Advertisement

Atypical hemolytic uremic syndrome and thrombotic thrombocytopenic purpura: Clinically differentiating the thrombotic microangiopathies

      Abstract

      The increased understanding of the pathophysiology of both atypical hemolytic uremic syndrome (aHUS) and thrombotic thrombocytopenic purpura (TTP) in recent years has led to significant therapeutic advances for both conditions. These advances have placed an increased emphasis on a more rapid differentiation of both disorders which remain clinical diagnoses. In particular, recent data demonstrating the effectiveness of complement inhibition in patients with aHUS have increased the need for a more rapid and accurate differentiation of aHUS and TTP. Previously utilized criteria have used the presence or absence of neurologic or renal injury and the pretreatment ADAMTS13 activity to differentiate aHUS from TTP. The use of presenting clinical symptoms and findings alone to differentiate these conditions is problematic given their overlapping clinical presentations. Similarly, the use of the pretreatment ADAMTS13 activity alone to differentiate aHUS from TTP is also problematic, and could lead to the inappropriate witholding of plasma exchange (PEX) therapy. However, when used collectively, the pretreatment clinical findings (symptoms and laboratory data) and ADAMTS13 activity in the context of the patient's response to PEX therapy can allow for a more effective differentiation of these two disorders in a timely fashion that will allow for the prompt initiation of the most appropriate therapy.

      Keywords

      1. Introduction

      Historically there have been numerous debates in the medical literature attempting to accurately diagnose and differentiate thrombotic thrombocytopenic purpura (TTP) and atypical hemolytic uremic syndrome (aHUS). Differentiating features of the two disorders that have been proposed previously suggested that the presence or severity of clinical symptoms including renal and neurologic injury can differentiate these two rare thrombotic microangiopathies. More recently following the discovery of the ADAMTS13 protease, it had been proposed that a severe deficiency (<5% or <10%) of the ADAMTS13 protease may correctly differentiate TTP from aHUS [
      • Tsai H.M.
      Is severe deficiency of ADAMTS-13 specific for thrombotic thrombocytopenic purpura? Yes.
      ,
      • Veyradier A.
      • Obert B.
      • Houllier A.
      • Meyer D.
      • Girma J.P.
      Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases.
      ,
      • Tsai H.M.
      • Lian E.C.
      Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
      ,
      • Tsai H.M.
      Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura.
      ]. However, despite these apparent advances in our ability to differentiate these conditions, the results of the debate were largely academic as plasma-based therapy was considered the first-line therapy for both disorders.
      This all changed however after the report regarding the effectiveness of eculizumab in a child with a diagnosis aHUS by Gruppo et al. [
      • Gruppo R.A.
      • Rother R.P.
      Eculizumab for congenital atypical hemolytic–uremic syndrome.
      ]. Despite plasma-based therapy that had been effective previously, hematologic abnormalities persisted and renal function worsened. The initiation of eculizumab therapy resulted in hematologic improvement, but more importantly improvement in renal function. Soon thereafter prospective studies of eculizumab in patients with a clinical diagnosis of aHUS were completed that resulted in the approval of eculizumab for the treatment of aHUS by the European Medicines Agency and the Federal Drug Administration (FDA) of the United States [
      • Licht C.
      • Muus P.
      • Legendre C.
      • Douglas K.W.
      • Hourmant M.
      • Delmas Y.
      • et al.
      Eculizumab is an effective long-term treatment in patients with atypical hemolytic uremic syndrome (aHUS) previously receiving chronic plasma exchange/infusion (PE/PI): extension study results.
      ,
      • Greenbaum L.
      • Babu S.
      • Furman R.R.
      • Sheerin N.
      • Cohen D.
      • Gaber O.
      • et al.
      Eculizumab is an effective long-term treatment in patients with atypical hemolytic uremic syndrome (aHUS) resistant to plasma exchange/infusion (PE/PI): results of an extension study.
      ]. These remarkable data have forced investigators to re-examine how to more accurately differentiate aHUS from TTP to allow for the prompt initiation of the most appropriate therapy for both disorders. There are certainly challenges though given that there is no objective diagnostic test to differentiate these disorders at the time of initial presentation. Both aHUS and TTP remain clinical diagnoses. The text that follows will review both the historic and more recent data regarding the use of the clinical presentation, ADAMTS13 activity, and the clinical laboratory studies that collectively may provide a framework for a more accurate differentiation of aHUS from TTP and allow for the prompt initiation of the most appropriate therapy for patients with aHUS.

      2. Clinical diagnosis and the impact on published studies

      The finding of a microangiopathic hemolytic anemia and thrombocytopenia, with or without end organ injury should alert the clinician to the possible diagnosis of a thrombotic microangiopathy (TMA) that may require the rapid initiation of PEX therapy. Historically physicians have used the findings of renal injury and neurologic abnormalities to differentiate aHUS from TTP. While to some extent this dogma may have been partially correct (more prominent renal injury in patients with aHUS), the reliance on clinical symptoms alone does not accurately differentiate these two disorders and should not be relied upon [
      • Remuzzi G.
      Is ADAMTS-13 deficiency specific for thrombotic thrombocytopenic purpura? No.
      ].
      Much of the difficulty with the published data that attempted to differentiate these two disorders stems from the inherent fallibility in applying clinical criteria to separate disorders that have overlapping presentations. This is especially true in the studies that attempted to determine the rates of severe ADAMTS13 deficiency seen in patients with aHUS or TTP. Depending upon the clinical definition used to define each disorder, the incidence of a particular finding could be relatively increased or decreased. This will always be the case when there are no objective diagnostic criteria to define each disorder.
      To avoid these potential pitfalls in this manuscript, data to be presented regarding TTP will be limited to reports from studies in which the data were analyzed in the context of the severely deficient ADAMTS13 activity at presentation. This is not meant to imply that the finding of severely deficient ADAMTS13 activity is diagnostic of acquired TTP, nor to say that non-deficient ADAMTS13 activity excludes the diagnosis, but rather to identify a homogenous population of TTP patients from which objective conclusions may be drawn. These same issues are present as well with the diagnosis of aHUS, and to a greater extent given the lack of any single objective diagnostic marker that can be reliably used to define aHUS patients. For this reason whenever possible, the criteria used to define the diagnosis of aHUS in any referenced study will be stated.

