Comparison of long-term outcomes of patients treated with nonthoracotomy and thoracotomy implantable defibrillators.

In 193 consecutive patients treated with implantable defibrillators at our institution, thoracotomy approaches were used in 87 patients and nonthoracotomy approaches in 106 patients. Long-term outcomes of the 2 groups were compared by the intention-to-treat analysis. Surgical mortality (30-day mortality) rates were 5.7% in the thoracotomy group and 0% in the nonthoracotomy group. Six of 106 patients who underwent nonthoracotomy implantation had a high defibrillation threshold and did not receive nonthoracotomy defibrillators. The duration of follow-up was 52 +/- 31 months in the thoracotomy group, and 23 +/- 15 months in nonthoracotomy group. Actuarial survival rates at 6 and 24 months were, respectively, 90% and 81% in nonthoracotomy patients and 89% and 80% in thoracotomy patients (p = NS). In patients with left ventricular ejection fraction <30%, surgical mortality was 0% by the nonthoracotomy and 10% by the thoracotomy approach. Despite the 10% difference in 30-day mortality, survival rates at 6 months were 85% in nonthoracotomy patients and 81% in thoracotomy patients. At 24 months they were 73% in nonthoracotomy patients and 74% in thoracotomy patients. Thus, this nonrandomized study suggests that while short-term survival is better in nonthoracotomy patients than thoracotomy patients, the difference in survival diminishes quickly during the first few months and disappears by 6 months. The results were similar in patients with severe ventricular dysfunction. Several important implantable-cardioverter defibrillator (ICD) trials initially utilized thoracotomy ICDs. Although questions may be raised with regard to applicability of such a trial in the era of nonthoracotomy ICDs, this study suggests that the results of such ICD trials will be largely applicable to patients treated with nonthoracotomy ICDs.