|
|||||||||
|
|
|
|
|
|
|
|
|
|
|
Prepublished online as a Blood First Edition Paper on March 13, 2003; DOI 10.1182/blood-2003-01-0193.
Blood, 1 July 2003, Vol. 102, No. 1, pp. 60-68 ADAMTS13 activity in thrombotic thrombocytopenic purpura–hemolytic uremic syndrome: relation to presenting features and clinical outcomes in a prospective cohort of 142 patientsFrom the Hematology-Oncology Section, Department of Medicine, College of Medicine, The University of Oklahoma Health Sciences Center, Oklahoma City; Department of Biostatistics and Epidemiology, College of Public Health, The University of Oklahoma Health Sciences Center, Oklahoma City; and Central Hematology Laboratory, University Hospital, Inselspital, University of Bern, Switzerland.
Initial management of patients with thrombotic thrombocytopenic purpura—hemolytic uremic syndrome (TTP-HUS) is difficult because of lack of specific diagnostic criteria, high mortality without plasma exchange treatment, and risks of plasma exchange. Although severe ADAMTS13 (a disintegrin-like and metalloprotease with thrombospondin type 1 repeats) deficiency may be specific for TTP, the role of ADAMTS13 activity measurements for initial management decisions is unknown. ADAMTS13 was measured before beginning plasma exchange treatment in 142 (88%) of 161 consecutive patients with clinically diagnosed TTP-HUS with assignment to 1 of 4 categories: less than 5% (severe deficiency), 5% to 9%, 10% to 25%, and more than 25%. Eighteen (13%) of 142 patients had severe ADAMTS13 deficiency. Among 6 predefined clinical categories (stem cell transplantation, pregnant/postpartum, drug association, bloody diarrhea, additional/alternative disorder, idiopathic), severe deficiency occurred only among pregnant/postpartum (2 of 10) and idiopathic (16 of 48) patients. The presenting features and clinical outcomes of the 16 patients with idiopathic TTP-HUS who had severe ADAMTS13 deficiency were variable and not distinct from the 32 patients with idiopathic TTPHUS who did not have severe ADAMTS13 deficiency. Many patients in all ADAMTS13 activity categories apparently responded to plasma exchange treatment. Therefore, severe ADAMTS13 deficiency does not detect all patients who may be appropriately diagnosed with TTP-HUS and who may respond to plasma exchange treatment. (Blood. 2003;102:60-68)
Thrombotic thrombocytopenic purpura (TTP) was fatal in 90% of patients prior to the availability of effective treatment with plasma exchange.1 Observations of the presenting features and clinical course of TTP suggested a pentad of clinical features for diagnosis: thrombocytopenia, microangiopathic hemolytic anemia, neurologic and renal abnormalities, and fever.1 The availability of plasma exchange treatment, allowing 81% to 96% of patients to survive,2-8 has created urgency to diagnose TTP. Now only thrombocytopenia and microangiopathic hemolytic anemia, without another apparent etiology, are sufficient criteria to establish a clinical diagnosis and begin treatment.2,3,5,8,9 These criteria do not distinguish TTP from hemolytic uremic syndrome (HUS); therefore, the comprehensive term TTP-HUS is appropriate. The decreased diagnostic threshold has resulted in a 7-fold increase of patients treated with plasma exchange for TTP-HUS from 1981 to 1997.10 However, plasma exchange has substantial risk of major complications,11,12 making the initial diagnostic and treatment decisions difficult. A severe deficiency of ADAMTS13 (a disintegrin-like and metalloprotease with thrombospondin type 1 repeats) as detected by current assays, less than 5% of normal activity, may be specific for TTP.13 Furthermore, it has been proposed that severe ADAMTS13 deficiency now defines TTP.14,15 Because ADAMTS13 deficiency caused by an autoantibody provides a possible explanation for the effectiveness of plasma exchange (removal of the autoantibody by apheresis; supply of ADAMTS13 by plasma replacement), a role for ADAMTS13 activity measurements to guide treatment decisions has been suggested.14,16-19 However, the sensitivity of severe ADAMTS13 deficiency for identifying all patients who have an appropriate clinical diagnosis of TTP and who respond to plasma exchange has not been evaluated. In 7 reports, 45% to 100% of patients described as having TTP were reported to be severely deficient in ADAMTS13 activity.19-25 Patients described as having HUS were not19,20 or were rarely23 severely deficient in ADAMTS13 activity. However, interpretation of these studies is limited by the absence of explicit criteria to distinguish patients with TTP from patients with HUS as well as the selection of many patients after their outcomes were known. Therefore, the role of ADAMTS13 activity measurements in initial diagnostic and treatment decisions for patients with suspected TTP or HUS remains unknown. Currently plasma exchange treatment is indicated for all adult patients with a clinical diagnosis of TTP or HUS, because its effectiveness was established in patients in whom the only diagnostic criteria were thrombocytopenia and microangiopathic hemolytic anemia (with or without renal failure or neurologic abnormalities).2,3 In this report we address the role of ADAMTS13 activity measurements in management decisions by describing the relation of ADAMTS13 activity to presenting features and clinical outcomes in an inception cohort of 142 patients with TTP-HUS who were identified at the time when the clinical diagnosis was established and the decision for plasma exchange treatment was made.
