icd system high voltage component failure (presented to fda)
DESCRIPTION
Discussion of mechanism of failure of ICD systems to deliver shocks. Detailed attention to St. Jude ICD leads.TRANSCRIPT
ICD Lead High-Voltage Component Failure
How to Identify ICD Leads and Systems That Will Not Deliver Shocks
Edward J. Schloss MD FACC FHRS
Presented to FDA
October 14, 2014
Disclosures
• Medtronic: Consultant, Speaker
• Boston Scientific: Consultant, Speaker
Components of ICD System
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
ICD Can
Low Voltage Function - Sensing
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
ICD Can
Low Voltage Function - Pacing
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
ICD Can
High Voltage Function
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
ICD Can
High Voltage Failure is Tough to Detect
• Impedance is only measured parameter– No sensing/noise alerts, no pacing threshold
• Testing amplitude is order of magnitude less than working stress on lead
• Failure may be asymptomatic and go undetectedfor years
• First presentation may be sudden cardiac arrest or death– High voltage failure is “quiet”– May go undiagnosed if no postmortem interrogation
How does HV Failure Present?
• Silent failure, picked up with remote notification
How does HV Failure Present?
• Silent failure, picked up with remote notification
• Silent failure, picked up in office
How does HV Failure Present?
• Silent failure, picked up with remote notification
• Silent failure, picked up in office
• Silent failure, seen w/hospital shock testing
“annual DFT testing, once a common clinical practice is performed infrequently today”
JACC Aug 2008
How does HV Failure Present?
• Silent failure, picked up with remote notification
• Silent failure, picked up in office
• Silent failure, picked up w/hospital HV testing
• Clinical Shock with manifest failure
– Inappropriate
How does HV Failure Present?
• Silent failure, picked up with remote notification
• Silent failure, picked up in office
• Silent failure, picked up w/hospital HV testing
• Clinical Shock with manifest failure
– Inappropriate
– Appropriate
Lee Oukrup with portrait of son Joshua www.nbcnews.com
How Do We Detect HV Component Failure?
• What can fail?
– Pulse Generator
How Do We Detect HV Component Failure?
• What can fail?
– Pulse Generator
– Generator/Lead Interface – i.e. header
How Do We Detect HV Component Failure?
• What can fail?
– Pulse Generator
– Generator/Lead Interface – i.e. header
– Lead
• Weakest Link in the Chain
• Types of Failures
– Conductor
– Insulator
Shock Lead Impedance Testing
• Pulse delivered automatically every 21 hours– Older generation devices 15 mA for 60 μsec
– Newer generation devices 80 μA for 156 μsec (10 Ω higher)
• Red Alert in Latitude if >125 Ω or <20 Ω(programmable)
• Measures each TRIAD vector separately and combines to single value
Shock Lead Impedance Testing
• Pulse delivered once a week
– 4V for 20 msec, Coil to can
• Normal defined as <400 Ω
• Abnormal will show up as alert on interrogation
• With shock, normal is <200 Ω
Shock Lead Impedance Testing
• Pulse delivered daily (up to 4 attempts)– 400 mV for 90 msec
• Two vectors reported– HV Impedance = RV coil to SVC coil+Can
– SVC Impedance = RV coil to SVC coil
• Alarm for Impedance < 20Ω, > 200Ω (programmable)
• Triggers clinician and patient alert (not LIA)
Shock Lead Impedance Testing
• Atlas/Epic have no daily measurements
• Physician generated HVLI test (12v)
• Later Generations HV pulse delivered daily
– All vectors (RV and SVC to can)
• Alert impedances are programmable
Evidence Base for Shock Lead Integrity Testing
Canine ModelLooked intact and disrupted (high Ω) lead configurationsPacing pulses and shocks measured over wide range of impedances from normal to highClose correlation suggests low voltage measurements predict high impedance failure
Did not test for low impedance failure
Evidence Base for Shock Lead Integrity Testing
Validation of newly developed HV testing technique29 pts with Guidant PRIZM ICDs at implantAll pts got 0.4 μ J pulse, 1.1 J, and 16-45 J shocksImpedances measured for all pulse strenthsLow energy pulses correlated with high energy shocks
Only new functional leads were tested
Evidence Base for Shock Lead Integrity Testing
"Partial insulation defects may not be identified by low-energy pulses that deliver insufficient current to activate the shorted high-output protection feature."
