Performance evaluations of continuous glucose monitoring systems: precision absolute relative deviation is part of the assessment

BACKGROUND

Even though a Clinical and Laboratory Standards Institute proposal exists on the design of studies and performance criteria for continuous glucose monitoring (CGM) systems, it has not yet led to a consistent evaluation of different systems, as no consensus has been reached on the reference method to evaluate them or on acceptance levels. As a consequence, performance assessment of CGM systems tends to be inconclusive, and a comparison of the outcome of different studies is difficult.

MATERIALS AND METHODS

Published information and available data (as presented in this issue of Journal of Diabetes Science and Technology by Freckmann and coauthors) are used to assess the suitability of several frequently used methods [International Organization for Standardization, continuous glucose error grid analysis, mean absolute relative deviation (MARD), precision absolute relative deviation (PARD)] when assessing performance of CGM systems in terms of accuracy and precision.

RESULTS

The combined use of MARD and PARD seems to allow for better characterization of sensor performance. The use of different quantities for calibration and evaluation, e.g., capillary blood using a blood glucose (BG) meter versus venous blood using a laboratory measurement, introduces an additional error source. Using BG values measured in more or less large intervals as the only reference leads to a significant loss of information in comparison with the continuous sensor signal and possibly to an erroneous estimation of sensor performance during swings. Both can be improved using data from two identical CGM sensors worn by the same patient in parallel.

CONCLUSIONS

Evaluation of CGM performance studies should follow an identical study design, including sufficient swings in glycemia. At least a part of the study participants should wear two identical CGM sensors in parallel. All data available should be used for evaluation, both by MARD and PARD, a good PARD value being a precondition to trust a good MARD value. Results should be analyzed and presented separately for clinically different categories, e.g., hypoglycemia, exercise, or night and day.

Evaluation of the performance of a novel system for continuous glucose monitoring

BACKGROUND

The performance of a continuous glucose monitoring (CGM) system in the early stage of development was assessed in an inpatient setting that simulates daily life conditions of people with diabetes. Performance was evaluated at low glycemic, euglycemic, and high glycemic ranges as well as during phases with rapid glucose excursions.

METHODS

Each of the 30 participants with type 1 diabetes (15 female, age 47 ± 12 years, hemoglobin A1c 7.7% ± 1.3%) wore two sensors of the prototype system in parallel for 7 days. Capillary blood samples were measured at least 16 times per day (at least 15 times per daytime and at least once per night). On two subsequent study days, glucose excursions were induced. For performance evaluation, the mean absolute relative difference (MARD) between CGM readings and paired capillary blood glucose readings and precision absolute relative difference (PARD), i.e., differences between paired CGM readings were calculated.

RESULTS

Overall aggregated MARD was 9.2% and overall aggregated PARD was 7.5%. During induced glucose excursions, MARD was 10.9% and PARD was 7.8%. Lowest MARD (8.5%) and lowest PARD (6.4%) were observed in the high glycemic range (euglycemic range, MARD 9.1% and PARD 7.4%; low glycemic range, MARD 12.3% and PARD 12.4%).

CONCLUSIONS

The performance of this prototype CGM system was, particularly in the hypoglycemic range and during phases with rapid glucose fluctuations, better than performance data reported for other commercially available systems. In addition, performance of this prototype sensor was noticeably constant over the whole study period. This prototype system is not yet approved, and performance of this CGM system needs to be further assessed in clinical studies.

Performance evaluation of a continuous glucose monitoring system under conditions similar to daily life

BACKGROUND

This study aimed at evaluating and comparing the performance of a new generation of continuous glucose monitoring (CGM) system versus other CGM systems, under daily lifelike conditions.

METHODS

A total of 10 subjects (7 female) were enrolled in this study. Each subject wore two Dexcom G4™ CGM systems in parallel for the sensor lifetime specified by the manufacturer (7 days) to allow assessment of sensor-to-sensor precision. Capillary blood glucose (BG) measurements were performed at least once per hour during daytime and once at night. Glucose excursions were induced on two occasions. Performance was assessed by calculating the mean absolute relative difference (MARD) between CGM readings and paired capillary BG readings and precision absolute relative difference (PARD), i.e., differences between paired CGM readings.

RESULTS

Overall aggregate MARD was 11.0% (n = 2392). Aggregate MARD for BG <70 mg/dl was 13.7%; for BG between 70 and 180 mg/dl, MARD was 11.4%; and for BG >180 mg/dl, MARD was 8.5%. Aggregate PARD was 7.3%, improving from 11.6% on day 1 to 5.2% on day 7.

CONCLUSIONS

The Dexcom G4 CGM system showed good overall MARD compared with results reported for other commercially available CGM systems. In the hypoglycemic range, where CGM performance is often reported to be low, the Dexcom G4 CGM system achieved better MARD than that reported for other CGM systems in the hypoglycemic range. In the hyperglycemic range, the MARD was comparable to that reported for other CGM systems, whereas during induced glucose excursions, the MARD was similar or slightly worse than that reported for other CGM systems. Overall PARD was 7.3%, improving markedly with sensor life time.

Performance evaluation of three continuous glucose monitoring systems: comparison of six sensors per subject in parallel

BACKGROUND

This study is aimed at comparing the performance of three continuous glucose monitoring (CGM) systems following the Clinical and Laboratory Standards Institute's POCT05-A guideline, which provides recommendations for performance evaluation of CGM systems.

