Author + information
- Received September 11, 1998
- Revision received March 25, 1999
- Accepted May 10, 1999
- Published online September 1, 1999.
- Heikki Swan, MDa,* (, )
- Matti Viitasalo, MDa,
- Kirsi Piippo, PhDa,
- Päivi Laitinen, PhDa,
- Kimmo Kontula, MDa and
- Lauri Toivonen, MDa
- ↵*Reprint requests and correspondence: Dr. Heikki Swan, Division of Cardiology, Department of Medicine, Helsinki University Hospital, Haartmaninkatu 4, FIN-00290 Helsinki, Finland
This study was performed to evaluate the QT interval and heart rate responses to exercise and recovery in gene and mutation type-specific subgroups of long QT syndrome (LQTS) patients.
Reduced heart rate and repolarization abnormalities are encountered among long QT syndrome (LQTS) patients. The most common types of LQTS are LQT1 and LQT2.
An exercise stress test was performed in 23 patients with a pore region mutation and in 22 patients with a C-terminal end mutation of the cardiac potassium channel gene causing LQT1 type of long QT syndrome (KVLQT1 gene), as well as in 20 patients with mutations of the cardiac potassium channel gene causing LQT2 type of long QT syndrome (HERG gene) and in 33 healthy relatives. The QT intervals were measured on electrocardiograms at rest and during and after exercise. QT intervals were compared at similar heart rates, and rate adaptation of QT was studied as QT/heart rate slopes.
In contrast to the LQT2 patients, achieved maximum heart rate was decreased in both LQT1 patient groups, being only 76 ± 5% of predicted in patients with pore region mutation of KvLQT1. The QT/heart rate slopes were significantly steeper in LQT2 patients than in controls during exercise. During recovery, the QT/heart rate slopes were steeper in all LQTS groups than in controls, signifying that QT intervals lengthened excessively when heart rate decreased. At heart rates of 110 or 100 beats/min during recovery, all LQT1 patients and 89% of LQT2 patients had QT intervals longer than any of the controls.
LQT1 is associated with diminished chronotropic response and exaggerated prolongation of QT interval after exercise. LQT2 patients differ from LQT1 patients by having marked QT interval shortening and normal heart rate response to exercise. Observing QT duration during recovery enhances the clinical diagnosis of these LQTS types.
The clinical presentation and QT interval duration is variable in different types of long QT syndrome (LQTS). Even within a genetically homogenous population with a specific mutation of cardiac potassium channel gene causing LQT1 type of long QT (KvLQT1), the spectrum of QT interval durations is wide (1). Observation of the distribution of corrected (QTc) intervals of LQTS patients with molecularly established diagnosis has indicated that the use of QT intervals measured in standard resting electrocardiogram (ECG) as diagnostic criterion may misclassify members of LQTS families (1–3). Exercise test has been suggested to enhance diagnostic accuracy in LQTS, as shortening of QT interval was shown to be inadequate in LQTS patients (4–7). Schwartz et al. (8)reported that patients with mutations of HERG (LQT2 patients) display a lesser degree of QT interval shortening than those with mutations of the SCN5A (LQT3 patients) in response to increase in heart rate. Besides the heart rate, the physiological state also has been shown to influence the duration of ventricular repolarization time in healthy subjects (9–13). In some LQTS patients, a submaximal heart rate response during exercise has been reported (7,14,15). No differences in heart rate responses to exercise between the LQTS types have been reported thus far.
In the present study, we evaluated how physical exercise influences sinus nodal rate and ventricular repolarization in three molecularly defined LQT1 and LQT2 patient cohorts. In addition, mutations in the transmembrane domains are more frequently associated with cardiac events than C-terminal missense mutations, which has been described to cause a forme fruste LQTS (16). This prompted us to compare two subsets of LQT1 patients with different types of mutations, one with a point mutation in the functionally important pore domain (3), and the other with a mutation in the C-terminal end of the KvLQT1 gene (17), in terms of QT interval behavior and sinus rate during exercise.
