Abstract

The guideline group was selected to be representative of UK-based medical experts. MEDLINE and EMBASE were searched systematically for publications in English, using the keywords: thrombotic thrombocytopenia purpura (TTP), ADAMTS13, plasma exchange (PEX) and relevant key words related to the subsections of this guideline. The writing group produced the draft guideline, which was subsequently revised by consensus by members of the Haemostasis and Thrombosis Task Force of the BCSH. The guideline was then reviewed by a sounding board of British haematologists, the BCSH and the British Society for Haematology Committee and comments incorporated where appropriate. The 'GRADE' system was used to quote levels and grades of evidence, details of which can be found at http://www.bcshguidelines.com. The objective of this guideline is to provide healthcare professionals with clear, up-to-date, and practical guidance on the management of TTP and related thrombotic microangiopathies, defined by thrombocytopenia, microangiopathic haemolytic anaemia (MAHA) and small vessel thrombosis. Thrombotic thrombocytopenic purpura (TTP) is rare, with a reported incidence of six cases per million per year in the UK (Scully et al, 2008). It is an important diagnosis to make because the untreated mortality is 90%, which can be reduced with the prompt delivery of plasma exchange (PEX). Early death still occurs: approximately half of the deaths in the regional UK registry occurred within 24 h of presentation, primarily in women (Scully et al, 2008). In the last 15 years there has been a marked increase in the understanding of the pathogenesis of TTP. It is now recognized that congenital and acute acquired TTP are due to a deficiency of von Willebrand factor (VWF) cleaving protein, also known as ADAMTS1, (a disintegrin and metalloproteinase with a thrombospondin type 1 motif, member 13 – von Willebrand factor cleaving protein) (Fujikawa et al, 2001; Levy et al, 2001). In the absence of ADAMTS13, ultra large multimers of VWF (ULVWF) released from endothelium are not cleaved appropriately, and cause spontaneous platelet aggregates in conditions of high shear, such as in the microvasculature of the brain, heart and kidneys. Congenital TTP is due to an inherited deficiency of ADAMTS13, but acquired immune TTP is due to the reduction of ADAMTS13 by autoantibodies directed against ADAMTS13 (Furlan et al, 1998a; Tsai & Lian, 1998). Other clinical forms of thrombotic microangiopathy (TMA) occur in the absence of severe deficiency. Diagnosis can be difficult, as there is clinical overlap with haemolytic uraemic syndrome (HUS), autoimmune disease and a spectrum of pregnancy-related problems. Thrombotic thrombocytopenic purpura was originally characterized by a pentad of thrombocytopenia, MAHA, fluctuating neurological signs, renal impairment and fever, often with insidious onset. However, TTP can present without the full pentad; up to 35% of patients do not have neurological signs at presentation and renal abnormalities and fever are not prominent features. The revised diagnostic criteria state that TTP must be considered in the presence of thrombocytopenia and MAHA alone (Galbusera et al, 2006). This can result in an increased referral of other TMAs (Table 1). TTP remains a diagnosis based on clinical history, examination of the patient and the blood film. ADAMTS 13 assays help to confirm the diagnosis and monitor the course of the disease and possible need for additional treatments. Presenting symptoms and signs are summarized in Table 2 and reflect widespread multi organ thromboses. Neurological impairment has multiple presentations including headache, altered personality, reduced cognition, transient ischaemic attacks, fits and fluctuating levels of consciousness including coma; the latter is a poor prognostic sign. Acute renal failure requiring haemodialysis is rare in TTP and more indicative of HUS (Coppo et al, 2006; Scully et al, 2008). Additional ischaemic complications may be seen, such as abdominal pain due to intestinal ischaemia. Consumption of platelets in platelet-rich thrombi results in thrombocytopenia. The median platelet count is typically 10–30 × 109/l at presentation (Dervenoulas et al, 2000; Vesely et al, 2003; Coppo et al, 2006; Tuncer et al, 2007; Scully