AV conduction with atrial rate adaptive pacing in the bradycardia tachycardia syndrome.

AV conduction with atrial rate adaptive pacing (AAIR) during exercise was investigated in 43 patients (28 men, 15 female, mean age 68 +/- 7 years) who were paced and medicated with antiarrhythmic drugs for the bradycardia tachycardia syndrome (BTS). Patients were included if they had no second- or third-degree AV block, no complete bundle branch or bifascicular block, and a PQ interval < or = 240 ms during sinus rhythm at rest. The interval between the atrial spike and the following Q wave (SQ) was measured in the supine position at rest (R) with maximum AAI pacing rate (Fmax) achieved below the Wenckebach point (SQ-R-Fmax). Bicycle ergometry was performed using the Chronotropic Assessment Exercise Protocol, and AAI pacing rate was increased stepwise by programming load-adapted increments. Seven patients showed intrinsic rhythm during exercise. In those 36 patients who were atrially paced throughout ergometry (E), SQ was measured with 70 beats/min on the lowest CAEP stage (SQ-E-70) and with Fmax at maximum work load (SQ-E-Fmax). During exercise, no second-degree AV block was observed, but 28 of 36 patients (78%) showed a nonphysiological increase of the SQ interval, and the average SQ-E-Fmax was significantly longer than SQ-E-70 (250 +/- 31 versus 228 +/- 32 ms, P < 0.01). There was only a weak correlation between SQ-R-Fmax and SQ-E-Fmax (r = 0.35824, P < 0.05). When Fmax obtained during exercise was kept during recovery, 14 patients (39%) developed a second-degree AV block between 15 and 240 seconds after ergometry, 8 patients within 90 seconds. Patients who had exhibited a P on T wave in the ECG with Fmax at the end of exercise (11 of 36 patients) were reevaluated by Doppler echocardiography. Using the same exercise protocol and identical, load-adapted rate increments, only 3 of 11 patients showed premature mitral valve closure. It is concluded that patients paced and medicated for BTS are prone to a nonphysiological prolongation of AV conduction with AAIR pacing during and after exercise. As this risk can hardly be predicted by rapid atrial pacing at rest, the pacing system should be dual chamber in this subset of patients. This especially applies to the patients in whom mechanical AV timing is affected by the conduction delay.