Accuracy and precision of the certified Glucometer-Strips Systems Galileo, Calla and Contour Plus One in clinical practice

doi:10.21203/rs.3.rs-4711593/v1

Download PDF

registered-report

Accuracy and precision of the certified Glucometer-Strips Systems Galileo, Calla and Contour Plus One in clinical practice

https://doi.org/10.21203/rs.3.rs-4711593/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background The purpose of this study was to revise the accuracy and precision of three glucometer-strips systems (GSS) in professional routine.

Methods Prospective single-centre study. Triplets of cPG were investigated on 3 GSS: 1. Galileo, 2. Calla, both employing glucoseoxidase strips (Wellion, Austria), 3. Contour Plus (Ascensia, Switzerland) employing glucosedehydrogenase strips. In each GSS three devices and one LOT of strips were employed. Following the cPG readings, venous blood was taken to measure venous plasma glucose (vPG) on Cobas Integra 400 analyzer employing hexokinase. A total of 258 pairs of cPG vs vPG were compared using Wilcoxon signed rank test with Bonferroni correction and software IBM SPSS Statistics for Windows, V. 23.0. Precision was estimated using SD of average differences between individual cPG measurements. P<0.05 was considered significant.

Results Deviations (mean±SD) from vPG≥5.55 mmo/L (100 mg/dL): Galileo 0.25±0.49 mmol/L (5±9 mg/dL), Calla -1.25±1.08 mmol/L (-23±19 mg/dL), Contour -0.63±0.71 mmol/L (-11±13 mg/dL). Number of cPG values within interval ±15 % of vPG: Galileo 196/207 (94.7 %), Calla 162/207 (78.3 %), Contour 200/207 (96.6%). For vPG values <5.55 mmol/L (100 mg/dL) the interval±0.83 mmol/L (15 mg/dL) reached: Galileo 11/51 (21.6 %), Calla 48/51 (94.1 %), Contour 51/51 (100 %). Galileo cPG <7.0 mmol/L (126 mg/dL) after subtraction of 0.7 mmol/L (12,4 mg/dL) reached 51/51 (100 %).

Conclusions Borderline accuracy of Galileo was found for cPG≥7 mmol/L (126 mg/dL) whereas for cPG<7 mmol/L (126 mg/dL) correction –0.7 mmol/L (13 mg/dL) was necessary. Accuracy of Contour but not of Calla was confirmed.

Accuracy

Cobas Integra analyzer

Glucosedehydrogenase strips

Glucoseoxidase strips

Precision

YSI 2300Stat Plus analyzer

Self-monitoring of glucose concentration in capillary blood was introduced into practice in the seventies of the last century. Today it belongs to the basic prerequisites for effective metabolic control in persons with type 1 and type 2 (pre)diabetes. In addition, HbA1c [1], Continuous Glucose Monitoring (CGM) [2–7], Flash Glucose Monitoring (FGM) [8] and Time in Range (TIR) [9–11] are modern attractive means of metabolic control. Nevertheless, glucometer-strip systems (GSS) remain the most widely used monitoring tools in diabetes care.

In the course of three decades, we tested the GSS Card (Medisense) [12], Optium (Abbott) [12, 13], Advance (Hypoguard) [12, 13], Linus (Agamatrix) [14, 15], Contourlink (Bayer) [15] and Calla (Wellion, MedTrust) [15].

In the year 2019, the perfect accuracy and precision GSS Galileo Glu/Ket (Wellion, Medtrust) was clearly demonstrated by comparing the readings from its display with the respective values from the laboratory analyzer YSI 2300StatPlus (YSI Ltd., Yellow Springs, UT) using glucoseoxidase [16]. The purpose of the present study was to compare the capillary plasma glucose (cPG) readings from the GSS Galileo, Calla and Contour Plus One with venous plasma glucose concentrations (vPG) measured on the analyzer Cobas Integra 400 employing the hexokinase method in order to revise the accuracy and precision of these GSS operated by health care professionals in real clinical practice.

Study design

This was a prospective single centre open label experimental and clinical study. In a period of 6 months (20th September 2021 to 20th March 2022) 210 triplets of capillary blood were investigated on 3 GSS each. In a respective GSS, three (previously non-employed) glucometers and one LOT of respective strips were used over the 6-month study period [Supplementary material SM 1]:

Galileo Glu/Ket (JJ023987, JJ024014, JJ023984, CE 0197, Wellion, MedTrust, Austria) employing glucoseoxidase-based sensors in disposable blood glucose test strips, LOT J1S2004028, Exp. date 10/22. No coding. Measuring range 1.1–33.3 mmol/L (20–600 mg/dL).

Calla Light (DC198190, DC198183, DC197277, CE 0197, Wellion, MedTrust, Austria) employing glucoseoxidase-based sensors in disposable blood glucose test strips, LOT TKS2012012, Exp. date 15.12.2022. No coding. Measuring range 1.1–33.3 mmol/L (20–600 mg/dL).

