Skip to main content
NCBI home page
Annals of Noninvasive Electrocardiology logoLink to Annals of Noninvasive Electrocardiology
. 2017 Apr 21;22(6):e12455. doi: 10.1111/anec.12455

Automated T‐wave analysis can differentiate acquired QT prolongation from congenital long QT syndrome

Alan Sugrue 1,2, Peter A Noseworthy 2, Vaclav Kremen 2,3, J Martijn Bos 4,6, Bo Qiang 2, Ram K Rohatgi 4, Yehu Sapir 5, Zachi I Attia 2,5, Peter Brady 2, Pedro J Caraballo 1, Samuel J Asirvatham 2,4, Paul A Friedman 2, Michael J Ackerman 2,4,6,✉
PMCID: PMC6931613  PMID: 28429460

Abstract

Background

Prolongation of the QT on the surface electrocardiogram can be due to either genetic or acquired causes. Distinguishing congenital long QT syndrome (LQTS) from acquired QT prolongation has important prognostic and management implications. We aimed to investigate if quantitative T‐wave analysis could provide a tool for the physician to differentiate between congenital and acquired QT prolongation.

Methods

Patients were identified through an institution‐wide computer‐based QT screening system which alerts the physician if the QTc ≥ 500 ms. ECGs were retrospectively analyzed with an automated T‐wave analysis program. Congenital LQTS was compared in a 1:3 ratio to those with an identified acquired etiology for QT prolongation (electrolyte abnormality and/or prescription of known QT prolongation medications). Linear discriminant analysis was performed using 10‐fold cross‐validation to statistically test the selected features.

Results

The 12‐lead ECG of 38 patients with congenital LQTS and 114 patients with drug‐induced and/or electrolyte‐mediated QT prolongation were analyzed. In lead V5, patients with acquired QT prolongation had a shallower T wave right slope (−2,322 vs. −3,593 mV/s), greater T‐peak‐Tend interval (109 vs. 92 ms), and smaller T wave center of gravity on the x axis (290 ms vs. 310 ms; p < .001). These features could distinguish congenital from acquired causes in 77% of cases (sensitivity 90%, specificity 58%).

Conclusion

T‐wave morphological analysis on lead V5 of the surface ECG could successfully differentiate congenital from acquired causes of QT prolongation.

Keywords: electrocardiogram, long QT syndrome, QT prolongation, T‐wave analysis, ventricular repolarization

1. Introduction

Long QT syndrome (LQTS) is characterized by prolongation of the QT interval on the electrocardiogram secondary to delayed cardiac repolarization. Congenital LQTS (cLQTS) is an inherited disorder for which approximately 17 known LQTS‐susceptibility genes have been identified, with LQT1 and LQT2 accounting for about two thirds of all congenital LQTS (Wilde & Bezzina, 2005). In acquired QT prolongation, drugs (both cardiac and noncardiac) account for a large number of cases, as well as electrolyte disturbances (hypomagnesemia, hypokalemia, and hypocalcaemia). A prolonged QT detected on a surface ECG significantly increases the predisposition for torsade de pointes (TdP) and subsequent risk of sudden cardiac death (SCD) (Straus et al., 2006; Yap & Camm, 2003). The identification of the underlying cause of QT prolongation (acquired vs. congenital) is of critical importance given that they harbor different prognosis and underlying management strategies. The ability to augment and empower clinical decision making when a physician first encounters a prolonged QT through facilitating etiological determination would provide valuable information for the clinician in determining underlying cause and possible management strategies, especially as many of the causes of acquired QT prolongation are modifiable to mitigate risk. We have shown previously that T‐wave morphology can help identify those at risk of either sotalol‐ or dofetilide‐induced TdP (Sugrue, Kremen et al., 2015) and in the identification of patients with congenital LQTS even when it is concealed electrocardiographically (Sugrue, Bos et al., 2015). There have also been reports of T‐morphology changes occurring in Ikr blockage (Couderc et al., 2007, 2011) and in those with a prolonged QT interval of any etiology (Pavri, Siu, Andrikopoulou, Ho, & DeCaro, 2014). Therefore, the aim of our study was to determine whether a quantitative automated T‐wave analysis program could provide a tool for the physician to differentiate between congenital and acquired causes of QT prolongation and thereby help augment clinical decision making.