The Oklahoma TTP-HUS Registry The Registry includes all consecutive patients for whom the Oklahoma Blood Institute (OBI) was requested to provide plasma exchange treatment for clinically diagnosed TTP or HUS by 11 hospitals in the Oklahoma City region and other hospitals in surrounding communities. Because the OBI is the sole provider of plasma exchange services for all hospitals in central-western Oklahoma, we identified an inception cohort of consecutive patients in whom the clinical diagnosis of TTP-HUS was established and a decision to initiate plasma exchange treatment was made. The standard practice in our region is to treat all adult patients who are diagnosed as either TTP or HUS with plasma exchange, as well as children with TTP and some children with atypical HUS. Therefore, all patients diagnosed with either syndrome in central-western Oklahoma are enrolled. The only patients described by the comprehensive term TTP-HUS who are excluded from the Registry are children with typical (diarrhea-positive) HUS who are not treated with plasma exchange. All patients had diagnostic criteria for TTP-HUS2,3,5,8,9: thrombocytopenia and microangiopathic hemolytic anemia without another apparent etiology. The validity and consistency of the initial evaluation and the decision for plasma exchange treatment was provided by one of the authors (J.N.G.) who saw and actively participated in the diagnosis and initial treatment decisions for 136 (96%) of the 142 patients. The Registry has enrolled and followed prospectively all 255 consecutive patients in central-western Oklahoma who had their first episode of clinically diagnosed TTP-HUS from January 1, 1989, to December 31, 2001. Complete demographic, clinical, and laboratory data, including all available data preceding the diagnosis of TTP-HUS, are collected prospectively on standardized forms and entered into a Microsoft Access database (Redmond, WA). Serum sample collection was initiated on November 13, 1995. In this report we describe 142 of 161 consecutive patients initially diagnosed between November 13, 1995, and December 31, 2001, for whom we have measured ADAMTS13 activity. Follow-up to the present time is complete on all 142 patients. The Oklahoma TTP-HUS Registry is approved by the institutional review boards of the University of Oklahoma Health Sciences Center and each participating community hospital. Definition of clinical categories Patients were assigned in a hierarchical order on the basis of the initial clinical assessment, usually within several days of beginning plasma exchange treatment, to 1 of 6 predefined categories related to associated conditions: (1) hematopoietic stem cell transplantation, (2) pregnant/postpartum, (3) drug association, (4) bloody diarrhea prodrome, (5) presence of an additional or alternative disorder that may have caused the presenting features, and (6) idiopathic (Table 1). Assignment to categories 1, 2, and 4 was self-evident. Patients were assigned to category 3 if they were currently taking a drug previously reported to be associated with TTP-HUS. All patients were specifically asked about use of drugs reported to be associated with TTP-HUS. Among the 21 patients in this report who were assigned to category 3, the associated drugs were quinine (n = 12), mitomycin C (n = 3), cyclosporine (n = 2), ticlopidine (n = 1), gemcitabine (n = 1), pentostatin (n = 1), and carmustine (n = 1). TTP-HUS was considered to be the only clinical diagnosis responsible for the presenting features in patients in categories 1 to 4 and 6. Patients in category 5 who were described as having an additional disorder also had an established diagnosis of an autoimmune disorder26 (n = 17) or HIV infection27 (n = 2) at the time the diagnosis of TTP-HUS was made. All 142 patients had diagnostic tests for HIV infection at the time of the first plasma exchange; only 2 (1%) patients had confirmed positive tests. In these patients TTP-HUS, rather than the autoimmune disorder or HIV infection, was considered to be the principal diagnosis and the indication for plasma exchange treatment. Patients in category 5 who were described as having an alternative disorder had an unexpected alternative diagnosis made after plasma exchange was initiated that explained the presenting features and resulted in discontinuing plasma exchange treatments. Alternative disorders included sepsis (n = 11), critical care patients with multiorgan failure (n = 8), systemic malignancy (n = 5), and malignant hypertension (n = 3). Patients not assigned to categories 1 to 5 were assigned to category 6 (idiopathic TTP-HUS). Three patients initially assigned to category 6, who died within 10 days of beginning treatment, were reclassified to category 5 when autopsy revealed an unexpected alternative diagnosis: disseminated aspergillosis (n = 2) and disseminated breast cancer (n = 1). Some patients had features of more than one clinical category, but on the basis of our hierarchical method each patient was assigned to only one category. For example, one postpartum woman with an established diagnosis of systemic lupus erythematosus was assigned to category 2, and 2 patients with quinine exposure who presented with bloody diarrhea were assigned to category 3. The clinical category only describes the initial episode of TTP-HUS. For example, patient 2 (Table 2), who initially had idiopathic TTP-HUS, had an acute relapse of apparent TTP-HUS and was discovered to have systemic lupus erythematosus, but he remains in category 6; patient 8 (Table 2) had her initial episode immediately after giving birth and then had 4 relapses when she was not pregnant, but she remains in category 2. At the time of the clinical diagnosis of TTP-HUS, plasma exchange was considered by the treating hematologists to be essential treatment for each patient in all categories. Assignments to clinical categories were made without knowledge of ADAMTS13 activity and prior to data analysis.
Laboratory data Laboratory data recorded as "at presentation" were the most abnormal values on the day of diagnosis +- 7 days, to avoid transient effects of transfusions and to document the frequent worsening of anemia and renal function following diagnosis. Lactic dehydrogenase (LDH) values were adjusted to an upper limit of normal value of 200 U/L to compare data from different laboratories. Definitions of acute renal failure and severe neurologic abnormalities We defined acute renal failure as either (1) an increasing serum creatinine (≥ 44.5 µmol/L [≥ 0.5 mg/dL] per day for 2 consecutive days) or (2) a serum creatinine of 353.6 µmol/L or more (≥ 4.0 mg/dL) plus dialysis that began within 7 days of diagnosis. We defined severe neurologic abnormalities as coma, stroke, seizure, or fluctuating focal signs, such as motor deficits, diplopia, or aphasia. These abnormalities were defined as presenting features if they occurred within 7 days of diagnosis or at any time during the plasma exchange treatment course. Less severe abnormalities, such as headache, blurred vision, ataxia, or mental status changes with transient confusion, were omitted from these analyses because of difficulty with reproducible documentation. These definitions were established without knowledge of ADAMTS13 activity and prior to data analysis. Measurement of ADAMTS13 activity Serum samples were obtained immediately before beginning the first plasma exchange treatment by the OBI in 142 of 161 consecutive patients. Six of the 19 patients for whom we had no samples died immediately after the clinical diagnosis was made, before plasma exchange could be begun and samples collected; in the 13 other patients, the samples were either not obtained or misplaced by error. ADAMTS13 activity was measured as previously described,13,20 and patients were assigned to 1 of 4 categories: less than 5%, severe deficiency; 5% to 9%, borderline severe deficiency; 10% to 25%, moderate deficiency; more than 25%, minimal deficiency or normal. ADAMTS13 inhibitor activity was estimated in all samples with ADAMTS13 activity less than 5%. Results of inhibitor assays were reported as absent, trace, mild, moderate, or strong. A strong inhibitor was defined as complete inhibition of ADAMTS13 activity in a 1:1 mixture of patient serum and normal plasma. No patients had a family history of TTP or HUS; no assays were done on family members. Assays were performed in 2001-2002 without knowledge of the clinical findings at presentation or the patients' long-term outcomes. Plasma exchange treatment Our standard treatment9 was to initiate daily exchange of one plasma volume, using either fresh frozen or cryosupernatant plasma, immediately on referral. Daily plasma exchange was continued until the platelet count was 150 x 109/L or more (≥ 150 x 103 µL) for 2 consecutive days, then treatments were discontinued either gradually or abruptly. At this time, LDH values in most patients had decreased to normal or nearly normal. Examination of the peripheral blood smears for schistocytes was not routinely performed after the initial evaluation. Definitions of clinical outcomes The day of the first plasma exchange treatment was designated as the day of diagnosis. Response to treatment was defined as the achievement of a platelet count 150 x 109/L or more (≥ 150 x103/µL) during plasma exchange treatment or within 1 week of stopping treatment. It is acknowledged that in some patients recovery from thrombocytopenia may have been coincidental, not a result of the plasma exchange treatment. Exacerbation was defined as recurrent thrombocytopenia following a response plus resumption of daily plasma exchange treatment after 1 day or more but less than 30 days of no plasma exchange treatment. Remission was defined as no plasma exchange treatment for 30 days or more. Relapse was defined as the recurrence of TTP-HUS following a remission. TTP-HUS—associated death is death that occurred within 30 days of completion of plasma exchange treatment. In addition all deaths that occurred during follow-up are reported. These definitions were established without knowledge of ADAMTS13 activity and prior to data analysis. Statistical analysis
