HemoCue Glucose 201 RT Systems Performance Report

HemoCue Glucose 201 RT Systems Performance Report

Performance Report HemoCue Glucose 201 RT Systems This document summarizes the performance of the HemoCue Glucose 201 RT System and the HemoCue Glucose 201 DM RT System. Summary The HemoCue Glucose 201 RT Systems are based on a glucose dehydrogenase method and consist of a small dedicated analyzer and a unique disposable microcuvette. These systems combine the precision and accuracy of a central laboratory with the speed and convenience of obtaining results at the point-of-care. With just a fraction of a drop of whole blood and three simple steps, the HemoCue Glucose 201 RT Systems will produce immediate, accurate results for screening, diagnosing or monitoring diabetes, as well as monitoring neonatal blood glucose levels. The HemoCue 201 RT Analyzer stores patient and QC results together with date and time, the results can be downloaded to a PC. The analyzer is equipped with an outlet for printing the results. The HemoCue 201 DM RT Analyzer can forward measurements to host system using CLSI POCT1A. GPM316 130617 2

Content Performance Report HemoCue Glucose 201 RT Systems... 2 Summary... 2 Content... 3 HemoCue Glucose 201 RT System specifications... 4 HemoCue Glucose 201 DM RT System specifications... 5 System technology... 6 HemoCue Glucose 201 RT Microcuvette... 6 HemoCue Glucose 201 RT Analyzer... 8 HemoCue Glucose 201 DM RT Analyzer... 9 Traceability of calibration HemoCue Glucose 201 RT System and HemoCue Glucose 201 DM RT System... 11 Environmental... 11 Interferences... 12 Studies... 14 1. Precision studies with control and patient material... 14 2. Accuracy studies venous sample... 15 Comparative study for venous whole blood samples between HemoCue Glucose 201 RT System and ID GC-MS... 15 3. Accuracy study with capillary sample... 16 Capillary blood measured with HemoCue Glucose 201 RT Analyzer versus mean values for YSI 2300 STAT Plus... 16 4. Accuracy study with venous and capillary samples... 17 Comparative study between the HemoCue Glucose 201 RT System and YSI 2300 STAT Plus.... 17 5. Accuracy study with arterial sample... 18 HemoCue Glucose 201 RT System compared with ABL 825 FLEX... 18 6. Accuracy and precision study with neonatal samples... 19 Clinical laboratory evaluation of neonatal samples... 19 Publications... 20 Comparison of the HemoCue Glucose 201 DM RT System with Cobas Modular (Roche)... 20 Comparison of the HemoCue Glucose 201 RT System with Architect ci8200 (Abbot)... 21 Comparison of the HemoCue Glucose 201 DM RT System with Vitros 5.1 (Ortho-Clinical Diagnostics)... 22 Comparison of the HemoCue Glucose 201 RT System with a hexokinase method.... 23 Validation of a new generation POCT glucose device with emphasis on aspects important for glycemic control in the hospital care... 25 GPM316 130617 3

HemoCue Glucose 201 RT System specifications Intended purpose/intended use The HemoCue Glucose 201 RT System is used for quantitative determination of glucose in whole blood, supplementing the clinical evidence in the diagnosis and treatment of patients with diabetes as well as monitoring of neonatal blood glucose levels. The system should not be used on critically ill neonates in neonatal intensive care settings. The HemoCue Glucose 201 RT System is for In Vitro Diagnostic use only. The HemoCue Glucose 201 RT Analyzer is only to be used with HemoCue Glucose 201 RT Microcuvettes. For professional use only. Category Principle Calibration Sample Material Displayed range Limit of Detection Results Sample volume Shelf life Package Size Dimensions Weight Storage Temperature Operating temperature Operating humidity HemoCue Glucose 201 RT System specifications Modified glucose dehydrogenase in which the total amount of glucose is measured at the end point photometrically Factory calibrated and traceable to the ID GC-MS method, needs no further calibration and no coding Capillary, venous or arterial whole blood Plasma equivalent values 0-31 mmol/l (0-560 mg/dl) Whole blood values 0-27.8 mmol/l (0-500 mg/dl) 0.56 mmol/l (10 mg/dl) Within 1 minute for normal glucose level <4 µl whole blood Hematocrit Range 20%-70% Altitude Memory function Power Interface Quality Control Data Management Each individually packed microcuvette: 12 months from manufacturing 25 single pack microcuvettes/box, 4 boxes/package 160 x 85 x 43 mm (6.3 x 3.35 x 1.69 inches) 350 g (0.77 pounds) including batteries Analyzer: 0-50 C (32-122 F) Microcuvette: 0-30 C (32-86 F) 15-27 C (59-80 F) Up to 90 % relative humidity Not affected by altitude 600 measurements AC adapter or 4 AA batteries Connectable to printer and HemoCue Basic Connect including barcode scanner Built-in self-test ; system can be verified using liquid control HemoCue Basic Connect software, store and view measurements and patient data GPM316 130617 4

