Author + information
- Leon Stephen Edwards,
- Isuru Ranasinghe,
- Fiona Turnbull,
- Mark Woodward,
- Derek Chew,
- Anushka Patel,
- David Brieger,
- Paul Antonis and
- Bernadette Costa
Current therapies used to treat Acute Coronary Syndrome are associated with increased bleeding. Bleeding is linked with greater cost, length of stay, morbidity and mortality. Several scores exist which predict a patients risk of bleeding after an Acute Coronary Syndrome. Our aim was to validate the performance of two internationally derived risk scores using Australia or New Zealand (ANZ) data, and compare this against a score developed specifically for this population
Data were analysed from 5,418 ANZ patients enrolled in the Global Registry of Acute Coronary Events. Using admission characteristics, stepwise logistic regression was used to construct a multivariable model and score to predict in-hospital major bleeding. This score was validated using 1,834 patients enrolled in Cooperative National Registry of Acute Coronary Care, Guideline Adherence and Clinical Event.
Were compared to the Can Rapid risk stratification of Unstable angina patients Suppress ADverse outcomes with Early implementation of the ACC/AHA guidelines (CRUSADE) and Reduction of Atherothrombosis for Continued Health (REACH) scores. Model performance was evaluated using the c-statistic and Hosmer-Lemeshow goodness-of-fit test.
Four independent baseline predictors of bleeding were identified and included in the score: admission diagnosis of myocardial infarction, history of a major bleed, diastolic blood pressure and age. Discrimination of the ANZ score in the development (0.67) and validation (0.65) cohort was acceptable and superior to its comparators (0.63 and 0.60). The ANZ score demonstrated good calibration (Hosmer: Lemeshow goodness of fit test; p > 0.9772).
This simple 4-variable risk is superior to existing scores at predicting a patient's baseline risk of bleeding at admission to hospital with Acute Coronary Syndrome in ANZ. This tool may aid clinicians in early stratification of bleeding risk when deciding upon appropriate therapy. Additionally, these results suggest that a model tailored to a specific population may more accurately predict risk.
Moderated Poster Contributions
Poster Sessions, Expo North
Sunday, March 10, 2013, 9:45 a.m.-10:30 a.m.
Session Title: Bleeding and ACS: Predicting Risk and Measuring Impact
Abstract Category: 1. Acute Coronary Syndromes: Clinical
Presentation Number: 1215M-196
- 2013 American College of Cardiology Foundation