Hemolytic Uremic Syndrome Revisited Shiga Toxin, Factor H, and Fibrin Generation
Transcription
Hemolytic Uremic Syndrome Revisited Shiga Toxin, Factor H, and Fibrin Generation
Pathology Patterns Reviews Hemolytic Uremic Syndrome Revisited Shiga Toxin, Factor H, and Fibrin Generation Douglas P. Blackall, MD,1 and Marisa B. Marques, MD2 Key Words: Factor H; HUS; Hemolytic-uremic syndrome; Microangiopathic hemolytic anemia; Renal failure; Shiga toxin; Thrombotic microangiopathy DOI: 10.1309/06W402EHNGVVB24C Abstract The hemolytic uremic syndrome (HUS) is a disease characterized by microangiopathic hemolytic anemia, thrombocytopenia, and renal failure. These features reflect the underlying histopathologic lesion: fibrin-rich thrombi that predominate in the renal microvasculature. HUS most commonly affects children younger than 5 years and is associated with Shiga toxin–producing enteric bacteria, the most important of which is Escherichia coli O157:H7. In this setting, HUS is epidemic and also might affect adults, particularly elderly people. Sporadic cases of HUS more commonly occur in adults and are associated with a wide variety of inciting agents and conditions. Although the disease manifestations might be similar and endothelial activation or injury likely represents a common etiologic event, differing responses to therapy suggest different pathogenic mechanisms. As more is understood about the underlying pathogenesis of the diseases that we now lump together as HUS, more efficacious and rational treatment and prevention strategies are likely to follow. © American Society for Clinical Pathology The hemolytic uremic syndrome (HUS) is a disease characterized by microangiopathic hemolytic anemia, thrombocytopenia, and acute renal failure. It was first described by Gasser et al in 1955 and is the most common cause of renal failure in early childhood.1,2 However, all age groups are affected by this disease process, and extrarenal manifestations of the disease may occur (eg, microthrombi in the brain).2 Vascular endothelial injury seems to have a central role in HUS pathogenesis, with resulting fibrin-rich thrombi responsible for the clinical manifestations of the disease.3 Hemolysis occurs as erythrocytes traverse occluded vessels, with schistocytes typically abundant on examination of the peripheral blood smear. Platelets are consumed at sites of vascular injury, resulting in thrombocytopenia. Finally, for reasons that are not completely understood, thrombus formation is most significant and obvious in the kidneys, with resulting renal failure. The clinical manifestations of HUS can be difficult to distinguish from thrombotic thrombocytopenic purpura (TTP). In fact, many have viewed these disorders as an overlap syndrome (TTP-HUS), with the predominant clinical manifestation determining the ascribed diagnosis. 4 Thrombotic microangiopathies with predominant neurologic manifestations might be diagnosed as TTP, while those with primary renal involvement are named HUS. However, the past decade has realized an explosion in our understanding of the pathogenic underpinnings of both illnesses, to the extent that it is becoming possible to differentiate at least 2 distinctive pathologic processes. This is an important consideration because therapy for each is quite distinct. This review focuses on the clinical manifestations of HUS, its laboratory findings and pathogenesis, and the implications for therapy that have resulted from a greater understanding of the disease and its causes. Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C S81 S81 Blackall and Marques / HEMOLYTIC UREMIC SYNDROME REVIEW Case Study A previously healthy 4-year-old girl was admitted to the hospital with a 1-week history of abdominal pain and watery diarrhea. Just before admission, bloody diarrhea was noted. The patient was admitted with a presumptive diagnosis of gastroenteritis and dehydration. Blood and stool cultures were obtained, and the patient was hydrated. Admitting laboratory data were all within defined reference ranges, including a hemoglobin of 12.9 g/dL (129 g/L; reference range, 11.5-13.5 g/dL [115-135 g/L]), a hematocrit of 37.4% (0.37; reference range, 34.0%-40.0% [0.34-0.40]), a platelet count of 306 × 103/µL (306 × 109/L; reference range, 150-400 × 103/µL [150-400 × 109/L]), and a creatinine level of 0.4 mg/dL (35 µmol/L; reference range, 0.1-0.7mg/dL [9-62 µmol/L]). A few days after admission, the patient’s laboratory values began to change: her hemoglobin, hematocrit, and platelet count fell, and her creatinine level rose. At this time, the microbiology laboratory reported a negative blood culture, but a stool culture was positive for Escherichia coli, serotype O157:H7. An enzyme immunoassay also was positive for Shiga toxin. One week after the onset of the patient’s bloody diarrhea, her hemoglobin had fallen to 6.9 g/dL (69 g/L), her hematocrit was 19.5% (0.20), her platelet count was 39 × 103/µL (39 × 109/L), and her creatinine level had risen to 4.9 mg/dL (433 µmol/L). She was also anuric, for which hemodialysis was started. During the next few days, the patient received supportive care and a total of 3 dialysis procedures. Almost immediately, her condition improved as evidenced by a falling creatinine level, a rising platelet