Validation of the PLASMIC score, a clinical prediction tool for thrombotic thrombocytopenic purpura diagnosis, in Chinese patients.

Background: The recently published PLASMIC score was shown to be an excellent diagnostic model to identify patients with severe ADAMTS-13 deficiency. However, it is unclear if this score is suitable for Chinese patients with suspected thrombotic thrombocytopenic purpura (TTP).
Methods: We retrospectively studied a Chinese cohort of 134 consecutively hospitalized patients with thrombotic microangiopathy, ADAMTS-13 test results, and clinical information between January 2015 and June 2018, the PLASMIC score's distinguishing ability in this cohort was evaluated. We also combined the lactate dehydrogenase (LDH)-to-aspartate aminotransferase (AST) ratio with the PLASMIC score to determine its predictive value for TTP in pregnancy.
Results: In this Chinese cohort, 39 patients were diagnosed with TTP using ADAMTS13 testing. When stratified at high (score, 6-7) vs. low-intermediate (score, 0-5) risk, the PLASMIC test predicted TTP with a positive predictive value of 51%, negative predictive value of 92%, sensitivity of 85%, and specificity of 58%. In pregnant patients, a combination of the PLASMIC score of 6-7 and LDH-to-AST ratio of ≥5.5 showed a higher positive predictive value for TTP compared with the PLASMIC score alone (71% vs. 35%, P < 0.001). Removing the component MCV < 90 fL from the PLASMIC score did not reduce its distinguishing power, as measured using the C statistic (0.837, 95%CI 0.767-0.907 vs. 0.797, 95%CI 0.718-0.877; P = 0.771).
Conclusions: Combination with a higher LDH-to-AST ratio improves the positive predictive value of the PLASMIC score in pregnant patients. A simpler score without MCV < 90 fL can be used to identify TTP in Chinese patients.
(Copyright © 2018 Elsevier Ltd. All rights reserved.)