CPB is a nonphysiological procedure that can induce the activation of platelet dysfunction, the fibrinolytic system, and the systemic inflammatory response, leading to an increased risk of perioperative bleeding[1, 35]. Tissue plasminogen activator (t-PA), as a major activator of plasminogen, is able to convert plasminogen to plasmin, in turn lysing fibrin. Prior studies have demonstrated that with the setting of CPB, t-PA secretion has an immediate sustained increase, leading to the generation of plasmin along with the degradation of fibrin, increasing 10- to 20-fold during CPB, which is associated with excessive bleeding. As an antifibrinolytic agent, TXA has the ability to mitigate the state of hyperfibrinolysis. Plasminogen activator inhibitor 1 (PAI-1), an inhibitor of plasmin formation, secretes much less plasmin than t-PA during CPB. Nevertheless, the secretion of PAI-1 increases rapidly after 30 min into CPB and continues to rise 15-fold compared to baseline by 2 h postoperatively, while the active t-PA level gradually falls [35–38]. These changes in t-PA and PAI-1 activities are likely to last from the first to the second postoperative day, thus leading to a hypofibrinolytic state and likely increasing the risk of thromboembolic events, such as early graft occlusion of the coronary artery [38, 39].
In our study, we found no statistically significant differences in terms of the TBL, postoperative blood loss, volumes of red blood cells and plasma transfusion, total drainage, days of drainage, length of ICU stay, duration of mechanical ventilation or duration of postoperative hospitalization between Groups C and T, which indicated that the additional TXA postoperatively did not improve bleeding on the basis of intraoperative TXA. As a result, we concluded that there was no need for postoperative TXA in heart valve replacement surgery with CPB, which was mainly related to the hypofibrinolytic state occurring after surgery, as mentioned above. A previous study found that perioperative use of TXA was protective against fibrinolysis and that fibrinolytic resistance continued for up to 6 h after cardiac surgery. The fibrinolytic activities of the majority of patients almost normalized 48 h postoperatively [40], which may also explain the lack of a need for additional TXA administration after heart valve replacement surgery with CPB. However, the fibrinolytic response to CPB is heterogeneous, with individual differences among patients. Approximately one-third of patients show no change in PAI-1 activity or t-PA activity postoperatively [41]. Consequently, monitoring the fibrinolysis level during the perioperative period is necessary.
The CRASH-2 trial [42] concluded that TXA should be administered as soon as possible to trauma patients with significant hemorrhage because the administration of TXA after 3 h seemed to be less effective and even increased the risk of death due to bleeding. In our study, the minimum lengths of surgery in Groups T and TH were 200 min and 180 min, respectively, both longer than 3 h. The results showed that additional TXA after cardiac valve replacement surgery did not improve bleeding, which was, to some extent, consistent with the results of the CRASH-2 study. However, owing to the limited conditions, our study did not take into consideration continuous massive hemorrhage after surgery, such as bright red fluid in a large amount that was discharged from the pericardium, which might influence our results. In this case, TXA can be administered as soon as possible for hemostasis according to the fibrinolysis level and severity of the patient’s state.
The instructions for hemocoagulase agkistrodon injection (BT) show that hemocoagulase agkistrodon with a strongly degradative effect of fibrinogen at large doses (50~100 KU) can markedly reduce the plasma fibrinogen level. Xu et al. [43] demonstrated that a significant decline in fibrinogen levels was observed in 11 patients with hematological disorders following extended exposure (longer than 7 days) to hemocoagulase agkistrodon. Furthermore, after the withdrawal of this drug, the low level of fibrinogen began to increase gradually. Nevertheless, Qi et al. [44] concluded that the rate of plasma fibrinogen reduction related to hemocoagulase agkistrodon was 14.28%, which mainly occurred 6 hours after the administration of this agent, even at a normal dose. Moreover, another study including 91 patients discovered that even a small dose (2 KU/d) or short duration (one day) of receiving hemocoagulase agkistrodon could result in hypofibrinogenemia [45].
The dose and time of hemocoagulase agkistrodon in the present study were less than 4 KU and 24 hours, respectively. The results showed that the fibrinogen levels on postoperative Days 1 and 2 in Group H were both lower than those in Group C. In addition, increased postoperative fibrinogen infusions were found in Group H (64.71%) than in Groups C (1.45%) and T (2.78%) (P < 0.05), which indicated that a small dose (2 KU/d) or short duration of hemocoagulase agkistrodon after surgery could contribute to a decrease in plasma fibrinogen, which was consistent with the results of previous studies. The mechanism is probably related to the process of fibrinolysis. Hemocoagulase agkistrodon is able to turn fibrinogen into fibrin monosomic I and accelerate the formation of blood clots, along with its ability to consume fibrinogen, which may lead to hypofibrinogenemia. Moreover, cardiac valve replacement surgery is a high-risk factor for bleeding that may aggravate the occurrence of hypofibrinogenemia. Consequently, frequent monitoring of plasma fibrinogen levels should be performed to decrease blood loss when prescribing hemocoagulase agkistrodon.
The fibrinogen level on postoperative Day 1 in Group TH was significantly lower than that in Group T but higher than that in Group H, although the difference was not significant (P > 0.05). Additionally, the patients in Group TH received a lower incidence of human fibrinogen infusions than patients in Group H (P < 0.05). Most patients in both groups recovered to the normal fibrinogen level (2~4 g/L) on postoperative Day 2 (P>0.05). This indicated that the antifibrinolytic effect of TXA could slightly improve the hypofibrinogenemia caused by hemocoagulase agkistrodon.
There are some limitations of our study. First, this was a retrospective study, and some underlying factors, such as the effect of coadministered drugs, the method of anesthesia and hemostasis during surgery, may influence the final outcomes. Second, the sample size was relatively small, especially in Group H (17 patients enrolled), which may also lead to bias and imprecision of the results. Finally, the long-term safety of TXA, hemocoagulase agkistrodon and their combination after surgery has not been investigated. For example, the occurrence of deep vein thrombosis within 3 months or 1 year postoperatively should be further followed up.