The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies


The Federal Drug Administration (FDA) approval of eculizumab (Soliris) for the treatment of patients with atypical haemolytic uremic syndrome (aHUS) has focused attention on the need for objective criteria to identify patients with aHUS who may benefit from therapy with eculizumab. Using ADAMTS13 activity alone to differentiate thrombotic thrombocytopenic purpura (TTP) from aHUS has yielded conflicting results due to differing and, admittedly, arbitrary clinical criteria to differentiate TTP from aHUS (Bianchi et al, 2002; Remuzzi et al, 2002; Remuzzi, 2003; Tsai, 2003). These studies shared the conclusion that symptoms alone cannot differentiate TTP from aHUS given the potential for overlapping clinical presentations.

Given the pathophysiology of aHUS that is distinct from TTP, identifying thrombotic microangiopathy (TMA) patients more consistent with aHUS would lead clinicians to consider therapy with eculizumab (terminal complement-binding antibody) rather than continued plasma exchange therapy (PEX). Although genetic studies documenting mutations in complement control proteins (CHF, CFI, MCP, CFB, CFHR5, and THBD) can identify patients with aHUS, these studies are not available in “real time” and do not identify all patients who may benefit from eculizumab (Mache et al, 2009; Legendre et al, 2010).

We hypothesized that those patients with a TMA unrelated to disseminated intravascular coagulation (DIC), malignancy, or transplantation, and ADAMTS13 activity (>10%), represent a population enriched for the diagnosis of aHUS. Based on this hypothesis, we reviewed the initial presentation of 54 consecutive patients referred for treatment of “TTP” between October 2002 and April 2011, with an intent to develop an algorithm to identify TMA patients in whom an alternate diagnosis to TTP [aHUS, pre-eclampsia, HELLP syndrome (Haemolytic anaemia, Elevated Liver enzymes, Low Platelet count] should be considered.

The diagnosis was based upon the finding of a microangiopathic haemolytic anaemia and thrombocytopenia (<100 × 109/l), without an alternative explanation. Patients were divided based upon pretreatment ADAMTS13 activity > or <10% and compared in terms of demographic, clinical, laboratory, and outcome data. All patients were treated with PEX therapy +/− the addition of immune suppressive therapy (corticosteroids or ciclosporin). Four patients in the ADAMTS13 activity >10% cohort also received eculizumab for a suspected diagnosis of aHUS.

The pretreatment demographic and clinical data are shown in aggregate for patients with ADAMTS13 activity <10% (n = 40) and ADAMTS13 >10% (n = 14) in Table 1. The mean platelet count and serum creatinine at presentation were significantly higher in the cohort with ADAMTS13 activity >10% compared patients with ADAMTS13 activity <10%. The need for haemodialysis (HD) at presentation was also significantly higher in the ADAMTS13 activity >10% cohort (P = 0·035), but neurological symptoms were more common in the cohort of patients with ADAMTS13 <10% (= 0·026). Genetic mutation studies were obtained in four patients in the ADAMTS13 >10% cohort; two patients tested positive for mutations in specific complement regulatory proteins, specifically CFH and CFI, and one patient had demonstrable antibodies against CFH.

Table 1. Pretreatment demographic, clinical, and laboratory data for all 54 patients at the time of their initial presentation with a diagnosis of TMA for patients based on the pretreatment ADAMTS13 activity.
 RaceSexPost-partumNeuro SymptomsAge (years)DialysisPlatelet count (150–400 × 109/l)LDH (100–190 u/l)Creatinine (μmol/l)ADAMTS13 activity
  1. LDH, lactate dehydrogenase; C, Caucasian; F, female.

ADAMTS13 <10% (n = 40)C 29/40F 26/401/4018/40420/40121262132·61·7%
ADAMTS13 >10% (n = 14)C 10/14F 12/144/141/143910/14661879512·765·1%

Of the 10 patients in the ADAMTS13 >10% cohort who required acute HD, one died and two patients failed to recover kidney function after PEX therapy. Seven patients became independent of HD, four after PEX, and three after therapy with eculizumab (900 mg IV, ×4 weeks, followed by 1200 mg every other week) after a suboptimal response to PEX.

These data presented support the hypothesis that, after exclusion of TMA due to DIC, malignancy, or transplantation, patients with ADAMTS13 activity >10% are clinically distinct from patients with a TMA and severely deficient ADAMTS13 activity. Our data are consistent with previous reports demonstrating higher platelet counts and renal failure in patients with a TMA and ADAMTS13 activity >10% (Coppo et al, 2004; Hovinga et al, 2010).

The finding of mutations of antibodies against complement proteins in 3/4 patients tested in the ADAMTS13 >10% cohort supports the hypothesis that aHUS may be overrepresented in the cohort of patients with a TMA and ADAMTS13 activity >10%. However, the fact that 4/14 patients with ADAMTS13 activity >10% presented post-partum requires consideration of additional causes of the TMA (pre-eclampsia, HELLP syndrome). A recent study in patients with autoimmune disease demonstrated mutations of MCP, CFI, and CFH in 18% of patients with pre-eclampsia and HELLP syndrome (Salmon et al, 2011). One of the four post-partum patients in the ADAMTS13 normal cohort had a mutation of CFH, raising the possibility of pre-eclampsia or HELLP as the cause for her presentation.

The finding of ADAMTS13 activity >10% alone is not diagnostic of aHUS, and should not be used as the basis to withhold PEX. Rather, measurable ADAMTS13 activity in a TMA patient should heighten a clinician's awareness to alternative diagnoses to TTP, including aHUS, pre-eclampsia, and HELLP syndrome. Currently available ADAMTS13 activity assays may not identify all patients with severely deficient ADAMTS13 activity. A report from Froehlich-Zahnd et al (2011) described a patient who, despite demonstrable ADAMTS13 autoantibodies in each of his six TTP episodes, had ADAMTS13 activity >10% at the time of his initial episode by three different methodologies. ADAMTS13 activity at subsequent episodes was later found to be severely deficient first by a flow-based methodology, and then at later episodes with an immunoblotting methodology and a FRETS-based assay, suggesting a pathophysiologic role for low titre inhibitors below the level of detection of presently available ADAMTS13 activity assays. For these reasons, the finding of in vitro ADAMTS13 activity >10% alone cannot absolutely differentiate aHUS from acquired TTP.

In the absence of an objective test to reliably diagnose aHUS, we have adopted a combination of laboratory and clinical response data to identify TMA patients who may be more clinically consistent with aHUS than TTP. At presentation, a more moderate thrombocytopenia (>30 × 109/l) and more pronounced abnormalities of renal function raises our clinical suspicion for the diagnosis of aHUS. Despite this suspicion however, PEX is still immediately initiated to avoid potentially withholding PEX therapy from a patient with acquired TTP but with an atypical presentation. A suboptimal response to PEX, defined as the absence of a steadily declining lactate dehydrogenase level and an increase in the platelet count after 4–5 days of daily PEX in the context of ADAMTS13 >10%, would lead us to consider therapy with eculizumab over intensified PEX or immune-based therapy as might be considered in TTP.


SRC and HMW designed the study, performed the research, analysed the data, and wrote the paper. SY performed the research and analysed the data.