ASCENSIA CONTOUR BLOOD GLUCOSE METER, MODEL 7151; REAGENT STRIP, MODEL 7080

K062058 · Bayer Healthcare · NBW · Sep 11, 2006 · Clinical Chemistry

Device Facts

• Record ID: K062058
• Device Name: ASCENSIA CONTOUR BLOOD GLUCOSE METER, MODEL 7151; REAGENT STRIP, MODEL 7080
• Applicant: Bayer Healthcare
• Product Code: NBW · Clinical Chemistry
• Decision Date: Sep 11, 2006
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 862.1345
• Device Class: Class 2
• Attributes: Pediatric

Indications for Use

The Shogun Blood Glucose Monitoring System is used for the measurement of glucose in whole blood. The Shogun Blood Glucose Monitoring System is an over-the-counter (OTC) device used by persons with diabetes and by healthcare professionals in home settings and in healthcare facilities. The Shogun Blood Glucose Monitoring System is indicated for use with capillary, venous, and arterial whole blood samples and neonatal blood samples. Capillary samples may be drawn from the fingertip, palm, forearm, and in the case of neonates, the heel. The frequent monitoring of blood glucose is an adjunct to the care of persons with diabetes.

Device Story

System measures glucose in whole blood; includes meter, test strips, control solution, lancing device, and lancets. Used by patients with diabetes or healthcare professionals in home/clinical settings. Provides blood glucose readings to assist in diabetes management. Modifications from predicate include meter design, software, and labeling changes; fundamental scientific technology remains unchanged.

Clinical Evidence

Laboratory and clinical evaluations compared the modified system to the original Ascensia CONTOUR system and a laboratory reference method. Results demonstrated equivalent performance.

Technological Characteristics

Glucose dehydrogenase-based blood glucose monitoring system. No changes to fundamental scientific technology. Includes meter, test strips, control solution, and lancing accessories.

Indications for Use

Indicated for persons with diabetes and healthcare professionals in home/clinical settings for glucose measurement in capillary, venous, arterial, and neonatal whole blood. Capillary sites include fingertip, palm, forearm, and neonatal heel.

Regulatory Classification

Identification

A glucose test system is a device intended to measure glucose quantitatively in blood and other body fluids. Glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus, neonatal hypoglycemia, and idiopathic hypoglycemia, and of pancreatic islet cell carcinoma.

Special Controls

*Classification.* Class II (special controls). The device, when it is solely intended for use as a drink to test glucose tolerance, is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.

Predicate Devices

• Ascensia® CONTOUR® Diabetes Care System (k023657, k060470)

Submission Summary (Full Text)

{0}

1

# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY AND INSTRUMENT COMBINATION TEMPLATE

A. 510(k) Number:
k062058

B. Purpose for Submission:
Change in read time, addition of correcting algorithms

C. Measurand:
Glucose

D. Type of Test:
Whole Blood Glucose Concentration through a Quantitative Amperometric Assay (Glucose Oxidase)

E. Applicant:
Bayer HealthCare, LLC.

F. Proprietary and Established Names:
Ascensia® CONTOUR® Blood Glucose Monitoring System

G. Regulatory Information:
1. Regulation section:
21 CFR § 862.1345, Glucose Test System
2. Classification:
Class II
3. Product code:
NBW, LFR
4. Panel:
75 (Clinical Chemistry)

H. Intended Use:
1. Intended use(s):
See indications for use below.
2. Indication(s) for use:
The Ascensia® CONTOUR® Blood Glucose Monitoring System is used for the measurement of glucose in whole blood. The Ascensia® CONTOUR® Blood Glucose Monitoring System is an over-the-counter (OTC) device used by persons with diabetes and by healthcare professionals in home settings and in healthcare facilities. The

{1}

Ascensia® CONTOUR® Blood Glucose Monitoring System is indicated for use with capillary, venous, and arterial whole blood samples and neonatal blood samples. Capillary samples may be drawn from the fingertip, palm, forearm, and in the case of neonates, the heel. The frequent monitoring of blood glucose is an adjunct to the care of persons with diabetes.