count, and stabilized hemoglobin and hematocrit values (although the patient required 1 transfusion of packed RBCs). After a 2-week hospitalization, the patient’s creatinine level and platelet count returned to normal, and she was discharged from the hospital. Clinical Manifestations The preceding case study represents the classic manifestations of HUS. The most common form of the syndrome is associated with a prodromal diarrhea (D+ HUS) that occurs in healthy young children between 6 months and 5 years of age.1 Initially, the diarrhea is watery but usually evolves to a hemorrhagic colitis. This is followed by evidence of hemolysis (falling hemoglobin and hematocrit values; schistocytes on peripheral blood smear) and thrombocytopenia within 5 to 7 days. Oliguria and anuria may follow several days later. Therapy is largely symptomatic, and the outcome for patients with D+ HUS generally is favorable, although the combined mortality rate and rate of end-stage renal disease is approximately 10%.5 In addition, long-term follow-up of S82 S82 Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C survivors has demonstrated that 25% will have renal dysfunction (as characterized by a glomerular filtration rate lower than 80 mL/min per 1.73 m 2 ), hypertension, or proteinuria.5 Fortunately, recurrence is uncommon. In contrast with diarrhea-associated HUS, the sporadic, nonprodromal form of HUS is not associated with a preceding diarrhea (D– HUS) and is rare in childhood. This form of HUS has a worse prognosis, is more likely to relapse, and sometimes is associated with a family history of disease.2 In addition, D– HUS is associated with certain drugs (eg, pentostatin, cyclosporine, mitomycin C), various malignant neoplasms (eg, pancreatic and lung cancers), solid organ and bone marrow transplantation, and vasculitic diseases.2 D– HUS is associated with a greater incidence of extrarenal disease (eg, neurologic complications, liver dysfunction, pancreatic and cardiac problems) and might be difficult to distinguish from TTP. Still, the hallmarks of HUS remain the same: microangiopathic hemolytic anemia, thrombocytopenia, and renal failure. Laboratory Findings HUS is largely a clinical diagnosis supported by laboratory abnormalities that reflect the underlying pathophysiologic process (intravascular fibrin-rich thrombus formation). A nonimmune, schistocytic, hemolytic anemia with prominent polychromatophilia is seen on the blood smear. Biochemical evidence supporting hemolysis also is routinely present, including an elevated lactate dehydrogenase level, a low serum haptoglobin concentration, and an unconjugated hyperbilirubinemia. Thrombocytopenia is a common finding in HUS and directly reflects the underlying pathophysiology (endothelial injury and platelet consumption). However, severe thrombocytopenia (<10 × 103/µL [<10 × 109/L]) is rare, with most patients having platelet counts in the 50 × 103/µL (50 × 109/L) range.1 Neither the severity nor the duration of thrombocytopenia correlates with overall disease severity or likelihood of long-term renal sequelae.2 However, persistent renal disease (decreased glomerular filtration rate, hypertension, proteinuria) after apparent renal recovery seems to be associated with the development of end-stage renal disease (10%-15% of patients with HUS).5 Therefore, extended follow-up is recommended for all patients. The results of routine coagulation studies (prothrombin time, partial thromboplastin time, fibrinogen) usually are normal in HUS, with only mildly elevated fibrin degradation product levels. However, a recently published study of children with E coli O157:H7 infections, in whom HUS subsequently developed, demonstrated prothrombotic coagulation abnormalities preceding clinical evidence of the syndrome.6 The authors found that these patients, in comparison with © American Society for Clinical Pathology Pathology Patterns Reviews infected children in whom HUS did not develop, had significantly higher median plasma concentrations of prothrombin fragment 1+2, tissue plasminogen activator antigen, tissue plasminogen activator inhibitor type 1 complex, and D-dimer levels. They also found that urinary concentrations of β2microglobulin and N-acetyl-β-glucosaminidase (early markers of renal injury) were not elevated until the obvious clinical onset of HUS. This implies that the tubular injury and renal insufficiency accompanying HUS result primarily from the formation of fibrin thrombi (rather than the direct toxic effect of circulating bacterial products such as Shiga toxin). Finally, in contrast with patients with TTP, patients with D+ HUS almost invariably do not have deficiencies of von Willebrand factor–cleaving protease (ADAMTS 13), a metalloproteinase important in the pathogenesis of TTP.7 However, this protease is much more likely to be deficient in children and adults with atypical, D– HUS and might define a true overlap syndrome with TTP.7,8 HUS is a thrombotic microangiopathy confirmed by histopathologic examination. Endothelial injury is evident (glomerular endothelial cell swelling), and thrombotic occlusion of the glomerular capillary lumen results in ischemic tissue damage.3 Tubular epithelial cell injury, mesangial expansion, and mesangiolysis also have been observed. In the most