TIMI Risk Score accurately predicts risk of death in 30-day and one-year follow-up in STEMI patients treated with primary percutaneous coronary interventions.Kardiol Pol. 2007 Jul; 65(7):788-95; discussion 796-7.KP
TIMI Risk Score for ST-elevation myocardial infarction (STEMI) was developed in a cohort of patients treated with fibrinolysis. It was though to predict in-hospital and short-term prognosis. Later studies validated this approach in large cohorts of patients, regardless of the applied treatment and presented its good power to predict 30-day mortality.
We applied the TIMI Risk Score to our registry of STEMI patients treated with primary percutaneous intervention (pPCI) to validate the possibility to predict one-year survival.
Our registry comprised 494 consecutive patients (mean age 58.5+/-11.3 years) with STEMI treated with pPCI who were followed for approximately one year. STEMI was diagnosed based on typical criteria: chest pain, ECG changes and rise in myocardial necrosis markers. In all patients TIMI Risk Score for STEMI was calculated and they were divided into three groups: low risk (0-5 points), medium risk (6-7) and high risk (>7 points). Multivariate logistic regression analysis, Kaplan-Meier survival analysis with Cox and log-rank tests as well as c statistics from receiver-operator curves (ROC) were used for statistical analysis.
TIMI 3 flow was obtained in 95.5% of patients. Median TIMI risk score was 4 (ranging from 0 to 10). During follow-up there were 47 deaths (9.5%). There was a statistically significant difference in survival between all risk groups both in 30-day and one-year follow-up (p <0.001 log-rank test). TIMI Risk Score had good power to predict 30-day (c statistic 0.834, 95% CI 0.757-0.91, p <0.0001) as well as one-year mortality (c statistic 0.809, 95% CI 0.739-0.878, p <0.0001). Interestingly, when we excluded from the analysis all patients who died during the first 30 days, TIMI Risk score maintained its very good prognostic value. All analysed risk groups significantly differed between each other with respect to mortality (p <0.05, log-rank test) and the c statistic was 0.745 (95% CI 0.612-0.879, p <0.0002). In multivariate logistic regression analysis TIMI Risk Score was one of the independent risk factors of death during one-year follow-up (OR 1.59, p <0.001).
TIMI Risk Score accurately defines the population of STEMI patients who are at high risk of death not only during the first 30 days, but also during a long-term follow-up. This simple score should be included in the discharge letters because it contains very useful information for further care.