3. Special conditions for use statement(s): For over-the-counter use and by healthcare professionals
4. Special instrument requirements: Ascensia® CONTOUR® Blood Glucose Monitoring System

I. Device Description: The Ascensia® CONTOUR® Blood Glucose Monitoring System is used for the measurement of glucose in whole blood. The system contains a blood glucose meter, a bottle of strips, a bottle of normal control solution, a lancing device and lancets, and instructions for use.

J. Substantial Equivalence Information:

1. Predicate device name(s): Ascensia® CONTOUR® Diabetes Care System
2. Predicate 510(k) number(s): k023657, k060470
3. Comparison with predicate:

| Similarities | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Detection Method | Amperometry | Amperometry |
| Enzyme | Glucose Dehydrogenase (FAD) | Glucose Dehydrogenase (FAD) |
| Sample Volume | 0.6 μL | 0.6 μL |
| Test Range | 10 – 600 mg/dL | 10 – 600 mg/dL |
| Differences | | |
| --- | --- | --- |
| Item | Device | Predicate |
| Test Time | 5 seconds | 15 seconds |
| Tests Stored in Memory | 480 | 240 |
| Correction factors | Added correcting algorithms | None. |

K. Standard/Guidance Document Referenced (if applicable): ISO 15197: In vitro diagnostic test systems - Requirements for blood-glucose monitoring systems for self-testing in managing diabetes mellitus.

{2}

3

L. Test Principle:

Once a whole blood sample is applied to the sample chamber of the test strip, glucose measurement commences. Glucose measurement is based on electrical potential caused by the reaction of glucose with the reagents contained on the strip's electrodes. The current resulting from this enzymatic reaction is measured and converted to glucose concentration by the meter.

M. Performance Characteristics (if/when applicable):

1. Analytical performance:

a. Precision/Reproducibility:

With-in Run Precision:

Human whole blood samples with hematocrit of 35%-50% were collected in tubes with sodium heparin anticoagulant. Using a 10% aqueous glucose solution, blood samples were prepared within the following ranges prescribed in ISO 15197 section 7.2.2.2:

30-50 mg/dL (actual value = 41.3 mg/dL)
51-110 mg/dL (actual value = 99.2 mg/dL)
111-150 mg/dL (actual value = 119.5 mg/dL)
151-250 mg/dL (actual value = 200.0 mg/dL)
251-400 mg/dL (actual value = 325.5 mg/dL)

A total of ten blood glucose meters were used to test one test sensor lot with the above blood samples. One bottle of test sensors was assigned to each instrument at the start of the study. One operator performed all the testing. Ten tests were performed with each blood sample on each instrument. Results are summarized below.

Pooled Statistics for the Shogun System

| Level (mg/dL) | Grand Mean (mg/dL) | Pooled Variance | Pooled %CV |
| --- | --- | --- | --- |
| 40* | 38 | 3.7 | 4.8 |
| 100 | 101 | 26.8 | 5.1 |
| 120 | 118 | 19.4 | 3.7 |
| 200 | 205 | 48.1 | 3.3 |
| 325 | 326 | 139.7 | 3.6 |

*Pooled SD=1.9

Day-to-Day Precision:

One bottle of Ascensia Contour test sensors was assigned to each of ten Ascensia Contour instruments. On each day of ten consecutive working days (no testing on weekends), one test sensor was tested on each instrument with each level of Low, Normal and High Ascensia Microfill control solution. One operator performed all of the testing.

{3}

The following summarizes the specific materials used in the study:

Control

| Level | Lot # | Control Ranges |
| --- | --- | --- |
| Low | 1197 | 32 – 44 mg/dL |
| Normal | V99216 | 99 – 136 mg/dL |
| High | V99207 | 288 – 398 mg/dL |

Results from the precision evaluation are summarized in the table below.

