Increased activation of the coagulation and fibrinolytic systems leads to hemorrhagic complications during left ventricular assist implantation.

BACKGROUND

Left ventricular assist devices (LVADs) have provided a new therapeutic option for patients with end-stage heart failure. Despite advances in device design, there remains an apparent bleeding diathesis, which leads to increased transfusion requirements and reoperative rates. The purpose of our study was to examine the abnormalities that might contribute to these clinical sequelae.

METHODS AND RESULTS

To separate the effects of cardiopulmonary bypass (CPB), eight patients undergoing coronary revascularization (CABG) were compared with seven LVAD (TCI HeartMate) recipients intraoperatively and 2 hours postoperatively. We evaluated several well-characterized indexes of platelet activation: platelet count, platelet factor 4 (PF4), beta-thromboglobulin (beta-TG), and thromboxane B2 (TXB2). We also measured activation of thrombin: thrombin-antithrombin III (TAT), prothrombin fragment 1 + 2 (F1 + 2), and fibrinopeptide A (FPA) as well as markers of fibrinolysis: plasmin-alpha 2-antiplasmin (PAP) and D-dimer. Patterns of intraoperative platelet adhesion and activation were not statistically different in the CABG control and LVAD groups. In the immediate postoperative period, however, there was significant release of PF4 and beta-TG and generation of TXB2. Compared with the CABG controls (TAT, 26 +/- 8 micrograms/L; F1 + 2, 4 +/- 1 nmol/L; mean +/- SEM), there was a significant increase in TAT (380 +/- 112 micrograms/L) and F1 + 2 (23 +/- 4 nmol/L) in LVAD patients 2 hours after surgery. Furthermore, a sharp rise in FPA was noted 20 minutes after LVAD initiation (CABG, 8 +/- 4 ng/mL; LVAD, 235 +/- 63 ng/mL; P < .05). A concomitant increase in both PAP (CABG, 987 +/- 129 micrograms/L; LVAD 3456 +/- 721 micrograms/L; P < .05) and D-dimer (CABG, 1678 +/- 416 ng/mL; LVAD, 15243 +/- 4682 ng/mL; P < .05) was observed.

CONCLUSIONS

The additive effects of CPB and LVAD lead to platelet activation as well as elevation of markers of in vivo thrombin generation, fibrinogen cleavage, and fibrinolytic activity. The etiology of these findings may be secondary to the LVAD surface, flow characteristics, and/or operative procedure. Nevertheless, platelet alterations and exaggerated activation of the coagulation and fibrinolytic systems may contribute to the clinically observed hemostatic defect.