HemoCue Glucose 201 DM RT System specifications Intended purpose/intended use The HemoCue Glucose 201 DM RT System is used for quantitative determination of glucose in whole blood, supplementing the clinical evidence in the diagnosis and treatment of patients with diabetes as well as monitoring of neonatal blood glucose levels. The system should not be used on critically ill neonates in neonatal intensive care settings. The HemoCue Glucose 201 DM RT System is for In Vitro Diagnostic use only. The HemoCue Glucose 201 DM RT Analyzer is only to be used with HemoCue Glucose 201 RT Microcuvettes. For professional use only. Category Principle Calibration Sample Material Displayed range Limit of Detection Results Sample volume Shelf life Package Size Dimensions Weight Storage Temperature Operating temperature Operating humidity HemoCue Glucose 201 DM RT System specifications Modified glucose dehydrogenase in which the total amount of glucose is measured at the end point photometrically Factory calibrated and traceable to the ID GC-MS method, needs no further calibration and no coding Capillary, venous or arterial whole blood Plasma equivalent values 0-31 mmol/l (0-560 mg/dl) Whole blood values 0-27.8 mmol/l (0-500 mg/dl) 0.56 mmol/l (10 mg/dl) Within 1 minute for normal glucose level <4 µl whole blood Hematocrit Range 20%-70% Altitude Memory function Barcode scanner Power Interface Quality Control Each individually packed microcuvette: 12 months from manufacturing 25 single pack microcuvettes/box, 4 boxes/package Analyzer: 170 x 93 x 50 mm (6.7 x 3.66 x 1.97 inches). Docking Station: 206 x 135 x 61 mm (8.1 x 5.3 x 2.4 inches) Analyzer: 350 g (0.77 pounds) including batteries Docking Station: 566 g (1.24 pounds) Analyzer: 0-50 C (32-122 F) Microcuvette: 0-30 C (32-86 F) 15-27 C (59-80 F) Up to 90% relative humidity Not affected by altitude 4000 test results, 500 QC results and 500 Analyzer log entries Built in barcode scanner Internal rechargeable batteries or docking station with AC adapter USB/LAN POCT1-A (CLIS standard) Built-in self-test ; system can be verified using liquid control GPM316 130617 5

System technology Test method The HemoCue Glucose 201 RT System and the HemoCue Glucose 201 DM RT System use the Glucose dehydrogenase method and photometric detection. The glucose reaction is an enzymatic method, in which tetrazolium salt (MTT) is used to obtain a quantification of glucose in visible light. At first, erythrocytes are hemolyzed by saponin. Mutarotase catalyzes the conversion of α- D-glucose to β-d-glucose. β-d-glucose reacts with NAD + (Nicotinamide Adenine Dinucleotide) in the presence of the enzyme glucose dehydrogenase. The β-d-glucose is oxidized to D-gluconolactone and NAD + is reduced to NADH. The color reaction occurs when NADH, catalyzed by the enzyme diaphorase, reduces MTT (Methylthiazolylphenyltetrazolium), to MTTH, a colored formazan. The amount of colored formazan is spectrophotometric measured at 670 nm, with a second measurement at 840 nm for background compensation. The analysis time depends on the glucose concentration in the sample, and is an end-point measurement of the chemical reaction. Start measurement no later than 40 seconds after filling the microcuvette. The measurement will take approximately 1 min for normal glucose levels. HemoCue Glucose 201 RT Microcuvette Microcuvette Glucose 201 RT Microcuvette is specially designed and serves both as a pipette and as a measuring microcuvette. It is for single use and made of copolyester plastic, which turns into H 2 O and CO 2 when incinerated. The cavity of the microcuvette is filled by capillary force with less than 4 µl capillary, venous or arterial whole blood. The cavity contains an exact amount of dry reagents for the enzyme reaction to take place. Individually wrapped microcuvettes Each disposable microcuvette is individually wrapped to avoid contamination and to offer maximum shelf life. GPM316 130617 6

Microcuvette lots A lot to lot comparison was performed using five different cuvette lots and three different EDTA whole blood samples prepared to glucose levels 2.2, 6.1 and 22.2 mmol/l, whole blood values (Table 1). Each sample and level was analyzed with five Glucose 201 RT Analyzers and with four replicates per cuvette lot in alternated order. Table 1 Total Range of mean values, for whole blood samples between five different microcuvette lots. Glucose level mmol/l (mg/dl ) Max Mean mmol/l (mg/dl) Min Mean mmol/l (mg/dl) Range mmol/l (mg/dl) Level 1 2.2 (40) 2.3 (40.9) 2.0 (36.8) 0.2 (4.1) Level 2 6.1 (110) 6.1 (110.1) 5.9 (106.6) 0.2 (3.5) Level 3 22.2 (400) 22.6 (406.6) 21.9 (394.0) 0.7 (12.6) Storage and expiry date The microcuvettes are to be stored at 0-30 C (32-86 F). Unused microcuvettes should be kept in the original package but once an individual package is opened, the microcuvette must be used immediately or discarded. Use the microcuvettes prior to the expiry date printed on the package. The expiry date is set 12 months from manufacturing date. GPM316 130617 7

