PAI-1 and t-PA/PAI-1 Complex Potential Markers of Fibrinolytic Bleeding After Cardiac Surgery Employing Cardiopulmonary Bypass

Agnese Ozolina; Eva Strike; Inta Jaunalksne; Angelika Krumina; Lars J Bjertnaes; Indulis Vanags


BMC Anesthesiol. 2012;12(27) 

In This Article


The present study revealed that lower levels of PAI-1 preoperatively and of t-PA/PAI-1 complex postoperatively, are associated with lower plasma concentrations of fibrinogen, higher levels of D-dimer and increased blood loss during the first 24 hours after the operation. Our results are consistent with several recent investigations showing that activation of the fibrinolytic system is associated with increased postoperative bleeding in cardiac surgery employing CPB.[1,5,12]

The plasma concentrations of PAI-1 and t-PA/PAI-1 complex that are supposed to be the main regulators of fibrinolysis in humans, are both characterized by wide variations that may explain the large inter - individual differences in fibrinolytic activity.[16] Several recent studies have described the influence of genetic factors, such as PAI-1 promoter – 675 (4G/5G) polymorphism, on the plasma levels of PAI-1, t-PA and t-PA/PAI-1 complex.[1,4,6,17–20] The 5G allele is associated with low levels of PAI-1.[4,6,17,19]

Our patients were cooled on CPB to a bladder temperature of 34–35°C with no intergroup difference. Some investigators claim that reduced temperature lowers endogenous production of PAI-1, resulting in enhanced fibrinolysis and increased per- and postoperative bleeding[21] whereas others refute this idea.[22] Our patients were rewarmed to normal body temperature before transfer to the recovery. Therefore, it is unlikely that temperature had any influence on the formation of t-PA/PAI-1 complex. Since PAI-1 is a more stable indicator of fibrinolysis, as compared to t-PA, whose concentration peaks during CPB, we determined PAI-1 before the operation and t-PA/PAI-1 complex after the surgery[23] as well as their associations with postoperative blood loss 24 hours after the surgery. Our results indicated that those presenting with higher preoperative plasma concentrations of PAI-1 had less blood loss, and conversely, those with a lower preoperative plasma level had a larger blood loss 24 hours after surgery. Other investigators have noticed similar results.[1,24] Recently, investigators have reported favorable effects of administration of very long half-life PAI-1 (> 700 hours) on bleeding time and total blood loss after tail clip in PAI-1 deficient mice.[25] However, so far, very long half-life PAI-1 is not available as a medicine to promote hemostasis after surgery, trauma, or PAI-1 deficiency in humans.

As to the best of our knowledge, the literature is scanty on reports focusing on the importance of t-PA/PAI-1 complex and its relationship with enhanced bleeding after CPB. Our notion that patients with an accumulated blood loss in excess of 500 ml 24 hours after CPB had lower levels of t-PA/PAI-1 complex is consistent with the findings reported by Rivera and coworkers.[1] In a subgroup of patients presenting with enhanced bleeding, these workers reported lower levels of PAI-1 both before - and after surgery and lower concentrations of t-PA/PAI-1 complex postoperatively. Surprisingly, we observed no significant correlation between the preoperative concentration of PAI-1 and the level of t-PA/PAI-1 complex 24 hours postoperatively. We speculate that the lack of such correlation can be explained by the fact that the increase in PAI-1 production culminates on the first postoperative day and usually returns towards normal on the second postoperative day.[23] The plasma levels of PAI-1 are known to increase immediately after CPB as part of the "fibrinolytic shut-down"[26] and afterwards, it decreases slowly over the subsequent days or weeks.[27]

Lower preoperative levels of PAI-1 and lower t-PA/PAI-1 complex ratio 24 hours after surgery might have led to higher levels of D-dimer immediately after the surgery.[28,29] Kuepper and co-workers[5] examined 120 patients scheduled for cardiac surgery who were randomized to an aprotinin group and a control group. D-dimers reached higher plasma levels in the control group indicating increased fibrinolysis. Consistently, in our study, the patients of Group I (with the highest blood loss) had significantly lower fibrinogen - and higher D-dimer levels after surgery and throughout the ensuing 24 hours (Figure 2). The significantly lower fibrinogen level in Group I immediately after the surgery might indicate increased consumption because of the hyperfibrinolytic state of this group. These findings are also supported by the contention of previous workers that plasmin generation and fibrin degradation is increased 10-to 20-fold during CPB, and moreover that fibrin formation and degradation rates are nearly equally affected by the CPB.[2,30]

Our study has limitations. That we found no associations between preoperative PAI-1 and postoperative t-PA/PAI-1 complex could be due to a small sample size. By considering the first 25 patients included in each group, calculation of sample size revealed that the correlation between t-PA/PAI-1 complex and blood loss after 24 hours (r = − 0.24, P = 0.08) was underpowered and might have reached statistical significance (P<0.05 and a power of 80%) first by increasing the total number of patients to 134 (n = 67 in each group). Another weakness is that we did not analyze patient outcome data.

An attractive idea for a future investigation would be to determine t-PA and PAI-1 activity separately and in concert with t-PA/PAI-1 complex concentrations. This would increase our understanding of the relationship between free PAI-1 antigen and its formation of t-PA/PAI-1 complex. Moreover, a multicenter study should be performed focusing on the influence of the fibrinolytic system on postoperative bleeding and its relation to outcome after cardiac surgery.