Objectives: Past studies have suggested the absence of lag between palm glucose and fingertip glucose, even when glucose levels are changing rapidly. However, at any given time point, there may be differences between palm and fingertip glucose values because of glycemic instability and/or test methodology. The objectives of this study included assessing the variability in fingertip blood glucose test results between two fingers, and establishing whether the variability in blood glucose test results obtained from the palm was clinically equivalent to that observed in fingertip-to-fingertip comparisons.
Methods: This multicenter trial was conducted on patients under both steady-state glycemic conditions and after meal and exercise challenges (to promote rapidly changing glucose). Sequential capillary glucose testing, performed with the One Touch Ultra Blood Glucose Monitoring System (LifeScan, Inc., Milpitas, CA), was allocated to two of four fingertip sites and one of two palm sites in each subject using a randomized, balanced, incomplete block design. One of the fingertips was designated the reference site. Fingertip-to-fingertip variability and fingertip- to-palm variability were assessed under these steady-state and dynamic testing conditions using error grid analysis and by comparing the proportion of clinically acceptable blood glucose tests at the palm site versus the fingertip site. Clinically acceptable agreement was defined as pairs of values (fingertip to reference, or palm to reference) within 15 mg/dL when reference glucose was < or = 75 mg/dL or within 20% when reference glucose was >75 mg/dL.
Results: One hundred eighty-one subjects with type 1 [n = 74 (40.9%)] or type 2 [n = 107 (59.1%)] diabetes at eight clinical sites completed the study. Overall, the proportion of clinically acceptable agreement was high for both palm (95.1%) and fingertip (97.5%) testing. The mean difference between palm and fingertip clinically acceptable agreement when done by healthcare professionals was -1.3% and -4.4%, under steady-state and dynamic glycemic conditions, respectively. Error grid analysis showed >97% of all palm and fingertip measurements fell in Zone A.
Conclusion: This study demonstrated that variability between fingertip-to-fingertip and palm-to-fingertip measurements was in the clinically acceptable range during steady-state conditions and when glucose was rapidly changing.