Impaired fibrinolysis in the hemolytic-uremic syndrome of childhood

EspP, an Extracellular Serine Protease from Enterohemorrhagic E. coli, Reduces Coagulation Factor Activities, Reduces Clot Strength, and Promotes Clot Lysis

BACKGROUND

EspP (E. coli secreted serine protease, large plasmid encoded) is an extracellular serine protease produced by enterohemorrhagic E. coli (EHEC) O157:H7, a causative agent of diarrhea-associated Hemolytic Uremic Syndrome (D+HUS). The mechanism by which EHEC induces D+HUS has not been fully elucidated.

OBJECTIVES

We investigated the effects of EspP on clot formation and lysis in human blood.

METHODS

Human whole blood and plasma were incubated with EspP(WT )at various concentrations and sampled at various time points. Thrombin time (TT), prothrombin time (PT), and activated partial thromboplastin time (aPTT), coagulation factor activities, and thrombelastgraphy (TEG) were measured.

RESULTS AND CONCLUSIONS

Human whole blood or plasma incubated with EspP(WT) was found to have prolonged PT, aPTT, and TT. Furthermore, human whole blood or plasma incubated with EspP(WT) had reduced activities of coagulation factors V, VII, VIII, and XII, as well as prothrombin. EspP did not alter the activities of coagulation factors IX, X, or XI. When analyzed by whole blood TEG, EspP decreased the maximum amplitude of the clot, and increased the clot lysis. Our results indicate that EspP alters hemostasis in vitro by decreasing the activities of coagulation factors V, VII, VIII, and XII, and of prothrombin, by reducing the clot strength and accelerating fibrinolysis, and provide further evidence of a functional role for this protease in the virulence of EHEC and the development of D+HUS.

Lipopolysaccharide from enterohemorrhagic Escherichia coli binds to platelets through TLR4 and CD62 and is detected on circulating platelets in patients with hemolytic uremic syndrome

This study presents evidence that human platelets bind lipopolysaccharide (LPS) from enterohemorrhagic Escherichia coli (EHEC) through a complex of toll-like receptor 4 (TLR4) and CD62, leading to their activation. TLR4 colocalized with CD62 on the platelet membrane, and the TLR4 specificity of LPS binding to platelets was confirmed using C57BL/10ScN mice lacking Tlr4. Only platelets from TLR4 wild-type mice bound O157LPS in vitro. After in vivo injection, O157LPS bound to platelets from wild-type mice, which had lower platelet counts than did mice lacking TLR4. Mouse experiments confirmed that O157LPS binding to TLR4 is the primary event leading to platelet activation, as shown by CD40L expression, and that CD62 further contributes to this process. Activation of human platelets by EHEC-LPS was demonstrated by expression of the activated GPIIb/IIIa receptor, CD40L, and fibrinogen binding. In perfusion experiments, platelet activation on endothelial cells was TLR4 and CD62 dependent. O157LPS was detected on platelets from 12 of 14 children with EHEC-associated hemolytic uremic syndrome (HUS) and on platelets from 2 children before the development of HUS but not on platelets of EHEC-infected children in whom HUS did not develop (n = 3). These data suggest that O157LPS on platelets might contribute to platelet consumption in HUS.

Activation of coagulation in C57BL/6 mice given verotoxin 2 (VT2) and the effect of co-administration of LPS with VT2

To obtain better insight into the pathogenesis of verotoxin-producing Escherichia coli-associated diseases, in this study, we explored the effect of verotoxin 2 (VT2) on coagulation in an animal model. After being given VT2 (50 ng/kg, lethal dose), C57BL/6 mice showed progressively increasing expression of TF mRNA in the kidney and brain and elevated plasma levels of thrombin-antithrombin III complex (TAT), normotest, fibrinogen, and PAI-1 paralleling the disease course over 24 hours; platelet counts were decreased at 48 hours with hemorrhage in the kidney and brain. Co-administration of lipopolysaccharide (LPS, 0.5 mg/kg) with VT2 (50 ng/kg) exhibited more prominant and/or prolonged increase in not only expression of TF and PAI-1 mRNAs in the kidney and brain but also plasma levels of TAT, fibrinogen, and PAI-1 and was associated with more remarkable hemorrhage in the tissues. Although VT2 (5 ng/kg) was not a lethal dose, co-administration of LPS (0.5 mg/kg) with VT2 (5 ng/kg) enhanced the susceptibility to VT2, resulting in more prolonged elevation of TAT levels during the first 24 hours than that in the LPS group and a second elevation at 72 hours, followed by death. Plasma IL-1beta level reached a maximum at 24 hours after VT2 (50 ng/kg) injection prior to the increase in TAT levels, whereas the increase in TNFalpha level immediately after injection was associated with the increase in PAI-1 mRNA. These observations indicate that the activation of coagulation by VT2 may occur through a mechanism different from that used by LPS, since plasma TAT levels rose in the mice immediately after LPS injection and returned to normal over 36 hours.