The first patient group consisted of 23 subjects with a mutation (Asp317Asn) in the pore domain of the KvLQT1 gene (3)(LQT1 pore region group). Seven of these patients were symptomatic. All symptoms were associated with physical exercise. The second patient group included 22 subjects with the Gly589Asp mutation close to the C-terminus of the KvLQT1 gene (17)(LQT1 C-terminus group). Four of the patients in the LQT1 C-terminus group had experienced syncopal spell; three of these were exercise related. Thus, a syncopal spell was associated with exercise in 10 out of 11 cases in LQT1 patients.
The third patient group consisted of 20 patients from 10 families with a variety of mutations in cardiac potassium channel gene causing LQT2 type of LQTS (HERG) (Arg176Trp, Leu552Ser, Tyr569His, Gly584Ser, Gly601Ser, 453delC and 1631delAG; unpublished data) (LQT2 group). Each of these mutations resulted in translation frameshift or a substitution of a conserved amino acid, and was found to be present in affected family members but absent in controls. In every family, at least one patient had experienced syncope and had QTc ≥480 ms. Together, eight out of the nine symptomatic LQT2 patients had their symptoms at rest or during night.
Thirty-three healthy relatives were included as a control group. No beta-adrenergic blocking agents or other medications known to affect the repolarization were used by patients or control subjects during the study. The study was approved by the institutional review committee and was in accordance with the Helsinki Declaration. An informed consent was obtained from all patients and controls.
QT interval measurement and exercise protocol
Standard 12-lead electrocardiogram (ECG) was recorded with a paper speed of 50 mm/s and amplification of 0.1 mV/mm. All subjects were in sinus rhythm and none had atrioventricular or bundle branch block. Heart rate was calculated from three RR intervals. QT interval was measured in lead V3 for exercise test data because it usually has the largest T-wave amplitude (18)and in lead II for baseline characterization of subjects (Table 1). QT intervals were measured manually from the onset of the QRS complexes to the end of T wave, defined as the intersection of isoelectric line and the tangent of maximal downward limb of the T wave. A mean of two consecutive QT intervals was used in rest ECG and of four QT intervals in exercise test ECG. If the amplitude of T wave was low (<0.1 mV), the lead was excluded from the analysis. Measurements were carried out by an investigator who was unaware of the subject’s genetic classification (LQTS or control).
Exercise test was performed with bicycle ergometer with continuous ECG recording. The initial load was 30 W, followed by increments of the load by 15 W each minute until exhaustion. Thereafter, ECG recording was continued at supine position for 8 min. QT interval (QT) was measured at specified heart rates from 100 to 130 beats/min by steps of 10 beats/min. Expected maximum heart rate was calculated as follows: expected maximum heart rate = 205 − (0.5 × age in years) beats/min, which follows the guidelines reviewed by Hammond and Froelicher (19).
QT adaptation to heart rate changes
The relationship of QT to heart rate in each subject was examined by plotting QT duration against heart rate (ms/min−1, unit later omitted) and calculating the slopes by least squares linear regression analysis in each individual. Slope relating QT to heart rate (QT/heart rate slope) was accepted if correlation coefficient was >0.70 (97% of all slopes met this criterion). The mean correlation coefficient in all study subjects was 0.96 ± 0.06. QT/heart rate slopes were used because the relationship between QT intervals and heart rate appears to be linear, whereas the relationship between QT and RR intervals is curvilinear (10).
Comparisons between phases were performed with Wilcoxon signed-rank test, paired ttest, and between groups by Mann-Whitney Uor Kruskal-Wallis test when appropriate. For dichotomous variables, chi-square test was used. Correlation between continuous variables was studied with Pearson’s correlation coefficient. Statistical analyses were carried out using the SPSS 7.5.1. statistical software package (SPSS Inc., Chicago, Illinois). Data are expressed as mean ± SD. A pvalue <0.05 was considered to signify a statistical significance.
