Skip Navigation
Skip to contents

Diabetes Metab J : Diabetes & Metabolism Journal



Page Path
HOME > Diabetes Metab J > Volume 34(6); 2010 > Article
Original Article
The Correlation and Accuracy of Glucose Levels between Interstitial Fluid and Venous Plasma by Continuous Glucose Monitoring System
Young Ha Baek1, Heung Yong Jin1, Kyung Ae Lee1, Seon Mee Kang1, Woong Ji Kim1, Min Gul Kim2, Ji Hyun Park1, Soo Wan Chae3, Hong Sun Baek1, Tae Sun Park1,3
Korean Diabetes Journal 2010;34(6):350-358.
Published online: December 31, 2010
  • 51 Download
  • 14 Crossref

1Division of Endocrinology and Metabolism, Department of Internal Medicine, Chonbuk National University Hospital, Chonbuk National University Medical School, Jeonju, Korea.

2Department of Pharmacology, Chonbuk National University Medical School, Jeonju, Korea.

3Clinical Trial Center, Chonbuk National University Hospital, Chonbuk National University Medical School, Jeonju, Korea.

Corresponding author: Tae Sun Park. Division of Endocrinology and Metabolism, Department of Internal Medicine, Chonbuk National University Medical School, 634-18 Geumam 2-dong, Deokjin-gu, Jeonju 561-712, Korea.
• Received: March 15, 2010   • Accepted: September 16, 2010

Copyright © 2010 Korean Diabetes Association

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License ( which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • Background
    Clinical experience with the continuous glucose monitoring systems (CGMS) is limited in Korea. The objective of this study is to evaluate the accuracy of the CGMS and the correlation between interstitial fluid and venous plasma glucose level in Korean healthy male subjects.
  • Methods
    Thirty-two subjects were served with glucose solution contained same amount of test food's carbohydrate and test foods after separate overnight fasts. CGMS was performed over 3 days during hopitalization for each subjects. Venous plasma glucose measurements were carried out during 4 hours (0, 0.25, 0.5, 0.75, 1, 2, 4 hours) just before and after glucose solution and test food load. The performance of the CGMS was evaluated by comparing its readings to those obtained at the same time by the hexokinase method using the auto biochemistry machine (Hitachi 7600-110). Also, correlations between glucose recorded with CGMS and venous plasma glucose value were examined.
  • Results
    CGMS slightly underestimated the glucose value as compared with the venous plasma glucose level (16.3 ± 22.2 mg/dL). Correlation between CGMS and venous plasma glucose values throughout sensor lifetime is 0.73 (regression analysis: slope = 1.08, intercept = 8.38 mg/dL). Sensor sensitivity can deteriorate over time, with correlations between venous blood glucose and CGMS values dropping from 0.77 during 1st day to 0.65 during 2nd and 3rd day.
  • Conclusion
    The accuracy of data provided by CGMS may be less than expected. CGMS sensor sensitivity is decreased with the passage of time. But, from this study, CGMS can be used for glucose variability tendency monitoring conveniently to the Korean.
For the past 15 years, the use of short acting insulin and long-acting insulin, insulin pump development, and regular self-monitoring of blood glucose have played positive roles in reaching the desired level of glycated hemoglobin in diabetics. However, the fear of patients who still have hypoglycemia is one complicated element of blood glucose level regulation that cannot be overlooked [1]. In other words, normal blood glucose regulation is becoming the limiting factor in hypoglycemia in patients with diabetes [2].
In this sense, compared to the usual method of checking blood glucose levels 3 to 4 times per day, a continuous glucose monitoring system (CGMS), a system that provides relatively accurate information regarding blood glucose patterns, duration, variations, maximum and minimum levels, allows patients to be aware of their blood glucose levels at all times. The continuous measurement throughout a single day or 2 to 3 days provides insight to the patterns of blood glucose levels and gives warnings to recognize sudden low or high blood glucose levels and helps in preventing fatal episodes [2-4]. Furthermore, the CGMS is easy to use for patients. A great benefit is insertion and wearing the equipment involves minimal discomfort.
