Comparison of in vitro effectiveness of mechanical thrombectomy devices

PURPOSE: To determine the in vitro efficacy of clot removal of the followin g hydrodynamic thrombectomy devices: the AngioJet (AJ), Hydrolyser (HL), Oa sis (OS), and Amplatz Thrombectomy Device (ATD). All devices have 6-F cathe ters. MATERIALS AND METHODS: Thrombectomy of 5-day-old porcine clots (n = 68; 8.5 g) was performed with the AJ without a guide wire [Aj(gw0)], with a coaxia l 0.016-inch guide wire [AJ(gw.016)], and with a 0.035-inch guide wire [AJ( gw.035)]) and with the HL, OS, and ATD in an artery flow model (pulsed flow : 700 L/min) simulating the superficial femoral artery (7-mm inner tube dia meter). The effluent was passed through a three-step filter system (10-1,00 0 gm; pressure drop: 35 mm Hg). RESULTS: Mean thrombectomy time ranged from 49 seconds (AJ(gw0)) to 88 seco nds (OS; P < .001). The fluid balance with use of the AJ(gw.035) was 0.89, whereas the mean ratio of applied saline solution to aspirated fluid for th e other devices was not isovolumetric (AJ(gw0), 0.8; AJ(gw.016), 0.78; HL, 0.73; OS, 0.62; P < .05). Remaining thrombus ranged in size from 8.2 mg (AJ (gw0)) to 27.3 mg (AJ(gw.035); P = .079). Hydrodynamic devices (0.6% OSI to 0.98% [AJ(gw.016)]) caused low amounts of added emboli greater 10 mum, 100 mum, and 1,000 mum. The ATD (5.19%) caused t4ew most extensive embolizatio n (P <.001). CONCLUSIONS: The tested mechanical thrombectomy devices have the power for sufficient thrombectomy in vitro; however, they showed moderate differences in performance. In contrast to hydrodynamic devices, the ATD fragmentation device showed a higher peripheral embolization rate of particles larger th an 1,000 <mu>m; adjunctive in vivo treatment would probably be required.