≥ Intermediate risk — designed for sensitivity
In prospective validation, the ≥intermediate-risk threshold achieved 95.9% sensitivity (95% CI, 86.3–98.9) and 58.8% precision (95% CI, 47.8–68.9).
Clinical predictive model
Rapid TTP risk stratification
This score is designed exclusively for confirmed TMA cases in patients aged 16 years or older.
ADAMTS13 testing remains essential and should not be omitted on the basis of the PURPURA Score. An unexpected result should prompt clinical reassessment rather than automatically determining whether treatment is started or withheld.
Certain conditions such as DIC, malignant hypertension, evidence of disseminated carcinomatosis, or positive Shiga toxin may be associated with a lower likelihood of TTP, although they were not directly evaluated during score development.
Classification
Score 0.000
The PURPURA Score should be used only when there is sufficient evidence of an active TMA. In the development study, TMA eligibility required ≥1% schistocytes, LDH ≥250 U/L, and anemia and/or thrombocytopenia. It is not intended for presentations in which TMA has not been adequately established or for a known TTP relapse.
Use clinical and laboratory values obtained before therapeutic plasma exchange, red-cell or platelet transfusion, or other treatments likely to substantially alter the recorded variables.
After selecting Calculate risk, the model generates a score and an initial low-, intermediate-, or high-risk classification. Intermediate-risk cases are then assessed using predefined clinical rules designed to reduce false-positive classifications. These rules can reclassify a case from intermediate to low risk. The category displayed in the result is the final classification.
Medication exposure before TMA onset is not included in the score. Drug history should be reviewed independently, as some treatments can cause non-TTP TMA, whereas severe ADAMTS13-deficient TTP has also been reported in association with specific agents, including immune checkpoint inhibitors and dasatinib.
The PURPURA Score was developed using 1,328 historical index TMA presentations and evaluated in an internal test cohort of 332 presentations. It was subsequently evaluated in a prospective temporal cohort of 244 consecutive presentations. Cases were referred through a Spanish central laboratory network involving 82 centers.
In prospective validation, the ≥intermediate-risk threshold achieved 95.9% sensitivity (95% CI, 86.3–98.9) and 58.8% precision (95% CI, 47.8–68.9).
High-risk precision was 95.7% (95% CI, 79.0–99.2) in the internal test cohort and 92.3% (95% CI, 75.9–97.9) in prospective validation. Prospective sensitivity was 49.0% (95% CI, 35.6–62.5).
Validation has so far been performed within the same Spanish central laboratory network. Independent external validation is required to assess generalizability to other settings.