Abstract
Controversies exist how to predict medium term mortality (Mt) in diabetics (DM) with 3 vessel (3VD) and/or left main (LM) disease undergoing myocardial revascularization ranging from Syntax Score II (SSII) where DM was not predicative variable up to FREEDOM formula which was derived, just from population with DM (without LM), having DM patients (Pts) requirement of insulin as one of predicative variable.
To compare predicative power of SSII, FREEDOM and formula developed in our institution in Pts post first isolated CABG with 3VD and/or LM with DM.
From our prospective data base of 2455 consecutive pts who had the first isolated CABG in the period 01/2012–12/2014 with complex Ischemic Heart Disease with 100% follow up of 4 years all-cause Mt we created by random sampling Training (1321; Mt:10.4%; DM 511; Mt:13.3%) and Validation (1134; Mt:10.0%; DM 414; Mt: 11.8%) sets. After deriving predicative formula (Cox regression) from training population we validated FREEDOM, SSII and Our Formula in 414 pts with DM from the Validation set.
Characteristics of pts, our formula, predicating power by C Statistics, Calibration plots and Brier scores are presented in Picture 1.
FREEDOM formula designed just for DM pts with complex Ischemic Heart Disease without LM had the smallest standard error in the estimate, but moderate C statistics as Syntax Score II and our formula which may be used for pts with and without DM and 3VD and/or LM.
Picture 1
Type of funding source: Public grant(s) – National budget only. Main funding source(s): Ministry of education, science and technology development, Republic of Serbia
