|Home | About | Journals | Submit | Contact Us | Français|
The aim of this study was to determine the accuracy of the Guardian® RT system (Medtronic Minimed, Northridge, CA) in young children and adolescents with type 1 diabetes (T1D) during different scenarios of glucose levels and sensor age.
At five clinical centers, 30 subjects between 4 and 17 years old with T1D were recruited. All subjects had a glycosylated hemoglobin level of ≤10.0% and were using an insulin pump. Subjects initially used a Guardian RT for approximately 1 week at home. Each subject was then hospitalized overnight for about 18h in a clinical research center, during which time insulin-induced hypoglycemia occurred, along with frequently sampled glucose.
There were 1,511 laboratory glucose measurements paired with glucose measurements from 48 Guardian RT sensors. Overall, the median absolute difference (AD) was 21mg/dL, and the median relative AD (RAD) was 14%, with 64% of sensor values meeting International Organization for Standardization home glucose meter criteria. The median AD was 27mg/dL for reference glucose values ≤60mg/dL and 25mg/dL for reference glucose values ≤70mg/dL. The median RAD was 19% for reference glucose values 71–120mg/dL, 14% for reference glucose values 121–180mg/dL, and 10% for reference glucose values >180mg/dL.
The Guardian RT appears to perform as well in children with T1D as it has been reported to perform in adults with diabetes. The Guardian RT has an accuracy similar to that of other available continuous glucose monitors and can give important and useful clinical information.
Continuous glucose monitors are becoming an extremely useful tool in the management of diabetes. These devices are able to display a glucose value in near real-time and have alarms for high and low blood sugars. Data are also stored for later analysis of glucose trends, which can be of great benefit in insulin dose adjustments for patients with diabetes. Accuracy remains a key issue for continuous glucose monitors, particularly in children and adolescents, who may have increased variability of blood glucose. Accuracy in the hypoglycemic ranges is of vital importance in the clinical use of real-time continuous glucose monitors.
The Guardian® RT system (Medtronic Minimed, Northridge, CA) uses a glucose oxidase-based electrochemical sensor. The system is designed to measure blood glucose levels in a range of 40–400mg/dL. The sensor is inserted subcutaneously and measures interstitial glucose every 10 s. These values are averaged in the monitor to provide a reading every 5min (or 288 readings a day), which is transmitted wirelessly to a receiver that can be kept up to 6 feet from the transmitter. Each sensor is designed to measure readings for up to 72 h. The system has alarms for hypoglycemia and hyperglycemia that can be adjusted by the user. Subjects can enter events, such as insulin administration, meals, or exercise. The sensor requires at least two capillary glucose readings each day (every 12 h) to validate sensor function and allow for development of a calibration equation. The Guardian RT system has been approved by the Food and Drug Administration for detecting trends and tracking patterns in adults (18 years of age and older) and is indicated as an adjunctive method rather than replacement of standard home glucose monitoring devices. The Guardian RT and the Paradigm® REAL-Time system (which includes both the continuous glucose monitor and a pump) (Medtronic Minimed) use the same sensor.
The aim of this study was to determine the accuracy of the Guardian RT in young children and adolescents during different scenarios of glucose levels and sensor age.
The study was conducted by the Diabetes Research in Children Network (DirecNet) at five clinical centers. Subjects were participating in another study, conducted in Clinical Research Centers (CRCs), that was evaluating the counter-regulatory response to hypoglycemia in younger (3 to <8 years old) and older (12 to <18 years old) children. Eligibility criteria included a clinical diagnosis of type 1 diabetes (T1D) of ≤1 year in duration, glycosylated hemoglobin level of ≤10.0%, and use of an insulin pump. The Institutional Review Boards at each center and a Data and Safety Monitoring Board approved the study protocol, consent form, and assent form. A parent or guardian and each subject 7 years and older gave written consent and assent, respectively.
Subjects initially used a Guardian RT for approximately 1 week at home. Approximately half of the subjects were asked to insert a new sensor 4 days before the CRC admission, and the other half 2 days before the admission. Glucose levels were checked at least four times per day with a new One Touch® Ultra® meter (LifeScan, Milpitas, CA) that was provided by the study.