HemoCue Glucose 201 RT Analyzer Calibration and quality control The HemoCue 201 RT Analyzer is factory calibrated and does not need any further calibration by the end user. It has an internal quality control, the selftest. At every start up the selftest will automatically verify the performance of the optronic unit. The selftest is performed every second hour when the analyzer is switched on. To ensure accuracy, the analyzer checks the stability of the blank value every fifth second. External quality control If an external quality control test is to be performed, GlucoTrol -NG from Eurotrol B.V. is recommended by HemoCue. GlucoTrol -NG is a glucose control that exhibits qualities similar to blood. It is available in five physiological levels and has 30-day open vial stability when refrigerated (2-8 C). Units The HemoCue Glucose 201 RT analyzer is available in 2 different units, mmol/l or mg/dl. This is factory set and can not be changed. Whole blood and plasma equivalent results Two versions of the HemoCue Glucose 201 RT Analyzer are available, displaying either whole blood results or plasma equivalent results. For plasma equivalent analyzers, the plasma conversion is made according to IFCC using the factor 1.11 to convert whole blood result to plasma equivalent result (Ref; D Orazio et al: Approved IFCC recommendation on Reporting Results for Blood Glucose, Clinical Chemistry 51:9 1593-1576, 2005). Limit of detection The limit of detection is defined as the lowest amount of an analyte in a sample that can be detected with stated probability. (Ref; Protocols for determination of limits of detection and limits of quantification, CLSI EP17-A) The limit of detection for the HemoCue Glucose 201 RT in whole blood is 0.56 mmol/l (10 mg/dl). Memory function The HemoCue Glucose 201 RT Analyzer automatically stores up to 600 results. When the memory is full the analyzer will automatically overwrite the oldest result. If needed, single or all results can be deleted. Storage The HemoCue Glucose 201 RT Analyzer can be stored and transported at 0-50 C (32-122 F). The operating temperature is 15-27 C (59-80 F) and the analyzer needs to reach ambient temperature before use. The analyzer operates at a humidity <90% relative humidity. Connectivity The HemoCue Glucose 201 RT Analyzer can be connected to a serial printer for immediate printout of measurement. When there is a need for improved patient tracking, HemoCue Basic Connect Software provides analyzer with basic data management functions and connectivity to EMR (Electronic Medical Record) systems. The software makes it possible to connect up to 6 analyzers to one computer, electronically transfer test results to computer and forwarding test results from computer to EMR systems. The data management contains optional prompting for operator ID, site ID, lab ID, patient ID, patient name and microcuvette lot no. The software enables users to initiate tests, add comments to the results and mark results, falling outside a user-defined range, with a flag. GPM316 130617 8

HemoCue Glucose 201 DM RT Analyzer The HemoCue Glucose 201 DM RT is the data management version of the HemoCue Glucose 201 RT. Calibration The HemoCue 201 DM RT Analyzer is factory calibrated and does not need any further calibration by end user. It has an internal quality control, the selftest. At every start up the selftest will automatically verify the performance of the optronic unit. The selftest is performed every eight hour when the analyzer is switched on. To ensure accuracy, the analyzer checks the stability of the blank value every fifth second. External quality control If an external quality control test is to be performed, GlucoTrol -NG from Eurotrol B.V. is recommended by HemoCue. GlucoTrol -NG is a glucose control that exhibits qualities similar to blood. It is available in five physiological levels and has a 30-day open vial stability when refrigerated (2-8 C). Units The HemoCue Glucose 201 DM RT Analyzer is available in 2 different units, mmol/l or mg/dl. This is factory set and can not be changed. Whole blood and plasma equivalent results Two versions of the HemoCue Glucose 201 DM RT Analyzer are available, either displaying whole blood results or plasma equivalent results. For plasma equivalent analyzers, the plasma conversion is made according to IFCC using the factor 1.11 to convert whole blood result to plasma equivalent result (Ref; D Orazio et al: Approved IFCC recommendation on Reporting Results for Blood Glucose, Clinical Chemistry 51:9 1593-1576, 2005). Limit of detection The limit of detection is defined as the lowest amount of an analyte in a sample that can be detected with stated probability (Ref; Protocols for determination of limits of detection and limits of quantification, CLSI EP17-A). The limit of detection for the HemoCue Glucose 201 DM RT in whole blood is 0.56 mmol/l (10 mg/dl). Memory function The HemoCue Glucose 201 DM RT Analyzer automatically stores up to 4 000 results. When the memory is full there are two options, either the analyzer stores new test result by overwriting the oldest result or lockout. At lockout no patient tests can be performed until the stored results have been deleted. If needed, single or all results can be deleted. Storage The HemoCue Glucose 201 DM RT Analyzer can be stored and transported at 0-50 C (32-122 F). The operating temperature is 15-27 C (59-80 F) and the analyzer needs to reach ambient temperature before use. The analyzer operates at a humidity <90% relative humidity. HemoCue DM docking station Full data management is possible when the Analyzer is placed in the docking station which is connected via USB or LAN to a PC hosting the DMS-Software. The battery is recharged while in the docking station. Up to five docking stations can be connected to each other with only one of them connected to an AC adapter. GPM316 130617 9