Plasminogen activator inhibitor type-1 and interleukin-6 in haemolytic uraemic syndrome

OBJECTIVE

Because haemolytic uraemic syndrome (HUS) is an important cause of renal dysfunction in children, the availability of prognostic markers of disease severity could assist in identifying those at risk of developing long-term sequelae. The aim of this study was to test the hypothesis that plasma levels of plasminogen activator inhibitor type-1 (PAI-1) and interleukin-6 (IL-6) in children at the time of diagnosis of HUS would predict renal function outcome in terms of glomerular filtration rate (GFR).

METHODOLOGY

Fourteen children suffering from diarrhoeal HUS were studied. Plasma samples were assayed for PAI-1 and IL-6, and GFR was measured at intervals after discharge from hospital. Twelve months following their recovery from HUS, the children were allocated to one of two outcome groups depending on whether GFR was above (Good Outcome, n = 9), or below (Poor Outcome, n = 5) 80 mL/min per 1.73 m2.

RESULTS

Elevated concentrations of PAI-1 were found in 4 of 5 Poor Outcome and 4 of 9 Good Outcome children. At the same time, increased concentrations of IL-6 were observed in 3 of 5 Poor Outcome and 3 of 9 Good Outcome children. Renal function continued to be compromised in four Poor Outcome children 36 months after diagnosis.

CONCLUSIONS

Our data show that PAI-1 and IL-6 are elevated in the plasma of some children at the time of diagnosis of HUS, but that neither is a definitive prognostic marker of poor outcome 3 years later.

Activation of both coagulation and fibrinolysis in childhood hemolytic uremic syndrome

BACKGROUND

Thrombotic microangiopathy is the fundamental lesion in diarrhea-associated hemolytic uremic syndrome. The extent of the lesion in the renal parenchyma determines the severity and outcome of the disorder, bilateral renal cortical necrosis being the worst end of the spectrum. In the early years, intravascular coagulation was considered the most important pathogenic mechanism. Yet, individual coagulation factors were normal in the vast majority of patients and therapy with anticoagulants did not alter the course. Recent studies indicate that impaired fibrinolysis might be of importance.

METHODS

We studied seven variables of the coagulation pathway (PT, aPTT, fibrinogen, FVIII:c, von Willebrand factor, thrombin-antithrombin complexes, prothrombin fragments 1+2) and seven parameters of the fibrinolytic system (plasminogen, alpha2-antiplasmin, C1-esterase inhibitor, tissue-type plasminogen activator, urokinase-type plasminogen activator, plasminogen activator inhibitor type 1, D-dimer) in 24 pediatric patients with diarrhea-associated hemolytic uremic syndrome and in 15 children with acute renal failure not due to hemolytic uremic syndrome. Samples were collected at diagnosis and every second day thereafter for a period of ten days. Additional samples were collected from patients who underwent dialysis, that is, before and after each session from those subjected to hemodialysis and every day from those subjected to peritoneal dialysis. The obtained data were compared with data from a control group consisting of healthy children.

