Hemolytic uremic syndrome: sound minds, sick kidneys



This commentary is on the original article by Gitiaux et al. on pages 758–765 of this issue.

In antiquity, there was great debate about the seat of the soul and whether it resided in the brain, heart, or kidneys. Had hemolytic uremic syndrome (HUS) been known to the ancients, they might have reconsidered the relevance of the dispute because the disease affects all three organs with equal virulence. Kidney involvement is a defining feature of HUS, but neurological and cardiovascular complications are the leading causes of mortality.

Most cases of HUS occur after a gastrointestinal illness caused by bacteria that elaborate Shiga toxin 1 and/or 2, predominantly E. coli O157:H7. As a consequence, it is now recommended to use the term Shiga toxin-producing E. coli (STEC)-HUS rather than the older phrase, diarrhea-associated (D+)-HUS. The monumental outbreak in 2011 that was centered in Hamburg, Germany and which was caused by a unique serotype, E. coli O104:H4, is a powerful reminder that STEC-HUS is not ready to fade into history.[1] Shiga toxin has an AB5 structure and binds via the B subunit to globotriaosylceramide (Gb3) on endothelial cells. The active A moiety is transported retrograde to the Golgi apparatus where it inhibits protein synthesis and triggers ribotoxic and inflammatory responses.[2]

To date, the neurological injury in STEC-HUS is based primarily on clinical observation. Gitiaux et al. report an imaging signature that may enable better characterization of brain injury in patients with STEC-HUS.[3] In a series of seven children with severe STEC-HUS who had neurological impairment and who underwent magnetic resonance imaging (MRI) within 24 hours, they documented a reduced apparent diffusion coefficient in diffusion-weighted images in the basal ganglia and white matter with minimal changes in T2-weighted and fluid-attenuated inversion recovery sequences. They attribute these alterations to vascular injury and ischemia, a scenario fully compatible with Shiga toxin-induced microangiopathy. Furthermore, when MRI was repeated 6 weeks later, there was normalization of cerebral abnormalities. Long-term follow-up in the five surviving patients revealed no neurological or cognitive deficits. Of note, all of the patients received eculizumab therapy. Interestingly, favorable neurological outcomes have also been reported in patients affected during the recent German outbreak.[4]

This paper raises a number of tantalizing questions. Should MRI be performed routinely at the time of diagnosis of STEC-HUS to assess neurological involvement and should the findings guide therapy? This approach may be logistically challenging but it might yield better estimates of how many children have brain disease during an episode of STEC-HUS. Moreover, it might clarify the patterns of injury due to specific disturbances including azotemia, hypertension, hyponatremia, and acidosis. In addition, is this MRI pattern observed in children with atypical HUS whose disease is secondary to genetic mutations leading to abnormal activation of the alternative pathway of complement?[5] STEC-HUS and genetic forms of HUS are characterized by thrombotic microangiopathy and the findings detailed by Gitiaux et al. should be relevant to both forms of disease.

I have saved the hardest question for last. All seven patients in this series were treated with eculizumab and the temptation is great to attribute the favorable outcome in the five survivors to this monoclonal antibody that targets C5a and inhibits the alternative complement pathway. The authors acknowledge that there were two other cases of STEC-HUS with central nervous system disease that were excluded because they did not receive eculizumab and both children recovered completely. Seven is a small number of cases on which to base conclusions about prognosis, good or bad, or efficacy of a novel treatment. I will simply state that it is premature to conclude that eculizumab had a favorable impact on the course of disease in these five patients, either on their renal or neurological outcome. There is evidence of activation of the alternative pathway of complement in STEC-HUS[6] and sound rationale to prescribe eculizumab. But STEC-HUS is vastly different from genetic forms of atypical HUS because most children with the former disease survive without permanent sequelae. I am confident that my peers in pediatric nephrology would welcome the opportunity to join forces with our colleagues in neurology to implement a randomized clinical trial to test the efficacy of eculizumab in the treatment of STEC-HUS.