PURPOSE: To compare the outcome of critically ill patients in relation to:admission blood glucose (A-BG) levels, mean blood glucose levels during the first 24 hours (M-BG-24), mean blood glucose levels during ICU stay (M-BG-ICU).
METHODS: Retrospective analysis of 150 Med/Surg ICU patients. Patients’ characteristics, admission diagnoses, comorbidities, length of stay (LOS), number of ventilator days, APACHE score, ICU complications and final outcome were recorded. BG levels were tabulated as A-BG, M-BG-24 and M-BG-ICU. BG of 150 was used as a cutoff to compare the ICU outcome of the three measurements using chi-square test, independent T-test and ANOVA test.
RESULTS: 150 patients, mean age 64, mean APACHE 22.3 and overall mortality 14%. Mortality rate of patients with A-BG of > 150 was 12% compared to 9% for < 150 (P=0.1). Mean ICU LOS and mean ventilator days were 9.15 and 8.91 for A-BG > 150 compared to 7.25 and 5.81 for A-BG < 150 (T = 1.5, not significant). A-BG > 150 was associated with increased number of complications 2.96 versus 2.34 (T = 2.63, significant). Using M-BG-24, no statistically significant difference in mortality rate, LOS, mean ventilator days and number of complications between > 150 and < 150 groups (11%, 8.95, 7.96 & 2.71 versus 10%, 7.31, 6.35 & 2.49). Similarly no significant difference observed for the four variables using M-BG-ICU of > 150 versus < 150 (10%, 9.31, 8.93 & 2.72 versus 11%, 7.38, 6.13 & 2.51). Using ANOVA test, mortality rates of BG > 150 for A-BG, M-BG-24 and M-BG-ICU were 12%, 11% & 10% (P=1.0). Also ICU-LOS, mean ventilator days and number of complications were not significantly different between these three groups.
CONCLUSION: There was no statistically significant correlation between elevated A-BG, M-BG-24, M-BG-ICU and ICU morbidity & mortality. Only elevated A-BG was associated with increased risk of complications during ICU stay.
CLINICAL IMPLICATIONS: Increased A-BG levels could be used as a predictor of increased risk of ICU complications. No correlation between BG levels and ICU mortality.
DISCLOSURE: Shadi Soufi, None.