Dataset Information

Predictability of Mortality in Patients With Myocardial Injury After Noncardiac Surgery Based on Perioperative Factors via Machine Learning: Retrospective Study.

ABSTRACT:

Background

Myocardial injury after noncardiac surgery (MINS) is associated with increased postoperative mortality, but the relevant perioperative factors that contribute to the mortality of patients with MINS have not been fully evaluated.

Objective

To establish a comprehensive body of knowledge relating to patients with MINS, we researched the best performing predictive model based on machine learning algorithms.

Methods

Using clinical data from 7629 patients with MINS from the clinical data warehouse, we evaluated 8 machine learning algorithms for accuracy, precision, recall, F1 score, area under the receiver operating characteristic (AUROC) curve, and area under the precision-recall curve to investigate the best model for predicting mortality. Feature importance and Shapley Additive Explanations values were analyzed to explain the role of each clinical factor in patients with MINS.

Results

Extreme gradient boosting outperformed the other models. The model showed an AUROC of 0.923 (95% CI 0.916-0.930). The AUROC of the model did not decrease in the test data set (0.894, 95% CI 0.86-0.922; P=.06). Antiplatelet drugs prescription, elevated C-reactive protein level, and beta blocker prescription were associated with reduced 30-day mortality.

Conclusions

Predicting the mortality of patients with MINS was shown to be feasible using machine learning. By analyzing the impact of predictors, markers that should be cautiously monitored by clinicians may be identified.

SUBMITTER: Shin SJ

PROVIDER: S-EPMC8554678 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundElevated plasma or serum troponin, indicating perioperative myocardial injury (PMI), is common after noncardiac surgery. However, underlying mechanisms remain unclear. Acute coronary syndrome (ACS) is associated with the early appearance of circulating microRNAs, which regulate post-translational gene expression. We hypothesised that if PMI and ACS share pathophysiological mechanisms, common microRNA signatures should be evident.MethodsWe performed a nested case control study of samples obtained before and after noncardiac surgery from patients enrolled in two prospective observational studies of PMI (postoperative troponin I/T>99th centile). In cohort one, serum microRNAs were compared between patients with or without PMI, matched for age, gender, and comorbidity. Real-time polymerase chain reaction quantified (qRT-PCR) relative microRNA expression (cycle quantification [Cq] threshold <37) before and after surgery for microRNA signatures associated with ACS, blinded to PMI. In cohort two, we analysed (EdgeR) microRNA from plasma extracellular vesicles using next-generation sequencing (Illumina HiSeq 500). microRNA-messenger RNA-function pathway analysis was performed (DIANA miRPath v3.0/TopGO).ResultsMicroRNAs were detectable in all 59 patients (median age 67 yr [61-75]; 42% male), who had similar clinical characteristics independent of developing PMI. In cohort one, serum microRNA expression increased after surgery (mean fold-change) hsa-miR-1-3p: 3.99 (95% confidence interval [CI: 1.95-8.19]; hsa-miR-133-3p: 5.67 [95% CI: 2.94-10.91]; P<0.001). These changes were not associated with PMI. Bioinformatic analysis of differentially expressed microRNAs from cohorts one (n=48) and two (n=11) identified pathways associated with adrenergic stress and calcium dysregulation, rather than ischaemia.ConclusionsCirculating microRNAs associated with cardiac ischaemia were universally elevated in patients after surgery, independent of development of myocardial injury.