DETERMINANTS OF VENTRICULAR ARRHYTHMIAS IN HEMODIALYSIS-PATIENTS - EVALUATION OF THE EFFECT OF ARRHYTHMOGENIC SUBSTRATE AND AUTONOMIC IMBALANCE

Background/Aims: In chronic hemodialysis patients, we evaluated determ inants of repetitive ventricular tachyarrhythmias which included late potentials and heart rate variability. Methods: We compared the presen ce of late potentials and heart rate variability obtained by ambulator y electrocardiogram (ECG), findings of echocardiography, and laborator y data between patients with and those without ventricular arrhythmias of Lown class 4A or 4B. Ambulatory ECG was recorded for 24 h from the beginning of hemodialysis. Heart rate variability was evaluated by th e standard deviation of the normal RR interval (SDNN). Results: Thirty patients (17%) had ventricular arrhythmias of Lown class 4A or 4B. Th ey were older than patients without such arrhythmias (p = 0.0021). Lef t-ventricular wall motion score (2.0 +/- 3.9 and 0.3 +/- 1.2, respecti vely, p < 0.0001) and left-ventricular mass index (167 +/- 59 and 140 +/- 44 g/m(2), respectively, p = 0.0053) were larger in patients with ventricular arrhythmias of Lown class 4A or 4B than in those without. Stepwise logistic regression analysis was performed to select variable s related to ventricular arrhythmias of Lown class 4A or 4B from the f ollowing 8 candidate variables; age, sex, presence of ischemic heart d isease, diabetic nephropathy as the primary renal disease, presence of late potentials, SDNN, left-ventricular wall motion score and left-ve ntricular mass index. Higher left-ventricular wall motion score (p < 0 .0001), older age (p = 0.0022) and male sex (p = 0.0235) were the vari ables associated with ventricular arrhythmias of Lown class 4A or 4B. Conclusion: In patients receiving hemodialysis, predominantly with chr onic glomerulonephritis, ventricular arrhythmias of Lown class 4A or 4 B were not associated with arrhythmogenic substrate revealed by late p otentials or autonomic dysfunction assessed by heart rate variability. Left-ventricular wall motion abnormalities, age and sex were signific ant factors.