When patient groups were compared, all data are only from the patients' initial episodes. To determine statistical differences across ADAMTS13 activity levels (Tables 3 and 4), we used the Kruskal-Wallis test for continuous data, such as age, and the chi-square test or Fisher exact test for categorical data, such as sex. For the ordinal categories of ADAMTS13 activity, location shifts or trends were examined for continuous data using a Jonckheere-Terpstra test and for categorical data using a Cochran-Mantel-Haenszel mean score statistic. All tests were 2-sided using an
Relation of ADAMTS13 activity to clinical categories of patients with clinically diagnosed TTP-HUS Severe ADAMTS13 deficiency was present in 18 (13%) of 142 patients (95% confidence interval [CI], 7.7%-19.3%), 16 patients with idiopathic TTP-HUS, and 2 women in whom TTP-HUS occurred after giving birth (Table 1). None of the patients with TTP-HUS following hematopoietic stem cell transplantation, with a drug association, following a prodrome of bloody diarrhea, or who had an additional or alternative diagnosis had severe ADAMTS13 deficiency. There were 7 patients with ADAMTS13 activity between 5% and 9%: 4 patients with idiopathic TTP-HUS, 1 patient following a prodrome of bloody diarrhea, and 2 patients who had an alternative diagnosis. The patient who presented with bloody diarrhea had ADAMTS13 activity of 5%; her stool cultures for Escherichia coli 0157:H7 were negative; she had multiple exacerbations requiring 54 plasma exchange treatments over 68 days to achieve remission. The 2 patients in the clinical category of additional/alternative disorders had ADAMTS13 activities of 8% and 9%. They had unexpected diagnoses of systemic infection following initiation of plasma exchange treatment that explained their presenting features, and plasma exchange treatment was stopped. Only 4 of the 142 patients had received prior plasma treatment. Three patients had been previously treated with infusion of 2 units plasma before the ADAMTS13 sample was drawn: 1 idiopathic patient had less than 5% activity, 2 patients with drug association had more than 25% activity. One patient who presented with bloody diarrhea had received 4 plasma exchanges and multiple plasma infusions in another state; her ADAMTS13 activity was more than 25%. Presenting features and clinical outcomes of patients with severe ADAMTS13 deficiency Individual data for the 18 patients with clinically diagnosed TTP-HUS who had severe ADAMTS13 deficiency are presented in Table 2. The median age was 36 years, 14 (78%) of the patients were women, 9 (50%) were African American, and 9 (50%) were obese (body mass index [BMI] ≥ 30 kg/m2).28 Among the 9 obese patients, 8 were women and 5 were African American. No patients with severe ADAMTS13 deficiency were identified during the years 1997 and 1998. During these 2 years serum samples from 34 patients, including 8 patients with idiopathic TTP-HUS, were collected. In the other 4 complete years of this study serum samples from 104 patients, including 38 patients with idiopathic TTP-HUS, were collected. Presenting symptoms were variable and nonspecific. One patient presented only with profound weakness; all others had multiple presenting symptoms. The most common presenting symptoms, occurring in 10 patients, were gastrointestinal, including abdominal pain, nausea, vomiting, and diarrhea. Five patients had severe neurologic abnormalities at the time the clinical diagnosis of TTP-HUS was made. Four additional patients had disorientation, confusion, and/or ataxia at the time the clinical diagnosis of TTP-HUS was made. Two of the 4 patients with less severe neurologic abnormalities (patients 6 and 7) and 1 patient who presented without neurologic abnormalities (patient 11) developed severe neurologic abnormalities 1 to 2 days after beginning plasma exchange treatment. The remaining 8 patients had no neurologic signs or symptoms at presentation or throughout their course. Three patients presented with chest pain as their major symptom, and 2 patients presented with gross hematuria. The time between the onset of the presenting symptoms and the clinical diagnosis of TTP-HUS ranged from several hours up to 3 weeks; the median duration was 5 days; in 4 patients symptoms had been present for 2 to 3 weeks. The mildness of the symptoms resulted in 7 patients being seen and released by their primary care physicians prior to