GEM Premier 5000 (Measured Parameters: pH, pCO2 and pO2)

K160412 · Instrumentation Laboratory CO · CHL · Dec 14, 2016 · Clinical Chemistry

Device Facts

• Record ID: K160412
• Device Name: GEM Premier 5000 (Measured Parameters: pH, pCO2 and pO2)
• Applicant: Instrumentation Laboratory CO
• Product Code: CHL · Clinical Chemistry
• Decision Date: Dec 14, 2016
• Decision: SESE
• Submission Type: Traditional
• Regulation: 21 CFR 862.1120
• Device Class: Class 2

Indications for Use

The GEM Premier 5000 system is a portable critical care system for use by health care professionals to rapidly analyze heparinized whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory. The instrument provides quantitative measurements of pH and pO₂ from venous, arterial and capillary heparinized whole blood, as well as quantitative measurements of pCO₂ from venous and arterial heparinized whole blood. These parameters, along with derived parameters, aid in the diagnosis of a patient’s acid/base status. pH, pCO₂ and pO₂ measurements in whole blood are used in the diagnosis and treatment of life-threatening acid-base disturbances.

Device Story

Portable critical care system; analyzes heparinized whole blood samples (venous, arterial, capillary) for pH, pCO2, and pO2. System comprises GEM Premier 5000 analyzer and disposable GEM Premier 5000 PAK cartridge. Cartridge contains sensors, reagents, CO-Ox optical cell, and process control solutions. Uses potentiometry for pH/pCO2 and amperometry for pO2. Operated by clinicians/technicians at point-of-care or central labs. Provides quantitative measurements and derived parameters (e.g., bicarbonate, oxygen saturation) to aid diagnosis of acid-base status. Results used by healthcare providers for rapid clinical decision-making in life-threatening conditions.

Clinical Evidence

Bench testing only. Performance validated via precision studies (internal and POC settings, n=120 replicates per level), linearity studies (CLSI EP06-A), interference studies (CLSI EP07-A2), and altitude studies (up to 10,000 ft). Method comparison against predicate (GEM Premier 4000) performed at POC sites using native whole blood (n=479-506 for syringe mode; n=167-171 for capillary mode). Results showed high correlation (r² ≥ 0.975) and acceptable bias across clinical decision levels.

Technological Characteristics

Portable analyzer with color touch screen. Disposable multi-use GEM PAK cartridge houses sensors, CO-Ox optical cell, process control solutions, sampler, pump, and waste bag. Measurement principles: Potentiometry (pH, pCO2) and Amperometry (pO2). Connectivity: EEPROM on PAK stores solution values/test counts. Operating system: Linux-based. Calibration: 2-point. System operating temperature: 12-32°C.

Indications for Use

Indicated for health care professionals to analyze heparinized whole blood (venous, arterial, capillary) for pH, pO2, and pCO2 (venous/arterial only) to aid in diagnosis/treatment of acid-base disturbances in clinical/laboratory settings.

Regulatory Classification

Identification

A blood gases (PCO2 , PO2 ) and blood pH test system is a device intended to measure certain gases in blood, serum, plasma or pH of blood, serum, and plasma. Measurements of blood gases (PCO2 , PO2 ) and blood pH are used in the diagnosis and treatment of life-threatening acid-base disturbances.

Predicate Devices

• GEM® Premier 4000 for pH, pCO2, and pO2 (k133407)

Submission Summary (Full Text)

{0}

1

# 510(k) SUBSTANTIAL EQUIVALENCE DETERMINATION DECISION SUMMARY ASSAY ONLY TEMPLATE

A. 510(k) Number:
k160412

B. Purpose for Submission:
New device

C. Measurand:
pH, pCO₂, and pO₂ in whole blood

D. Type of Test:
Potentiometry for pH and pCO₂
Amperometry for pO₂

E. Applicant:
Instrumentation Laboratory Co.

F. Proprietary and Established Names:
GEM® Premier 5000 (Measured Parameters: pH, pCO₂, and pO₂)

G. Regulatory Information:
1. Regulation section:
21 CFR § 862.1120 Blood gases (pCO₂ and pO₂) and blood pH test system.
2. Classification:
Class II
3. Product code:
CHL

{1}

4. Panel:

Chemistry (75)

H. Intended Use:

1. Intended use(s):

See indications for use below.