severe cases, cortical necrosis is seen. In addition to these changes, inflammatory cells (neutrophils and macrophages) infiltrate the kidneys, and apoptosis of renal cortical glomerular and tubular cells has been documented.3 The thrombotic microangiopathic changes of D+ HUS also can be seen in other organ systems (eg, the central nervous system). In contrast with the aforementioned findings, the microangiopathic changes of TTP and D– HUS are similar. They are arteriolar, and the associated thrombi are rich in platelets and von Willebrand factor rather than fibrin.9 These findings suggest that TTP/D– HUS cases have a pathogenesis that is different from that of D+ HUS. Pathogenesis D+ Hemolytic Uremic Syndrome Epidemic and endemic D+ HUS is associated with infections caused by verocytotoxin (Shiga toxin)–producing organisms, including E coli O157:H7, E coli O26:H11, and Shigella dysenteriae type 1.1 In North America and Western Europe, infection with E coli O157:H7 is the predominant cause of HUS.3 Food and water are the most common modes of infection, although person-to-person transmission has been described.10 After a mean incubation period of 3 days, infected patients develop a watery diarrhea with cramping abdominal pain that may evolve into hemorrhagic colitis. A © American Society for Clinical Pathology week after the onset of diarrhea, HUS will develop in 15% of patients.6 Shiga toxins (Stx 1 and Stx 2) are considered the major virulence factors of the organisms involved in HUS. However, a wide variety of other elaborated bacterial products, including lipopolysaccharide (LPS) and the adhesins— intimin and E coli–secreted proteins A, B, and D—also might have important pathogenic roles.3 In addition, the host inflammatory response to infection and a prothrombotic state might contribute to the development of HUS.11,12 As previously mentioned, Shiga toxins seem to be the major virulence factor responsible for the pathogenesis of HUS. This is highlighted by the fact that enteropathogenic strains of E coli, although they possess many of the same virulence factors as enterohemorrhagic strains (eg, O157:H7), have never been associated with HUS, presumably because they do not elaborate Shiga toxin.3 Human isolates of Shiga toxin–producing E coli express Stx 1 or Stx 2 encoded on a bacteriophage. These toxins are 60% homologous, but Stx 1 from E coli is identical to the toxin produced by S dysenteriae, except for a single amino acid substitution.3 Stx 2 seems to be more virulent than Stx 1 with respect to the development of hemorrhagic colitis and HUS.13-15 Experimental data also support this clinical observation, as Stx 2 is 400 times more potent than Stx 1 in mice. 16 In addition, human intestinal and glomerular endothelial cells are more sensitive to the in vitro cytotoxic effects of Stx 2.17,18 Although Shiga toxin virulence is not understood completely, its biochemistry, cell biology, and molecular effects have been well studied. Shiga toxin is a holotoxin containing 1 A subunit and 5 B subunits.3 The B subunit is responsible for toxin binding and has specificity for the glycosphingolipid membrane receptor globotriaosylceramide (Gb3).3 This molecule is the biochemical equivalent of the Pk blood group antigen. 19 Gb3 facilitates endocytosis and translocation into intestinal epithelial cells. The A subunit has N-glycosidase activity and contributes to cell death by inhibiting protein synthesis at the level of 28S ribosomal RNA. 3 A toxic effect on both intestinal epithelial and endothelial cells might facilitate access of the toxin to the bloodstream with subsequent systemic pathologic effects. This is important in HUS pathogenesis because there is no evidence for bacteremia in human disease except for 2 isolated case reports.20,21 After gaining access to the circulation, there is evidence that the toxin is transported to distant sites through an interaction with neutrophils, monocytes, and platelets.3,22 Once in the circulation, the toxin binds vascular endothelial cells rich in Gb3. Each B subunit binds with high affinity to terminal galactose α1,4β galactose disaccharides in Gb3 membrane receptors.11 Glomerular, colonic, and cerebral endothelial cells are rich in Gb3 receptors, as are renal mesangial and Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C S83 S83 Blackall and Marques / HEMOLYTIC UREMIC SYNDROME REVIEW tubular epithelial cells.11 This might account for the localization of tissue damage in HUS (ie, renal predominant). Shiga toxin induces a wide variety of effects that are directly related to the toxin but also result from the host inflammatory response to the toxin and toxin-induced injury. The direct toxic effect results primarily from an inhibition of protein synthesis.3 Stx subunit A undergoes partial proteolysis and disulfide bond reduction in kidney cells. These processes generate an enzyme (N-glycosidase) that cleaves an adenine from 28S ribosomal RNA and inhibits the elongation of peptide chains. In so doing, protein synthesis is disrupted and cell death ensues. Stx also has been shown to induce apoptosis in endothelial cells and in renal tubular and intestinal epithelial cells, both in vitro and in vivo.3,23,24 In addition, Stx has been shown to up-regulate a number of transcription factors