Published online March 1, 2005
PEDIATRICS Vol. 115 No. 3 March 2005, pp. 662-666 (doi:10.1542/peds.2004-0820)
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hathout, E.
Right arrow Articles by Eastman, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hathout, E.
Right arrow Articles by Eastman, R.
Related Collections
Right arrow Endocrinology

Home Use of the GlucoWatch G2 Biographer in Children With Diabetes

Eba Hathout, MD*, Nina Patel, MD*, Christina Southern, MD*, Julie Hill, RN*, Reginald Anderson, MA*, Jeannine Sharkey, CDE*, Merrilee Hadley-Scofield, RD*, Long Tran, MA*, Amy Leptien, BA{ddagger}, Margarita Lopatin, MS{ddagger}, Betty Wang, MS{ddagger}, John Mace, MD* and Richard Eastman, MD{ddagger}

* Pediatric Diabetes Center, Loma Linda University, Loma Linda, California
{ddagger} Cygnus Inc, Redwood City, California


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Objective. To evaluate usability, accuracy, and hypoglycemia detection of the GlucoWatch G2 Biographer (GW2B) in children aged 1 to 17 years.

Methods. After a 15-hour study of device accuracy, 46 children (15 <7 years, 31 ≥7 years) with type 1 diabetes were enrolled for an extended-wear outcome study: 2 daytime and 2 nighttime 15-hr wear periods each week and blood glucose monitoring 4 times daily for 3 months.

Results. A total of 531 paired GW2B/meter readings were available for accuracy assessment. The correlation coefficients were 0.58 and 0.74 (ages <7 and ≥7 years, respectively). There was no significant change in hemoglobin A1C or weight-adjusted insulin dose at 3 months after biographer use. Forty-two episodes of hypoglycemia were detected by the GW2B, 33 of which were confirmed by blood glucose meters. Sensitivity and specificity of audible low-glucose alerts were ~79% and 83%, respectively. No significant side effects were reported.

Conclusion. The GW2B is usable and safe in children who are <7 years or older in the home setting. The GW2B can detect asymptomatic nocturnal hypoglycemia in younger children.

Key Words: continuous glucose monitoring • noninvasive glucose monitoring • pediatric diabetes

Abbreviations: GW2B, GlucoWatch G2 Automatic Glucose Biographer • HbA1c, hemoglobin A1c

Continuous blood glucose monitoring is a promising method for the management of type 1 diabetes, as it may provide means of ensuring continuous normoglycemia.1 The glucose concentration of interstitial fluid has been shown to reflect blood glucose concentration.2 Thus, continuous noninvasive extraction of interstitial fluid for glucose measurement can obviate the need for frequent skin-puncture blood sampling, currently recommended for intensive diabetes control.3 The GlucoWatch G2 Automatic Glucose Biographer (GW2B) (Cygnus Inc, Redwood City, CA) is an automatic noninvasive device that provides accurate glucose readings 6 times per hour. The main hypothesis of this project was that use of the GW2B is more effective than frequent skin-puncture sampling alone for the development of optimal therapy toward normoglycemia in children who are younger or older than 7 years and have type 1 diabetes.4 Infants and young children who are unable to verbalize symptoms are at particular risk for missed repeated hypoglycemia with potential long-term cognitive defects5 and, thus, could greatly benefit from near-continuous glucose monitoring.

The proposed device for continuous blood glucose monitoring in this experimental model was the GW2B, which is based on noninvasive sampling of subcutaneous interstitial fluid extracted by generating a transdermal low-level electric current.6 The principle, known as iontophoresis,7,8 involves enhanced transport of neutral molecules by current-induced solvent flow (electro-osmosis). Glucose is collected and measured by a replaceable AutoSensor. Six glucose readings are displayed each hour for up to 12 hours after calibration with a conventional capillary blood glucose measurement. Individual setting of target glucose ranges allows triggering of audible alarms on recognition of hypoglycemic (below-target) and hyperglycemic (above-target) values, as well as rapidly falling (>35% decrease from previous) glucose values. The device's lower and upper limit sensitivities are 40 mg/dL and 400 mg/dL, respectively. Out-of-range results trigger an alarm and display "low" (<40 mg/dL) or "high" (>400 mg/dL) messages.