2. Indication(s) for use:

The GEM Premier 5000 system is a portable critical care system for use by health care professionals to rapidly analyze heparinized whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory. The instrument provides quantitative measurements of pH and pO₂ from venous, arterial and capillary heparinized whole blood, as well as quantitative measurements of pCO₂ from venous and arterial heparinized whole blood. These parameters, along with derived parameters, aid in the diagnosis of a patient’s acid/base status.

pH, pCO₂ and pO₂ measurements in whole blood are used in the diagnosis and treatment of life-threatening acid-base disturbances.

3. Special conditions for use statement(s):

For prescription use only.

For clinical laboratory and point-of-care use.

4. Special instrument requirements:

GEM Premier 5000 analyzer

I. Device Description:

The GEM Premier 5000 system contains two key components: the GEM Premier 5000 analyzer and the GEM Premier 5000 PAK (cartridge):

GEM Premier 5000 PAK

The GEM® Premier 5000 PAK is a disposable, multi-assay cartridge which includes measurements of pH, pCO₂ and pO₂. The cartridge is used exclusively with the GEM Premier 5000 analyzer. The cartridge houses all required components to test heparin whole blood samples for pH, pCO₂ and pO₂. The components include the sensors, reagents, CO-Ox optical cell, Process Control Solutions, sampler, pump tubing, distribution valve, and waste bag. The cartridge contains sufficient materials to operate the analyzer for 75 to 600 tests.

2

{2}

The five Process Control Solutions (A, B, C, D, and E) contained on the GEM Premier 5000 PAK are control materials and calibration verifiers. These materials are tonometered to specific levels of $\mathrm{pO}_2$ and $\mathrm{pCO}_2$ , and contain known pH values tested using CLSI and NIST traceable reference standards. The solutions are sealed in gas impermeable bags with no headspace.

## GEM Premier 5000 analyzer

The GEM Premier 5000 analyzer is a portable critical care system for use by health care professionals to rapidly analyze whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory.

## J. Substantial Equivalence Information:

1. Predicate device name(s):

GEM® Premier 4000 for pH, $\mathrm{pCO}_2$ , and $\mathrm{pO}_2$

2. Predicate 510(k) number(s):

k133407

3. Comparison with predicate:

| Similarities | | |
| --- | --- | --- |
| Item | Candidate Device k160412 GEM Premier 5000 for pH, pCO2, and pO2 | Predicate k133407 GEM Premier 4000 for pH, pCO2, and pO2 |
| Intended use | The GEM Premier 5000 is a portable critical care system for use by health care professionals to rapidly analyze whole blood samples at the point of health care delivery in a clinical setting and in a central laboratory. The instrument provides quantitative measurements of pH, pCO2, and pO2. | Same |
| Intended user | Central Laboratory and Point-of-Care sites | Same |
| Type of Measurement | Potentiometry: pH and pCO2 | Same |

{3}

| Similarities | | |
| --- | --- | --- |
| | Amperometry: pO2 | |
| Sample type | Lithium heparin whole blood | Same |
| Sampling devices | Heparinized syringe and heparinized capillary tube | Same |
| Measuring modes and sample volumes | Normal Mode for syringe: 150 μL | Same |
| | Normal Mode for capillary tube: 150 μL | |
| | tBili/CO-Ox for capillary tube: 100 μL | |
| | Micro Mode for capillary tube: 65 μL | |
| Time to analysis | Analyze samples within 15 minutes from draw. | Same |
| Calibration | 2-point calibration | Same |
| Differences | | |
| --- | --- | --- |
| Item | Candidate Device k160412 GEM Premier 5000 for pH, pCO2, and pO2 | Predicate k133407 GEM Premier 4000 for pH, pCO2, and pO2 |
| Measurement ranges | pH 7.00 to 7.92 | pH 7.00 to 8.00 |
| | pCO2 6 to 125 mmHg | pCO2 6 to 125 mmHg |
| | pO2 6 to 690 mmHg | pO2 5 to 690 mmHg |
| Specimens | pH: Arterial, capillary, and venous | pH: Arterial, capillary, and venous |
| | pCO2: Arterial and venous | pCO2: Arterial, capillary, and venous |
| | pO2: Arterial, capillary, and venous | pO2: Arterial, capillary, and venous |

# K. Standard/Guidance Document Referenced (if applicable):

CLSI EP05-A3: Evaluation of Precision of Quantitative Measurement Procedures; Approved Guideline, 3rd Edition.

{4}

CLSI EP06-A: Evaluation of Linearity of Quantitative Measurement Procedures; A Statistical Approach; Approved Guideline. First Edition.

CLSI EP07-A2. Interference Testing in Clinical Chemistry; Approved Guideline. Second Edition.

CLSI EP09-A3 Measurement Procedure Comparison and Bias Estimation Using Patient Samples; Approved Guideline.