hospitalization. Platelet counts, hematocrits, and LDH values at presentation were all abnormal but of variable severity. Eight patients had elevated serum creatinine values (≥ 132.6 µmol/L [≥ 1.5 mg/dL]); one of these patients (patient 11) had acute renal failure but did not require dialysis. All patients were in the idiopathic clinical category except patients 8 and 17, who were diagnosed in the immediate postpartum period (postpartum days 1 and 7) of their first pregnancies. Patient 8 had 4 relapses during 9 months following her initial presentation but then had an uncomplicated, successful second pregnancy in 2001. Patient 17 had an uncomplicated pregnancy with delivery of a healthy infant in January 2003, 14 months after her presentation with TTP-HUS. Two patients (patients 7 and 13) died during their initial hospitalization; patient 3 died of an acute myocardial infarction during a plasma exchange treatment 78 days into her second episode. The 16 patients who survived their initial episode required 5 to 74 plasma exchange treatments over 5 to 137 days to achieve a remission. Nine of these 16 patients had one or more exacerbations of TTP-HUS activity when daily plasma exchange treatments were stopped, requiring resumption of daily plasma exchange. Among the 8 surviving patients who had any neurologic abnormalities, only patient 11 has persistent symptoms (mild aphasia). Seven of the 8 surviving patients who were diagnosed prior to 2000 have relapsed, 6 within a year of achieving remission. None of the 8 surviving patients diagnosed in 2000-2001 have relapsed, and all have been followed for at least 1 year after remission was achieved. Four of 8 patients diagnosed prior to 2000 received glucocorticoids; 3 of 8 patients diagnosed in 2000-2001 received glucocorticoids. ADAMTS13 inhibitor activity was demonstrable in the sera of 17 (94%) of the 18 patients with severe ADAMTS13 deficiency, suggesting an acquired disorder; 8 patients had strong inhibitor activity. There were no apparent patterns relating the strength of the ADAMTS13 inhibitor activity to the presenting features and clinical outcomes. The patient with no demonstrable inhibitor activity (patient 5) required the longest time to achieve remission (71 plasma exchanges over 136 days), suggesting that he also had acquired rather than congenital TTP-HUS. He had received 2 units of plasma before ADAMTS 13 activity was measured, which may have neutralized an inhibitor. Relation of presenting features and clinical outcomes of 142 patients with clinically diagnosed TTP-HUS to categories of ADAMTS13 activity Ninety-eight (69%) of the 142 patients were women. The racial distribution (white, 71%; African American, 19%; Native American, 6%; other, 4%) of all 142 patients was different from the Oklahoma population (white, 76%; African American, 8%; Native American, 10%; other, 6%)29 (P < .01). The relative frequency of obesity (BMI ≥ 30 kg/m2)28 of all 142 patients was 31%, greater than the relative frequency in the Oklahoma population (22.1%; 95% CI, 20.1%-23.7%, in 2001, the year with the highest reported rate of obesity in our time period30,31) (P = .01). Patients in the category of severe ADAMTS13 deficiency were younger, and the relative frequencies of African Americans and obesity were greater than in patients in the other categories of ADAMTS13 activities (Table 3). There was no difference in the relative frequency of patients with severe neurologic abnormalities across different categories of ADAMTS13 activity. There was a significant linear trend for a greater relative frequency of acute renal failure in patients with higher levels of ADAMTS13 activity. Although platelet counts were significantly different among patients with different levels of ADAMTS13 activity, 41 (44%) of 94 patients with ADAMTS13 activity more than 25% had platelet counts less than 20 x 109/L (< 20 x 103/µL). There were no differences in hematocrit and LDH values across different categories of ADAMTS13 activity. There were significant linear trends for more frequent responses, more frequent exacerbations, more plasma exchange treatments required to achieve a remission, and more frequent relapses in patients with more severe ADAMTS13 deficiency. However, exacerbations requiring prolonged plasma exchange treatment as well as subsequent relapse occurred in all categories of ADAMTS13 activity. Relation of presenting features and clinical outcomes of 142 patients with clinically diagnosed TTP-HUS to clinical categories Patients in the idiopathic clinical category were compared with the combined groups of patients in all other categories (Table 4). There was no difference in the relative frequency of severe neurologic abnormalities. The relative frequency of acute renal failure was less among the patients with idiopathic TTP-HUS. Patients with idiopathic TTP-HUS had a higher response