1. Transcatheter Cardiovascular Therapeutics 2008 (October 12-17, 2008 · Washington, DC) First-in-Human Report: Initial Experience with a Stentless and Retrievable Percutaneous Aortic Valve Prosthesis Joachim Schofer Medical Care Center Prof. Mathey, Prof. Schofer Hamburg University Cardiovascular Center Hamburg, Germany Joachim Schofer Medical Care Center Prof. Mathey, Prof. Schofer Hamburg University Cardiovascular Center Hamburg, Germany

2. Disclosure Statement of Financial Interest I, Joachim Schofer, DO NOT have a financial interest/arrangement or affiliation with one or more organizations that could be perceived as a real or apparent conflict of interest in the context of the subject of this presentation.

3. The Direct Flow Medical (DFM) Aortic Valve Prosthesis Consists of a trileaflet valve made of bovine pericardium encased in a slightly tapered, conformable polyester fabric cuff Independently inflatable balloon rings constitute the upper (aortic) and lower (ventricular) margins of the cuff Transfemoral implantation Consists of a trileaflet valve made of bovine pericardium encased in a slightly tapered, conformable polyester fabric cuff Independently inflatable balloon rings constitute the upper (aortic) and lower (ventricular) margins of the cuff Transfemoral implantation Investigational device currently in European clinical trial. Not available for sale in or outside the United States.

4. The DFM Aortic Valve Prosthesis To expand the valve, the prosthesis rings are pressurized (via positioning/fill lumens) with a mix of contrast agent and saline Upon expansion of the ventricular ring, the valve is immediately functioning (no rapid pacing needed during implantation) To expand the valve, the prosthesis rings are pressurized (via positioning/fill lumens) with a mix of contrast agent and saline Upon expansion of the ventricular ring, the valve is immediately functioning (no rapid pacing needed during implantation)

5. The DFM Aortic Valve Prosthesis

6. To expand the valve, the prosthesis rings are pressurized (via positioning/fill lumens) with a mix of contrast agent and saline Upon expansion of the ventricular ring, the valve is immediately functioning (no rapid pacing needed during implantation) 3-D alignment of the prosthesis is achieved by moving the 3 positioning/fill lumens After expansion of the aortic ring, the implant can be checked for correct subcoronary position and paravalvular leaks To expand the valve, the prosthesis rings are pressurized (via positioning/fill lumens) with a mix of contrast agent and saline Upon expansion of the ventricular ring, the valve is immediately functioning (no rapid pacing needed during implantation) 3-D alignment of the prosthesis is achieved by moving the 3 positioning/fill lumens After expansion of the aortic ring, the implant can be checked for correct subcoronary position and paravalvular leaks

7. The DFM Aortic Valve Prosthesis

8. Deflation/re-inflation of the prosthesis rings allows repeated attempts at positioning and even withdrawal and exchange for another DFM valve Once the position and function of the prosthesis is deemed satisfactory, the saline/contrast mix is replaced – while maintaining a pressure of 8-10 atmospheres – with a polymer The polymer becomes solid in less than 10 minutes and cures completely within 24 hours Deflation/re-inflation of the prosthesis rings allows repeated attempts at positioning and even withdrawal and exchange for another DFM valve Once the position and function of the prosthesis is deemed satisfactory, the saline/contrast mix is replaced – while maintaining a pressure of 8-10 atmospheres – with a polymer The polymer becomes solid in less than 10 minutes and cures completely within 24 hours

9. The DFM AV Prosthesis European Clinical Trial Objectives Feasibility = accurate placement of the valve in the subcoronary position with associated improvement of hemodynamic parameters (aortic gradient) and absence of aortic regurgitation Safety = major adverse cardiac and cerebro- vascular events (death, myocardial infarction, emergent cardiac surgery, major or minor stroke) or other valve-related adverse events at 30 days

10. The DFM AV Prosthesis European Clinical Trial Inclusion criteria Contraindication to surgery (determined by consensus of two experienced cardiac surgeons) Symptomatic valvular aortic stenosis with an aortic valve area ≤0.8 cm 2 Dimensional requirements  Aortic annular diameter 19–23 mm  Average LVOT >18 mm to aortic root <36 mm Logistic EuroSCORE ≥20% Valvular and peripheral anatomy appropriate to accommodate study device and its delivery system Age ≥70 years

