Prolonged QT Interval in Athletes: Distinguishing between Pathology and Physiology

Chronic exercise training in athletes has been associated with longer QT interval, whereas detraining has been found to induce a normalization of electrocardiogram (ECG) with reappearance of repolarization abnormalities after resumption of exercise training [1‒3]. QT prolongation in athletes has been linked with decreased QT dispersion and thus reduced arrhythmic potential, implying the benign nature of this physiological adaptation to chronic exercise training [1]. On the contrary, QT prolongation in individuals with congenital long QT syndrome (LQTS) is associated with increased incidence of sudden cardiac death (SCD) [4]. Therefore, identification of prolonged QT should prompt further diagnostic evaluation to rule out LQTS. In this respect, the physiological QT prolongation in athletes is expected to widen the gray zone between physiology and pathology of QT, increasing the diagnostic challenges encountered in athletes with QT prolongation [2].

Apart from corrected QT (QTc) ≥500 ms, diagnostic challenges arise in borderline cases of QTc prolongation, where further clinical investigations are needed to be performed to clarify whether LQTS exists [4, 5]. Clinical diagnostic investigations are more readily available, convenient, and easily interpretable, as well as less costly than genetic testing for LQTS. Exercise testing has an important role in the clinical evaluation of individuals with QT prolongation. Thus, this article presents the diagnostic strategy that is recommended to be employed in athletes with prolonged QT interval. Moreover, we aimed to explore the usefulness of exercise testing in the diagnostic evaluation of individuals with suspected LQTS.

According to international recommendations for ECG in athletes, further evaluation for LQTS is indicated in male athletes with QTc ≥470 ms and female athletes with QTc ≥480 ms [6]. Figure 1 shows an algorithm for the diagnostic evaluation of athletes with suspected LQTS. When QTc ≥500 ms, there is unequivocal diagnosis of LQTS, whereas for borderline cases of QTc prolongation (i.e., 470 ms ≤ QTc <500 ms for males, 480 ms ≤ QTc <500 ms for females) further clinical investigations are needed to be performed to clarify whether LQTS exists [6, 7]. In cases of athletes with borderline QTc prolongation, personal and family history should be evaluated and, if they are not suggestive of LQTS, a second ECG should be performed on a different day to confirm QTc prolongation. When the personal or family history is positive, further evaluation of the athlete is needed, including calculation of Schwartz score for LQTS, exercise testing, and ambulatory ECG monitoring, along with screening of the athlete’s relatives using resting ECG [6, 8]. Specifically, in case of Schwartz score ≥3.5, there is high probability for LQTS [8]. With regard to exercise testing, paradoxical prolongation of QT with increasing HR during exercise and QTc ≥480 ms at the fourth min of recovery suggest the presence of LQTS [8‒10]. Detection of excessive prolongation of QTc (i.e., ≥500 ms), T wave abnormalities, or polymorphic ventricular tachycardia (VT) on 12-lead ECG monitoring is indicative of LQTS [11]. Identification of prolonged QTc in a relative is suggestive of the presence of LQTS in this family. If a least one of these investigations is positive in an athlete, the diagnosis of LQTS can be reasonably confirmed. When all these investigations are negative, genetic testing for the three major LQTS-associated genes of KCNQ1, KCNH2, and SCN5A can be performed [12]. A pathogenic or likely pathogenic mutation for these genes confirms diagnosis of LQTS type 1, 2, and 3, respectively, that collectively comprise 75% of LQTS etiology [4, 5, 12].

With regard to personal history, an athlete with prolonged QTc should be asked whether there is history of syncope, pre-syncope, palpitations, or epileptic seizures [6]. Syncopal episodes in an individual with LQTS may give rise to epileptic seizures, due to cerebral ischemia, making the false diagnosis of epilepsy. Questions about family history should include any occurrence of the following in a relative: unexplained SCD at young age, one-sided motor vehicle accident, or drowning.