rate, a lower death rate, and a higher relative frequency of relapse than patients in all other categories. Relation of presenting features and clinical outcomes of 48 patients with clinically diagnosed idiopathic TTP-HUS to the presence or absence of severe ADAMTS13 deficiency The presenting features of patients with severe ADAMTS13 deficiency (Table 5) were not distinct from the presenting features of patients without severe deficiency, about half of whom had ADAMTS13 activities more than 25% (Table 1). The ranges for all continuous variables overlapped. Among presenting features, there were statistically significant differences only for younger age and less frequent acute renal failure in the patients with severe ADAMTS13 deficiency. Among clinical outcomes, there was a statistically significant difference only for the greater relative frequency of relapses in the patients with severe ADAMTS13 deficiency. The severity of thrombocytopenia and microangiopathic hemolysis (the essential diagnostic criteria for TTP-HUS), the relative frequency of severe neurologic abnormalities, the responses to plasma exchange, and the number of plasma exchange treatments required to achieve a remission were not different. In both groups, some patients required few plasma exchanges to achieve remission, whereas others had multiple exacerbations requiring prolonged courses of treatment.
We describe the relation of ADAMTS13 activity to presenting features and clinical outcomes in an inception cohort of 142 patients with clinically diagnosed TTP-HUS. The 142 patients in whomADAMTS13 activity was measured represent 88% of all 161 consecutive patients within a defined geographic region and time period in whom a clinical diagnosis of TTP-HUS was established and a decision for plasma exchange treatment was made. Identification of all patients with clinically diagnosed TTP-HUS was ensured because our region has only one provider of plasma exchange services, and because the practice in our region is to treat all adult patients who have clinically diagnosed TTP or HUS with plasma exchange. The validity and consistency of the clinical diagnosis was provided by one hematologist who saw and actively participated in the diagnosis and initial treatment decisions for 96% of these patients. This patient cohort is different from previously reported case series describing ADAMTS13 activity in patients with TTP and/or HUS.19-25 In the previous case series19-25 patients may have been selected by referral to specialized centers and may have been selected for ADAMTS13 measurements after their clinical outcomes were known. Differences in patient selection between our cohort of consecutive patients and previous case series may be responsible for the lower relative frequency of severe ADAMTS13 deficiency reported here. Our cohort represents the complete spectrum of patients diagnosed with TTPHUS. Other selection criteria, including clinical outcomes, that are not available at the time of initial diagnosis may lead to a higher relative frequency of severe ADAMTS13 deficiency. Although our data are not comparable to previous case series, our data should be generalizable to initial management decisions in community practice32 because (1) our patients represent all consecutive patients within a defined region and time period who had a clinical diagnosis of TTP-HUS, and (2) they were identified at the time the clinical diagnosis of TTP-HUS was first established and the decision for plasma exchange treatment was made. Therefore, our patient cohort reflects routine clinical practice. In the patients who are described as having additional disorders, the additional disorders were not thought to exclude the diagnosis of TTP-HUS. However, the alternative disorders were only appreciated after plasma exchange treatment had begun, by continuing diagnostic studies, and then plasma exchange treatment was stopped. In 3 patients, the alternative diagnosis was only apparent at autopsy. In all of our patients, the clinical diagnosis of TTP-HUS was sufficiently firm to justify intervention with plasma exchange, a procedure recognized to have substantial risk of major complications.11,12 The recent descriptions of ADAMTS13 have increased our understanding of the pathogenesis of the TTP-HUS syndromes.14,33 Severe ADAMTS13 deficiency (< 5% activity) is now recognized as an abnormality specific for patients described as TTP.13-15 Therefore, it is often assumed that most patients with TTP have severe ADAMTS13 deficiency. However, only 18 (13%) of 142 patients (95% CI, 7.7%-19.3%) in our cohort were severely deficient. Sixteen (89%) of the 18 severely deficient patients were in the idiopathic clinical category, consistent with previous reports that patients with TTP-HUS following hematopoietic stem cell transplantation34 or, with rare exceptions, a prodrome of bloody