CLSI EP17-A2: Evaluation of Detection Capability for Clinical Laboratory Measurement Procedures; Approved Guideline. Second Edition.

CLSI EP25-A: Evaluation of Stability of In Vitro Diagnostic Reagents; Approved Guideline.

## L. Test Principle:

### Assay principles:

pH
pH is measured using an ion selective electrode on the GEM Premier 5000 PAK. The result may be corrected for actual patient temperature if different than 37°C.

pCO₂
pCO₂ is measured by potentiometry using a sensor on the GEM Premier 5000 PAK. The result may be corrected for actual patient temperature if different than 37°C.

pO₂
pO₂ is measured amperometrically using a three electrode cell configuration with platinum working electrode. The result may be corrected for actual patient temperature if different than 37°C.

### Derived parameters:

The following are derived parameters which utilize measured pH, pCO₂, and pO₂ values:

Total Carbon Dioxide, Base Excess of Extracellular Fluid, Base Excess of Blood, Arterial partial pressure / inspired oxygen ratio, Alveolar oxygen partial pressure, Arterial oxygen content, Oxygen content, Partial pressure of O₂ in a hemoglobin solution, Alveolar-arterial oxygen gradient, Arterial-alveolar oxygen ratio, Respiratory index, End pulmonary capillary oxygen content, Arterial-mixed venous oxygen gradient mixed venous oxygen content, Estimated shunt, Physiological shunt, Actual bicarbonate, and Calculated oxygen saturation.

5

{5}

M. Performance Characteristics (if/when applicable):

1. Analytical performance:

a. Precision/Reproducibility:

Four separate precision studies were conducted to evaluate the precision of the GEM Premier 5000 system for pH, pCO₂, and pO₂ and were conducted according to CLSI EP05-A3 guidance document.

#1 Internal precision study using control material – GEM System Evaluators

An internal precision study was performed on the GEM Premier 5000 for pH, pCO₂, and pO₂ using GEM System Evaluators 1, 2 and 3 control materials.

The GEM System Evaluators 1, 2, 3 (previously cleared in k093623) are ampules of three levels of assayed quality control material intended for evaluating performance characteristics of pH, pCO₂, pO₂, and other analytes on the GEM Premier 5000 analyzer.

Each of the control levels was run on three GEM Premier 5000 analyzers for 20 days, with two runs per day and one replicate measured per run per level for a total of n=120 replicates. The samples were introduced into the analyzer using the QC sample drawn mode. The results are given as follows:

| Analyte | Level | Mean | N | Within Analyzer SD | Within Analyzer %CV | Total SD | Total %CV |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pH | 1 | 7.14 | 120 | 0.008 | 0.1% | 0.008 | 0.1% |
| | 2 | 7.38 | 120 | 0.004 | 0.1% | 0.006 | 0.1% |
| | 3 | 7.57 | 120 | 0.003 | 0.0% | 0.003 | 0.0% |
| pCO₂ (mmHg) | 1 | 87 | 120 | 2.3 | 2.7% | 2.3 | 2.7% |
| | 2 | 35 | 120 | 0.7 | 1.9% | 0.8 | 2.3% |
| | 3 | 14 | 120 | 0.3 | 2.2% | 0.3 | 2.3% |
| pO₂ (mmHg) | 1 | 31 | 120 | 1.9 | 6.1% | 2.4 | 7.9% |
| | 2 | 88 | 120 | 1.2 | 1.4% | 2.2 | 2.4% |
| | 3 | 370 | 120 | 4.8 | 1.3% | 5.9 | 1.6% |

#2 Internal precision study using whole blood samples

Note: Capillary not claimed for pCO₂ on the GEM Premier 5000.

An internal precision study was performed on the GEM Premier 5000 for pH, pCO₂, and pO₂ using lithium heparin whole blood samples.

{6}

Donor venous blood was collected in balanced lithium heparinized blood collection tubes. For the $\mathrm{pO}_2$ and $\mathrm{pCO}_2$ , five levels were prepared by tonometrization. For pH, whole blood was adjusted to the five levels using carbonate.

For pH and the $\mathrm{pO_2}$ each sample was tested in two sampling modes: normal using 150 $\mu$ L with syringe and micro using $65~\mu \mathrm{L}$ with capillary tube. For $\mathrm{pCO_2}$ each sample was tested in normal using $150~\mu \mathrm{L}$ with syringe. The samples were run on three GEM Premier 5000 analyzers per sample mode for five days, with one run per day and 8 replicates measured per run per level for a total of 120 replicates per level.

