Admission glucose level and stress hyperglycemia ratio (SHR) are both common indicator of stress hyperglycemia. However, it is controversial whether SHR has better prognostic value for long-term mortality than admission glucose in patients with acute myocardial infarction (AMI).

A total of 5,832 AMI patients from a large nationwide prospective cohort enrolled between January 2015 and July 2019 were included [median follow-up period: 2.1 years (interquartile range: 2.0-2.1 years)]. SHR was calculated by the following formula: SHR=admission glucose (mmol/L)/[1.59×HbA1c (%)-2.59]. The primary endpoint was all-cause mortality during follow-up.

After multivariable adjustment, the highest tertiles of SHR and admission glucose were both significantly associated with an increased long-term mortality risk irrespective of diabetic status in patients with AMI. According to restricted cubic spline curves, positive linear associations between admission glucose and risk of long-term mortality were observed in subjects with diabetes and without diabetes (Figure 1). In contrast, the association between SHR and long-term mortality appeared to be J-shaped in patients with diabetes (Figure 1). The area under curves (AUC) for admission glucose were higher than AUC for SHR in both diabetic (0.621 vs. 0.599) and non-diabetic patients (0.597 vs. 0.566) (Figure 2).

Elevated admission glucose and stress hyperglycemia ratio were both significantly associated with the long-term mortality risk. Compared with admission glucose, stress hyperglycemia ratio has no superiority in long-term mortality prediction of AMI patients.