Pooled Statistics for the Shogun System

| Control Solution | N | Grand Mean | Pooled Variance | Pooled %CV |
| --- | --- | --- | --- | --- |
| Low* | 100 | 40 | 0.5 | 1.7 |
| Normal | 100 | 124 | 2.7 | 1.3 |
| High | 100 | 368 | 14.9 | 1.0 |

*Pooled SD=0.7

b. Linearity/assay reportable range:

To establish the linearity of the Contour system throughout the entire reportable range of 10 to 600 mg/dL, data from three studies were combined. In one study, blood with 40% hematocrit was adjusted to plasma glucose concentrations of 10, 20, 30, 40, 50, and 60 mg/dL and tested with four Contour lots, n=16 per lot. In a second study, 12 Contour lots were tested with blood adjusted to 50, 120, and 300 mg/dL glucose, n = 24 per lot. In a third study, 10 Contour lots were tested with blood adjusted to 43, 62, 127, 331, and 609 mg/dL, n = 20 per lot. Regression analysis (using the proportionally weighted least-squares model) conducted with the combined lot means from all three studies (N = 110) yields the following statistics:

| N | 110 |
| --- | --- |
| Slope | 0.987 |
| Intercept | -0.7 |
| r² | 0.996 |

c. Traceability, Stability, Expected values (controls, calibrators, or methods): Traceability is referenced to the NIST SRM 917a (dry D-glucose).

d. Detection limit:

10 mg/dL. This level was determined to be detectable by the linearity study (above) and by the hematocrit sensitivity study (see Analytical Specificity below).

e. Analytical specificity:

The interference effect of oxidizable substances (acetaminophen, uric acid, ascorbic acid) were tested at the following levels and were found to meet the sponsors acceptance criteria that the bias at the following limiting plasma concentrations be

{4}

less than 15%:

Bilirubin: 20 mg/dL

Acetaminophen: 20 mg/dL

Uric Acid: 15 mg/dL

Ascorbic Acid: 5 mg/dL

To verify that the Contour system provides accurate results at the lower end of the reportable range (where actual clinical specimens are rare and clinical trials cannot be expected to provide extensive data), a dose response study was conducted with blood at three hematocrit levels (40%, 55%, and 70%) adjusted to plasma glucose concentrations of 10, 20, 30, 40, 50, and 60 mg/dL. Four Contour lots (SN05M26A, SN06A06C, 6EC3C01 and 6EC3C02) were each tested with eight Contour meters, two replicates per meter, for a total of 16 replicates per lot per sample. Readings were compared to YSI plasma glucose. The sponsor's acceptance criterion was 95% of results within ±15 mg/dL of the YSI mean which 100% of the samples met. Regression statistics are summarized below.

| | 40% Hematocrit | 55% Hematocrit | 70% Hematocrit |
| --- | --- | --- | --- |
| N | 384 | 384 | 384 |
| Regression equation | y = 0.880(x) + 1.1 | y = 0.788(x) + 10.1 | y = 0.892(x) + 7.3 |
| r² | 0.985 | 0.956 | 0.975 |

To test the accuracy of the hematocrit correction algorithm, three lots, n = 20 per lot, with blood adjusted to hematocrit levels of 0% (pure plasma), 20%, 35%, 45%, 55%, and 70% at glucose concentrations of 80 and 350 mg/dL were tested. The sponsor's acceptance criteria were was a difference between mean at 45% hematocrit (normal) and means at 0% and 70% hematocrit < 10% or 7.5 mg/dL. Results are summarized below.

Hematocrit Effect

| | | | 0% Hct | 20% Hct | 35% Hct | 45% Hct | 55% Hct | 70% Hct |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Percent Deviation from 45% Hct | 80 mg/dL | SN05L10C | -1.8% | 2.3% | 2.2% | 0.0% | -1.4% | -2.8% |
| | | SN05L10D | -0.5% | 4.1% | 4.7% | 0.0% | -3.1% | -9.5% |
| | | SN05L15 | 1.8% | 5.4% | 6.3% | 0.0% | -2.7% | -8.3% |
| | | Mean | -0.2% | 3.9% | 4.4% | 0.0% | -2.4% | -6.8% |
| | 350 mg/dL | SN05L10C | -1.9% | -4.2% | -0.4% | 0.0% | 0.5% | -4.4% |
| | | SN05L10D | -1.4% | -3.4% | 1.0% | 0.0% | 0.9% | -3.8% |
| | | SN05L15 | 0.1% | -2.8% | 0.9% | 0.0% | 1.8% | -3.3% |
| | | Mean | -1.1% | -3.5% | 0.5% | 0.0% | 1.1% | -3.8% |