2. Methods

2.1. Study population (ECG selection)

The study population for this study was derived from a previous study evaluating Mayo Clinic's QT‐alert system (Haugaa et al., 2013). For that study, over the time period of November 10, 2010 and June 30, 2011 there were 52,579 unique electrocardiograms (ECGs) performed at Mayo Clinic Rochester (Figure 1). All ECGs were automatically analyzed by an institution‐wide QT‐alert system which screens all ECGs performed at our institution and automatically alerts the physician if the QTc is 500 ms or greater, as previously described (Haugaa et al., 2013). Overall, 1145 ECGs received a “QT alert” each of which was manually reviewed to determine the presence bundle branch block, ventricular pacing, atrial fibrillation, atrial flutter or other supraventricular tachycardias, ST‐T changes in typical ischemic origin, and left ventricular hypertrophy which could impact upon the QT interval and were excluded if any of these were present. If none of these ECG diagnoses above were present, the ECG was subsequently defined as isolated QTc > 500 ms (n = 470, Figure 1). Next, and for this study, we obtained only those ECGs from patients with isolated QTc > 500 ms where the QT prolongation could be attributed to congenital LQTS, QT prolonging drugs, or QT prolonging electrolyte abnormalities (n = 417, Figure 1) and excluded those that had identified other etiologies of QT prolongation (n = 53). We then performed T‐wave analysis using our novel software program. The study protocol was approved by the hospital's institutional review board.

Figure 1.

Figure 1

Open in a new tab

Flowchart of electrocardiograms (ECGs) performed at Mayo Clinic during the study period and identification of study cohort. ECG, electrocardiogram, LQTS, long QT syndrome

2.2. Clinical data

From these 417 subjects, the electronic medical record was reviewed and baseline clinical data, laboratory data (in particular, potassium, magnesium, and calcium), and medications were extracted. Hypokalemia was defined as <3.6 mm/L, hypomagnesemia <1.7 mg/dl, and hypocalcaemia <4.65 mg/dl (only ionized calcium was used). The values closest to the time of the recorded ECG were used. All medications within 7 days before the alerted ECG were reviewed. QT prolonging medications were defined by its inclusion on the Arizona CredibleMeds QT drug list (CredibleMeds).

2.3. T‐wave analysis

The alerted 12‐lead surface ECGs were analyzed using our novel, proprietary T‐wave analysis program as previously described (Sugrue, Kremen et al., 2015). Briefly, the raw, 12‐lead ECG tracings (10 s duration, fs = 500 samples/second) collected from the General Electric MUSE ECG management system (GE Healthcare, Waukesha, WI) were uploaded into our software tool. Preprocessing procedures were applied to enable de‐noising and baseline correction. This was followed by ECG feature extraction using our automated software program. ECG T wave features from the ECG are detected by a Bayesian statistical peak delineation algorithm (Lin, Mailhes, & Tourneret, 2010). Multiple beats over a 10 s ECG strip are used for analysis. The T wave features analyzed by the software are illustrated in Figure 2. Specific T wave features used for this analysis wereT wave left/right slope, T‐wave amplitude, T wave enclosed area, x/y coordinates of the center of gravity (COG) of the first 25% of a T wave (T1), x/y coordinates of COG of the last 25% of a T wave (T4), and Tpeak‐Tend (Tpe) interval.

Figure 2.

Figure 2

Open in a new tab

Illustration of the features quantified by our novel T‐wave analysis program

Further detailed information can be reviewed in supplementary file. Leads were excluded from analysis if the T wave was of low amplitude (<0.1 mV), if we were unable to interpret the T wave from a poor tracing due to interference, or if the T wave was biphasic. To preserve what we would expect in an everyday ECG recording, if the lead was not interpretable, this lead was excluded from the analysis.

