Ventricular late potentials among thalassemia patients

Results Mean QRS duration increased from 89.23 (± 10.60) ms in 1997 to 94.27 (± 10.91) in 2004 ( p < 0.01), mean late amplitude signal (LAS) duration increased from 23.04 (± 7.68) ms in 1997 to 27.69 (± 6.82) ms in 2004 ( p = 0.01), whereas mean root mean square voltage RMS decreased from 80.85 (± 51.19) mV in 1997 to 45.12 (± 21.42) mV in 2004 ( p < 0.01). Changes in QRS duration and RMS voltage were found to be linearly correlated with average of ferritin over years ( r = 0.38, p = 0.03 and r = − 0.47, p = 0.01 respectively); and only 1 patient developed VLP over 7 years. Conclusion The incidence of VLP is 3.8% in the TM population over 7 years, despite the presence of significant changes in all SAECG criteria. RMS voltage and QRS duration changes over time seem to be related to iron overload measured by ferritin level. Keywords Thalassemia major Ventricular late potentials Iron overload Incidence Introduction Thalassemia is a congenital hemoglobinopathy leading to anemia because of impaired erythropoeisis and peripheral hemolysis. Thalassemia major (TM) patients are a subset of patients who are transfusion dependent since infancy, and subsequently develop iron overload related organ dysfunctions [1,2] . To date, iron induced cardiac dysfunction (pump and conduction abnormalities) remains the number one cause of death. Signal averaged ECG (SAECG) was suggested to predict ventricular tachycardia as the underlying substrate for up to 5% incidence of sudden cardiac death among TM patients. However, reports from Lebanon and Italy described a major difference in prevalence (3% and 31% respectively) of ventricular late potentials (VLP) among their TM populations [3,4] , with no clear reason for this difference. We sought to describe the incidence of VLP and changes in ECG and SAECG parameters among 26 TM patients (randomly selected from a group of 240 TM patients followed up regularly at the Chronic Care Center, Hazmieh, Lebanon in 1997) over a 7 year follow up period (1997 to 2004). The patients underwent signal averaged SAECG and regular ECG as described in earlier study [3] . In 2004 the same group of patients was requested to present for a follow up SAECG and had an echocardiography-Doppler performed [5] . Paired sample t -test was conducted to test for time difference in the SAECG and ECG values and independent t -test was conducted to test for association with gender. Statistical significance is considered when a p value is less than 0.05. Fourteen (53.8%) of the patients were females and none were lost to follow up. The average age was 23.0 ± 8.0 years at the end of the study period. One patient (3.8%) qualified by the ACC/AHA criteria to have developed VLP over this period of time. All the SAECG criteria had significantly changed over the observation period as depicted in Fig. 1 and gender did not seem to affect these changes ( Fig. 2 ). The mean noise level and mean heart rate were 0.38 ± 0.17 and 83.1 ± 8.8 respectively. Comparing the SAECG criteria to various echocardiography-Doppler parameters, only a trend towards positive significance between change in RMS and time velocity integral (Pearson r = 0.37; p = 0.07) was found. The annual ferritin of this patient fell well below the group's mean annual values. Studying the relationship of changes in ECG and SAECG parameters over time and average ferritin level showed that (assuming that there is no change in each of the SAECG for a zero ferritin level) only the changes in QRS (ΔQRS; ΔQRS = QRS (in year 2004) − QRS (in year 1997))and RMS (ΔRMS; ΔRMS = RMS (in year 2004) − RMS (in year 1997)) over seven years could be modeled by a linear correlation with the average ferritin level ( r = 0.38 and − 0.47; p = 0.03 and 0.01 respectively ( Fig. 3 a and b)). Iron overload mediated derangements in the heart are the focus of intense research currently since cardiac causes remain the number one cause of death in Thalassemia major patients [2] . In this study we showed over a 7 year follow up period that the three parameters to determine the presence of VLP (QRS duration, LAS duration, and RMS voltage) significantly changed in our cohort. Due to several limitations in our study design we cannot be certain in attributing these changes to iron overload alone. These limitations include that the cohort was in a growing age group and thus the chamber sizes of the left ventricle could have increased secondary to either an increase in body mass index, or because of the long standing (7 years) dilutional anemia effect. The relationship of left ventricle chamber sizes and SAECG parameters has been described in earlier studies showing significant correlation between them [3] . Alternatively, the role of iron overload in causing conduction delays in the thalassemic heart is well documented and thus extrapolating this effect to explain the prolongation in the QRS duration and LAS duration remains plausible [6] . This is further evident from the significant — albeit weak-linear correlation found between average ferritin level over 7 years and changes in RMS voltage and QRS duration. Unfortunately due to the low incidence of VLP in our cohort it is statistically not possible to conclude any effect of ferritin level–as a marker of iron overload–over the years on this event occurrence. This is not withstanding that recently most of the thalassemia literature undermines the role of ferritin in predicting iron overload and emphasizes cardiac MRI T2 ⁎ as an alternative specific method [7] . In conclusion, this study has demonstrated low incidence (3.8%) of VLP in Lebanese TM cohort over 7 years. The individual parameters of SAECG significantly changed over this period of time and the changes in QRS duration and RMS voltage are related to ferritin levels. However, a causal relationship between this change and iron overload independent of other factors can only be hypothesized due to absence of a matched control arm. References [1] C. Hershko E.A. Rachmilewitz Mechanism of desferrioxamine-induced iron excretion in thalassaemia Br J Haematol 42 1979 125 132 [2] A. Inati N. Zeineh H. Isma'eel S. Koussa W. Gharzuddine A. Taher Beta-thalassemia: the Lebanese experience Clin Lab Haematol 28 4 2006 Aug 217 227 Review [3] Isma'eel H, Taher A, Shamseddeen W, Khoury M, Gharzuddine W, Arnaout MS, Alam S. SAECG parameters and left ventricular chamber sizes: lesson from anemia conditions in thalassemia major patients. Int J Cardiol. 2006 Nov 18;113(3):E102-4. Electronicpublication 2006 Aug 2. [4] F. Franzoni F. Galetta C. Di Muro G. Buti F. Pentimone G. Santoro Heart rate variability and ventricular late potentials in beta-thalassemia major Haematologica 89 2 Feb 2004 233 234 [5] H. Dokainish Tissue Doppler imaging in the evaluation of left ventricular diastolic function Curr Opin Cardiol 19 5 2004 Sep 437 441 [6] C. Chrysohoou D.B. Panagiotakos Y. Barbetseas S. Brilli S. Lambrou M. Karagiorga V.A. Ladis C. Pitsavos C. Stefanadis Echocardiographic and electrocardiographic prognostic factors of heart failure in young patients with beta-thalassemia major: a 10-year (1995–2004) follow-up Int J Hematol 80 4 2004 Nov 336 340 [7] A. Kolnagou C. Economides E. Eracleous G.J. Kontoghiorghes Low serum ferritin levels are misleading for detecting cardiac iron overload and increase the risk of cardiomyopathy in thalassemia patients The importance of cardiac iron overload monitoring using magnetic resonance imaging T2 and T2 ⁎ Hemoglobin vol. 30(2) 2006 219 227