Association between stress hyperglycemia ratio and short-term all-cause mortality in non-diabetic critically ill patients: a multi-database cohort study

Stress hyperglycemia is common in non-diabetic critically ill patients. However, the association between the stress hyperglycemia ratio (SHR) and short-term mortality in this population remains incompletely characterized in large cohorts.

This retrospective cohort study used data from MIMIC-IV and MIMIC-III. Non-diabetic adults with available admission glucose and HbA1c measurements were included. The primary outcomes were 28-day and 90-day all-cause mortality. Patients were grouped by SHR quartiles. Cox proportional hazards models with progressive adjustment were used to assess the associations between SHR and 28-day and 90-day all-cause mortality. Restricted cubic spline (RCS), subgroup, and sensitivity analyses were performed. A total of 6994 non-diabetic patients were included. In the MIMIC-IV cohort, 28-day and 90-day mortality rates were 9.30% and 13.22%, respectively; corresponding rates in MIMIC-III were 4.75% and 5.47%. In the primary adjusted models, the highest SHR quartile (Q4) was associated with higher 28-day and 90-day all-cause mortality compared with Q1 in both cohorts (MIMIC-IV: 28-day HR 3.07, 95% CI 2.56-3.68; 90-day HR 3.34, 95% CI 2.84-3.94; MIMIC-III: 28-day HR 4.26, 95% CI 2.95-6.14; 90-day HR 4.39, 95% CI 3.06-6.30). RCS analyses demonstrated significant nonlinear associations between SHR and both 28-day and 90-day all-cause mortality in both cohorts, with the increase in mortality hazard becoming more pronounced at higher SHR levels.

Higher SHR was associated with increased short-term all-cause mortality in non-diabetic critically ill patients, with significant nonlinear associations observed in both cohorts.