RESUMO
AIM: To study variations in glucose levels over 48 hours in critically ill patients by capillary blood glucose done on glucometer and compare the same in different categories of patients based on various diseases, as well as their correlation with sepsis and diabetes mellitus. To compare the same results in a subset of patients with the readings of continuous glucose monitoring. MATERIAL AND METHODS: We studied 50 critically-ill patients (Age≥18 years), admitted in medical ICU (on mechanical ventilation/ionotropic supports/in sepsis) in a teaching hospital in semi-urban Maharashtra. Critical illness was defined as any physiological instability leading to disability or death within minutes or hours, based on neurological assessment, respiratory system involvement and cardiovascular involvement. Capillary blood sugar levels were done 4 hourly using 'NIPRO' glucometer. Site was rotated. 5 patients had simultaneous continuous glucose monitoring, using I-Pro bio-sensor. RESULTS: Total 50 patients were included in the study. The data was collected and tabulated. Analysis showed that all critically ill patients showed some higher than normal recordings of blood sugar, which till now has been attributed to 'stress-hyperglycaemia'. This may be absent or blunted in sepsis. In the criticallyill patients with primary involvement of gastrointestinal tract, meal-unrelated fluctuations were seen. In critically-ill patients with CNS and CVS involvement, lowest BSL recordings were seen (meal unrelated) at 2 am. CONCLUSIONS: We concluded that that patients who develop hypoglycaemia may have an equally bad prognosis or even worse than those who develop hyperglycaemia during the period of critical illness. CGM devices record tissue glucose levels continuously, and may be useful as a 'tissue hypoglycaemia' alert.