Contour Plus One (SN6241926, SN6241930, SN6242113, CE 2797, Ascensia Diabetes Care Holding, AG, Switzerland) employing glucosedehydrogenase-based sensors in disposable blood glucose test strips, LOT DPOLQHH03A, Exp. Date 11/22. No coding. Measuring range 0.7–33.3 mmol/L (13–600 mg/dL).

Fingerpricks were performed after routine swab skin disinfection using Softasept N, B.Braun – ethanol + propane-2-ol solution. In an individual investigation adequate amount of blood from one fingerprick was transferred on all respective strips in nine glucometres. The time from the first till the last of nine capillary blood readings did not exceed 1 min. [SM 2]. The individual cPG readings were manually transferred to an MS Excel file.

Following the cPG readings the venous blood from the other upper limb was transferred to a Grainer sample tube for vPG analysis and into two other tubes for investigation of additional routine parameters. Venous blood collection was completed in the course of 2 min and this was immediately followed by 5 -minute centrifugation (3 000 rpm, centrifuge 5702 Eppendorf).

Three experienced health care professionals and one undergraduate student were in charge of all the scheduled procedures which were carried out between 6:00–8:00 a.m. Analyses including 3 subsequent vPG estimations (from one sample of plasma) were carried out within 4 h after blood collection on the Cobas Integra 400 analyzer employing the accredited hexokinase method. The room temperature in the research area was always 23 ± 5°C.

Subjects

Blood was taken from out-patients attending a routine diabetes examination (n = 162). Informed consent was obtained from all of them. There weren´t concomitant diseases nor relevant medication influencing the function of GSS. The exclusion criteria were identified according to international standards ISO 15197 [17] and FDA [18], namely, packed cell volume < 35%, incomplete data sets, the respective bin had already been filled by the required number of triplets (Table 1). Such as a total of 86 out of 210 triplets were included in the statistical evaluation (Table 2) leaving the bins 1, 2 and 6 incomplete due to missing probands with bin-required vPG concentrations.

Table 1

Amount of required-, collected- and assessed- capillary blood samples/triplets in respective bins (see ISO 15197 for details)
Bin No	Intervals of vPG concentration in bin	By ISO 15197 required percentage of samples in bin	No of collected triplets	No of assessed persons (triplets) in bin	No of assessed cPG samples (= triplets*3)	Real percentage of assessed cPG samples in bins
1	<= 2.77 mmol/L <= 50 mg/dL	5%	0	0	0	0
2	> 2.77–4.44 mmol/L > 50–80 mg/dL	15%	10	10	30	12%
3	> 4.44-6,66 mmol/L > 80–120 mg/dL	20%	77	20	60	23%
4	> 6.66-11,10 mmol/L > 120–200 mg/dL	30%	88	30	90	35%
5	> 11.10-16.65 mmol/L > 200–300 mg/dL	15%	24	15	45	17%
6	> 16.65–22.20 mmol/L > 300–400 mg/dL	10%	6	6	18	7%
7	> 22.20 mmol/L > 400 mg/dL	5%	5/5	5	15	6%
N	Total	100%	210	86*	258 = 100%	100%
n	Cobas Integra 400 PG ≥ 5.55 mmol/L PG ≥ 100 mg/dL				207
n₁	Cobas Integra 400 PG < 5.55 mmol/L PG < 5.55 100 mg/dL				51
• ^{See Table 2 for PWD characteristics}

Statistical analysis

Software IBM SPSS Statistics for Windows, V. 23.0 (IBM Corp., Armonk, NY, USA) was used. The selected 86 cPG triplets containing 3x86 = 258 individual cPG values from each GSS were categorized into seven bins, compared with respective means of 3 vPG values from COBAS Integra 400 using the Wilcoxon signed rank test with Bonferroni correction, plotted in the Error Grid diagram [19] and assessed separately for the group with vPG ≥ 5.55 mmol/L (100 mg/dL) (n = 207) and for the group with vPG < 5.55 mmol/L (100 mg/dL) (n₁ = 51) as required by ISO 15197 [17]. When the accuracy of a tested GSS did not reach the criteria recommended by ISO 15197, an adequeate correction factor was sought for. A complex assessment of the accuracy of GSS values in both groups considered the respective evaluation criteria for each group. Intraclass Correlation Coeficient (ICC) was calculated for all vPG values (N = 258) and separately vPG values ≥ 5.55 mmol/L (n = 207). The precision of the individual GSS was estimated at the outset using the SD of the differences between individual cPG measurements in the first 107 triplets. P < 0.05 was considered significant.