Demographic characteristics of the patients and controls are summarized in Table 1. As expected, the mean QT and QTcintervals at rest were prolonged in all patient groups; no differences were found between patient groups. Resting heart rates were similar in controls and all patient groups (Table 1). QT and QTcintervals did not differ significantly between genders in any group (data not shown). A total of 12 (18%) of all 65 LQTS patients had QTcvalues less than the longest QTcduration in controls (447 ms). Conversely, 19 (58%) of the 33 controls had QTcvalues longer than the shortest QTcobserved in LQTS patients.
Maximal heart rate
Maximal heart rate was 140 ± 13 in the LQT1 pore region group and 161 ± 7 in the LQT1 C-terminus group. In the LQT2 group, maximal heart rate was 187 ± 14, and in controls, 181 ± 13 beats/min (p < 0.001 between all groups except LQT2 and controls). These were 76 ± 5%, 86 ± 4%, 99 ± 6% and 96 ± 7% of the expected age-related maximal heart rate in respective groups (p < 0.001 between groups except LQT2 patients and controls). No difference was found in the achieved heart rate between symptomatic and asymptomatic LQT1 patients (LQT1 pore region and C-terminus groups combined). The maximal load achieved was 236 ± 46, 248 ± 46, 187 ± 48, and 237 ± 46 W for men, and 135 ± 40, 149 ± 31, 161 ± 17 and 152 ± 32 W for women of each group, respectively (p = NS). LQT1 patients and controls were void of arrhythmias during the exercise test, but two of the LQT2 patients exhibited frequent ventricular premature complexes during exercise.
Combined, 56% of the LQT1 patients, but only 3% of the controls, failed to reach the provisional limit of 85% of the expected maximal heart rate, whereas all the LQT2 patients exceeded this limit. Impairment in heart rate was related to QT interval duration, examined as correlation between achieved/expected heart rate ratio and QT interval (Fig. 1). The predictive value (r2) was 28% during exercise (r= 0.53) and 30% (r= 0.55) during recovery at heart rate 130 beats/min.
Rate adaptation of QT interval
QT intervals were significantly longer in both LQT1 groups and LQT2 group than in the control group at all heart rates during exercise and recovery (Fig. 2, Table 2). Comparison between mutation types showed that QT intervals among LQT1 patients were longer in patients with the mutation in the pore region than in those with the mutation in the C-terminal region of KvLQT1 gene throughout the exercise phase, except at the lowest heart rate studied (100 beats/min). LQT2 patients had QT interval significantly longer than that of LQT1 C-terminus patients at the lowest heart rate, but thereafter exhibited shortening to values less than in either of the LQT1 groups. During recovery phase, LQT1 patient groups did not differ from each other but both showed significantly longer QT intervals than LQT2 patients (Fig. 2).
During exercise, the heart rate adaptation of QT interval expressed as QT/heart rate slopes was significantly steeper in the LQT2 group than in the LQT1 C-terminus group or controls (Table 3). The QT/heart rate slopes did not differ significantly between LQT1 groups and controls during exercise. In contrast to this, during the recovery phase, QT/heart rate slopes were steeper in all LQTS patient groups than in controls (Table 3).
In comparison between test phases, LQT1 patients showed steeper slopes during recovery than during exercise, whereas the slopes during the corresponding phases did not differ in LQT2 patients and in controls (Table 3). The QT/heart rate slopes did not differ between symptomatic and asymptomatic LQT1 or LQT2 patients during exercise (data not shown).
Comparison of the QT intervals between exercise and recovery phases
In the control group, the QT interval was at all heart rates significantly shorter during recovery than during exercise (Table 2). In LQT1 patient groups, QT intervals were equal to or longer than during exercise at heart rates of 100 and 110 beats/min (Table 2). In LQT2 group, QT intervals at the lowest examined heart rate were shorter during recovery than during exercise. At higher heart rates, there was no difference between exercise and recovery phases in LQT2 patients. During exercise, the QT intervals of the patients and the controls overlapped at all analyzed heart rates (Fig. 3). During recovery at heart rates of 110 beats/min, and 100 beats/min, LQT1 patients and controls showed no overlapping, but two (10%) LQT2 patients had QT values within the range of control subjects (Fig. 3).
