| Summary: | Objective The aim in this study was to stratify maximum blood glucose levels to identify the the best cut-off value of glucose levels to predict mortality in acute coronary syndrome (ACS) patients, regardless of whether they had diabetes.Design A retrospective cohort study.Setting All clinical data were obtained from the ‘Medical Information Mart for Intensive Care III’ database.Participants A total of 3078 patients with ACS were included in the study. We divided the patients into four levels based on their maximum blood glucose levels (glucosemax), then analysed the relationship between each group with mortality.Results Among enrolled patients, 2780 and 298 were survivors and non-survivors, respectively. Blood glucose levels and mortality showed a ‘tick’ type relationship, with levels 3 and 4 found to be closely associated with increased hospital mortality (p<0.05), relative to level 1 (<6.1 mmol/L), used as the reference group. No significant association was observed in mortality between level 2 and level 1 (p=0.095). In addition, we found a gradual increase in OR for level 2 (OR: 2.42, 95% CI 0.86 to 6.80, p=0.095), level 3 (OR: 4.33, 95% CI 1.55 to 12.13, p=0.005) and level 4 (OR: 7.27, 95% CI 2.56 to 20.62, p<0.001), relative to level 1. Based on receiver operating characteristic curves, the optimal cut-off value for predicting mortality were 11.5 (area under curve (AUC)=0.724), 11.2 (AUC=0.729), 13.4 (AUC=0.638), 15.8 (AUC=0.717) and 11.3 mmol/L (AUC=0.764) in all ACS, acute myocardial infarction, unstable angina, diabetes and non-diabetes patients, respectively. The results of subgroup analysis suggested that in patients with significantly elevated blood glucose, the mortality of non-diabetes was higher than patients with diabetes (OR: 0.42, 95% CI 0.31 to 0.57, p<0.001).Conclusion Overall, glucosemax ≥11.5 mmol/L had a significant association with increased mortality in patients with ACS. Non-diabetes ACS patients need a more robust blood glucose management strategy compared with diabetes counterparts.
|
|---|