An important requirement for correct identification of prolonged QTc in an athlete is accurate measurement of QT and correction for HR [6]. Specifically, a prolonged computer-derived QT should be confirmed with manual measurement in leads II and V5. Measurement of QT interval can be performed with two methods: the “tangent method” and the “threshold method” with the latter providing a mean of 10 ms greater QT values [13]. The end of QT by the “tangent method” is defined as the point where the tangent on the steepest part of the descending limb of T wave intersects with the isoelectric baseline. Regarding the “threshold method,” the end of QT is defined as the intersection of descending limb of T wave with the isoelectric baseline. According to international recommendations for ECG in athletes, the “tangent method” should be preferred for QT measurement in athletes [6]. QT correction for HR is recommended to be performed with the use of Bazett’s formula, in the HR range of 50–90 bpm, since significant underestimation or overestimation can result in lower or higher HR, respectively [6]. If HR <50 bpm, calculation of QTc can be performed after mild physical activity to increase HR appropriately, whereas in case of HR >90 bpm additional rest will allow a correct calculation of QTc after a drop of HR below 90 bpm. When sinus arrhythmia is present, average QTc should be calculated from an ECG rhythm strip [6]. The most widely applied form of Bazett’s formula QTc = QT/√RR represents a simplified version of the original and most correct form of QTc = QT/√(RR/1 s), corresponding to the QTc interval at a HR of 60 beats/min when RR interval is expressed in sec [14]. Using dimensional analysis only the latter form is mathematically correct determining the units for QTc to be the same as for QT, as opposed to the most popular form of Bazett’s formula, in which the QTc can be dimensionally expressed as [T^1/2] resulting in bizarre units devoid of any physiological rationale [14]. In clinical practice, the simplified version of Bazett’s formula yields the same numerical value for QTc as the original form provided that RR interval is measured in sec [14].

Considering the diurnal and day-to-day variation of QTc, international recommendations for ECG in athletes state that for an athlete with negative personal and family history and borderline QTc prolongation, a repeat ECG should be performed to confirm QTc prolongation [6, 15, 16]. Before evaluation of an athlete with prolonged QTc, reversible extrinsic factors should be excluded, such as electrolyte abnormalities (i.e., hypocalcaemia, hypokalemia, hypomagnesaemia) and QT-prolonging medications [17].

Dagradi et al. [2] shed light on the impact of practicing sports on QTc interval and the influence of exercise training on the diagnostic evaluation of athletes with QTc prolongation. Specifically, all athletes with QTc prolongation, including genotype-positive and genotype-negative athletes, exhibited reduction of QTc after detraining [2]. The genotype-negative (i.e., without firm diagnosis of LQTS based on genetic testing) athletes with QTc prolongation were demonstrated to consist of two subgroups with markedly different responses of QTc to detraining [2]. The subgroup with normalization of QTc and reversion of repolarization abnormalities along with conversion of Schwartz score to almost zero after detraining was considered as normal, whereas the subgroup with much less decrease in QTc and Schwartz score and persistence of repolarization abnormalities after detraining was deemed as affected by LQTS [2]. Resumption of exercise training by the athletes of the former subgroup induced reappearance of QTc prolongation and repolarization abnormalities in a subset of them [2]. High intensity of retraining was reported to be the common denominator of the individuals that experienced reappearance of the QTc prolongation and repolarization abnormalities after resumption of exercise training [2]. International recommendations for ECG in athletes were formulated without taking into account the results of Dagradi et al. [2] that suggest detraining in all genotype-negative athletes with QTc prolongation to reclassify them as affected or not by LQTS according to the attendant changes in QTc and Schwartz score. Despite a seeming inconsistency in the literature about the comparison of QTc between athletes and nonathletes, the majority of well-designed studies displayed increased QTc in athletes [1, 3, 18‒23]. Moreover, the subgroup of athletes in these studies was characterized by much lower HR compared to nonathletes with a considerable proportion of athletes in the zone of less than 50 bpm that is associated with significant underestimation of QTc, tending to decrease the significance of the difference in QTc between athletes and nonathletes. Apart from one study that showed a downregulation of QTc in power athletes abusing androgenic anabolic steroids, the relevant studies did not report any information about the use of androgenic anabolic steroids (based either on history taking or on physical signs indicative of concealed use) that can considerably reduce QTc in athletes, counterbalancing the QTc prolonging effect of exercise training [1, 24]. The impact of the type of sports on QTc has not been adequately investigated, though the studies that found upregulation of QTc in athletes included mainly athletes with high endurance component of exercise training [1, 3, 19, 23]. Although the underlying mechanisms of the QTc prolongation due to exercise training have not been elucidated yet, a plausible mechanism involves the chronic upregulation of stretch-activated sarcolemmal Ca²⁺ channels of cardiomyocytes in the context of exercise training-induced membrane deformation, leading to prolonged ventricular repolarization and thus QTc prolongation on the surface ECG [25, 26].