Table 2

Clinical and fasting laboratory characteristics in PWD (N = 86) included into statistical analysis
Parameter	Mean	SD	Median	Minimum	Maximum
Age (years)	63.2	13.1	65.0	32.0	90.0
BP syst (mmHg)	144	20	147	96	198
BP diast (mmHg)	86	10	86	49	106
Heart rate (beats/min)	82.9	12.6	84.0	52	118
Body mass (kg)	86.5	15.9	85.7	60	125
Height (cm)	167	10	168	147	192
BMI (kg/m²)	30.8	4.8	30.7	22.3	45.1
HbA1c (mmol/mol)	60	26	49	37	149
HbA1c (% NGSP)	7.6	4.5	6.6	5.5	15.8
vP glucose (mmol/L) vP glucose (mg/dL)	10.0 180	6.3 113	7.6 137	3.2 58	30.4 547
Hb (g/L)	143	13.5	144	112	173
Packed cell volume (PCV) (%)	43.2	3.5	42.5	35.8	51.6
eGF (ml/s)	1.449	0.333	1.448	0.364	1.983

Accuracy of cPG readings of individual GSS for Cobas Integra values ≥ 5.55 mmol/L (100 mg/dL) (n = 207)

Absolute and relative cPG deviations (n = 207) of GSS Galileo, Calla and Contour Plus One from mean vPG Cobas ≥ 5.55 mmol/L (100 mg/dL) see Fig. 1 – Fig. 4 and [Supplementary material SM 3]

Relation between cPG of individual GSS and mean vPG of Cobas Integra 400 are shown in Fig. 2 (Galileo), Fig. 3 (Calla), and Fig. 4 (Contour Plus One). Deviations are assessed for vPG ≥ 5.55 mmol/L (100 mg/dL).

• Accuracy of cPG readings of individual GSS for Cobas Integra < 5.55 mmol/L (100 mg/dL)

Accuracy of GSS for mean vPG < 5.55 mmol/L (100mg/dL) is acceptable if at least 95% of deviations are within the area ± 0.83 mmol/L (15 mg/dL). Apropos the primary results (Table 3) a correction of cPG Galileo and of cPG Calla was necessary (Table 3, Fig. 5, Fig. 6).

Table 3

Accuracy of cPG on GSS Galileo, Galileo corrected, Calla, Calla corrected and Contour Plus One in comparison to mean vPG on Cobas Integra. Assessed for mean vPG < 5.55 mmol/L (100 mg/dL) (n = 51). The acceptable deviation area ± 0.83 mmol/L (15 mg/dL) is highlighted
Deviation area from mean vPG	No of cPG within deviation area ± 0.28 mmol/L (± mg/dL)	No of cPG within deviation area ± 0.56 mmol/L (± 10 mg/dL)	No of cPG within deviation area ± 0.83 mmol/L (± 15 mg/dL)
Galileo (no correction)	1/51 (2.0%)	5/51 (9.8%)	11/51 (21.6%)
Galileo cPG < 7.0 mmol/L (126 mg/dL) reduced by 0.7 mmol/L	21/51 (41.2%)	42/51 (82.4%)	51/51 (100%)
Calla (no correction)	27/51 (52.9%)	42/51 (82.4%)	48/51 (94.1%)
Calla cPG > 5,55 mmol/L increased by 1.0 mmol/L	25/51 (49.0%)	40/51 (78.4%)	46/51 (90.2%)
Calla cPG > 7.0 mmol/L (126 mg/dL) increased by 1.0 mmol/L	27/51 (52.9%)	42/51 (82.4%)	48/51 (94.1%)
Calla cPG 5.5–9.9 mmol/L increased by 0.5 mmol/L, cPG ≥ 10.0 mmol/L increased by 2 mmol/L	24/51 (47.1%)	40/51 (78.4%)	48/51 (94.1%)
Contour Plus One	25/51 (49.0%)	43/51 (84.3%)	51/51 (100%)

Given the acceptable area of deviations and the real distribution of cPG Galileo < 7.0 mmol/L (123 mg/dL) (Table 3, Fig. 5) we found the correction factor − 0.7 mmol/L (-12,6 mg/dL) to be an adequate mean of “calibration” bringing the Galileo cPG < 7.0 mmol/L (126 mg/dL) readings into the acceptable area of deviations from respective vPG Cobas Integra values (Table 3, Fig. 6).

System accuracy of cPG readings of individual GSS for the entire spectrum of Cobas Integra values (N = 258)

Table 4

Complex assessment of the accuracy of cPG on GSS Galileo, Galileo corrected, Calla, Calla corrrected and Contour Plus in comparison for mean vPG on Cobas Integra 400.over the entire range of measurments (N = 258)
Acceptable deviation area	No of cPG within the both acceptable deviation areas
Galileo (no correction)	207/258 (80.2%)
Galileo cPG < 7.0 mmol/L (126 mg/dL) corrected/reduced by 0.7 mmol/L (12,6 mg/dL)	245/258 (95.0%)
Calla (no correction)	210/258 (81.4%)
Calla cPG > 7.0 mmol/L (126 mg/dL) corrected/increased by 1.0 mmol/L (18 mg/dL)	231/258 (89.5%)
Calla cPG 5.5–9.9 mmol/l (= 100–180 mg/dL) corrected/increased by 0.5 mmol/L (= 9 mg/dL), cPG ≥ 10.0 mmol/L (180 mg/dL) corrected/increased by 2 mmol/L (36 mg/dL)	241/258 (93.4%)
Contour Plus (no correction)	251/258 (97.3%)