Performance of QT measures in revealing a gene carrier status
We evaluated the sensitivity and specificity of QT interval duration in standard resting ECG and in ECG obtained during recovery phase of exercise stress test to correctly diagnose the carriers of the KvLQT1 and HERG mutations. First, using QT intervals obtained in standard resting ECG, adjusted for heart rate according to Bazett’s square root formula, we studied the sensitivity and specificity of QTcinterval limit of 470 ms. Second, the sensitivity and specificity of the provisional normal upper limits for QT interval obtained in recovery phase at heart rates 110 or 100/min were explored. The value observed at the lowest heart rate achieved during 8 min after cessation of exercise was studied. Unadjusted QT intervals exceeding 350 ms at heart rate 110 beats/min or 360 ms at heart rate 100 beats/min were considered abnormal (normal upper limits defined as mean QT of control group ± 2 SD).
The division of gene carriers and controls into affected and nonaffected according to the selected criteria are summarized in Table 4. For LQT1, diagnostic sensitivity and specificity of QTclimit of 470 ms in resting ECG were 67% and 100% and for LQT2 65% and 100%, respectively. Assessment by recovery phase QT interval yielded a sensitivity and specificity of 100% for LQT1 and 89% and 100% for LQT2 patients.
The major findings of this study were the differential responses of heart rates and QT intervals during exercise stress test in patients with different cardiac potassium channel defects of LQT1 and LQT2 types of long QT syndrome. Heart rate response was impaired in LQT1 patients. In addition, QT interval shortened less during exercise in LQT1 than in LQT2 patients. The impaired heart rate response correlated with QT interval duration during exercise. Furthermore, different mutations of the same potassium channel gene showed different effects on QT interval behavior and heart rate response. The longest QT intervals as well as the most impaired heart rate responses to exercise were observed in the patient group with a mutation in the pore region of cardiac potassium channel gene KvLQT1.
Heart rate response of QT interval during and after exercise
Our data suggest that QT interval shortening during heart rate increase is similar in LQT1 patients as in controls, and even faster in LQT2 patients. This is in contrast to some previous findings obtained from studies on molecularly undefined LQTS populations in which inadequate shortening or even lengthening of the Bazett’s QTc interval during exercise has been observed (4–7,20,21). Vincent et al. (7)showed that the absolute duration QT interval in LQTS patients shortened as long as their heart rate increased. Our results in LQT1 patients are thus in accordance with the results of Vincent et al. (7), whose study pedigree was later demonstrated to represent LQT1 type of LQTS (Vincent GM, personal communication, 1998).
Regardless of the normal extent of QT shortening, most LQT1 patients have throughout the exercise abnormally long ventricular repolarization time, and thus would remain at risk of arrhythmia. However, the more rapid QT shortening with heart rate increase in LQT2 patients suggests that the most aberrant repolarization in these patients is encountered at rest. This may have connections to the more frequent symptoms of LQT1 patients during effort and LQT2 patients during night or at arousal.
Our results demonstrate that the difference in QT intervals between LQT1 patients and healthy subjects usually attains its maximum after physical effort. The same has also been observed in children with LQTS of unknown genotypes, most of which were likely to represent LQT1 and LQT2 types of LQTS (22,23).
The provisional upper limit for normal QT interval during recovery phase seems to offer better discrimination between LQTS patients and healthy relatives than resting ECG. Although the control group consisted of noncarrier relatives, their QTc intervals were normal, and in a prior study, the distribution of QTc intervals in noncarrier relatives and in unrelated healthy control subjects was shown to be similar (2). It must be noted, however, that the number of control subjects was relatively small and that they were void of structural heart diseases and medications, factors that may cause false-positive diagnoses.