The 470 and 480 ms QTc cut-offs of normality for male and female athletes, respectively, that are mentioned in international recommendations for ECG in athletes represent the 99th percentile for healthy population [6, 27]. Notably, an online QTc calculator can provide an age- and sex-specific LQTS probability calculation [13]. For athletes with borderline QTc prolongation, further evaluation is needed to differentiate between pathology and physiology, whereas detection of QTc ≥500 ms on repeated measurements and in the absence of QT-prolonging conditions appears to be highly specific for LQTS, practically establishing the diagnosis of LQTS [7]. Although among individuals with genetically confirmed LQTS, QTc <470 ms can be found in 40% of males and QTc <480 ms in 20% of females, the very low prevalence (1/2,500) of LQTS results in very high negative predictive values (males: 99.98%, females: 99.99%) [28, 29]. It should be underlined that the physiological prolongation of QTc due to chronic exercise training in athletes may result in the false diagnosis of congenital LQTS based solely on resting ECG [2]. Indeed, Dagradi et al. [2] demonstrated that individuals engaging in chronic exercise training may be more common to fulfil diagnostic criteria for LQTS, whereas detraining can induce a normalization of ECG with reappearance of repolarization abnormalities after resumption of exercise training [2]. Therefore, a more comprehensive approach should be applied in athletes with QTc prolongation, including other clinical information and, if deemed appropriate, genetic testing.

T wave morphology, including notched T waves in precordial leads and T wave alternans, may suggest the presence of LQTS, irrespective of QT duration [30‒33]. The morphology of ST-T on ECG of individuals with LQTS can be genotype-specific, since type 1 is characterized by broad-based T waves, type 2 by notched T waves, whereas type 3 by a long isoelectric ST segment and a peaked T wave [31, 34]. Notably, biphasic anterior T wave inversions have been reported in individuals with LQTS [30]. These T wave inversions were found to be much more frequent in symptomatic individuals with LQTS compared to the asymptomatic ones [30]. These pathological T wave inversions encountered in individuals with LQTS usually attract most attention of an ECG reader, potentially leading the reader to overlook the QT prolongation, if a systematic approach to ECG interpretation is not applied. However, increased prevalence of T wave abnormalities has been normally reported in athletes, especially in endurance athletes, further increasing the diagnostic challenges in ECG interpretation of athletes with T wave abnormalities and QTc prolongation [23].

Findings on exercise testing that are suggestive of LQTS can be the following: paradoxical prolongation of QTc during exercise testing, QTc ≥480 ms at fourth min of recovery, polymorphic VT and T wave alternans [8, 10, 35, 36]. Caution should be applied in calculating QTc during exercise, since existing formulas become inaccurate at high HR. Thus, calculation of QTc during exercise is usually performed up to 100–120 bpm. As opposed to prolongation of QTc during exercise that may be subject to some inaccurate calculation, paradoxical prolongation of QT during exercise is far more diagnostic for LQTS, since it contradicts to the normal shortening of QT at higher HR [35]. The detection of QTc ≥480 ms at fourth min of recovery is included in Schwartz diagnostic criteria and was associated with 100% specificity for LQTS in a nonathletic population [8, 10]. This finding is useful for diagnosis of LQTS type 1 and 2 [8]. Considering that the criterion of QTc ≥480 ms at fourth min of recovery has not been validated in athletes, at present diagnosis of LQTS in athletes should not be based solely on this finding. Instead, a comprehensive diagnostic approach is preferred in athletes, taking into account the consistency of different diagnostic investigations to confirm the diagnosis of LQTS in an integrative manner. The appearance of polymorphic VT may be the least specific finding, whereas T wave alternans during exercise, though it may be uncommonly detected, is more specific for LQTS [36].