2.4. ECG feature selection/statistical approach

To identify ECG features (mentioned above) which could differentiate congenital from acquired QT prolongation in each ECG lead, we performed a univariate two‐sample t‐test to preselect features that showed statistically significantly different means by case status. Features whose mean for a specific feature significantly (p < .05) differed between cases (congenital LQTS) and controls (acquired QT prolongation secondary to drugs/electrolytes) underwent further analysis. To select independent features, Pearson's correlation coefficient was used to examine correlation between preselected features. A filter approach was used to find the features with lowest p‐value with case status while having low mutual correlation between features. A mutual correlation threshold of features was chosen to be |ρ < 0.6|. To evaluate and compare performance of features selected by the filter approach, a 10‐fold cross‐validation using the LDA classifier was used. Given that the final cohort was likely to contain more patients with acquired QT prolongation than congenital LQTS, we randomly matched with a ratio of 1:3 (congenital: acquired).

Although all 12 leads were initially analyzed, a single lead was selected for analysis. The decision behind this was for potential ease of use by the physician without relying on multiple leads for determination. Lead V5 provided the greatest ability to discriminate acquired from congenital and so was used for analysis. Using the ECG features from lead V5, we also calculated the diagnostic accuracy, sensitivity, specificity, negative predictive value (NPV), and positive predictive values (PPV) for these features to accurately identify the presence of acquired LQTS. We performed a sensitivity analysis to test the robustness of our findings given that demographics (in particular age) between the acquired and congenital group could be potential confounders. To do this, we planned an analysis on those patients <30 years of age with either congenital LQTS or acquired QT prolongation, using the variables that the primary analysis would conclude were the best predictors. We also examined for a sex confounder in the T wave features by performing linear discriminant by gender. In addition, we also tested for statistical significant differences in the selected T wave features between the acquired and congenital QT prolongation groups.

3. Results

3.1. Study population

In our cohort of 417 patients, with electrocardiographically isolated QT prolongation from either electrolytes and/or QT prolonging medications or from congenital LQTS, 311 ECGs were of sufficient quality to be analyzed using our algorithms (74%). Among these, there were 38 patients with congenital LQTS and 286 with acquired QT prolongation. Using the 38 congenital LQTS cases, we randomly selected an acquired QT prolongation case in a 1:3 ratio, leaving 38 cases (congenital LQTS; 15 LQT1, 23 LQT2) and 114 controls (acquired QT prolongation) as our analysis cohort. Table 1 describes the baseline characteristics of this cohort.

Table 1.

Baseline Demographics

Congenital LQTS Acquired QT Prolongation p‐Value
Number 38 114
Male, n (%) 8 (25%) 45 (47%) p < .001
Age, years ± SD 15 ± 12 66 ± 14 p < .001
QTc, ms ± SD 500 ± 30 520 ± 29 p = .004
Etiology (%)
Drugs 78 (81%)
Hypocalcemia 33 (34%)
Hypokalemia 25 (26%)
Hypomagnesemia 25 (26%)
Open in a new tab

LQTS, long QT syndrome. QTc, heart rate‐corrected QT interval by Bazett's formula.

3.2. T‐wave analysis

Univariate analysis demonstrated Lead V5 provided the greatest discrimination ability between congenital LQTS and acquired QT prolongation, and this was used subsequently for analysis. Unfortunately, this lead was not readable in 6 patients with LQTS leaving our final cohort of 32 congenital LQTS cases versus 96 with acquired QT prolongation. In lead V5, we observed that patients with acquired QT prolongation had a significantly shallower T wave right slope (−2,322 ± 2,400 vs. −3,593 ± 1212 mV/s, p < .001), greater Tpeak‐Tend interval (109 ± 29 vs. 92 ± 31 ms, p < .001), and smaller T wave center of gravity of x axis (0.29 ± 4 s vs. 0.31 ± 4 s, p < .001).

3.3. Prediction

Use of these three T wave features enabled us to successful identify congenital versus acquired LQTS in 77% of cases: with a sensitivity of 90%, a specificity of 58%, a positive predictive value of 83%, and a negative predictive value of 71% (Figure 3). When examining those that had QT prolongation only due to drugs (n = 40), classification error was similar, 78%.

Figure 3.