Data management features The HemoCue Glucose 201 DM RT features to ensure patient safety: Barcode scanning of patient identification and comparison with current patient lists. Performing measurement can be set to only be accessible by certified operators/analyzer (up to 1 000 operators/analyzers) Allows for Short-Turn-Around-Test in emergency situations. Storage of 4 000 Patient measurements. Quality Control storage of 500 measurements. Quality Control management, QC scheduling and lock-out. The analyzer is designed to provide optimal workflow for the operator with built-in barcode scanner, touch screen and the possibility to enter comments for the results. The storage and possibility to transfer patient and QC results, ensures that no results are lost. The results are transferred automatically when the analyzer is docked in the docking-station. The workflow for the Point-Of-Care coordinator is supported by connectivity to the HemoCue 201 DM - DMS software and to all major data management server software (observation reviewers). These software applications will grant control of all analyzers, cuvette and QC lots used. The analyzer allows the software to perform QC scheduling centrally and in addition, the analyzers can be locked remotely by using supervisory lockout. The analyzer also has linearity and proficiency functions. The daisy-chain function of the docking-station makes it possible to connect up to five (5) docking-stations, only using one AC adapter and one network (LAN) connection. Training HemoCue provides a web-based interactive e-learning application that can be used for certification/re-certification of operators. The HemoCue e-learning can be integrated with the HemoCue 201 DM - DMS software and to selected middleware to enable an automatic workflow for certification. The workflow is as follows; the operator receives a mail with a link to e-learning - the operator takes the tailored certification course - the e-learning updates the operator list in the software - finally the software updates the operator list in the analyzer. GPM316 130617 10

Traceability of calibration HemoCue Glucose 201 RT System and HemoCue Glucose 201 DM RT System The calibration of the HemoCue Glucose 201 RT Analyzer and the HemoCue Glucose 201 DM RT Analyzer and HemoCue 201 RT Microcuvettes are fully traceable to ID GC-MS. HemoCue calibrates reference analyzers and reference lot of microcuvettes 4 to 5 times a year with the ID GC-MS. These analyzers and lots are used as reference for all manufactured analyzers and microcuvettes at HemoCue. The ID GC-MS system is calibrated with a NIST standard. The ID GC-MS method is controlled in an external IFCC control program for reference labs. HemoCue also controls the YSI used internally at HemoCue with ID GC-MS (Figure 1). Figure 1 Traceability of Calibration, HemoCue Glucose 201 RT Systems. Environmental HemoCue has an environmental management system that fulfills the requirements of SS- EN-ISO 14001:2004 with respect to design and development, production and distribution of test system for In-Vitro Diagnostics. Further information can be seen at www.sp.se According to REACH (EG 1907/2006), HemoCue is a downstream user of chemicals and a manufacturer of articles. HemoCue Microcuvettes is defined as a product where no Material Safety Data Sheet needs to be sent to customer. GPM316 130617 11

Interferences Interference studies have been performed for HemoCue Glucose 201 RT (Ref; Interference testing in Clinical Chemistry CLSI protocol EP7-A2, Vol.25 No. 27, Paired-difference testing, Worst Case, section 7.1). Potentially interfering substances were added to samples and bias relative to a control of the same samples was evaluated. The potential interfering substances were tested at relatively high concentrations to simulate worst case and were generally tested at 2-3 different glucose levels. If interference was found or known from earlier studies, dose-response experiments, (Ref; CLSI EP7-A2 section 7.2) were performed. In some cases the level of interference of a substance had to be determined with patient samples (Ref; CLSI EP7-A2, section 8). These studies were based on the analysis of two groups of patient specimen, i.e. a test group and a control group (Table 2). Table 2 Interferences Interferent Highest concentration tested without interference Anti-coagulants EDTA 5 x normal level Heparin 2.8 x normal level (8000U/L) Lithium-Heparin 5 x normal level Lithium-Iodo-Acetate 5 x normal level Sodium fluoride/ EDTA 5 x normal level Sodium fluoride/ Potassium oxalate 2 x normal level Sodium fluoride/ Heparin 3 x normal level Exogenous analyte Salicylic Acid 4.34 mmol/l (60 mg/dl ) Ibuprofen 2.19 mmol/l (50 mg/dl) Tetracycline 20 mg/dl Dopamine 13 mg/dl Ephedrine 1 mg/dl L-Dopa 50 mg/dl Methyldopa 2.5 mg/dl Tolazamide 25 mg/dl Tolbutamide 3.7 mmol/l (100 mg/dl) Acetaminophen 1.32 mmol/l (20 mg/dl) Ascorbic Acid 0.34 mmol/l (6 mg/dl) Level defined by EP7, 0.57 mmol/l (10 mg/dl) tested with no interference. Caffeine 0.31 mmol/l (6 mg/dl) Endogenous analyte Creatinine 2655 µmol/l (30 mg/dl) Uric acid 1050 µmol/l (20 mg/dl) Urea 83 mmol/l (500mg/dL) Unconjugated Bilirubin 342 µmol/l (20 mg/dl) Conjugated Bilirubin 475 µmol/l (40 mg/dl) ph 6.5-6.6/8.1-8.6 Leukocytes (WBC)** 50-205 10 9 /L HHb (Reduced Hemoglobin) >75% HbCO* 30% Hematocrit* 70% Cholesterol** 9.6 mmol/l (350 mg/dl) *Data from Dose Response studies **Data from Patient studies GPM316 130617 12