Using CGMS, when implemented previous prospective studies were examined, the goal glycated hemoglobin level was reached more quickly than before CGMS introduction [2,5-10], and complications were prevented. However, this benefit was only reported in one study. Precisely assessed studies on blood glucose regulation were very uncommon [1]. Looking at the situation in South Korea, only 17% of diabetics measure their blood glucose regularly at medical institutions [11]. Recently, raised interest or initiation of maintenance of diabetes has been occurring. Owing to the increased education of patients and medical professionals about diabetes, it is estimated that more patients are measuring their blood glucose regularly at medical institutions. In addition, in some diabetes centers, using CGMS, diurnal changes in interstitial fluid levels were measured. However limitations include: the increase in elderly, eastern traditions, difficulty mounting a new medical device, financial problems, and domestic problems which make it difficult for patients to use. Therefore, clinical experience for the use of CGMS is limited, so the reality is that various experiences and clinical applications are very difficult.
In this study, the authors evaluated the accuracy of CGMS and the relationship between the interstitial fluid and venous blood glucose level in healthy adults. Additionally, we compared differences of the blood glucose patterns according to the experimental diets, with the use and confirmation of flaws in the CGMS.
The subjects selected for this clinical trial were healthy Korean volunteers who satisfied all the selection criteria: 1) male aged 20 to 40 years, 2) body weight is 50 kg or more and whithin 30% of their ideal body weight {ideal body weight = (height [cm] - 100) × 0.9}, and 3) agreement of study enrollment. Informed consent was obtained from each subject. Exclusion criteria include the following: 1) individuals with illness of the liver, kidneys, nervous system, respiratory, endocrine, blood, tumors, mental illness, or a history of cardiovascular disease, 2) individuals with a history of gastrointestinal diseases (e.g., Crohn's disease) that can influence nutrient absorption, gastrointestinal surgery (excluding simple appendectomy or hernia surgery), 3) blood test findings aspartate transaminase (AST)/alanine transaminase (ALT) > 1.25 times the upper limit of the normal range, 4) individuals having a systolic blood pressure under 100 mm Hg and over 150 mm Hg, or individuals having a diastolic blood pressure under 65 mm Hg and over 95 mm Hg, 5) individuals with hypersensitivity to experimental diets or having a medical history of clinical allergic reactions, 6) individuals who took prescribed drugs or eastern medicines two weeks before the beginning of the study, or individuals who took over-the-counter (OTC) medication or vitamin supplements (However, candidates may be able to participate at the discretion of the examinees), 7) individuals who have participated in other clinical trials two months prior to the beginning of this study, 8) individuals who have donated blood up to one month prior to the study, or individuals who have donated plasma within the two weeks prior to the beginning of the study, 9) individuals who drink 21 drinks/week or individuals who cannot abstain from alcohol consumption or more during clinical trials, for any other reason that the examiner thought that participants' participation would be irrelevant. A 4 × 4 cross-examination was carried out on a total of 32 patients. Statistical administration occurred in groups of eight.
The study was performed in open-label, 4-treatment, 4-order, 4-time frame crossover design (Table 1); appropriate candidates were selected through an interview, physical exam, and disease screening 4 weeks (D-28 to D-1) prior to the clinical trial and a random assignment was carried out one day prior (D-1) to the first day of the study.
For the first period of clinical trial (period 1), study subjects were admitted to the Clinical Trial Center for Functional Foods in Chonbuk National University Hospital by 7:00 PM on the 1st day. The subjects fasted with the exception of drinking water after the initial dinner that was provided by the center. At approximately 9:00 AM on the 2nd day, all the subjects who had been fasting consumed a control diet based on their randomly assigned group. A control diet was composed of the same amount of glucose with the carbohydrate content in corresponding test meal. Caloric and nutritional composition of each test meal is as follows: Korean 1 (Bibimbap) 588 kcal (carbohydrates 86.1 g, protein 24.6 g, and fat 16.53 g), Korean 2 (Kimbap) 598.7 kcal (carbohydrates 83.1 g, protein 27.6 g, and fat 17.14 g), Western 1 (pork cutlet) 602.2 kcal (carbohydrates 66.4 g, protein 28.4 g, and fat 23.2 g), Western 2 (hamburger) 586 kcal (carbohydrates 63.2 g, protein 30.6 g, and fat 23.2 g). The ingredients of the test meal were produced in the same region to ensure the consistency of the quality. They were also stored for a minimal amount of time within the expiration date. The cooking methods were also kept consistent by establishing a standard method for meal preparation regarding intensity of the cooking fire and cooking time. This standard was used for each preparation. The study subjects consumed the provided control diet within 2 minutes.