METHODS

A total of 12 subjects with type 1 diabetes were enrolled in this study. Each subject wore six CGM systems in parallel, two sensors of each CGM system [FreeStyle Navigator™ (Navigator), MiniMed Guardian® REAL-Time with Enlite sensor (Guardian), DexCom™ Seven® Plus 3rd generation (Seven Plus)]. Each sensor was used for the lifetime specified by the manufacturer. To follow POCT05-A recommendations, glucose excursions were induced on two separate occasions, and venous and capillary blood glucose (BG) concentrations were obtained every 15 min for five consecutive hours. Capillary BG concentrations were measured at least once per hour during the day and once at night. Parameters investigated were CGM-to-BG differences [mean absolute relative difference (MARD)] and sensor-to-sensor differences [precision absolute relative difference (PARD)].

RESULTS

Compared with capillary BG reference readings, the Navigator showed the lowest MARD, with 12.1% overall and 24.6% in the hypoglycemic range; for the Guardian and the Seven Plus, MARD was 16.2%/34.9% and 16.3%/32.7%, respectively. PARD also was lowest for the Navigator (9.6%/9.8%), followed by the Seven Plus (16.7%/25.5%) and the Guardian (18.1%/20.2%). During induced glucose excursions, MARD between CGM and BG was, again, lowest for the Navigator (14.3%), followed by the Seven Plus (15.8%) and the Guardian (19.2%).

CONCLUSIONS

In this study, two sensors of each of the three CGM systems were compared in a setting following POCT05-A recommendations. The Navigator CGM system achieved more accurate results than the Guardian or the Seven Plus with respect to MARD and PARD. Performance in the hypoglycemic range was markedly worse for all CGM systems when compared with BG results.

Pharmacological profile and toxicity of fluorescein-labelled sinistrin, a novel marker for GFR measurements

There is an evident and growing medical need for an accurate determination of kidney function for a broad spectrum of indications. The glomerular filtration rate (GFR) is the most accepted indicator of renal function. Due to difficulties in performing the test, GFR is currently determined rarely in clinical practice. A procedure for such GFR determination has to be safe, accurate and easy to handle. By using the new compound fluorescein isothiocyanate-sinistrin (FS) these requirements are met. The pharmacological profile and tolerability of FS, selected from among various newly synthesized, labelled compounds intended for use as GFR markers, was characterized in male Sprague-Dawley rats following i.v. application. Using the newly described fluorometric method, FS can be determined much more easily in serum and urine than with the established enzymatic method. After i.v. dosing, FS concentrations in serum declined rapidly in various experimental groups to a comparable extent (t (1/2), mean+/-SD: 22.4+/-8.3 to 26.2+/-5.4 min). Its increase after unilateral nephrectomy reflects the loss of filtration capacity. Comparable concentration-time curves of FS in serum measured fluorometrically and enzymatically suggest no relevant alteration of pharmacokinetic behaviour by the labelling. This notion is supported by the high urinary excretion rate and absence of biliary excretion. The higher sensitivity of the fluorometric method suggests a dose of FS of 100 mg in humans compared with 5 g of sinistrin or inulin. FS was well tolerated after single and multiple applications. On the basis of these results, the kinetics of FS are comparable with the gold standard inulin or sinistrin, but FS is superior in handling. Providing the data can be transferred from rat to human, determination of GFR using the new method should result in an improvement of acceptance by both physicians and patients.

Noninvasive glucose measurement by monitoring of scattering coefficient during oral glucose tolerance tests. Non-Invasive Task Force

BACKGROUND

Continuous glucose monitoring by means of optical glucose sensors would allow patients with diabetes to check their metabolic control to their convenience. In an earlier study, we showed that noninvasive glucose monitoring is feasible for rapid changes in blood glucose by means of measuring the scattering coefficient of human skin. In this study, we investigated whether also slower changes in blood glucose, this time induced by an oral glucose load, can also be monitored by this approach.

METHODS

Five healthy subjects and 13 patients with type 2 diabetes have been given a 75-g oral glucose load. Portable noninvasive systems were used to measure the skin tissue scattering coefficient. For this purpose, two optical sensor heads were attached directly to the skin of each volunteer. Light was applied to the skin and the reflected light intensity was registered.

RESULTS

In 8 of 10 measurements, correlation of changes in scattering coefficient with changes in glycemia was acceptable. In 19 of 26 measurements (73%) of patients with type 2 diabetes the observed changes in the scattering coefficient also correlated in acceptable manner. The accordance between the simultaneous measurements of the two sensor heads was acceptable in 13 of 18 volunteers and patients studied. There were virtually no differences in the quality of the measurements between healthy volunteers and patients with diabetes.

CONCLUSIONS

This study shows that also slow changes in blood glucose induced by an oral glucose load can be monitored by registration of scattering coefficient changes. It remains to be elucidated why this has not been possible in all experiments.

?IR-Spectroscopy? ? computer-program for the interpretation of infrared spectra

A computer-program "IR-Spectroscopy" for the interpretation of infrared spectra of organic compounds is described. It contains the following basic items for preprocessing and execution of the computer based interpretation process: handling of the rule-base CorTab, routines for automatic rule generation and interpretation modules on the basis of the programming languages PROLOG and C.