Pediatric Nephrology

, Volume 18, Issue 9, pp 949–951

Hemolytic uremic syndrome associated with group A beta-hemolytic streptococcus


  • Amy B. Shepherd
    • Department of PediatricsUniversity of Mississippi Medical Center
  • April L. Palmer
    • Department of PediatricsUniversity of Mississippi Medical Center
  • Steven A. Bigler
    • Department of PathologyUniversity of Mississippi Medical Center
    • Department of PediatricsUniversity of Mississippi Medical Center
    • Department of Pediatrics, Division of NephrologyUniversity of Mississippi Medical Center
Brief Report

DOI: 10.1007/s00467-003-1191-4

Cite this article as:
Shepherd, A.B., Palmer, A.L., Bigler, S.A. et al. Pediatr Nephrol (2003) 18: 949. doi:10.1007/s00467-003-1191-4


Group A beta-hemolytic streptococcal (GABS) hemorrhagic colitis due to Streptococcus pyogenes is extremely rare and its association with hemolytic uremic syndrome (HUS) in children has not been described. We report a 9-year-old white male who developed biopsy-proven HUS while continuing to have GABS-positive bloody diarrhea. Renal function deteriorated rapidly requiring intermittent hemodialysis. Three months following discharge, his renal function is normal for age except for significant proteinuria.


Hemolytic uremic syndromeGroup A beta-hemolytic streptococcusHemorrhagic colitisDialysis


Hemolytic uremic syndrome (HUS) is a common cause of acute renal failure (ARF) in children [1]. Typically, the syndrome occurs about a week following an episode of bloody diarrhea secondary to Escherichia coli O157:H7 even though other etiologies have been described [1, 2]. Streptococcus pyogenes, a Group A beta-hemolytic streptococcus (GABS), as a cause of bloody diarrhea, is extremely rare [3, 4]. To our knowledge the association between HUS and GABS in children has not been described. We report a child who developed sudden onset of ARF while having bloody diarrhea due to GABS and whose renal biopsy showed findings consistent with HUS.

Case report

A 9-year-old white male was transferred to the Children's Hospital with a 2-day history of crampy abdominal pain associated with bloody diarrhea. He had no history of fever or vomiting and had been previously healthy with no prior medical problems. There was no familial history of HUS. On admission, pulse was 102 bpm, blood pressure 104/66 mmHg, and temperature 98.2°F. The abdomen was diffusely tender with increased bowel sounds. Complete blood cell count, smear, urinalysis, renal function, and electrolytes were normal. Blood and urine cultures obtained on admission showed no growth. One stool culture done on admission and one stool culture from the outside hospital grew pure growth of GABS. No Salmonella sp., Shigella sp., Campylobacter sp., Vibrio sp., Yersinia sp., or E. coli O157:H7 were isolated. Stools for Shiga toxins 1 and 2 were negative by EIA (Meridian Bioscience, Cincinnati, OH), and serum antibody titers to 0157 LPS by ELISA were non-significant (Centers for Disease Control). The patient was admitted for intravenous hydration. The following day, despite hydration, his urine output decreased to less than 1 ml/kg/h, and he soon became anuric. BUN and serum creatinine (SCr) had risen to 35 and 2.1 mg/dl. He was transferred to the intensive care unit on hospital day 2. His renal function continued to deteriorate. BUN and SCr peaked at 119 and 11.2 mg/dl by the 10th day. Hemoglobin and platelet count decreased to values of 6.6 g/dl and 34,000/mm3 by the fifth hospital day. Blood smear showed polychromasia with the presence of schistocytes. Haptoglobin was less than 14 mg/dl (normal 30–200 mg/dl), C3 was 67 mg/dl (normal 90–180 mg/dl), and streptozyme was negative. Serum fibrinogen levels, prothrombin time, and partial thromboplastin time were normal, and the Coombs was negative. A percutaneous renal biopsy was done on hospital day 10 (Fig. 1). He continued to be anuric for a period of 11 days during which time he received intermittent hemodialysis. At discharge on day 25, his BUN and SCr were 17 and 1.1 mg/dl, urinalysis showed 1+ for protein and small for blood, hemoglobin was 7.8 g/dl, platelet count 236,000/mm3, and C3 98 mg/dl. While in the hospital, he had blood pressures as high as 150/100 mmHg, requiring therapy with a beta-blocker. One month following discharge, his blood pressure was 102/67 mmHg and he was off beta-blocker. His BUN and SCr were 18 and 0.7 mg/dl, hemoglobin 11.2 g/dl, and platelets 196,000/mm3. Urinalysis showed 2+ for protein and negative for blood. A random sample urine protein to creatinine ratio was 1.3 (normal <0.2).
Fig. 1.