To test the accuracy of the hematocrit correction algorithm at higher levels closer to the claimed range, three lots, n = 5 per lot, with blood adjusted to hematocrit levels of 0% (pure plasma), 20%, 45%, 60%, and 70% at glucose concentrations of 450 and

{5}

550 mg/dL were tested. The sponsor’s acceptance criteria were was a difference between mean at 45% hematocrit (normal) and means at 0% and 70% hematocrit <10% or 7.5 mg/dL. Results are summarized below.

| | | | 0% Hct | 20% Hct | 45% Hct | 60% Hct | 70% Hct |
| --- | --- | --- | --- | --- | --- | --- | --- |
| Percent Deviation from 45% Hct | 450 mg/dL | 6JC3C07 | 5.2 | 2.2 | 0.0 | 3.7 | 2.5 |
| | | 6JC3C05 | 2.9 | 3.9 | 0.0 | 2.0 | 0.3 |
| | | 6HC3C09 | 0.9 | 3.4 | 0.0 | 3.2 | 0.5 |
| | | Mean | 3.0 | 3.2 | 0.0 | 3.0 | 1.1 |
| | 550 mg/dL | 6JC3C07 | 1.0 | -2.8 | 0.0 | 3.4 | 1.3 |
| | | 6JC3C05 | 4.8 | -0.3 | 0.0 | 2.8 | -0.3 |
| | | 6HC3C09 | 0.1 | -1.2 | 0.0 | 3.0 | -1.1 |
| | | Mean | 2.0 | -1.4 | 0.0 | 3.1 | -0.1 |

In addition to testing compounds that are known to interfere with electrochemical glucose monitoring systems (above), testing was also conducted with a variety of common compounds found in medications or food or occurring naturally in the blood. The following substances were tested and found to either have no effect trend at any concentration or to have a limiting concentration (interpolated or extrapolated concentration creating a bias of 15% at either 80 or 300 mg/dL glucose) that was significantly higher than the upper limit of the therapeutic or reference range.

{6}

| Compound | High Normal / Therapeutic Concentration | Test Levels | Limiting Concentration |
| --- | --- | --- | --- |
| Acetone | 2 mg/dL | 17, 34, 67 | N.S. |
| Acetylsalicylic Acid | 2 – 10 mg/dL | 8, 17, 34, 67 | 74 mg/dL |
| Albumin | 3.4 – 5.4 g/L | 17, 32, 56 g/L | N.S. |
| β-hydroxybutyric Acid | 32 mg/dL | 14, 28, 56 | N.S. |
| Caffeine | 1.5 mg/dL | 1.7, 3.4, 6.7 | N.S. |
| Creatinine | 1.5 mg/dL | 8,17,34 | N.S. |
| Dobutamine | 0.07 mg/dL | 9, 17, 34 | 16 mg/dL |
| Dopamine | 0.03 mg/dL | 1.1, 2.2, 4.5, 9.0 | 3.7 mg/dL |
| Ethanol | 300 mg/dL | 98, 196, 392 | N.S. |
| Fructose | 7 mg/dL | 84, 168, 336 | N.S. |
| Gentisic Acid | 5.0 mg/dL | 28, 56, 112 | 15 mg/dL |
| Glipizide | 0.8 mg/dL | 0.8, 1.7, 3.4 | 8.4 mg/dL |
| Glucosamine | 30 mg/dL | 56, 112, 224 | N.S. |
| Glutathione | 68.5 ± 10.7 mg/dL in RBC | 35, 70, 140, 280 | 28.8 mg/dL in plasma |
| Glyburide | 0.4 mg/dL | 0.6, 1.1, 2.2 | N.S. |
| Heparin | 4000 units/dL | 3808, 6048, 10528 | N.S. |
| Ibuprofen | 4.2 mg/dL | 11, 22, 45 | N.S. |
| L-Dopa | 0.3 mg/dL | 0.3, 0.7, 1.3 | 1.9 mg/dL |
| Metformin | 4.0 mg/dL | 1.1, 2.2, 4.5 | N.S. |
| Methyldopa | 0.75 mg/dL | 0.8, 1.7, 3.4 | 3.6 mg/dL |
| Naproxen | 12 mg/dL | 28, 56, 112 | 208 mg/dL |
| Precose | 6 mg/dL | 5.6, 11.2, 22.4 | N.S. |
| Sodium lactate | 20 mg/dL | 7, 14, 56, 112 | N.S. |
| Sodium salicylate | 30 mg/dL | 28, 56, 112 | 41 mg/dL |
| Sucrose | 2.7 mg/dL | 5.6, 11.2, 22.4 | N.S. |
| Tetracycline | 0.4 mg/dL | 1.1, 2.2, 4.5 | N.S. |
| Triglycerides | 190 mg/dL | 2580, 5160 | 5070 mg/dL |

