Increasing understanding of abnormalities within the alternative complement pathway in atypical hemolytic uremic syndrome (aHUS) is changing the way the disease is both diagnosed and treated. It is rarely possible to definitively diagnose aHUS at the time of initial acute presentation and treatment, with plasma exchange, is initiated on clinical grounds. With the risk of further acute episodes and increasing availability of terminal complement inhibitors accurate molecular diagnosis is imperative.
to determine the clinical phenotype and nature of complement abnormalities within a cohort of aHUS patients referred to a large thrombotic thrombocytopenic purpura (TTP) referral centre.
Data from 14 patients with a clinical diagnosis of aHUS was retrospectively analysed. 13 patients were referred with thrombotic microangiopathy not initially requiring renal replacement therapy (RRT). 1 patient presented to another institution requiring urgent RRT and was subsequently transferred to our care following recovery of renal function. All patients had ADAMTS13 levels above 30% and negative anti-ADAMTS13 antibody levels at presentation to exclude a diagnosis of TTP. 3 patients had diarrhoea at presentation; all were enterotoxin negative. Patients were subsequently assessed for mutations within complement factors H (CFH), I (CFI), B (CFB), C3 and membrane cofactor protein (MCP), at risk haplotypes and CFH antibodies. A control group of 14 acute acquired TTP patients with confirmed ADAMTS13 levels <5% were assessed for the same abnormalities within complement regulatory proteins.
In the aHUS cohort, the median age of presentation was 25.5 years (11 months to 72 years). The median serum creatinine was 275 µmol/l (range 79-1812 µmol/l), platelet count 27 x109/l (10-115) and LDH was 2016 IU (342-4621). In the TTP group, presenting creatinine was 106 µmol/l (61-353) µmol/l, platelets 13 x109/l (5-74) and LDH 1954 IU (756-3518). aHUS precipitants at initial presentation or relapse included pregnancy (n=2), upper respiratory tract infection (n=6), vaccination (n=1), abdominal sepsis (n=4). In 3 cases, there was no identified trigger. Headache was a common presenting symptom; only one hypertensive patient (72 years) had a transient ischemic attack; no other neurological events were documented in the aHUS group.
In 57% (8/14 patients) variants of the alternative complement pathway were identified; 5 with MCP mutations, encoding p.R59X, p.C157Y (present in 2 brothers), p.C64F and c.286+2T>C/c.286+2T>G (both present in the same patient); 2 with CFH mutations, encoding c.3134-5T>C and p.R1215X; and 1 with a CFB mutation (p.D371G). All of the mutations identified, except CFH c.3134-5T>C, are of clear functional significance. 2 of the patients with MCP mutations had a normal serum creatinine at presentation. C3/4 levels were low in 3/8 patients. In the control group of TTP patients with ADAMTS13 <5% no complement mutations were identified.
13/14 aHUS patients were treated initially with plasma exchange; 1 received eculizumab subsequently. 3 patients required temporary RRT and 1 died within 24 hours of presentation with progressive cardiorespiratory failure. At follow-up, all patients had platelet counts >150 x109/l and 12/13 had normal serum creatinine levels; one patient had a creatinine of 122 µmol/l. 5/13 patients had recurrent episodes, 4 of whom had confirmed complement pathway abnormalities (3 MCP mutated, 1 CFB mutated). None required long-term RRT or progressed to end-stage renal failure (ESRD) at a median follow-up of 2 years (range 0.25-28 years).
This aHUS cohort, without ESRD, demonstrates the difficulty in clinically differentiating TTP from complement mediated TMAs. We demonstrate that diagnostic differentiation based on platelet count, renal function and serum C3/C4 levels is insufficient to predict an underlying complement mutation. This distinction is increasingly important with the proven efficacy of complement inhibitor therapy in targeting complement activation in aHUS. Specifically, we demonstrate a very high frequency of functionally significant MCP mutations which mimic relapsing/remitting TTP. An ADAMTS13 activity >5% in a patient with a TMA should necessitate genetic screening for complement gene mutations prior to consideration of complement inhibitor therapy.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.