Results of the Loma Linda Pediatric Accuracy Study were encouraging.9 Data suggested that the GW2B, whether worn on the forearm or at alternative sites, was safe and effective for diabetic children and adolescents of multiple ethnic backgrounds. Subsequent studies of the GW2B in pediatric patients with diabetes confirmed safety and effectiveness of the device in children and adolescents.10 Other reports of pediatric trials of the GW2B, including a multicenter trial, have been either encouraging11–13 or inconclusive.14,15


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
A convenience sample of 46 patients with type 1 diabetes at Loma Linda University Pediatric Diabetes Center were selected on the basis of expressed interest in trying a new noninvasive glucose monitoring device. The GW2B use protocol included 2 daytime and 2 nighttime 15-hour wear periods per week. GW2B wear sites included the forearm, ankle, and upper arm. Low and high glucose alert functions of the GW2Bs were set at 90 mg/dL and 220 mg/dL, respectively. Hypoglycemia was defined as meter-confirmed blood glucose readings <70 mg/dL, and its frequency was obtained from uploading of home glucose meters at baseline and weekly up to 3 months. Only 1 GW2B low glucose event was counted for each hour of low glucose reading. Finger-prick monitoring was done using the patient's own home glucose monitor (One Touch Ultra, Lifescan Inc, Milpitas CA; Precision Xtra, Abbott, Bedford, MA; or AccuCheck Complete, Roche, Indianapolis, IN). Home blood glucose monitoring was done at least 4 times a day in addition to calibration, recalibration, and/or alert time points of the GW2B. The majority of patients were on multiple daily injections, including neutral protamine Hagedorn (46%) and glargine (33%); 22% were on continuous subcutaneous insulin infusion via the Minimed 508 pump (Medtronics Inc, Sylmar, CA). Blood glucose and GW2B readings were uploaded using device-specific software weekly for 1 month, then monthly. Appropriate insulin adjustments were made on the basis of meter glucose values weekly by the study physician or coordinator. Withdrawal from the study was tracked, and reasons were documented. No detailed data regarding compliance with study instructions or lack thereof or number of wear periods per week before withdrawal were collected. Hemoglobin A1c (HbA1c) was measured at baseline, 6 weeks, and 3 months using the DCA2000 machine (Bayer Inc, Tarrytown, NY). Statistical analysis was done by 2-sided paired Wilcoxon signed-rank test, consensus error grid, and Deming regression analysis. All patients and parents signed the appropriate consent and assent forms approved by Loma Linda University institutional review board.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Two thirds of patients were white, and 57% were female (Table 1). Age range was 2.4 to 18.5 years, mean age was 9.6 years ± 4.7, and 33% of patients were younger than 7 years. Average diabetes duration was 4.5 years. At baseline, before wearing the GW2B, mean HbA1c was 8.0% ± 0.7, BMI was 19 kg/m2± 4, and total daily insulin was 0.79 units/kg/day ± 0.2. Baseline frequency of hypoglycemia was 13.7% ± 0.7. Peak times of GW2B wear start (calibration) were 8:00 to 10:00 AM and 6:00 to 8:00 PM (Fig 1). All devices could be calibrated at the beginning of a wear period, although not necessarily on the first attempt. More than half of the patients used the GW2B 4 or more times per week (Fig 2).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Patient Demographics

 

Figure 1
View larger version (20K):
[in this window]
[in a new window]
  		Fig 1. Distribution of time of day when the GW2B started (accumulated uses over 3 months).

 

Figure 2
View larger version (18K):
[in this window]
[in a new window]
  		Fig 2. Frequency of GW2B use (accumulated uses over 3 months).

 
There were 531 paired GW2B-meter readings for accuracy assessment at the end of 3 months (Table 2). GW2B calibration was successful on the first attempt in 84% of uses; calibration yielded on average 35 glucose readings of 78 potential readings per GW2B use period. Devices provided 9.5 and 9.7 hours of reading per wear period. Twenty-nine percent and 32% of readings were skipped as a result of data integrity checks. Correlation coefficients between GW2B and blood glucose meter were 0.74 and 0.58 for children older and younger than 7 years, respectively. Consensus error grid showed 91% of points within the A+B regions (Fig 3).


View this table:
[in this window]
[in a new window]
  		TABLE 2. Three-Month Data Analysis Summary

 

Figure 3
View larger version (33K):
[in this window]
[in a new window]
  		Fig 3. Consensus error grid analysis.

 
Of the 42 hypoglycemia episodes detected by the GW2B low alert set at 90 mg/dL, 33 were confirmed by the blood glucose meter, suggesting a GW2B sensitivity of 79% with specificity of 83% (Table 3). Sixty percent of hypoglycemic alerts were false alarms with a normal home glucose meter value. The decreased frequency of hypoglycemia from 13.7% to 11.2% ± 7 after GW2B use did not reach statistical significance (P = .1); neither did the reduction in HbA1c (8.02–7.96%). There were no significant changes in daily weight-adjusted insulin used or BMI at 3 months after GW2B use. Approximately 35% of patients withdrew from the study after 6 weeks (Table 4) as a result of irritation, distraction, or restlessness from audible low/high glucose alerts or inability to meet the study time commitment.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Hypoglycemia Alert Function

 

View this table:
[in this window]
[in a new window]
  		TABLE 4. Patient Enrollment Status by End of 3 Months

