Author's Accepted Manuscript
Diagnostic Value of T-Wave Morphology Changes During “Qt-Stretching” in Patients With Long Qt Syndrome. Ehud Chorin M.D. PhD, Ofer Havakuk M.D., Arnon Adler M.D., Arie Steinvil M.D., UriTE: this author affilation is missing plz.check nd provide the correct affilation. Rozovski M.D., Christian van der Werf M. D., Ph.D., Pieter G. Postema M.D., Ph.D., Guy Topaz M.D., Arthur A.M. Wilde M.D., Ph.D., Sami Viskin M.D, Raphael Rosso M.D.
PII: DOI: Reference:
S1547-5271(15)00814-0 http://dx.doi.org/10.1016/j.hrthm.2015.06.040 HRTHM6339
To appear in:
Heart Rhythm
www.elsevier.com/locate/buildenv
Cite this article as: Ehud Chorin M.D. PhD, Ofer Havakuk M.D., Arnon Adler M.D., Arie Steinvil M.D., UriTE: this author affilation is missing plz.check nd provide the correct affilation. Rozovski M.D., Christian van der Werf M.D., Ph.D., Pieter G. Postema M.D., Ph.D., Guy Topaz M.D., Arthur A.M. Wilde M.D., Ph.D., Sami Viskin M.D, Raphael Rosso M.D., Diagnostic Value of T-Wave Morphology Changes During “QtStretching” in Patients With Long Qt Syndrome., Heart Rhythm, http://dx.doi.org/ 10.1016/j.hrthm.2015.06.040 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Diagnostic value of T-wave morphology changes during “QT-stretching” in patients with long QT syndrome.
Ehud Chorin,1 M.D. PhD, Ofer Havakuk,1 M.D., Arnon Adler,1M.D., Arie Steinvil,1 M.D., Uri Rozovski, M.D., Christian van der Werf,3 M.D., Ph.D. Pieter G. Postema,3 M.D., Ph.D., Guy Topaz,2 M.D., Arthur A.M. Wilde,3 M.D., Ph.D., Sami Viskin,1 M.D. and Raphael Rosso,1 M.D.
1
Department of Cardiology and 2Internal Medicine D, Tel Aviv Sourasky Medical Center and Sackler School of Medicine, Tel Aviv University, Israel and 2Heart Center, Department of Clinical and Experimental Cardiology, Academic Medical Center, Amsterdam, The Netherlands.
Running head: T-wave morphology changes in long QT syndrome
Corresponding author: Sami Viskin, M.D. Department of Cardiology. Tel Aviv Sourasky Medical Center. Weizman 6, Tel Aviv 64239. Israel. Tel: +972536973311 Fax: +97236972749 Mobile: +972524266859 E-mail: [email protected]
Conflicts of interest: None. Financial Disclosures: None.
Abstract Background. Specific T-wave patterns in the resting electrocardiogram (ECG) aid in diagnosing long QT syndrome (LQTS) and identifying the specific genotype. However,
provocation tests are often required to establish a diagnosis when the QT interval is borderline at rest.
Objectives. To determine if the T-wave morphology changes provoked by standing aid in the diagnosis of LQTS and determining the genotype.
Methods. The quick standing test was performed by 100 LQTS patients (40 LQT1, 42 LQT2 and 18 LQT3) and 100 controls. We used logistic regression to determine if T-wave morphology changes provoked by standing add to the already established diagnostic value of QTc-stretching for identifying LQTS.
Results: During maximal QT stretching, the T-wave morphologies that best discriminated LQTS from controls included: “notched,” “late-onset” and “biphasic” T-waves. These 3 categories were grouped into a category named “abnormal T-wave response to standing”. During quick standing, a QTc-stretched ≥ 490 msec increased the odds for correctly identifying LQTS. T-wave morphology changes provoked by standing were most helpful for identifying LQT2, less helpful for LQT1 and least helpful for LQT3. Conclusion: The sudden heart rate acceleration produced by abrupt standing, not only increases the QTc but also exposes abnormal T-waves that are valuable for diagnosing LQTS.