Swerdlow/Friedman Pacing Clin Electrophysiol. 2006;29(1):70-96.
Response to Failed Shock
Conductor Failure – High impedance
– Cable fracture, Set screw issue
• Fidelis
– Current delivery attenuated to affected limb of circuit
• RV coil likely ineffective shock
• SVC coil likely effective shock unless high DFT
– Device will be unharmed and report out a high impedance alert
Response to Failed Shock Insulator Failure – Low impedance
– Abrasion: can/coil , coil/coil, Header short• Riata, PRIZM 2 DR
– HV Current pulse generator as a short• Older devices: current shunted from heart , electrical
overstress may damage PG
• Newer devices: current shunted, PG protected.
– If PG is still intact, low Ω alert issued.
– St. Jude Dynamic Tx• Reconfigures shock to remove SVC coil or can from
circuit if breach detected.
HV Component Failure by Lead Model
• Evolution to cable design associated with improved lead performance
www.bostonscientific.com
Medtronic Transvene
Ellenbogen et al. JACC Jan 1, 2003
Medtronic Fidelis
High voltage performance not broken out in PPR data
Medtronic Fidelis
• HV Failures: 5% of overall failures in series
Circulation. 2011;123:358-363
Medtronic Fidelis
• HV Failures: 5% of overall failures in series
• HV Failures: 6.1% of overall failures in RPA
Medtronic Fidelis
• HV Failures: 5% of overall failures in series
• HV Failures: 6.1% of overall failures in RPA
• Conductor fracture (abrupt >100 Ω) in all RPA
Medtronic Fidelis
Medtronic Field Communication April 2011
• HV Failures: 5% of overall failures in series
• HV Failures: 6.1% of overall failures in RPA
• Conductor fracture (abrupt >100 Ω) in all RPA
• Failure of pace-sense electrode predicts failure of high voltage conductors
Medtronic Field Communication April 2011
Medtronic Fidelis
Conductor fracture,New pace/sense lead 5076
SVC cable fracture
RV cable fracture
St. Jude Riata/Riata ST
WSJ.com
Cardiobrief.org
Riata High Voltage Shock Delivery
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
ICD Can
Riata HV Insulation Shorts
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
z12
ICD Can
3
3
RV HV cable/coil
shorts to
SVC cable/coil or can
Riata HV Insulation Shorts
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can
z1
ICD Can
Programming the can out of the system may mitigate (if dual coil lead)
Riata HV Insulation Shorts
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil
2
ICD Can
Programming SVC Coil out of system may mitigate failure
Riata HV Insulation Shorts
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to can
ICD Can
3
3
Other variations of #3 exist with alternative return circuit
Detection of Riata/ST Failures
• Earlier generation ICDs (Atlas/Epic) do not have automatic HV testing
High voltage testing with Atlas/Epic must be performed manually
Detection of Riata/ST Failures
• Earlier generation ICDs do not have automatic HV testing (Atlas/Epic)
• Automated HVLI testing in later generations
St. Jude Medical. Riata Lead Summit 1/12
Detection of Riata/ST Failures
• Earlier generation ICDs do not have automatic HV testing (Atlas/Epic)
• Automated HVLI testing in later generations
St. Jude Medical. Riata Lead Summit 1/12
St. Jude Medical. Riata Lead Summit 1/12
St. Jude Medical. Riata Lead Summit 1/12
Riata/ST High Voltage FailureWhat do the large trials tell us?
Trial #patients #/% ElectricalFailures
#/% of electrical that are HV
VA1 1403 47 (3.4%) 3 (6.3%)
Canadian2 4358 201 (4.6%) ?? >= 1 HV failure
US Multicenter3 1081 67 (6.2%) 3 (4.4%)
St. Jude Riata LES4 776 10 (1.3%) 5 (50%)
1. Sung et al. Heart Rhythm 2012 9:1954-19612. Parkash et al. Heart Rhythm 2013 10: 692-6953. Abdelhadi et al Heart Rhythm 2013 10: 361-3654. Hayes et al. Heart Rhythm 2013 10: 1778-1782
50% of detected failures were in HV circuit
Heart Rhythm 2013 10: 1778-1782
• Prospective 782 pts: followed avg 9.8 months • Baseline fluoroscopy and serial electrical testing• Average lead age--Riata 6.5 yr, Riata ST 4.8 yr• “Potential” HV failure defined as
• HVLI change >25 Ωor
• HVLI >125Ω or <20Ω • 5 of 10 adjudicated electrical failures were HV• Additional 4 leads had HVLI change >25 Ω but
adjudicated as normal function
Danish Riata Registry
• 295 mature Riata Leads (5.1 +/- 1.1 yr)
• Baseline fluoro, prospective F/U (1.1 yr)
• 6.5% electrical abnl (21.9% w/EC, 5.3% wo/EC)
Heart Rhythm, In Press
- Low incidence of HV failures (high impedance)
- No instances of HV insulation (low impedance) failure detected
- No low impedance pace failures in EC group
Only 51% of leads had high voltage testing performed
How Well are we Detecting Riata High Voltage Failures?
