The aim of this study was to identify clinical characteristics that distinguish patients with high DFTs and assess the prognostic implication. DFTs testing is a lengthy, potentially painful, and a hazardous process. Little information is available concerning the identification of patients with high DFT who undergo ICD surgery with transvenous leads and biphasic energy. This study analyzed 968 patients from two separate clinical studies who received a Medtronic cardioverter defibrillator from January 1995 through November 1999 and who had DFT testing measured by a binary search protocol. Compared to 865 patients with low defibrillation thresholds (< 18 J), the 103 patients with high thresholds (> or = 18 J) had a lower LVEF (34 +/- 16.7 vs 38.3 +/- 16.2%, P = 0.01), a worse NYHA functional class (23% Class I, 43% Class II, 29% Class III, 5% Class IV vs. 27% Class I, 55% Class II, 17% Class III, 1% Class IV, P < 0.0001), had bypass surgery less often (10.7 vs 27.5%, P < 0.0001), used amiodarone within the past 6 weeks (42.7 vs 27.2%, P = 0.002), and had a history of ventricular fibrillation more often (44.7 vs 33.1%, P = 0.02). Information concerning the number of shocks delivered was available in 345 (35%) patients; 23 were in the high DFT group and 322 were in the low DFT group. Twelve (52%) of the 23 patients in the high DFT arm received 3.6 +/- 2.7 shocks (median 2.5) and 106 (33%) of the 322 patients with low DFT received 4.9 +/- 9.5 shocks (median 2). After 6 months the mortality rate of patients with high thresholds was 11.7 vs 7.8% in patients with low thresholds (P = 0.118). Using a multivariate logistic regression model the significant predictors of death were older age, higher NYHA class, lower LVEF, amiodarone use, had a presenting arrhythmia of ventricular fibrillation and CHF but not initial high defibrillation thresholds. The study found that (1) 11% of patients have high DFTs, (2) clinical characteristics that identify high defibrillation thresholds are NYHA Class III, IV, low ejection fraction, no previous history of bypass surgery, prior amiodarone use preoperatively, and presenting with ventricular fibrillation, and (3) while high DFTs were associated with a more ill patient population, there was no difference in survival in a 6-month follow-up. Patients with a predicted low DFTs may be eligible for abbreviated ICD testing while high risk patients require formal testing.