 

   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Our data show that the GW2B device is usable and tolerable in toddlers and children who are younger than 7 years. Despite a slight decrease in accuracy in this age group compared with older children, improvement with time of successful first calibration attempts is promising. The availability of an audible low alarm is a particularly valuable feature for this age group, in whom parental anxiety over nocturnal asymptomatic hypoglycemia can constitute a major obstacle to diabetes control. Data suggesting similar accuracy of alternative wear sites10 allow for enhanced usability through attachment to sites that are out of reach for the avid toddler. Although data on the device use in adults may seem more encouraging with a median correlation coefficient of 0.9 between device and standard methods,11,16,17 usability of the device with a dropping or low alert may be more significant in very young children, who are particularly susceptible to diabetes-related morbidity and mortality.5 The GW2B was approved for use in adults who have diabetes (patients older than 18 years) by the US Food and Drug Administration in March 2001 and in children who have diabetes and are older than 7 years in August 2002 but has not yet been approved in children who are younger than 7 years.

In terms of pediatric use, this study contrasts with a previous one11 in which use of the GW2B in children with type 1 diabetes was associated with a significant reduction in HbA1c and an increase in the frequency of hypoglycemia detection. Differences may be a function of the different nature of the 2 studies. The study by Chase et al11 compared an intervention with a randomized control group of patients, whereas this study analyses paired data with each patient serving as his or her own control. Patients who were selected for this study were different from those who enrolled in the 15-hour accuracy study. Thus, the training time on use of the GW2B was more adequate in the controlled study, which also tested tolerability and compliance with using the device a priori as part of a 15-hour accuracy study. Our study allowed patients to continue to use their own (1 of 3 brands) home glucose meter, such that intermeter variability could not be eliminated as a confounding variable. Compliance levels, in terms of duration of intensive home monitoring for our study patients, were not assessed before enrollment, and our data regarding postenrollment compliance and number of wear periods per week over time have not been analyzed. Despite the potential importance of the device in younger children, particularly at night, their withdrawal did not differ from older children, and their wear periods were similarly more adhered to in the daytime. Anecdotally, parents reported being frustrated with false alarms at night in a group that already had interrupted nocturnal sleep as a result of need for feeding or checking because of parental anxiety. Thus, poor compliance, including inconsistent use of the device over 3 months, is a potential negative compounding variable in assessing the usefulness of the study device. Moreover, no quality-of-life assessments were included in this analysis. A negative impact from false alarms is a potential factor. As withdrawal of patients continued in the younger age, smaller numbers yielded less statistical power.

It is noteworthy to mention that the true incidence of hypoglycemia detection of the GW2B's alert cannot be determined from this study because of the lack of continuous measurements of actual (capillary or venous) blood glucose. A recent multicenter inpatient study of children who were older than 7 years suggested a benefit of the device's dropping alert, especially at night, despite lesser accuracy for detection of low compared with high blood glucose readings.13

As for the lack of decrease in HbA1c in our study, this may be a reflection of the lower, although insignificant, decline in the frequency of hypoglycemia. The study adhered to US Food and Drug Administration regulations in terms of not using readings or alerts for the device as a basis for insulin adjustments. Whether parents were subjectively inclined to make minor modifications such as dietary ones on the hearing of a low alert has not been documented adequately; neither has the number of missed (unheard) low alerts, especially during sleep.

The compliance factor before and at the end of the study is an important parameter to be considered. It would have been valuable, although not part of the study design, to record and analyze frequency of blood glucose checks for the 3 months before and after study initiation. The initial fascination by a new device might be countered by frustration over the device's not meeting expected standards, thus compromising general compliance. Quality-of-life measures also would have been helpful in this respect. As the focus of the study was tolerability in the very young and utility in detection of hypoglycemia, data regarding high alerts have not been collected adequately for analysis.

Because of the short-term duration of our study, performance of the GW2B has not been studied sufficiently in conditions that lead to excess perspiration, such as vigorous exercise or severe heat. Effect of long-term continuous glucose monitoring via the GW2B on diabetes control, frequency, and severity of hypoglycemia and on the quality of life in children with diabetes and their parents was not evaluated in this study. Moreover, comparative standardized simple-task performance for detection of subtle cognitive deficits associated with hypoglycemia was not addressed therein The impact of using such an adjunct noninvasive device on the frequency and the severity of long-term microvascular complications requires more reliable use and prospective follow-up not permitted by the scope of this study. Ethnicity- and gender-related differences in diabetes management and control using the GW2B and differences in results in relation to parental involvement and motivation have not been examined. The ability of the GW2B to predict future blood glucose on the basis of glycemic pattern analysis remains to be tested to allow development of algorithms of trend-based management for optimal control.