      3. Neurologic injury and the differentiation of atypical hemolytic uremic syndrome and thrombotic thrombocytopenic purpura

      Neurologic injury is a common finding at presentation in patients with TTP. In patients with a diagnosis of TTP and severely deficient ADAMTS13 activity, neurologic injury has been reported in 25%–79% of patients at presentation [
      • Zheng X.L.
      • Kaufman R.M.
      • Goodnough L.T.
      • Sadler J.E.
      Effect of plasma exchange on plasma ADAMTS13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura.
      ,
      • Coppo P.
      • Bengoufa D.
      • Veyradier A.
      • Wolf M.
      • Bussel A.
      • Millot G.A.
      • et al.
      Severe ADAMTS13 deficiency in adult idiopathic thrombotic microangiopathies defines a subset of patients characterized by various autoimmune manifestations, lower platelet count, and mild renal involvement.
      ,
      • Kremer Hovinga J.A.
      • Vesely S.K.
      • Terrell D.R.
      • Lammle B.
      • George J.N.
      Survival and relapse in patients with thrombotic thrombocytopenic purpura.
      ,
      • Jang M.J.
      • Chong S.Y.
      • Kim I.H.
      • Kim J.H.
      • Jung C.W.
      • Kim J.Y.
      • et al.
      Clinical features of severe acquired ADAMTS13 deficiency in thrombotic thrombocytopenic purpura: the Korean TTP registry experience.
      ,
      • Matsumoto M.
      • Bennett C.L.
      • Isonishi A.
      • Qureshi Z.
      • Hori Y.
      • Hayakawa M.
      • et al.
      Acquired idiopathic ADAMTS13 activity deficient thrombotic thrombocytopenic purpura in a population from Japan.
      ]. While there may be some subjectivity in the specific definition, these data support the relatively common finding of neurologic abnormalities in TTP patients. A recent report however from Coppo et al. from the French TMA registry demonstrated that the rate of CNS involvement in TMA patients was not significantly different in the ADAMTS13 deficient and non-deficient patients. If severely deficient ADAMTS13 activity in a TMA patient is used as a surrogate definition for the diagnosis of TTP, these data would suggest that rates of neurologic injury are not significantly different in TTP patients versus TMAs from other etiologies [
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      ].
      It is accepted that aHUS is a disorder with end-organ damage that primarily targets the kidney, but extrarenal involvement including neurologic injury has been reported in 10–30% of patients [
      • Sellier-Leclerc A.L.
      • Fremeaux-Bacchi V.
      • Dragon-Durey M.A.
      • Macher M.A.
      • Niaudet P.
      • Guest G.
      • et al.
      Differential impact of complement mutations on clinical characteristics in atypical hemolytic uremic syndrome.
      ,
      • Noris M.
      • Caprioli J.
      • Bresin E.
      • Mossali C.
      • Pianetti G.
      • Gamba S.
      • et al.
      Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype.
      ]. In recent years, a better understanding of aHUS as a disorder of complement dysregulation, coupled with the ability to more easily obtain complement control protein mutation analysis have been able to provide confirmation that neurologic injury can be seen in patients with aHUS. As a disorder of complement dysregulation, it is certainly plausible that there could be neurologic injury in a disease characterized by widespread complement-mediated microvascular injury. Proof of this hypothesis has come from recent case reports that have demonstrated profound neurologic injury in patients with a TMA and documented mutations of complement regulatory proteins [
      • Noris M.
      • Caprioli J.
      • Bresin E.
      • Mossali C.
      • Pianetti G.
      • Gamba S.
      • et al.
      Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype.
      ,
      • Gulleroglu K.
      • Fidan K.
      • Hancer V.S.
      • Bayrakci U.
      • Baskin E.
      • Soylemezoglu O.
      Neurologic involvement in atypical hemolytic uremic syndrome and successful treatment with eculizumab.
      ,
      • Povey H.
      • Vundru R.
      • Junglee N.
      • Jibani M.
      Renal recovery with eculizumab in atypical hemolytic uremic syndrome following prolonged dialysis.
      ]. In a more striking case, Salem et al. reported the case of a 66 year-old female with a clinical diagnosis of aHUS, with renal failure and profound neurologic injury that did not respond to PEX therapy. The diagnosis of aHUS was confirmed by the documentation of a mutation of C3 and the subsequent complete recovery after therapy with eculizumab [
      • Salem G.
      • Flynn J.M.
      • Cataland S.R.
      Profound neurological injury in a patient with atypical hemolytic uremic syndrome.
      ]. It is for these reasons that the presence or absence of neurologic findings or injury at presentation alone cannot be used to differentiate TTP from aHUS.