Epinephrine stress testing can be used as a diagnostic tool for LQTS, primarily type 1 [37]. Specifically, paradoxical prolongation of QT during intravenous infusion of epinephrine is indicative of LQTS type 1, whereas appearance of notched T waves suggests LQTS type 2 [37]. This test does not aid to diagnosis of LQTS type 3 [37]. Epinephrine testing is prone to false positives and is associated with suboptimal positive predictive value [37]. This test is not usually preferred for the evaluation of athletes with suspected LQTS, as it is classically reserved for individuals unable to perform exercise testing, which is not the case for healthy athletes.

The transition from supine to standing position can normally induce a mild QTc prolongation during HR acceleration, since the RR interval shortens more than the QT interval [38]. Although the exaggerated QTc prolongation upon standing of individuals with LQTS has been demonstrated to confer incremental diagnostic value over supine ECG, specific cut-offs of the QTc increase on supine-standing test have not been proposed yet [38].

Ambulatory 12-lead ECG monitoring in athletes with borderline QTc prolongation on resting ECG can potentially detect excessive prolongation of QTc (i.e., ≥500 ms) or T wave abnormalities indicative of LQTS, including occasional biphasic T waves and episodes of T wave alternans [39]. Considering that the normal QTc values relevant for the baseline resting ECG may not be pertinent to the ambulatory ECG monitoring, only detection of excessive QTc prolongation on ambulatory ECG monitoring could be considered diagnostic. According to European Society of Cardiology (ESC) guidelines, athletes with borderline QTc prolongation should be subjected to ECG monitoring for detection of subclinical arrhythmias [11]. Documentation of episodes of polymorphic VT renders diagnosis of LQTS very likely.

Both ESC and Heart Rhythm Society (HRS)/European Heart Rhythm Association (EHRA)/Asia Pacific HRS (APHRS) guidelines state that documentation of a pathogenic mutation in a LQTS gene is sufficient for diagnosis of LQTS irrespective of QT duration [4, 5]. Genetic testing for LQTS is highly indicated, since the signal/noise ratio is high (19:1), attributed to a high yield of genetic testing (80%) and low percentage of controls with rare variants of uncertain significance [12]. Genetic testing should be reserved for cases with a high probability diagnosis of LQTS and not as an attempt to clarify an uncertain diagnosis [40, 41]. Variant-specific genetic testing can be performed in relatives after the identification of the disease-causing variant [40]. Approximately 75% of individuals with a clinically definite LQTS diagnosis have mutations in one of the following three genes: KCNQ1 (LQTS type 1), KCNH2 (LQTS type 2), and SCN5A (LQTS type 3) [12]. Thus, genetic testing for LQTS classically involves these three genes. However, 15–20% of LQTS may remain genetically elusive after genetic testing of all the known genes [12].

Diagnosis of LQTS can be based on either HRS/EHRA/APHRS or ESC guidelines [4, 5]. Diagnostic criteria of both guidelines include Schwartz score ≥3.5 or pathogenic mutation in a LQTS gene. Diagnostic criteria of HRS/EHRA/APHRS guidelines also include QTc ≥500 ms or QTc = 480–499 ms with unexplained syncope, whereas the respective QTc values of ESC guidelines are 20 ms lower. Taking into account the normal prolongation of QTc in athletes, the higher cut-offs of HRS/EHRA/APHRS recommendations may be more relevant for athletes and more consistent with international recommendations for ECG in athletes.

Both guidelines mention that diagnosis of LQTS can be based on identification of pathogenic mutations for LQTS genes. The more recent American College of Medical Genetics and Genomics (ACMG) guidelines report that both pathogenic and likely pathogenic mutations can be used in clinical decision-making [42]. In current clinical practice, both pathogenic and likely pathogenic variants are classically used to set the diagnosis of LQTS, in accordance with ACMG guidelines.

