Development and Validation of a Predictive Model for Death in Acquired Severe ADAMTS13 Deficiency-Associated Idiopathic Thrombotic Thrombocytopenic Purpura: The French TMA Reference Center Experience

Acquired thrombotic thrombocytopenic purpura (TTP) is still associated with a 10–20% death rate, which did not significantly improve for more than 20 years despite a better awareness about this diagnosis. Usually, death occurs within the first days of management and so far, early prognostic factors of death could not be clearly identified. In this context, the accurate understanding of factors associated with a fatal outcome at the acute phase of the disease would help to better tailor initial treatment and further improve these results.

To identify prognostic factors associated with 1-month death in TTP patients with acquired severe (< 10% of normal activity) ADAMTS13 deficiency.

Prospective national cohort of adult (≥ 18 year-old) patients included between October, 2000, and December, 2008. A validation cohort of patients was set up from January, 2009 to March, 2010.

248 (analysis cohort) and 39 (validation cohort) consecutive adult TTP patients with acquired severe ADAMTS13 deficiency from 39 French centers.

30-day mortality after treatment initiation according to characteristics at inclusion.

When compared to survivors, non-survivors (11%) were older (54.0 ± 19.4 versus 39.0 ± 15.5 year-old, respectively, P <.001) and had more frequently a past history of arterial hypertension (37% versus 10%, respectively, P <.001) and ischemic heart disease (19% versus 4%, respectively, P =.002). Prognosis was increasingly poor with age (p <.004), particularly in patients ≥ 60 year-old. On presentation, cerebral manifestations were more frequent in non-survivors (81% versus 55%, respectively, P =.009) and serum creatinine level was higher (172 ± 123 versus 117 ± 91 μmol/L, respectively, P =.037). Death occurred after a few days (mean 7 days, interquartile range = [3, 12]), in a context of one or multiple organ failure in relation with an uncontrolled TTP. Platelet count between diagnosis and death did not significantly increase (20 ± 25×10⁹/L versus 28 ± 46×10⁹/L, respectively, P =.44). The most significant independent variables for determining death were age (P <.001), cerebral involvement (stupor and seizure) and LDH level > 10 times normal value (P <.04 for both). After computing the risk scores using these 3 variables (Table), patients were stratified into 3 distinctive risk groups regarding death: low, 0 and 1; intermediate, 2; and high, 3 and 4. The proportion of positive predictive value for 30-day death was 11–13% in the low risk group, 20% in the intermediate group, and 39–50% in the high risk group. This score was confirmed in the validation cohort using these variables, with higher values corresponding to increased risk of early death (P <.01).

A risk score for early death was defined in patients with TTP and validated on an independent cohort. This score should help to stratify early treatment and intensify patients with a worse prognosis. Importantly, we provide clear evidence that age is an important prognostic factor of TTP. Consequently, old patients with a diagnosis of TTP should benefit from more intensive supportive care and should be monitored more closely in intensive care units with more aggressive attention to cardiac and renal function.

Rottensteiner:Baxter Innovations GmbH: Employment.