To assess the performance of endocardial pacemaker leads and to identify factors associated with structural lead failure, medical records of 2,611 endocardial pacing leads (in 1,518 patients) implanted between 1980 and 1991, having at least 1 month of follow-up, were reviewed. Leads without structural failure had normal function at the last follow-up date, or were discontinued for reasons other than structural failure (patient death, infection, dislodgment, lead-pacemaker incompatibility, operative complication, or abandonment by telemetry not related to failure). Leads with suspected structural failures were invasively or noninvasively disconnected because of clinical malfunction (loss of capture or sensing, oversensing, elevated thresholds, or skeletal muscular stimulation). Leads with verified structural failures met the criteria for suspected lead failure and also had a visible defect seen in the operating room or on chest roentgenograms, a change in the impedance interpreted by the physician as lead disruption, or a manufacturer's return product report that confirmed structural failure. Variables analyzed included patients' age and gender, paced chamber, venous access, insulation materials, fixation mechanism, coaxial design, polarity, and different lead models. The cumulative lead survival at 5 and 10 years were 97.4% and 92.9%, respectively, for suspected failures; and 98.7% and 97.3%, respectively, for verified failures. Leads in older patients (> or = 65 years old), and leads in atrial position had fewer verified failures (P = 0.014 and P = 0.007, respectively). Unipolar leads also tended to perform better according to the verified definition (P = 0.07).(ABSTRACT TRUNCATED AT 250 WORDS)