They were prohibited from consuming anything besides water after dinner on the 2nd day and provided the test meal on 9 AM in the next morning according to the group. The menu of each test meal was designed to include representative dishes for common types of cuisine: Korean 1 (Bibimbap), Korean 2 (Kimbap), Western 1 (pork cutlet), and Western 2 (hamburger) (the study subjects were instructed to consume the test meal within 15 minutes). After the end of the period 1, all the subjects were discharged and 6-day's resting period was followed by the period 2. For the period 2, subjects were admitted again and provied the control diet and test meal according to the same method in period 1. Period 3 and 4 were carried out in sequence with intervening 6-day's resting period.
The self-monitoring of blood glucose unit used for continuous blood glucose measurement needed for the study was a CGMS model from Medtronic (Northridge, CA, USA). Data was collected by inserting the blood sugar sensor under the skin using plasma calibration with an auto biochemistry machine (Hitachi 7600-11; Hitachi, Tokyo, Japan) which was measured and analyzed using the hexokinase method. For this process, blood glucose measurements were taken 4 times a day and inputted into the CGMS. Continuous blood glucose measurements were taken on day one (and on the 2nd, 8th, 9th, 15th, 16th, 22nd, 23rd day) between hour 0 (before the meal) and hour 4 every 5 minutes. Blood samples for measurement of blood glucose and insulin was taken on day one (and 2nd, 8th, 9th, 15th, 16th, 22nd, 23rd day) at hour 0 (before the meal), hour 0.25, hour 0.5, hour 0.75, hour 1, hour 2, and hour 4. A blood sample of 1 mL was taken from brachial veins without the use of a catheter due to the characteristics of the research methods.
Statistical analysis
Statistical analysis was performed using the following methods using a significance level of 0.05 (average ± standard deviation) with SPSS (SPSS Inc., Chicago, IL, USA), SAS (SAS Institute, Cary, NC, USA) and/or S-plus (TIBCO, Palo Alto, CA, USA) statistical software. Demographic statistical analysis of participants of the clinical trial was carried out to ensure random sampling. Participant age, height, body weight, and alcohol and caffeine consumption information was collected for demographic statistical analysis.
Analysis of variance (ANOVA) was used for continuous variables, and Fisher's exact test was used for categorical variables. The correlation and significance between the glucose level measured from the reference sample of venous blood, and that from CGMS were analyzed using linear regression analysis.
Each measured level was marked as a distribution point using a Clarke error grid analysis; areas A and B are clinically acceptable regions and areas C, D, and E are regions which are considered unsafe or unacceptable.
Clinical characteristics of the participants
Thirty-two adult males were randomly assigned to participate in this study. Mean age was 23.9 ± 1.5 years, mean height was 173.9 ± 5.3 cm, mean body weight was 64.8 ± 6.3 kg, fasting plasma glucose/insulin was 88.3 ± 5.6 mg/dL/3.6 ± 1.7 uU/mL, respectively, blood urea nitrogen (BUN)/Creatinine was 13.7 ± 2.8 mg/dL/0.97 ± 0.1 mg/dL, AST/ALT was 20.3 ± 8.6 IU/L/17.9 ± 13.4 IU/L, and the lipid profile (cholesterol/triglyceride/HDL/LDL) was 159.6 ± 26.6 mg/dL/103.3 ± 40.9 mg/dL/51.9 ± 8.0 mg/dL/96.4 ± 26.9 mg/dL, respectively. The mean value for body mass index was 20.8 ± 1.5 (Table 2). Personal medical history, physical examination, and baseline blood examination showed no associated diseases.
