed stimulation failed to cause an arrhythmia, isoproterenol was given (1 ng/g) and the pacing protocol repeated to assess the effects of an increased catecholamine state on cardiac conduction and arrhythmia inducibility. The dose was titrated to effect, aiming for an increase of 25% above the resting heart rate.

Measurements. The QRS, RR, and QT intervals were measured in six surface limb ECG leads by two independent observers blinded to the animals' genders and genotypes. The QT interval was measured from the beginning of the QRS complex to the end of the T wave. If the T-wave offset was not easily seen, the tangent of the slope to the isoelectric baseline was used to determine the T-wave endpoint. The QTd was defined as the difference between the maximum and minimum QT interval in any of the frontal plane ECG leads. The mean of the QTd measured by each observer yielded a QTd value for each mouse. A QT interval corrected for heart rate was also calculated using the formula proposed by Mitchell et al.(10): QTc = QTo/(RR/100)e1/2. This was used instead of Bazett's formula because it is meant for higher heart rates, as found in mice. Dispersion of the QTc values was then calculated as described for the uncorrected QT intervals.

Statistical analysis. Data were analyzed by a two-sample Wilcoxon rank sum test for nonparametric data using a computerized statistical analysis program (STATA). Statistical significance was defined as a p value <0.05.

RESULTS

Seventy-one mice completed full ECG and electrophysiology testing. Sixteen mice were excluded (7 wild-type, 9 FHC) because of either death before study completion or uninterpretable surface ECG data. Possible causes of death include an adverse reaction to anesthesia, cardiac perforation, cardiac arrhythmia, bleeding, and air embolism. The FHC group as a whole had greater QTd than the wild-type group, although this did not achieve statistical significance (p = 0.1). When analyzed by gender within each genotype, however, female wild-type mice had a significantly increased QT interval (p = 0.005), QTc interval (p = 0.04), and QTd (p = 0.005) compared with male wild-types, and male FHC mice had a significantly increased QTd compared with female FHC mice (p = 0.02) (Figs. 1 and 2). Male FHC mice also had significantly increased QT intervals (p = 0.02) and QTd (p = 0.001) when compared with male wild-type mice. In addition, 12 of 23 FHC mice had inducible ventricular arrhythmias, whereas only two of 32 wild-type mice were inducible (p = 0.004). There was no statistically significant difference in QTd between FHC mice that developed arrhythmias and those that did not, even when subgrouped by gender (Figs. 3 and 4). Male FHC mice with inducible arrhythmias had greater QTd than female FHC mice with arrhythmias (p = 0.05). There was no difference in the incidence of arrhythmias between male and female mutants, although the small number of mice in each subgroup may be too small to demonstrate such a trend. In no instance was analysis of the QTd statistically significant when the corrected QTd was not (Table 1).

Figure 1
figure 1

Mean QT interval in milliseconds ± SEM for male and female wild-type (wt) and familial hypertrophy (403/+) subgroups of mice.

Figure 2
figure 2

Mean QT interval dispersion values (QTd) in ms ± SEM compared with the mean for male and female wild-type (wt) and FMC (403/+) subgroups of mice.

Figure 3
figure 3

QT interval in milliseconds plotted against QT interval dispersion value (QTd) for each study mouse. ×, mice with no inducible ventricular arrhythmia during electrophysiology study with standard programmed stimulation. •, mice with inducible ventricular arrhythmias. VT, ventricular tachycardia.

Figure 4
figure 4

QT interval dispersion values (QTd) in milliseconds for each mouse by gender and genotype. ×, mice with no inducible arrhythmia during programmed stimulation. •, mice with inducible ventricular arrhythmias. wt, wild-type; 403/+, FHC mutation; VT, ventricular tachycardia.

Table 1 Electrophysiologic interval values

DISCUSSION

Patients with FHC are at risk for ventricular arrhythmias and early sudden death. Variability in the QT interval on surface ECG, or interlead QTd, is a marker of inhomogeneous recovery of myocardial excitability, which suggests electrical instability. This has been shown to correlate with the incidence of ventricular arrhythmias in humans affected with FHC(4,5,11,12). In 1993, Buja et al.(4) reported such findings after studying 26 patients with hypertrophic cardiomyopathy-13 who had experienced ventricular arrhythmias and 13 who had not. A significantly increased QT, QTc, and QTd were found in patients with a history of arrhythmia versus those without arrhythmias. In addition, hypertrophic cardiomyopathy patients without arrhythmias had increased QT intervals and QTd when compared with normal subjects. They asserted that QTd on surface ECG may be used as a marker to identify patients with the disease who are at higher risk for ventricular arrhythmias and, therefore, sudden death(4). QTd is also increased in patients with LQTS(13), those with ischemic heart disease(14,15), and those taking class Ia antiarrhythmic medications(16), all of whom are at increased risk for ventricular arrhythmias. Linker et al.(17) investigated QTd in patients with congenital QT prolongation as a predictor of arrhythmia risk. They found that although patients with LQTS had longer QT and QTc intervals and greater QTd than controls, this was not predictive of the severity of symptoms.