Figure 3

Open in a new tab

Scatter plot showing linear discriminate analysis results for acquired QT prolongation and congenital Long QT syndrome, using two T wave features

3.4. Sensitivity analysis

We performed an additional sensitivity analysis to exclude the possibility of an age‐related and sex confounder within our cohort. For age we exclusively looked at those patients under the age of 30 years (n = 50 patients, 29 cLQTS, 21 acquired QT prolongation) and observed the results remained robust. Using the determined features above (T wave right slope, Tpeak–Tend Interval, and T wave center of gravity x axis), we could successful predict the underlying etiology of the prolonged QT (acquired vs. congenital) in 78.3%. For gender there was no statistically significant difference detected between male and females across all three selected T wave features (data not shown) and the ability to predict the underlying etiology of QT prolongation using only male and females patients was similar to each other and the overall cohort (Males 76%, Females 78%, and whole cohort 77%).

4. Discussion

When a clinician is faced with a patient who has QT prolongation, particularly in those cases where the QT has crossed the proarrhythmic threshold (QTc > 500 ms), the underlying etiology may be unclear at the time the QT prolongation is first detected. Utilizing automated T‐wave analysis, we detected subtle variations in cardiac repolarization that can differentiate and identify the underlying etiology (acquired QT prolongation vs. cLQTS). The ability to distinguish acquired from congenital QT prolongation is of critical importance and can inform subsequent management strategies and prognosis, especially as many of the underlying causes for acquired QT prolongation are treatable or modifiable.

Our group has used quantitative T‐wave analysis to detect changes in repolarization previously (Sara et al., 2016; Sugrue, Bos et al., 2015; Sugrue, Kremen et al., 2015). Identification and differentiation of acquired QT prolongation and cLQTS is important as although genotype‐specific patterns of T‐wave morphology have been well described in cLQTS (Dausse et al., 1996; Lehmann et al., 1994; Moss et al., 1995; Zhang et al., 2000), they are observed inconstantly and are often subtle (Sy et al., 2011) making the detection of cLQTS difficult. To the best of our knowledge, there are only a handful of studies that have described T wave changes with inhibition of the rapidly acting delayed rectifier potassium channel (I Kr) (Couderc et al., 2007, 2011; Graff et al., 2009). We have expanded upon this and demonstrate that subtle changes in repolarization can be detected, quantified, and utilized to differentiate between causes of QT prolongation. Couderc et al. observed an increased Tpeak–Tend (TpTe) interval (105 ± 35 vs. 75 ± 13 ms, p < .001) and flatter right slope (2.4 ± 1.6 vs. 4.5 ± 2.2 μV/ms, p < .001) when comparing patients with moxifloxacin‐induced QT prolongation (pharmacologic block of I Kr) and patients with LQT2 which is caused by genetically mediated loss‐of‐function in I Kr (Couderc et al., 2011). This data is similar to ours in that TpTe and T wave right slope were selected ECG features. However, we observed a flatter T wave and longer TpTe in the acquired group. This is attributed to the fact that we examined all LQTS genotypes rather than LQT2 exclusively, which would likely influence the overall average values.

The mechanism behind acquired QT prolongation is largely due to changes in the I Kr channel (a critical channel in the phase 3 repolarization of the cardiac action potential), which manifests in T‐wave repolarization changes. Drugs that are known to cause prolongation of the QT are often related to the blockage of the I Kr channel, although there is some evidence that I Ks (Veerman et al., 2013) and I NA may also be involved. Hypokalemia predisposes to prolongation through modification in the function of the I Kr channel. Specifically, this modification is a decrease in I Kr by enhanced inactivation (Yang, Snyders, & Roden, 1997) or exaggerated competitive block by sodium (Numaguchi, Johnson, Petersen, & Balser, 2000). Interesting, extracellular potassium is a critical determinant of drug block of I Kr (Yang & Roden, 1996 ), which has significant implications for clinical practice and why potassium replacement in those receiving drugs causing I Kr blockade is important. Hypomagnesemia is related due to its ability to directly cause hypokalemia. However, it also has its own potential mechanisms by influencing the inward rectification of the potassium channels (Matsuda, 1991; Vandenberg, 1987), as well as its potential impact upon the L‐type calcium channels (Kannankeril, Roden, & Darbar, 2010). From our data, we have successfully detected different changes in these channels (especially I Kr), which manifest as T wave repolarization abnormalities. Particularly, it seems that T wave right slope and Tpeak Tend are likely to be markers of I Kr channel dysfunction.