diarrhea35 do not have severe ADAMTS13 deficiency. Also none of our patients with drug-associated TTP-HUS had severe ADAMTS13 deficiency. Severe ADAMTS13 deficiency has been described in patients whose TTP-HUS was reported to be associated with ticlopidine36 or clopidogrel37; in our cohort, only 1 patient had ticlopidine-associated TTP-HUS and none had clopidogrel-associated TTPHUS. In 12 of our patients with drug-associated TTP-HUS the implicated drug was quinine. ADAMTS13 activity measurements have not been previously reported in patients with quinine-associated TTP-HUS. Quinine may cause TTP-HUS by the formation of quinine-dependent antibodies to endothelial cells, platelets, and other cells.38 Although none of the patients who were discovered to have alternative diagnoses after plasma exchange had begun had severe ADAMTS13 deficiency, 2 who were diagnosed with systemic infections had borderline severe deficiency (ADAMTS13 activities of 8% and 9%). Even within the category of idiopathic TTP-HUS, only 33% of patients had severe ADAMTS13 deficiency. The presenting features of patients with severe ADAMTS13 deficiency were heterogeneous. Furthermore, most presenting features of patients with idiopathic TTP-HUS who had severe ADAMTS13 deficiency were not different from the presenting features of patients with idiopathic TTP-HUS who did not have severe deficiency. The severity of thrombocytopenia and microangiopathic hemolysis and the relative frequency of severe neurologic abnormalities were not different between these 2 groups. The observation that 8 (44%) of 18 patients with severe ADAMTS13 deficiency had no neurologic abnormalities supports the current practice requiring only thrombocytopenia and microangiopathic hemolytic anemia without another apparent cause to make a clinical diagnosis of TTP-HUS.2,3,5,8,9 These observations suggest that severe ADAMTS13 deficiency does not detect all patients who may be appropriately diagnosed as TTP-HUS. Plasma exchange may be effective treatment for TTP because autoantibodies that inhibit ADAMTS13 activity are removed and ADAMTS13 protease is replaced.14 Therefore, it has been postulated that plasma exchange may not be effective in patients without ADAMTS13 deficiency and that measurements of ADAMTS13 activity may be able to guide treatment decisions.14,16-19 However, in our cohort, patients at all levels of ADAMTS13 activity apparently responded to plasma exchange treatment. In all ADAMTS13 activity categories, some patients responded with few plasma exchange treatments, whereas others required prolonged treatment to achieve a remission. Although relapses occurred in patients in all ADAMTS13 activity categories, the relative frequency of relapses was much greater in patients with severe ADAMTS13 deficiency. Among patients with idiopathic TTPHUS, the responses to plasma exchange treatment and durations of treatment required to achieve remission were not different between patients with or without severe ADAMTS13 deficiency. Among patients with severe ADAMTS13 deficiency, clinical outcomes did not appear to be related to the strength of the ADAMTS13 inhibitor activity. These observations suggest that severe ADAMTS13 deficiency does not detect all patients who may be effectively treated with plasma exchange. Therefore, the absence of severe ADAMTS13 deficiency should not be used as a basis for withholding plasma exchange treatment. When patients with severe ADAMTS13 deficiency were compared with the other patients with a clinical diagnosis of TTP-HUS, they were younger, were more likely to be African American, had a greater relative frequency of obesity, and only one (6%) of 18 patients had acute renal failure. This low relative frequency of acute renal failure is consistent with previous reports that severe ADAMTS13 deficiency may distinguish TTP from HUS.20,23 However, our patient and one other report that described acute renal failure in patients with severe ADAMTS13 deficiency39 demonstrate that patients who may be described as HUS can have severe ADAMTS13 deficiency. Patients with severe ADAMTS13 deficiency were also more likely to respond to plasma exchange treatment, similar to a previous report,19 although they required more treatments to achieve a remission and then had a greater relative frequency of relapses after achieving a remission. However, these differences in presenting features and clinical outcomes were characterized by linear trends across different levels of ADAMTS13 activity, not by distinct differences between patients with or without severe deficiency. This demonstration of clinical differences with increasing severity of ADAMTS13 deficiency supports the role of ADAMTS13 deficiency as an important pathogenetic factor in the development of TTP-HUS.14 However, other pathogenetic factors may also contribute to the etiology of TTP-HUS, such