• Intraclass Correlation Coefficient (ICC)

The Intraclass Correlation Coefficient (ICC) demonstrated a high correlation (> 0.98) between the cPG-values of GSS and vPG values of Cobas and also between the cPG values of individual GSS.[SM 5]

• Precision of GSS Galileo, Calla, Contour Plus One and analyzer Cobas Integra 400

The precision of GSS and of the analyzer Cobas Integra 400 was assessed using SD of triplet means in a group of the first 107 tested PWD. (Table 5), [SM 6]

Table 5

Characteristics of standard deviations (SD in mmol/L) of three cPG measurements on each GSS and on analyzer COBAS INTEGRA (n = 107). NA – non applicable
	Mean	SD	Median	Minimum	Maximum	P (GSS vs Cobas)	P (GSS vs Contour
sd Galileo	0.308	0.252	0.252	0.000	1.343	< 0.0001	< 0.0001
sd Calla	0.294	0.326	0.208	0.000	2.800	< 0.0001	< 0.0001
sd Contour	0.204	0.247	0.153	0.000	2.136	< 0.0001	NA
sd COBAS	0.046	0.035	0,035	0,006	0,185	NA	< 0.0001

The purpose of this trial was to compare the results of our measurments on three GSS performed under the routine conditions of a diabetes practice with the a priori accuracy of GSS estimated using certified methods. Special attention was paid to the recent GSS Galileo Glu/Ket as we are going to include betahydroxybutyrate in the monitored parameters in selected PWD [20].

ISO 15197 [17] defines for GSS the following accuracy criteria: Ninety-five percent (95%) of the individual glucose results will fall within ± 0,83mmol/L (15mg/dL) of the Reference at glucose concentration less than 5.55 mmol/L (100 mg/dL) and within ± 15% at glucose concentration equal to or more than 5.55 mmol/L (100 mg/dL). Evaluation of the system accuracy will be performed with fresh capillary blood samples of with sufficient volume for the blood glucose monitoring system measurements and reference measurements specified in the study protocol. At least 100 pairs of glucose concentrations (GSS value vs Reference value) should be estimated whereas the requested distribution of samples with specified glucose concentrations should be considered (Table 1). Laboratory adjustment for low and high glucose concentrations is possible. The procedures should be repeated for the other 2 reagents‘ LOTs.

Our trial revealed significant downward deviations of cPG Calla readings exceeding the 5,55 mmol/L (100 mg/dL) (Fig. 3) and upward deviations of cPG Galileo readings below 7,0 mmol/L (126 mg/dL) compared to reference vPG on laboratory analyzer COBAS Integra 400 (Table 3, Table 4, Fig. 5). We aimed to define and assesse a factor to correct the Galileo cPG values (< 7.0 mmol/L = 126 mg/dL) into the acceptable area. (Table 3, Table 4, Fig. 6) Nevertheless, we were unable to find a correction factor for the cPG values of GSS Calla. (Table 3, Table 4, SM4) The primary cause of these deviations remains to be clarified.

There are concerns about the design of our trial and the accuracy outcomes of previous studies.

Our present trial on Galileo, Calla and Controur Plus One 2021–2022

We compared readings of the cPG from GSS Galileo Glu/Ket, Calla Light and Contour Plus One displays with a mean value of three vPG measurements.
Diabetes-related diagnostic and therapeutic recommendations in the Czech Republic are based on the glucose concentration in venous plasma which is assumed to be strongly related to capillary plasma. We applied vPG as a reference parameter.
GSS Galileo and GSS Calla employ glucoseoxidase-based sensors; our analyzer Cobas Integra 400 estimates vPG employing the spectrophotometric hexokinase-based accredited method and its accuracy is regularly re-tested. [SM 7] Capillary and venous samples were acquired by health care professionals.
The weak points of this study are related to two deviations from the ISO 15197/2013: (1) the number of required vPG concentrations less than 2,77 mmol/L (50 mg/dL) (bin 1), and of vPG values > 2,77 − 4,44 mmol/L (50–80 mg/dL) (bin 2) and of vPG values > 16,65–22,20 mmol/L (300–400 mg/dL) (bin 6) were not reached (Table 1). As this study aimed to reflect the real clinical practice, we did not carry out any laboratory adjustment for low and high cPG. (2) In each GSS we tested only one LOT of strips (instead of three LOTs) which was available at the trial period.