in human vascular endothelial cells that belong to the tumor necrosis factor (TNF)-stress-related signaling pathway and the NFκB pathway.3 These transcription factors have important roles in apoptosis and in immune regulation, cell proliferation, and the regulation of proinflammatory cytokines. Recent evidence strongly suggests that Shiga toxin works in concert with other agents to induce HUS and upregulates the expression of additional factors that might have important roles in HUS pathogenesis.3 For example, animal models (mouse, rabbit, baboon) indicate that LPS might be involved in the pathogenesis of HUS.25 In baboons, the previous administration of LPS can prime animals to develop severe HUS following the administration of otherwise subtoxic doses of Shiga toxin.26 Similarly, pretreatment of rabbits and mice with Stx enhanced the lethal effects of LPS.25 This same synergistic effect also has been demonstrated in vitro using human vascular endothelial cells.27 In humans, however, the pathophysiologic role of LPS and HUS is less clear, because endotoxemia has not been demonstrated in patients with E coli–associated HUS.3 However, there is at least indirect evidence that LPS enters the human circulation and participates in HUS pathogenesis, because antibodies to LPS can be demonstrated in up to 90% of patients with Stx-mediated HUS.28 Regardless of the role of LPS in HUS, it is becoming clear that a variety of proinflammatory and procoagulant mediators might be important causes of tissue injury in HUS. Children with D+ HUS have increased circulating levels of granulocyte colony-stimulating factor, TNF-α, interleukin (IL)-1β, IL-6, IL-8, and monocyte chemoattractant protein1.3,12 Although these findings do not prove that elevated levels of inflammatory mediators have a pathogenic role in HUS, they indicate a marked host inflammatory response. Shiga toxins might be responsible for up-regulating the genes that give rise to these mediators and might provoke the direct release of these mediators from the cells to which S84 S84 Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C Shiga toxins bind (eg, colonic epithelial cells and multiple cell types in the kidney). In addition to recruiting inflammatory cells (neutrophils) to tissue sites that have experienced Shiga toxin–induced injury, it has been shown that certain cytokines (TNF-α, IL6, and IL-8) up-regulate the expression of Gb3 on renal endothelial cells, potentially promoting further Shiga toxin binding and subsequent injury.11,29 A recent study also showed that Shiga toxin induces expression of P-selectin and intercellular adhesion molecule-1 on human umbilical vein endothelial cells.12 These adhesion molecules are important in tethering circulating leukocytes to endothelium before extravasation to sites of tissue injury. Interestingly, this study demonstrated adhesion molecule up-regulation in response to Stx 2 but not to Stx 1. This might represent at least 1 reason for the well-established fact that systemic complications more often occur with Stx 2–producing strains of enterohemorrhagic E coli than with Stx 1 producers. There is growing evidence that Shiga toxin, either directly or through the elaboration of a mediator, might induce a prothrombotic state in patients with HUS. Stx 1, through its B (binding) subunit, binds Gb3 receptors on platelets, with subsequent internalization of the toxin and platelet activation.22 Stx 1 also has been shown to increase fibrinogen binding to platelets, which might further promote the development of the platelet-fibrin thrombi found in HUS.22 Stx 1 also induced the binding of platelets to human vascular endothelial cells pretreated with TNF-α.30 Through a direct toxic effect on the endothelium, Shiga toxin might expose the subendothelium, which contains collagen and von Willebrand factor. This would further enhance platelet adhesion, platelet aggregation, and fibrin formation (through the local exposure of tissue factor and resultant downstream effects). Thus, it is easy to see how Shiga toxin could directly contribute to thrombosis via endothelial injury and platelet activation. However, it is also important to note that patients with E coli infections in whom HUS ultimately develops might have prothrombotic coagulation abnormalities that precede a clinical diagnosis of HUS, when the hematocrit, platelet count, and serum creatinine concentration are normal.6 At this point, the mechanism for this prothrombotic state has not been elucidated but might represent a target to prevent the progression of HUS in patients with diagnosed Shiga toxin–associated enteric infections. D– Hemolytic Uremic Syndrome Non–Shiga toxin–associated, D– HUS accounts for 95% of all cases of HUS that occur in adults.2 Unlike the more common childhood form of the disease, D– HUS is associated with a higher mortality rate, a higher rate of treatment failure and relapse, and a high rate of progression to endstage renal disease (50%-100% depending on the case © American Society for Clinical Pathology Pathology Patterns Reviews series).1,2 In addition, patients with atypical D– HUS who receive a renal transplant have a very high rate of disease recurrence, in some series close to 100%. 