Current evidence for the role of complement in the pathogenesis of Shiga toxin haemolytic uraemic syndrome

Shiga toxin-associated haemolytic uraemic syndrome (Stx HUS) is the leading cause of paediatric acute kidney injury. This toxin-mediated disease carries a significant morbidity and mortality but has no direct treatments. Rare familial atypical HUS (aHUS) is now understood to result from over-activation of the alternative complement pathway causing glomerular endothelial damage. By understanding the pathogenic mechanisms of this disease, the monoclonal antibody eculizumab, which blocks the final common pathway of complement, is now being used to treat aHUS. For this reason, clinicians and scientists are studying the role of the alternative complement pathway in Stx HUS with the aim of targeting treatment in a similar way. There is some evidence suggesting that complement plays a role in the pathogenesis of Stx HUS, but other mechanisms may also be important. Clinically, modulating the complement system using plasma exchange provides no proven benefit in Stx HUS, and the use of eculizumab has provided conflicting results. Understanding the local effect of Stx on the glomerulus, in particular regulation of the complement and coagulation systems, may lead to advances in defining the precise pathogenesis of this disease. Then, targeted treatment strategies could be devised and clinical trials undertaken.     

Plasma therapy for atypical haemolytic uraemic syndrome associated with heterozygous factor H mutations

Atypical haemolytic uraemic syndrome (aHUS) is frequently associated with mutations in the gene encoding complement factor H (CFH). The clinical response to plasma therapy in aHUS is variable. We present here our experience of plasma therapy in three aHUS patients with CFH mutations. Three children presented aged 4, 22 and 6 months (patients 1–3 respectively) in acute kidney injury requiring dialysis. Plasma therapy consisting of plasma filtration (patient 1) or plasma exchange (PEX; patients 2 and 3) was commenced early following presentation. This resulted in aHUS remission and cessation of dialysis after 2 weeks, 9 days and 2 weeks respectively. Relapses were common and associated with increasing the interval between PEX, but all responded to intensification of PEX therapy. Patient 1 recovered 50% of renal function after first presentation. She had four relapses and started peritoneal dialysis 41 months after presentation. Mutation screening of CFH showed a missense mutation (c.3546 G > T, p.Arg1182Ser) in exon 23. PEX in patient 2 was slowly tapered over 4 months to fortnightly sessions, but she relapsed when PEX was extended to every 4 weeks. Renal function remained normal 12 months post-presentation. Mutation screening of CFH showed a mutation in exon 23 (c.3590 T > C, p.Val1197Ala) and two additional sequence variants in exons 3 and 4. Patient 3 had two relapses associated with intercurrent illnesses concurrent with reducing PEX to weekly doses. Renal function was normal 5 months post-presentation. All three patients showed a good response to PEX with improved renal function both initially and following a relapse. Further research is necessary to determine the best maintenance strategy to delay or prevent end-stage kidney disease.     

The haemolytic uraemic syndrome and anaesthesia

The Haemolytic Uraemic Syndrome is the most important cause of renal failure in infancy and childhood. The disease usually consists of a typical triad of renal failure, haemolytic anaemia, thrombocytopenia. It is, however, a multisystem disorder which may also involve the liver as well as cardiovascular, pulmonary and central nervous systems. We present a case of a female child with Haemolytic Uraemic Syndrome who presented for placement of an arteriovenous shunt to facilitate haemodialysis. The clinical presentation and management of the Haemolytic Uraemic Syndrome is discussed. Anaesthetic management of patients with the Haemolytic Uraemic Syndrome is discussed and recommendations are made.     

Recurrent haemolytic uraemic syndrome and acquired hypomorphic variant of the third component of complement

In a girl with recurrent haemolytic uraemic syndrome (HUS), persistently low serum levels of C3 were found. Analysis of complement phenotype revealed a hypomorphic variant of C3 Fast in the patient (C3fS) and a normal heterozygous pattern in both parents and the brother (C3FS). Other complement aberrations in the patient were: the presence of a null gene for C4A and C4B and low serum levels of factor H. The father also had partial factor H deficiency. It is hypothesized that the hypomorphic C3 variant may predispose to recurrent HUS. In the acquired forms the role of uraemia in alteration of C3F should be considered.     

Prostacyclin in diarrhoea-associated haemolytic uraemic syndrome

The role of prostacyclin (PGI₂) in the pathogenesis of haemolytic uraemic syndrome (HUS) is controversial. In part, confusion has been caused by failure to distinguish between two main sub-types of the syndrome: extrinsic, diarrhoea-associated HUS (D+ HUS), usually caused by infection with verocytotoxin-producingEscherichia coli orShigella dysenteriae, and the heterogeneous group of non-prodromal forms where intrinsic factors predominate (D– HUS). This paper critically reviews data confined to D+ HUS. Two methods have been used to assess PGI₂ synthesis; the generation of PGI₂ from endothelium in the presence of HUS plasma in vitro and the measurement of stable metabolites in body fluids. No concensus could be reached with regard to the former. The reported increase of PGI₂ stable metabolites in plasma may represent reduced clearance or increased carriage by plasma lipids. Apparent differences between studies of urinary excretion of PGI₂ metabolites may reflect the way excretion was expressed. If the metabolite concentration is factored for urinary creatinine, it appears that renal excretion and thus renal synthesis of PGI₂ is reduced. However, these are insufficient data on which to attribute the pathogenesis of D+ HUS to disordered PGI₂ metabolism.Presented at the Festschrift for Professor R. H. R. White on March 6, 1992, Birmingham, UK     