      4. Renal injury and the differentiation of atypical hemolytic uremic syndrome and thrombotic thrombocytopenic purpura

      As with neurologic injury previously, it has also been suggested that severe renal injury is isolated to the diagnosis of aHUS, and may be used to differentiate aHUS from TTP. While renal failure that requires hemodialysis during the acute presentation is one of the prominent findings that favors a diagnosis of aHUS over TTP, renal injury severe enough to require hemodialysis may still be seen in patients with TTP mediated by a severe deficiency of the ADAMTS13 protease. In the report by Hovinga et al., 10% of TTP patients with severely deficient ADAMTS13 activity were found to have acute renal failure (increasing serum creatinine for 2 consecutive days or increased serum creatinine and requiring hemodialysis within 7 days of diagnosis) [
      • Hovinga J.A.
      • Vesely S.K.
      • Terrell D.R.
      • Lammle B.
      • George J.N.
      Survival and relapse in patients with thrombotic thrombocytopenic purpura.
      ]. Similarly, as initially reported by Coppo et al., renal failure requiring hemodialysis in patients with ADAMTS13 deficient TTP was less common than in the cohort of patients with ADAMTS13 activity >5% (9.7% vs. 46.7%), but it still occurred in 3/31 patients [
      • Coppo P.
      • Bengoufa D.
      • Veyradier A.
      • Wolf M.
      • Bussel A.
      • Millot G.A.
      • et al.
      Severe ADAMTS13 deficiency in adult idiopathic thrombotic microangiopathies defines a subset of patients characterized by various autoimmune manifestations, lower platelet count, and mild renal involvement.
      ]. A recent update of the data from the French TMA Registry also showed that end-stage renal disease was also significantly more common in the ADAMTS13 detectable group than the ADAMTS13 deficient group (21% vs. 0%, p < 0.0001) [
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      ]. In contrast, data from Zheng et al. was consistent with the report from our institution demonstrating that renal failure at presentation requiring hemodialysis was not seen in patients with severely deficient ADAMTS13 activity, but was restricted to TMA patients with detectable ADAMTS13 activity [
      • Zheng X.L.
      • Kaufman R.M.
      • Goodnough L.T.
      • Sadler J.E.
      Effect of plasma exchange on plasma ADAMTS13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura.
      ,
      • Cataland S.R.
      • Yang S.
      • Wu H.M.
      The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies.
      ]. These data support the hypothesis that severe renal injury requiring hemodialysis is more common in patients with TMAs with detectable ADAMTS13 activity, but this criteria cannot be used alone to differentiate TTP from other TMAs given that acute renal failure can rarely be seen in ADAMTS13 deficient TTP.

      5. ADAMTS13 activity and the differentiation of atypical hemolytic uremic syndrome and thrombotic thrombocytopenic purpura

      Beginning with the discovery of the ADAMTS13 protease [
      • Tsai H.M.
      • Lian E.C.
      Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
      ,
      • Furlan M.
      • Robles R.
      • Galbusera M.
      • Remuzzi G.
      • Kyrle P.A.
      • Brenner B.
      • et al.
      von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic–uremic syndrome.
      ], there was hope that the ADAMTS13 activity could be used to objectively differentiate TTP from aHUS. And while initial reports suggested that this may be the case [
      • Veyradier A.
      • Obert B.
      • Houllier A.
      • Meyer D.
      • Girma J.P.
      Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases.
      ,
      • Tsai H.M.
      • Lian E.C.
      Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
      ,
      • Tsai H.M.
      Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura.
      ] (Table 1), a closer look at the studies and their reliance on the clinical differentiation of the cohorts of TTP and aHUS reveals the potential fallibility of relying on the ADAMTS13 activity alone to differentiate these two TMAs [
      • Remuzzi G.
      Is ADAMTS-13 deficiency specific for thrombotic thrombocytopenic purpura? No.
      ,
      • Remuzzi G.
      • Galbusera M.
      • Noris M.
      • Canciani M.T.
      • Daina E.
      • Bresin E.
      • et al.
      von Willebrand factor cleaving protease (ADAMTS13) is deficient in recurrent and familial thrombotic thrombocytopenic purpura and hemolytic uremic syndrome.
      ].
      Table 1Reported rates of severely deficient ADAMTS13 activity in patients clinically diagnosed with aHUS or TTP across different studies.
      Number of patientsPatients with severe ADAMTS13 deficiency
      Clinical categorizationaHUS/TTPaHUSTTP
      Veyradier et al.
      • Veyradier A.
      • Obert B.
      • Houllier A.
      • Meyer D.
      • Girma J.P.
      Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases.
      45/6613%89%