31 This is in contrast with children with end-stage renal disease secondary to D+ HUS: the rate of HUS recurrence is extremely rare. All of these differences suggest that the underlying pathogenesis of D+ and D– HUS is different. At the same time, it is clear that activation of microvascular endothelium has a central role in HUS irrespective of the underlying cause of disease. Furthermore, the clinical disease is essentially indistinguishable: microangiopathic hemolytic anemia, thrombocytopenia, and renal failure. Non–Shiga toxin–associated, D– HUS can be sporadic and idiopathic or familial and relapsing or occur secondary to a wide variety of conditions, including pregnancy (postpartum HUS), HIV infection, autoimmune disorders (antiphospholipid syndrome, systemic lupus erythematosus, scleroderma), cancer, and transplantation (hematopoietic progenitor cell and solid organ).2 Certain drugs also have been associated with the development of HUS, particularly in transplant recipients (eg, cyclosporine A, tacrolimus, methotrexate).31 In general, the pathogenesis of D– HUS is not well understood and has not been studied extensively. The one exception is cases that are familial and relapsing. These cases are highly associated with mutations in a complement regulatory protein termed factor H.32 Of all cases of atypical D– HUS, 10% to 30% are associated with factor H mutations; consequently, the term factor H–associated HUS has been coined.33 Factor H is a 150-kd multifunctional plasma glycoprotein that has a pivotal role in regulating the alternative pathway of complement activation.32 It acts as a cofactor for factor I in the degradation of newly formed C3b molecules and exhibits decay-accelerating activity in controlling the formation and stability of C3 convertase (C3bBb). The secreted plasma protein is organized into 20 homologous units termed short consensus repeats. In addition to its role as a complement regulatory protein, factor H has anti-inflammatory properties, is a ligand for a number of plasma proteins (C-reactive protein, adrenomedullin, osteopontin), and binds extracellular matrix proteins. Detailed structurefunction studies have defined 3 distinct binding regions for the ligands: C3b, glycosaminoglycans, and heparin.34 A deficiency of functional factor H in plasma is associated with recurrent microbial infections, membranoproliferative glomerulonephritis type II (MPGN II) and HUS. 32 Recent evidence strongly suggests that factor H mutations are associated with atypical D– HUS and that these mutations are almost entirely restricted to short consensus repeat 20, a “hot spot” for the disease process.32 These gene mutations have been identified in sporadic and familial cases of HUS. It is not completely clear how deficiencies of factor H result in disease, but it is known that in MPGN II, both factor © American Society for Clinical Pathology H alleles are inactivated and that the protein is absent from the plasma.35 In contrast, in atypical HUS, suboptimal factor H activity is the result of 1 intact and 1 defective allele.32 Thus, it seems that the local level of factor H is highly relevant to the manifest disease process. In MPGN II, the glomerular membrane is damaged owing to the absence of endogenous complement regulators, and there is no plasma factor H to serve this function. In atypical HUS, suboptimal factor H activity is sufficient to maintain the integrity of the glomerular and subendothelial membranes under normal conditions. However, upon insult (any of the conditions associated with D– HUS), endothelial cells are injured, complement is activated, and there is insufficient factor H to dampen down the local inflammatory response. Thus, the activity of factor H becomes pivotal in determining the balance between tissue repair and tissue damage. Although the specific mechanisms that give rise to factor H–associated HUS are unknown, knowledge of this factor in HUS pathogenesis might serve as a springboard for the development of rational approaches to therapy. Treatment Standard therapy for Shiga toxin–associated HUS is primarily supportive and designed to reverse renal failure and control hypertension (as it occurs).1 Peritoneal dialysis or hemodialysis should be considered when fluid and electrolyte imbalances cannot be corrected by conservative measures or when fluid overload compromises cardiac function. As a general principle, when the serum urea nitrogen level exceeds 100 mg/dL (35.7 mmol/L), dialysis should be considered, even in the absence of fluid and electrolyte imbalances. Hypertension should be treated (eg, short-acting calcium channel blockers, angiotensin-converting enzyme inhibitors) to prevent congestive heart failure and encephalopathy. Nonstandard therapies for HUS include antiplatelet agents, anticoagulants, thrombolytic agents, prostacyclin, intravenous immune globulin, and corticosteroids.2 To date, these therapies have not proven beneficial. There is a limited role for transfusion medicine intervention in cases of D+ HUS. RBC transfusion might be necessary for patients with more severe hemolysis and symptomatic anemia. Platelet transfusion typically is not indicated because these patients generally do not have a generalized bleeding diathesis. In addition, there is at least a theoretical risk that platelet transfusion could contribute to enhanced