One potential application of this technology is drug screening for potential I Kr activity. Although we did not look at arrhythmogenic outcomes (namely, TdP) in this study, the T‐wave morphological feature we have identified could have a potential role in the future identification and risk assessment of drug arrhythmogenesis. QT prolongation is an imperfect surrogate for torsadogenic potential (Antzelevitch & Shimizu, 2002; Hondeghem, 2006; Thomsen, Volders, Beekman, Matz, & Vos, 2006; Yap & Camm, 2003) as the risk of TdP is neither a linear function of the baseline QT interval nor of the extent of QT interval prolongation during drug administration (Roden, 2004). There is a need for a better/complementary marker of risk, potentially T wave right slope and TpTe interval. T wave right slope as identified in this study was also previously shown by our group to be correlated with TdP in patients loaded with sotaolol and dofetilide (Sugrue, Kremen et al., 2015). In addition, the TpTe interval is considered a good marker for arrhythmogenic risk (Gupta et al., 2008; Letsas, Weber, Astheimer, Kalusche, & Arentz, 2010). Prolongation of this interval increases the period when potential fatal reentry ventricular tachycardias can occur and has been linked to arrhythmogenesis in long QT syndromes (Topilski et al., 2007), hypertrophic cardiomyopathy (Shimizu et al., 2002), patients receiving primary percutaneous coronary intervention for an MI (Haarmark et al., 2009), and Brugada syndrome (Letsas et al., 2010). Further work identifying the role of these repolarization features in arrhythmogenic risk in those with both acquired QT prolongation and cLQTS is important. It could provide an additional tool for drug testing, particularly given the concerns surrounding the current FDA thorough QT/QTc (TQT) drug‐testing protocol.

5. Limitations

While internally validated with a 10‐fold cross‐validation, we do not have a separate validation cohort for which to run the analysis. With the small sample size and many features searched, the model may be overtrained and therefore caution should be applied in interpreting our results. Acquired QT prolongation and cLQTS patients were not matched for demographic characteristics that could also be potentially associated with repolarization; however, our more focused sensitivity analysis shows our results were very robust. Some patients with acquired causes may harbor functional common polymorphisms that predispose them to QT prolongation and risk of sudden death (Lehtonen et al., 2007; Yang et al., 2002). Unfortunately, we have no data on the genetic background in this population. Although we view this as a potential limitation, it is important to also note its potential strength. If our population did harbor these mutations, the T‐wave analysis was still able to differentiate between those with manifest cLQTS and those with potential polymorphism. We used the Bazett's formula to calculate the QTc, which is the most commonly used in clinical practice, however, there is the possibility of false prolongations with this formula (Luo, Michler, Johnston, & Macfarlane, 2004). Lastly, it is known that different T wave features exist for each LQTS phenotype, our analysis did not stratify per genotype so while these T wave changes may help you decide if it is cLQTS it will not specifically address the underlying phenotype.

6. Conclusion

We demonstrate that a quantitative T‐wave analysis tool can successfully differentiate acquired and congenital causes of QT prolongation through surface ECG analysis. This tool may have implications for clinical practice in informing decision making regarding idiopathic QT prolongation.

Disclosures

Dr Ackerman is a consultant for: Boston Scientific, Medtronic, St. Jude Medical, Transgenomic. Royalties and Intellectual Property from Transgenomic.

Dr Asirvatham is a consultant for Abiomed, Atricure, Biotronik, Biosense Webster, Boston Scientific, Medtronic, Spectranetics, St Jude Medical, Sanofi‐Aventis, Wolters Kluwer, and Elsevier.

The other authors report no disclosures.