      Remuzzi et al.
      • Remuzzi G.
      • Galbusera M.
      • Noris M.
      • Canciani M.T.
      • Daina E.
      • Bresin E.
      • et al.
      von Willebrand factor cleaving protease (ADAMTS13) is deficient in recurrent and familial thrombotic thrombocytopenic purpura and hemolytic uremic syndrome.
      9/1255%92%
      Tsai et al.
      • Tsai H.M.
      • Lian E.C.
      Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
      ,
      • Tsai H.M.
      Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura.
      NA/127NA100%
      NA, not applicable.
      Veyradier et al. reported in 2001 a greater frequency of a deficiency of the von Willebrand factor (VWF) cleaving protease in patients with TTP compared to those with HUS (89% vs. 13%) [
      • Veyradier A.
      • Obert B.
      • Houllier A.
      • Meyer D.
      • Girma J.P.
      Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases.
      ]. While these data seem encouraging, the difficulties arise in the fact that the clinical definition of TTP or HUS was dependent upon the classification of the referring site. With the rates of VWF protease deficiency dependent upon the clinical classification, and without clear and objective criteria to clinically define TTP vs. aHUS, the results of this study and the rates of respective vWF protease deficiency could be significantly affected by the differing criteria used by the referring sites. Similarly, the ADAMTS13 activity in 127 patients clinically classified as TTP was reported by Tsai et al. to all have severely deficient ADAMTS13 activity [
      • Tsai H.M.
      • Lian E.C.
      Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
      ,
      • Tsai H.M.
      Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura.
      ]. Assumptions made though to clinically classify patients as TTP included a restriction in the level of serum creatinine, the need for hemodialysis, and age over 10 years, criteria that do not differentiate TTP and aHUS with absolute certainty. In contrast, Remuzzi et al. reported the finding of deficient ADAMTS13 activity in 5/9 (55%) patients clinically diagnosed with HUS, and 11/12 (92%) of patients characterized clinically as TTP [
      • Remuzzi G.
      • Galbusera M.
      • Noris M.
      • Canciani M.T.
      • Daina E.
      • Bresin E.
      • et al.
      von Willebrand factor cleaving protease (ADAMTS13) is deficient in recurrent and familial thrombotic thrombocytopenic purpura and hemolytic uremic syndrome.
      ], highlighting the same issues of clinical classification and their impact on reported rates of ADAMTS13 deficiency in TTP and aHUS. Despite the limitations of these studies imposed by the reliance on a clinical classification, these data support the hypothesis that a severe deficiency of the ADAMTS13 protease is more consistent with a diagnosis of TTP over aHUS. However, these data should not be interpreted to mean that measurable ADAMTS13 activity alone excludes the clinical diagnosis of TTP.
      Despite these data that suggest that deficient ADAMTS13 activity is more consistent with a diagnosis of TTP over aHUS, the absolute reliance on these data for clinical decision-making is problematic. Specifically, the use of measurable or non-deficient ADAMTS13 activity at presentation to justify withholding PEX therapy, where it may be presumed to be less effective, is a potentially dangerous undertaking. This is highlighted quite well by a retrospective study by Froehlich-Zahnd et al. who described multiple recurrent acute TTP episodes from a patient that was clinically diagnosed with HIV-associated TTP that responded to PEX therapy during each acute episode [
      • Froehlich-Zahnd R.
      • George J.N.
      • Vesely S.K.
      • Terrell D.R.
      • Aboulfatova K.
      • Dong J.F.
      • et al.
      Evidence for a role of anti-ADAMTS13 autoantibodies despite normal ADAMTS13 activity in recurrent thrombotic thrombocytopenic purpura.
      ]. Study of the ADAMTS13 biomarkers from each of these episodes demonstrated the patient to have measurable/non-deficient ADAMTS13 activity at the time of presentation by three different ADAMTS13 testing methodologies (immunoblotting-based assay, modified FRETS-VWF73 assay, and a flow-based assay), despite having demonstrable anti-ADAMTS13 antibodies. ADAMTS13 activity at the time of presentation during subsequent acute episodes was eventually found to be severely deficient by all three methodologies that coincided with a rising concentration and titer of ADAMTS13 inhibitory antibodies. These data suggest that the presence of very low levels of ADAMTS13 inhibitory antibodies may be sufficient to lead to an acute episode of TTP, but may not be sufficient to make in vitro assays of ADAMTS13 activity demonstrate severe ADAMTS13 deficiency. These data illustrate the limitations of using an in vitro assay in the diagnosis of TTP, and a potential pitfall in using the ADAMTS13 activity alone as a reason to withhold PEX therapy.