Interferent Sugar Maltose Lactose Mannose Fructose Galactose Dextran* (Replacement Agent) Highest concentration tested without interference 15 mmol/l (540 mg/dl) 15 mmol/l (540 mg/dl) 15 mmol/l (270 mg/dl) 15 mmol/l (270 mg/dl) 15 mmol/l (270 mg/dl) 26 g/l Interfering substances with limitations in operating manual and package Insert Glucosamine is a natural monosaccharide used to treat osteoarthritis. Blood concentration >1.44 mmol/l (26 Glucosamine* mg/dl) will interfere with the glucose measurement. See limitations in operating manual and package insert. Orally administrated ascorbic acid will not affect the measurement. Intravenously administrated, ascorbic acid Ascorbic acid* resulting in blood concentration >0.57 mmol/l (10 mg/dl) will interfere with the measurement. See limitations in operating manual and package insert. During a xylose absorption test the concentration of xylose will likely exceed >1.44mmol/L (22 mg/dl) and Xylose* interfere with the glucose measurement. See limitations in operating manual and package insert. High levels of triglyceride can cause turbidity in the blood Triglycerides** sample and disturb the measurement. See limitations in operating manual and package insert. Intralipid (fat emulsion for intravenous use) that give >500 mg/l could interfere with the measurement. If a specimen is taken in this situation the firmware could Intralipid* detect it as a deviation from the normal pattern and an E70 or E71 information message will be displayed. See limitations in operating manual and package insert. The range of normal levels of Met Hemoglobin is 0-1%. Met Hemoglobin* Levels >7% will interfere with the analysis. See limitations in operating manual and package insert. *Data from Dose Response studies **Data from Patient studies Limitations in operating manual and package insert (extraction) The measured blood glucose value of, grossly lipemic specimens (intralipids >500 mg/l), turbid specimens, specimens containing >7 % methemoglobin and high concentrations of glucosamine (>1.44 mmol/l, >26 mg/dl) may give incorrect results and should be interpreted with caution. At normal oral administration of xylose the HemoCue Glucose 201 RT System does not have any interference (Xylose <1.44 mmol/l, <22 mg/dl). However the high dosage used during a xylose absorption test may give incorrect results and should be interpreted with caution. GPM316 130617 13

Studies The HemoCue Glucose 201 RT System and the HemoCue Glucose 201 DM RT System are based on the same platform and have identical performance. 1. Precision studies with control and patient material Precision study with control material The precision for Glucose 201 RT, repeatability and within laboratory precision, was determined according to EP5-A2 Ref; (CLSI Document EP5-A2, Vol. 24, No. 25, Evaluation of Precision Performance of Quantitative Measurement Methods, Approved Guideline Second Edition) and ISO15197. Three different levels of Eurotrol, GlucoTrol-NG control solutions were analyzed in duplicate twice a day during 20 days on ten different analyzers. Two different lots of microcuvettes were used in the study. In total 800 samples/level (Table 3). Table 3 Precision study with control material. Repeatability Within-laboratory precision Control level N Mean mmol/l SD pooled mmol/l CV % SD pooled mmol/l (mg/dl) CV % (mg/dl) (mg/dl) 1 800 2.5 (45.9) 0.06 (1.15) 2.5 0.07 (1.28) 2.8 2 800 5.7 (101.8) 0.08 (1.36) 1.3 0.09 (1.62) 1.6 4 800 17.0 (305.7) 0.18 (3.26) 1.1 0.21 (3.72) 1.2 Precision study, with patient sample material Precision data is obtained from duplicate samples in validation studies at four different sites. Two different analyzers and three microcuvette lots were used. Standard deviation (SD), for duplicate samples, within each group was calculated (Table 4). Table 4 Precision study with patient sample material. SD mmol/l (mg/dl) Glucose range mmol/l (mg/dl) Capillary samples n=222 Venous samples n=180 All samples n=402 0-11.1 (0-200) 0.14 (2.60) 0.13 (2.32) 0.14 (2.49) 11.2-22.2 (201-400) 0.19 (3.50) 0.31 (5.60) 0.27.(4.81) 22.3-27.8 (401-500) 0.35 (6.05) 0.35 (6.36) 0.34 (6.22) GPM316 130617 14

2. Accuracy studies venous sample Comparative study for venous whole blood samples between HemoCue Glucose 201 RT System and ID GC-MS Patient samples, reference method ID GC-MS The study was performed according to CLSI EP9-A2. The evaluation time was 7 working days. Each sample was analyzed in duplicate with the HemoCue Glucose 201 RT and in parallel with the comparative reference method ID GC-MS. Human venous whole blood samples from 104 patients at Linköping University Hospital were used. Glucose solution was used to spike some samples to higher glucose levels. The user in this evaluation is one employee at the site, who is usually assigned to perform analysis of glucose samples with the ID GC-MS. The study shows very good agreement with ID GC-MS (Figure 2 and Figure 3). Figure 2 Venous whole blood measured at the HemoCue Glucose 201 RT compared to ID GC-MS. Difference (Glucose 201RT first replicate - ID GC-MS mean values), mg/dl 0 5 10 15 20 100 50 0-50 Mean values ID GC-MS, mmol/l -100 0 100 200 300 400 500 Figure 3 Mean values ID GC-MS, mg/dl Bland-Altman Difference plot for Glucose 201 RT System versus ID GC-MS, mg/dl and mmol/l. 4 2 0-2 -4 Difference (Glucose 201RT first replicate - ID GC-MS mean values), mmol/l GPM316 130617 15

