Session SB1.4

Role of Hemodialysis in Atrial Fibrillation Onset: Preliminary Results from a Combined Computational and Experimental Analysis

S Severi*, G Fantini, C Corsi, A Vincenti, S Genovesi

Università di Bologna
Cesena, Italy

Atrial fibrillation (AF) is the most common and troublesome arrhythmia in clinical practice and is a significant contributor to cardiovascular morbidity and mortality. It has been demonstrated that in end stage renal disease (ESRD) patients AF prevalence is extremely high. Moreover, the hemodialysis (HD) session can trigger paroxysmal AF episodes. AF onset is determined by two important phenomena: a) extrasystolic firing from atrial ectopic foci, b) structural and electrical atrial remodeling. Supraventricular ectopic beat occurrence increases in the last stage of the HD session and structural remodeling is often present in ESRD patients. There are two electrophysiological aspects of electrical atrial remodeling: a) action potential (AP) shortening and consequent reduction of the atrial cell refractory period, b) slowing of intra-atrial electrical conduction. The ECG P wave duration reflects the atrial conduction velocity, whereas through computational analysis the effects of electrolytes variations on AP morphology and duration can be quantified. Here we present the preliminary results of a combined experimental and computational study aiming to highlight the role of hemodialysis-induced acute electrical remodeling on the AF onset.
In l1 ESRD patients, a signal averaged P wave recording was performed before and after the HD session. For each recording 100 to 200 beats were averaged, and P wave duration was automatically calculated. Heart rate and Na+, K+ and Ca++ plasma values were also measured. The Courtemanche model of human atrial myocyte was used to simulate the effects of dialysis on AP, by imposing extracellular electrolyte concentrations and heart rate to the values measured in vivo. Variations in the maximum upstroke velocity (Vmax) of the action potential, which is one determinant of conduction velocity, was compared with variations in experimentally measured P wave duration.
The P wave duration slightly but systematically increased after dialysis (131±9 vs 125±12 ms, p<0.05), indicating a dialysis-induced slowing of intra-atrial conduction. Coherently, the model-based analysis indicated a similar reduction in the maximum AP upstroke velocity. The simulated action potential duration (APD) was also reduced at the end of dialysis (279 vs 298 ms) but it is worth to note that the Courtemanche model does not consider time-varying extracellular ion concentrations. In particular the dependence of IKr and IK1 potassium currents on extracellular K+ concentration is neglected. Incorporation of such dependence, as described in most ventricular AP models, resulted in an almost constant APD (311 vs 306 ms, +1.6% end vs beginning of dialysis).
These results indicate slowing of the intra-atrial conduction as a possible consequence of hemodialysis-induced changes in electrolyte concentrations, whereas eventual shortening of the refractory period should be further investigated.

(Abstract Control Number: 120)