Supporting information

 

Click here for additional data file. (45.6KB, docx)

Sugrue A, Noseworthy PA, Kremen V, et al. Automated T‐wave analysis can differentiate acquired QT prolongation from congenital long QT syndrome. Ann Noninvasive Electrocardiol. 2017;22:e12455 10.1111/anec.12455

Funding information

VK was supported by institutional resources for research by Czech Technical University in Prague, Czech Republic

References

  1. Antzelevitch, C. , & Shimizu, W. (2002). Cellular mechanisms underlying the long QT syndrome. Current Opinion in Cardiology, 17, 43–51. [DOI] [PubMed] [Google Scholar]
  2. Couderc, J. P. , Vaglio, M. , Xia, X. , McNitt, S. , Wicker, P. , Sarapa, N. , … Zareba, W. (2007). Impaired T‐amplitude adaptation to heart rate characterizes I(Kr) inhibition in the congenital and acquired forms of the long QT syndrome. Journal of Cardiovascular Electrophysiology, 18, 1299–1305. [DOI] [PubMed] [Google Scholar]
  3. Couderc, J. P. , Xia, X. , Peterson, D. R. , McNitt, S. , Zhao, H. , Polonsky, S. , … Zareba, W. (2011). T‐wave morphology abnormalities in benign, potent, and arrhythmogenic I(kr) inhibition. Heart Rhythm: the Official Journal of the Heart Rhythm Society, 8, 1036–1043. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. CredibleMeds . Retrived from https://www.crediblemeds.org/index.php[registration required] Access Date: July 18, 2012.
  5. Dausse, E. , Berthet, M. , Denjoy, I. , Andre‐Fouet, X. , Cruaud, C. , Bennaceur, M. , … Guicheney, P. (1996). A mutation in HERG associated with notched T waves in long QT syndrome. Journal of Molecular and Cellular Cardiology, 28, 1609–1615. [DOI] [PubMed] [Google Scholar]
  6. Graff, C. , Andersen, M. P. , Xue, J. Q. , Hardahl, T. B. , Kanters, J. K. , Toft, E. , … Struijk, J. J. (2009). Identifying drug‐induced repolarization abnormalities from distinct ECG patterns in congenital long QT syndrome: A study of sotalol effects on T‐wave morphology. Drug Safety, 32, 599–611. [DOI] [PubMed] [Google Scholar]
  7. Gupta, P. , Patel, C. , Patel, H. , Narayanaswamy, S. , Malhotra, B. , Green, J. T. , & Yan, G. X. (2008). T(p‐e)/QT ratio as an index of arrhythmogenesis. Journal of Electrocardiology, 41, 567–574. [DOI] [PubMed] [Google Scholar]
  8. Haarmark, C. , Hansen, P. R. , Vedel‐Larsen, E. , Pedersen, S. H. , Graff, C. , Andersen, M. P. , … Kanters, J. K. (2009). The prognostic value of the Tpeak‐Tend interval in patients undergoing primary percutaneous coronary intervention for ST‐segment elevation myocardial infarction. Journal of Electrocardiology, 42, 555–560. [DOI] [PubMed] [Google Scholar]
  9. Haugaa, K. H. , Bos, J. M. , Tarrell, R. F. , Morlan, B. W. , Caraballo, P. J. , & Ackerman, M. J. (2013). Institution‐wide QT alert system identifies patients with a high risk of mortality. Mayo Clinic Proceedings, 88, 315–325. [DOI] [PubMed] [Google Scholar]
  10. Hondeghem, L. M. (2006). Thorough QT/QTc not so thorough: Removes torsadogenic predictors from the T‐wave, incriminates safe drugs, and misses profibrillatory drugs. Journal of Cardiovascular Electrophysiology, 17, 337–340. [DOI] [PubMed] [Google Scholar]
  11. Kannankeril, P. , Roden, D. M. , & Darbar, D. (2010). Drug‐induced long QT syndrome. Pharmacological Reviews, 62, 760–781. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Lehmann, M. H. , Suzuki, F. , Fromm, B. S. , Frankovich, D. , Elko, P. , Steinman, R. T. , … Taggart, R. T. (1994). T wave “humps” as a potential electrocardiographic marker of the long QT syndrome. Journal of the American College of Cardiology, 24, 746–754. [DOI] [PubMed] [Google Scholar]