Keywords: Long QT syndrome; Electrocardiogram; QT interval; T-wave morphology
Abbreviations. CI = confidence intervals; ECG = electrocardiogram; LQTS = long QT syndrome; LQT1, LQT2 and LQT3 = LQTS type I, II and III, respectively; PPV and NPV = positive and negative predictive value.
2
I find it tempting to say that, when dealing with a possible case of long QT syndrome, one does not measure the QT interval, one looks at it. Peter J. Schwartz, 1989 1.
Timely diagnosis of congenital long QT syndrome (LQTS) is crucial because of the malignant nature of the disease and the existence of effective preventive therapy.2 However, accurate diagnosis is often difficult because >30% of the patients with genetically confirmed LQTS have QTc intervals within the range also seen in the healthy population.3 Consequently, many patients with suspected LQTS require additional tests for establishing or excluding the diagnosis. Such tests expose the abnormal response of the QT interval to spontaneous or provoked changes in heart rate throughout Holter recordings,4 exercise examinations5-8 or during epinephrine9-11 or adenosine12 challenge tests. We13,14 and others15,16 recently proposed the “quick standing” test as a simple bedside test that facilitates the diagnosis of congenital LQTS. The test takes advantage of the fact that as one stands up, the heart rate acceleration is abrupt while the associated QT-interval shortening is gradual.13,14 As the R-R interval shortens faster than the QT interval, the QT appears to “stretch” toward the next P wave and the corrected QT interval (QTc) for heart rate actually increases momentarily. The phenomenon of “QT stretching” is universal but is exaggerated in patients with LQTS, allowing for a simple but accurate diagnostic test.13,14 The duration of the QT interval is not the only diagnostic clue. Analysis of the T-wave morphology in the resting ECG of patients with LQTS not only increases diagnostic accuracy but also provides useful information regarding the specific genotype involved. Indeed, broad Twaves are common in LQT1 whereas notched T-waves are typically seen in LQT2 and lateonset peaked T-waves are characteristic of LQT3.17-20 Further characterization of T-wave morphology changes in response to epinephrine infusion was done by Ackerman’s group.10 They demonstrated that the appearance of specific forms of T-wave notching during low-dose epinephrine identifies patients with LQT2 with the test loosing specificity as higher epinephrine 3
doses are used.10 Finally, we reported specific T-wave patterns that predict acquired LQTS during bradyarrhythmias.21 Yet, there is no data on the effects of quick standing on T-wave morphology. We therefore performed the present analysis, to determine if the T-wave morphology changes provoked by the sudden acceleration triggered by standing aid in diagnosing LQTS or in determining the genotype.
Methods Population. The study cohort consisted of all the LQTS patients and controls included in the original studies of the “quick standing” test13,14 except for 6 patients for whom genetic confirmation remains elusive. In addition, all the LQTS patients studied in Tel Aviv since our last publication,14 all with genetic confirmation of the diagnosis, were included here, bringing the total of LQTS patients to 100. The control group was also enlarged to a matching number of 100 healthy individuals. “Obvious QT prolongation at baseline” was defined as QTc >470 msec for males and QTc >480 msec for females because such values are not observed in large cohorts of healthy adults.22
Abrupt standing test (QT stretching). The test protocol was previously described.13 Subjects rested for ≥5 minutes in the supine position as a baseline electrocardiogram (ECG) was recorded. They were then asked to stand up quickly and remain standing still for 5 minutes during continuous ECG recording. QT measurements and evaluation of T-wave morphology were performed at 4 points in time: 1) baseline (at the slowest sinus rhythm recorded during the resting period), 2) maximal heart rate (at the fastest sinus rate during heart rate acceleration provoked by standing), 3) point of maximal “QT stretching” (defined as the time when the end of the T-wave gets nearest to the next P wave due to R-R–interval shortening without sufficient QT-interval shortening)13 and 4) upon return of the heart rate to its baseline value (QTstunning)14 (the first R-R interval during heart rate deceleration that is within 40 msec of the 4
baseline R-R interval). At all these stages, the QT interval was corrected using the Bazett formula. For the present study, we focused on the effects of the abrupt heart rate changes provoked by quick standing on the morphology of the T-wave at each of the above-mentioned points in time. Two cardiologists (E.C. and O.H.) who did not participate in our previous studies13,14 and were blinded to patient allotting classified T-wave morphology changes by consensus. Our classification of T-wave morphology integrated the classifications proposed by Moss,17 Ackerman10 and our group21 for T-wave morphology at rest, during epinephrine test, and during bradyarrhythmias, respectively. The resulting classification is shown in Figure 1. Assessment of T-wave morphology was done independently in three groups of leads: (1) II, III, aVF, (2) V1-V3 and (3) V4-V6 (leads I and aVL were not recorded by the analog ECG recorder used for the initial tests performed in Tel Aviv). For each patient, T-wave abnormalities were counted in all 3 lead-groups separately, leading to a total of 300 patient-leads and 300 control-leads. Frequencies of T-wave abnormalities are presented as percentages from 100 patients (vs. 100 controls) and percentage from 300 patient-leads (vs. 300 control-leads). The study was approved by our institutional review committee.