The comparison of the glucose profiles from CGMS and venous plasma glucose values of each group
Fig. 1 shows the glucose profiles from CGMS and venous plasma glucose for each group, and the patterns and relationship between them. Values of postprandial venous plasma glucose levels of all groups were between 130 to 190 mg/dL. Plasma glucose 4 hours after eating was measured to be 60 to 90 mg/dL. The highest blood glucose measurement was 171 mg/dL, and the lowest blood glucose measurement was 71 mg/dL. The CGMS showed that peak postprandial glucose level was 120 to 140 mg/dL which was lower than that from plasma and the time to the peak glucose level was 45 to 75 minutes which was longer than 30 to 60minutes in plasma value. With the CGMS, the highest value was measured to be 131 mg/dL, and the lowest value was measured to be 71.6 mg/dL. The overall distribution pattern of intravenous glucose and CGMS values were similar in each group.
The comparison of the glucose level according to the control diet and test meal
Looking at the CGMS and venous plasma glucose patterns from different types of diets (Fig. 2), the pattern of venous plasma glucose in all control diet and test meal reached approximately 100 to 190 mg/dL 30 to 45 minutes after consumption and then decreased in 60 minutes after meal, and normal fasting glucose level were seen 4 hours after meal. Additionally, the control diet resulted in the highest postprandial blood glucose level, and also the lowest postprandial venous plasma glucose level in all four groups, but it had no statistical significance.
Additionally, glucose value distribution measured with CGMS showed the similar pattern of postprandial increament in plasma glucose level, however, time to peak level was 45 to 60 minutes. Four hours after eating, the glucose level were similar to that from plasma again (Fig. 3). The highest and lowest glucose level in CGMS were also seen in the control diet and there was no significant difference in the value from CGMS and plasma in each group.
The relationship and correlation between the glucose profiles from CGMS and venous plasma glucose measurement values
During the entire experimental period for all of the groups, when comparing the glucose level measured with CGMS and venous plasma glucose level, postprandial blood glucose patterns and even fasting plasma glucose level patterns decreased in a similar fashion. The time lag phenomenon in the peak postprandial glucose level between the CGMS value and plasma glucose were seen in gll group and all kinds of diet. Additionally, the difference between each of the CGMS and venous plasma glucose values for hyperglycemia period (45 to 60 minutes after eating), and hyperglycemia period (0 to 30 minuites after eating and 2 to 4 hours after eating) 34 ± 8.23 mg/dL, and 2.9 ± 4.94 mg/dL respectively, which confirmed that the accuracy in the hyperglycemic period was declining. The value from CGMS was lower than venous plasma glucose in all the time-section (16.3 ± 22.2 mg/dL). The correlation value between CGMS and venous plasma glucose was 0.73 (Fig. 4; regression analysis: slope = 1.08, intercept = 8.38 mg/dL). Additionally, correlation of these values, on the 1st day of clinical trials was 0.77, and on the 2nd day, it decreased to 0.65 (Figs. 5 and 6). These results were analyzed by error grid analysis (n = 1,568). Area A and area B are located at 95.4%, and CGMS's usefulness was confirmed (Fig. 7).
One benefit of the CGMS is that it can provide more information on the fluctuations of blood glucose. In other words, when blood glucose cannot be pin pointed, CGMS provides more benefits than intermittent methods of measuring blood glucose [10]. This benefit can greatly reduce the incidence of complications in uncontrolled blood glucose, and patients' concerns are minimized, and it allows for more active blood glucose regulation. However, as previously mentioned, utilizing CGMS to care for patients may not be practical. In reality, clinical experience in utilizing CGMS is minimal in Korea.
Through this clinical trial, authors discovered that the result of CGMS was very similar to the venous plasma glucose level, and the correlation between CGMS and the value of venous plasma glucose level was relatively accurate. These points adequately explained the general benefits of the CGMS. As a result, previous reports regarding the accuracy of the CGMS are considered to be similar [12-16], and it appears that using the CGMS in the future in Korean patients with diabetes can occur without difficulty.
There are reports that state there is no proof that CGMS is better at regulating blood glucose and metabolism when measuring and comparing venous blood glucose levels [17], but there are reports that show the benefits of CGMS in practice in reaching target glycated hemoglobin levels and adjusting it to appropriate levels [1,2,5-10]. Therefore, it is thought that the domestic application of CGMS should be more active, so that research and reporting similar to this can be more active.