as drug-dependent antibodies to endothelial cells, platelets, and other cells38; factor V Leiden24; pregnancy40; and perhaps also female sex, African American race, and obesity. Fourteen of 18 patients (78%) with severe ADAMTS13 deficiency were women, and the relative frequency of women was not different across different categories of ADAMTS13 activity. Two patients with severe ADAMTS13 deficiency were diagnosed in the immediate postpartum period of their first pregnancy. Previous case series have reported that 66% to 71% of patients with TTP-HUS were women and that 12% to 25% of women presented during pregnancy or after giving birth.4-6,8,40 ADAMTS13 deficiency has been previously reported in patients with TTP-HUS occurring during pregnancy,25,33 and pregnancy has also been noted to provoke acute episodes of TTP-HUS in women with familial ADAMTS13 deficiency.33,40 These observations suggest that pregnancy and the postpartum period are risk factors for TTP-HUS, perhaps because of a physiologic decrease of ADAMTS13 activity during the second and third trimesters of normal pregnancies.41 The explanation for the predominance of women among patients with clinically diagnosed TTP-HUS is unknown. Half of patients with severe ADAMTS13 deficiency were African American, and the relative frequency of African Americans was increased among all patients with clinically diagnosed TTPHUS. Only 2 previous reports have addressed racial differences among patients with TTP-HUS, and both described an increased relative frequency of African Americans.7,42 Similar to our observations, an additional study reported that 12 (29%) of 42 patients with TTP-HUS were African American; 11 (37%) of 30 patients who had ADAMTS13 deficiency were African Americans, and only 1 (8%) of 12 patients who did not have ADAMTS13 deficiency was African American.24 The explanation for the increased relative frequency of African Americans among patients with clinically diagnosed TTP-HUS is unknown. Half of patients with severe ADAMTS13 deficiency were obese, and the relative frequency of obesity among all patients with clinically diagnosed TTP-HUS was increased. An association of obesity with TTP-HUS has not been previously reported. Obesity is a well-recognized risk factor for cardiovascular disease43 and may be a risk factor for thrombosis by causing vascular endothelial dysfunction,44 enhancing tissue factor expression,45 impairing fibrinolysis,46 causing platelet activation,47 or causing low-grade systemic inflammation.48 Therefore, obesity may be an additional contributing risk factor for the development of TTP-HUS. The clinical diagnosis of TTP-HUS is often uncertain. The initial symptoms are variable, not specific, and may not be severe. The diagnostic laboratory criteria of thrombocytopenia and microangiopathic hemolytic anemia are also not specific. Because of the urgency to initiate plasma exchange treatment and because plasma exchange has substantial risk of major complications,11,12 efforts to establish a more certain diagnosis and a more clear indication for plasma exchange treatment are important. However, data from our cohort of consecutive patients suggest that severe ADAMTS13 deficiency does not detect all patients who may be appropriately diagnosed as TTP-HUS. Because plasma exchange treatment has documented efficacy in patients with clinically diagnosed TTPHUS, using current diagnostic criteria,2,3 and because some patients in our cohort at all levels of ADAMTS13 activity appeared to benefit from plasma exchange, plasma exchange treatment remains indicated for all patients with clinically diagnosed TTP-HUS.
We thank Ronald O. Gilcher, MD, James W. Smith, MD, PhD, their staff at the Oklahoma Blood Institute, and the physicians of Oklahoma for their continuing cooperation and support of these studies.
Submitted January 22, 2003; accepted February 22, 2003.
Prepublished online as Blood First Edition Paper, March 13, 2003; DOI 10.1182/blood-2003-01-0193.
Supported by Swiss National Science Foundation (grant 32-066756.01).
The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked "advertisement" in accordance with 18 U.S.C. section 1734.
Reprints: James N. George, The University of Oklahoma Health Sciences Center, Hematology-Oncology Section, PO Box 26901, Oklahoma City, OK 73190; e-mail: jim-george{at}ouhsc.edu.
  CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati What's this?
Related Letter in Blood Online:
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
 |
 |
|||||||
 |
 |
 |
 |
 |
 |
 |
 |
||
| Copyright © 2003 by American Society of Hematology Online ISSN: 1528-0020 | |||||||||
Top
Abstract
Introduction
of 0.05. Confidence intervals around a single proportion were calculated using exact binomial formulas. SAS software, release 8.01 (SAS Institute, Cary, NC), and SPSS for Windows, release 10.0, were used. 