      6. Clinical laboratory data and ADAMTS13 activity

      The rapid initiation of PEX therapy remains the first course of action in any patient with a clinical diagnosis of a TMA. The ADAMTS13 activity, although not diagnostic of TTP, should be considered an important adjunct to the clinical diagnosis of TTP, just as the laboratory testing for heparin-induced thrombocytopenia (HIT) is considered an adjunct to the clinical diagnosis of HIT. The combination of both the turnaround-time for most clinicians to obtain the results of ADAMTS13 testing from reference labs and the reluctance to rely solely on the ADAMTS13 activity alone for clinical decision-making has led to approaches to utilize pre-treatment laboratory data to predict which subjects may be more likely to have severely deficient ADAMTS13 activity. These data should not be used in lieu of ADAMTS13 testing, but should be used in concert with the ADAMTS13 activity testing to aid in its interpretation.
      Going back more than 10 years ago it was reported that the pre-treatment platelet count and serum creatinine may identify patients more likely to have severely deficient ADAMTS13 activity. Several groups have reported a significantly lower platelet count and lower serum creatinine in patients with severely deficient ADAMTS13 activity compared to patients who were not deficient (Table 2). Using a threshold of 5% to define severe ADAMTS13 deficiency, Coppo et al. reported a significantly lower platelet count (median 12 × 109/L vs. 49.5 × 109/L, p = 0.0004) and a significantly lower serum creatinine (median 98 μmol/L vs. 329 μmol/L, p < 0.0001) in the cohort of patients with severely deficient ADAMTS13 activity [
      • Coppo P.
      • Bengoufa D.
      • Veyradier A.
      • Wolf M.
      • Bussel A.
      • Millot G.A.
      • et al.
      Severe ADAMTS13 deficiency in adult idiopathic thrombotic microangiopathies defines a subset of patients characterized by various autoimmune manifestations, lower platelet count, and mild renal involvement.
      ]. A recent update of these data from the French TMA Registry again confirmed these same findings [
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      ]. In data initially reported in 2003 and updated in 2010 Kremer Hovinga et al. found the same results, with a significantly lower median platelet count (11 × 109/L v. 22 × 109/L, p = 0.0004) and lower median serum creatinine (1.6 mg/dL vs. 4.6 mg/dL) in those patients with a pre-treatment ADAMTS13 activity <10% [
      • Hovinga J.A.
      • Vesely S.K.
      • Terrell D.R.
      • Lammle B.
      • George J.N.
      Survival and relapse in patients with thrombotic thrombocytopenic purpura.
      ]. Raife et al. using an ADAMTS13 threshold of 15% found the same association of a lower platelet count and serum creatinine in patients with deficient ADAMTS13 activity compared to the non-deficient patients [
      • Raife T.
      • Atkinson B.
      • Montgomery R.
      • Vesely S.
      • Friedman K.
      Severe deficiency of VWF-cleaving protease (ADAMTS13) activity defines a distinct population of thrombotic microangiopathy patients.
      ]. While it could be argued that differing thresholds for the ADAMTS13 may affect the conclusions drawn, a more recent publication from our group addressed this issue [
      • Cataland S.R.
      • Yang S.
      • Wu H.M.
      The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies.
      ]. Utilizing a threshold for the ADAMTS13 activity of 10%, the same association of a significantly lower platelet count and serum creatinine was seen in the ADAMTS13 deficient group. In comparing the ADAMTS13 activity between the two groups in the report by Cataland et al., the mean ADAMTS13 activity in the “deficient” group was 1.7% compared to 65.1% in the non-deficient group suggesting that the groups are quite divergent in their ADAMTS13 activity, with the results of such analyses not likely to be affected by the use of 5%, 10%, or 15% thresholds for ADAMTS13 deficiency. This was also confirmed by Coppo et al. who performed a similar analysis, finding that a mean ADAMTS13 activity of 51.4% in the ADAMTS13 detectable group when using a threshold of 5% was used for to categorize as undetectable ADAMTS13 activity [
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      ].
      Table 2Reported platelet count and serum creatinine at presentation from studies of thrombotic microangiopathy patients with severely deficient and non-deficient ADAMTS13 activity.
      ADAMTS13 thresholdADAMTS13 activity
      Severely deficientNon-deficient
      Platelets (×109/L)Creatinine (mg/dL)Platelets (×109/L)Creatinine (mg/dL)
      Raife et al.
      • Raife T.
      • Atkinson B.
      • Montgomery R.
      • Vesely S.
      • Friedman K.
      Severe deficiency of VWF-cleaving protease (ADAMTS13) activity defines a distinct population of thrombotic microangiopathy patients.
      15%131.2442.7
      Coppo et al.
      • Coppo P.
      • Bengoufa D.
      • Veyradier A.
      • Wolf M.
      • Bussel A.
      • Millot G.A.
      • et al.
      Severe ADAMTS13 deficiency in adult idiopathic thrombotic microangiopathies defines a subset of patients characterized by various autoimmune manifestations, lower platelet count, and mild renal involvement.
      5%121.3504.3
      Kremer et al.
      • Kremer Hovinga J.A.
      • Vesely S.K.
      • Terrell D.R.
      • Lammle B.
      • George J.N.
      Survival and relapse in patients with thrombotic thrombocytopenic purpura.
      10%111.6224.6
      Cataland et al.
      • Cataland S.R.
      • Yang S.
      • Wu H.M.
      The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies.
      10%121.7666.7
      Extending their earlier observations, Coppo et al. performed additional statistical analysis of the presenting clinical characteristics of patients with a clinically diagnosed TMA (Table 3). The primary objective of their analysis was to establish a predictive score for severe ADAMTS13 deficiency based upon standard laboratory criteria [
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      ,
      • Noris M.
      • Caprioli J.
      • Bresin E.
      • Mossali C.
      • Pianetti G.
      • Gamba S.
      • et al.
      Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype.
      ]. In their analysis, the presenting laboratory data from 214 patients (160 deficient, 54 non-deficient) were studied using an ADAMTS13 activity threshold of 5%. As before, they found a significantly lower platelet count, serum creatinine, and higher GFR in patients with severe ADAMTS13 deficiency. They also reported the finding of a positive ANA to be significantly higher in severely deficient patients compared to those with detectable ADAMTS13 activity (53% vs. 24%, p <  0.001). The association of the platelet count (≤30 × 109/L) and serum creatinine (≤200 μmol/L) with severe ADAMTS13 deficiency was the strongest. When all three criteria were present (platelet count, serum creatinine, and ANA), the specificity was 98.1% and the positive predictive value was 98.7%. When at least one criterion was present, the sensitivity was 98.8% and the negative predictive value was 93.3%. The conclusion of the authors was that by the use of readily available clinical laboratory data, an accurate prediction of the patients to ADAMTS13 activity could be made. These data should not be used to replace ADAMTS13 activity testing, but rather to validate the initial clinical management and predict the response to PEX therapy.
      Table 3Data from Coppo et al. showing the pre-treatment laboratory characteristics and their ability to predict severe ADAMTS13 activity. The sensitivity and specificity for the finding of at least 1 or all 3 criteria are shown below with 95% confidence intervals in parentheses
      • Coppo P.
      • Schwarzinger M.
      • Buffet M.
      • Wynckel A.
      • Clabault K.
      • Presne C.
      • et al.
      Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
      .
      Deficient ADAMTS13 activityNon-deficient ADAMTS13 activityp-Value
      N = 160N = 54
      Platelet count (×109/L)1767<.0001
      Serum creatinine (mg/dL)1.35.1<.0001
      Antinuclear antibody85/160 (53%)13/54 (24%)<.001
      1 criteria presentAll 3 criteria present
      Sensitivity98.8 (96.9–100)46.9 (41.3–53.1)
      Specificity48.1 (38.9–59.3)98.1 (94.4–100)
      These associations of the presenting platelet count and serum creatinine with the pre-treatment ADAMTS13 could also be considered as confirmation of the reported ADAMTS13 activity in the context of the patient's response to PEX therapy. While the finding of a higher platelet count and serum creatinine at presentation is not diagnostic of aHUS, in the context of non-deficient ADAMT13 activity and a poor response to PEX therapy, the diagnosis of aHUS should be considered in the differential diagnosis in a patient presenting with a TMA. Support for this hypothesis comes from 4/14 patients from the ADAMTS13 >10% cohort in the report from Cataland et al. who had testing for mutations or autoantibodies against the complement regulatory proteins [
      • Cataland S.R.
      • Yang S.
      • Wu H.M.
      The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies.
      ]. Three of the 4 patients tested indeed were shown to have either mutations in or autoantibodies against complement regulatory proteins supporting the hypothesis that this subset of patients is enriched for the diagnosis of aHUS.