Day-to-day variability was not assessed due to the in vitro instability of whole blood. Fresh blood samples at each level were prepared daily. The results are given as follows:

| Analyte | Mode | Level | Mean | N | Within Run SD | Within Run %CV | Total SD* | Total %CV* |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| pH | Normal | 1 | 7.11 | 120 | 0.004 | 0.1% | 0.005 | 0.1% |
| | | 2 | 7.33 | 120 | 0.007 | 0.1% | 0.007 | 0.1% |
| | | 3 | 7.35 | 120 | 0.004 | 0.1% | 0.005 | 0.1% |
| | | 4 | 7.42 | 120 | 0.005 | 0.1% | 0.006 | 0.1% |
| | | 5 | 7.68 | 120 | 0.012 | 0.1% | 0.014 | 0.2% |
| | Micro | 1 | 7.10 | 120 | 0.006 | 0.1% | 0.007 | 0.1% |
| | | 2 | 7.32 | 120 | 0.004 | 0.1% | 0.006 | 0.1% |
| | | 3 | 7.35 | 120 | 0.004 | 0.1% | 0.005 | 0.1% |
| | | 4 | 7.41 | 120 | 0.005 | 0.1% | 0.006 | 0.1% |
| | | 5 | 7.67 | 120 | 0.013 | 0.2% | 0.015 | 0.2% |
| pCO2** (mmHg) | Normal | 1 | 112 | 120 | 2.7 | 2.4% | 2.9 | 2.3% |
| | | 2 | 70 | 120 | 1.1 | 1.5% | 1.5 | 2.0% |
| | | 3 | 50 | 120 | 0.6 | 1.2% | 0.8 | 1.5% |
| | | 4 | 36 | 120 | 0.5 | 1.3% | 0.8 | 1.9% |
| | | 5 | 10 | 120 | 0.5 | 4.6% | 0.8 | 7.7% |
| pO2 (mmHg) | Normal | 1 | 32 | 120 | 0.4 | 1.2% | 0.9 | 2.7% |
| | | 2 | 62 | 120 | 0.7 | 1.1% | 0.9 | 1.2% |
| | | 3 | 204 | 120 | 2.3 | 1.1% | 3.1 | 1.4% |
| | | 4 | 415 | 120 | 8.6 | 2.1% | 13.0 | 2.8% |
| | | 5 | 722 | 120 | 18.6 | 2.6% | 32.7 | 4.3% |
| | Micro | 1 | 31 | 120 | 0.9 | 3.0% | 1.6 | 5.3% |
| | | 2 | 62 | 120 | 0.7 | 1.1% | 0.9 | 1.4% |
| | | 3 | 204 | 120 | 4.3 | 2.1% | 5.5 | 2.5% |
| | | 4 | 402 | 120 | 15.3 | 3.8% | 16.4 | 3.8% |
| | | 5 | 693 | 120 | 26.5 | 3.8% | 34.0 | 4.8% |

* The day-to-day contribution was excluded in total precision evaluation for whole blood samples since different whole blood samples were prepared each day.

{7}

** The candidate device is not intended for the measurement of $\mathrm{pCO_2}$ in capillary samples; therefore, only $\mathrm{pCO_2}$ data from syringe samples tested under normal mode is shown in the table.

# 3 Point of Care setting using control material - GEM System Evaluators

A reproducibility study at three point of care sites in hospital settings was performed on the GEM Premier 5000 for pH, $\mathrm{pCO_2}$ , and $\mathrm{pO_2}$ using the GEM System Evaluators 1, 2 and 3 control material.

The studies were conducted by 9 different operators (perfusionists and respiratory therapists) on three different GEM Premier 5000 instruments using a single lot of GEM Premier 5000 PAKs. Each site used the same lot of GEM System Evaluators and running each control level in triplicate, twice a day for 5 days, for a total of 30 replicates per level per site. The samples were introduced into the analyzer using the QC sample drawn mode. The results are given as follows:

Point of care site 1:

| Analyte | Level | N | Mean | Repeatability | | Within Site | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | | | SD | %CV | SD | %CV |
| pH | 1 | 30 | 7.14 | 0.003 | 0.0% | 0.003 | 0.0% |
| | 2 | 30 | 7.39 | 0.005 | 0.1% | 0.007 | 0.1% |
| | 3 | 30 | 7.57 | 0.000 | 0.0% | 0.000 | 0.0% |
| pCO2(mmHg) | 1 | 30 | 87 | 0.6 | 0.7% | 1.1 | 1.3% |
| | 2 | 30 | 34 | 0.7 | 1.9% | 0.7 | 2.0% |
| | 3 | 30 | 13 | 0.4 | 3.1% | 0.4 | 3.3% |
| pO2(mmHg) | 1 | 30 | 30 | 0.8 | 2.6% | 1.0 | 3.3% |
| | 2 | 30 | 87 | 1.5 | 1.7% | 1.6 | 1.8% |
| | 3 | 30 | 351 | 8.2 | 2.3% | 8.6 | 2.5% |