microthrombosis. Platelet transfusions are best reserved for thrombocytopenic patients who have demonstrable bleeding or who must have an invasive surgical procedure (eg, catheter placement for hemodialysis or peritoneal dialysis). Finally, unlike TTP, there does not seem to be a role for Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C S85 S85 Blackall and Marques / HEMOLYTIC UREMIC SYNDROME REVIEW plasma exchange in the treatment of children with D+ HUS, although it is important to note that there has not been a controlled prospective study in this patient population. In contrast, there is at least suggestive evidence that therapeutic plasma exchange might be of benefit to elderly patients with D+ HUS.36 These patients tend to have disease with more significant extrarenal (ie, neurologic) manifestations, and there is a high overall mortality. In such patients, plasma exchange might represent a rational therapeutic approach. On the research front, there has been interest in determining whether a Shiga toxin binding agent could favorably affect the course and outcome of D+ HUS. A recent report described the use of orally administered silicon particles linked to the Gb3 molecule, the receptor for Shiga toxin.37 The rationale for this study was that intestinal absorption of Shiga toxin does not occur as a one-time event but, rather, continues after the onset of HUS. Binding and elimination of elaborated Shiga toxin could have a beneficial effect. Unfortunately, this did not seem to be the case. It simply might be that there is little free Shiga toxin available owing to the tight binding of E coli to the intestinal epithelial surface. Alternatively, it might be that once HUS becomes evident, the underlying disease process (and the related host response) is too advanced to be altered by simply removing the inciting agent (Shiga toxin). It might be that an orally administered Shiga toxin binding agent would have greatest benefit in preventing the progression to HUS in patients with documented Shiga toxin–associated enteric infections, but such a study has not been reported. The optimal management for patients with D– HUS is much less clear than that for patients with Shiga toxin–associated disease. Patients with D– HUS represent a heterogeneous population, from the standpoint of inciting etiologic agent and, possibly, pathogenesis. These patients have significantly higher chronic morbidity (eg, end-stage renal disease) and mortality in comparison with patients with D+ HUS. Historically, patients with D– HUS have been treated with intensive therapeutic plasma exchange; however, relapse rates are high, and this therapy is of unproven benefit in preventing chronic renal failure.2 Unfortunately, and as previously mentioned, patients with D– HUS who receive renal transplants also have a very high rate of disease relapse with subsequent loss of the transplanted organ.31 Plasma therapy (infusion or exchange) would seem to be a rational therapeutic approach in the subset of patients with D– HUS found to have a factor H abnormality. However, there are no controlled studies that provide proof for this therapy. Prevention and New Horizons It is clear that the best way to prevent D+ HUS is to prevent primary infection with Shiga toxin–producing S86 S86 Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C organisms capable of causing the disease. This is a pressing and important public health concern. Just as clear is the fact that D– HUS prevention strategies will not likely be forthcoming until more is known about the pathogenesis of these heterogeneous disorders. Thus, this discussion will be limited to interventions to prevent D+ HUS. In patients who already are infected with Shiga toxin–producing bacteria, a number of strategies have been studied to prevent the development of HUS (known to occur in about 15% of all who are infected). Of these, antibiotic treatment of E coli O157:H7 infections is the best studied and, perhaps, most controversial. 38 The rationale for providing antibiotics to prevent HUS progression is obvious: eliminate the bacteria that produce the disease-causing toxin. On the other hand, there is equally valid concern that antibiotic therapy might, in fact, promote progression to HUS: (1) Bacteria might release toxin as a result of antibiotic-induced injury. (2) Treatment with antibiotics might give E coli O157:H7 a selective growth advantage if these organisms are eliminated less readily by treatment in comparison with normal gut flora. (3) Several antimicrobial agents (quinolones, trimethoprim, furazolidone) are potent inducers of Shiga toxin 2 gene expression, which could increase toxin levels in the intestine. A number of studies have demonstrated a positive effect or no association between antibiotic treatment of infection and the subsequent development of HUS.38 A recent metaanalysis also demonstrated that antibiotic therapy was not associated with HUS.39 However, a prospective cohort study of 71 children with documented E coli O157:H7 infections (14% of whom developed HUS) showed that in 56% of children treated with antibiotics, HUS developed, whereas it developed in only 8% of those who were not treated.40 It seems that a randomized trial will be required to conclusively determine whether antibiotic