1. 1 Is It Time to Redefine ICD Therapy ? The Importance of Reducing Shocks

2. 2 30 Years Ago…

3. 3 Secondary Prevention Primary Prevention AVID 1 CASH 2 CIDS 3 MADIT 4 MUSTT 5 MADIT II 6 SCD-HeFT 7 Mortality Reduction with ICDs 1. The AVID Investigators. N Engl J Med. 1997;337:1576-1583. 2. Kuck KH. Circulation. 2000;102:748-754. 3. Connolly. Circulation. 2000;101:1297-1302. 4. Moss AJ, et al. N Engl J Med. 1996;335:1933-1940. 5. Buxton AE, et al. N Engl J Med. 1999;341:1882-1890. 6. Moss AJ, et al. N Engl J Med. 2002;346:877-883. 7. Bardy GH, et al. N Engl J Med. 2005;352:225-237.

4. 4 Marked Over-Treatment in Primary and Secondary Prevention Patients 1. Wathen M. Am Heart J. 2007;153(4 suppl):44-52. 2. Buxton AE. N Engl J Med. 1999;341:1882-1890. 3. Bardy GH. N Engl J Med. 2005;352:225-237. 4. The AVID Investigators. N Engl J Med. 1997;337:1576-1583. Does ICD Therapy = A Life-Saving Event? 2.7X 5.7X 5 yr 37% SCD 5 yr 60% VT/VF shocks 6.6X 2 3 4

5. 5 ICD Therapy: different types of shocks Appropriate shocks: Shocks triggered by potentially life-threatening arrhythmias –Unnecessary shocks: Other painless therapy can be used to terminate arrhythmia –Necessary shocks : Shock for arrhythmias not terminated by other means Inappropriate Shocks: Shocks triggered by an inappropriate detection

6. 6 1 Kadish A, et al. N Engl J Med. 2004;350:2151-2158. 2 Daubert JP, et. J Am Coll Cardiol. 2008;51:1357-1365. 3 Poole JE, et al. N Engl J Med. 2008;359:1009-1017. 4 Saxon A. L. Circulation. 2010;122:2359-2367. Up to 21% of ICD Patients Receive Inappropriate Shocks 1-4 % of patients shocked n.a 2 1 3 4

7. 7 ICD Shocks in Atrial Fibrillation 913 consecutive ICD pts: 73% had no AF, 9% paroxysmal AF, 7% persistent AF, 11% permanent AF Permanent AF: mortality risk doubled, higher risk any therapy Paroxysmal/persistent AF: 3 times higher risk of inappropriate shocks Borleffs J. J Am Coll Cardiol 2010; 55: 879–885 Appropriate device shockInappropriate device shock Black line: no AF Green line: Paroxysmal AF Orange line: Persistent AF Red line: Permanent AF

8. 8 Prognostic Importance of ICD Shocks SCD-HeFT patients who received and ICD (n=811) 33.2% received shocks: 15.8% only appropriate, 10.7% only inappropriate and 6.7% both Patients who receive shocks for any arrhythmia have a higher risk of death than those who do not receive such shocks Poole JE. N Engl J Med 2008; 359: 1009-1017

9. 9 Shocks but not ATP are associated with Higher Mortality Retrospective analysis of pooled data –PainFREE I and II, EMPIRIC and PREPARE –2,135 pts, EF 31%, 87% CAD, 55% NYHA II/III, 42% NYHA I/ no CHF Sweeney M. Heart Rhythm 2010; 7: 353 - 360

10. 10 Shocks Affect Patients’ Quality of Life Psychological impact of shocks on patients has been clinically studied For patients, simply the fear of shocks can be disruptive to a normal, active life Shock reduction has been shown to improve ICD patient’s quality of life and, in the process, may reduce a patient’s greatest fear: fear of getting shocked Sears SE, et al. Clin Cardiol. 2003;26:107-111. Irvine J, et al. Am Heart J. 2002;144:282-289. Wathen MS, et al. Circulation. 2004;110:2591-2596.

