OBJECTIVE To evaluate the results of permanent epicardial pacing in children. RESULTS All consecutive patients from one country (n = 119, period 1977-2009) undergoing permanent epicardial pacemaker implantation at <18 years of age (median 1.8 years, inter-quartile range 0.3-6.4 years) were studied retrospectively. Median patient follow up was 6.4 years (inter-quartile range 2.9-11.1 years); 207 generators, 89 atrial and 153 ventricular pacing leads were implanted. The probability of absence of any pacing system dysfunction was 70.1 and 47.2% at 5 and 10 years after implantation, respectively. Overall probability of continued epicardial pacing was 92.8 and 76.1% at 5 and 10 years, respectively, and increased in the recent implantation era (post-2000, P = 0.04). The use of steroid-eluting leads decreased the risk of exit block with a hazard ratio (HR) of 0.20 [95% confidence interval (CI) 0.09-0.44, P < 0.001)]. The use of bipolar Medtronic 4968 leads reduced the risk of surgical reintervention because of fracture, insulation break, outgrowth or exit block in comparison to the unipolar 4965 lead design (HR 0.12, 95% CI 0.04-0.40, P < 0.001). The AutoCapture™ feature (HR 0.08, 95% CI 0.02-0.36, P < 0.001) and steroid-eluting leads (HR 0.30, 95% CI 0.11-0.84, P = 0.021) decreased the risk of battery depletion. CONCLUSIONS The probability of continued epicardial pacing in children was 76% at 10 years after implantation, increased for implantation in recent years, and allowed transvenous pacing to be deferred to a significantly greater age. The use of bipolar steroid-eluting leads and of a beat-to-beat capture tracking feature significantly increased pacing system longevity and decreased the need for surgical reinterventions.