However, when this study was compared with the results of the main study that was published prior to this, some were able to confirm the significance level and the decline of the accuracy on the error grid in zones A and B at 98.4% [12] vs. 95.4%. However, there were a relatively small number of subjects and a short follow-up period. A study involving more patients would be required for further explanation.
When the CGMS measured values and venous plasma glucose concentration of each group is compared, no significant difference was observed in the overall changes of blood glucose. Although blood glucose values measured with CGMS was not statistically significant, it showed a lower trend, which in this study; target participants were not patients with chronic diseases including diabetes, but healthy adult male patients. In addition, a comparison with diabetes patients must also be made. CGMS did not show statistically significant differences between implementing control diet and experimental diet for each diet group and intravenous glucose. Also, we think this because this is a study implemented by targeting non-diabetes healthy male subjects. In the future, the importance of the glycemic index and glycemic load of this diet group is studied, but it is thought to be utilizable.
When CGMS results were compared with the venous plasma glucose concentrations, CGMS's time lag was observed. Glucose achieves equilibrium through interstitial fluid in the blood through this, and at this time, it is a physiological process [18]. It is assumed that CGMS measurement values during hypoglycemia or sudden pattern changes may be imprecise [19].
In addition, the sensor's sensitivity decreased over time, and the accuracy decline was observed. These results have been reported in previous CGMS clinical trials [12,13]: bleeding and irritation of the skin between sensors [18], adsorption of tissue around the sensor, localized accumulation of inflamed cells, and fibrous encapsulation are thought to have caused these results [20-22]. Using corticosteroid spray on the sensors for all treatments can reduce these problems [9]. Further clinical trials are required to confirm this.
In conclusion, this study investigated the data obtained from a CGMS and venous plasma glucose level, which was confirmed to have a high correlation. There was no statistical difference observed in the changing patterns in blood glucose from food types and experimental diets, and it was confirmed that there are still problems such as the decrease in time lag and sensor sensitivity. Additionally, it is reported that CGMS was performed in many previous studies involving diabetes patient groups; especially type 1 diabetics, pregnant patients, and special cases such as patients on dialysis. In this clinical trial, healthy adults were used because the accuracy of the correlation between changes in glucose is more credible. However, since it is used as such on diabetes patients, there are limits to the interpretation of the results. Additionally, the number of participants in this study was small. For increased reliability and representation of the clinical trials, more personnel are needed. Others cannot comment on the accuracy of all CGMS's between CGMS machines; and because differences could not be assessed, future studies should include the CGMS used in this study so each machine's correlation and accuracy can be compared. However, through this clinical trial, the authors were able to confirm that CGMS would be very useful in the domestic market to figure out changes in blood glucose and distribution states as levels of blood glucose measured by CGMS were very similar to venous blood. Therefore, continuous research should be conducted so that CGMS can be used as a method to increase the success of reaching target blood glucose level and also to determine appropriate controlling methods for blood glucose.