treatment of E coli O157:H7 enteritis increases the risk of HUS. In addition to antibiotics, other HUS prevention strategies are being studied, including Shiga toxin vaccines. A recent report described the development of DNA vaccines to prevent HUS in a murine model of the disease.41 Eukaryotic plasmids expressing the B subunit of Stx 2 or the B and A subunits together were evaluated for their ability to elicit in vitro humoral responses and a protective in vivo immune response. The B+A plasmid was most efficacious and was potentiated when a plasmid encoding granulocytemacrophage colony-stimulating factor was coadministered. This strategy underscores the fact that it may be possible to prevent HUS or alter the course of established disease by inducing a potent neutralizing humoral immune response. This work follows other animal studies demonstrating that either the active induction of Shiga toxin antibodies or their passive provision can prevent or ameliorate HUS.42,43 © American Society for Clinical Pathology Pathology Patterns Reviews Conclusions The classic clinical presentation of HUS is characterized by microangiopathic hemolytic anemia, thrombocytopenia, and renal failure. These features reflect the underlying histopathologic lesion: platelet-fibrin thrombi that predominate in the renal microvasculature. HUS most commonly affects children younger than 5 years and is associated with Shiga toxin–producing enteric bacteria, the most important of which is E coli O157:H7. In this setting, HUS is epidemic and also might affect adults, particularly elderly people. Sporadic cases of HUS (D– disease) more commonly occur in adults and are associated with a wide variety of inciting agents and conditions. Although the disease manifestations are similar, different pathogenic mechanisms and responses to therapy indicate that they might represent 2 separate diseases, with D– HUS sometimes difficult to distinguish from TTP. As more is understood about the underlying mechanisms giving rise to the diseases that we now lump together as HUS, more efficacious and rational treatment and prevention strategies are likely to follow. From the Departments of Pathology, 1University of Arkansas for Medical Sciences, Little Rock, and 2University of Alabama at Birmingham. Address reprint requests to Dr Blackall: Dept of Pathology, Arkansas Children’s Hospital, 800 Marshall St, Slot 820, Little Rock, AR 72202. References 1. Corrigan JJ, Boineau FG. Hemolytic-uremic syndrome. Pediatr Rev. 2001;22:365-369. 2. Allford SL, Hunt BJ, Rose P, et al, and the Haemostasis and Thrombosis Task Force, British Committee for Standards in Haematology. Guidelines on the diagnosis and management of the thrombotic microangiopathic haemolytic anaemias. Br J Haematol. 2003;120:556-573. 3. Proulx F, Seidman EG, Karpman D. Pathogenesis of Shiga toxin–associated hemolytic uremic syndrome. Pediatr Res. 2001;50:163-171. 4. Kelton JG. Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome: will recent insight into pathogenesis translate into better treatment? Transfusion. 2002;42:388-392. 5. Garg AX, Suri RS, Barrowman N, et al. Long-term renal prognosis of diarrhea-associated hemolytic uremic syndrome: a systematic review, meta-analysis, and meta-regression. JAMA. 2003;290:1360-1370. 6. Chandler WL, Jelacic S, Boster DR, et al. Prothrombotic coagulation abnormalities preceding the hemolytic-uremic syndrome. N Engl J Med. 2002;346:23-32. 7. Studt J-D, Kremer Hovinga JA, Alberio L, et al. von Willebrand factor–cleaving protease (ADAMTS-13) activity in thrombotic microangiopathies: diagnostic experience 2001/2002 of a single research laboratory. Swiss Med Wkly. 2003;133:325-332. © American Society for Clinical Pathology 8. Veyradier A, Obert B, Haddad E, et al. Severe deficiency of the specific von Willebrand factor–cleaving protease (ADAMTS 13) activity in a subgroup of children with atypical hemolytic uremic syndrome. J Pediatr. 2003;142:310317. 9. Hosler GA, Cusumano AM, Hutchins GM. Thrombotic thrombocytopenic purpura and hemolytic uremic syndrome are distinct pathologic entities: a review of 56 autopsy cases. Arch Pathol Lab Med. 2003;127:834-839. 10. Carter AO, Borczyk AA, Carlson JA, et al. A severe outbreak of Escherichia coli O157:H7–associated hemorrhagic colitis in a nursing home. N Engl J Med. 1987;316:1496-1500. 11. Moake JL. Thrombotic microangiopathies. N Engl J Med. 2002;347:589-600. 12. Matussek A, Lauber J, Bergau A, et al. Molecular and functional analysis of Shiga toxin–induced response patterns in human vascular endothelial cells. Blood. 2003;102:13231332. 13. Boerlin P, McEwen SA, Boerlin-Petzold F, et al. Associations between virulence factors of Shiga toxin–producing Escherichia coli and disease in humans. J Clin Microbiol. 1999;37:497-503. 14. Ostroff SM, Tarr PI, Neill MA, et al. Toxin genotypes and plasmid profiles as determinants of systemic sequelae in Escherichia coli O157:H7 infections. J Infect Dis. 1998;160:994-998. 15. Scotland SM, Willshaw GA, Smith HR, et al. Properties of strains of Escherichia coli belonging to serogroup O157 with special reference to production of Vero cytotoxins VT1 and VT2. Epidemiol Infect. 1987;99:613-624. 16. Tesh VL, Burris JA, Owens JW, et al. Comparison of the relative toxicities of Shiga-like toxins type I and type II for mice. Infect Immun. 