      7. Complement activation biomarkers and the diagnosis of atypical hemolytic uremic syndrome

      Similar to the attempts to use ADAMTS13 activity to help differentiate aHUS from TTP, it is reasonable to presume that the use of markers of complement activation might be helpful in the diagnosis of aHUS. Activation of the alternative pathway of complement activation and dysregulation of the complement activity are central to the pathophysiology of aHUS, with low levels of C3 and normal C4 providing evidence of activation of the alternative pathway. That being said, while decreased levels of C3 and/or C4 provide evidence for complement activation, normal levels of C3 and C4 do not rule out the possibility of a complement-mediated thrombotic microangiopathy [
      • Loirat C.
      • Fremeaux-Bacchi V.
      Atypical hemolytic uremic syndrome.
      ,
      • Fremeaux-Bacchi V.
      • Fakhouri F.
      • Garnier A.
      • Bienaime F.
      • Dragon-Durey M.A.
      • Ngo S.
      • et al.
      Genetics and outcome of atypical hemolytic uremic syndrome: a nationwide French series comparing children and adults.
      ]. C3 levels may also vary depending on the complement regulatory protein mutation present [
      • Noris M.
      • Caprioli J.
      • Bresin E.
      • Mossali C.
      • Pianetti G.
      • Gamba S.
      • et al.
      Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype.
      ,
      • Loirat C.
      • Fremeaux-Bacchi V.
      Atypical hemolytic uremic syndrome.
      ,
      • Fremeaux-Bacchi V.
      • Fakhouri F.
      • Garnier A.
      • Bienaime F.
      • Dragon-Durey M.A.
      • Ngo S.
      • et al.
      Genetics and outcome of atypical hemolytic uremic syndrome: a nationwide French series comparing children and adults.
      ]. Many other biomarkers of complement activation are available that may allow for a detailed evaluation of complement activation, including markers of the initiation of complement activity (Factor Bb, C4d) as well as activation of the terminal complement pathways and the formation of the membrane attack complex (C5a and C5b9). These assays however require greater expertise and standardization of sample processing to provide reliable and reproducible data that are not available routinely in clinical laboratories [
      • Harboe M.
      • Thorgersen E.B.
      • Mollnes T.E.
      Advances in assay of complement function and activation.
      ].

      8. Complement activation biomarkers and differentiating thrombotic thrombocytopenic purpura from atypical hemolytic uremic syndrome

      With the understanding of aHUS as a disorder of complement activation, it is reasonable to presume that more specific biomarkers of complement activation may be able to help in the differentiation of aHUS from TTP and other TMAs. This hypothesis however assumes that complement activation and activation of the alternative pathway are unique to aHUS. Recently published data however suggests that activation of complement activity may also play a role in the pathophysiology of ADAMTS13 deficient TTP. Réti et al. studied 23 patients with a diagnosis of TTP and ADAMTS13 activity <5% at presentation and found evidence for activation of complement as defined by significantly increased levels of C3a and C5b-9 compared to healthy controls, indicative of common and terminal pathway complement activation respectively [
      • Reti M.
      • Farkas P.
      • Csuka D.
      • Razso K.
      • Schlammadinger A.
      • Udvardy M.L.
      • et al.
      Complement activation in thrombotic thrombocytopenic purpura.
      ]. Wu et al. have also recently presented data supporting a role for complement activation in patients with acquired TTP. In 38 patients presenting with an acute TTP episode, 33/38 patients demonstrated complement activation at presentation as defined by increased levels of C3a and C5a (Wu et al., submitted for publication). Patient episodes were also divided between those that the patient survived, and those that they did not survive. The acute episodes from which the patients did not survive were characterized by significantly increased factor Bb (alternative pathway activation), C5a, and C5b-9 (terminal pathway activation) compared to those episodes from which the patients survived. An additional link between complement activation and severe ADAMTS13 deficiency was reported by Turner et al. who studied the role of ultra-large von Willebrand factor multimers (ULVWF) in complement activation. C4 released from human umbilical vein endothelial cells (HUVEC) did not attach to the ULVWF multimer strings, ruling out activation of the classical and lectin complement pathways by ULVWF. In contrast, components of the alternative pathway attached to the ULVWF multimer strings in quantitative patterns consistent with the assembly of the alternative pathway components into active complexes. This would be consistent with the data from Wu et al. and provide a potential mechanism for the activation of the alternative pathway in the non-surviving patients, and provide a plausible link between complement activation and a severe deficiency of the ADAMTS13 protease. As we continue to learn more regarding complement activation in patients with ADAMTS13 deficient-TTP, we may gain a greater understanding of the complement activation “footprint” that may be able to help differentiate aHUS from TTP, but it is presently not possible to use biomarkers of complement activation to accurately differentiate these two disorders.

