Unequal atrial stretch in dogs increases dispersion of refractoriness conducive to developing atrial fibrillation

We have shown previously that acute atrial dilation prolonged atrial refractoriness. We hypothesized that this increase in refractoriness might be heterogeneous and could create an electrophysiologic substrate leading to atrial fibrillation. The purpose of the present study was to test that hypothesis.

We studied 23 anesthetized open chest dogs. Bipolar plunge electrodes were placed in the medial free wall of the right atrium (thin region) and in the lower crista terminalis of the right atrium (thick region). Two bipolar plunge electrodes were also placed in the left ventricular apex to stimulate and record. Atrial effective refractory period (ERP) was measured in a group of nine dogs using the atrial extrastimulus method (A1A2) in two ways: during atrial pacing (AP) and during simultaneous atrioventricular (AV) pacing that achieved and AV interval of 0 msec (AV = 0). One liter/hour of normal saline was infused intravenously to elevate right atrial pressure and produce right atrial stretch. Atrial ERPs were measured before and after the normal saline infusion. To compare the extent of atrial stretch produced by volume overload, two pairs of sonomicrometer transducers were implanted in the thick and thin regions in a separate group of six dogs. The area encompassed by sonomicrometers was measured before and after saline infusion. The inducibility of atrial fibrillation was compared before and after saline infusion using rapid AP in another group of five dogs. Atrial pressure during sinus rhythm increased from 5.1 +/- 0.96 mmHg to 6.3 +/- 0.93 mmHg after normal saline infusion (P < 0.01). ERP increased in the thin free wall from 151 +/- 14.3 to 172 +/- 14.7 msec (AV = 0), and from 169 +/- 12.0 to 170 +/- 14.3 msec (AP). ERP increased in the thick crista terminalis from 134 +/- 9.9 to 147 +/- 10.2 msec (AV = 0), and from 133 +/- 7.9 to 146 +/- 9.8 msec (AV) (P < 0.01). The increase in ERP in the thin free wall exceeded that in the thick crista terminalis (P < 0.01), increasing the dispersion of atrial ERP. After 500-mL saline infusion for 30 minutes, the increase of area in the thin region was 12.8% +/- 3.7%, and that in the thick was 3.5% +/- 3.2%. The increase of the area in the thin region after 1000 mL for 1 hour was 18.8% +/- 6.2%, and that in the thick region was 6.3 +/- 5.1% (P < 0.01). Atrial fibrillation was not induced in any dog before infusion, but induced in all five dogs after saline infusion.

Atrial ERP in the thin right atrial free wall exceeds the ERP of the thick crista terminalis, and an increase in atrial pressure produced by saline infusion exaggerates this difference by stretching thin segments of the atrial myocardium more than it stretches thick regions. Thus, atrial stretch, by increasing the dispersion of atrial ERP, may be conducive to the development of atrial fibrillation.