  • 1. Juvenile Diabetes Research Foundation Continuous Glucose Monitoring Study Group. The effect of continuous glucose monitoring in well-controlled type 1 diabetes. Diabetes Care 2009;32:1378-1383. ArticlePubMedPMCPDF
  • 2. Cryer PE. Banting lecture. Hypoglycemia: the limiting factor in the management of IDDM. Diabetes 1994;43:1378-1389. ArticlePubMed
  • 3. Chico A, Vidal-Rios P, Subira M, Novials A. The continuous glucose monitoring system is useful for detecting unrecognized hypoglycemias in patients with type 1 and type 2 diabetes but is not better than frequent capillary glucose measurements for improving metabolic control. Diabetes Care 2003;26:1153-1157. ArticlePubMedPDF
  • 4. Ludvigsson J, Hanas R. Continuous subcutaneous glucose monitoring improved metabolic control in pediatric patients with type 1 diabetes: a controlled crossover study. Pediatrics 2003;111(5 Pt 1):933-938. ArticlePubMedPDF
  • 5. Chase HP, Kim LM, Owen SL, MacKenzie TA, Klingensmith GJ, Murtfeldt R, Garg SK. Continuous subcutaneous glucose monitoring in children with type 1 diabetes. Pediatrics 2001;107:222-226. ArticlePubMedPDF
  • 6. Boland E, Monsod T, Delucia M, Brandt CA, Fernando S, Tamborlane WV. Limitations of conventional methods of self-monitoring of blood glucose: lessons learned from 3 days of continuous glucose sensing in pediatric patients with type 1 diabetes. Diabetes Care 2001;24:1858-1862. PubMed
  • 7. Danne T, Deiss D, Hopfenmuller W, von Schutz W, Kordonouri O. Experience with insulin analogues in children. Horm Res 2002;57(Suppl 1):46-53. ArticlePubMedPDF
  • 8. Weintrob N, Schechter A, Benzaquen H, Shalitin S, Lilos P, Galatzer A, Phillip M. Glycemic patterns detected by continu-ous subcutaneous glucose sensing in children and adolescents with type 1 diabetes mellitus treated by multiple daily injections vs continuous subcutaneous insulin infusion. Arch Pediatr Adolesc Med 2004;158:677-684. ArticlePubMed
  • 9. Bolinder J, Ungerstedt U, Arner P. Long-term continuous glucose monitoring with microdialysis in ambulatory insulin-dependent diabetic patients. Lancet 1993;342:1080-1085. ArticlePubMed
  • 10. Klonoff DC. Continuous glucose monitoring: roadmap for 21st century diabetes therapy. Diabetes Care 2005;28:1231-1239. PubMed
  • 11. Rhee SY, Kim YS, Oh S, Choi WH, Park JE, Jeong WJ. Diabcare Asia 2001-Korea: country report on outcome data and analysis. Korean J Intern Med 2005;20:48-54. ArticlePubMedPMC
  • 12. Weinstein RL, Schwartz SL, Brazg RL, Bugler JR, Peyser TA, McGarraugh GV. Accuracy of the 5-day Free Style Navigator Continuous Glucose Monitoring System: comparison with frequent laboratory reference measurements. Diabetes Care 2007;30:1125-1130. PubMed
  • 13. Gross TM, Mastrototaro JJ. Efficacy and reliability of the continuous glucose monitoring system. Diabetes Technol Ther 2000;2(Suppl 1):S19-S26. ArticlePubMed
  • 14. Gross TM, Ter Veer A. Continuous glucose monitoring in previously unstudied population subgroups. Diabetes Technol Ther 2000;2(Suppl 1):S27-S34. ArticlePubMed
  • 15. Gross TM, Bode BW, Einhorn D, Kayne DM, Reed JH, White NH, Mastrototaro JJ. Performance evaluation of the MiniMed continuous glucose monitoring system during patient home use. Diabetes Technol Ther 2000;2:49-56. ArticlePubMed
  • 16. Bode BW, Gross TM, Thornton KR, Mastrototaro JJ. Continuous glucose monitoring used to adjust diabetes therapy improves glycosylated hemoglobin: a pilot study. Diabetes Res Clin Pract 1999;46:183-190. ArticlePubMed
  • 17. Tanenberg R, Bode B, Lane W, Levetan C, Mestman J, Harmel AP, Tobian J, Gross T, Mastrototaro J. Use of the Continuous Glucose Monitoring System to guide therapy in patients with insulin-treated diabetes: a randomized controlled trial. Mayo Clin Proc 2004;79:1521-1526. ArticlePubMed
  • 18. Sternberg F, Meyerhoff C, Mennel FJ, Mayer H, Bischof F, Pfeiffer EF. Does fall in tissue glucose precede fall in blood glucose? Diabetologia 1996;39:609-612. ArticlePubMedPDF
  • 19. Kulcu E, Tamada JA, Reach G, Potts RO, Lesho MJ. Physiological differences between interstitial glucose and blood glucose measured in human subjects. Diabetes Care 2003;26:2405-2409. ArticlePubMedPDF
  • 20. Gerritsen M, Jansen JA, Lutterman JA. Performance of subcutaneously implanted glucose sensors for continuous monitoring. Neth J Med 1999;54:167-179. ArticlePubMed
  • 21. Gerritsen M. Problems associated with subcutaneously implanted glucose sensors. Diabetes Care 2000;23:143-145. ArticlePubMedPDF
  • 22. Gerritsen M, Jansen JA, Kros A, Vriezema DM, Sommerdijk NA, Nolte RJ, Lutterman JA, Van Hovell SW, Van der Gaag A. Influence of inflammatory cells and serum on the performance of implantable glucose sensors. J Biomed Mater Res 2001;54:69-75. ArticlePubMed
Fig. 1
(A) group 1, (B) group 2, (C) group 3, and (D) group 4. The relation between continuous glucose monitoring system (CGMS) and venous blood glucose levels according to the subject groups. CGMS slightly underestimated the glucose value as compared with the venous plasma glucose level. "Time-lag" is observed. However, between each group were similar to venous glucose and CGMS aspects. Data was shown as mean ± standard deviation. Rc: Control , Rt: Test meal (R: Glucose from CGMS vs. Blood plasma glucose).
