Europace Advance Access originally published online on May 1, 2008
Europace 2008 10(6):726-728; doi:10.1093/europace/eun113
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ICDs
Inappropriate implantable cardioverter defibrillator shocks in fractured Sprint Fidelis leads associated with ‘appropriate’ interrogation
University of Ottawa Heart Institute, Canada
Manuscript submitted 27 March 2008. Accepted after revision 31 March 2008.
* Corresponding author. Tel: +1 6138824670.E-mail address: djfarwell{at}bigfoot.com
 |
Abstract |
|---|
 Top Abstract IntroductionDiscussionReferences |
|---|
We present two patients with fractures within the pace–sense circuit of their Medtronic Sprint Fidelis leads who received inappropriate shocks from their Medtronic defibrillators during device interrogation. This was not simply a coincidence, but due to electromagnetic interference induced within the Sprint Fidelis lead by the device programmer during two-way communication with the defibrillator. Our subsequent investigations have uncovered at least two other similar incidents in Canada. We have also discovered that the Medtronic ‘Auto-resume’ feature may leave future patients uniquely vulnerable to such inappropriate shocks in the future.
Key Words: Implantable cardioverter defibrillator, Implantable cardioverter defibrillator lead, Inappropriate shocks, Electromagnetic interference
|
Introduction |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
Sprint Fidelis (models 6930, 6931, 6948, and 6949) is a 6.6 Fr bipolar high-voltage implantable cardioverter defibrillator (ICD) lead (Medtronic Inc., Minneapolis, MN, USA). The lead was approved by the US FDA in September 2004 and to date
268 000 leads had been implanted world-wide. Concerns regarding the early fracture rate of the lead were first reported in April 2007.1
On 15 October 2007, the manufacturer suspended the distribution of Sprint Fidelis leads.2 We report two patients implanted with Sprint Fidelis 6949 leads, presenting with a suspected lead fracture, who received inappropriate shocks during appropriately conducted interrogation of their devices.
Case 1
Patient 1, a 54-year-old woman with a dilated cardiomyopathy (left ventricular ejection fraction 30%, New York Heart Association II), had received an EnTrust single-chamber ICD implanted 26 months earlier for primary prevention. The device had never delivered any therapies. She presented after hearing an audible alarm from her device. In the device clinic, the programmer head was positioned over her device and the ‘find patient’ option was selected. During initial interrogation, the patient received a shock. When the interrogation was completed, all therapies were immediately suspended. The Interrogation indicated the presence of a lead fracture [sudden right ventricular (RV) impedance rise to >3000 
, 1274 short v–v intervals]. The analysis of electrograms demonstrated that the shock was inappropriate, caused by electrical artefact ‘noise’ (Figure 1). The ‘noise’ was only present during the initial interrogation period and not during subsequent programming of the device. The ‘noise’ was not reproduced by pocket manipulation. A chest radiography showed the lead pins to be well seated.
|
Device interrogation was repeated immediately before the implantation of a new ICD lead. With the patient lying still on the operating table, the programmer head was placed over the ICD (therapies off/detections on) with the device software already loaded. Normal device telemetry was observed with no evidence of oversensing. The device interrogation function was executed. A period of oversensing was observed
3 s after the initiation of device interrogation, lasting until interrogation was complete. This observation was reproduced precisely on multiple occasions with the patient lying still and the programmer head simply resting on her chest. No oversensing was observed during the remainder of programmer/ICD telemetry communication. The ICD was then explanted. Inspection and lead manipulation during explanation showed the lead pins to be well seated and set screws were tight. The lead was detached from the device and intracardiac electrograms were recorded (RV tip/ring), demonstrating low-voltage high-frequency noise. Other electrograms (HVB/RV ring) remained sharp, implying fracture within the RV tip electrode or the conductor. A new lead was implanted and the post-procedural chest X-ray clearly demonstrated fracture of the RV tip conductor, which had been previously concealed behind the device (Figure 2).
|
Case 2
Patient 2, a 57-year-old man with coronary artery disease and prior myocardial infarction, had been implanted with an EnTrust single-chamber device 30 months earlier following an episode of sustained VT. He presented after being awoken by a single shock from his device. No alarm tone had been heard. The programmer head was positioned over the device and the ‘find patient’ function was executed. Device interrogation commenced automatically. Almost immediately the patient received three shocks in rapid succession. During the interrogation, clinic staff were locked out of the programmer and thus unable to use the suspend therapy option. Device data and electrogram analysis demonstrated inappropriate therapies caused by lead fracture (sudden rise in lead impedance to >3000
, 2766 short v–v intervals). Shocks occurred because of a ‘noisy’ signal during programmer interrogation. Chest radiography showed the lead pins to be well seated, with no evidence to suggest a loose set screw.
