SP243
PROGNOSTIC SIGNIFICANCE OF LOW LEVEL OF BLOOD GLUCOSE IN SEVERE ACUTE KIDNEY INJURY: A SECONDARY ANALYSIS FROM THE RENAL STUDY

INTRODUCTION AND AIMS: We aimed to determine associations between baseline blood glucose levels (BGL) and clinical outcomes in patients with severe acute kidney injury (AKI) receiving continuous renal replacement therapy (CRRT).

METHODS: A secondary analysis from the Randomized Evaluation of Normal versus Augmented Level of RRT (RENAL) study was performed. The primary endpoint was all-cause mortality at 90 days after randomization. The secondary outcomes included duration of hospital and ICU stay. The multivariate Cox regression model adjusted for baseline variables was used to assess the association of baseline BGL (measured prior to patient randomization) and mortality.

RESULTS: Baseline BGL data was available in 1404 out of 1508 patients from the RENAL study. The patients were divided into 4 groups using quartiles of baseline BGL (group 1, BGL <5.8mmol/L, group 2, BGL 5.8-7.2mmol/L, group 3, BGL 7.3-9.1mmol/L, and group 4, BGL>9.2mmol/L). A total of 627 patients died in 4 groups (51.7% in group 1, 38.7% in group 2, 44.5% in group 3, and 44.0% in group 4). The Cox regression model showed that the baseline BGL in patients in the group 1 (<5.8 mmol/L) was associated with an increased risk of death at 90 days (HR 1.47, 95% CI 1.13-1.90, p = 0.0036), compared with the group 2 (BGL 5.8-7.2mmol/L), while BGL in the group 3 and 4 did not show a significant impact on 90 days mortality. Furthermore, there were no significant differences in the duration of hospital stay and ICU stay among these 4 groups (p=0.55 and p=0.30, respectively).

CONCLUSIONS: Blood glucose levels within the normal physiological range at baseline appear to be associated with higher mortality. This effect may be due to other unmeasured comorbidities but may warrant further study in the setting of severe AKI.