Fig. 2
(A) Korean meal 1, (B) Korean meal 2, (C) Western meal 1, and (D) Western meal 2. The relation between continuous glucose monitoring system (CGMS) and venous plasma glucose levels according to the meals. When compared by each diet, venous glucose and CGMS levels showed similar patterns that did not differ significantly. Data was shown as mean ± standard deviation. Rc: Control , Rt: Test meal (R: Glucose from CGMS vs. Blood plasma glucose).
Fig. 3
The glucose levels measured by continuous glucose monitoring system between controls (O) and treatment (Δ) groups according to the time passage.
Fig. 4
Correlation between continuous glucose monitoring system (CGMS) and venous plasma glucose levels (regression analysis: slope = 1.08, intercept = 8.38 mg/dL).
Fig. 5
Correlation between continuous glucose monitoring system (CGMS) and venous plasma glucose levels (1st day, regression analysis: slope = 1.08, intercept = 8.38 mg/dL).
Fig. 6
Correlation between continuous glucose monitoring system (CGMS) and venous plasma glucose levels (2nd day, regression analysis: slope = 1.08, intercept = 8.38 mg/dL).
Fig. 7
Correlation between continuous glucose monitoring system (CGMS) and venous plasma glucose levels using the Clarke error grid analysis (n = 1,568); 95.4% measures were within zones A and B.
Table 1
Open label, 4-treatment, 4-sequence, 4-perioid crossover study
Table 2
Baseline characteristics of the patients

BUN, blood urea nitrogen; AST, aspartate transaminase; ALT, alanine transaminase; TG, triglyceride; BMI, body mass index.

Figure & Data



    Citations to this article as recorded by  
    • Evaluation of the performance and usability of a novel continuous glucose monitoring system
      Li Yan, Qiang Li, Qingbo Guan, Mingsong Han, Yu Zhao, Junfei Fang, Jiajun Zhao
      International Journal of Diabetes in Developing Countries.2023; 43(4): 551.     CrossRef
    • Correlation between short- and mid-term hemoglobin A1c and glycemic control determined by continuous glucose monitoring
      Jen-Hung Huang, Yung-Kuo Lin, Ting-Wei Lee, Han-Wen Liu, Yu-Mei Chien, Yu-Chun Hsueh, Ting-I Lee, Yi-Jen Chen
      Diabetology & Metabolic Syndrome.2021;[Epub]     CrossRef
    • Accuracy of Flash Glucose Monitoring During Postprandial Rest and Different Walking Conditions in Overweight or Obese Young Adults
      Xiaoyuan Zhang, Fenghua Sun, Waris Wongpipit, Wendy Y. J. Huang, Stephen H. S. Wong
      Frontiers in Physiology.2021;[Epub]     CrossRef
    • The MEDGICarb-Study: Design of a multi-center randomized controlled trial to determine the differential health-promoting effects of low- and high-glycemic index Mediterranean-style eating patterns
      Robert E. Bergia, Izabela Biskup, Rosalba Giacco, Giuseppina Costabile, Savanna Gray, Amy Wright, Marilena Vitale, Wayne W. Campbell, Rikard Landberg, Gabriele Riccardi
      Contemporary Clinical Trials Communications.2020; 19: 100640.     CrossRef
    • A genetic programming-based regression for extrapolating a blood glucose-dynamics model from interstitial glucose measurements and their first derivatives
      I. De Falco, A. Della Cioppa, A. Giugliano, A. Marcelli, T. Koutny, M. Krcma, U. Scafuri, E. Tarantino
      Applied Soft Computing.2019; 77: 316.     CrossRef
