Incidence and Prediction of Unstable Blood Glucose Level among Critically Ill Patients: A Cohort Study.

Abstract

To evaluate the incidence and the prediction of unstable blood glucose level among critically ill patients hospitalized in an intensive care unit. A cohort study was conducted with 62 adult patients hospitalized at an intensive care unit of a hospital located in Minas Gerais, Brazil, between March and July of 2017. Patient's demographic information, along with scores for Simplified Acute Physiology Score III, primary medical diagnosis, discharge status, diagnosis of diabetes and/or sepsis, length of stay, glycemic variability, type of nutrition, types of medications and treatments, and oxygen therapy were collected daily. A daily venous blood sample was collected to measure blood glucose levels during the patient's hospitalization period. Bivariate analysis was used to explore the association among the potential diagnostic indicators and the outcome of unstable blood glucose levels. Multivariate Cox regression was used to identify the potential predictors for the outcome. Of the total of 62 participants, 45.1% (n=28) had unstable blood glucose level. Among the 28 patients with unstable blood glucose levels, half of them (n=14, 50%) had hypoglycemia and the other half had hyperglycemia (n=14, 50%). Decreased number of days hospitalized and the use of intensive glucose control with regular insulin were associated with decreased odds of developing hyperglycemia. The presence of mechanical ventilation was associated with a higher risk for the development of hypoglycemia. This study provides knowledge and evidence of diagnostic indicators for unstable blood glucose levels that are not currently included in the NANDA-International terminology for the nursing diagnosis Risk for unstable blood glucose level (00179). This study identified important diagnostic indicators that nurses can observe during the assessment to identify patients that are at risk for developing unstable blood glucose level and provide the appropriate care.