OBJECTIVETo develop an inflow cannula of left ventricular assist implanted by blood vessel.

METHODSThe maximum inflow and properties against folding of 8 sorts of cannulae were measured in mimic extracorporeal circulation appliances and canines.

RESULTSThe maximum flow of the cannula increased, as the inner diameter became greater (P < 0.01) compared with each group. The maximum flow rate was (1.82 +/- 0.03) L/min, (2.44 +/- 0.03) L/min, (3.02 +/- 0.04) L/min, (3.31 +/- 0.03) L/min respectively for polyvinyl cannulae with wall thickness of 0.5 mm (PV 0.5 cannula) and inner diameter of 3 mm, 4 mm, 5 mm, 6 mm; (1.83 +/- 0.03) L/min, (3.07 +/- 0.04) L/min respectively for the polyvinyl chloride cannula with wall thickness of 1.0 mm imbedded by spring wire (PVCSW 1.0) and inner diameter of 3 mm and 5 mm; (1.82 +/- 0.02) L/min, 1.84 +/- 0.02 L/min for strengthened polyvinyl cannula with wall thickness of 0.8 mm (SPV 0.8) and inner diameter of 3 mm and polyvinyl cannula with wall thickness of 1.0 mm (PV 1.0 cannula) of inner diameter of 3 mm. There was no remarked statistical difference in vitro maximum flow among the four cannulae of 3 mm inner diameter in vitro. PVCSW 1.0 was showed the best antifolding property, PV 1.0 cannula good and SPV 0.8 and PV 0.5 unsatisfactory in properties against fold. There was no significant statistical difference between in vivo and in vitro maximum flow for PVCSW 1.0 and PV 1.0 cannulae of 3 mm inner diameter. But for SPV 0.8 and PV 0.5 cannulae of 3 mm inner diameter, there was a significant difference between in vivo and in vitro.

CONCLUSIONSPV 0.5 cannula and SPV 0.8 cannula are not suitable to clinical use. PV 1.0 cannula can be used in clinics. PVCSW 1.0 cannula is fully qualified for inflow conduit of left ventricular assist in surgery.