The purpose of this study was to evaluate the effectiveness and safety of temporary VDD pacing using an esophageal electrode for sensing of the atrial electrogram. We studied 15 patients, 8 men and 7 women, aged 77 +/- 2 years (mean +/- SE, range 61-90), with severe atrioventricular (AV) conduction disturbances. A 24-hour beat-to-beat ECG analysis was used to evaluate the effectiveness of the pacing system and special tests were performed to test the stability of pacing and sensing. The system performed satisfactorily in 12 of the 15 patients. The 24-hour Holter ECG monitoring revealed the following percentages of beats: 96.32 +/- 0.5 VDD, 2.92 +/- 0.6 VVI, and 0.14 +/- 0.05 paced beats resulting from pseudosensing. All the latter were single, with no bigeminy or salvos. The results of the stability tests were as follows: the percentage of VDD beats was significantly lower than the 24-hour mean when the patient lay on his right side (92.8 +/- 0.5, P less than 0.001), during the swallowing of liquids (91.26 +/- 0.4, P less than 0.001) and soft foods (84.2 +/- 1.4, P less than 0.001), and during coughing (94.2 +/- 0.6, P less than 0.001). The percentage of VVI type beats increased in these four cases (6.7 +/- 0.5, 7.2 +/- 0.3, 13.2 +/- 1.2 and 4.8 +/- 0.4, respectively, P less than 0.001 in each case). The percentage of ectopic beats due to pseudosensing did not change significantly during any of the tests. These results indicate that the method described is a safe and effective technique for temporary VDD pacing.