To support the reliability of our conclusions, in 6 tested persons we repeated the vPG estimation 24 h after the blood collection and centrifugation. In between, the Grainer tube was kept in the refrigerator at 5°C. Even though the reduction of vPG after 24 h was only slight (0.13–1.04 mmol/L (2–20 mg/dL), average 0.33 mmol/L (6 mg/dL), it is worth mentioning in relation to the comparison of immediate cPG readings on individual GSS and up to 4 h delayed vPG readings on Cobas. [SM 8] Whether this finding could be a factor influencing the deviation of readings on the Galileo from Cobas, is open to question. A decline in the accuracy of GSS Calla over the last 8 years is also suspected. [15], [SM 9], [SM 10]

Previous clinical and laboratory trial on GSS Galileo by Weissenbacher et al 2019 [16]

The system accuracy evaluation was performed with three glucose test strip LOTs, six devices and with 100 samples that were distributed over the blood glucose measurement range with a distribution of glucose values as recommended by the ISO standard. The extremely low and high values were obtained by laboratory manipulations. Capillary samples were acquired by a health care professional.
The YSI 2300Stat Plus device (YSI Ltd., Yellow Springs, UT) served for the glucose test strip comparison.
A total of 600 data-pairs distributed according to ISO guidelines over the measurement range (1.89–27.22 mmol/L = 34–500 mg/dL) were obtained.
The ISO acceptance criteria were met by 100% (186/186) of the readings < 5.55 mmol/L (100 mg/dL) and also by 100% of the readings (414/414) with glucose values ≥ 5.55 mmol/L (100 mg/dL).

Previous trials on GSS Contour Plus

Bailey et al [21] in 2017 found high accuracy of Contour Plus One: 99.2% (133/134) of subject-obtained capillary fingertip results and 100% (132/132) of study staff–obtained venous results met ISO 15197:2013 accuracy criteria. Plasma samples were tested in duplicate on a YSI 2300 STAT Plus™ laboratory glucose analyzer.

Klaff et al [22] in 2020 assessed the accuracy and user performance of a new GSS CONTOUR® Plus Elite which applies the same strips as Contour Plus One. Capillary fingertip blood samples from 100 PWD were tested with test strips from three different LOTs. The results are compatible with the results of our present study: minimum ISO 15197:2013- specified standards for accuracy (± 0,83 mmol/L = 15 mg/dL) at BG < 5.55 mmol/L = 100 mg/dL); ±15% at BG ≥ 5.55 mmol/L = 100 mg/dL) were exceeded. Error grid analysis showed that 100% of results were within zone A.

The following questions arise:

Would different oxygen tension in capillary and venous blood significantly affect cPG measured with glucoseoxidase-based amperometric Galileo teststrips eliminating any interference with the glucosedehydrogenase-based amperometric Contour Plus teststrips? The observations of Tang et al. [23] support this sophisticated explanation by finding that increases in oxygen tension lowered glucose measured with glucoseoxidase-based amperometric test strips and did not significantly affect glucose measured with the glucosedehydrogenase-based amperometric test strips. Nevertheless, we have not measured the oxygen tension in our specimens and this possibility remains speculative.
Would a 2 to 4 h time lag between measurement of cPG using GSS Galileo and measurement of vPG on Cobas Integra 400 result in an observed PG difference? This potential factor appears to be of little importance as the Grainer sample tube is heparinized and contains lithium fluoride, EDTA, citrate and sodium azide; centrifugation was performed within 20 min after blood collection. In addition, should this be really relevant we might expect similar differences in all GSS and over the whole range of PG concentrations. We have found a potentially suspicious marker, namely, a slight reduction in vPG concentration 24 h after the blood collection (Avg 0.33 mmol/L = 6 mg/dL per 24 h, n = 6), which might support this hypothesis [SM 8].
Would potential specific differences between reference methods employed on COBAS Integra 400 analyzer (using hexokinase) in our study and on YSI 2300Stat Plus (using glucoseoxidase) employed by Bailey [21] and Weissenbacher [16] help to solve the problém of different accuracy of GSS Galileo [24]?

In recent years the accuracy of many GSS has been assessed. Outcomes of various studies and metaanalyses are open to question [25, 26]. The requirements for ISO standards are increasing. [27]

Repeated evaluation of certified systems may become mandatory. [28] Karon et al. [29] studied the accuracy of point-of-care glucometers and the impact of sample source on the accuracy of glucometer results. There were no significant differences between median capillary whole blood (8.3 mmol/L = 149 mg/dL) and laboratory plasma (8.4 mmol/L = 151 mg/dL) glucose concentrations. The medians of arterial (8.9 mmol/L = 160 mg/dL) and of venous (9.0 mmol/L = 162 mg/dL) whole blood glucose concentrations were significantly higher than the median of laboratory PG concentration. Adnan et al. [30] showed a strong association between capillary and venous glucose measurements. It may be difficult to assess the effect of factors that can impact the glucose measurement, such as sample type and measuring procedure for a given PG concentration [31]. PG concentration in fasting capillary plasma is found to be 0.1–0.3 mmol/L (2–6 mg/dL) higher than in venous plasma. [32]

The performance of a quantitative monitoring device can be based on analytical accuracy or clinical accuracy. Analytical accuracy quantitatively expresses how close the measurement by the index system is to the same measurement by a reference method. Clinical accuracy is a qualitative measure expressing the clinical outcome resulting in a treatment decision. The error grids for glucose self-measurement devices have five zones—Zone A through Zone E—with A indicating the least risk and E indicating the most risk. [33]

In our study, by subtraction of 0.7 mmol/L (13 mg/dL), the upward deviation of low cPG Galileo (Fig. 5) could be brought into the required “accuracy area” of vPG (Fig. 6) whereas 99% of results reached the zone A/B as it is required by the standard of ISO 15197. Given these observations, people using Galileo are trained in corrections of PG readings and in adequate adaptations of diabetes management. A potential perspective of glucosedehydrogenase method [34], analytical and clinical performance of GSS, education and communication of the data between user and physician should be brought in mind.