et al, 2008). Mechanical fragmentation of erythrocytes during flow through partially occluded, high shear small vessels causes a MAHA. Median haemoglobin levels on admission are typically 80–100 g/l, with schistocytes in the film, low haptoglobin levels and raised reticulocyte counts due to haemolysis. The direct Coombs test is negative. The combination of haemolysis and tissue ischaemia produces elevated lactate dehydrogenase (LDH) values. The clotting screen (prothrombin time, activated partial thromboplastin time and fibrinogen) is usually normal. A virology screen pre-treatment is necessary to exclude human immunodeficency virus (HIV) and other viral-associated TTP, and as a baseline prior to plasma exposure. Troponin T levels are raised in 50% of acute idiopathic TTP cases (Hughes et al, 2009), highlighting that cardiac involvement is common. Raised troponin levels are a sinister finding, for coronary artery occlusion is a common mode of early death. The incidence of symptomatic heart failure is increased in patients who have been given a recent platelet transfusion (Gami et al, 2005) (Table 3). Blood must be taken prior to treatment to assess baseline ADAMTS13 activity. Severely reduced ADAMTS13 activity (5%) has been reported in a wide variety of non-TTP conditions such as uraemia, inflammatory states, post-operatively and during pregnancy (Loof et al, 2001; Mannucci et al, 2001; Moore et al, 2001). The specificity of severe ADAMTS13 deficiency (5%, who have an increased mortality (Malak et al, 2008). Pregnancy can be the initiating event for approximately 5–25% of TTP cases (Ridolfi & Bell, 1981; Vesely et al, 2004; Scully et al, 2008), which are late onset adult congenital TTP or acute idiopathic TTP. Differentiating TTP from the more common pregnancy-related TMAs, such as pre-eclampsia, HELLP syndrome (haemolysis, elevated liver enzymes, low platelets) and HUS is difficult, especially if TTP presents post-partum (Table 4). Thrombosis occurs in the placenta in untreated TTP pregnancies and results in fetal growth restriction, intrauterine fetal death and pre eclampsia. There is a continued risk of relapse during subsequent pregnancies. Women with normal levels of ADAMTS13 pre-pregnancy have a lower risk of relapse (Ducloy-Bouthors et al, 2003; Scully et al, 2006b). Drugs appear to be responsible for 150× 109/l). Tapering (reducing frequency and/or volume of PEX) has not been shown to reduce relapse rates (Bandarenko & Brecher, 1998). Cryosupernatant is at least as efficacious as standard fresh frozen plasma (FFP) (Rock et al, 1996; Brunskill et al, 2007). The UK Department of Health recommends the use of solvent/detergent-treated (S/D) plasma (O'Shaughnessy, 2006) in TTP patients to reduce the risk of transfusion-transmitted infection and adverse immune responses (Scully et al, 2007b). S/D plasma contains reduced levels of protein S, but an increased thrombotic rate has not been reported in cases where thromboprophylaxis with low molecular weight heparin (LMWH) and low dose aspirin was used routinely once the platelet count was >50 × 109/l (Scully et al, 2007b). ADAMTS13 activity is present in normal amounts in FFP, S/D plasma, methylene blue-treated FFP (MB-FFP) and psoralen-treated FFP (Yarranton et al, 2005). In the UK, single donor MB-FFP is the recommended plasma for use in all indications in those born after 1st January 1996 to minimize the risk of prion transmission (O'Shaughnessy et al, 2004). However MB-FFP has been associated with increased numbers of PEX and longer hospital stay in TTP (de la Rubia et al, 2001; Rio-Garma et al, 2008). A prospective study using psoralen–FFP compared to standard FFP showed equal efficacy and safety (Mintz et al, 2006). Plasma-related adverse events, such as allergic reactions, anaphylaxis and central venous catheter thrombosis, appeared to be more frequent prior to the use of S/D plasma (Scully et al, 2007b). 1 PEX should be started with 1·5 PV exchanges, using S/D plasma in all age groups and reassessed daily (1B). 2 The volume of exchange can be reduced to 1·0 PV when the clinical condition and laboratory test results are stabilizing (2C). 3 Intensification in frequency and or volume of PEX procedures should be considered in life-threatening cases (2B). 