Should unipolar leads be implanted in the atrium? A Holter electrocardiographic comparison of threshold adapted...

The accuracy of atrial sensing plays a central role in dual chamber pacing. Recent Holter electrocardiographic studies showed a high incidence of atrial malsensing. We investigated the efficacy of bipolar atrial sensing at high sensitivity compared to threshold adapted unipolar sensing. One hundred consecutive patients with identical dual chamber pacemakers and bipolar atrial leads were investigated. Mean and individual range of 40 unipolar and bipolar telemetered atrial potentials were calculated; sensing threshold was determined by a semiautomatic sensing test. Oversensing was investigated with the help of a muscle provocation test. Twenty-four-hour Holter monitoring was performed at the highest bipolar sensitivity as well as at a unipolar sensitivity of half the measured sensing threshold. Mean atrial potential was significantly lower during bipolar mode compared to the unipolar sensing configuration, 3.66 +/- 1.75 versus 3.85 +/- 1.62 mV, P = 0.02. The bipolar atrial potentials showed a higher individual range than the unipolar signals, 2.44 +/- 2.62 versus 1.79 +/- 0.92 mV, P < 0.01. Sensing threshold did not differ significantly, 2.76 +/- 1.33 versus 2.67 +/- 1.29 mV. Mean oversensing threshold was 1.21 mV at unipolar configuration, whereas oversensing could not be provoked at a bipolar sensitivity of 0.5 mV. The incidence of atrial undersensing was significantly higher at threshold adapted unipolar sensing compared to bipolar sensing at highest atrial sensitivity, 35% versus 22%, P = 0.04. Oversensing did not occur at bipolar sensing, but was observed in 56% of patients at unipolar mode. Thirty-two percent of patients showed both atrial undersensing and oversensing at the unipolar sensing configuration. The muscle provocation test reached a sensitivity of 89% and a specificity of 95% in prediction of atrial oversensing during daily life. In conclusion, unipolar atrial potentials are more stable than bipolar ones. On the other hand, bipolar atrial sensing is less prone to the perception of myopotentials. Programming a high bipolar sensitivity significantly improves atrial sensing. Thus, bipolar leads should generally be implanted in the atrium.