• 77 year old man, ischemic cardiomyopathy• EPIC DR ICD with Riata 1580, externalized, No HVLIs• Gen change to Fortify DR. HVLI normal at 39 Ω• VF induced with 1 J T shock• 20 J shock failed, external defib, ICD system error• HVLI repeated, still normal at 35 Ω• 20 J shock failed, external defib, ICD system error• HVLI now <10 Ω, no sensing issues
Short evident between SVC coil and RV shock cable
Doshi Case - Riata HV Insulation Short #2
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
RV HV cable to SVC coil short
2
ICD Can
Lead failure repeatedly not evident by low voltage testing
Tip of the Iceberg?
• 3 similar case reports in the literature1,2,3 – HV failure at shock with normal low voltage testing
• Hauser analysis of MAUDE database4 reported 91% of Riata related deaths were due to HV failure
• Hauser reported abrasion failures of Durata5
1) Leong et al, PACE, June 20122) Shah, et al, JCE January 20133) Marenco PACE 20134) Hauser Heart Rhythm August 20125) Hauser Europace, February 2013
Thomas Hawk: Flikr.com shared under Creative Commons license
Tip of the Iceberg?“Crash Testing”
AHA Scientific Sessions 2012
• 90 Riata/ST pts underwent fluoro and HV shock• 51% externalized conductors• 9% (8 pts) failed shock testing (8/9 not externalized)
• HV failure defined as failure to induce/sense/shock VF
Tip of the Iceberg?“Crash Testing”
HRS Sessions 2013
• 289 Riata/ST patients evaluated• Mean implant time 5.7 years• Externalization by CXR: 13.1% Riata, 4.8% Riata ST• 116 patients underwent DFT test at max output• 1 Riata patient (1%) failed to rescue with initial shock
(HV <10%)
RV ring P/S Cable
Riata Evolution to Durata
SVC HV Cable
RV HV Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to can
z12
3
ICD Can
3
Optim outer insulation added with Durata
Durata ICD Lead Performance
Durata ICD Lead Performance
• Cairns et al report results of three active St. Jude managed registries on Optim ICD leads
• 11,016 leads
• Mean Follow up 3.2 years
• Independent review by Canadian Investigators
• 51 mechanical failures (0.46%), 99% survival
• No externalized leads
Criticisms of Optim ICD Lead Registry
• 22.9% drop out due to missing data
• 3.3% of leads taken out of service seems high compared to 0.46% failure. Why?
• 19.2% death or transplant. No report of post mortem interrogations. No way to exclude lead failure as cause of death.
• 39% of the leads with mechanical failure had returned product analysis. Is this too high?