Point of care site 2:

| Analyte | Level | N | Mean | Repeatability | | Within Site | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | | | SD | %CV | SD | %CV |
| pH | 1 | 30 | 7.14 | 0.002 | 0.0% | 0.003 | 0.0% |
| | 2 | 30 | 7.39 | 0.010 | 0.1% | 0.011 | 0.1% |
| | 3 | 30 | 7.57 | 0.002 | 0.0% | 0.005 | 0.1% |
| pCO2(mmHg) | 1 | 30 | 87 | 0.2 | 0.2% | 1.2 | 1.4% |
| | 2 | 30 | 34 | 1.2 | 3.5% | 1.3 | 3.8% |
| | 3 | 30 | 13 | 0.4 | 3.0% | 0.4 | 3.0% |
| pO2(mmHg) | 1 | 30 | 28 | 0.6 | 2.1% | 1.0 | 3.6% |
| | 2 | 30 | 87 | 1.7 | 2.0% | 1.7 | 2.0% |
| | 3 | 30 | 360 | 9.1 | 2.5% | 9.0 | 2.5% |

{8}

Point of care site 3:

| Analyte | Level | N | Mean | Repeatability | | Within Site | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | | | SD | %CV | SD | %CV |
| pH | 1 | 30 | 7.14 | 0.000 | 0.0% | 0.000 | 0.0% |
| | 2 | 30 | 7.38 | 0.003 | 0.0% | 0.005 | 0.1% |
| | 3 | 30 | 7.57 | 0.003 | 0.0% | 0.003 | 0.0% |
| pCO2 (mmHg) | 1 | 30 | 87 | 0.4 | 0.5% | 0.8 | 1.0% |
| | 2 | 30 | 34 | 0.5 | 1.5% | 0.6 | 1.6% |
| | 3 | 30 | 13 | 0.5 | 3.6% | 0.5 | 3.9% |
| pO2 (mmHg) | 1 | 30 | 27 | 0.6 | 2.1% | 1.0 | 3.6% |
| | 2 | 30 | 85 | 1.2 | 1.4% | 1.3 | 1.6% |
| | 3 | 30 | 359 | 6.3 | 1.7% | 6.5 | 1.8% |

All 3 point of care sites combined:

| Analyte | Level | N | Mean | Repeatability | | Reproducibility | |
| --- | --- | --- | --- | --- | --- | --- | --- |
| | | | | SD | %CV | SD | %CV |
| pH | 1 | 90 | 7.14 | 0.002 | 0.0% | 0.002 | 0.0% |
| | 2 | 90 | 7.39 | 0.007 | 0.1% | 0.008 | 0.1% |
| | 3 | 90 | 7.57 | 0.002 | 0.0% | 0.003 | 0.0% |
| pCO2 (mmHg) | 1 | 90 | 87 | 0.4 | 0.5% | 1.1 | 1.2% |
| | 2 | 90 | 34 | 0.9 | 2.5% | 0.9 | 2.7% |
| | 3 | 90 | 13 | 0.4 | 3.3% | 0.5 | 3.5% |
| pO2 (mmHg) | 1 | 90 | 28 | 0.7 | 2.3% | 1.9 | 6.7% |
| | 2 | 90 | 87 | 1.4 | 1.7% | 1.9 | 2.2% |
| | 3 | 90 | 357 | 7.9 | 2.2% | 9.3 | 2.6% |

# #4 Point of Care setting using whole blood samples

Note: Capillary not claimed for pCO₂ on the GEM Premier 5000.

A precision study at two central laboratories, one company internal customer simulation laboratory, and three point of care sites in hospital settings was performed on the GEM Premier 5000 for pH, pCO₂, and pO₂ using lithium heparinized whole blood patient samples.

All samples were arterial or venous native whole blood patient samples from individual donors. At least two whole blood specimens were analyzed in triplicate daily for 5 days in two sampling modes: normal using 150 μL with syringe and micro using 65 μL with capillary tube. A new specimen was used for each sample mode.

For the central laboratory setting, the studies were performed by 3 operators on 3 GEM Premier 5000 instruments using a single lot of GEM Premier 5000 PAK

{9}

(cartridge). For the POC Setting, the studies were performed by 11 operators on 3 GEM Premier 5000 instruments, using a single lot of GEM Premier 5000 PAK.

