I was delighted to read the interesting study by Rosa et al.,1 in which the authors evaluated the prognostic value of the corrected global (mean of the 12-lead electrocardiogram [ECG] values) Tpeak-Tend interval (Tpeak-Tend) at 48hours from admission in 87 consecutive patients, aged 72±12 years, with tako-tsubo syndrome (TTS). The authors found that a Tpeak-Tend of>108ms was an independent predictor of subacute (beyond 48hours after admission) ventricular arrhythmias (VAs), defined as premature ventricular contractions ≥ 2000 within a 24-hour window of telemetry monitoring, ventricular fibrillation, sustained ventricular tachycardia (VT), polymorphic VT, and nonsustained VT. Such VAs, detected during a median of 8 days of hospitalization, were found to be associated with greater in-hospital mortality. The predictive performance of Tpeak-Tend was found to be superior to that of the standard corrected QT interval (QTc), currently used in patients monitored after TTS.1 The authors emphasize the advantages of employing all 12 ECG leads, rather than the limb or the precordial leads, in calculating the Tpeak-Tend; they also allude to the hypothesis that myocardial edema (ME) may be at the root of the repolarization aberrations and subacute VAs, via a re-entry or afterdepolarization mechanism due to delayed and dispersed ventricular apico-basal, interventricular and transmural repolarization gradients,1 as previously proposed.2–4 Unfortunately, cardiac magnetic resonance data to evaluate ME were available in only 21% of their patients, and thus the authors could not evaluate the relationship of repolarization aberrations (Tpeak-Tend and QTc) and VAs with the inflammatory ME.1

I would appreciate the authors’ response to the following points: a) the calculation of T-peak-Tend is labor-intensive; perhaps the global Tpeak-Tend can be electronically calculated, since many other ECG calculations (eg, QT) are currently automatically available upon ECG recording in many commercially available contemporary electrocardiographs; b) to evaluate the feasibility of such an undertaking, perhaps the authors could compare their QTc values, as manually measured, with those calculated by their ECG recording equipment (EL 280 Resting Electrocardiograph, Welch Allyn, United States), as has been done previously; c) although the authors carried out an impressively comprehensive analysis, using a very large array of variables,1 there is no information on the different morphological variants (ie, apical, mid-ventricular, basal/inverse, or focal) encountered in their 87 patients with TTS; d) the topography of ME would most probably be expected to differ in intensity in the apical and the basal/inverse TTS morphological phenotypes; indeed, in mid-ventricular TTS, it has been found to be more intense in the mid-lateral wall, with corresponding T-wave inversion/QTc prolongation confined to the lateral ECG leads3; e) accordingly, it would be of interest for the authors1 to explore whether the global Tpeak-Tend and QTc differed in patients with apical vs basal/inverse TTS; f) of even greater interst is the question of whether the Tpeak-Tend and QTc, derived from limb, or precordial, or individual (eg, lateral)3 ECG leads, were more prolonged in patients with apical, basal/inverse, or mid-ventricular TTS.