Purpose: To assess in vitro the performance of 5 distal protection devices (DPDs) by evaluating the capture efficiency, pressure gradient, volume flow rate, and vascular resistance in the internal carotid artery (ICA).
Methods: The time-averaged mean peak velocity in the common carotid artery and a blood-mimicking solution were used to simulate physiologICAl conditions in a silicone carotid phantom representing average human carotid artery geometry with a 70% symmetrICAl ICA stenosis. Five milligrams of dyed 200-microm nominal diameter polymer microspheres (larger than the pore size of the devices, except Spider RX, which was tested with 300-microm-diameter particles) were injected into the ICA. The percentages of particles missed after injection and lost during device retrieval were measured for the 5 devices (Spider RX, FilterWire EZ, RX Accunet, Angioguard XP, and Emboshield). The normalized pressure gradient, fraction of the volume flow rate, and vascular resistance in the ICA were calculated.
Results: Spider RX captured the most particles (missing 0.06%, p<0.05) and yielded the smallest normalized pressure gradient increase (4.2%), the largest volume flow rate fraction (0.40), and the smallest vascular resistance in the ICA (272 mmHg/L x min(-1), a 5.4% increase with respect to initial conditions). Angioguard XP captured the fewest particles (missing 36.3%, p<0.05 except Emboshield) and resulted in the largest normalized pressure gradient increase (37%) in the ICA. RX Accunet produced the smallest volume flow rate fraction in the ICA (0.30) and the largest vascular resistance in the ICA (470 mmHg/ L x min(-1), an 82.2% increase). Emboshield migrated approximately 6 cm distal to the original position after particle injection. FilterWire EZ lost the fewest particles during retrieval (0.45%, p<0.05 except Accunet RX and Spider RX) and had the best overall performance with 200-microm emboli (p<0.05 except Accunet RX).
Conclusion: None of the devices tested completely prevented embolization. Overall, Spider RX had the best performance and is conjectured to have the best wall apposition of the devices tested. Vascular resistance should be considered a key filter design parameter for performance testing since it represents a quantitative estimation of the "slow-flow phenomenon." Our findings should be extrapolated cautiously to help interventionists choose the best device.