4 Daily PEX should continue for a minimum of 2 d after platelet count has been >150 × 10 9 /l and then stopped (2B). Plasma-derived or recombinant concentrates of ADAMTS13 are not yet available. Therefore current treatment consists of plasma infusion/exchange or the use of a virally-inactivated intermediate purity factor VIII concentrate containing ADAMTS13, such as 8Y (BPL; BioProducts Laboratory, Elstree, Herts) (Allford et al, 2000), which has a small infusion volume and can be given in the outpatient or home setting. 15–30 u/kg of 8Y has been used with reported success, although there is no guaranteed constant quantity of ADAMTS13 in such concentrates. Antibodies to ADAMTS 13 have not been detected following the use of 8Y. Despite that ADAMTS13 has a half-life of only 2–3 d (Furlan et al, 1999; Suzuki et al, 2004), the clinical effect of infusions of plasma (10–15 ml/kg) or BPL 8Y are such that infusions are required only every 10–20 d, to achieve a normal platelet and haemoglobin level. Ultimately the frequency of treatment depends on the patient's phenotype. Some require regular 'prophylactic' therapy to keep the platelet count normal and avoid relapses at times of infection and other stress situations. The phenotypically mildly affected, who have a normal platelet count most of the time, only require occasional treatment. 1 S/D plasma infusion or intermediate purity Factor VIII (eg BPL 8Y) should be used to treat congenital TTP (1C). 2 Treatment regimens for congenital TTP should be individualized according to the patient's phenotype (1A). Diagnosis of pregnancy-associated TTP is especially difficult if it develops postnatally. In any mother with a TMA, and uncertainty as to the diagnosis (and recognizing that pre-eclampsia and HELLP can present in the postnatal period), PEX should be considered. If TTP develops in the first trimester, regular PEX may allow continuation of pregnancy with delivery of a live infant (Ambrose et al, 1985; Rozdzinski et al, 1992; Mokrzycki et al, 1995; Scully et al, 2006b). Delivery is the definitive treatment of choice for pregnancy-associated TMA, although delivery does not guarantee remission of TTP. Pre-treatment ADAMTS13 assays will distinguish congenital and acquired TTP from other pregnancy-associated TMAs. In pre-eclampsia and HELLP syndrome ADAMTS13 activity is reduced (median 31% range 12–43%) but antibodies to ADAMTS13 are not found. Close liaison with an obstetrician with expertise in thrombosis and fetomaternal medicine is required. Serial fetal monitoring with uterine artery dopplers should be used to assess if there is adequate fetal growth and to assess placental blood flow. Plasma infusions alone may be sufficient in mothers with congenital TTP. However, at delivery PEX may be required to ensure adequate levels of ADAMTS13. The ideal frequency of plasma replacement during pregnancy is unknown. In acquired TTP, it is difficult to predict future relapse in pregnancy. A reduction in ADAMTS13 activity (50 × 10 9 /l) (2B). Red cell transfusion and folic acid supplementation are required during active haemolysis. It has been shown that transfusion in the critically ill is safe using a transfusion trigger of 70 g/l. However this trigger was not applicable to those with cardiac disease (Hebert et al, 1999) and, as cardiac microvascular thrombosis is a feature of TTP, a higher haemoglobin level may be required in those with evidence of cardiac involvement and acute haemolysis. Due to the risk of precipitating further thrombotic events, platelet transfusions are contra-indicated unless there is life-threatening haemorrhage. The risk of venous thromboembolism has never been formally quantified in acute TTP but is likely to be increased due to immobility and acute illness. Therefore routine LMWH thromboprophylaxis should be given once the platelet count has recovered to >50 × 109/l (Yarranton et al, 2003). Hepatitis B vaccination should be considered in TTP, once a platelet threshold of 50 × 109/l has been achieved, but studies of efficacy are required in the face of continued PEX and/or immunosuppression with rituximab. 1 Red cell transfusion should be administered according to clinical need especially if there is cardiac involvement (1A). 2 Folate supplementation is required during active haemolysis (1A). 