RESULTS

Our data show four important features. (1) A significant increase in both thrombin-antithrombin complexes (P < 0.005) and prothrombin fragments 1 + 2 (P < 0.001) is observed in hemolytic uremic patients as compared to patients with acute renal failure of other causes. This finding is clearly indicative for an activation of the coagulation pathway. (2) Patients with the hemolytic uremic syndrome have significantly higher D-dimer levels, a sensitive marker of fibrin-specific fibrinolysis, as compared to patients with acute renal failure of other causes (P < 0.005). (3) Levels of plasminogen activator inhibitor-1 (active antigen as well as plasminogen activator inhibitor-1 activity) are not different in both patient groups. In contrast, plasma levels of tissue-type plasminogen activator and urokinase-type plasminogen activator are significantly higher in the hemolytic uremic patients than in those with acute renal failure of other causes (P < 0.01 and P < 0.05 respectively). (4) Hemodialysis leads to an increase in tissue-type plasminogen activator antigen and a decrease of plasminogen activator inhibitor-1 activity levels.

CONCLUSIONS

Our data demonstrate that in children with diarrhea-associated hemolytic uremic syndrome, limited intravascular coagulation occurs, without evidence of impaired fibrinolysis.

Plasminogen activator inhibitor (PAI-1) antigen levels in primary TTP and secondary TTP post-bone marrow transplantation

Our objectives were to measure and compare plasminogen activator inhibitor levels (PAI-1) in primary adult thrombotic thrombocytopenic purpura (TTP) and in secondary TTP associated with bone marrow transplantation (BMT)-TTP. PAI-1 antigen levels were measured by an enzyme linked immunosorbent assay on platelet poor plasma samples obtained from patients at the time of diagnosis of the TTP disorder and from a group of normal volunteers. The samples were frozen at -70 degrees C. Patients with TTP secondary to bone marrow transplantation had their grade determined by percentage fragmented cells and lactate dehydrogenase levels. The primary TTP samples were contributed by investigators in the multi-institutional North American TTP Group, and the bone marrow transplant samples were obtained from an adult bone marrow transplant program. Nineteen patients with adult TTP, and 47 patients with bone marrow transplant-TTP were evaluated. Of the latter, 14 had Grade 2, 13 had Grade 3, and 20 had Grade 4 BMT-TTP. PAI-1 levels were elevated compared to control volunteers in both primary adult TTP and BMT-TTP, P < 0.001. Levels did not differ from normal in Grade 2 BMT-TTP (median = 16 ng/ml; quartiles = 9-20). PAI-1 levels were similar in primary TTP (median = 32 ng/ml; quartiles = 25-51) and Grade 3 BMT-TTP (median = 35 ng/ml; quartiles = 19-48 ng/ml), P = 0.7. However, PAI-1 levels were significantly higher in Grade 4 BMT-TTP (median = 83 ng/ml; quartiles = 60-143) than Grade 3 BMT-TTP, and primary TTP, P < 0.001. PAI-1 levels are high in primary TTP and secondary bone marrow transplant-TTP (Grades 3-4). In contrast, normal levels are seen in Grade 2 BMT-TTP, which is a self-limited disorder. Therefore, high PAI-1 levels may contribute to hypofibrinolysis in the pathogenesis of primary TTP and of moderate to severe TTP (Grades 3-4) following bone marrow transplantation.

Hemostatic complications in renal disorders of the young

This review focuses on the hemorrhagic and thrombotic complications sometimes associated with the most common renal disorders in children. A Medline search of the literature was conducted from 1966 to January 1995, using combinations of key words appropriate for each disorder. Additional references were located through the bibliographies of the publications and recent journals were searched independently. The most common renal disorders with hemostatic complications in children were: renal vein thrombosis (268 children in 80 publications), hemolytic uremic syndrome (473 children in 29 publications), nephrotic syndrome (4,158 children in 51 publications), renal transplantation (3,976 children in 14 publications), glomerulonephritis (20 publications), end-stage renal disease, and dialysis (22 publications). The age distribution, clinical presentation, etiology, diagnosis, treatment, and outcome of the affected children were analyzed for each disorder. Children with inherited pre-thrombotic disorders usually do not present during childhood unless there is a secondary risk factor. Similarly, most children with renal disease do not develop thromboembolic complications. Therefore, when a child with a renal disorder develops a thromboembolic event, evaluation for an inherited pre-thrombotic disorder should be seriously considered. Guidelines for the use of heparin and warfarin in these children (both therapeutically and prophylactically) are given. At this time, the risk/benefit of thrombolytic therapy in children is not known and a general recommendation for thrombolytic therapy cannot be made.