Statistics. All data were summarized and displayed as mean ± standard deviation for continuous variables and as number (and percentage) of patients in each group for categorical variables. The p values for the categorical variables were calculated with the chi-square and fisher's exact tests. Continuous variables were compared using the independent sample t-test. In our original studies13,14 receiver operator characteristics curves identified cut-off values of QTc 423 msec at baseline and QTc of 489 msec during maximal QT stretching in response to standing, as the values that distinguished between LQT-patients and controls with a sensitivity of 90%. For the 5
present study, we rounded those numbers to a cut-off value of baseline QTc 425 msec and QTcstretched of 490 msec. We then used logistic regression to determine if T-wave morphology changes at baseline or provoked by standing improved to the diagnostic value of the test for correctly identifying LQTS. In this model, the Exp (ß) was used to estimate the odds ratio (OR) and 95% confidence intervals (CI) compared to the reference group of individuals with normal T wave and QTc-stretched
Diagnostic Value of T-Wave Morphology Changes During “Qt-Stretching” in Patients With Long Qt Syndrome. Ehud Chorin M.D. PhD, Ofer Havakuk M.D., Arnon Adler M.D., Arie Steinvil M.D., UriTE: this author affilation is missing plz.check nd provide the correct affilation. Rozovski M.D., Christian van der Werf M. D., Ph.D., Pieter G. Postema M.D., Ph.D., Guy Topaz M.D., Arthur A.M. Wilde M.D., Ph.D., Sami Viskin M.D, Raphael Rosso M.D.
PII: DOI: Reference:
S1547-5271(15)00814-0 http://dx.doi.org/10.1016/j.hrthm.2015.06.040 HRTHM6339
To appear in:
Heart Rhythm
www.elsevier.com/locate/buildenv
Cite this article as: Ehud Chorin M.D. PhD, Ofer Havakuk M.D., Arnon Adler M.D., Arie Steinvil M.D., UriTE: this author affilation is missing plz.check nd provide the correct affilation. Rozovski M.D., Christian van der Werf M.D., Ph.D., Pieter G. Postema M.D., Ph.D., Guy Topaz M.D., Arthur A.M. Wilde M.D., Ph.D., Sami Viskin M.D, Raphael Rosso M.D., Diagnostic Value of T-Wave Morphology Changes During “QtStretching” in Patients With Long Qt Syndrome., Heart Rhythm, http://dx.doi.org/ 10.1016/j.hrthm.2015.06.040 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Diagnostic value of T-wave morphology changes during “QT-stretching” in patients with long QT syndrome.
Ehud Chorin,1 M.D. PhD, Ofer Havakuk,1 M.D., Arnon Adler,1M.D., Arie Steinvil,1 M.D., Uri Rozovski, M.D., Christian van der Werf,3 M.D., Ph.D. Pieter G. Postema,3 M.D., Ph.D., Guy Topaz,2 M.D., Arthur A.M. Wilde,3 M.D., Ph.D., Sami Viskin,1 M.D. and Raphael Rosso,1 M.D.