Reproducibility was not assessed due to the in vitro instability of whole blood. Fresh blood samples were collected daily. The repeatability results are given as follows:

| Analyte | Sample Mode | Site | N | Mean | Min | Max | Within sample SD |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pH | Normal Mode | POC1 | 54 | 7.36 | 7.21 | 7.47 | 0.008 |
| | | POC2 | 42 | 7.33 | 7.14 | 7.43 | 0.009 |
| | | POC3 | 30 | 7.35 | 7.06 | 7.51 | 0.008 |
| | | POC-All | 126 | 7.35 | 7.06 | 7.51 | 0.008 |
| | | CSL | 30 | 7.32 | 7.28 | 7.39 | 0.009 |
| | | Lab1 | 30 | 7.36 | 7.20 | 7.44 | 0.009 |
| | | Lab2 | 30 | 7.32 | 7.24 | 7.44 | 0.007 |
| | | Lab-All | 90 | 7.33 | 7.20 | 7.44 | 0.008 |
| | Micro Mode | POC1 | 30 | 7.29 | 7.17 | 7.42 | 0.009 |
| | | POC2 | 36 | 7.32 | 7.18 | 7.45 | 0.012 |
| | | POC3 | 36 | 7.28 | 6.97 | 7.45 | 0.012 |
| | | POC-All | 102 | 7.30 | 6.97 | 7.45 | 0.011 |
| | | CSL | 30 | 7.30 | 7.26 | 7.37 | 0.008 |
| | | Lab1 | 30 | 7.33 | 7.21 | 7.48 | 0.007 |
| | | Lab2 | 30 | 7.26 | 6.94 | 7.46 | 0.009 |
| | | Lab-All | 90 | 7.30 | 6.94 | 7.48 | 0.008 |
| Analyte | Sample Mode | Site | N | Mean | Min | Max | Within sample SD |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pCO2* mmHg | Normal Mode | POC1 | 48 | 43 | 33 | 54 | 1.1 |
| | | POC2 | 39 | 42 | 28 | 56 | 0.8 |
| | | POC3 | 30 | 49 | 40 | 61 | 1.7 |
| | | POC-All | 117 | 44 | 28 | 61 | 1.2 |
| | | CSL | 24 | 53 | 48 | 61 | 1.3 |
| | | Lab1 | 30 | 46 | 40 | 58 | 1.3 |
| | | Lab2 | 24 | 46 | 28 | 56 | 0.8 |
| | | Lab-All | 78 | 48 | 28 | 61 | 1.2 |

{10}

| Analyte | Sample Mode | Site | N | Mean | Min | Max | Within sample SD or CV% |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pO2(mmHg) | Normal Mode | POC1 | 27 | 52 | 35 | 85 | 1.5 |
| | | POC2 | 12 | 63 | 55 | 70 | 0.4 |
| | | POC3 | 12 | 71 | 51 | 86 | 2.0 |
| | | POC-All | 51 | 59 | 35 | 86 | 1.5 |
| | | CSL | 30 | 52 | 32 | 79 | 0.6 |
| | | Lab1 | 21 | 55 | 42 | 85 | 0.8 |
| | | Lab2 | 15 | 50 | 26 | 95 | 3.9 |
| | | Lab-All | 66 | 53 | 26 | 95 | 0.8 |
| | | POC1 | 27 | 147 | 82 | 191 | 6.0% |
| | | POC2 | 30 | 175 | 97 | 402 | 2.6% |
| | | POC3 | 18 | 154 | 103 | 218 | 3.4% |
| | | POC-All | 75 | 160 | 82 | 402 | 4.0% |
| | | Lab1 | 9 | 110 | 92 | 130 | 1.1% |
| | | Lab2 | 15 | 145 | 112 | 195 | 2.1% |
| | | Lab-All | 24 | 132 | 92 | 195 | 1.7% |
| pO2(mmHg) | Micro Mode | POC1 | 21 | 53 | 46 | 65 | 1.5 |
| | | POC2 | 6 | 45 | 33 | 59 | 1.2 |
| | | POC3 | 18 | 58 | 43 | 82 | 1.1 |
| | | POC-All | 45 | 54 | 33 | 82 | 1.3 |
| | | CSL | 30 | 54 | 31 | 83 | 0.8 |
| | | Lab1 | 24 | 60 | 40 | 89 | 1.0 |
| | | Lab2 | 18 | 52 | 39 | 67 | 1.1 |
| | | Lab-All | 72 | 56 | 31 | 89 | 1.0 |
| | | POC1 | 9 | 130 | 94 | 186 | 2.3% |
| | | POC2 | 30 | 216 | 90 | 413 | 4.4% |
| | | POC3 | 18 | 135 | 83 | 199 | 4.2% |
| | | POC-All | 57 | 177 | 83 | 413 | 4.0% |
| | | Lab1 | 6 | 124 | 119.0 | 128.0 | 3.3% |
| | | Lab2 | 12 | 149 | 100.0 | 214.0 | 2.6% |
| | | Lab-All | 18 | 141 | 100 | 214 | 2.8% |