If an athlete is diagnosed with LQTS, lifestyle changes are recommended by both HRS/EHRA/APHRS and ESC guidelines, including avoidance of QT-prolonging drugs, as well as early identification and correction of electrolyte abnormalities leading to QT prolongation [4, 5]. Although beta blockers are recommended in essentially all individuals with LQTS, particularly for types 1 and 2, athletes are usually not eligible for this therapy, due to the commonly encountered low HR [4, 5, 12, 43]. This consideration is more relevant for endurance athletes, who are characterized by even lower HR. Thus, tolerability of b-blocker therapy in athletes is questionable. However, the decrease in volume of exercise training of an athlete following the recommended disqualification from competitive sports may result in such an increase in HR, rendering this athlete eligible for b-blocker therapy. Left cardiac sympathetic denervation can be applied as add-on therapy in athletes with LQTS. Finally, sodium channel blockers, such as mexiletine, can have a role in LQTS type 3 [4].

Genetic testing for LQTS in an athlete can aid to exercise recommendations. According to ESC guidelines, if diagnosis of LQTS is genetically confirmed in an athlete, restriction from competitive sports will be indicated, especially for LQTS type 1, that carries the highest risk of exercise-related SCD [11, 44]. Considering that swimming appears to convey the highest risk of exercise-related SCD in individuals with LQTS, restriction of an athlete with LQTS from not only competitive, but also recreational swimming is reasonable [44]. Swimming without diving or without head under water in any other way may be associated with relatively lower arrhythmic risk, since face immersion in cold water has been found to induce QT prolongation and T wave changes in children with LQTS [45]. Even more, early defibrillation can be impractical in water sports, mandating a stricter approach for swimming [11]. In this respect, recreational exercise, apart from swimming, is permitted for an athlete with LQTS.

A more lenient approach is adopted by American Heart Association (AHA)/American College of Cardiology (ACC) guidelines, mentioning that athletes with LQTS can participate in competitive sports, provided they have been asymptomatic on treatment for at least 3 months and all the necessary precautionary measures are applied, including avoidance of electrolyte abnormalities and dehydration, availability of a personal automatic external defibrillator, and formation of an emergency action plan [46]. However, efficacy of automatic external defibrillator may not be high enough in athletes suffering cardiac arrest, even when applied early enough [47]. In support of a more liberal management of athletes with LQTS, a recent study by Tobert et al. [48] demonstrated that optimally treated athletes with LQTS experienced a low event rate of breakthrough cardiac events (i.e., <2% risk per annum) with no deaths during an average follow-up of 4 years. Specifically, breakthrough cardiac events occurred in 5.9% of athletes with LQTS with only 1.0% experiencing sports-related cardiac events, while the cardiac event rate was similar between the period of the athletic career and outside of this period [48]. In this respect, AHA/ACC guidelines appear to be more evidence-based and may be more reasonable than ESC guidelines regarding the management of athletes with LQTS.

A literature search based on PubMed listings up to November 21, 2021, using “long QT” AND (“exercise testing” OR “stress test” OR “exercise test”) as the search terms identified 265 articles (Fig. 2). Furthermore, the reference list of these articles was examined for eligible articles to be included in the review. Finally, only 17 studies were included in this review article.

1. Original research articles written in English.

2. Studies evaluating the diagnostic utility of exercise testing to discriminate between individuals with genetically confirmed LQTS and controls.

The most commonly used method for confirming diagnosis of LQTS on exercise testing is paradoxical prolongation of QT during exercise. As opposed to prolongation of QTc during exercise that may be subject to some inaccurate calculation, paradoxical prolongation of QT during exercise is far more diagnostic for LQTS, since it contradicts to the normal shortening of QT at higher HR. Although there are reports of abnormal QT prolongation at peak exercise in LQTS, the ambiguity in determining the true end of T wave at very high HR possibly challenges the accuracy of this approach [10, 49]. Even more, individuals with LQTS type 2 appear to exhibit abnormal lengthening of QT in early stages of exercise testing, rather than at peak exercise [9, 50, 51]. On the contrary, a progressive lengthening of QTc from rest to peak exercise has been found to occur in LQTS type 1 [9]. Abnormal prolongation of QTc during exercise can be detected more commonly in individuals with LQTS type 1 (96%) compared to type 2 (80%), whereas patients with LQTS type 3 appear to exhibit normal QTc shortening [35]. B-blocker therapy may not influence the abnormal QT prolongation during exercise in LQTS, obviating the need for b-blocker wash-out [9, 49, 52].

