A technique for internal mammary artery to coronary artery anastomoses.

A "no-touch" technique for anastomosing the internal mammary artery to the anterior descending artery is described. This technique has been used in more than 3000 patients, and enables one to construct a long, patulous anastomosis in which all suturing is forehand.

Perioperative paraplegia and multiorgan failure from heparin-induced thrombocytopenia.

Heparin-induced thrombocytopenia and thrombosis syndrome is a rare but devastating complication. We report a patient with heparin-induced thrombocytopenia in whom heparin-induced thrombocytopenia and thrombosis syndrome developed after a cardiac operation, complicated by acute thrombosis of the aorta followed by renal failure, paralysis, and ischemic necrosis of the lower extremities. The literature suggests aspirin, dipyridamole, and iloprost as effective prophylactic agents for perioperative heparin-induced thrombocytopenia and thrombosis syndrome. This unfortunate complication underscores the importance of close platelet count monitoring in all preoperative patients undergoing prolonged heparin therapy.

Long-term results of St. Jude Medical valve in the tricuspid position.

Little is known of the St. Jude Medical heart valve's long-term clinical performance in the tricuspid position. Fourteen patients underwent tricuspid valve replacement between 1981 and 1984. Ages of the patients ranged from 32 to 81 years. Eight patients (68%) were in New York Heart Association class IV preoperatively. Twelve patients had other concomitant cardiac operations. One patient (8%) died within 30 days after operation, due to cerebrovascular accident. Postoperatively, all patients were anticoagulated with warfarin sodium and showed improvement in functional class. There were no mechanical valve failures or thromboses in patients on the anticoagulation regimen. Follow-up in all patients extended over 7 to 10 years with a mean of 8 years. There were six late deaths unrelated to valve function. The actuarial survival curve with these patients is 50% over 10 years. In conclusion, the St. Jude Medical valve is a good mechanical substitute in the tricuspid position, and we recommend its use when reconstruction is not possible and there is no contraindication for anticoagulation.

Allograft vein patency in a canine model. Additive effects of cryopreservation and cyclosporine.

Autogenous saphenous vein is the preferred conduit for many cardiovascular operations. Attempts to use allograft veins for arterial reconstruction have had poor results. To define circumstances under which allograft veins might prove to be acceptable vascular conduits, dogs underwent femoral artery bypass using reversed saphenous veins. Veins were transplanted fresh or after cryopreservation. Group I dogs received a fresh autograft to replace one femoral artery (group I F) and a cryopreserved (CP) autograft (group I C) to replace the other. Group II dogs received fresh allograft veins, and group III received CP allograft veins, neither group receiving additional treatment. Group IV received fresh allograft veins and Group V received CP allograft veins; both groups received cyclosporine 15 mg/kg. Animals were maintained until grafts occluded or until six months elapsed. Patency was observed in all group I F grafts throughout the observation period. Six-month patency rates in the other groups were: group I C, 9/10 (P = NS vs. group I F); group II, 0/10 (P less than 0.01), group III, 0/10 (P less than 0.01), group IV, 1/10 (P less than 0.01), group V, 7/11 (P = NS). In a separate series of observations 10 cryopreserved allograft veins were implanted in 10 dogs that received CsA for 30 days. CsA was then discontinued. All of these grafts occluded within 30 days of discontinuing the CsA. Long-term patency of saphenous vein allografts was achieved only with the combination of cryopreservation and immunosuppression with continued CsA.

Production of endothelium-dependent relaxation responses by saphenous vein grafts in the canine arterial circulation.

To determine if venous endothelium can acquire the ability to elicit endothelium-dependent relaxation responses, five dogs underwent femoral artery bypass with autogenous saphenous vein. The veins were harvested 15 to 17 months later. Endothelium-dependent relaxation was determined by measuring tension of deendothelialized coronary arteries mounted on a tensiometer and superfused with the effluent of the vein grafts. These grafts were perfused with acetylcholine and calcium ionophore A23187, which cause case vascular smooth muscle relaxation by means of endothelium-dependent relaxing factor production. Control arteries and veins were obtained from other dogs for comparison. In response to acetylcholine from 10(-9) to 10(-4) mol/L, the final cumulative relaxation produced in the detector coronary artery (mean +/- SD) was 64.2% +/- 25.7% by the control arteries, 14.2% +/- 5.5% by the vein bypass graft, and 5.3% +/- 5.6% by the control veins. In response to A23187 from 10(-8) to 10(-4) mol/L, the final cumulative relaxation was 66.2% +/- 19.0% by the control arteries, 30.6% +/- 8.9% by the vein bypass grafts, and 5.3% +/- 5.6% by the control veins. The differences were significant between the vein bypass grafts and the control arteries (p less than 0.04 for acetylcholine; p less than 0.04 for A23187) and the control veins (p less than 0.03 for acetylcholine; p less than 0.02 for A23187). Perfusion of saphenous veins used as chronic arterial bypass grafts with either acetylcholine or A23187 produced detector vessel relaxation, consistent with endothelium-dependent relaxing factor production. The magnitude of the relaxation response did not approach that from perfusion of control arteries.

The epithelial origin of a stromal cell population in adenocarcinoma of the rat prostate.

Dunning R3327-H rat prostate adenocarcinoma cells, when grown in syngeneic (Copenhagen) rats or nude mice, produce tumors with prominent hypercellular stroma. The authors have previously demonstrated the presence of anomalous steroid-sensitive cells in both the epithelium and stromal compartments of this model system. In order to better understand the histogenesis of these cells, the authors studied samples of the tumor which were radiolabeled overnight with tritiated dihydrotestosterone (3H-DHT). Frozen sections of the tissues were thaw-mounted onto autoradiographic emulsion-coated slides to permit silver grain identification in association with nuclei of androgen-sensitive cells. Surprisingly, numerous silver grains were found to be associated with nuclei of large cells within the stroma. Therefore, these cells were termed "epithelioid" pending confirmation of their origin. To further define these cells and their relationship to the surrounding matrix, autoradiograms have now been examined immunohistochemically with antibodies directed against the basement membrane glycoprotein, laminin, as well as antibodies specific for intermediate cytoskeletal filaments. Following identification of acinar basement membranes, epithelioid cells were identifiable both in the stroma and in the acinar epithelial cell layer. Histochemical staining with acid phosphatase, a marker for prostatic epithelium, was performed and shown to be present in acinar epithelial cells as well as in epithelioid cells. Additionally, fluorescence-activated cell sorting was employed to characterize the DNA content of cell types within the H tumor. Epithelioid cells were found to be in highest concentration in an aneuploid peak with a ploidy of approximately 6N. The autoradiographic, immunohistochemical, cytometric, and ultramicroscopic studies suggest that 1) epithelioid cells are epithelial derived stromal cells; 2) these epithelioid cells arise by pathologic division of aneuploid neoplastic precursor cells of approximately 3N ploidy, which are found within the prostatic epithelium; and 3) the resulting 6N cells degrade the basement membrane locally, invade the stroma, and populate it. Here, they can be distinguished from fibroblasts by their size, acid phosphatase activity, and hormone receptor content. Thus, the term "epithelioid" is inappropriate; and these cells should be regarded simply as large neoplastic epithelial (LNE) cells. The presence of this cell type suggests that this tumor subline represents a useful naturally occurring model for the study of the initial stages of neoplastic transformation.