1
Department of Cardiology and 2Internal Medicine D, Tel Aviv Sourasky Medical Center and Sackler School of Medicine, Tel Aviv University, Israel and 2Heart Center, Department of Clinical and Experimental Cardiology, Academic Medical Center, Amsterdam, The Netherlands.
Running head: T-wave morphology changes in long QT syndrome
Corresponding author: Sami Viskin, M.D. Department of Cardiology. Tel Aviv Sourasky Medical Center. Weizman 6, Tel Aviv 64239. Israel. Tel: +972536973311 Fax: +97236972749 Mobile: +972524266859 E-mail: [email protected]
Conflicts of interest: None. Financial Disclosures: None.
Abstract Background. Specific T-wave patterns in the resting electrocardiogram (ECG) aid in diagnosing long QT syndrome (LQTS) and identifying the specific genotype. However,
provocation tests are often required to establish a diagnosis when the QT interval is borderline at rest.
Objectives. To determine if the T-wave morphology changes provoked by standing aid in the diagnosis of LQTS and determining the genotype.
Methods. The quick standing test was performed by 100 LQTS patients (40 LQT1, 42 LQT2 and 18 LQT3) and 100 controls. We used logistic regression to determine if T-wave morphology changes provoked by standing add to the already established diagnostic value of QTc-stretching for identifying LQTS.
Results: During maximal QT stretching, the T-wave morphologies that best discriminated LQTS from controls included: “notched,” “late-onset” and “biphasic” T-waves. These 3 categories were grouped into a category named “abnormal T-wave response to standing”. During quick standing, a QTc-stretched ≥ 490 msec increased the odds for correctly identifying LQTS. T-wave morphology changes provoked by standing were most helpful for identifying LQT2, less helpful for LQT1 and least helpful for LQT3. Conclusion: The sudden heart rate acceleration produced by abrupt standing, not only increases the QTc but also exposes abnormal T-waves that are valuable for diagnosing LQTS.
Keywords: Long QT syndrome; Electrocardiogram; QT interval; T-wave morphology
Abbreviations. CI = confidence intervals; ECG = electrocardiogram; LQTS = long QT syndrome; LQT1, LQT2 and LQT3 = LQTS type I, II and III, respectively; PPV and NPV = positive and negative predictive value.
2
I find it tempting to say that, when dealing with a possible case of long QT syndrome, one does not measure the QT interval, one looks at it. Peter J. Schwartz, 1989 1.
Timely diagnosis of congenital long QT syndrome (LQTS) is crucial because of the malignant nature of the disease and the existence of effective preventive therapy.2 However, accurate diagnosis is often difficult because >30% of the patients with genetically confirmed LQTS have QTc intervals within the range also seen in the healthy population.3 Consequently, many patients with suspected LQTS require additional tests for establishing or excluding the diagnosis. Such tests expose the abnormal response of the QT interval to spontaneous or provoked changes in heart rate throughout Holter recordings,4 exercise examinations5-8 or during epinephrine9-11 or adenosine12 challenge tests. We13,14 and others15,16 recently proposed the “quick standing” test as a simple bedside test that facilitates the diagnosis of congenital LQTS. The test takes advantage of the fact that as one stands up, the heart rate acceleration is abrupt while the associated QT-interval shortening is gradual.13,14 As the R-R interval shortens faster than the QT interval, the QT appears to “stretch” toward the next P wave and the corrected QT interval (QTc) for heart rate actually increases momentarily. The phenomenon of “QT stretching” is universal but is exaggerated in patients with LQTS, allowing for a simple but accurate diagnostic test.13,14 The duration of the QT interval is not the only diagnostic clue. Analysis of the T-wave morphology in the resting ECG of patients with LQTS not only increases diagnostic accuracy but also provides useful information regarding the specific genotype involved. Indeed, broad Twaves are common in LQT1 whereas notched T-waves are typically seen in LQT2 and lateonset peaked T-waves are characteristic of LQT3.17-20 Further characterization of T-wave morphology changes in response to epinephrine infusion was done by Ackerman’s group.10 They demonstrated that the appearance of specific forms of T-wave notching during low-dose epinephrine identifies patients with LQT2 with the test loosing specificity as higher epinephrine 3
doses are used.10 Finally, we reported specific T-wave patterns that predict acquired LQTS during bradyarrhythmias.21 Yet, there is no data on the effects of quick standing on T-wave morphology. We therefore performed the present analysis, to determine if the T-wave morphology changes provoked by the sudden acceleration triggered by standing aid in diagnosing LQTS or in determining the genotype.