  13. Lehtonen, A. , Fodstad, H. , Laitinen‐Forsblom, P. , Toivonen, L. , Kontula, K. , & Swan, H. (2007). Further evidence of inherited long QT syndrome gene mutations in antiarrhythmic drug‐associated torsades de pointes. Heart Rhythm: the Official Journal of the Heart Rhythm Society, 4, 603–607. [DOI] [PubMed] [Google Scholar]
  14. Letsas, K. P. , Weber, R. , Astheimer, K. , Kalusche, D. , & Arentz, T. (2010). Tpeak‐Tend interval and Tpeak‐Tend/QT ratio as markers of ventricular tachycardia inducibility in subjects with Brugada ECG phenotype. Europace, 12, 271–274. [DOI] [PubMed] [Google Scholar]
  15. Lin, C. , Mailhes, C. , & Tourneret, J.‐Y. (2010). P‐ and T‐wave delineation in ECG signals using a Bayesian approach and a partially collapsed Gibbs sampler. IEEE Transactions on Bio‐Medical Engineering, 57, 2840–2849. [DOI] [PubMed] [Google Scholar]
  16. Luo, S. , Michler, K. , Johnston, P. , & Macfarlane, P. W. (2004). A comparison of commonly used QT correction formulae: The effect of heart rate on the QTc of normal ECGs. Journal of Electrocardiology, 37(Suppl), 81–90. [DOI] [PubMed] [Google Scholar]
  17. Matsuda, H. (1991). Magnesium gating of the inwardly rectifying K+ channel. Annual Review of Physiology, 53, 289–298. [DOI] [PubMed] [Google Scholar]
  18. Moss, A. J. , Zareba, W. , Benhorin, J. , Locati, E. H. , Hall, W. J. , Robinson, J. L. , … Lehmann, M. H. (1995). ECG T‐wave patterns in genetically distinct forms of the hereditary long QT syndrome. Circulation, 92, 2929–2934. [DOI] [PubMed] [Google Scholar]
  19. Numaguchi, H. , Johnson, J. P. Jr , Petersen, C. I. , & Balser, J. R. (2000). A sensitive mechanism for cation modulation of potassium current. Nature Neuroscience, 3, 429–430. [DOI] [PubMed] [Google Scholar]
  20. Pavri, B. B. , Siu, H. , Andrikopoulou, E. , Ho, R. , & DeCaro, M. (2014). T Wave Slopes: A novel method for assessment OF repolarization dispersion from surface ECGS with prolonged QT as compared to normal ECGS. Journal of the American College of Cardiology, 63, A1638. [Google Scholar]
  21. Roden, D. M. (2004). Drug‐induced prolongation of the QT interval. The New England Journal of Medicine, 350, 1013–1022. [DOI] [PubMed] [Google Scholar]
  22. Sara, J. D. , Sugrue, A. , Kremen, V. , Qiang, B. , Sapir, Y. , Attia, Z. I. , … Noseworthy, P. A. (2016). Electrocardiographic predictors of coronary microvascular dysfunction in patients with non‐obstructive coronary artery disease: Utility of a novel T wave analysis program. International Journal of Cardiology, 203, 601–606. [DOI] [PubMed] [Google Scholar]
  23. Shimizu, M. , Ino, H. , Okeie, K. , Yamaguchi, M. , Nagata, M. , Hayashi, K. , … Mabuchi, H. (2002). T‐peak to T‐end interval may be a better predictor of high‐risk patients with hypertrophic cardiomyopathy associated with a cardiac troponin I mutation than QT dispersion. Clinical Cardiology, 25, 335–339. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Straus, S. M. , Kors, J. A. , De Bruin, M. L. , van der Hooft, C. S. , Hofman, A. , Heeringa, J. , … Witteman, J. C. (2006). Prolonged QTc interval and risk of sudden cardiac death in a population of older adults. Journal of the American College of Cardiology, 47, 362–367. [DOI] [PubMed] [Google Scholar]
  25. Sugrue, A. , Bos, J. M. , Kremen, V. , Qiang, B. , Friedman, P. A. , Ackerman, M. J. , & Noseworthy, P. A. (2015). Identification of genotype‐specific electrocardiogram patterns in long QT syndrome using a novel. Automated T wave analysis program. Circulation, 132, A17604. [Google Scholar]