      9. Clinical symptoms, laboratory data, and the ADAMTS13 and the response to plasma exchange therapy

      Despite our increased understanding of the differing pathophysiology of aHUS and TTP, both remain clinical diagnoses that are challenging to accurately differentiate. While it is not possible to differentiate these disorders with any certainty at the time of presentation, the collective use of presenting clinical symptoms, laboratory data, and the ADAMTS13 activity in the context of their response to PEX may allow for a more accurate diagnosis. Given the uncertainty of the diagnosis in most cases, and to prevent the potentially catastrophic event of withholding PEX therapy from a patient with TTP, PEX should be rapidly initiated in all patients presenting with a suspected diagnosis of aHUS or TTP. This is not undertaken with the thought that PEX is the most effective therapy for aHUS, but rather an acceptance of the limitations of our ability to differentiate aHUS from TTP with certainty. It still remains important to differentiate these two disorders in a timely manner, given recent data showing a direct relationship in aHUS patients between the time to initiate therapy with eculizumab and the time to respond to therapy [
      • Zuber J.
      • Fakhouri F.
      • Roumenina L.T.
      • Loirat C.
      • Fremeaux-Bacchi V.
      French Study Group for a HCG. Use of eculizumab for atypical haemolytic uraemic syndrome and C3 glomerulopathies.
      ].
      We have standardized our approach to the differentiation of the TMAs at our institution and our approach to therapy based on the published data presented in this review (Fig. 1). At the time of presentation, patients with a suspected TMA, a platelet count ≥30 × 109/L, an elevated serum creatinine (≥2.5 mg/dl), or the need to initiate hemodialysis at presentation raises our clinical suspicion for the diagnosis of aHUS over TTP. ADAMTS13 activity is measured at presentation, but the result is not typically known for 24 h. PEX is rapidly initiated in all patients with a plan for at least 4–5 daily procedures, with the response (or lack of response) to PEX therapy being potentially both diagnostic and therapeutic. As a surrogate marker for ongoing microthrombotic disease, the hematologic (platelet count, lactate dehydrogenase) parameters are important markers of response to therapy. While the hematologic parameters are important, of greater importance are the marker of end-organ damage (serum creatinine), neurologic findings, and symptoms when present. The response to PEX is typically assessed at day 5, with a goal of seeing both hematologic improvement and stabilization or improvement in end-organ injury. In patients clearly responding to PEX therapy, it is continued daily until normalization of both the platelet count and the LDH and then stopped without prolonged tapering.
      Figure thumbnail gr1
      Fig. 1Presented is one proposed diagnostic approach to the initial therapy in patients with a suspected TMA. Pre-treatment ADAMTS13 activity should be obtained prior to the initiation of PEX therapy. The findings of thrombocytopenia and a microangiopathic hemolytic anemia should prompt the rapid initiation of PEX therapy. The pre-treatment platelet count and serum creatinine are used to help predict the ADAMTS13 activity and the expected response to PEX. If after 4–5 days of daily PEX the response is poor, the ADAMTS13 activity is used to help determine the next course of therapy. The finding of severely deficient ADAMTS13 activity should lead to considering the addition or intensification of immunosuppressive therapy. The finding of non-deficient ADAMTS13 activity, in the context or a poor response to PEX, should lead to the consideration of an alternative cause for the TMA including aHUS.
      A poor response to PEX can be characterized by the lack of a hematologic response and worsening of end-organ injury, but also an improvement in the hematologic parameters but clearly worsening kidney injury and/or neurologic symptoms despite daily PEX therapy. In a recent review published by Zuber et al. on behalf of the French Study Group for aHUS/C3G, plasma resistance was defined as the failure to normalize the platelet count or LDH or reduce the creatinine level by at least 25% [
      • Zuber J.
      • Fakhouri F.
      • Roumenina L.T.
      • Loirat C.
      • Fremeaux-Bacchi V.
      French Study Group for a HCG. Use of eculizumab for atypical haemolytic uraemic syndrome and C3 glomerulopathies.
      ]. In these cases the pre-treatment ADAMTS13 activity can be quite helpful in directing future therapy. In poorly responding patients with severely deficient ADAMTS13 activity, this serves as confirmation of the clinical diagnosis of TTP, and provides the rationale for the addition or the intensification of immune suppressive therapy targeting the inhibitor of ADAMTS13. If however the ADAMTS13 activity is not severely deficient (with no evidence of an antibody inhibitor of ADAMTS13) in a patient responding poorly to PEX therapy, the consideration of an alternative etiology of the TMA (including aHUS) and the initiation of therapy with eculizumab should be considered.

      Learning points

      • Despite our increased understanding of the pathophysiology of aHUS and TTP, both remain clinical diagnoses.
      • Clinical symptoms alone cannot accurately differentiate aHUS from TTP as both conditions may present with overlapping findings of renal and neurologic injury.
      • Clinical findings, laboratory data, and ADAMTS13 activity data in the context of a patient's response to plasma exchange therapy will allow for the accurate and timely differentiation of TTP from other thrombotic microangiopathies.
      • The differentiation of aHUS from TTP is essential to allow for the most appropriate therapy of each disorder to improve clinical outcomes and preserve end organ function.

      Conflict of interests

      SRC and HMW have received research funding from Alexion and have received honorarium for speaking.

