The Old “VT vs SVT” Dilemma in a Patient with ICD


Dr Rajiv Sankaranarayanan MBBS MRCP(UK)
Cardiology Speciality Registrar in EP and Devices and BHF Clinical Research Fellow, Institute of Cardiovascular Sciences, University of Manchester, Oxford Road, Manchester

A 70 year old man with known ventricular tachycardia (VT) who had an ICD implanted, presented with palpitations. His ICD had not been activated to his knowledge. He was pain-free, not in cardiac failure clinically and his blood pressure was 110/70 mm Hg (normally 140/90). His ECG (Fig 1) showed a broad complex tachycardia at a rate of 115 bpm. However, his previous resting ECG showed a similar broad complex  rhythm (Fig 2), indeed apart from a slight increase in QRS duration (164ms vs. 198ms), the ECGs were morphologically identical. The fact that the ECGs were so similar and the ICD had not functioned, led the admitting team to diagnose supraventricular tachycardia (SVT) with mild rate related intra-ventricular conduction delay. However, there was no response to several doses of adenosine. In view of his haemodynamic stability, he was simply monitored on coronary care unit overnight.

Analysis of his ICD the following morning revealed that it had not diagnosed an arrhythmia because the rate was below the detection zone. Slow VT (rate 101-148 beats per minute) has a 30% incidence[1] in patients with ICD. We analysed his electrograms and performed an EP study through his ICD (demonstrated in the figures) to ascertain if the arrhythmia was in fact VT.

Click here to view figures.


  1. What does the electrogram (Figure 3) show?
  2. If we analyse the electrogram (Figure 3) by p-r logic, what does it show?
  3. We decided to do an EP study (Figure 4) through the device. What does this show?
  4. What does analysis of one of the failed ATPs (Fig 5) show? What could be the explanation?

Answers to the Questions and Discussion

The electrogram confirmed a tachycardia with 1:1 association between atrial and ventricular activity. It was unclear whether the atrial activity was retrograde, anterograde or reciprocating, but he had no prior history of SVT. On the other hand, he had never demonstrated retrograde conduction during previous episodes of VT either.

Analysis of the electrogram using p-r logic would indicate that this was an SVT as the V-A time was longer than the A-V time. PR logic algorithm has been shown to have a sensitivity of 100% and specificity of 67% in distinguishing VT from SVT [2].

Acceleration of the atrial rate by pacing from 561ms to 398ms led to disassociation of A and V activity and no change in V rate (Fig 4). We concluded that this was VT and activated the device’s pre-programmed ATP. This failed to terminate the VT.

During V pacing, Wenkebach conduction occurred in the retrograde direction. We took this as confirmation of VT and revised the ventricular ATP to a burst at a faster rate. This terminated the VT; during faster ATP there was no retrograde conduction. We concluded that the retrograde atrial activation had occurred on this occasion because of the much slower rate of VT, but the prolonged retrograde activation time reflected his poor retrograde conductivity.

We also concluded that previous ECGs were not always helpful in the diagnosis of tachycardia.


  1. Aliot E, Nitzsche R, Ripart A.Arrhythmia detection by dual-chamber implantable cardioverter defibrillators. A review of current algorithms. Europace. 2004 Jul;6(4):273-86.
  2. Sadoul N, Mletzko R, Anselme F, Bowes R, Schöls W, Kouakam C, Casteigneau G, Luise R, Iscolo N, Aliot E; Slow VT Study Group. Incidence and clinical relevance of slow ventricular tachycardia in implantable cardioverter-defibrillator recipients: an international multicenter prospective study. Circulation. 2005 Aug 16;112(7):946-53.

Click here to download the editorial as a PDF file.