* The candidate device is not intended for the measurement of $\mathrm{pCO}_2$ in capillary samples; therefore, only $\mathrm{pCO}_2$ data from syringe samples tested under normal mode is shown in the table.

# b. Linearity/assay reportable range:

The linearity studies were performed on GEM Premier $5000\mathrm{pH}$ , $\mathrm{pO}_2$ and $\mathrm{pCO}_2$ following CLSI EP06-A guidance.

{11}

To challenge the measuring range, pH was tested for linearity at 8 levels, and $\mathrm{pO_2}$ and $\mathrm{pCO_2}$ were tested at 9 levels. The samples were venous whole blood from healthy volunteers that were adjusted to different levels by addition of carbonate for pH samples, and tonometrization for $\mathrm{pO_2}$ and $\mathrm{pCO_2}$ samples. The concentrations for pH were assigned using the Radiometer ABL 837 as a reference device. For $\mathrm{pO_2}$ and $\mathrm{pCO_2}$ the concentrations were assigned by tonometry using commercially available gases.

The whole blood concentration pools were adjusted so that one pool exceeded the upper end of the range by $20 - 30\%$ and one pool exceeded the lower end of the range by $20 - 30\%$ , wherever possible. For $\mathsf{pH}$ , the upper end of the reference analyzer measuring range was the same as the anticipated GEM Premier 5000 claimed measuring range, thus the sample level exceeding the upper limit was omitted. For $\mathsf{pO}_2$ , test samples could not be adjusted beyond $756~\mathrm{mmHg}$ under ambient atmospheric conditions, thus the upper end of the claimed measuring range was not exceeded. Each blood sample was analyzed in triplicate on three GEM Premier 5000 analyzers and results compared to the reference measurements.

For pH, regression analysis found a statistically significant coefficient for a second order polynomial fit which indicated possible non-linearity. Linearity was further assessed by comparing the deviation of the regression line from linearity. At each level, the deviation from linearity was less than $\pm 0.04$ pH units.

For $\mathrm{pCO_2}$ , regression analysis found a statistically significant coefficient for a third order polynomial fit which indicated possible non-linearity. Linearity was further assessed by comparing the deviation of the regression line from linearity. At each level, the deviation from linearity was less than $\pm 5\mathrm{mmHg}$ .

For $\mathrm{pO}_2$ , regression analysis found that the coefficients for a second or third order polynomial fit were not statistically significant. Therefore, the $\mathrm{pO}_2$ results demonstrate a satisfactory linear response over the measuring range.

The linear regression results are given in the table below:

| Analyte | Slope | Intercept | R2 | Tested Range | GEM Premier 5000 Claimed Measuring Range |
| --- | --- | --- | --- | --- | --- |
| pH | 0.972 | 0.191 | 0.998 | 6.67 to 7.97 | 7.00 to 7.92 |
| pCO2(mmHg) | 1.045 | -2.027 | 0.998 | 1 to 149 | 6 to 125 |
| pO2(mmHg) | 1.028 | -4.069 | 0.995 | 5 to 727 | 6 to 690 |

Based on these results, the sponsor concluded that the linearity data support the claimed measuring range for $\mathsf{pH}$ , $\mathsf{pCO}_2$ , and $\mathsf{pO}_2$ .

{12}

c. Traceability, Stability, Expected values (controls, calibrators, or methods):

Traceability:
The GEM Premier 5000 PAK measurement for pCO₂ is traceable to tonometry at 37°C using NIST traceable gas mixtures.

The GEM Premier 5000 PAK measurement for pO₂ is traceable to tonometry at 37°C using NIST traceable gas mixtures.

The GEM Premier 5000 PAK measurement for pH is traceable to NIST Standard Reference Material 186g.

Stability:
Stability studies were performed with GEM Premier 5000 PAK for pH, pCO₂, and pO₂. The protocols for stability and acceptance criteria were reviewed and found to be adequate.