      References

        • Tsai H.M.
        Is severe deficiency of ADAMTS-13 specific for thrombotic thrombocytopenic purpura? Yes.
        J Thromb Haemost. 2003; 1: 625-631
        • Veyradier A.
        • Obert B.
        • Houllier A.
        • Meyer D.
        • Girma J.P.
        Specific von Willebrand factor-cleaving protease in thrombotic microangiopathies: a study of 111 cases.
        Blood. 2001; 98: 1765-1772
        • Tsai H.M.
        • Lian E.C.
        Antibodies to von Willebrand factor-cleaving protease in acute thrombotic thrombocytopenic purpura.
        N Engl J Med. 1998; 339: 1585-1594
        • Tsai H.M.
        Advances in the pathogenesis, diagnosis, and treatment of thrombotic thrombocytopenic purpura.
        J Am Soc Nephrol. 2003; 14: 1072-1081
        • Gruppo R.A.
        • Rother R.P.
        Eculizumab for congenital atypical hemolytic–uremic syndrome.
        N Engl J Med. 2009; 360: 544-546
        • Licht C.
        • Muus P.
        • Legendre C.
        • Douglas K.W.
        • Hourmant M.
        • Delmas Y.
        • et al.
        Eculizumab is an effective long-term treatment in patients with atypical hemolytic uremic syndrome (aHUS) previously receiving chronic plasma exchange/infusion (PE/PI): extension study results.
        in: ASH annual meeting abstracts. 118. 2011: 3303
        • Greenbaum L.
        • Babu S.
        • Furman R.R.
        • Sheerin N.
        • Cohen D.
        • Gaber O.
        • et al.
        Eculizumab is an effective long-term treatment in patients with atypical hemolytic uremic syndrome (aHUS) resistant to plasma exchange/infusion (PE/PI): results of an extension study.
        in: ASH annual meeting abstracts. 118. 2011: 193
        • Remuzzi G.
        Is ADAMTS-13 deficiency specific for thrombotic thrombocytopenic purpura? No.
        J Thromb Haemost. 2003; 1: 632-634
        • Zheng X.L.
        • Kaufman R.M.
        • Goodnough L.T.
        • Sadler J.E.
        Effect of plasma exchange on plasma ADAMTS13 metalloprotease activity, inhibitor level, and clinical outcome in patients with idiopathic and nonidiopathic thrombotic thrombocytopenic purpura.
        Blood. 2004; 103: 4043-4049
        • Coppo P.
        • Bengoufa D.
        • Veyradier A.
        • Wolf M.
        • Bussel A.
        • Millot G.A.
        • et al.
        Severe ADAMTS13 deficiency in adult idiopathic thrombotic microangiopathies defines a subset of patients characterized by various autoimmune manifestations, lower platelet count, and mild renal involvement.
        Medicine. 2004; 83: 233-244
        • Kremer Hovinga J.A.
        • Vesely S.K.
        • Terrell D.R.
        • Lammle B.
        • George J.N.
        Survival and relapse in patients with thrombotic thrombocytopenic purpura.
        Blood. 2010; 115: 1500-1511
        • Jang M.J.
        • Chong S.Y.
        • Kim I.H.
        • Kim J.H.
        • Jung C.W.
        • Kim J.Y.
        • et al.
        Clinical features of severe acquired ADAMTS13 deficiency in thrombotic thrombocytopenic purpura: the Korean TTP registry experience.
        Int J Hematol. 2011; 93: 163-169
        • Matsumoto M.
        • Bennett C.L.
        • Isonishi A.
        • Qureshi Z.
        • Hori Y.
        • Hayakawa M.
        • et al.
        Acquired idiopathic ADAMTS13 activity deficient thrombotic thrombocytopenic purpura in a population from Japan.
        PLoS One. 2012; 7: e33029
        • Coppo P.
        • Schwarzinger M.
        • Buffet M.
        • Wynckel A.
        • Clabault K.
        • Presne C.
        • et al.
        Predictive features of severe acquired ADAMTS13 deficiency in idiopathic thrombotic microangiopathies: the French TMA reference center experience.
        PLoS One. 2010; 5: e10208
        • Sellier-Leclerc A.L.
        • Fremeaux-Bacchi V.
        • Dragon-Durey M.A.
        • Macher M.A.
        • Niaudet P.
        • Guest G.
        • et al.
        Differential impact of complement mutations on clinical characteristics in atypical hemolytic uremic syndrome.
        J Am Soc Nephrol. 2007; 18: 2392-2400
        • Noris M.
        • Caprioli J.
        • Bresin E.
        • Mossali C.
        • Pianetti G.
        • Gamba S.
        • et al.
        Relative role of genetic complement abnormalities in sporadic and familial aHUS and their impact on clinical phenotype.
        Clin J Am Soc Nephrol. 2010; 5: 1844-1859
        • Gulleroglu K.
        • Fidan K.
        • Hancer V.S.
        • Bayrakci U.
        • Baskin E.
        • Soylemezoglu O.
        Neurologic involvement in atypical hemolytic uremic syndrome and successful treatment with eculizumab.
        Pediatr Nephrol. 2013; 28: 827-830
        • Povey H.
        • Vundru R.
        • Junglee N.
        • Jibani M.
        Renal recovery with eculizumab in atypical hemolytic uremic syndrome following prolonged dialysis.
        Clin Nephrol. 2013; ([epud ahead of print])
        • Salem G.
        • Flynn J.M.
        • Cataland S.R.
        Profound neurological injury in a patient with atypical hemolytic uremic syndrome.
        Ann Hematol. 2013; 92: 557-558
        • Hovinga J.A.
        • Vesely S.K.
        • Terrell D.R.
        • Lammle B.
        • George J.N.
        Survival and relapse in patients with thrombotic thrombocytopenic purpura.
        Blood. 2010; 115: 1500-1511
        • Cataland S.R.
        • Yang S.
        • Wu H.M.
        The use of ADAMTS13 activity, platelet count, and serum creatinine to differentiate acquired thrombotic thrombocytopenic purpura from other thrombotic microangiopathies.
        Br J Haematol. 2012; 157: 501-503
        • Furlan M.
        • Robles R.
        • Galbusera M.
        • Remuzzi G.
        • Kyrle P.A.
        • Brenner B.
        • et al.
        von Willebrand factor-cleaving protease in thrombotic thrombocytopenic purpura and the hemolytic–uremic syndrome.
        N Engl J Med. 1998; 339: 1578-1584
        • Remuzzi G.
        • Galbusera M.
        • Noris M.
        • Canciani M.T.
        • Daina E.
        • Bresin E.
        • et al.
        von Willebrand factor cleaving protease (ADAMTS13) is deficient in recurrent and familial thrombotic thrombocytopenic purpura and hemolytic uremic syndrome.
        Blood. 2002; 100: 778-785
        • Froehlich-Zahnd R.
        • George J.N.
        • Vesely S.K.
        • Terrell D.R.
        • Aboulfatova K.
        • Dong J.F.
        • et al.
        Evidence for a role of anti-ADAMTS13 autoantibodies despite normal ADAMTS13 activity in recurrent thrombotic thrombocytopenic purpura.
        Haematology. 2012; 97: 297-303
        • Raife T.
        • Atkinson B.
        • Montgomery R.
        • Vesely S.
        • Friedman K.
        Severe deficiency of VWF-cleaving protease (ADAMTS13) activity defines a distinct population of thrombotic microangiopathy patients.
        Transfusion. 2004; 44: 146-150
        • Loirat C.
        • Fremeaux-Bacchi V.
        Atypical hemolytic uremic syndrome.
        Orphanet J Rare Dis. 2011; 6: 60
        • Fremeaux-Bacchi V.
        • Fakhouri F.
        • Garnier A.
        • Bienaime F.
        • Dragon-Durey M.A.
        • Ngo S.
        • et al.
        Genetics and outcome of atypical hemolytic uremic syndrome: a nationwide French series comparing children and adults.
        Clin J Am Soc Nephrol. 2013; 8: 554-562
        • Harboe M.
        • Thorgersen E.B.
        • Mollnes T.E.
        Advances in assay of complement function and activation.
        Adv Drug Deliv Rev. 2011; 63: 976-987
        • Reti M.
        • Farkas P.
        • Csuka D.
        • Razso K.
        • Schlammadinger A.
        • Udvardy M.L.
        • et al.
        Complement activation in thrombotic thrombocytopenic purpura.
        J Thromb Haemost. 2012; 10: 791-798
        • Zuber J.
        • Fakhouri F.
        • Roumenina L.T.
        • Loirat C.
        • Fremeaux-Bacchi V.
        French Study Group for a HCG. Use of eculizumab for atypical haemolytic uraemic syndrome and C3 glomerulopathies.
        Nat Rev Nephrol. 2012; 8: 643-657