Aim: Inappropriate therapy, due to poor discrimination of supraventricular tachycardia (SVT) from ventricular tachycardia (VT) remains a major problem in patients with an implantable cardioverter defibrillator (ICD). Theoretically, the addition of atrial sensing in discrimination algorithms should improve this differentiation. The aim of the study is to evaluate the performance of a new tachycardia discrimination algorithm, SMART Detection.
Methods and results: Twenty-six patients received a non-thoracotomy ICD system (Phylax AV, Biotronik, Germany). All documented spontaneous arrhythmia episodes were analyzed. During a mean follow-up of 8 months, a total number of 139 events with stored electrograms were recorded in 12 patients. The final diagnosis was ventricular fibrillation (VF) or polymorphic VT (n=20), monomorphic VT (n=69), SVT (n=26), other ventricular arrhythmia (n=3) and T wave oversensing (n=21). In 6 episodes a dual tachycardia was present. Considering SVT episodes, inappropriate therapy occurred in 2 cases of atrial flutter due to stable ventricular rate (<30 ms), 1 case of atrial tachycardia and 2 cases of sinus tachycardia due to a sudden onset (> 10%).
Conclusion: With the SMART Detection algorithm, discrimination of VT from SVT achieved a sensitivity of 100%, with an accuracy of 95.6% for all ventricular arrhythmias. In the case of SVT, the algorithm appropriately detected and inhibited therapy in 88% of atrial fibrillation.