Summary Objective A number of short-term studies using the continuous glucose monitoring system (CGMS) indicate that improved metabolic control can be observed in patients with type 1 diabetes when CGMS is applied in clinical practice. Data have also accumulated to suggest that spot measurements of glucose performed four times a day would not detect as much as 70% of all hypoglycaemic episodes registered by CGMS. When more frequent reference values were obtained however it was inferred that nighttime hypoglycaemia reported by CGMS may be spurious. As most assessments with CGMS have been utilizing abdominal subcutaneous tissue, we were interested to evaluate whether differences between blood glucose and sensor readings obtained from different sites exist. Research design and methods Two viscometric affinity glucose sensors, applicable to subcutaneous tissue of both forearm and abdomen, were inserted subcutaneously in 12 non-diabetic subjects. Sensors generated glucose data at 3 min intervals and venous blood glucose was determined in duplicates by HemoCue at 15-90 min intervals for 24 hours. Each subject consumed three carbohydrate-rich meals, performed an exercise test, and was observed during nocturnal bed-rest at the research center. Results The initial decrease of blood glucose during exercise was not fully detected by the sensors. Otherwise, no significant differences between sensor values and blood glucose were observed during day-time. During nocturnal bed-rest abdominal sensor values came approximately 20% lower than blood glucose ( P Conclusion It is concluded that a difference between glucose values obtained from abdominal and forearm subcutaneous fat can be observed during nocturnal bed-rest in non-diabetics.