A section of the renal biopsy stained with hematoxylin and eosin, demonstrating variable engorgement of the glomerular capillary loops with red blood cells and flattening of the tubular epithelium with regenerative changes

Renal pathology

The biopsy findings included 26 glomeruli, eight of which were necrotic. Seven of the glomeruli were engorged with red blood cells, and many of the glomeruli displayed double contouring of the capillary basement membranes when viewed with the periodic acid methenamine silver (PAMS) stain. One afferent arteriole contained a fibrin thrombus, while other vessels were unremarkable. The tubules were lined by flattened, regenerative epithelium. Immunofluorescence studies were essentially negative in the glomeruli, except that fibrin/fibrinogen was identified within the glomerular capillary walls and lumen. By electron microscopy, the glomeruli demonstrated prominent lucent widening of the subendothelial portion of the capillary basement membranes, with complete effacement of podocyte foot processes, fragmentation of erythrocytes, and a moth-eaten appearance of the mesangium.


Hemorrhagic colitis as a presenting symptom of GABS disease is extremely rare [3, 4]. Its occurrence should normally suggest enteric pathogens such as Shigella sp., Campylobacter sp., and E. coli O157:H7, the clinical presentation of which would be similar to our case. We believe that the bloody diarrhea in our patient was due to GABS as pure growth was isolated from the stool cultures on two occasions. Moreover, the stool was negative for Shiga toxin and the serum antibody titers to 0157 LPS by ELISA were non-significant. The source of GABS in our patient is not known. Presence of systemic GABS infection does at times result in diarrhea, but this was not the case in our patient. Positive cultures for GABS have been found in the stool and rectum of patients with streptococcal pharyngitis [5]. Our patient did not have any preceding sore throat or perianal disease.

The sudden onset of microangiopathic hemolytic anemia, thrombocytopenia, and ARF in our patient along with the renal biopsy picture certainly fits with the diagnosis of HUS. The unusual features were the age of onset and the initial normal laboratory findings followed by the sudden onset of ARF and microangiopathic hemolytic anemia while continuing to have bloody diarrhea. These findings prompted an early renal biopsy. Despite a period of anuria and marked deterioration of renal function requiring intermittent hemodialysis, his renal function is currently normal for age and he is off all antihypertensives although he continues to have 2+ proteinuria.

Low serum levels of C3 have been described in the acute phase of the illness both in typical and atypical HUS [6, 7]. The C3, however, returned to normal levels with the remission of the disease. Remuzzi et al. [8] have recently shown that the serum C3 levels in cases of familial HUS were consistently decreased as compared to the controls even during remission of the disease. They believe that the reduced C3 serum levels may indicate increased C3 consumption secondary to a genetically determined factor H deficiency, thus predisposing to microvascular thrombosis in familial HUS. Our patient had no family history of HUS and the C3 levels promptly returned to normal with the remission of the disease.

The association of GABS with HUS is not well known. GABS, however, has several virulence factors that may predispose to microangiopathy. Burns et al. [9] have described activation of an endothelial cell matrix metalloprotease by GABS extracellular cysteine protease in human umbilical vein endothelial cells. Activation of this metalloprotease resulted in endothelial cell damage. Pandiripally et al. [10] demonstrated that GABS M1 and Fba proteins bind Factor H. Decreased circulating Factor H levels may result in complement mediated endothelial cell damage secondary to decreased clearance of C3b and uncontrolled formation of the C3bBb complex [11].

To our knowledge, this is the first case report of HUS associated with GABS hemorrhagic colitis with or without bacteremia in a child. We do emphasize that GABS-associated bloody diarrhea, although extremely rare, does occur in children. Physicians taking care of such children should follow them closely for the development of HUS.

Copyright information

© IPNA 2003