3. Accuracy study with capillary sample Capillary blood measured with HemoCue Glucose 201 RT Analyzer versus mean values for YSI 2300 STAT Plus A total of 222 human capillary blood samples were collected and analysed (some capillary samples collected in EDTA-microtainer tubes were spiked with glucose standards to obtain high levels of glucose). The test users in this evaluation were assistant nurses at three different sites. The test user analysed the samples by using two HemoCue Glucose 201 RT Analyzers and in parallel the samples were analysed with a comparative method, YSI 2300 STAT Plus, in duplicate, by a professional user (figure 4 and figure 5). YSI, mean values (mmol/l) Glucose 201RT, 1st repl. (mg/dl) 500 400 300 200 0 5 10 15 20 25 30 Coefficients: y 0 =1.6919 ±1.804 a = 1.0069 ±0.011 r 2 = 0.9936 n = 222 A 30 25 20 15 10 Glucose 201RT, 1st repl. (mmol/l) 100 5 0 0 100 200 300 400 500 YSI, mean values (mg/dl) 0 Figure 4 Capillary blood measured at the Glucose 201 RT System versus mean values for YSI 2300 Stat Plus, mg/dl and mmol/l. Differences (Glucose 201RT, 1st repl. - YSI, mean values), mg/dl 0 5 10 15 20 25 100 50 0-50 -100 Figure 5 Mean of Glucose 201RT and YSI, mmol/l 0 100 200 300 400 500 Mean of Glucose 201RT and YSI, mg/dl Bland-Altman difference plot for Glucose 201 RT System versus YSI 2300 STAT Plus, mg/dl and mmol/l. 4 2 0-2 -4 Differences (Glucose 201RT, 1st repl. - YSI, mean values), mmol/l GPM316 130617 16

4. Accuracy study with venous and capillary samples Comparative study between the HemoCue Glucose 201 RT System and YSI 2300 STAT Plus Patient samples, reference method YSI 2300 STAT Plus The study was performed according to CLIA Guidance. A total of 400 human capillary and venous whole blood samples were collected and analysed. Some venous samples and some capillary samples collected in EDTA-microtainer tubes were spiked with glucose standards to obtain high levels of glucose. In total 60 samples were spiked. The test users in this evaluation were assistant nurses at the 4 sites. Before performing analysis the test users read the Instructions for Use, Quick Reference Guide and the Package Insert. The samples were analysed by the test user with one sample/analyzer, in total two HemoCue Glucose 201 RT Analyzers and in parallel the samples were analysed with a comparative method, YSI 2300 STAT Plus, in duplicate, by a professional user (Fig 6). YSI, mean values (mmol/l) 550 500 0 10 20 30 30 450 D B 400 E C A A Glucose 201RT, 1st repl. (mg/dl) 350 300 250 200 150 C B 20 10 Glucose 201RT, 1st repl. (mmol/l) 100 50 D 0 0 50 100 150 200 250 300 350 400 450 500 550 YSI, mean values (mg/dl) 0 Figure 6 Consensus grid for first replicate for Glucose 201 RT System versus mean values for YSI 2300 STAT Plus, mg/dl and mmol/l, venous and capillary samples (n = 400). The study was performed according to CLIA Guidance. The grid is divided into zones signifying the degree of risk posed by the incorrect measurement: Zone A no effect on clinical action Zone B altered clinical action little or no effect on clinical outcome Zone C altered clinical action likely to affect clinical outcome Zone D represents altered clinical action could have significant medical risk Zone E represents altered clinical action could have dangerous consequences GPM316 130617 17

5. Accuracy study with arterial sample HemoCue Glucose 201 RT System compared with ABL 825 FLEX Human arterial heparin blood samples, leftover samples from 140 patients at the intensive care unit (ICU) at Lund University Hospital. The study was performed according to CLSI EP9- A2. Each sample was analysed in single with the HemoCue Glucose 201 RT, in parallel with one single replicate with a comparative method, ABL 825 FLEX (Radiometer). The users in this evaluation were employees at the site, who are usually assigned to perform analysis of glucose samples with the ABL 825 FLEX (figure 7 and figure 8). ABL 825 FLEX, single replicate (mmol/l) 500 0 5 10 15 20 25 30 Coefficients: y 0 = 3.62 a = 1.040 r 2 = 0.971 n = 140 30 25 Glucose 201RT, 1st repl (mg/dl) 400 300 200 20 15 10 Glucose 201RT, 1st repl (mmol/l) 100 5 0 0 100 200 300 400 500 ABL 825 FLEX, single replicate (mg/dl) 0 Figure 7 First replicates for Glucose 201 RT System versus single replicate for ABL 825 FLEX, mg/dl and mmol/l. Bland-Altman difference plot for Glucose 201 RT compared with ABL 825 FLEX, arterial samples Mean values [ABL and Glucose 201 RT], mmol/l 0 5 10 15 20 25 100 Difference (Glucose 201RT - ABL), mg/dl 50 0-50 4 2 0-2 -4 Difference (Glucose 201RT - ABL), mmol/l -100 0 100 200 300 400 500 Mean values [ABL and Glucose 201 RT], mg/dl Figure 8 Bland-Altman difference plot for Glucose 201 RT System (first replicate) and ABL 825 FLEX versus mean of ABL and Glucose 201 RT, mg/dl and mmol/l. GPM316 130617 18