It was found that the anticoagulant EDTA, especially when occurring at concentrations higher than normal due to under-filling an EDTA tube, is electrochemically active enough to produce a significant negative bias. For this reason, EDTA tubes will be restricted in the Limitations section of the labeling.

To investigate the effect of altitude on the system, three lots of reagent were tested across eight (8) meters both inside and outside a hypoxic chamber simulating an altitude of 12,095 feet. A maximum of three (3) repetitions on each meter were performed (n=24). Time limitations in some cases allowed for only two (2) repetitions on each meter (n=16). Whole blood samples with plasma glucose values of 50, 100, and 400 mg/dL were tested in the above manner, at both 40% and 60% hematocrits. The acceptance criteria defined by the sponsor was <10% difference between means in the low oxygen chamber and means outside the chamber. No systematic response to altitude is seen at either Hematocrit level, and percent bias at all levels was found to be less than 10%. Therefore the system is not significantly affected by lack of atmospheric oxygen up to 12,000 feet above sea level.

f. Assay cut-off: Not Applicable.

{7}

8

2. Comparison studies:

a. Method comparison with predicate device:

Fingerstick testing was conducted in two four-day studies. Data collected in four separate studies with contrived blood specimens were also used in order to bring the distribution of samples into compliance with the protocol:

Distribution of Plasma Glucose Values

| Glucose Range (mg/dL) | <50 | 50 - 80 | 81 - 120 | 121 - 200 | 201 - 300 | 301 - 400 | >400 |
| --- | --- | --- | --- | --- | --- | --- | --- |
| % of Samples | 5% | 15% | 20% | 30% | 15% | 10% | 5% |

Each specimen was tested with three Contour lots, $n = 2$ per lot for a total of six readings per specimen. After filling each bin in the ISO distribution, no additional specimens falling in that bin were included in the analysis. A total of 105 diabetic volunteers participated in the two fingerstick studies after giving informed consent, and of these donors, 74 met the distribution requirements. Since no diabetic subjects presented with extremely low or high blood glucose concentrations in either of the two fingerstick studies, readings from contrived samples were used to fill the extreme bins. Low glucose specimens were obtained from a dose response study designed to simulate neonatal specimens. In this study, blood was adjusted to glucose concentrations between 10 and $60\mathrm{mg/dL}$. To fill the lowest bin, the five lower levels (11.4, 20.5, 30.7, 40.6, and 49.8) were used (taking the two replicates generated with the first meter used with each lot). To help fill the second lowest bin, three samples at the $60\mathrm{mg/dL}$ level, each with a different hematocrit ($40\%$, $55\%$, and $70\%$) were used. Data from a second study was used to obtain two samples at $63\mathrm{mg/dL}$ (two testers), four samples at $301\mathrm{mg/dL}$ (two testers, two meters per tester), two samples at 405 mg/dL (two testers), and one sample at $503\mathrm{mg/dL}$. Data from a third study was used to obtain two samples at $51\mathrm{mg/dL}$ (two meters), two samples at $79\mathrm{mg/dL}$ (two meters), and one sample at 201, 301, 404, and $554\mathrm{mg/dL}$. A fourth study was used to obtain one sample at $300\mathrm{mg/dL}$.

A total of 13 lots were represented in the studies. The following table summarizes the distribution of specimens and lots in each study.