Serum proteins in the haemolytic uraemic syndrome

In 122 patients with the haemolytic uraemic syndrome (HUS), serum proteins were analysed in the acute phase of the disease (n = 122) and 6 weeks (n = 57) and 6 months (n = 84) later. Total serum protein levels were significantly lower on admission than 6 weeks and 6 months later (P < 0.0001). The same was true for median values of serum albumin (P < 0.0001), ₂-globulins (P < 0.0001) and -globulins (P < 0.001). There was no difference in -globulins, whereas the ₁-globulins were significantly higher in the acute phase (P < 0.0001). There was a significant positive correlation between age and total protein and -globulin levels. Serum total protein and albumin levels displayed a significant positive correlation with serum sodium levels and a significant negative correlation with urinary protein excretion. Patients with oligoanuria had significantly lower serum albumin and significantly higher ₁-globulin levels than those with preserved urine production. Marked differences were observed between patients with (D+) and patients without (D–) prodromal diarrhoea. In D(–) HUS, only albumin and total protein levels were lower on admission, but to a lesser degree than in D(+) HUS. Serum at-globulin levels were significantly higher and (₂-globulin levels significantly lower in D(+) HUS than D(–) HUS. In the D(+) subgroup of patients, by far the largest, there was a significant positive correlation between serum albumin and total protein on the one hand and the duration of the prodrome on the other. Patients with bloody stools had significantly lower serum albumin and total protein levels than those without. Faecal al-antitrypsin concentration measured in 12 HUS patients on admission was found to be significantly increased compared with age-matched controls. This study confirms the existence of hypoproteinaemia in childhood HUS and indicates that intestinal protein loss is an important, albeit not the only, physiopathological mechanism.     

Clinico-pathological findings in diarrhoea-negative haemolytic uraemic syndrome

This is a retrospective, national clinico-pathological study of past and current patients with haemolytic uraemic syndrome not associated with diarrhoea (D– HUS). Thirty-four patients were analysed and notified by members of the British Association for Paediatric Nephrology in 1998–1999. There was a 2:1 excess of males. Ten presented in infancy. The aetiology included 5 patients with complement abnormalities, 2 patients with complications of pneumococcal infection, and 2 with malignancies. Parental consanguinity was noted in 6 patients. Five children died, 9 developed chronic renal failure, and 10 end-stage renal failure. Only 7 made full recoveries. With a single exception, the pathological findings were unlike the previously reported glomerular thrombosis that is characteristic of diarrhoea-associated HUS, or HUS complicating verocytotoxin-producing Escherichia coli infection. Early and late glomerulopathy could be distinguished. Arteriolar and arterial disease was observed in 8 and 7 patients, respectively. Arterial disease correlated with a poor outcome. The pathology of D– HUS is of prognostic value, but this study was not powered to identify specific aetiological/pathological correlations.This paper was written on behalf of the British Association for Paediatric Nephrology     

Plasma renin activity in haemolytic uraemic syndrome

Plasma renin activity (PRA) was measured in 50 consecutive patients (aged 4 months to 12 years) admitted during the acute phase of the haemolytic uraemic syndrome (HUS). Blood samples were taken as soon as the diagnosis was made and prior to any diuretic, anti-hypertensive or dialysis treatment. Prodromal diarrhoea was present in all but 3 patients, 17 were anuric and 12 were oliguric. PRA ranged from 0.3 to 24.2 ng/ml per hour and was low compared with values in normal infants and children: in 13 HUS patients PRA was above the median and in 37 it was below the median. PRA was significantly, independently and inversely related to age. There was no correlation, however, with blood pressure, urine output, volume status and serum levels of sodium, potassium and creatinine. Moreover, no relationship was found between PRA and the course of the disease. Our findings do not support the idea that renin activation plays a role in the pathophysiology of the haemolytic uraemic syndrome.     

Oesophageal and severe gut involvement in the haemolytic uraemic syndrome.

Between 1982 and 1989, 78 children with diarrhoea-associated haemolytic uraemic syndrome (HUS) were referred to this hospital. Most presented with abdominal pain, bloody diarrhoea and vomiting. Seven had severe gastrointestinal involvement, four of whom required resection for bowel perforation or necrosis. One also developed an oesophageal stricture, a previously unreported complication of HUS. These seven children had a high incidence of other complications including hypertension, and cerebral and pancreatic involvement. One died from severe cerebral involvement, one has a residual neurological deficit and one has residual renal impairment. Severe gastrointestinal involvement did not significantly affect the long-term outcome. Simple haematological indices helped predict severe gut involvement. Four of the 78 children had undergone appendicectomy before the diagnosis of HUS was made. The operative findings were in no case typical of primary acute appendicitis, although histological examination did confirm inflammation of the appendix in two patients. Diagnosis is difficult in early disease, but increased awareness may help prevent unnecessary appendicectomy.