Each subject was then hospitalized overnight for about 18h in a CRC. For subjects of sufficient size to accommodate another device, a second Guardian RT sensor was inserted. During insulin-induced hypoglycemia in which insulin was infused subcutaneously from the subject's insulin pump, the glucose level was checked with the Ultra meter using venous blood every 15min until the glucose level reached 100mg/dL and then every 5–10min. The test ended when the glucose level was <60mg/dL, and then the subject was treated with intravenous glucose. Blood samples were collected at the beginning of the test, when the meter glucose level was <90, <80, < 70, and <60mg/dL, and 15min following treatment. For subjects of sufficient weight to accommodate the volume of blood required, venous blood samples for laboratory serum glucose concentration determinations were taken every 30min during the hospitalization and every 15min for 2h after dinner. Serum glucose concentrations from these samples were measured at the DirecNet Central Biochemistry Laboratory at the University of Minnesota using a hexokinase enzymatic method.1,2
Accuracy analyses were performed separately for inpatient and outpatient Guardian RT use. Laboratory serum glucose values were used as the reference during the inpatient CRC visit, and Ultra meter glucose measurements were used as the reference during home use. Suspected outliers in the meter values where the rate of change was >5mg/dL/min or where a repeat test was done within 10min were excluded from analysis. In a prior study, we found that the Ultra meter had a high degree of accuracy.3 Each reference glucose value was paired to the closest Guardian RT reading within 2.5min. The following were computed for each Guardian–reference pair: difference (Guardian RT value minus reference value), absolute difference (absolute value of difference; AD), and relative AD (AD divided by reference value, expressed as a percentage; RAD). The difference measure incorporates the direction of the error so that pairs with the sensor reading high cancel out pairs with the sensor reading low. The median difference therefore evaluates whether there is any bias for the sensor to read systematically high or low. The AD and RAD values use the absolute value of the difference between the sensor value and the reference value, ignoring the direction of the error. These measures reflect the magnitude of the error without regard to whether the sensor value was higher or lower than the reference value. Each pair was also evaluated to determine whether the sensor value met the International Organization for Standardization (ISO) criteria for home glucose meters (for reference glucose value ≤75mg/dL, meter value within 15mg/dL; for reference glucose value >75mg/dL, meter value within 20%; hereafter referred to as the “ISO criteria”).4 Summary statistics (e.g., median and percentages) were calculated by pooling all paired values. Median values were reported instead of means because of the skewed distribution. The bootstrap technique (resampling subjects with replacement)5 was used to account for the within-subject correlation in the statistical comparisons and calculation of confidence intervals.
The glucose excursions during the insulin-induced hypoglycemia test (drop from baseline to nadir) for the laboratory reference values and the Guardian RT were compared. The Guardian RT nadir glucose was the extrapolated nadir in 20min following treatment and was calculated using 15-min rate of change of the Guardian RT values. The rate of change for laboratory reference values was defined from the blood sample at the meter glucose level <90mg/dL to the first time the laboratory glucose value dropped below 70mg/dL (approximately 90mg/dL to <70mg/dL) and was compared with the rate of change of Guardian RT values in the same period.
In the post-dinner glucose excursion (rise from baseline to peak), the Guardian RT peak glucose was defined as the maximum glucose value from baseline until 30min following the laboratory peak (to allow for a possible lag). The rate of change was defined from baseline until the peak.
For subjects using two Guardian RT sensors simultaneously in the CRC, precision was evaluated by pairing each Guardian RT sensor value with the closest value that was within±2.5min and computing the RAD.
The study included 30 subjects: 15 subjects between 4 and 7 years old and 15 subjects between 12 and 17 years old. Thirty-three percent were female; 93% were Caucasian, 3% Hispanic, and 3% with more than one race. The mean duration of diabetes was 5.0±2.9 years, and the mean glycosylated hemoglobin level was 7.7±0.8%.
The Guardian RT download during CRC admission was unavailable for one subject. Of the remaining 29 subjects, 25 inserted a Guardian RT sensor at home prior to CRC admission (5 days prior for one subject, 4 days prior for eight, 3 days prior for three, 2 days prior for six, and 1 day for seven). During the CRC admission, 19 subjects simultaneously wore two Guardian RT sensors (one inserted at home and one inserted at admission for 16 subjects and both inserted at admission for three), and 10 wore one (inserted at home for nine subjects and at admission for one).