6. Accuracy and precision study with neonatal samples Clinical laboratory evaluation of neonatal samples Human venous EDTA whole blood or capillary heparin blood samples from 103 neonates in Linköping and Holbæk were collected and analyzed. The users in the study were employees at these sites, and usually assigned to perform glucose measurements with the comparative laboratory method. Since sufficient samples with glucose levels of less than or equal 55 mg/dl (plasma values) were not obtained from neonates, the study was complemented with 21 umbilical cord blood samples analyzed at HemoCue. The glucose levels in most of the samples were natural. However, some samples from the Linköping site were spiked with glucose solution in order to cover a larger glucose range. Each sample was analyzed in duplicate with the HemoCue Glucose 201 RT System, in parallel in single replicate with the comparative method ABL 835 FLEX/ID-GCMS/YSI 2300 STAT Plus. All values are recalculated and presented in plasma values (Table 5, Fig 9). The system should not be used on critically ill neonates in neonatal intensive care settings. Table 5 Glucose mmol/l (mg/dl) 0-11.1 (0-200) Mean value, precision, SD and CV for Glucose 201 RT for neonatal samples from non-neonatal intensive care units. No of samples Mean value Glucose 201 RT mmol/l (mg/dl) SD duplicates mmol/l (mg/dl) CV duplicat e % Mean value comparative method mmol/l (mg/dl) 124 4.9 (87.6) 0.24 (4.4) 5.1 4.7 (85.5) Comparative method (YSI, ABL, ID GC-MS), single, mmol/l 0 2 4 6 8 10 12 Glucose 201 RT, single, mg/dl 200 150 100 50 ABL 835 FLEX (neonatal patient samples) ID GC-MS (neonatal patient samples) YS 2300 STAT Plus (Umbilical cord samples) x=y 12 10 8 6 4 2 Glucose 201 RT, single, mmol/l 0 0 50 100 150 200 0 Comparative method (YSI, ABL, ID GC-MS), single, mg/dl Figure 9 Neonatal samples from non-neonatal intensive care units. Scatterplot of Glucose 201 RT first replicates versus comp. method first replicate (Plasma equivalent values), mg/dl and mmol/l. GPM316 130617 19

Publications Comparison of the HemoCue Glucose 201 DM RT System with Cobas Modular (Roche) Ref: Poster by Jukka Saarimies 1, Anja Rautajoki 2 Charlotta Roos 3 1) Hospital District of Southwest Finland, TYSKLAB, Department of Clinical Chemistry, Turku Finland. 2) HemoCue Finland, Espoo, Finland. 3) HemoCue AB Ängelholm, Sweden This study compared the HemoCue Glucose 201 RT System performance against two different lab systems; Glucose 201 DM RT vs. blood gas Cobas b 221(Roche) and vs. plasma base Cobas Modular (Figure 10), a hexokinase method. The study shows good agreement between the HemoCue Glucose 201 DM RT and the two comparative methods, measuring glucose over a wide measuring range. This indicates that the methods can be used interchangeably for clinical purposes within the healthcare settings. Evaluating the variation of lot-to-lot and instrument-to-instrument confirmed that the system is robust and the precision is high (Table 6). Figure 10 Comparison of the HemoCue Glucose 201 DM RT System with Cobas Modular (Roche). Analysis between glucose measured with the HemoCue Glucose 201 DM RT System, individual replicates, and glucose measured with the Cobas Modular, mean values. No. of replicates = 100. Table 6 Evaluation of lot to lot variation and instrument to instrument variation regarding the precision. Pooled SD was calculated as well as 95% confidence level (CI) for these values. Type of comparison Pooled SD 95% Cl for SD Cuvette lot a 0.20 0.17-0.24 Cuvette lot b 0.25 0.22 0.29 Instrument 1 0.24 0.21 0.28 Instrument 2 0.22 0.19 0.26 GPM316 130617 20

Comparison of the HemoCue Glucose 201 RT System with Architect ci8200 (Abbot) Ref: Poster by Eva Olofsson 1, Lena Piscator 2 Charlotta Roos 2 1) Avdelningen för klinisk kemi/kliniklärarenheten, Falu lasarett, Falun, Sverige 2) HemoCue AB Ängelholm, Sweden The study compares the HemoCue Glucose 201 RT System to Architect ci8200 (Abbot) which uses a hexokinase method. The populations studied was diverse, some objects had pronounced hypo- and hyperglycemia as noted from the glucose values. There is a very good agreement between the HemoCue Glucose 201 RT and Architect ci8200 in a broad range of glycaemia. This indicates that this method can be used interchangeably for clinical purpose within the healthcare settings. (Figure 12) Figure 12 Scatter plot between glucose measured with the HemoCue Glucose 201 RT System and glucose measured with the Architect ci8200, y = 0.921x + 0.4253, R2 = 0.9914, R = 0.9957, N=100. GPM316 130617 21