11. 11 Potential Benefits of Shock Reduction Improved patient quality of life Increased ICD therapy acceptance Extended ICD longevity Less demand for post-shock care May improve survival benefits of ICDs Reduced healthcare spending Wathen MS. Circulation 2001; 104: 796-801. Wathen MS. Circulation 2004; 110: 2591-2596 Sears SE Jr. Clin Cardiol. 2003; 26(3): 107-111. Ahmad M. PACE. 2000; 23(6): 934-938. Sweeney M. Heart Rhythm 2010; 7: 353 - 360

12. 12 Wathen MS. Am Heart J. 2007;153(4 Suppl):44-52. Rethinking the Job of the ICD “For every episode of sudden death, cardiac arrest, or sustained VT, the ICD should produce a therapy. In other words, in a population of patients who receive an ICD, if there is a 5% annual risk of sudden death, then 5% of patients should be treated.”

13. 13 Reducing inappropriate and unnecessary shocks

14. 14 What Is the Job of the ICD? Detection –Reliable detection of VT/VF Therapy –Reliable termination of potentially lethal ventricular tachyarrhythmias – either through the use of ATP (antitachycardia pacing) or if needed, delivery of a shock

15. 15 What Are the Challenges to Shock Reduction Programming? “Changing habits” ATP (antitachycardia pacing) is often neglected: Why? –Efficacy questioned – especially for Fast VTs (> 188 bpm) –Concern about VT acceleration –Concern about syncope due to delayed therapy Detect times are too short: –Over-treatment of VT (ICD Rx is delivered for NSVTs)

16. 16 Causes of Shocks Poole JE. Presented at the Heart Rhythm Society 2004.

17. 17 Programming Strategies to Safely Reduce Shocks PREPARE Programming Charge time < 10 s even at end of life ATP Before and During Charging Poole JE. Presented at the Heart Rhythm Society 2004.

18. 18 ATP for Fast VTs Reduces Shocks Wathen MS. Circulation 2001; 104: 796-801 Note: ATP success was 89% with 3 ATP; 85% with up to 2 ATP (protocol), 77% adjusted efficacy rate ATP success 85% ATP failed 15% FVT 40% VT 57% VF 3% PainFREE Rx I: 220 ICD pts. with CAD received empirical ATP (up to 2) for fast VTs (188-250 bpm), NID 12/16 ATP terminated 396 out of 446 FVT episodes (89%) VT acceleration and FVT syncope were rare (4 and 2% resp.)

19. 19 ATP for Fast VTs Reduces Shocks II Wathen MS. Circulation 2004; 110: 2591-2596 ATP success 72% ATP failed 28% ATP Arm n=284 episodes Shocked 64% Spontaneous Termination 34% ATP 2% Shock Arm n=147 episodes PainFREE Rx II: 634 prim./sec. prevention ICD pts. randomized to empirical ATP or shock for fast VTs (188-250 bpm), NID 18/24 ATP successfully terminated 3 out of 4 Fast VTs ATP is highly effective, equally safe and improves QoL

20. 20 “These observations, combined with the established efficacy of ATP for slower VT, reposition the ICD as primarily an ATP device with only occasional backup defibrillation.” Wathen MS. Circulation 2004; 110: 2591-2596 PainFREE Rx II Lessons

21. 21 ATP During Charging ™ in PainFREE Study Made ATP nominal in ICDs Deliver single sequence of ATP while the capacitor charges for a shock No delay in delivery of shock therapy Medtronic Protecta™ and Protecta™ XT system reference guides.