Patient 2 was scheduled for implantation of a new ICD lead. With the patient lying still, the programmer head was placed over the ICD (therapies off/detections on) with the device software already loaded. Normal device telemetry was observed. The device interrogation function was executed. Once again a period of oversensing was observed 3 s after the initiation of device interrogation, lasting until interrogation was complete. This observation was reproducible. No oversensing was observed during the remainder of programmer/ICD telemetry communication. The ICD was then explanted. Inspection showed the lead pins to be well seated and set screws were tight. The lead was detached from the device and intracardiac electrograms were recorded (RV tip/ring), demonstrating low-voltage high-frequency noise similar to that observed in Patient 1. Other electrograms (HVB/RV ring) remained sharp, implying fracture within the RV tip electrode or conductor. Right ventricular tip/ring intracardiac electrograms were also recorded during device interrogation, with the device juxtaposed to the exposed lead and only low-voltage noise, as before, was observed. The fractured lead was not extracted. The site of the fracture was not visible on radiography.
|
Discussion |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
We report two cases of Sprint Fidelis lead fracture who received inappropriate shocks because of sensing of electrical noise artefacts during device interrogation. We suspect the underlying mechanism to be due to electromagnetic interference (EMI) provoked by active device/programmer communication. To investigate further, we placed a Medtronic analyser over a surface ECG lead and executed the analyse function. Electromagnetic interference was observed at a characteristic cycle length of
200 ms (Figure 3). A similar frequency of interference was observed in the electrograms prior to the inappropriate shocks in both patients (Figure 1). We understand that this frequency is similar to the carrier frequency used by the Medtronic programmer.
|
In a previous case report, Pickett et al.3
observed inappropriate shocks in a Medtronic Marquis device during interrogation. They suggested that devices may become sensitized to EMI by an unbalanced input. We propose that, in our patients, a relatively proximal fracture of the tip electrode conductor led to an unbalancing of EMI produced by the normal functioning of the Medtronic programmer (Figure 4). In our first case, the fracture was clearly proximal. We do not know the site in the second patient.
|
This problem is compounded by the Auto-resume feature of Marquis and newer Medtronic ICDs. This function was designed for safety reasons to ensure that therapies would not be withheld during the potentially long interrogation process. However, unless therapies are suspended prior to interrogation (using the suspend function of the analyser), detection will be resumed during interrogation even with a donut magnet in place. Furthermore, the use of the ‘find patient function’ or positioning the programmer head over the device, before the correct software has fully loaded, will result in automatic initiation of interrogation before the suspend option can be used.
It is our understanding that if we had followed the following interrogation protocol, we might have avoided inappropriate therapies:
- Keep programmer head distant from patient.
- Fully load device model-specific software onto Medtronic programmer.
- Place programmer head over the device.
- Immediately suspend therapies before interrogating device.
- Take care to avoid losing telemetry communication with the device during interrogation.
Medtronic data suggest that 90% of 6949 fractures involve the pace–sense circuit and that of these, 40% are proximal.4 We therefore estimate that 36% of 6949 fractures may be vulnerable to this lead–programmer interaction. We contacted all 21 ICD implantation centres in Canada. About 6181 Fidelis leads have been implanted in Canada with 80 failures reported up to 20 December 2007. In four of 80 cases (our two and two others of which we know), patients received inappropriate shocks during device interrogation.
As fractures with these leads will continue to occur and as the first action by device care centres will be to interrogate the device, we consider this information to be of some urgency. We have contacted the manufacturer and urged them to notify physicians of this potentially harmful interaction and to issue formal recommendations to try to prevent it.
Conflict of interest: The authors (D.B., R.L., M.H.G., and M.S.G.) have received research support from Medtronic, Boston Scientific, and St Jude Medical. D.B. is on a Medtronic Speaker's Bureau and M.S.G. has received speaker's honoraria from Medtronic and Boston Scientific.
|
References |
|---|
|
TopAbstractIntroductionDiscussionReferences |
|---|
[1] Hauser RG, Kallinen LM, Almquist AK, Gornick CC, Katsiyiannis WT. Early failure of a small-diameter high-voltage implantable cardioverter-defibrillator lead. Heart Rhythm (2007) 4:892–6.[CrossRef][Web of Science][Medline]
[2] Patient management recommendations for Sprint Fidelis® leads. (2007) http://www.medtronic.com/fidelis/appendixC.html.
[3] Pickett RA III, Saavedra P, Ali MF, Darbar D, Rottman JN. Implantable cardioverter-defibrillator malfunction due to mechanical failure of the header connection. J Cardiovasc Electrophysiol (2004) 15:1095–9.[CrossRef][Web of Science][Medline]
[4] Medtronic. Sprint Fidelis physician letter. March 2007.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. Goette, F. Cantu, L. van Erven, P. Geelen, F. Halimi, J. L. Merino, J. M. Morgan, and on behalf of the Scientific Initiative Committee o Performance and survival of transvenous defibrillation leads: need for a European data registry Europace, January 1, 2009; 11(1): 31 - 34. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||