Shiga-toxin-producing haemolytic uraemic syndrome affects ~100 United Kingdom children each year. Around half need dialysis, a quarter develop serious complications with long-term consequences and ~3% die. No effective intervention is known; however, some studies report eculizumab, effective in atypical haemolytic uraemic syndrome, may be effective.
To determine whether the severity of Shiga-toxin-producing haemolytic uraemic syndrome is less in those given eculizumab.
Randomised, double-blind, placebo-controlled, parallel-group trial with internal pilot phase and nested mechanistic laboratory studies.
Paediatric nephrology units in 12 United Kingdom hospitals.
Children aged 6 months to < 19 years weighing ≥ 5 kg, with presumed Shiga-toxin-producing haemolytic uraemic syndrome, including ‘injury’ or ‘failure’ category of the acute kidney injury paediatric risk/injury/failure/loss/end criteria.
Participants were randomised in a 1 : 1 ratio to receive intravenous eculizumab or placebo on day 1 and 8. All received meningococcal vaccination and prophylactic antibiotics.
The primary outcome measure was a multidomain clinical severity score, reflecting morbidity until day 60. Secondary outcome measures included survival, duration of renal replacement therapy, persistent neurological defect (day 60) and presence of chronic kidney disease at 1 year. Mechanistic studies assessed complement activation and vascular endothelial growth factor profiles in plasma ± urine samples. In vitro cell co-culture work assessed the effect of Shiga toxin on endothelial cells.
Thirty-six participants from 10 sites were randomised: 17 to eculizumab and 19 to placebo. The target sample size was 134 participants – recruitment stopped early due to low recruitment (factors included reduced incidence and limited out-of-hours research infrastructure) and the COVID-19 pandemic. The mean clinical severity score for participants randomised to eculizumab was 11.5 (standard deviation 8.4) compared to 14.6 (standard deviation 7.7) for participants randomised to placebo (adjusted mean difference: −2.5, 95% confidence interval −7.8 to 2.8, = 0.3). Five participants (three eculizumab, two placebo) experienced an adverse event; there were seven serious adverse events in six participants (five eculizumab, one placebo). Urinary complement factor H and vascular endothelial growth factor levels were high initially and subsequently normalised. Shiga toxin caused a podocyte-dependent decrease in endothelial cell factor H levels.
There was no significant difference in mean clinical severity score between eculizumab and placebo groups – since the trial was underpowered, this cannot be interpreted as evidence of no effect. No significant safety concerns were observed. With further validation, the Eculizumab in Shiga-toxin-producing Haemolytic Uraemic Syndrome clinical severity score may be an outcome measure for future trials. Our results imply that Shiga toxin causes complement-dependent glomerular endothelial cell injury through its action on podocytes and subsequent cellular cross-talk.
We will continue to investigate cross talk between podocytes and endothelial cells after exposure to Shiga toxin and further develop plasma/urine biomarkers for diagnosis of Shiga-toxin-producing haemolytic uraemic syndrome.
This trial is registered as EudraCT-2016-000997-39 and ISRCTN89553116.
This award was funded by the National Institute for Health and Care Research (NIHR) Efficacy and Mechanism Evaluation (EME) programme (NIHR award ref: 14/48/43) and is published in full in ; Vol. 11, No. 11. See the NIHR Funding and Awards website for further award information.
