Accuracy of Two Continuous Glucose Monitors Differs after Hydroxyurea in Pediatric Patients Undergoing Total Pancreatectomy with Islet Autotransplantation.
Jennifer M Ladd, Kajal Gandhi, Kelly Friesner-Gephart, Robert P Hoffman, Mary Joy Okafor, Christie Heinzman, Cheryl E Gariepy, Sara K Rasmussen, Jaimie D Nathan, A Jay Freeman
{"title":"Accuracy of Two Continuous Glucose Monitors Differs after Hydroxyurea in Pediatric Patients Undergoing Total Pancreatectomy with Islet Autotransplantation.","authors":"Jennifer M Ladd, Kajal Gandhi, Kelly Friesner-Gephart, Robert P Hoffman, Mary Joy Okafor, Christie Heinzman, Cheryl E Gariepy, Sara K Rasmussen, Jaimie D Nathan, A Jay Freeman","doi":"10.1089/dia.2025.0010","DOIUrl":null,"url":null,"abstract":"<p><p><b><i>Background:</i></b> Total pancreatectomy with islet autotransplantation (TPIAT) requires strict glycemic management for islet survival using insulin pumps and continuous glucose monitors (CGMs). Hydroxyurea prevents reactive thrombocytosis but interferes with the accuracy of the Dexcom CGM. Hydroxyurea is reported to not interfere with the Libre CGM but has not been studied after TPIAT. <b><i>Methods:</i></b> Seven patients wore both Dexcom and Libre starting approximately a week after TPIAT. Dexcom and Libre values were obtained with point-of-care testing blood glucose (POCT BG) at 560 unique time points. Descriptive statistics included median, interquartile range (IQR), absolute difference between CGM and POCT, and mean absolute relative difference (MARD) for each Dexcom and Libre. Wilcoxon-Mann-Whitney tests were performed to compare parameters between Dexcom and Libre, with two-sided significance of <i>P</i> < 0.05. Clarke error grids and boxplots were constructed. <b><i>Results:</i></b> In the 9 h after hydroxyurea, median POCT BG was 110 mg/dL (IQR 88-143), median Dexcom BG was 172 mg/dL (135-219), and median Libre BG was 106 mg/dL (76-138). MARD for Dexcom was 59.5% and for Libre was 14.8% (<i>P</i> < 0.001). Median absolute difference between Dexcom and POCT BG (56 mg/dL [32-88]) was greater than that for Libre (12 mg/dL [6-23]; <i>P</i> < 0.001). In Clarke error grids, 98.3% of values fell within clinically acceptable Zones A/B for Libre; 77.9% of values fell within these zones for Dexcom. At all other times, median POCT BG was 110 mg/dL (86-133), median Dexcom BG was 124 mg/dL (97-154), and median Libre BG was 104 mg/dL (76-128). MARD for Dexcom was 19.8% and for Libre was 14.7% (<i>P</i> < 0.001). Median absolute difference between Dexcom and POCT BG (18 mg/dL [9-30]) was clinically similar to that for Libre (13 mg/dL [6-23], <i>P</i> < 0.001). <b><i>Conclusion:</i></b> Hydroxyurea does not seem to interfere with the accuracy of Libre in contrast to Dexcom. Use of Libre after TPIAT could facilitate improved glycemic management.</p>","PeriodicalId":11159,"journal":{"name":"Diabetes technology & therapeutics","volume":" ","pages":""},"PeriodicalIF":5.7000,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Diabetes technology & therapeutics","FirstCategoryId":"3","ListUrlMain":"https://doi.org/10.1089/dia.2025.0010","RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"ENDOCRINOLOGY & METABOLISM","Score":null,"Total":0}
引用次数: 0
Abstract
Background: Total pancreatectomy with islet autotransplantation (TPIAT) requires strict glycemic management for islet survival using insulin pumps and continuous glucose monitors (CGMs). Hydroxyurea prevents reactive thrombocytosis but interferes with the accuracy of the Dexcom CGM. Hydroxyurea is reported to not interfere with the Libre CGM but has not been studied after TPIAT. Methods: Seven patients wore both Dexcom and Libre starting approximately a week after TPIAT. Dexcom and Libre values were obtained with point-of-care testing blood glucose (POCT BG) at 560 unique time points. Descriptive statistics included median, interquartile range (IQR), absolute difference between CGM and POCT, and mean absolute relative difference (MARD) for each Dexcom and Libre. Wilcoxon-Mann-Whitney tests were performed to compare parameters between Dexcom and Libre, with two-sided significance of P < 0.05. Clarke error grids and boxplots were constructed. Results: In the 9 h after hydroxyurea, median POCT BG was 110 mg/dL (IQR 88-143), median Dexcom BG was 172 mg/dL (135-219), and median Libre BG was 106 mg/dL (76-138). MARD for Dexcom was 59.5% and for Libre was 14.8% (P < 0.001). Median absolute difference between Dexcom and POCT BG (56 mg/dL [32-88]) was greater than that for Libre (12 mg/dL [6-23]; P < 0.001). In Clarke error grids, 98.3% of values fell within clinically acceptable Zones A/B for Libre; 77.9% of values fell within these zones for Dexcom. At all other times, median POCT BG was 110 mg/dL (86-133), median Dexcom BG was 124 mg/dL (97-154), and median Libre BG was 104 mg/dL (76-128). MARD for Dexcom was 19.8% and for Libre was 14.7% (P < 0.001). Median absolute difference between Dexcom and POCT BG (18 mg/dL [9-30]) was clinically similar to that for Libre (13 mg/dL [6-23], P < 0.001). Conclusion: Hydroxyurea does not seem to interfere with the accuracy of Libre in contrast to Dexcom. Use of Libre after TPIAT could facilitate improved glycemic management.
期刊介绍:
Diabetes Technology & Therapeutics is the only peer-reviewed journal providing healthcare professionals with information on new devices, drugs, drug delivery systems, and software for managing patients with diabetes. This leading international journal delivers practical information and comprehensive coverage of cutting-edge technologies and therapeutics in the field, and each issue highlights new pharmacological and device developments to optimize patient care.