    • Genetic Programming-based induction of a glucose-dynamics model for telemedicine
      Ivanoe De Falco, Antonio Della Cioppa, Tomas Koutny, Michal Krcma, Umberto Scafuri, Ernesto Tarantino
      Journal of Network and Computer Applications.2018; 119: 1.     CrossRef
    • A high-accuracy measurement method of glucose concentration in interstitial fluid based on microdialysis
      Dachao Li, Qingmei Xu, Yu Liu, Ridong Wang, Kexin Xu, Haixia Yu
      Measurement Science and Technology.2017; 28(11): 115701.     CrossRef
    • Effects of Higher Dietary Protein and Fiber Intakes at Breakfast on Postprandial Glucose, Insulin, and 24-h Interstitial Glucose in Overweight Adults
      Akua Amankwaah, R. Sayer, Amy Wright, Ningning Chen, Megan McCrory, Wayne Campbell
      Nutrients.2017; 9(4): 352.     CrossRef
    • High Surface Area Electrodes Generated via Electrochemical Roughening Improve the Signaling of Electrochemical Aptamer-Based Biosensors
      Netzahualcóyotl Arroyo-Currás, Karen Scida, Kyle L. Ploense, Tod E. Kippin, Kevin W. Plaxco
      Analytical Chemistry.2017; 89(22): 12185.     CrossRef
    • Hyperglycemia-Induced Changes in Hyaluronan Contribute to Impaired Skin Wound Healing in Diabetes: Review and Perspective
      Sajina Shakya, Yan Wang, Judith A. Mack, Edward V. Maytin
      International Journal of Cell Biology.2015; 2015: 1.     CrossRef
    • Hypoglycemia in everyday life after gastric bypass and duodenal switch
      Niclas Abrahamsson, Britt Edén Engström, Magnus Sundbom, F Anders Karlsson
      European Journal of Endocrinology.2015; 173(1): 91.     CrossRef
    • The use of reinforcement learning algorithms to meet the challenges of an artificial pancreas
      Melanie K Bothe, Luke Dickens, Katrin Reichel, Arn Tellmann, Björn Ellger, Martin Westphal, Ahmed A Faisal
      Expert Review of Medical Devices.2013; 10(5): 661.     CrossRef
    • Continuous glucose monitoring: current clinical use
      Hun‐Sung Kim, Jeong‐Ah Shin, Jin‐Sun Chang, Jae‐Hyoung Cho, Ho‐Young Son, Kun‐Ho Yoon
      Diabetes/Metabolism Research and Reviews.2012; 28(s2): 73.     CrossRef
    • Correlations of Glucose Levels in Interstitial Fluid Estimated by Continuous Glucose Monitoring Systems and Venous Plasma
      Byung-Joon Kim
      Korean Diabetes Journal.2010; 34(6): 338.     CrossRef

    • PubReader PubReader
    • Cite this Article
      Cite this Article
      export Copy Download
      Download Citation
      Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

      • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
      • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
      • Citation for the content below
      The Correlation and Accuracy of Glucose Levels between Interstitial Fluid and Venous Plasma by Continuous Glucose Monitoring System
      Korean Diabetes J. 2010;34(6):350-358.   Published online December 31, 2010
    • XML DownloadXML Download
    Baek YH, Jin HY, Lee KA, Kang SM, Kim WJ, Kim MG, Park JH, Chae SW, Baek HS, Park TS. The Correlation and Accuracy of Glucose Levels between Interstitial Fluid and Venous Plasma by Continuous Glucose Monitoring System. Diabetes Metab J. 2010;34(6):350-358.
    Received: Mar 15, 2010; Accepted: Sep 16, 2010

    Diabetes Metab J : Diabetes & Metabolism Journal
    Close layer