Leveraging machine learning to enhance postoperative risk assessment in coronary artery bypass grafting patients with unprotected left main disease - A retrospective cohort study.

BACKGROUND: Risk stratification for patients undergoing coronary artery bypass surgery (CABG) for left main coronary artery (LMCA) disease is essential for informed decision-making. This study explored the potential of machine learning (ML) methods to identify key risk factors associated with mortality in this patient group. METHODS: This retrospective cohort study was conducted on 866 patients from the Gulf Left Main Registry who presented between 2015 and 2019. The study outcome was hospital all-cause mortality. Various machine learning models [logistic regression, random forest (RF), k-nearest neighbor, support vector machine, naïve Bayes, multilayer perception, boosting] were used to predict mortality, and their performance was measured using accuracy, precision, recall, F1 score, and area under the receiver operator characteristic curve (AUC). RESULTS: Nonsurvivors had significantly greater EuroSCORE II values (1.84 (10.08-3.67) vs. 4.75 (2.54-9.53) %, P<0.001 for survivors and nonsurvivors, respectively). The EuroSCORE II score significantly predicted hospital mortality (OR: 1.13 (95% confidence interval: 1.09-1.18), P<0.001), with an AUC of 0.736. RF achieved the best ML performance (accuracy=98, precision=100, recall=97 and F1 score=98). Explainable artificial intelligence using SHAP demonstrated the most important features as follows: preoperative lactate level, emergency surgery, chronic kidney disease (CKD), NSTEMI, nonsmoking status, and sex. QLattice identified lactate and CKD as the most important factors for predicting hospital mortality this patient group. CONCLUSION: This study demonstrates the potential of ML, particularly the Random Forest, to accurately predict hospital mortality in patients undergoing CABG for LMCA disease and its superiority over traditional methods. The key risk factors identified, including preoperative lactate levels, emergency surgery, chronic kidney disease, NSTEMI, nonsmoking status, and sex, provide valuable insights for risk stratification and informed decision-making in this high-risk patient population. Additionally, incorporating newly identified risk factors into future risk scoring systems can further improve mortality prediction accuracy.

Impact of preoperative intra-aortic balloon pump on outcomes in coronary artery bypass grafting for unprotected left-main coronary artery disease.

Background: Preoperative intra-aortic balloon pump (IABP) before coronary artery bypass grafting (CABG) could improve operative outcomes by augmenting the diastolic coronary blood flow. Data on preoperative IABP use in patients with left-main coronary artery (LMCA) disease are limited. This study aimed to characterize patients who received preoperative IABP before CABG for LMCA and evaluate its effect on postoperative outcomes. Methods: This multicenter retrospective cohort study that included consecutive 914 patients who underwent CABG for unprotected LMCA disease from January 2015 to December 2019 in 14 tertiary referral centers. Patients were grouped according to the preoperative IABP insertion into patients with IABP (n=101) and without IABP (n=813). Propensity score matching adjusting for preoperative variables, with 1:1 match and a caliber of 0.03 identified 80 matched pairs. The primary outcomes used in propensity score matching were cardiac mortality and major adverse cardiac and cerebrovascular events (MACCE). Results: IABP was commonly inserted in patients with previous myocardial infarction (MI), chronic kidney disease, peripheral arterial disease, and congestive heart failure. IABP patients had higher EuroSCORE [ES >8%: 95 (11.86%) vs. 40 (39.60%), P<0.001] and SYNTAX {29 [interquartile range (IQR) 25-35] vs. 33 (IQR 26-36); P=0.02} scores. Preoperative cardiogenic shock and arrhythmia were more prevalent in patients with IABP, while acute coronary syndrome was more prevalent in patients without IABP. After matching, there was no difference in vasoactive inotropic score between groups [3.5 (IQR 1-7.5) vs. 6 (IQR 1-13.5), P=0.06], and lactate levels were nonsignificantly higher in patients with IABP [2.4 (IQR 1.4-4.5) vs. 3.1 (IQR 1.05-7.75), P=0.05]. There were no differences between groups in acute kidney injury [20 (25%) vs. 26 (32.5%), P=0.34], cerebrovascular accidents [3 (3.75%) vs. 4 (5%), P>0.99], heart failure [5 (6.25%) vs. 7 (8.75%), P=0.75], MI [7 (8.75%) vs. 8 (10%), P>0.99], major adverse cardiac and cerebrovascular events [10 (12.5%) vs. 17 (21.25%), P=0.21], and cardiac mortality [6 (7.50%) vs. 14 (17.50%), P=0.09]. Patients who received IABP had longer ventilation times [8.5 (IQR 6-23) vs. 15.5 (IQR 5-50.5) h, P=0.03] and intensive care unit (ICU) stays [3 (IQR 2-5) vs. 4 (IQR 2-7.5) days, P=0.01]. Conclusions: Preoperative IABP in patients with LMCA might not be associated with reduced cardiac mortality or hospital complications. IABP could increase the duration of mechanical ventilation and ICU stay, and its use should be individualized for each patient.

