A second look at the mortality in thrombotic thrombocytopenic purpura score: An external validation study.

BACKGROUND: Despite high effectiveness of therapeutic plasma exchange as the first-line therapy, thrombotic thrombocytopenic purpura (TTP) remains a life-threatening condition and may require utilization of adjunct modalities in certain patients. Mortality In TTP Score (MITS) is a prognostic risk stratified scoring tool designed to predict mortality in hospitalized patients with TTP. There has not been an external validation of MITS to date. STUDY DESIGN AND METHODS: We performed an external validation of MITS in patients hospitalized with TTP using the National Inpatient Sample database from 2016 to 2019. We identified 4589 patients who met the selection criteria. Univariate and multivariable logistic regression models were run based on the MITS parameters of arterial thrombosis, intracranial haemorrhage, age, renal failure, ischemic stroke, platelet transfusion, and myocardial infarction to evaluate prognostic performance and discriminatory power of this tool. RESULTS: All-cause mortality was reported more frequently in female subjects (57.8%), Caucasian race (58.9%), and ages 60 years, and above (52.3%). In a multivariable analysis, the variables included in the MITS criteria remained significant predictors of mortality. Moreover, MITS correlated with mortality risk. Our model's area under the receiving operator character curve was 71%, compared to 78.6% in the derivation cohort study. DISCUSSION: In this external validation cohort, performance of MITS was similar to the derivation cohort, validating it as a valuable clinical tool that may guide management of TTP patients.