A recent study of children with congenital LQTS showed that their QTd and JT dispersion were both significantly prolonged, and correlated with those of patients who later developed ventricular tachyarrhythmias or sudden death(18). These human clinical studies on familial childhood arrhythmia substrates indicate that dispersion of repolarization ECG markers may be valuable for risk-stratification purposes.

The current study demonstrates that measurement of QTd on the mouse ECG is feasible. In this set of experiments, male FHC mice have greater QTd than wild-type controls, with a significant gender difference in QTd within each genotype group; female control mice had greater QTd than male controls, and male FHC mice had greater QTd than their female counterparts. This gender difference is consistent with data from Berul et al.(6,8), who found that male FHC mice had significantly more electrophysiologic abnormalities and greater histopathologic disease progression than female FHC mice. Males had right-axis deviation, sinus node dysfunction, prolonged ventricular repolarization, and an increased incidence of inducible ventricular arrhythmias, findings strikingly similar to those in humans with FHC. It is conceivable that, in the setting of myocyte disarray and intercellular fibrosis, there is increased regional variability in repolarization in mice with more significant histopathology. This would account for the greater QTd in male mice in the current study.

Berul et al.(6) also found that female wild-type mice had slower heart rates and longer QT and JT intervals compared with male wild-type mice and found no significant difference in surface ECG parameters between female wild-type and mutant mice. They proposed potential modulation by sex hormones or receptors to account for the gender-specific differences. In accordance with this, the current study found longer repolarization times in female wild-type mice compared with male wild-type mice. It is difficult to account for the fact that wild-type females had greater QTd than mutant females. Possible explanations include 1) modulation of repolarization heterogeneity by female hormones such as estrogens, 2) hormonal influences on mutation-specific ECG abnormalities, or 3) gender-specific differences in ECG interpretability, such as flattening of T waves in female mice. We found no difference in QTd between mice carrying the Arg403Gln mutation that had inducible ventricular arrhythmias and Arg403Gln mice that were noninducible during the electrophysiology study (Figs. 5 and 6). The failure to show a significant difference in QTd between mice with and without inducible arrhythmias may be due to the lack of statistical power secondary to the relatively small number of mice in each group. Other possibilities include the inherent imprecision in measuring QTd, such as where the precise termination of the T wave is, and which surface leads yield the most valid QTd calculation. Finally, the particular FHC genotype may give rise to a phenotype that is prone to ventricular arrhythmias and histopathology that are unrelated to repolarization heterogeneity as measured by QTd.

Figure 5
figure 5

Lead I surface ECG during sinus rhythm in a male FHC mouse with inducible ventricular tachycardia during programmed stimulation from the right ventricle.

Figure 6
figure 6

Ventricular tachycardia, cycle length 55 ms, in a male FHC mouse (surface ECG pictured above) induced with burst ventricular pacing at cycle length 150 ms. The sinus cycle length is 160 ms.

The importance of this study is that it not only validates this mouse model for further investigation of cardiac conduction abnormalities and arrhythmogeneity in familial hypertrophic cardiomyopathy, it also suggests that gender in addition to genotype influences phenotype in this important human disease. This study did not, however, demonstrate that QTd measurements are predictive for the development of ventricular arrhythmias. Because QTd has been shown to be a useful predictor of arrhythmias in humans, and there are many mutations that cause FHC, it is possible that surface ECG measurements are more predictive in families with certain mutations and not others. Alternatively, the extrapolation of a mouse model of human electrophysiologic disorders may limit the power of risk stratification. Also, the small size of the murine ventricle may make assessment of repolarization electrical heterogeneity by interlead QTd on ECG more challenging. Finally, the disparate directions of QTd abnormalities between genders in mutant and wild-type mice suggest a modifying factor, such as a sex-chromosome modifier gene, or hormonally mediated modulation of repolarization. Further studies will be needed to evaluate the reasons for gender differences, such as a possible protective role of female reproductive hormones, the effects of oophorectomy, hormone replacement, estrogen-receptor blockade, and assessment of mice with both mutations of estrogen-receptor gene and FHC. These manipulations can be performed to study the gender-specific electrophysiologic attributes of mice affected with FHC, in an effort to identify a predictor of those at greatest risk for arrhythmias and further improve extrapolation to allow for better risk stratification in humans.