22. 22 Empiric Programming is Effective 900 ICD pts. randomized to Tailored or Empiric (ATP for fast VTs (200-250bpm), NID 18/24) programming Empiric was non-inferior to Tailored programming, had less unscheduled hospitalization and reduction of pts. with 5 or more shocks for all-cause and true VT/VF Wilkoff BL. J Am Coll Cardiol 2006; 48: 330-339

23. 23 Strategic Programming Reduces Shocks PREPARE: Prospective, cohort controlled study 700 primary prevention ICD or CRT-D patients programmed to ATP for fast VT (182-250 bpm), NID 30/40, VT monitor (<182 bpm) Reduction of unnecessary and inappropriate shocks, improved survival Wilkoff BL. J Am Coll Cardiol 2008; 52: 541-550

24. 24 CRT-D with NID programmed to 30/40 RELEVANT: 324 primary prevention pts. with non-ischemic etiology with CRT-D programmed to: NID 30/40 or 12/16 (control) Study arm showed: –Better event-free survival to first delivered therapy for total, appropriate and inappropriate episodes –Lower total number of delivered shocks –Reduced HF hospitalization Gasparini M. Eur Heart J 2009; 30(22): 2758-2767

25. 25 Longer NID – In Real Life It Works!! #1#12 #18 #24 #28 Completely asymptomatic patient. Seen for routine follow-up.

26. 26 Lots of Data and Experience Now Suggest Use of ATP Prolonged Detection Reduced Shocks in ICD patients For Primary Prevention Patients:

27. 27 New Algorithms to Reduce Shocks Lead noise discrimination and alert T Wave OS Discrimination Programmable RV Sensing Programmable Sensitivity Combine Wavelet + PR Logic in dual/triple chamber Apply SVT discrimination in the VF zone More reliable confirmation at the end of the charge PREPARE Programming Charge time < 10 s ATP Before and During Charging Poole JE. Presented at the Heart Rhythm Society 2004.

28. 28 Protecta ™ SmartShock ™ Technology Six exclusive shock reduction algorithms that do not reduce sensitivity (nominally programmed ON): Reduced shocks for SVT –PR Logic ® + Wavelet (d ual and triple chamber devices) –SVT limit nominally in the VF zone Reduced shocks for T-wave oversensing –T Wave Discrimination Reduced shocks for Lead fracture and lead noise –Lead noise discrimination and alert –Lead integrity alert Reduced shocks for non sustained VT –Confirmation + Medtronic Protecta™ and Protecta™ XT system reference guides.

29. 29 Reducing Shocks Due to SVT 20% of inappropriate shocks due to SVTs in VT/VF zone Avoid shocks for rapidly-conducted AF in VF zone and better discriminate sudden onset SVT in VT zone Poole JE. Presented at the Heart Rhythm Society 2004.

30. 30 Improved SVT Discrimination PR Logic ® + Wavelet –Combines morphology and A-V pattern recognition to better discriminate against all types (in dual and triple chamber devices) –PR Logic: Effectively discriminates Sinus Tach and AF/Aflutter –Wavelet: Uses EGM morphology to improve SVT discrimination (i.e., conducted AF and sudden onset SVT) Wavelet Medtronic Protecta™ and Protecta™ XT system reference guides.

31. 31 SVT Limit in the VF Zone PR Logic ® and Wavelet are nominally ON with an SVT limit zone set to 260 ms (231 bpm) Only Medtronic can program discriminators in the VF zone – down to 240 ms (250 bpm) Discrimination in the VF Zone (nominal) SVT Limit = 260 ms Medtronic Protecta™ and Protecta™ XT system reference guides.

32. 32 Reduce Shocks Due to T-wave Oversensing Algorithms to address TWOS may require manual sensitivity adjustment after shocks for TWOS This has the potential for undersensing R-waves and affecting ventricular detection sensitivity 2. Gunderson BD, et al. Heart Rhythm 2004; 1: S244.

33. 33 First and Only T-Wave Discriminator New approach to T-wave OS: Frequency analysis versus manual sensitivity adjustment Fully automatic Does not require an initial shock followed by reprogramming No compromise on VF detection sensitivity: –Assures that all R-waves are appropriately sensed, minimizing the chance of impacting ventricular sensitivity Medtronic Protecta™ and Protecta™ XT system reference guides.