Sinus node function
We observed that sinus rate response was decreased in LQT1 patients whose QT interval shortened less than in LQT2 patients and who had normal heart rate response. The present results may also explain why studies focusing on the maximal heart rate during exercise have yielded controversial results in LQTS. Despite this phenomenon being mentioned in reports reviewed by Schwartz et al. (14)in 1975, its consistency was questioned 10 years later (24). Eggeling et al. (25)reported no difference in the maximal heart rate between 14 LQTS patients and healthy controls during exercise test. In the series of Kugler (15), six out of 14 LQTS patients had abnormally low maximum heart rate during exercise. Genetic heterogeneity may explain the variability of these findings. Recently, Schott et al. (26)reported sinus bradycardia and unexpectedly low maximum heart rate during exercise in a few of the patients with LQTS type 4; the underlying molecular defect in this form of LQTS remains to be explored.
The diminished heart rate response of LQT1 patients to exertion could be due to decreased sympathetic stimulation, impaired sinus nodal response to normal autonomic stimuli or lower intrinsic sinus rate. Because the degree of sinus rate impairment was associated with the QT interval duration, this relationship makes it tempting to speculate that the ion channel defect present in ventricular myocytes also prolongs the action potential duration in sinus pacemaker cells. The consequent delay in spontaneous sinus nodal excitation would then limit the heart rate increase.
Importance of mutation site
Donger et al. (16)have shown that a mutation in the C-terminal end of the KvLQT1 gene is relatively infrequently associated with cardiac events. Despite the fact that QT intervals in standard resting ECGs do not necessarily differ between patients with mutations in pore region and in C-terminal end of KvLQT1 gene, as shown in the present study, more severe disturbance of repolarization could be demonstrated in patients with mutations in the pore region of KvLQT1 gene by exercise stress test. Furthermore, during recovery, C-terminal and pore region mutations had equally prolonged QT intervals. This, together with considerable proportion of symptomatic carriers, suggests that also C-terminal mutations of KvLQT1 cause the clinical disease.
Conclusions and clinical implications
The most common form of LQTS, LQT1, is characterized by two important features revealed by exercise stress test: an inadequate sinus rate response to exercise and an exaggeration of the QT interval prolongation after physical effort. These observations seem to apply to different types of KvLQT1 gene mutations to dissimilar extent. In contrast, in LQT2, the QT interval shortens more than in LQT1 when heart rate increases and the sinus nodal rate response is normal. The postexercise QT interval prolongation may enhance the diagnostic accuracy in potential type 1 and 2 LQTS carriers when QT interval measurements in standard resting ECGs are equivocal. The clinical utility of these features, however, needs to be verified in population studies.
☆ This study was supported by a grant from the Finnish Foundation for Cardiovascular Research, Helsinki, Finland.
- electrocardiogram, electrocardiographic
- cardiac potassium channel gene causing LQT2 type of long QT syndrome
- cardiac potassium channel gene causing LQT1 type of long QT syndrome
- long QT syndrome
- LQT1 type of long QT syndrome
- LQT2 type of long QT syndrome
- QT interval corrected with the square root formula (ms) (QT/RR1/2)
- Received September 11, 1998.
- Revision received March 25, 1999.
- Accepted May 10, 1999.
- American College of Cardiology
- Swan H,
- Saarinen K,
- Kontula K,
- et al.
- Garza L.A,
- Vick R.L,
- Nora J.J,
- McNamara D.G
- Schwartz P.J,
- Priori S.G,
- Locati E.H,
- et al.
- Viitasalo M,
- Rovamo L,
- Toivonen L,
- et al.
- Bexton R.S,
- Vallin H.O,
- Camm A.J
- Molnar J,
- Weiss J,
- Ehlert F,
- Rosenthal J.E
- Donger C,
- Denjoy I,
- Berthet M,
- et al.
- Saarinen K,
- Swan H,
- Toivonen L,
- et al.
- Lisker S.A,
- Finkelstein D
- Weintraub R.G,
- Gow R.M,
- Wilkinson J.L
- Swan H,
- Toivonen L,
- Viitasalo M
- Schwartz P.J
- Bazett H.C