A seasonal association of incident cases of thrombotic thrombocytopenic purpura was not observed in the Oklahoma TTP‐HUS Registry

In this issue of TRANSFUSION, Park and colleagues describe the seasonal association for all cases of suspected thrombotic thrombocytopenic purpurahemolytic uremic syndrome (TTP-HUS) treated with plasma exchange (PEX) at the University of North Carolina Healthcare, 1999 to 2008. They report that the occurrence of cases was significantly greater in summer months (June-August) than in fall (SeptemberNovember) and winter (December-February) for all 97 cases (including relapses) defined as idiopathic TTP and for 54 individual patients with an initial presentation of idiopathic TTP. They also state that more cases occurred in the summer months for all 139 suspected TTP-HUS cases and 31 cases with ADAMTS13 activity of less than 10%. Therefore, we analyzed the seasonal association of patients enrolled in the Oklahoma TTP-HUS Registry. The registry is a population-based inception cohort of all patients for whom PEX treatment is requested for a clinical diagnosis of TTP or HUS. All patients within a defined geographic area, 58 of Oklahoma’s 77 counties, are identified because the Oklahoma Blood Institute is the sole provider of PEX for this region. All 415 consecutive individual patients with an initial episode of clinically diagnosed TTP or HUS identified, 1989 to 2010, were enrolled in the registry. Patients are assigned in a hierarchal order to one of six categories based on the clinical features of their initial episode. A total of 162 patients were described as idiopathic as they did not qualify for one of the five previous categories: stem cell transplantation, pregnancy or postpartum, drug-associated, bloody diarrhea prodrome, or presence of an additional or alternative diagnosis. ADAMTS13 activity has been measured in the Central Hematology Laboratory of the Inselspital, Berne, Switzerland, by Drs Johanna Kremer Hovinga and Bernhard Lämmle on serum samples collected on 281 (93%) of 301 patients enrolled since November 13, 1995, by two methods: quantitative immunoblotting of plasma-derived von Willebrand factor substrate and a fluorogenic assay using FRETS-VWF73 substrate. Severe ADAMTS13 deficiency was defined as less than 10% activity documented by either method. To analyze patients identified during complete calendar years, the 67 patients with ADAMTS13 activity of less than 10% from 1996 through 2010 were identified; six patients in whom an alternative diagnosis (systemic infection or malignancy) was discovered to be responsible for their presenting clinical features after PEX had been started were excluded from this analysis. All 61 patients were determined to have acquired severe ADAMTS13 deficiency: three patients were postpartum, two presented with bloody diarrhea, three had an additional autoimmune disorder, and 53 were categorized as idiopathic. To determine if the number of incident cases was different across months or seasons, a goodness-of-fit chi-square was calculated; the null hypothesis was that the number of cases was equally distributed across all months or seasons. An alpha of 0.05 was used. Table 1 demonstrates that the initial presentations of patients in all three analyzed groups were not different across the four seasons. In addition, the initial presentations were also not different across all 12 months: all patients (p = 0.62), idiopathic patients (p = 0.79), and ADAMTS13-deficient patients (p = 0.36). The data for the 61 patients with ADAMTS13 activity of less than 10% are presented in Fig. 1. There was no increased frequency during summer months. The reason for the difference between the Oklahoma and North Carolina data is not clear. Both states have similar climates. Oklahoma data are based on the initial episodes of all consecutive individual patients from a defined geographic region across 22 years. North Carolina data represent both initial and relapsed episodes as well as multiple episodes for some individual patients. There also may be referral bias of the North Carolina patients. In conclusion, we did not observe a seasonal association of TTP.