Thrombotic phenomena with nonanticoagulated, composite-strut aortic prostheses.

Twenty-nine patients received no anticoagulant therapy after aortic valve replacement with a Starr-Edwards Model 2400 prosthesis. Hemodynamic studies were performed at 3 to 26 months (average 12 months) in 18 of 29 patients. Mean aortic valve gradients ranged from 14 to 62 mm. Hg and averaged 34 mm. Hg. Calculated aortic valve area varied from 0.20 to 1.75 sq. cm. and averaged 0.98 sq. cm. Thirteen of 18 patients had critically stenotic valve orifices. At reoperation or autopsy, examination of the prostheses consistently revealed pannus and thrombus which narrowed the inflow orifice and usually extended to the struts. Of the remaining 11 patients, 3 have died (2 suddenly and one after a cerebrovascular accident), 2 have had embolic episodes, and 6 have refused a repeat study but are being given anticoagulant therapy. Clinical examination, serum lactic dehydrogenase (LDH) levels, phonocardiography, echocardiography, and fluoroscopy of the prosthesis were often unrevealing. Cardiac catheterization was the only reliable method for critically evaluating prosthetic function. In conclusion, close follow-up, preferably with cardiac catheterization, is recommended in any patient who received a Starr-Edwards Model 2400 aortic valve prosthesis without anticoagulation. Long-term anticoagulation with sodium warfarin is indicated in all patients with a Model 2400 aortic valve prosthesis unless there is a contraindication to such therapy.