SAT-061 THE COMPARATIVE IMPACT OF DIALYTIC MODALITIES ON ELECTRIC MYOCARDIAL REPOLARIZATION INHOMOGENEITIES

Abstract

Patients with end stage renal disease (ESRD) on dialysis have increased all cause and CV mortality, a significant proportion of this being represented by sudden unexplained (cardiac) death, most probably secondary to malignant ventricular arrhythmias. QT dispersion (QTd), defined as the difference in duration between the longest and the shortest QT interval on electrocardiogram (ECG), may be considered as a direct measure of the regional heterogeneity of myocardial repolarization and also an approximation for repolarization abnormalities. The aim of our study was to investigate the effect of the dialytic modalities on ECG abnormalities (QT interval, QT dispersion) and to find its important modulators, as markers of arrhythmia and sudden death in a large population of dialysis patients. We compared 68 HD patients (4 h/HD session, F100 dialysers); mean age 41.5 ±11.2 years; mean dialysis vintage 55.2 ± 38.3 months, with 63 PD patients (4 exchanges of 2000 ml/day, glucose/Na/Ca/K/Mg = 1.5/134/1.75/2/0.5 mmol/l), mean age 50.9 ± 15.2 years; mean dialysis vintage 34.7 ± 20 months. The QT interval was calculated as an average of three consecutive complexes, and corrected for heart rate using the Bazett's formula (QTc=QT/(R–R)1/2). Echocardiography was performed in all patients; Na+, K+, Ca2+, PO4, pH and BP levels were also determined. Follow-up was performed to analyze the overall survival and to determine the potential impact of inhomogeneities in myocardial repolarization and dialytic method on patient survival. The mean values for QTc and QTdc were 428.8 ± 34.1 ms and 39.3 ± 32.1 ms. Overall, QTc significantly correlated (univariate analysis) with the LVM and QTdc with serum ionized Ca and LVM (p < 0.05 for all). Stepwise multiple linear regressions showed Ca2+ to be the only independent predictor for QTdc. In CAPD patients, the mean value for QTc was 426.8 ± 40.7 ms (27% patients with abnormal QTc interval) and for QTdc = 43.2 ± 42.7 ms. In the HD subgroup, the mean value for QTc was 430.3 ± 27.2 ms (33.8 % abnormal QTc interval), p = NS vs CAPD; QTdc valuewas 35.7 ± 16.9 ms. In both dialysis groups (univariate analysis) QTc significantly correlated with LVM; QTd and QTdc only correlated with serum ionized Ca (p < 0.05). In Cox analysis model significant predictors for survival were gender (p=0.009), serum potassium (p = 0.007) and QTc (p = 0.005). The difference in QTc between HD and PD, unadjusted for confounders, was 14.9 % greater in PD vs HD patients (p = NS). QTc significantly correlated with LVM (r = 0.223, p = 0.039), QTd and QTdc correlated with serum ionized Ca2+ (r = 0.335, p = 0.000; r = 0.375, p = 0.000 respectively) and LVM (r = 0.240, p = 0.026; r = 0.23, p = 0.032 respectively). Stepwise multiple linear regression showed Ca2+ to be the only independent predictor for both QTd (adjusted R2 = 0.18; beta = 0.44) and QTdc (adjusted R2 = 0.19; beta = 0.45). Our study, the largest to date and with follow-up survival analysis, thoroughly describes the characteristics of CAPD and HD patients, supporting the lack of a significant different impact between these two dialytic modalities on ECG abnormalities. We also confirm the paramount influence of subjacent structural cardiac abnormalities on arrhythmia risk. Calcium represents probably the most relevant electrolyte for this pathology, in ESRD subjects on renal replacement therapy.