34. 34 T Wave Oversensing Algorithm Uses both signal-frequency content and pattern analysis to distinguish R-T pattern Sense EGM: in current devices, signal is filtered to isolate R waves. May oversense T waves. In Protecta ™, differentiation of sense EGM enlarges the ratio of R-to-T– wave amplitudes, enabling R-T pattern recognition. Medtronic Protecta™ and Protecta™ XT system reference guides.

35. 35 Protecta ™ TWOS Algorithm The algorithm is withholding the therapy Medtronic Protecta™ and Protecta™ XT system reference guides.

36. 36 Reducing Shocks Due to RV Lead Noise Cumulative lead malfunction incidence is 4.6% at 10 years across manufacturers Lead malfunction resulted in inappropriate shocks in 76% of the cases 1 1. Eckstein J, et al. Circulation. 2008;117:2727-2733. 2. Gunderson BD, et al. Heart Rhythm. 2004;1:S244.

37. 37 Lead Integrity Suite Combines two algorithms that detect, alert, and withhold inappropriate therapy for lead failure: Lead Integrity Alert Provides advance warning for lead fracture and extends the VF detection time Lead Noise Discriminator + Alert Identifies oversensing due to noise artifacts and provides ability to withhold therapy delivery No compromise of VT/VF detection sensitivity Notifies clinician of potential lead noise Medtronic Protecta™ and Protecta™ XT system reference guides.

38. 38 CareLink functionality with wireless devices Lead Integrity Suite on the MDT CareLink ® Network CareLink website sends a CareAlert Notification (text message, or email) Clinic chooses whether this alert should be red, yellow, or website only up to a patient level

39. 39 Lead Integrity Alert Avoids Inappropriate Shocks in Lead Failure Fractures in pace-sense electrodes of ICDs often lead to inappropriate shocks due to oversensing LIA is based on lead impedance and an oversensing trigger When triggered it extends the NID to 30 out of 40 LIA provided at least 3-day warning of inappropriate shocks in 76% of pts. Swerdlow CD. Circulation. 2008; 118: 2122-2129

40. 40 Lead Integrity Alert Reduces Inappropriate Shocks in Lead Failure Evaluation of LIA algorithm by comparing pts. with Sprint Fidelis lead failure without versus with LIA Pts. with LIA had: –Less often inappropriate shocks as first sign of failure (p=0.0006) –Less inappropriate shocks (p=0.017) Kallinen LM. Heart Rhythm 2010; 7: 1048 –1055

41. 41 LIA Reduces Inappropriate Shocks 213 pts. with lead fracture and LIA were compared to 213 pts. with lead fracture but without LIA LIA group had 46% reduction in percentage of pts. receiving inappropriate shocks Swerdlow CD. Circulation. 2010; 122: 1449-1455

42. 42 LIA: Solution for Proactive Protection Reduces the chance of inappropriate shocks –Increases sensitivity of detecting lead fractures compared with fixed impedance –Automatically and safely extends the VF detection NID to 30/40 when the alert is triggered Helps you to manage your patient –Monitors lead integrity continuously –Provides improved diagnostics triggered by oversensing –Allows easy access to the information via CareLink™ Increases the chance to react early enough

43. 43 RV Lead Noise Discrimination Lead-noise oversensing is typically isolated to the near field sensing signal Compare near-field sensing signal: –Far field EGM used to confirm senses Compares far- field cardiac activity to sensed event Medtronic Protecta™ and Protecta™ XT system reference guides.

44. 44 Additional Safety Features: Programmable Timeouts Timeout addresses concerns about sustained high rates for long period of time and allows options for individual programming Allow to override the discrimination features and deliver therapy when a VT continues beyond a programmed length of time Timeouts available in Protecta: –Lead noise discriminator Timeout: Nominally ON (45 seconds) –SVT discriminators (PR Logic ® and Wavelet) timeout: High-Rate Timeout Applies to VT and VF Zone Nominally OFF VF High- Rate Timeout Applies only to the VF Zone Nominally ON in VR (45 seconds) Nominally OFF in DR/CRT-D Medtronic Protecta™ and Protecta™ XT system reference guides.

