Coronary artery bypass surgery grafting (CABG) is a commonly efficient treatment for coronary artery disease patients. Even if we know the underlying disease, and advancing age is related to survival, there is no research using the one year before surgery and operation-associated factors as predicting elements. This research used different machine-learning methods to select the features and predict older adults’ survival (more than 65 years old). This nationwide population-based cohort study used the National Health Insurance Research Database (NHIRD), the largest and most complete dataset in Taiwan. We extracted the data of older patients who had received their first CABG surgery criteria between January 2008 and December 2009 (n = 3728), and we used five different machine-learning methods to select the features and predict survival rates. The results show that, without variable selection, XGBoost had the best predictive ability. Upon selecting XGBoost and adding the CHA2DS score, acute pancreatitis, and acute kidney failure for further predictive analysis, MARS had the best prediction performance, and it only needed 10 variables. This study’s advantages are that it is innovative and useful for clinical decision making, and machine learning could achieve better prediction with fewer variables. If we could predict patients’ survival risk before a CABG operation, early prevention and disease management would be possible.

No abstract

Objective: The optimal selection of prosthetic heart valve for dialysis-dependent patients remains controversial. We investigated the comparative effectiveness and safety of mechanical prosthesis (MP) and bioprosthesis (BP) for these patients.Methods: After the systematic review, we included studies that involved patients on dialysis undergoing aortic valve replacement or mitral valve replacement (MVR) and reported comparative outcomes of MP and BP. Meta-analysis was performed using random-effects model. We conducted a subgroup analysis based on the valve position and postoperative international normalized ratio (INR), which was extracted from either tables or methods of each study. A meta-regression was used to examine the effects of study-level covariates.Results: We included 24 retrospective studies without randomized-controlled trials, involving 10,164 participants (MP ¼ 6934, BP ¼ 3230). Patients undergoing aortic valve replacement with MP exhibited a better long-term survival effectiveness (hazard ratio, 0.64; 95% confidence interval [CI], 0.47-0.86). Conversely, studies including MVR demonstrated little difference in survival (hazard ratio, 0.90; 95% CI, 0.73-1.12). A meta-regression revealed that age had little effect on long-term survival difference between MP and BP (b ¼ -0.0135, P ¼ .433). MP had a significantly greater bleeding risk than did BP when INR was above 2.5 (incidence rate ratio, 10.58; 95% CI, 2.02-55.41). However, when INR was below 2.5, bleeding events were comparable (incidence rate ratio, 1.73; 95% CI, 0.78-3.82). The structural valve deterioration rate was significantly lower in MP (risk ratio, 0.24; 95% CI, 0.14-0.44).Conclusions: MP is a reasonable choice for dialysis-dependent patients without additional thromboembolic risk requiring aortic valve replacement, for its better long-term survival, durability, and noninferior bleeding risk compared with BP. Conversely, BP might be an appropriate selection for patients with MVR, given its similar survival rate and lower bleeding risk. Although our meta-regression demonstrates little influence of age on long-term survival difference between MP and BP, further studies stratifying patients based on age cut-off are mandatory.