Techniques to Treat Slow-Flow/No-Reflow During Primary Percutaneous Coronary Intervention.

OBJECTIVE: The aim of this study was to compare TIMI flow after administering intracoronary (IC) medications through various routes for the treatment of slow flow/no-reflow during primary PCI. METHODS: Two independent parallel cohorts of the patients who underwent primary PCI for STEMI and developed slow/no-reflow were recruited. Selection of cohort was based on the route of administration of IC medications as proximal or distal. Post administration TIMI follow was compared between the two cohorts. RESULTS: A total of 100 patients were included in both, proximal and distal, cohort. Distribution of angiographic, clinical and demographic characteristics was not significant between the two cohorts except prevalence of hypertension, and diabetes mellitus. Frequency of hypertension, and diabetes mellitus were 45 % vs.70 %; p < 0.001 and 28 % vs. 44 %; p = 0.018 among patients in distal and proximal cohort respectively. Final TIMI III flow was achieved in significantly higher number of patients in distal cohort with the frequency of 88 % vs. 76 %; p = 0.027 as compared to proximal cohort. CONCLUSION: Administration of IC medication via distal route is observed to be more effective for the treatment of slow flow/no-reflow during primary PCI. Distal route via export catheter or perforated balloon technique should be preferred wherever feasible.

Real-World Experience with a 60-mm-Long Stent in the Setting of Primary Percutaneous Coronary Intervention.

Introduction: Primary percutaneous coronary intervention (PPCI) represents a timely procedure that requires speedy revascularization. Moreover, PPCI in diffuse coronary lesions remains to be challenging even in the hands of experienced operators as the use of a long stent may increase the difficulty of the procedure in terms of stent delivery, deployment, and optimization. However, the practicability and clinical outcomes of deployment of a 60-mm-long stent in the setting of PPCI remain to be determined. Methods: The study is a retrospective observational analysis in a prospective cohort. The prospectively gathered data of consecutive patients from June 2016 to December 2019, who underwent PPCI with BioMime sirolimus-eluting stents 2.5-3.0/60 mm or 3.0-3.5/60 mm were analyzed at 1 year regarding the primary outcome of major adverse cardiovascular and cerebrovascular events (MACCE) and target lesion revascularization (TLR). Results: A total of 88 cases were included in the study; 23 cases underwent PPCI, whereas 65 underwent nonPPCI. The PPCI group had a mean age of 65.7 ± 10.9 years compared with 63.3 ± 9.6 years (P = 0.34) in the nonPPCI group. Eighty-three percentage of PPCI were males compared with 94% of their nonPPCI counterparts (P = 0.20). In addition, the prevalence of hypertension was more common in the PPCI group (87% vs. 63%, P = 0.03). There was no statistically significant difference between the two groups regarding other comorbidities. The most common culprit vessel was the left anterior descending artery (57%) in the PPCI group and the right coronary artery (58%) in the nonPPCI. The use of a stent with a diameter of 2.5-3.0 mm was more common in both groups (61% in PPCI vs. 66% in nonPPCI, P = 0.8). MACCE occurred in four patients during a year of follow-up. One occurred in the PPCI group (4%) compared with three in the nonPPCI group (5%) (P = 1.00). TLR was required in two cases, one in each group (4% vs. 2%, P = 0.46). Conclusion: The use of a 60-mm-long stent in the setting of PPCI has an excellent 12-month outcome in procedural success, MACCE, and TLR. Large randomized studies are required to confirm these results.