45. 45 Reducing Shocks Due to Non Sustained Ventricular Arrhythmias Patients can receive unnecessary shocks for rhythms that are detected but terminate prior to end of the charge, or for PVCs that occur once the arrhythmia terminates Poole JE. Presented at the Heart Rhythm Society 2004.

46. 46 Confirmation + Better identifies tachycardia termination with ATP or spontaneously during charge and aborts the shock Avoids inappropriate shocks for single PVC or single fast events at the end of the charge Confirmation + Modify determination of persisting arrhythmia based on intrinsic detected rate vs. programmed lower therapy rate (sync interval) Confirmation now available after ATP During Charging sequence CONFIRMATION Medtronic Protecta™ and Protecta™ XT system reference guides.

47. 47 Confirmation + Rx1 confirmed during redetection Abort #1 Abort #2 Medtronic Protecta™ and Protecta™ XT system reference guides.

48. 48 Safe Shock Reduction Strategies ATP reduces shocks for VT Longer detection interval – coupled with shorter charge times – allows more episodes to terminate spontaneously Combined approach to reduce all types of inappropriate shocks: –Improved SVT discrimination –Better TWOS and lead noise algorithms –Improved confirmation algorithm reduces shocks for NSVT

49. 49 Clinical Experience: Computer Modeling Computer modeling addresses some of the limitations of randomized controlled trials: –Fast and early results –Individual algorithms can be tested separately –Possibility to repeat the test many times –Cost-saving Virtual ICD study predicts shock reduction results for a combination of strategies/features using a computer model and real ICD episode data from a long-term clinical study Combining Shock Reduction Strategies to Enhance ICD Therapy: A Role for Computer Modeling Volosin K. J Cardiovasc Electrophysiol 2010. Advance access published on Oct. 11, 2010

50. 50 Computer Modeling: Approach Build the model - Virtual ICD –Based on data from prior studies (PainFree II, WAVE, ENTRUST) Validate the model –Adjudicated episodes from the EMPIRIC trial (ATP, PR-Logic) Apply the model –Adjudicated SCD-HeFT episodes were used to establish clinically understandable performance predictions Combining Shock Reduction Strategies to Enhance ICD Therapy: A Role for Computer Modeling Volosin K. J Cardiovasc Electrophysiol 2010. Advance access published on Oct. 11, 2010

51. 51 Computer Modeling: Results of New Algorithms With Protecta TM family of devices, 98% of patients are free of inappropriate shocks at 1 year, and 92% at 5 years. Combining Shock Reduction Strategies to Enhance ICD Therapy: A Role for Computer Modeling Volosin K. J Cardiovasc Electrophysiol 2010. Advance access published on Oct. 11, 2010 † All shocked episodes from SCD-HeFT that had protocol defined programming and were determined not to be continuations of previous true VT/VF were included for these calculations (736/1,233, 59.7% of total SCD-HeFT treated episodes). * Only the 786 subjects whose devices were initially programmed to SCD-HeFT protocol parameters were included in the analysis. ** Derived from the subset of cohort episodes (650/736) for which NID of 30/40 could be evaluated. With SmartShock™ Technology

52. 52 Future Clinical Evidence The PainFree SST – a prospective clinical study – is conducted to further understand the utility of shock reduction strategies and anticipated to provide additional evidence

53. 53 “Near total reliance on shocks may have underestimated the ICD survival benefit in clinical trials. Strategies to minimize shocks may further improve survival in ICD patients.” 1 Dr. Michael O. Sweeney Director, Cardiac Pacing and Implantable Device Therapies Bigham and Women’s Hospital, Boston, MA “The reputation of the ICD as a “shock box” is a significant source of anxiety to potential patients. Today, third generation ICDs are much improved in their sensing and tiered therapy options to reduce shocks and their resulting distress.” Dr. Sam Sears Professor and Director of Health Psychology East Carolina University, Greenville, NC Sears SF, et al. Heart. 2002; 87: 488-493. Sweeney MO, et al. Heart Rhythm. 2010;7:353-360.