   ACKNOWLEDGMENTS
 
The GlucoWatch G2 Biographer devices and AutoSensors in this study were provided by Cygnus, Inc. No additional funding was provided from the manufacturer.


   FOOTNOTES
 
Accepted Jul 26, 2004.

Reprint requests to (E.H.) Pediatric Diabetes Center, Loma Linda University, 11175 Campus St, CP A1120R, Loma Linda, CA 92354. E-mail: ehathout{at}ahs.llumc.edu

No conflict of interest declared.


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Klonoff D. The importance of continuous glucose monitoring in diabetes. Diabetes Technol Ther. 2000;1 :S1 –S4[CrossRef]
  2. Rebrin K, Steil G. Can interstitial glucose assessment replace blood glucose measurements? Diabetes Technol Ther. 2000;2 :461 –472[CrossRef][Medline]
  3. The Diabetes Control and Complication Trial Research Group. The Effect of intensive treatment of diabetes on the development and progression of long-term complications of insulin-dependent diabetes mellitus. N Engl J Med. 1993;329 :977 –986[Abstract/Free Full Text]
  4. Garg S, Potts R, Ackerman N, et al. Correlation of finger-stick blood glucose measurements with GlucoWatch Biographer glucose results in young subjects with type 1 diabetes. Diabetes Care. 1993;16 :1391 –1393[Abstract]
  5. Matyka KA, Wigg L, Pramming S, Stores G, Dunger DB. Cognitive function and mood after profound nocturnal hypoglycaemia in prepubertal children with conventional insulin treatment for diabetes. Arch Dis Child. 1999;81 :138[Abstract/Free Full Text]
  6. Tamada J, Bohannon N, Potts R. Measurement of glucose in diabetic subjects using noninvasive transdermal extraction. Nat Med. 1995;1 :1198 –1201[CrossRef][ISI][Medline]
  7. Pott R. Blood glucose monitoring by inverse iontophoresis. J Annu Diabetol Hotel Dieu. 2000:67 –74
  8. Glikfeld P. Noninvasive sampling of biological fluids by iontophoresis. Pharm Res. 1989;6 :989 –990
  9. Eastman RC, Hathout EH, Fuller-Byk L, et al. Accuracy of the GlucoWatch automatic glucose biographer in subjects 7–17 years of age with diabetes. Diabetes. 2001;50 :A378
  10. Eastman RC, Chase P, Buckingham B, et al. Use of the GlucoWatch Biographer in children and adolescents with diabetes. Pediatr Diabetes. 2002;3 :127 –134[CrossRef][Medline]
  11. Chase HP, Roberts MD, Wightman C, et al. Use of the Glucowatch Biographer in children with type 1 diabetes. Pediatrics. 2003;111 :790 –794[Abstract/Free Full Text]
  12. Gandrud L, Paguntalan H, VanWyhe M, et al. Use of the Cygnus GlucoWatch Biographer at diabetes camp. Pediatrics. 2004;113 :108 –111[Abstract/Free Full Text]
  13. Eastman RC, Leptien AD, Chase HP. Cost-effectiveness of the use of the GlucoWatch in children and adolescents under real-life conditions. Pediatr Diabetes. 2003;4 :82 –86[Medline]
  14. The Diabetes Research in Children Network (DirecNet) Study Group. Accuracy of the GlucoWatch G2 Biographer and the continuous glucose monitoring system during hypoglycemia. Diabetes Care. 2004;27 :722 –726[Abstract/Free Full Text]
  15. Bozzetti V, Viscardi M, Bonfanti R, et al. Results of continuous glucose monitoring by GlucoWatch Biographer in a cohort of diabetic children and adolescents under real-life conditions. Pediatr Diabetes. 2003;4 :57 –58[Medline]
  16. Cox DJ, Clarke WL, Gonder-Fredrick L, et al. Accuracy of perceiving blood glucose in IDDM. Diabetes Care. 1985;8 :529 –536[Abstract]
  17. Tamada J, Garg S, Jovanovic L, Pitzer K, Fermi S, Potts R. The Cygnus Research Team. Non-invasive glucose monitoring: comprehensive clinical results. JAMA. 1999;282 :1839 –1844[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2005 by the American Academy of Pediatrics



This article has been cited by other articles:


Home page
 The Journal of School NursingHome page
T. H. Wyatt, P. B. Krauskopf, and R. Davidson
Using Focus Groups for Program Planning and Evaluation
The Journal of School Nursing, April 1, 2008; 24(2): 71 - 77.
[Abstract] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow P3Rs: Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when P3Rs are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow E-mail this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My File Cabinet
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via CrossRef
Right arrow Citing Articles via ISI Web of Science (3)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hathout, E.
Right arrow Articles by Eastman, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hathout, E.
Right arrow Articles by Eastman, R.
Related Collections
Right arrow Endocrinology