Abnormal QT prolongation during recovery of exercise testing has been found to characterize individuals with LQTS type 1 and 2, whereas QT changes during recovery appear not to differ between LQTS type 3 and controls [49, 50, 53]. Individuals with LQTS type 1 display a sustained QTc prolongation throughout the whole recovery with decreasing magnitude of QTc prolongation from early to late recovery, while for LQTS type 2 the magnitude of QTc prolongation increases from early to late recovery [52, 54]. The abovementioned data imply that the diagnostic performance of abnormal QTc prolongation during recovery may be enhanced at late rather than early recovery, at least for LQTS type 2. The inclusion of QTc at fourth min of recovery in the Schwartz diagnostic criteria for LQTS was actually based on the findings of a well-designed study performed by Sy et al. [8, 10]. This study evaluated relatives of LQTS probands and found that among the ones with normal or borderline QT prolongation the detection of QTc ≥480 ms at fourth min of recovery was 100% specific for LQTS [10]. The aforementioned results appear to be broadly consistent between treadmill and cycle ergometers. B-blocker therapy has been shown not to influence the abnormal QT prolongation during recovery in LQTS [10, 49, 54].

QT hysteresis, defined as the difference in QT between exercise and recovery at similar HR, appears to be abnormally prolonged in LQTS type 2 [10, 55]. QT hysteresis is classically assessed at a HR of 100 bpm, with an abnormal response of >25 ms corresponding to maximum discriminatory capacity [10, 55].

Manifestation of broad-based T waves on exercise testing can be diagnostic for LQTS type 1, while of notched T waves for type 2 [31]. The proportion of patients with LQTS of each subtype that develops these characteristic ECG patterns during exercise appears to be much greater than the respective proportion on resting ECG [31]. Episodes of T wave alternans can uncommonly occur on exercise testing in LQTS, further strengthening the diagnosis of LQTS [36].

Single premature ventricular contractions can be detected in approximately 10% of individuals with LQTS, whereas episodes of nonsustained VT in a far lower proportion [36]. Although these ventricular arrhythmias represent nonspecific findings for LQTS, appearance of polymorphic VT on exercise testing can reinforce diagnostic suspicion for LQTS.

A relative chronotropic incompetence has been demonstrated in LQTS type 1, as indicated by decreased peak HR during exercise compared to healthy controls [56, 57]. This abnormal regulation can be clinically evidenced by a lower peak HR during exercise testing until exhaustion than the age-predicted value [57]. Considering that peak HR during exercise was reported to be less than 95% of age-predicted maximum HR in all individuals with LQTS type 1, the identification of maximum HR exceeding the respective age-predicted value may rule out LQTS type 1 [57].

• The diagnostic performance of the criterion QTc ≥480 ms at fourth min of recovery on exercise testing has not been evaluated in athletes.

• The role of exercise testing in detecting individuals with LQTS type 3 remains to be clarified.

In conclusion, exercise testing plays an important role in the diagnostic evaluation of individuals with borderline QTc prolongation. The most widely accepted criteria for diagnostic confirmation of LQTS on exercise testing include paradoxical prolongation of QTc during exercise and detection of QTc ≥480 ms at fourth min of recovery. Further well-designed studies are needed to explore the diagnostic performance of exercise testing in various subpopulations with QT prolongation, including athletes.

There is no conflict of interest to declare.

There was no funding source.

Georgios A. Christou contributed to the conception and design of the work, interpretation of data, drafting the work, and final approval. Antonios P. Vlahos and Konstantinos A. Christou contributed to the interpretation of data, drafting the work, and final approval. Stefanos Mantzoukas, Chronis A. Drougias, and Dimitrios K. Christodoulou contributed to drafting the work and final approval.