Scoring system to guide decision making for the use of bilateral internal mammary arteries: The BIMA score

Abstract

ObjectiveCurrently, surgeons are unable to estimate whether the survival benefit from bilateral internal mammary artery (BIMA) grafting outweighs the increased deep sternal infection (DSWI) risk in the individual cases. As a consequence BIMA grafting is still largely underutilized. We aimed to develop and validate a scoring system to predict the individual risk for DSWI and 10 year survival probability to guide surgeons in decision making process for the use of BIMA over single internal mammary artery (SIMA). MethodsFast backward elimination on predictors was performed using Lawless and Singhal method. BIMA score was reported as a partial nomogram that can be used to manually obtain predicted individual risk of DSWI and 10 year survival probability from the regression models. Bootstrapping validation of the regression models was performed. ResultsThe study population consisted of 5234 patients of which a total of 960 (18%) received BIMA and the remaining 4274 (82%) underwent conventional strategy using SIMA. A total of 119 patients experienced DSWI (2.3%). A total of 489 deaths were recorded after a median follow-up time of 5.68 years (interquartile range 2.7–8.8, range 0–12.1). BIMA grafting was found to have effect on both risk of DSWI and 10 year survival probability and was included into partial nomograms. Bootstrapping validation confirmed a good discriminative power of the models. ConclusionThe BIMA score provides an impartial assessment of the decision making process for clinicians to establish the optimum revascularization strategy for individual patients.