There were 1,511 laboratory glucose measurements paired with glucose measurements from 48 Guardian RT sensors. The median number of paired values per sensor was 37 (interquartile range 32–38), ranging from five to 41. As seen in Table 1, there was no significant tendency for the Guardian RT to read systematically higher or lower than the reference glucose (median difference=−1mg/dL; 95% confidence interval −9 to +7). Overall, the median AD was 21mg/dL, and the median RAD was 14%, with 64% of sensor values meeting ISO home glucose meter criteria (Table 1). The median AD was 27mg/dL for reference glucose values ≤60mg/dL and 25mg/dL for reference glucose values ≤70mg/dL. The median RAD was 19% for reference glucose values 71–120mg/dL, 14% for reference glucose values 121–180mg/dL, and 10% for reference glucose values >180mg/dL. Accuracy measures did not improve substantially when incorporating a 10–30-min sensor lag. Among the 40 sensors with at least 10 Guardian RT–reference pairs, 11 (28%) had a median RAD ≤10%, 11 (28%) a median RAD of 10.1–15%, seven (18%) a median RAD of 15.1–20%, and 11 (28%) a median RAD of >20%.
After adjustment for glucose level, accuracy did not substantially vary by sensor location (buttock/hip vs. abdomen), sensor age, sex, age, body mass index percentile, and time of day (day versus night).
Prior to starting the insulin-induced hypoglycemia test, the median Guardian RT glucose concentration was 121mg/dL (interquartile range 106–132mg/dL), and median reference glucose concentration was 111mg/dL (interquartile range 105–123mg/dL). The median fall in reference glucose was 51mg/dL (interquartile range 44–62mg/dL), and the median AD in this fall between the Guardian RT and reference was 12mg/dL (interquartile range 5–22mg/dL). The ability of the Guardian RT to track changes in drops in glucose compared to the laboratory glucose is displayed in Figure 1. The rate of change in glucose as it dropped from approximately 90mg/dL to <70mg/dL is displayed in Figure 2. The reference glucose fell to ≤60mg/dL during the test for 16 subjects who were wearing 27 sensors; in these 27 sensors, sensor glucose fell to ≤60mg/dL with seven sensors, between 61 and 70mg/dL with nine sensors, and between 71 and 80mg/dL with five sensors, and sensor nadir ranged from 84 to 102mg/dL in the other six sensors.
Prior to starting the dinner meal test, the median Guardian RT glucose concentration was 146mg/dL (interquartile range 116–206mg/dL), and median reference glucose concentration was 156mg/dL (interquartile range 126–219mg/dL). The median rise in reference glucose was 26mg/dL (interquartile range 11–100mg/dL), and the median AD in this rise between the Guardian RT and reference was 19mg/dL (interquartile range 10–32mg/dL). The rise in glucose after a meal was compared between the laboratory glucose and the Guardian RT (Fig. 3). The rate of change in most cases was fairly similar between the sensor and reference glucose measurements (Fig. 4). The median time to the peak was 43min for the Guardian RT and 38min for the reference. The reference glucose rose to ≥200mg/dL during the test for 10 subjects who were wearing 18 sensors, while in 17 of these sensors the glucose values rose to ≥200mg/dL.
At home, there were 661 Guardian RT–Ultra meter pairs from 26 subjects (Guardian RT downloads from home use were unavailable for two subjects, and the ultra meter download was unavailable for two subjects). The median number of pairs per subject was 23 (interquartile range 16–37). The median RAD was 17%, with 56% of sensor values meeting ISO criteria (Table 1). Among the 25 subjects with at least 10 Guardian RT–reference paired values, one (4%) had a median RAD ≤10%, seven (28%) a median RAD of 10.1–15%, five (20%) a median RAD of 15.1–20%, and 12 (48%) a median RAD of >20%.
During the CRC admission, 19 subjects simultaneously used two Guardian RT sensors, resulting in 3,790 Guardian RT–Guardian RT pairs. The median RAD between two simultaneous (within ±2.5min) Guardian RT measurements was 9% (interquartile range 4–19%), median AD was 19mg/dL for values ≤70mg/dL (average of the two Guardian RT values), and median RAD was 9% for values 71–180mg/dL and 7% for values >180mg/dL.