{8}

| Study | Lots* | <50 | 50 – 80 | 81 – 120 | 121 – 200 | 201 – 300 | 301 – 400 | >400 |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Fingerstick Study 1 | A, B, C | 0 | 3 | 11 | 29 | 7 | 4 | 0 |
| Fingerstick Study 2 | A, D, E | 0 | 3 | 9 | 1 | 6 | 1 | 0 |
| Contrived Blood Study 1 | A, F, G | 5 | 3 | 0 | 0 | 0 | 0 | 0 |
| Contrived Blood Study 2 | A, D, E | 0 | 2 | 0 | 0 | 0 | 4 | 3 |
| Contrived Blood Study 3 | H, I, J | 0 | 4 | 0 | 0 | 1 | 1 | 2 |
| Contrived Blood Study 4 | K, L, M | 0 | 0 | 0 | 0 | 1 | 0 | 0 |
| | Total | 5 | 15 | 20 | 30 | 15 | 10 | 5 |

* A: SN05M26A; B: SN05M26B; C: SN06A30A; D: 6FC3B01; E: 6FC3B02; F: 6EC3C01; G: 6EC3C02; H: SN06A17D; I: SN06A17E; J: SN06A17F; K: SN06A06C; L: SN06A06D; M: SN0617A

Contrived specimens were prepared with heparinized venous blood that was allowed to glycolyze to produce low glucose levels or that were supplemented with 20% glucose stock solution to produce high glucose levels. For fingerstick comparison readings, approximately 200 µL of fingerstick blood was collected into a heparinized micro-collection tube. To obtain comparison glucose values, all specimens were centrifuged to separate the plasma from the blood cells, and the plasma portion was tested on the YSI STAT Plus Glucose Analyzer. Plasma glucose levels ranged from 11 to 554 mg/dL. Hematocrit levels ranged from 26% to 70%. The tables below include the proportionally weighted regression statistics and the percentage of readings within several error limits around the YSI plasma glucose comparison values (±5, 10, 15, and 20 mg/dL for samples &lt; 75 mg/dL, ±5%, 10%, 15%, and 20% for samples ≥ 75 mg/dL).

Contour vs. YSI Plasma

Regression Equations
(Proportionally Weighted Least Squares Model)

| Regression | y = 1.002(x) – 1.1 |
| --- | --- |
| 95% CI of Slope | 0.994 to 1.010 |
| 95% CI of Intercept | -1.48 to -0.72 |
| S_{yx} (proportional to YSI) | 7.25% |

Bias at Key Glucose Levels

| Plasma Glucose | 60 mg/dL | 126 mg/dL | 200 mg/dL | 400 mg/dL |
| --- | --- | --- | --- | --- |
| %Bias | -1.6% | -0.7% | -0.3% | -0.1% |

{9}

Contour Accuracy Assessment (ISO 15197 Section 7.4)
Contour vs. YSI Plasma

Accuracy Results for Glucose Concentration &lt; 75 mg/dL

| N | Within ±5 mg/dL | Within ±10 mg/dL | Within ±15 mg/dL | Within ±20 mg/dL |
| --- | --- | --- | --- | --- |
| 108 | 87 (80.6%) | 108 (100%) | 108 (100%) | 108 (100%) |

Accuracy Results for Glucose Concentrations ≥ 75 mg/dL

| N | Within ±5 % | Within ±10 % | Within ±15 % | Within ±20 % |
| --- | --- | --- | --- | --- |
| 492 | 261 (53.0%) | 413 (83.9%) | 467 (94.9%) | 488 (99.2%) |

Summary Assessment of Accuracy
Readings within ISO Minimum Acceptable Performance Criterion
(±15 mg/dL if &lt;75 mg/dL, ±20% if ≥75 mg/dL)

596 of 600 (99.3%)

The ISO 15197 standard specifies that at least 95% of readings should fall within ±20% of the standing comparison method for levels ≥ 75 mg/dL and within ±15 mg/dL for levels &lt; 75 mg/dL, and this limit is highlighted in all plots. The system meets the ISO 15197 criterion of at least 95% of readings falling within ±20% (or ±15 mg/dL at levels below 75 mg/dL) of the sponsor’s comparative method, the YSI Stat Plus Glucose Analyzer.

b. Matrix comparison:
See Method Comparison section above.