In conclusion: (1) Borderline ISO standard requested accuracy of GSS Galileo was found for cPG ≥ 7 mmol/L (126 mg/dL) whereas for cPG < 7 mmol/L (126 mg/dL) a correction factor − 0.7 mmol/L (13 mg/dL) is necessary. The accuracy of GSS Contour Plus One but not of Calla was confirmed. (2) In clinical practice, in addition to sustainable reviews of GSS also data communication between health care professionals and GSS users should be considered.

blood pressure

body mass

BPM

beats per minute

cPG

capillary plasma glucose

eGFR

estimated glucose filtration rate

FDa

Food and Drug Administartion

GSS

glucometer–strips–system

HbA1c

hemoglobin A1c

ISO

International Standards Organization

PCV

packed cells volume

PWD

person/people with diabetes

rpm

rotations per minute

supplementary materiál

vPG

venous plasma glucose

Ethics Approval

The Ethics Committee of the Teaching Hospital Olomouc and Faculty of Medicine and Dentistry, Palacký University Olomouc, Czech Republic approved the work. Reference Number 30/23. Only accredited methods and certified devices, equipment and facilities comprising routine clinical and laboratory investigations following the valid guidelines were employed.

Consent to participate and consent to publish

All participants gave written and verbal voluntary consent to participate as well as leave the project.

Authors' contributions

R.C. suggested the study protocol and wrote the main text and revised the manuscript, H.Z. and B.D and E.D. were responsible for venous blood collections and cPG estimations on GSS Galileo, Calla and Contour Plus One, M.N. supervised the exact timing of the respective procedures, data files and supplementary material, J.G. was in charge of laboratory investigations on analyser Cobas Integra 400, J.Z. carried out the final statistical assessment and prepared Tables and Figures.

Availability of data and materials

The data can be assessed in the study protocols and in patient files in the Institute of Specialized Treatments Paseka. The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Funding

No funds or grants were available for this study.

Competing interests

The authors received financial support from Elekta s.r.o. – a MedTrust Company, for participation at the National Congress of the Czech and Slovak Diabetes Association and at other postgraduate educational activities. Elekta, s.r.o., supplied glucometer strips systems Galileo and Calla; Promedica Praha Group, a.s., supplied glucometer strips systems Contour Plus One.

Acknowledgements

Support provided by the 2nd Dept. of Medicine, Gatroenterology and Geriatrics, Teaching Hospital Olomouc and Dept. of Medical Biophysics, Faculty of Medicine and Dentistry, Palacký University Olomouc and by the Institute of Specialized Treatments Paseka should be acknowledged.

