Comparison of programmed stimulation and Holter monitoring for predicting long-term efficacy and inefficacy of amiodarone used alone or in combination with a class 1A antiarrhythmic agent in patients with ventricular tachyarrhythmia

The values of two Holter ambulatory electrocardiographic monitoring criteria and one programmed stimulation efficacy criterion reported to be predictive of the efficacy of amiodarone were compared in 70 patients taking amiodarone for sustained ventricular tachyarrhythmias. At baseline, all patients had ventricular tachycardia inducible by programmed stimulation. After amiodarone loading (935 ± 271 mg for 16 ± 7 days), efficacy was determined by a programmed stimulation criterion (ventricular tachycardia no longer inducible or ≤15 beats) and two Holter monitoring criteria (Holter I = ≥85% reduction of ventricular premature complexes and abolition of couplets and triplets in 64 patients who had ≥ 10 ventricular premature complexes/h or couplets or triplets or both before therapy; Holter II = abolition of triplets in 41 patients who had triplets before therapy). Amiodarone was effective in 12 of 70 patients by the programmed stimulation criterion, in 49 of 64 patients by Holter criterion I and in 37 of 41 patients by Holter criterion II. In assessing efficacy of amiodarone, programmed stimulation and Holter criteria were discordant in 69% of patients or more (p < 0.001). There were 16 recurrences or sudden deaths during the entire follow-up period (19 ± 19 months). Arrhythmia-free survival rates at 24 months of patients with efficacy and inefficacy by each criterion, respectively, were 90 and 78% by programmed stimulation, 84 and 62% by Holter criterion I (p < 0.05) and 73 and 50% by Holter criterion II (p < 0.05). At 24 months, sensitivities of programmed stimulation and Holter I and II criteria were 92, 42 and 18%, respectively; specificities were 17, 86 and 94%, respectively, and predictive accuracies were 43, 71 and 63%, respectively. In patients with discordance between Holter monitoring and programmed stimulation findings, the specificity of programmed stimulation was less than 10% and the sensitivity of Holter monitoring was less than 20% at 12 months. Conclusions: 1) Treatment with amiodarone in a significant number of patients cannot be managed by Holter monitoring. 2) Inefficacy by Holter criteria predicts poor outcome but efficacy does not preclude poor outcome (insensitive). 3) Inefficacy by programmed stimulation does not preclude good outcome (nonspecific). 4) Many patients have inefficacy by programmed stimulation despite efficacy by Holter criteria; this is due to both insensitivity of Holter monitoring and nonspecificity of programmed stimulation. 5) Further studies should be conducted to identify predictors of efficacy of amiodarone in patients with discordant results.