Durata High Voltage Failure
• 55 yo male with Current DR, Durata lead, 12/09• NS VF episode detected, therapy diverted. EGM
shows noise on RV P/S and RV HV, not on RA• To ER w/vibrating ICD in backup mode. HV
therapies disabled and no programmability
PACE. September 2013
Durata High Voltage Failure
• In lab: Normal sensing/threshold/Ω/fluoro• No noise could be reproduced, appearance OK• New Fortify DR ICD:
• 20 J synchronized test shock successful• VF, 36 J shock failed, audible “pop”• Still in VF, 40J shock failed, backup mode• New lead & 2nd PG placed, VF test successful
Durata High Voltage Failure
RV coil-can
A sense
RV sense
Simultaneous noise on RV shock coil-can and RV sense electrodes with normal atrial sensing
RV shock coil to RV cable short
Durata High Voltage Failure #3
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to can
ICD Can
3
3
Optim Insulation
Durata High Voltage Failure
https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfmaude/Detail.cfm?MDRFOI__ID=3669333
• 49 yo woman w/hypertrophic cardiomyopathy
• Current DR with Durata 7120 placed 7/08
• Normal office follow up (low and high voltage)
Heart Rhythm, October 2014
Durata High Voltage Failure
• 12/13, arrest at work. 15 min bystander CPR, Squad shocked VF multiple times, hypothermia
• ICD evaluation: – HV impedance <10 Ω (vector not specified)
– HVLI and pace/sense evaluation normal
• Durata Lead extracted
• No visual or fluoroscopic abnormalities in lab
• New Medtronic ICD lead and PG placed
Heart Rhythm, October 2014
Durata HV Failure
Electrically silent inside out abrasion of Durata RV conductor cable under SVC coil near fatal shock failure
Personal communication and chart review, M. Gupta MD. The Christ Hospital, Cincinnati, OH
Durata HV FailureElectrically silent inside out abrasion of Durata RV
conductor cable under SVC coil near fatal shock failure
Heart Rhythm, October 2014
RV conductor CableSVC Coil
Durata High Voltage Failure #2
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
RV HV cable to SVC Coil
ICD Can
2
New St Jude System Enhancements
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
2
ICD Can
3
1
A) Parylene Coating on ICD can with Elipse PGB) Additional Optim layer from proximal end to
SVC coil with Optisure LeadC) Dynamic Tx provides shock vector switching
with detection of insulation breach
3
High Voltage ICD Lead Failure
• Underappreciated and potentially fatal
• Can present with no change in painless HV impedance or any other parameters measured in ICD follow up
• Presentation varied and at times very complex
• Existing means of detection are inadequate
• Open question: Do we need to change our follow up strategies?
High Voltage ICD Lead FailureSuggestions
• We need to encourage post-mortem interrogation ICD patients. This should be required in registries and pre market trials.
• Support investigation of St. Jude leads for high voltage performance - “Crash Test” trial.
• Encourage a protocol for assessment of St. Jude ICD leads at device replacement.
Thank you!
Backup Slides
Riata Failures in Danish Registry
Larsen et al HRS abstract 2014
• No low impedance high voltage failures in EC or no EC groups• Low impedance failure in pace/sense electrode seen only in no EC group
EC= Externalized Conductors
Thomas Hawk: Flikr.com shared under Creative Commons license
Airbag Reliability
• 100 drivers
– 10 get in accident
• 9 effective airbag deployment
• 1 driver dies with ineffective airbag
• What is the airbag failure rate?
– 1% or 10%?
Riata HV Insulation Shorts
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to can
z12
ICD Can
3
3
RV ring P/S Cable
Riata Evolution to Durata
SVC HV Cable
RV HV Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to can
z12
3
ICD Can
3
Optim outer insulation added with Durata
New St Jude System Enhancements
31
SVC HV Cable
2 RV HV Cable
RV ring P/S Cable
RV tip filament 3
ICD Can
A) Paraylene Coating on ICD can with Elipse PGB) Additional Optim layer from proximal end to
SVC coil with Optisure Lead
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable and RV
ring cable to can
Riata HV Insulation Shorts #3A
3
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to SVC coil/cable
ICD Can
3
Riata HV Insulation Short #3b
33
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
1) RV HV cable to can2) RV HV cable to SVC cable or coil3) RV Ring P/S cable to RV HV cable
and RV ring to RV tip and RV tip to can
ICD Can
3
Mitigation of Failure #2 with SVC Reprogramming
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
RV HV cable to SVC cable or coil
2
ICD Can
Mitigation of Failure #2 with SVC reprogramming
SVC HV Cable
RV HV Cable
RV ring P/S Cable
RV tip filament
RV HV cable to SVC cable or coil
2
ICD Can
SVC Programmed off Shorted SVC current blocked at can.Shock impulse arrives at heart, grounds to can.
JCE January 2013
• 72 year old with CAD, sustained VT• ICD 2004: 1580 Riata. Gen change 2009 Current Plus,
VF induced & terminated with 20 J shock • 2012 SVT episode. 25 J charge but shock truncated
with HV Ω <10J. SVT terminated spontaneously• All HV parameters normal – HVLI 44Ω
JCE January 2013
• Fluoro showed externalization between shock coils• DFT test: 20 J shock truncated with HV <10 Ω. Pt
externally defibrillated
Riata/ST High Voltage FailureWhat do the large trials tell us?
• VA Trial Sung et al– Total electrical Failure 3.7% (47/1403). HV failure 3 (6%/0.2%)*
* (% of population/% of failed leads)