3. Clinical studies:
a. Clinical Sensitivity:
Not Applicable.
b. Clinical specificity:
Not Applicable.
c. Other clinical supportive data (when a. and b. are not applicable):

Capillary Blood

Performance of the blood glucose monitoring system with capillary blood specimens and an assessment of the User Guide and Quick Reference Guide to show how well untrained subjects could perform a fingerstick, obtain an accurate blood glucose measurement, and perform control solution assays was examined at one site by 109

{10}

subjects with diabetes. Health care professionals (HCP) tested in parallel. Meters referenced to whole blood and plasma laboratory glucose and three test strip lots were used. The blood glucose results obtained with the system by the subjects and health care professionals were compared to laboratory glucose results using linear regression and are summarized below.

| Capillary Fingerstick vs. YSI | | |
| --- | --- | --- |
| glucose range 75-397 mg/dL
hematocrit: 31-59% | | |
| | Lay User | HCP |
| N | 108 | 108 |
| Slope | 0.9585 | 0.9479 |
| y-intercept | 3.7876 | 6.4176 |
| r² | 0.9661 | 0.9674 |

## Alternate Site Testing

For alternate site testing, the average of two fingertip test results with the system was used as the comparative method for all alternative site lancing results. A hematocrit determination for each subject was performed from fingertip blood. Results using linear regression are summarized below:

| AST vs. Fingerstick | | |
| --- | --- | --- |
| | Palm | Forearm |
| glucose range: | 41-374 mg/dL | 39-394 mg/dL |
| hematocrit: | 31-53% | 31-53% |
| N | 50 | 47 |
| Slope | 0.9993 | 1.0126 |
| y-intercept | 2.5104 | -7.8537 |
| r² | 0.9721 | 0.9497 |

## Neonatal Blood

Performance of the system with neonatal blood specimens was examined by health care professionals (HCP) at 2 sites using blood samples from 124 subjects (for the regression calculation 2 samples were removed as outliers meeting the NCCLS guideline for outliers: Method Comparison and Bias Estimation Using Patient Samples (EP9-A)). Three test strip lots were used. The blood glucose results obtained with the system were compared to laboratory glucose results. Results using linear regression are summarized below:

{11}

| Neonatal vs. YSI | |
| --- | --- |
| glucose range: 27-131 mg/dL | |
| hematocrit: 27-70% | |
| N | 122 |
| Slope | 1.0065 |
| y-intercept | -0.0522 |
| r² | 0.8952 |

The sponsor also demonstrated that these results met the ISO accuracy criteria of 95% of the meter results for all lots falling within ±15 mg/dL for samples &lt; 75 mg/dL and within ±20% for specimens &gt; 75 mg/dL of a laboratory reference method. Results are summarized in the tables below.

## AGREEMENT OF NURSERY SAMPLES TO THE SITE LABORATORY GLUCOSE METHODS

| Test Strip Lot | Meter Reference | Results Within Given Limits of the Laboratory Glucose Method | | |
| --- | --- | --- | --- | --- |
| | | ±7.5 mg/dL or 10% | ±15 mg/dL or 20% | ±22.5 mg/dL or 30% |
| 05M26A | Plasma | 79.5% (35/44) | 97.7% (43/44) | 97.7% (43/44) |
| 05M26B | Plasma | 73.8% (31/42) | 97.6% (41/42) | 97.6% (41/42) |
| 06A30A | Plasma | 71.1% (27/38) | 89.5% (34/38) | 100% (38/38) |
| Combined | Plasma | 75.0% (93/124) | 97.6% (121/124) | 98.4% (122/124) |

Thirty-two samples were obtained (based on the average of the laboratory analyzer results) ranging from 10-50 mg/dL. The table below shows that the Contour system's accuracy is adequate for samples ranging from 10-50 mg/dL, meeting ISO accuracy criteria of 95% of the meter results for all lots falling within ±15 mg/dL for samples &lt; 75 mg/dL and within ±20% for specimens &gt; 75 mg/dL of a laboratory reference method.

