Early Differentiation of Shiga Toxin–Associated Hemolytic Uremic Syndrome in Critically Ill Adults With Thrombotic Microangiopathy Syndromes

OBJECTIVES:Thrombotic microangiopathy syndromes are a heterogeneous group of severe diseases that often require ICU admission. Prompt initiation of targeted therapies is required for atypical hemolytic uremic syndrome and thrombotic thrombocytopenic purpura, whereas there is no specific consensus therapy for Shiga toxin–associated hemolytic uremic syndrome. We sought to compare the characteristics of Shiga toxin–associated hemolytic uremic syndrome, atypical hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura patients at admission in the ICU to allow early differentiation of Shiga toxin–associated hemolytic uremic syndrome from other thrombotic microangiopathy syndromes and help to tailor initial treatment. DESIGN:Retrospective cohort study. SETTING:Two ICUs part of the French reference center for thrombotic microangiopathy syndromes. PATIENTS:Adult patients presenting with features of thrombotic microangiopathy syndromes. Other causes than Shiga toxin–associated hemolytic uremic syndrome, atypical hemolytic uremic syndrome, and thrombotic thrombocytopenic purpura were excluded. INTERVENTIONS:None. MEASUREMENTS AND MAIN RESULTS:From September 2003 to January 2017, 236 thrombotic microangiopathy syndrome patients were admitted, including 12 Shiga toxin–associated hemolytic uremic syndrome, 21 atypical hemolytic uremic syndrome, and 91 thrombotic thrombocytopenic purpura. Shiga toxin–associated hemolytic uremic syndrome patients were older than other thrombotic microangiopathy syndromes patients (64 yr [interquartile range, 50–72 yr] vs 42 yr [31–54 yr]; p = 0.007) and presented with more frequent digestive symptoms (92% vs 42%; p < 0.001), especially nonbloody diarrhea and vomiting. Biologically, Shiga toxin–associated hemolytic uremic syndrome patients displayed higher fibrinogen (490 mg/dL [460–540 mg/dL] vs 320 mg/dL [240–410 mg/dL]; p = 0.003) and creatinine levels (2.59 mg/dL [2.12–3.42 mg/dL] vs 1.26 mg/dL [0.61–1.90 mg/dL]; p < 0.001), and less marked anemia (hemoglobin level, 9.7 g/dL [8.7–11.9 g/dL] vs 7.7 g/dL [6.3–9.1 g/dL]; p < 0.001). Forty-two percent (n = 5) required renal replacement therapy, and 83% (n = 10) were treated with plasma exchange before the distinction from other thrombotic microangiopathy syndromes could be made. CONCLUSIONS:Adult Shiga toxin–associated hemolytic uremic syndrome patients are older, present more frequently with digestive symptoms and display higher hemoglobin and fibrinogen levels than other throm