The overall RAD between Guardian RT measurements of interstitial glucose concentrations and reference serum glucose levels was 14% during inpatient and 17% during outpatient assessments, with 64% and 56% meeting ISO criteria, respectively. As will always be the case in accuracy analyses such as this, the RAD was greater at lower glucose levels. Prior studies in adults evaluating the accuracy of the Guardian RT reported a median RAD of 17% during outpatient use.6 Piper et al.7 reported a mean RAD of 18% compared to a laboratory reference in 20 children <3 years old undergoing cardiac bypass surgery. In this study the overall median RAD was similar for the outpatient portion of the study at 17% and was 14% for the inpatient portion. Thus the Guardian RT appears to perform as well in children with T1D as it has been reported to perform in adults with diabetes. The Guardian RT has a similar accuracy to that of other available continuous glucose monitors. Weinstein et al.8 demonstrated that comparison of the FreeStyle® Navigator® Continuous Glucose Monitoring System (Abbott, Abbbott Park, IL) measurements with a laboratory reference method demonstrated mean and median RAD of 12.8% and 9.3%, respectively. The DirecNet Study Group has also studied the FreeStyle Navigator in a pediatric population and found the median AD and RAD were 17mg/dL and 12%, respectively, for inpatient sensor–reference pairs and 20mg/dL and 14%, respectively, for outpatient pairs.9
The Guardian RT system tracked the drop in blood glucose induced by insulin reasonably well in most cases. However, sensor interstitial fluid glucose levels tended to lag behind the blood glucose levels, causing the device to underestimate the true rate of fall in glucose insulin-induced hypoglycemia in some subjects. Similarly, the Guardian RT tracked the magnitude and rate of rise in plasma glucose following the evening test meal fairly well, although the peak postprandial value tended to be delayed in comparison to reference plasma glucose concentrations.
The results of this study of Guardian RT performance will also likely apply to the Paradigm system since both devices use the same sensor. While the accuracy of this sensor system does not yet approach that of the current generation of home glucose meters, it is sufficient to believe that the device has the potential to be an important adjunct to treatment of youth with T1D. Important and useful clinical information can also be gained from the change in glucose with continuous glucose monitors. Predictions for high and low glucose can be determined even if the RAD of a single glucose is not as accurate as a home glucose meter. Further clinical trials are needed, however, to truly demonstrate the clinical utility of the Guardian RT as well as other glucose sensors.
Michael Tansey, M.D.; William Tamborlane, M.D.; Craig Kollman, Ph.D.; Larry Fox, M.D.; Stuart Weinzimer, M.D.; H. Peter Chase, M.D.; Dongyuan Xing, M.P.H.; Bruce Buckingham, M.D.; Roy Beck, M.D., Ph.D.; Katrina Ruedy, M.S.P.H.; and the Diabetes Research in Children Network (DirecNet) Study Group.
Clinical Centers. listed in alphabetical order with clinical center name, city, and state. Personnel are listed as (PI) for Principal Investigator, (I) for co-Investigator, and (C) for Coordinators:
Coordinating Center. Jaeb Center for Health Research, Tampa, FL: Roy W. Beck, M.D., Ph.D.; Katrina J. Ruedy, M.S.P.H.; Craig Kollman, Ph.D.; Dongyuan Xing, M.P.H.; Judy Jackson
University of Minnesota Central Laboratory. Michael W. Steffes, M.D., Ph.D.; Jean M. Bucksa, C.L.S.; Maren L. Nowicki, C.L.S.; Carol A. Van Hale, C.L.S.; Vicky Makky, C.L.S.
National Institutes of Health. Gilman D. Grave, M.D.; Mary Horlick, Ph.D.; Karen Teff, Ph.D.; Karen K. Winer, M.D.
Data and Safety Monitoring Board. Dorothy M. Becker, M.B.B.Ch.; Patricia Cleary, M.S.; Christopher M. Ryan, Ph.D.; Neil H. White, M.D., C.D.E.; Perrin C. White, M.D.
A listing of the DirecNet Study Group appears in the Appendix.
Appreciation is expressed for the work performed by the CRC Nurses at the five clinical centers. This research was supported by the following National Institutes of Health/National Institute of Child Health and Human Development grants: HD041919-01, HD041915-01, HD041890, HD041918-01, HD041908-01, and HD041906-01. Clinical Centers also received funding through the following GCRC grants: M01 RR00069, RR00059, RR 06022, and RR00070-41.