## AGREEMENT OF NURSERY SAMPLES TO THE SITE LABORATORY GLUCOSE METHODS - Specimens 10 to 50 mg/dL

| Test Strip Lot | Meter Reference | Results Within Given Limits of the Comparative Glucose Method | | |
| --- | --- | --- | --- | --- |
| | | ±7.5 mg/dL or 10%^{a} | ±15 mg/dL or 20%^{b} | ±22.5 mg/dL or 30%^{c} |
| Combined | Plasma | 68.8% (22/32) | 96.9% (31/32) | 96.9% (31/32) |

The Contour's meter bias was compared to the hematocrit level for each blood specimen. Correlation coefficients were determined using linear regression shown in the table below.

{12}

ASSOCIATION RESULTS FOR METER BIAS VS. SPECIMEN HEMATOCRIT LEVEL

| Plasma or WB | Strip Lot | Site | N | Slope | Slope 95% C.I. | Intercept (mg/dL) | Intercept 95% C.I. | Sy.x | Correlation Coefficient |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| Plasma | All lots | 1&2 | 122 | 0.44 | 0.14 to 0.73 | -21.76 | -37.4 to -6.12 | 11.47 | 0.255 |

Since the sponsor's neonatal clinical samples did not cover the claimed range of the meter, contrived samples were used to test the low and high ranges at high levels of Hematocrit. These studies are fond in the above in the Analytical Specificity section.

# Venous Blood

Performance of the system with venous blood specimens was examined at one site by a medical technologist who measured the glucose in 169 samples. Venous blood samples were collected into tubes containing heparin. The material assayed in this study consisted of excess volume that remained after routine laboratory blood work had been completed. Plasma and whole blood referenced meters and three test strip lots were used. The glucose concentration in the samples was adjusted, as necessary, to span the operating range of the system. Results were compared to glucose results obtained at the site with a laboratory analyzer. Results using linear regression are summarized below:

| Venous Blood vs. YSI | |
| --- | --- |
| glucose range: 12-609 mg/dL | |
| hematocrit: 15-53% | |
| N | 169 |
| Slope | 0.8927 |
| y-intercept | 3.614 |
| r² | 0.9884 |

The bias of results is within the ISO accuracy limits of $\pm 15\mathrm{mg / dL}$ or $20\%$ of the laboratory glucose method. The results are shown below.

| Median Differences From The Laboratory Method | | |
| --- | --- | --- |
| Lot | <75 mg/dL (n=26) | ≥ 75 mg/dL (n=143) |
| 05M26A | -2.0 mg/dL | -7.4% |
| 05M26B | -2.0 mg/dL | -8.3% |
| 06A30A | -3.0 mg/dL | -11.0% |

4. Clinical cut-off: Not Applicable.

{13}

14

5. Expected values/Reference range:
The sponsor included the following Expected Values for normal glucose levels in their meter’s user manual:

| Status | Range (mg/dL) |
| --- | --- |
| Before meals | 90-130 |
| 2 hours after meals | <180 |

American Diabetes Association: Standards of Medical Care in Diabetes (Position Statement). Diabetes Care 29 (Suppl. 1):S10, 2006.

N. Instrument Name:
Ascensia® CONTOUR ® Blood Glucose Monitoring System

O. System Descriptions:

1. Modes of Operation:
Each test strip is single use and must be replaced with a new strip for additional readings.

2. Software:
FDA reviewed applicant’s Hazard Analysis and software development processes for this line of product types in k023657 and updated functions in this submission.

3. Specimen Identification:
There is no sample identification function with this device. Samples are applied directly to the test strip as they are collected.

4. Specimen Sampling and Handling:
This device is intended to be used with capillary whole blood from the finger, the palm, or the forearm only. Since the whole blood sample is applied directly to the test strip there are no special handling or storage issues.

5. Calibration:
Calibration of the meter occurs by insertion of the test strip and the meter’s recognition of the strip configuration.

6. Quality Control:
The sponsor provides a glucose control solution with the test strips. The meter recognizes the sample as a control solution which prevents control results from being stored in the internal memory. An acceptable range for each control level is printed on the test strip vial label and box. The user is referred to the troubleshooting section of the owner’s manual if control results fall outside these ranges.

P. Other Supportive Instrument Performance Characteristics Data Not Covered In The “Performance Characteristics” Section above:
None.

{14}

Q. Proposed Labeling:
The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.

R. Conclusion:
The submitted information in this premarket notification is complete and supports a substantial equivalence decision.

15