Hanas R, John G on behalf of the International HbA1c Consensus Committee. Consensus statement on the worldwide standardization of the hemoglobin A1C measurement. Diabetes Care 2010;33:1903–4.
Mlčák P, Fialová J, Trnková K, Chlup R. A continuous glucose monitoring system (CGMS) – a promising approach for improving metabolic control in persons with type 1 diabetes mellitus treated by insulin pumps. Biomed Pap 2004;148(1):33–8.
Peterson K, Zapletalova J, Kudlova P, Matuskova V, Bartek J, Novotny D, Chlup R. Benefits of three-month continuous glucose monitoring for persons with diabetes using insulin pumps and sensors. Biomed Pap Med Fac Univ Palacky Olomouc Czech Repub 2009;153(1):47–52.
Cohen O, Körner A, Chlup R, Zoupas C, Ragozin A, Wudi K, Bartášková D, Pappas A, Niederland T, Taybani Z, Barák L, Vazeou AS. Improved glycemic control through continuous glucose sensor–augmented insulin pump therapy: Prospective results from a community and academic practice patient registry. J Diab Sci Techn 2009;3(4):804–11.
Liebl A, Henrichs HR, Heinemann L, Freckmann G, Biermann E, Thomas A, for the Continuous Glucose Monitoring Working Group of the Working Group Diabetes Technology of the German Diabetes Association. Continuous glucose monitoring: evidence and consensus statement for clinical use. J Diabetes Sci Technol 2013;7(2):500–19. www.journalofdst.org
Šoupal J, Petruželková L, Flekač M, Pelcl T, Matoulek M, Daňková M, Škrha J, Svačina Š, Prázný M. Comparison of different treatment modalities for type 1 diabetes, including sensor-augmented insulin regimens, in 52 weeks of follow-up: A COMISAIR Study. Diabetes Technology & Therapeutics 2016;18(9):532–38. http://doi.org/10.1089/dia.2016.0171
Šoupal J, Petruželková L, Grunberger G et al. Glycemic outcomes in adults with T1D are impacted more by continuous glucose monitoring than by insulin delivery method: 3 years of follow-up from the COMISAIR study. Diabetes Care 2020;43: 37–43.
Edelman SV, Argento NB, Pettus J, Hirsch IB. Clinical implications of real-time and intermittently scanned continuous glucose monitoring. Diabetes Care 2018;41(11):2265–74.
Advani A. Positioning time in range in diabetes management. Diabetologia 2020;63:242–52. https://doi.org/10.1007/s00125-019-05027-0
Battelino T, Danne T, Bergenstal RM, et al. Clinical targets for continuous glucose monitoring data interpretation: recommendations from the international consensus on time-in-range. Diabetes Care 2019;42:1593–603.
Gómez AM, Henao C DC, Rebolledo M, et al. Determination of Time in Range Associated With HbA1c ≤ 7% in a Prospective Cohort of Patients With Type 1 Diabetes Using CGM for Three Months. Journal of Diabetes Science and Technology. 2022;0(0). https://doi:10.1177/19322968221108424
Chlup R, Bartek J, Malá E, Doubravová B, Pukowietz L, Zatloukal P, Chlupová L, Zapletalová J. Uživatelská studie o správnost a přesnosti měření glukometrů Advance, Card a Optium. (User-oriented study on the accuracy and precision of the glucometers Advance, Card and Optium. Czech.) Klin Bioch Metabol 2004;12 (3):171–78.
Chlup R, Payne M, Zapletalová J, Doubravová B, Řezníčková M, Chlupová L, Sečkař P. Results of selfmonitoring on glucometer systems Advance and Optium in daily routine. Biomed Papers 2005;149 (1):127–39.
Chlup R, Doubravova B, Peterson K, Zapletalova J, Bartek J. Wavesense technology glucometer Linus for routine self-monitoring and clinical practice. Acta Diabetol 2011;48(1):35–40.
Chlup R, Doubravová B, Bartek J, Zapletalová J, Krystyník O, Procházka V. Effective assessment of diabetes control using personal glucometers (Contourlink, Bayer, Germany; Calla, Wellion, Austria; Linus, Agamatrix, USA). Disease Markers 2013;35(6):895–905.
Weissenbacher S, Chen-Yu Yang, Tang-Chin Kuan, Demircik F, Hanna M, Pfützner A. System accuracy assessments with a blood glucose meter with combined glucose and ß-hydroxybutyrate measurement capabilities, Expert Review of Molecular Diagnostics 20119;19(11):1043-48, https://DOI:10.1080/14737159.2019.1662300
International Organization for Standardization (ISO). ISO 15197:2013(E): in vitro diagnostic test systems—requirements for blood-glucose monitoring systems for self-testing in managing diabetes mellitus. Published May2013. https://www.iso.org/standard/54976.html Last reviewed and confirmed 2018. Geneva, Switzerland: International Organization for Standardization.
Food and Drug Administration (2016) guidance for Industry and Food and Drug Administration staff self-monitoring blood glucose test systems for over-the-counter use. [updated 2016 Oct 11]. Available from: https://www.fda.gov/downloads/ucm380327.pdf
Parkes JL, Slatin SL, Pardo S, et al. A New Consensus Error Grid to Evaluate the Clinical Significance of Inaccuracies in the Measurement of Blood Glucose. Diabetes Care 2000;23:1143–48.
Pfützner A, Klonoff D, Heinemann L, et al. Euglycemic ketosis in patients with type 2 diabetes on SGLT2-inhibitor therapy - an emerging problem and solutions offered by diabetes technology. Endocrine 2017;56:212–16.
Bailey TS, Wallace JF, Pardo S, Warchal-Windham ME, Harrison B, Morin R, Christiansen M. Accuracy and User Performance Evaluation of a New, Wireless-enabled Blood Glucose Monitoring System That Links to a Smart Mobile Device. J Diabetes Sci Technol. 2017;11(4):736–743. https://doi:10.1177/1932296816680829 Epub 2017 Feb 1. PMID: 28617617; PMCID: PMC5588817.
Klaff L, Shelat P, Zondorak D, Wayland-Smith A, Vernes P, Richardson JM (2021) Accuracy and User Performance of a New Blood Glucose Monitoring System. J Diabetes Sci Technol. 15(6):1382–89. https://doi:10.1177/1932296820974348 Epub 2020 Nov 26. PMID: 33243009; PMCID: PMC8655271
Tang Z, Louie RF, Lee JH, Lee DM, Miller EE, Kost GJ (2001) Oxygen effects on glucose meter measurements with glucosedehydrogenase-and oxidase-based test strips for point-of-care testing. Critical Care Medicine 2001;29(5):1062-70.
Han J, Nichols JH, Rice M, Klonoff DC. The End of the Road for the YSI 2300 Analyzer: Where Do We Go Now? J Diabetes Sci Technol. 2020;14(3):595–600. doi: 10.1177/1932296819886603. Epub 2019 Nov 11. PMID: 31711305; PMCID: PMC7576956.
Scott MG, Bruns DE, Boyd JC, Sacks DB. Tight glucose control in the intensive care unit. Are glucose meters up to the task? Clin Chem 2009;55(1):18–20.
Freckmann G, Baumstark A, Jendrike N et al. System accuracy evaluation of 27 blood glucose monitoring systems according to DIN EN ISO 15197. Diabetes Technol Ther 2010;12:221–31.
Freckmann G, Schmid C, Baumstark A, Rutschmann M, Haug C, Heinemann L. Analytical Performance Requirements for Systems for Self-Monitoring of Blood Glucose With Focus on System Accuracy: Relevant Differences Among ISO 15197:2003, ISO 15197:2013, and Current FDA Recommendations. J Diabetes Sci Technol 2015; 9:885–94. https://DOI:10.1177/1932296815580160
King F, Ahn D, Hsiao V, Porco T, Klonoff DC. A Review of Blood Glucose Monitor Accuracy. Diabetes Technol Ther 2018;20:843 – 56 Published Online: 4 Dec 2018 https://doi.org/10.1089/dia.2018.0232
Karon BS, Gandhi GY, Nuttall GA, Bryant SC, Schaff HV, McMahon MM, Santrach, PJ. Accuracy of Roche accu-chek inform whole blood capillary, arterial, and venous glucose values in patients receiving intensive intravenous insulin therapy after cardiac surgery. Am J Clin Pathol 07;127(6):919–26 https://doi:10.1309/6RFQCKAAJGKWB8M4
Adnan M, Imam, F, Rahat T, et al. Correlation between capillary and venous blood glucose levels in diabetic patients. Asian Biomedicin 2015;9(1):50–9.
Blaurock MG, Kallner A, Menzel S, Masuch A, Nauck M, Petersmann A. Impact of Glucose Measuring Systems and Sample Type on Diagnosis Rates of Diabetes Mellitus. Diabetes Ther 2018;9:2029–2041 https://doi.org/10.1007/s13300-018-0495-0
Friedecký B, Kratochvíla J, Springer D, Prázný M, Pelikánová T, Zima T, Racek J for the Czech Society of Clinical Biochemistry and for the Czech Diabetes Society CLS JEP. Diabetes mellitus – laboratorní diagnostika a sledování stavu pacientů. Diabetes mellitus – laboratory diagnostics and observation of patients´condition. Klin Biochem Metab 2019;27(48);1:32–47.
Tian T, Aaron RE, Kohn MA, Klonoff DC. The Need for a Modern Error Grid for Clinical Accuracy of Blood Glucose Monitors and Continuous Glucose Monitors. Journal of Diabetes Science and Technology 2024;18(1):3–9. https://doi:10.1177/19322968231214281
Pfützner A, Kalasauske D, Hanna M, Sachsenheimer D, Raab G, Weissenbacher S, Thomé N. System Accuracy and Interference Evaluation of a New Glucose Dehydrogenase-Based Blood Glucose Meter for Patient Self-Testing. Journal of Diabetes Science and Technology. 2023;0(0). https://doi:10.1177/19322968231201862