Comparison of the HemoCue Glucose 201 DM RT System with Vitros 5.1 (Ortho-Clinical Diagnostics) Ref: Poster by Ingvor Kasselstrand 1, Eva Drott Heideman 2 Elin Johansson 2 1) Klinisk kemi, Västerviks sjukhus, Sverige. 2) HemoCue AB Ängelholm, Sweden The study compared the HemoCue Glucose 201 DM RT System Vitros 5.1 (Ortho-Clinical Diagnostics) which uses a glucose oxidase method. Samples were analyzed in duplicates on both systems. There is a very good agreement between the HemoCue Glucose 201 DM RT and Vitros 5.1 when measuring glucose over a wide measuring range. (Figure 13) By providing a direct transfer of the results to the hospitals laboratory information system or medical record, the HemoCue Glucose DM RT reduces risk of pre- and post-analytical errors and improves the patient treatment. Figure 13 Analysis between glucose measured with the HemoCue Glucose 201 DM RT System, individual replicates, and glucose measured with Vitros 5.1, mean values. No. of replicates = 114. GPM316 130617 22

Comparison of the HemoCue Glucose 201 RT System with a hexokinase method. Ref: Dr ir. R.J. Slingerland (PhD) at Isala klinieken, Zwolle, Netherlands. The evaluation of the HemoCue Glucose 201 RT plasma equivalent analyzers was performed to compare the HemoCue Glucose 201 RT to a hexokinase method, plasma (Figure 14). There is a very good agreement between the HemoCue Glucose 201 RT and the hexokinase method in a broad range of glycaemia. The evaluation was also made to confirm the hematocrit range (Figure 15) as well as the lack of dependency of maltose (Figure 16). Figure 14 Method comparison in capillary whole blood compared with calculated plasma reference values. 55 capillary samples were measured. Figure 15 Hematocrit dependence in capillary whole blood. GPM316 130617 23

Figure 16 Maltose dependence in capillary whole blood GPM316 130617 24

Validation of a new generation POCT glucose device with emphasis on aspects important for glycemic control in the hospital care Kos, Snježana 1, van Meerkerk, Arie 1, van der Linden, Joke 2, Stiphout, Theo 1, Wulkan, Remi 1. 1 Central Laboratory of the Hospital (Maasstad Lab), Department of Clinical Chemistry, Maasstad Hospital, Rotterdam, The Netherlands 2 Department of Internal Medicine, Maasstad Hospital, Rotterdam, The Netherlands 2012 Ref; Kos, S et al: Validation of a new generation POCT glucose device with emphasis on aspects important for glycemic control in the hospital care, Clin Chem Lab Med 2012; 50 (9):1573-1580 This article in full is available for download from our website www.hemocue.com In this study a new generation of POCT glucose device, the HemoCue Glucose 201 DM RT has been evaluated for its suitability for (tight) glycemic control. This study was performed according to the CLSI/STARD criteria. The HemoCue Glucose 201 DM RT was compare to the laboratory hexokinase glucose method, Siemens Dimension Vista (Fig 17). The variation among different POC devices and microcuvette lot numbers was examined. Additionally, the influence of the partial pressure of oxygen and hematocrit on glucose measurement was investigated. Figure 17 Comparison experiments (A) Comparison of the HemoCue Glucose 201 DM RT device with the laboratory reference method. (B) Comparison of the HemoCue Glucose 201 DM RT device with the HemoCue Glucose 201 DM analyzer. (C) Device to device comparison. (D) Cuvette lot-to-lot comparison. (A) and (B) Presented by means of Bland Altman plot, while (C) and (D) are presented by means of Bland-Altman absolute bias plot due to a very small deviation expressed as percentage of the mean. The dashed lines in (C) and (D) indicate 5% deviation from the mean concentration over the whole tested concentration range. GPM316 130617 25

Results The HemoCue Glucose 201 DM RT showed a good agreement with the laboratory reference method. This was examined using Deming regression analysis, percentage Bland-Altman plot and a modified Clarke-error grid. The total analytical error at the clinically critical glucose concentrations of 5.6, 7.0 and 11.1 mmol/l (101, 126 and 200 mg/dl) was 6.4%, 4.3% and 3.0%, respectively. The total error among the different POCT devices and among different microcuvette lot numbers was <6.5%. Glucose measurements on the HemoCue Glucose 201 DM RT were not affected by changes in partial pressure of oxygen; whereas changes in hematocrit had influence on the results (3.4% for every 0.10 L/L change in hematocrit). Conclusion (extraction) The HemoCue Glucose 201 DM RT shows a very good agreement with the reference hexokinase method. The total analytical error at clinically critical glucose concentrations was < 6.5%. The results were not influenced by changes in partial oxygen pressure, although influenced by changes in hematocrit in a predictable fashion. GPM316 130617 26

HemoCue is a global leader in a field of diagnostics known as near patient, or pointof-care, testing. In 1982, HemoCue AB, based in Ängelholm, Sweden, introduced the first system making accurate hemoglobin testing possible in near-patient settings. Since then, more than 350 000 HemoCue Systems have been sold worldwide. The company also sells point-of-care tests for glucose, urine albumin, HbA1c, total and differential white blood cell count. HemoCue subsidiaries, affiliates, franchises, and third-party distributors generate revenue in more than 130 countries worldwide. HemoCue is owned by Radiometer, a Danaher company. Additional company information is available at www.hemocue.com GPM316 130617 27

GPM316 130617 28