  26. Sugrue, A. , Kremen, V. , Qiang, B. , Sheldon, S. H. , DeSimone, C. V. , Sapir, Y. , … Noseworthy, P. A. (2015). Electrocardiographic predictors of torsadogenic risk during dofetilide or sotalol initiation: Utility of a novel T wave analysis program. Cardiovascular Drugs and Therapy, 29, 433–441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Sy, R. W. , van der Werf, C. , Chattha, I. S. , Chockalingam, P. , Adler, A. , Healey, J. S. , … Krahn, A. D. (2011). Derivation and validation of a simple exercise‐based algorithm for prediction of genetic testing in relatives of LQTS probands. Circulation, 124, 2187–2194. [DOI] [PubMed] [Google Scholar]
  28. Thomsen, M. B. , Volders, P. G. , Beekman, J. D. , Matz, J. , & Vos, M. A. (2006). Beat‐to‐Beat variability of repolarization determines proarrhythmic outcome in dogs susceptible to drug‐induced torsades de pointes. Journal of the American College of Cardiology, 48, 1268–1276. [DOI] [PubMed] [Google Scholar]
  29. Topilski, I. , Rogowski, O. , Rosso, R. , Justo, D. , Copperman, Y. , Glikson, M. , … Viskin, S. (2007). The morphology of the QT interval predicts torsade de pointes during acquired bradyarrhythmias. Journal of the American College of Cardiology, 49, 320–328. [DOI] [PubMed] [Google Scholar]
  30. Vandenberg, C. A. (1987). Inward rectification of a potassium channel in cardiac ventricular cells depends on internal magnesium ions. Proceedings National Academy Science United States of America, 84, 2560–2564. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Veerman, C. C. , Verkerk, A. O. , Blom, M. T. , Klemens, C. A. , Langendijk, P. N. , van Ginneken, A. C. , … Tan, H. L. (2013). Slow delayed rectifier potassium current blockade contributes importantly to drug‐induced long QT syndrome. Circulation: Arrhythmia and Electrophysiology, 6, 1002–1009. [DOI] [PubMed] [Google Scholar]
  32. Wilde, A. A. , & Bezzina, C. R. (2005). Genetics of cardiac arrhythmias. Heart, 91, 1352–1358. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Yang, P. , Kanki, H. , Drolet, B. , Yang, T. , Wei, J. , Viswanathan, P. C. , … Roden, D. M. (2002). Allelic variants in long‐QT disease genes in patients with drug‐associated torsades de pointes. Circulation, 105, 1943–1948. [DOI] [PubMed] [Google Scholar]
  34. Yang, T. , & Roden, D. M. (1996). Extracellular potassium modulation of drug block of IKr. Implications for torsade de pointes and reverse use‐dependence. Circulation, 93, 407–411. [DOI] [PubMed] [Google Scholar]
  35. Yang, T. , Snyders, D. J. , & Roden, D. M. (1997). Rapid inactivation determines the rectification and [K+]o dependence of the rapid component of the delayed rectifier K+ current in cardiac cells. Circulation Research, 80, 782–789. [DOI] [PubMed] [Google Scholar]
  36. Yap, Y. G. , & Camm, A. J. (2003). Drug induced QT prolongation and torsades de pointes. Heart, 89, 1363–1372. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Zhang, L. , Timothy, K. W. , Vincent, G. M. , Lehmann, M. H. , Fox, J. , Giuli, L. C. , … Medina, A. (2000). Spectrum of ST‐T‐wave patterns and repolarization parameters in congenital long‐QT syndrome: ECG findings identify genotypes. Circulation, 102, 2849–2855. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

 

Click here for additional data file. (45.6KB, docx)

Articles from Annals of Noninvasive Electrocardiology : The Official Journal of the International Society for Holter and Noninvasive Electrocardiology, Inc are provided here courtesy of International Society for Holter and Noninvasive Electrocardiology, Inc. and Wiley Periodicals, Inc.

RESOURCES