P wave signal averaged ECG and chemoreflexsensitivity in paroxysmal atrial fibrillation.

Detailed analysis of the QRS-complex and autonomic dysfunction can identify patients at risk to suffer from ventricular arrhythmias. To determine whether patients at risk for paroxysmal atrial fibrillation (PAF) could be identified while in sinus rhythm, a P wave triggered signal averaged ECG and an analysis of the autonomic function by chemoreflexsensitivity (CHRS) were examined. The ratio between the difference of RR intervals in the ECG and the venous partial pressure of oxygen before and after 5-min oxygen inhalation was measured for the determination of CHRS. We examined 224 patients (group A) who suffered from PAF, 250 patients (group B) without arrhythmic history and 30 young volunteers (group C). The filtered P wave duration (FPD) was significantly longer in group A than in group B (140.9+/-21.0 vs. 118.2+/-9.4 ms, p<0.0001) or C (105.2+/-14.1 ms, p<0.0001) while the root mean square voltage of the last 20 ms of the P wave (RMS 20) was significantly lower in group A than in group B (2.68+/-1.12 vs. 4.06+/-1.57 microV, p<0.0001) or C (3.97+/-1.36 microV, p<0.0001). Atrial late potentials (ALP) were defined as a FPD>120 ms and a RMS 20< or =3.5 microV. ALP could identify patients of group A with a specificity of 78% and a sensitivity of 83%. Patients with PAF (2.32+/-1.15 ms/mm Hg) showed a significantly lower CHRS than group B (4.14+/-1.58 ms/mm Hg, p<0.0001) or group C (4.98+/-1.51 ms/mm Hg, p<0.0001). The sensitivity for the presence of atrial fibrillation was 71% for a CHRS below 3.0 ms/mm Hg with a specificity of 70%. A combination of both methods showed a specificity of 85% and a sensitivity of 65% when ALP and pathological CHRS were present. The results of our study suggest that risk of atrial fibrillation could be detected by P wave signal averaged ECG and CHRS. An analysis of CHRS seems to be an appropriate method to demonstrate a neurovegetative imbalance, which might be one possible trigger mechanism.