Acute obstetric coagulopathy is associated with excess plasmin generation and proteolysis of fibrinogen and factor V

Abstract

Hemostatic impairment may exacerbate postpartum hemorrhage (PPH). Previously, we described a distinct coagulopathy, associated with multiple causes of PPH including placental abruption and amniotic fluid embolus, termed acute obstetric coagulopathy (AOC). AOC is characterized by very high plasmin/antiplasmin complexes and rapid depletion of functional fibrinogen and factor V (FV). To determine mechanisms underlying AOC, we investigated the plasma from 12 women with AOC (defined by raised plasmin/antiplasmin) and 21 with severe PPH (blood loss >2000 mL or placental abruption) without AOC. Plasma from patients with AOC had a fourfold increased ability to generate plasmin compared with those with severe PPH without AOC (P < .0002). AOC was associated with fibrinogen cleavage in the circulation, demonstrated by fragment D and other breakdown products (P < .0001). D-dimer was increased 36-fold in AOC compared with severe PPH without AOC, thrombin/antithrombin complexes were not raised. FV was reduced on western blot in AOC but not severe PPH without AOC (P < .001) suggesting FV cleavage. Confocal microscopy revealed similar clot structure between AOC and non-AOC samples, but both groups differed from nonbleeding pregnant controls. These data suggest that in AOC an excess of plasmin cleaves fibrinogen and FV in the circulation causing a specific, pathognomonic depletion of coagulation factors. Fibrin(ogen) breakdown products have cofactor function for tissue plasminogen activator, and these data are consistent with these breakdown products, enhancing plasmin generation and potentially driving aberrant plasmin generation in AOC. These results have implications for the clinical management of coagulopathy during PPH.

Graphical abstract

Figure 1.

AOC is characterized by increased plasmin and loss of functional fibrinogen. Fibrinogen levels in the plasma samples from patients with AOC, patients with severe PPH without AOC, and nonbleeding pregnant controls were measured by (A) Clauss assay and as (B) total antigen, and (C) the ratio of Clauss:antigenic fibrinogen was then calculated. (D) PAP complexes and (E) thrombin/antithrombin complexes are illustrated. (F) D-dimer levels in cases of PPH with and without AOC grouped based on those that scored positive (score ≥ 26) or negative (score ≤25) on the ISTH criteria for obstetric DIC. Data are median (IQR). Analysis was by Kruskal-Wallis 1-way analysis of variance and Mann-Whitney U test with Bonferroni correction.

Figure 2.

Evidence of fibrinogenolysis and fibrinolysis in AOC. (A) A representative western blot run under nonreducing conditions using an anti-human fibrinogen detection antibody of samples from patients with AOC (red) and severe PPH without AOC (blue); the nonbleeding pregnant control sample (green) was taken before an elective cesarean delivery. Fragment D indicated is a specific marker of fibrinogen proteolysis; fragment Y is ∼160 kDa. (B) A reducing western blot with samples as in panel A, with breakdown products less than ∼30 kDa are detected in AOC but not in PPH without AOC. (C) A representative nonreducing gel of sequential samples over time from a single patient with AOC, revealing fibrinogen in the earliest sample on the left and the latest on the right. A confocal microscopy image of the plasma fibrin clot at the earliest time point (lane 3) is illustrated. (D) D-dimer levels in nonbleeding pregnant controls, AOC, and severe PPH without AOC. (E) Representative confocal microscopy images of plasma fibrin clots for pooled normal plasma (PNP), nonbleeding pregnant control taken before an elective cesarean delivery (control) PPH without AOC (PPH non-AOC), and a case of AOC (AOC). (F) The relationship between Clauss fibrinogen and INTEM A5, as a measure of clot strength, performed on a ROTEM Sigma device. Blue dots are cases of AOC and grey dots of PPH without AOC; individual cases might contribute more than 1 data point. There was no difference in clot firmness, as measured by ROTEM, between the AOC and PPH without AOC groups at similar Clauss fibrinogen levels. Dashed line is the linear correlation for all non-AOC cases (R² = 0.24), and the solid line is the linear correlation for all AOC cases (r² = 0.63).

Figure 3.

Depletion and proteolysis of FV and enhanced plasmin generation in AOC. (A) FV activity in nonbleeding pregnant controls, AOC, and severe PPH without AOC. (B) FV degradation was analyzed by western blotting. Samples were run under reducing conditions and FV detected by a monoclonal antibody directed to the heavy chain. Two patients with AOC are found to have reduced or undetectable FV compared with PNP and a nonbleeding pregnant control at the time of elective cesarean delivery (illustrated in green). The positive controls are PNP plus activated protein C (aPC) for 15 minutes or PNP plus plasmin for 10 minutes; in both cases, the FV is reduced. (C) Plasmin generation was analyzed in the plasma from the nonbleeding pregnant controls, AOC, and severe PPH without AOC. (D) The relationship between D-dimer and tPA-initiated plasmin generation capacity in all cases of PPH, including AOC and non-AOC cases. (E) The relationship between plasmin generation capacity and the amount of fragment D in the blood of a single patient with AOC at different time points who is described in the case report. Fibrinogen degradation in these patient samples is found in Figure 2C.

Figure 4.

Dynamic changes in hemostatic parameters in the case study of AOC. (A) Clauss and antigenic fibrinogen levels measured during the bleed and the response to an infusion of 6 g of fibrinogen concentrate. The increment for Clauss fibrinogen (orange arrow) is approximately half that found for antigenic (blue arrow). The measured blood loss at the time of each sample is found in pink. (B) The levels of PAP complexes, D-dimer, fragment D, and plasmin generation capacity at each time point. The time of the tranexamic acid (TXA) infusion is illustrated.