A multicenter study evaluated the performance of atrial and ventricular unipolar leads with porous steroid-eluting and platinized grooved electrodes. A total of 563 leads were implanted in 451 patients. These included 311 ventricular and 97 atrial steroid-eluting electrodes; and 112 ventricular and 43 atrial leads with platinized electrodes. Mean follow-up was > or = 1 year. At implant there were no significant differences in threshold parameters in either chamber. Chronically, however, the steroid-eluting lead consistently had significantly lower pacing thresholds in both chambers. For example, after 360 days implant, steroid-eluting electrodes had 0.23 +/- 0.10 msec ventricular thresholds at 0.8 V compared to 0.45 +/- 0.3 msec in the platinized group (P < 0.0001). In the atrium, the steroid-eluting lead's 6-month thresholds at 0.8 V were 0.15 +/- 0.06 msec compared to 0.9 +/- 0.8 msec for the platinized electrode (P < 0.01). The chronic ventricular QRS amplitudes were significantly greater for the steroid-eluting electrode (P < 0.0005). There were no significant differences in atrial sensing and no incidence of atrial undersensing in the study. The low and consistent thresholds of the steroid-eluting electrodes would have permitted pacing in the ventricle at < or = 2.5 V without compromising safety factor in 99.4% of the patients. The other 0.6% required 5 V temporarily. In the atrium, 100% of the patients could have been paced safely at reduced output. In spite of this, 63% of the implanters lacked the confidence to use reduced outputs.