1993;61:3392-3402. 17. Jacewicz MS, Acheson DWK, Binion DG, et al. Response of human intestinal microvascular endothelial cells to Shiga toxins 1 and 2 and pathogenesis of hemorrhagic colitis. Infect Immun. 1999;67:1439-1444. 18. Louise CB, Obrig TG. Specific interactions of Escherichia coli O157:H7–derived Shiga-like toxin II with human renal endothelial cells. J Infect Dis. 1995;172:1397-1401. 19. Cooling LL, Walker KE, Gille T, et al. Shiga toxin binds human platelets via globotriaosylceramide (Pk antigen) and a novel platelet glycosphingolipid. Infect Immun. 1998;66:43554366. 20. Krishnan C, Fitzgerald VA, Dakin SJ, et al. Laboratory investigation of an outbreak of hemorrhagic colitis caused by Escherichia coli O157:H7. J Clin Microbiol. 1987;25:1043-1047. 21. Emil S, Rockstad R, Vannix D. Hemolytic uremic syndrome in a child with burn injury. J Burn Care Rehabil. 1998;19:135-137. 22. Karpman D, Papadopoulou D, Nilsson K, et al. Platelet activation by Shiga toxin and circulatory factors as a pathogenetic mechanism in the hemolytic uremic syndrome. Blood. 2001;97:3100-3108. 23. Karpman D, Hakansson A, Perez MT, et al. Apoptosis of renal cortical cells in the hemolytic-uremic syndrome: in vivo and in vitro studies. Infect Immun. 1998;66:636-644. 24. Smith WE, Kane AV, Campbell ST, et al. Shiga toxin 1 triggers a ribotoxic stress response leading to p38 and JNK activation and induction of apoptosis in intestinal epithelial cells. Infect Immun. 2003;71:1497-1504. 25. Barrett TJ, Potter ME, Wachsmuth IK. Bacterial endotoxin both enhances and inhibits the toxicity of Shiga-like toxin II in rabbits and mice. Infect Immun. 1989;57:3434-3437. Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C S87 S87 Blackall and Marques / HEMOLYTIC UREMIC SYNDROME REVIEW 26. Siegler RL, Pysher TJ, Lou R, et al. Response to Shiga toxin1, with and without lipopolysaccharide, in a primate model of hemolytic uremic syndrome. Am J Nephrol. 2001;21:420-425. 27. Louise CB, Obrig TG. Shiga toxin–associated hemolytic uremic syndrome: combined cytotoxic effects of Shiga toxin and lipopolysaccharide (endotoxin) on human vascular endothelial cells in vitro. Infect Immun. 1992;60:1536-1543. 28. Bitzan M, Moebius E, Ludwig K, et al. High incidence of serum antibodies to Escherichia coli O157 lipopolysaccharide in children with hemolytic-uremic syndrome. J Pediatr. 1991;119:380-385. 29. Morigi M, Galbusera M, Binda E, et al. Verotoxin-1–induced up-regulation of adhesive molecules renders microvascular endothelial cells thrombogenic at high shear stress. Blood. 2001;98:1828-1835. 30. Keusch GT, Acheson DWK, Aaldering L, et al. Comparison of the effects of Shiga-like toxin I on cytokine- and butyratetreated human umbilical and saphenous vein endothelial cells. J Infect Dis. 1996;173:1164-1170. 31. Ruggenenti P. Post-transplant hemolytic-uremic syndrome. Kidney Int. 2002;62:1093-1104. 32. Zipfel PF. Hemolytic uremic syndrome: how do factor H mutants mediate endothelial damage? Trends Immunol. 2001;22:345-348. 33. Manuelian T, Hellwage J, Meri S, et al. Mutations in factor H reduce binding affinity to C3b and heparin and surface attachment to endothelial cells in hemolytic uremic syndrome. J Clin Invest. 2003;111:1181-1190. 34. Sanchez-Corral P, Perez-Caballero D, Huarte O, et al. Structural and functional characterization of factor H mutations associated with atypical hemolytic uremic syndrome. Am J Hum Genet. 2002;71:1285-1295. 35. Ault BH, Schmidt BZ, Fowler NL, et al. Human factor H deficiency: mutations in framework cysteine residues block factor H protein secretion and intracellular catabolism. J Biol Chem. 1997;272:25168-25175. S88 S88 Am J Clin Pathol 2004;121(Suppl 1):S81-S88 DOI: 10.1309/06W402EHNGVVB24C 36. Dundas S, Murphy J, Soutar RL, et al. Effectiveness of therapeutic plasma exchange in the 1996 Lanarkshire Escherichia coli O157:H7 outbreak. Lancet. 1999;354:13271330. 37. Trachtman H, Cnaan A, Christen E, et al, for the Investigators of the HUS-SYNSORB Pk Multicenter Clinical Trial. Effect of an oral Shiga toxin–binding agent on diarrheaassociated hemolytic uremic syndrome in children: a randomized controlled trial. JAMA. 2003;290:1337-1344. 38. Zimmerhackl LB. E coli, antibiotics, and the hemolyticuremic syndrome. N Engl J Med. 2000;342:1990-1991. 39. Safdar N, Said A, Gangnon RE, et al. Risk of hemolytic uremic syndrome after antibiotic treatment of Escherichia coli O157:H7 enteritis: a meta-analysis. JAMA. 2002;288:9961001. 40. Wong CS, Jelacic S, Habeeb RL, et al. The risk of the hemolytic-uremic syndrome after antibiotic treatment of Escherichia coli O157:H7 infections. N Engl J Med. 2000;342:1930-1936. 41. Capozzo AVE, Pistone Creydt V, Dran G, et al. Development of DNA vaccines against hemolytic-uremic syndrome in a murine model. Infect Immun. 2003;71:3971-3978. 42. Bosworth BT, Samuel JE, Moon HW, et al. Vaccination with genetically modified Shiga-like toxin IIe prevents edema disease in swine. Infect Immun. 1995;64:55-60. 43. Yamagami S, Motoki M, Kimura T, et al. Efficacy of postinfection treatment with anti-Shiga toxin (Stx) 2 humanized monoclonal antibody TMA-15 in mice lethally challenged with Stx-producing Escherichia coli. J Infect Dis. 2001;184:738-742. © American Society for Clinical Pathology