11. The DFM AV Prosthesis European Clinical Trial Patients enrolled n31 Age, years82 ± 4 Men15 (48%) NYHA functional class  I1 (3%)  II8 (26%)  III21 (68%)  IV1 (3%) LVEF, % 53 ± 15 Logistic EuroSCORE, % 28 ± 7 Mean pressure gradient, mmHg 50 ± 13 Aortic valve area, cm 2 0.55 ± 0.16

12. The DFM AV Prosthesis European Clinical Trial Intention-to-treat population n = 31 Device implanted n = 22 (71%) Permanent implant n = 20 (65%) Surgical conversion (n=2) No iliac access (n=2) Sites: Hamburg, Germany (n=25)Hamburg, Germany (n=25) Siegburg, Germany (n=6)Siegburg, Germany (n=6) Functionally bicuspid valve (n=2) Excessive LVOT calcification (n=3) Annular Ø , excessive calcification (n=1) Excessive valvular calcification (n=1)

13. Procedural failures secondary to native valve limitations (n=7) Functionally bicuspid native valve (n=2) The DFM AV Prosthesis European Clinical Trial LVOT calcification (n=3)LVOT calcification (n=3)  Cannot be adequately ballooned pre-implantation  Positioning difficulties Severe valvular calcification (n=2)Severe valvular calcification (n=2)  Does not fully open during valvuloplasty

14. The DFM AV Prosthesis European Clinical Trial Mean transvalvular pressure gradient in patients with a permanent implant 0 10 20 30 40 50 60 70 mm Hg Baseline (n=20) 52 Post (n=20) 14 30 Days (n=13) 17 90 Days (n=11) 20 P<0.05*P=n.s.* *Holm test P<0.001 (repeated-measures ANOVA)

15. The DFM AV Prosthesis European Clinical Trial Aortic orifice area in patients with a permanent implant 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 cm 2 P<0.001 (repeated-measures ANOVA) 0.60 Baseline (n=20) 1.57 30 Days (n=13) 1.34 90 Days (n=11) Post (n=20) 1.50 P<0.05*P=n.s.* *Holm test

16. The DFM AV Prosthesis European Clinical Trial NYHA class in 14 patients with permanent implant P=0.0002 vs. Baseline

17. The DFM AV Prosthesis European Clinical Trial – Outcomes Aortic regurgitation by echo (Hamburg patients) Post-implantation (n=18) 30 Days (n=13) 03-01-008 @ 30d

18. The DFM AV Prosthesis European Clinical Trial Major Adverse Events Deathn = 4  30-day mortality 13% [95% CI, 4%–30%]  Myocardial infarction @ day 2  Pulmonary embolism 1h after failed attempt at implantation  Septal rupture during valvuloplasty  Decompensated congestive heart failure Major stroke (@ 12h)n = 1 Surgical conversionn = 2 Totaln = 7 (23% [10%–41%] ) AV conduction block 3°n = 3 (1 after surgical conversion)

19. The DFM aortic valve prosthesis gives the operator unprecedented freedom of handling the device during implantation In the FIM experience with 31 patients, permanent implantation was achieved in 65% of patients with good hemodynamic results (mean gradient ≤20 mmHg, excellent sealing of native annulus) Despite the patients’ high surgical risk profile, implantation without hemodynamic compromise during the procedure appears safe The amount and distribution of leaflet and LVOT calcification impacts procedural outcome Patient selection is crucial! The DFM aortic valve prosthesis gives the operator unprecedented freedom of handling the device during implantation In the FIM experience with 31 patients, permanent implantation was achieved in 65% of patients with good hemodynamic results (mean gradient ≤20 mmHg, excellent sealing of native annulus) Despite the patients’ high surgical risk profile, implantation without hemodynamic compromise during the procedure appears safe The amount and distribution of leaflet and LVOT calcification impacts procedural outcome Patient selection is crucial! The DFM AV Prosthesis Conclusions