3 Platelet transfusions are contra-indicated in TTP unless there is life-threatening haemorrhage (1A). 4 Thromboprophylaxis with LMWH is recommended once platelet count has reached >50 × 10 9 /l (1B). There is a subgroup of patients who present with TTP who subsequently show a slow or incomplete response to PEX ± corticosteroids. Refractory disease was previously arbitrarily defined as persistent thrombocytopenia or LDH elevation after a total of seven daily PEX procedures. LDH is not however, a reliable marker of disease activity. We have therefore redefined refractory disease as progression of clinical symptoms or persistent thrombocytopenia despite PEX. Intensification of PEX with the introduction of 12-hourly or double PV exchanges and the addition of further steroids have provided some benefit (Shumak et al, 1995; Bobbio-Pallavicini et al, 1997; Bandarenko & Brecher, 1998; Kahwash & Lockwood, 2004; Nguyen et al, 2008). Rituximab is the current agent of choice in refractory disease (Scully et al, 2007a). Increased frequency of PEX and addition of rituximab can be considered in refractory TTP (1B). Relapse is defined as an episode of acute TTP more than 30 d after remission, and occurs in 20–50% of cases (Shumak et al, 1995; Bandarenko & Brecher, 1998; Willis & Bandarenko, 2005). The Canadian Apheresis Group estimated that over a 10-year follow up, 36% of patients would relapse (Shumak et al, 1995). Patients with ADAMTS13 activity 15%, only 5% relapsed (Ferrari et al, 2007). The use of rituximab in an acute epsiode reduces and delays the incidence of relapse (Scully et al, 2011). Prior to discharge all patients should be counselled regarding the risk and the symptoms and signs of relapse. In patients who have had previous TTP episodes and where a reduction of ADAMTS 13 activity from detectable levels to <5% is demonstrated, elective rituximab therapy has been successfully used, with normalization of ADAMTS 13 activity (Scully et al, 2007a; Bresin et al, 2009). Patients require long-term follow up with ADAMTS 13 assay monitoring. 1 Increased PEX and/or rituximab therapy are the agents of choice in relapsing disease (1B). 2 Patients should be counselled about symptoms, signs and risk of relapse before discharge with verbal and written information (1A). 3 In patients with a documented reduction of ADAMTS 13 activity to <5%, elective therapy with rituximab can be considered (1B). Haemolytic uraemic syndrome is characterized by MAHA, thrombocytopenia and acute renal failure. It maybe associated with extensive multi-organ involvement, e.g. neurological, hepatic complications, and cardiac problems and therefore diagnostic overlap with TTP can occur. It is important to differentiate between D + HUS, atypical HUS and TTP because the prognosis and management are different (Table 5). The reader is referred to (Ariceta et al, 2009) and (Taylor et al, 2010) for further guidance in children and adults, respectively. TTP and other TMAs remain diagnostically difficult. The current challenge is to ensure that haematologists, physicians, obstetricians and paediatricians are aware of the need to treat acute TTP as a medical emergency to prevent unnecessary early mortality. The development of new drugs and recombinant proteins, trialled in the developing networks should lead to better treatments in the future. While the advice and information in these guidelines is believed to be true and accurate at the time of going to press, neither the authors, the British Society for Haematology nor the publishers accept any legal responsibility for the content of these guidelines . In 2003, the British Society for Haematology (BCSH) published the first evidence-based guidelines for the diagnosis and management of thrombotic microangiopathies (Allford et al, 2003). We have revised these based on new evidence available between 2003 and 2011. Separate guidelines for atypical haemolytic-uraemic syndrome (HUS) (Taylor et al, 2010) and diarrhoea-positive HUS (Ariceta et al, 2009) are now available, so these sections have been reduced. Dr Michael J Desborough, Academic Clinical Fellow, Oxford Deanery UK for review of Fig 1 and Ms Lucy MacKillop, Obstetric Physician, Oxford Universities Hospital Trust for review of Table 4. The Haemostasis Research unit, UCL has received an unrestricted educational grant from Octapharma, UK.