Competing interest reported. The authors received financial support from Elekta s.r.o. – a MedTrust Company, for participation at the National Congress of the Czech and Slovak Diabetes Association and at other postgraduate educational activities. Elekta, s.r.o., supplied glucometer strips systems Galileo and Calla; Promedica Praha Group, a.s., supplied glucometer strips systems Contour Plus One.

20240709DiscChlupSM1SM10.docx

Download PDF

Reviews received at journal
28 Jul, 2024
Reviewers agreed at journal
28 Jul, 2024
Reviewers agreed at journal
23 Jul, 2024
Reviewers invited by journal
23 Jul, 2024
Editor assigned by journal
23 Jul, 2024
Submission checks completed at journal
17 Jul, 2024
First submitted to journal
09 Jul, 2024

You are reading this latest preprint version

Accuracy and precision of the certified Glucometer-Strips Systems Galileo, Calla and Contour Plus One in clinical practice

Status:

Version 1

Abstract

Figures

Introduction

Methods and subjects

Study design

Subjects

Statistical analysis

Results

• Accuracy of cPG readings of individual GSS for Cobas Integra < 5.55 mmol/L (100 mg/dL)

• Intraclass Correlation Coefficient (ICC)

• Precision of GSS Galileo, Calla, Contour Plus One and analyzer Cobas Integra 400

Discussion

Our present trial on Galileo, Calla and Controur Plus One 2021–2022

Previous clinical and laboratory trial on GSS Galileo by Weissenbacher et al 2019 [16]

Previous trials on GSS Contour Plus

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1