ABSTRACT
Background: Coronary re-interventions after CABG are generally preferably percutaneous, and may be related to progression of atherosclerosis due to pre-existing risk factors and may be influenced by operator experience, type of surgery-(off pump or on-pump), and conduits used. We analysed the demographics, patient and operative characteristics, clinical features of patients undergoing early re-interventions - arbitrarily defined as within 2 years of index "primary" CABG for this study (i.e. no prior percutaneous coronary interventions) to determine predictors of the same. Method(s): We collected data on 1367 patients who underwent primary CABG over a decade from Jan 1,2010 to Jan 1 2020( pre-COVID).Demographic and clinical risk factors for CAD, angiographic characteristics, pattern of CAD, electrocardiographic (ECG) changes, and prevalent LV function were evaluated at baseline, immediate post-op and on follow ups till the need of the next intervention. Patients who underwent re-intervention in the said period were compared with an age- and gender- matched population who did not undergo re interventions to determine predictors for re-intervention by both Logistic regression and Machine learning analysis using SVM, KNN and Naive Bayes Classifier Results: 160 patients ( 11.7%) patients underwent re-intervention within 2 years of the primary CABG. Multivariate backward logistic regression analysis and Machine learning analysis with three models revealed that re-intervention was performed significantly more frequent in patients with : Index admission with uncontrolled diabetes mellitus (strongest risk factor), patient age < 50 years, emergency CABG -Both as primary CABG or bail-out for PCI complications, Use of saphenous vein graft v/s total arterial revascularisation ( except RIMA usage), those undergoing off- pump CABG, failure of DAPT to continue beyond three months, CABG following recent acute coronary syndrome, CABG in Multivessel disease with Syntax score> 27 need of mechanical support following CABG ( IABP), higher ventilation requirement with delayed sternal course, patients with post-op renal insufficiency with /without undergoing assisted renal replacement, statin intolerance/ lower doses, and CABG with more than 4 grafts(all p < 0.001). Patients with carotid disease and peripheral disease were also higher in number in the re-intervened arm. Re-interventions were more common in patients with LVEF > 50% at time of need of re-intervention, but this may be attributed to selection bias due to preference for conservative management if the LVEF was too low on follow up. Conclusion(s): Predictors for early re-intervention after CABG are a pointer towards more aggressive "malignant" form of atherosclerosis. There is actually higher graft loss in younger age especially in emergency CABG, patients with uncontrolled DM, renal insufficiency and high SS, and inadequate statin usage- all of which may contribute to( or indicate towards) an inflammatory process of atherosclerosis. Knowledge of these risk factors may guide the surgeon in counselling the patient for possible graft loss as well as suitably plan the surgical course in reducing re-intervention. Copyright © 2022