Methods Population. The study cohort consisted of all the LQTS patients and controls included in the original studies of the “quick standing” test13,14 except for 6 patients for whom genetic confirmation remains elusive. In addition, all the LQTS patients studied in Tel Aviv since our last publication,14 all with genetic confirmation of the diagnosis, were included here, bringing the total of LQTS patients to 100. The control group was also enlarged to a matching number of 100 healthy individuals. “Obvious QT prolongation at baseline” was defined as QTc >470 msec for males and QTc >480 msec for females because such values are not observed in large cohorts of healthy adults.22
Abrupt standing test (QT stretching). The test protocol was previously described.13 Subjects rested for ≥5 minutes in the supine position as a baseline electrocardiogram (ECG) was recorded. They were then asked to stand up quickly and remain standing still for 5 minutes during continuous ECG recording. QT measurements and evaluation of T-wave morphology were performed at 4 points in time: 1) baseline (at the slowest sinus rhythm recorded during the resting period), 2) maximal heart rate (at the fastest sinus rate during heart rate acceleration provoked by standing), 3) point of maximal “QT stretching” (defined as the time when the end of the T-wave gets nearest to the next P wave due to R-R–interval shortening without sufficient QT-interval shortening)13 and 4) upon return of the heart rate to its baseline value (QTstunning)14 (the first R-R interval during heart rate deceleration that is within 40 msec of the 4
baseline R-R interval). At all these stages, the QT interval was corrected using the Bazett formula. For the present study, we focused on the effects of the abrupt heart rate changes provoked by quick standing on the morphology of the T-wave at each of the above-mentioned points in time. Two cardiologists (E.C. and O.H.) who did not participate in our previous studies13,14 and were blinded to patient allotting classified T-wave morphology changes by consensus. Our classification of T-wave morphology integrated the classifications proposed by Moss,17 Ackerman10 and our group21 for T-wave morphology at rest, during epinephrine test, and during bradyarrhythmias, respectively. The resulting classification is shown in Figure 1. Assessment of T-wave morphology was done independently in three groups of leads: (1) II, III, aVF, (2) V1-V3 and (3) V4-V6 (leads I and aVL were not recorded by the analog ECG recorder used for the initial tests performed in Tel Aviv). For each patient, T-wave abnormalities were counted in all 3 lead-groups separately, leading to a total of 300 patient-leads and 300 control-leads. Frequencies of T-wave abnormalities are presented as percentages from 100 patients (vs. 100 controls) and percentage from 300 patient-leads (vs. 300 control-leads). The study was approved by our institutional review committee.
Statistics. All data were summarized and displayed as mean ± standard deviation for continuous variables and as number (and percentage) of patients in each group for categorical variables. The p values for the categorical variables were calculated with the chi-square and fisher's exact tests. Continuous variables were compared using the independent sample t-test. In our original studies13,14 receiver operator characteristics curves identified cut-off values of QTc 423 msec at baseline and QTc of 489 msec during maximal QT stretching in response to standing, as the values that distinguished between LQT-patients and controls with a sensitivity of 90%. For the 5
present study, we rounded those numbers to a cut-off value of baseline QTc 425 msec and QTcstretched of 490 msec. We then used logistic regression to determine if T-wave morphology changes at baseline or provoked by standing improved to the diagnostic value of the test for correctly identifying LQTS. In this model, the Exp (ß) was used to estimate the odds ratio (OR) and 95% confidence intervals (CI) compared to the reference group of individuals with normal T wave and QTc-stretched
