Anzeige
Erschienen in:
11.11.2015 | Original Research
Continuous glucose monitoring system in the operating room and intensive care unit: any difference according to measurement sites?
Erschienen in: Journal of Clinical Monitoring and Computing | Ausgabe 1/2017
Einloggen, um Zugang zu erhaltenAbstract
Given the benefit of glucose control in the perioperative period, we evaluated the accuracy and performance of the continuous glucose monitoring system (CGMS) depending on different measurement sites in the operating room (OR) and in the intensive care unit (ICU). Patients over 18 years of age scheduled for elective surgery and ICU admission were enrolled prospectively. Two CGMS sensors were inserted into the subcutaneous tissue of the proximal lateral thigh and the lateral abdomen. The rate of successful measurements from thigh and abdomen in the OR and in the ICU were calculated separately. Each CGMS values were compared with the time-matched arterial blood glucose measurements. CGMS values from both measurement sites were also compared. A total of 22 patients undergoing cardiac surgeries were studied. The rate of successful measurements was higher in the ICU (73.2 %) than in the OR (66.0 %) (P = 0.01); however, that from thigh (72.9 %) and from abdomen (58.7 %) showed statistically significant difference only in the OR (P = 0.04). The Pearson correlation coefficient of thigh and abdomen versus arterial values was 0.67 and 0.60, respectively (P < 0.001). In Clarke error grid analysis, 94.6 % (89.3 % in the OR and 96.1 % in the ICU) of values from thigh fell into clinically acceptable zones compared to 93.7 % (89.0 % in the OR and 95.4 % in the ICU) from abdomen. There were no statistically significant differences in the accuracy according to measurement sites. The CGMS showed high measurement failure rate, especially in the OR. In the OR, the rate of successful measurement was higher from thigh than from abdomen. The CGMS showed low accuracy compared to arterial reference values. Nevertheless, there was no difference in the accuracy of the CGMS between two measurement sites. Perioperative performance of the CGMS still needs to be improved considering relatively low successful measurement rates.