Time in Range: A Fourth Domain in Glycemic Control or a Glucose Variability Alternative?

The outcome of glucose control in critically ill patients has received significant attention. Krinsley has stated that hyperglycemia, hypoglycemia, and high glucose variability (GV) are the 3 domains that are independently associated with high mortality in critically ill patients. Moving away from tight to safe glucose control provides an optimum goal. The interest in measuring GV has been raised after multiple studies demonstrated its association with mortality in intensive care units (ICUs).^,  Multiple approaches had to be used to measure GV, including average daily risk range; however, these measures seem to be too sophisticated to be implemented in each ICU. Time in range (TIR), defined as a measure of time where the blood glucose remains within the proposed target range, has been investigated by Omar et al. Our group found it to be a simple parameter that could be measured without specific requirements; therefore, it could be simply applied in any ICU. We calculated the whole time of insulin infusion (A) and the time being within the proposed target range (B) during insulin infusion, and expressed TIR as TIR = (B/A) × 100. We found that patients with more than 80% TIR, whether or not they had diabetes, had better outcomes than those with less than 80% TIR, as determined by wound infection, lengths of ventilation, and ICU stay. In addition, they were not subject to frequent hypoglycemic events.

Interestingly, Krinsley and Preiser followed the same technique in stratifying mortality in critically ill patients without diabetes in a retrospective descriptive study. The authors concluded that survival in critically ill patients without diabetes is strongly associated with a TIR 70 to 140 mg/dL value of more than 80%, independent of the ICU length of stay and severity of the individual's illness. Individualized algorithms for patients with and without diabetes, as mentioned by Tafelski et al, could replace published working guidelines. Therefore, management of blood glucose in the ICU by a single target looks unnecessarily restrictive. The clinical settings could mandate a target to fit, as in cardiac surgery where 6.0 to 8.1 mmol/L seems to be an acceptable goal.

In the view of the reports relating the TIR value to mortality and morbidity in those with and without diabetes, TIR value emerges. In addition to its simplicity, TIR could provide a possible alternative to GV measurements. Even if TIR and GV are mathematically and conceptually linked, they are not interchangeable. Research in ICU glucose control could move a step forward considering the proper intervention in intensive insulin therapy.