We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Glycaemic Variability as a Predictor of Mortality in Intensive Care Patients: A Retrospective Cohort Study.
- Authors
Hancı, Pervin; Cengizhan, Serdar; Yıldız, Çağla; İnal, Volkan
- Abstract
Introduction: Glycaemic variability (GV) refers to the fluctuations in blood glucose levels over time. In intensive care patients, glycaemic variability can have significant clinical implications, including increased mortality and morbidity 1. This study was conducted to contribute to accumulating data on the subject and to reveal the situation in our intensive care patient population. Materials and Methods: Patients admitted to the Internal Medicine Intensive Care Unit between January 2015 and August 2020 were retrospectively screened through the hospital database. Demographic characteristics of patients, comorbidities, APACHE II scores, and estimated mortality rates according to this score; the mean amplitude of glycaemic excursions (MAGE) 2 and SOFA scores from the daily blood glucose measurement values from the first day of ICU admissions to discharge or 28th day of the intensive care unit stay; usage of insulin, corticosteroid, vasopressor, and beta-blocker therapy, length of stay and 28-day mortality status has been recorded. Results: One hundred thirty-six patients were enrolled in the study and grouped into high GV (n=70) and low GV (n=66) using mean MAGE 65 mg/dL as the cut-off point. There wasn’t any difference between groups in age, gender, comorbidity, APACHE II, mean SOFA scores, applied corticosteroid, vasopressor, and beta-blocker treatments, intensive care unit length of stay, and mortality (Table 1). Insulin therapy was used more frequently in patients with high GV (74.3%) than in patients with low GV (36.4) (p<0.001). When the mean MAGE was compared according to 28-day mortality, MAGE was found to be higher in nonsurvivors (78.8±32.2) than in survivors (65.4±22.5) (t=-2.78, p=0.005). In the ROC analysis performed for mortality estimation of GV, AUC was 0.611 (p=0.02) (MAGE >61 mg/dL, sensitivity 68.5%, specificity 50%). Conclusion: In our study, glycemic variability predicted mortality moderately. Tight glycemic control is an important goal in critical care management that improves outcomes and reduces complications.
- Subjects
INTENSIVE care patients; COHORT analysis; INTENSIVE care units; AGE groups; BLOOD sugar measurement
- Publication
Turkish Journal of Intensive Care, 2023, Vol 21, p7
- ISSN
2146-6416
- Publication type
Abstract