Leann M Lesperance1, Anna Spektor, Kenneth J McLeod. 1. Clinical Science and Engineering Research Center, Binghamton University, Binghamton, New York 13902-6000, USA. llespera@binghamton.edu
Abstract
BACKGROUND: Oral glucose tolerance testing (OGTT) remains the gold standard for diagnosis of diabetes and is used commonly in the research laboratory. The Medtronic MiniMed (North-ridge, CA) Continuous Glucose Monitoring System (CGMS Gold), developed for long-term monitoring of glycemic levels, could provide a convenient means for tracking OGTT results during research protocols; however, the sensor demonstrates glucose and time dependencies that preclude direct application of the company-provided conversion algorithm in the first 12-24 h after sensor insertion. Here, we propose an alternative conversion algorithm that permits utilization of the CGMS monitor for glucose monitoring during this time. METHODS: Healthy female participants underwent CGMS monitoring during OGTT or fasting sessions in combination with finger-stick blood glucose measurements. Logarithmic transformations and multiple regression analysis were used to quantify the time and glucose dependence of the sensors. RESULTS: Sensor performance, as characterized by sensitivity (ratio of sensor current to capillary blood glucose levels), was shown to vary logarithmically with glucose levels as well as time after sensor insertion. A conversion algorithm developed on the basis of these observations was tested on 17 subjects during OGTT. A mean absolute relative difference between capillary blood glucose and CGMS values of 11.6 +/- 6.5% for the new algorithm was seen, compared to 20.6 +/- 5.9% with the Medtronic Solutions version 3.0c algorithm. CONCLUSIONS: Incorporation of the glucose and time dependence in CGMS sensor data yields an improved mean absolute difference between actual and estimated blood glucose values compared to the Medtronic-supplied algorithm in the immediate time period following sensor insertion.
BACKGROUND: Oral glucose tolerance testing (OGTT) remains the gold standard for diagnosis of diabetes and is used commonly in the research laboratory. The Medtronic MiniMed (North-ridge, CA) Continuous Glucose Monitoring System (CGMS Gold), developed for long-term monitoring of glycemic levels, could provide a convenient means for tracking OGTT results during research protocols; however, the sensor demonstrates glucose and time dependencies that preclude direct application of the company-provided conversion algorithm in the first 12-24 h after sensor insertion. Here, we propose an alternative conversion algorithm that permits utilization of the CGMS monitor for glucose monitoring during this time. METHODS: Healthy female participants underwent CGMS monitoring during OGTT or fasting sessions in combination with finger-stick blood glucose measurements. Logarithmic transformations and multiple regression analysis were used to quantify the time and glucose dependence of the sensors. RESULTS: Sensor performance, as characterized by sensitivity (ratio of sensor current to capillary blood glucose levels), was shown to vary logarithmically with glucose levels as well as time after sensor insertion. A conversion algorithm developed on the basis of these observations was tested on 17 subjects during OGTT. A mean absolute relative difference between capillary blood glucose and CGMS values of 11.6 +/- 6.5% for the new algorithm was seen, compared to 20.6 +/- 5.9% with the Medtronic Solutions version 3.0c algorithm. CONCLUSIONS: Incorporation of the glucose and time dependence in CGMS sensor data yields an improved mean absolute difference between actual and estimated blood glucose values compared to the Medtronic-supplied algorithm in the immediate time period following sensor insertion.
Authors: Claudio Cobelli; Chiara Dalla Man; Giovanni Sparacino; Lalo Magni; Giuseppe De Nicolao; Boris P Kovatchev Journal: IEEE Rev Biomed Eng Date: 2009-01-01