Shelf life stability studies demonstrated that GEM Premier 5000 PAK is stable for 180 days when stored in the claimed range of 15-25°C. The on-board stability of the GEM Premier 5000 PAK is 31 days or for 600 sample usage. The cartridges are stable when exposed to transport conditions for days at 10-38°C and up to 10,000 feet altitude.

d. Detection limit:

Linearity studies were used to support the lower end of the measuring range for pH, pCO₂, and pO₂.

e. Analytical specificity:

In accordance with EP07-A2, an interference study was conducted on the GEM Premier 5000 for pH, pCO₂, and pO₂.

Heparinized whole blood from healthy volunteers was collected daily for the samples. With these samples, an interference screening test was conducted at 2 levels of pH, pCO₂, and pO₂. At each level, a potential interfering substance was spiked at a worse case high concentration. The substances tested represent anesthetics and medications often prescribed for critically ill patients and endogenous substances. Interference was calculated as the bias between the average test result with interfering substance and average control measurement across the three GEM Premier 5000 analyzers. The control sample was prepared by spiking diluent only.

The sponsor stated that interference was considered to be non-significant if the bias between the test and control sample was within the following criteria:

13

{13}

| Analyte | Unit | Significant interference criteria, within: |
| --- | --- | --- |
| pH | | ±0.02 |
| pCO2 | mmHg | ±8% |
| pO2 | mmHg | ±10% |

No significant interference was observed with the substances tested on pH, $\mathrm{pCO}_2$ , and $\mathrm{pO}_2$ . The table below lists substances that were tested:

Endogenous substances:

| Substance | Highest conc. tested without significant interference in pH, pO2, and pCO2 |
| --- | --- |
| Albumin (Human) | 60 g/L |
| Bilirubin | 20 mg/dL |
| Hemoglobin (Hemolysis) | 2000 mg/dL (20%) |
| Triglycerides (as INTRALIPID® 20%) | 4012 mg/dL |

Exogenous substances:

| Substance | Highest conc. tested without significant interference in pH, pO2, and pCO2 |
| --- | --- |
| Acetaminophen | 1324 μmol/L |
| Amoxicillin | 206 μmol/L |
| Aprotinin | 50 mg/L |
| Atracurium | 50 mg/L |
| Benzalkonium (Chloride) | 5 mg/L |
| Ceftriaxone | 1460 μmol/L |
| Ciprofloxin | 30.2 μmol/L |
| Diazepam | 18 μmol/L |
| Epinephrine | 0.5 μmol/L |
| Ethanol | 86.8 mmol/L |
| Etomidate | 50 mg/L |
| Fentanyl | 0.02 μg/ml |
| Gentamycin | 21 μmol/L |
| Halothane | 759 μmol/L |
| Lithium (Chloride) | 3.2 mmol/L |
| Methadone | 6.46 μmol/L |
| Midazolam | 0.5 μg/mL |
| Morphine | 1.75 μmol/L |
| Omeprazole | 17.4 μmol/L |
| Phenobarbital | 431 μmol/L |
| Propofol | 0.05 mg/mL |
| Suxamethonium | 68 μmol/L |
| Thiopental | 248 μmol/L |
| Thyroxine | 1.29 μmol/L |
| Tetracyclopentadiene | 0.02 μg/mL |
| Toluidine | 0.02 μg/mL |
| Toluidine | 0.02 μg/mL |

{14}

# Hematocrit:

The sponsor conducted a screening interference study to demonstrate that changes in hematocrit do not affect the pH, $\mathrm{pCO_2}$ , and $\mathrm{pO_2}$ measurements. Whole blood samples at two levels of hematocrit (25% and 75%) representing the lowest and highest hematocrit levels were tested at two levels for each of the blood gases and compared to the same samples with normal hematocrit. The results demonstrated bias less than the significant interference criteria (see table above). Based on the hematocrit study results, the sponsor concluded that hematocrit levels between 25% to 75% do not significantly interfere with the pH, $\mathrm{pCO_2}$ , and $\mathrm{pO_2}$ measurements.

# Altitude study:

A study was conducted to demonstrate that altitudes between sea level and 10,000 feet (3050 meters) have no significant effect on the GEM Premier 5000 for pH, $\mathrm{pCO_2}$ and $\mathrm{pO_2}$ measurements. Two sites were selected with altitudes of 10,152 feet (high altitude site) and 135 feet (low altitude site). At each site, samples were prepared from heparinized whole blood using gas mixtures to create different pH and blood gas levels. Since identical analyte levels could not be achieved at each of the two sites, the altitude study analysis consisted of: (1) fitting a linear regression line to the values obtain with the six levels for each site, (2) deriving the measurement bias between the two sites by calculating the difference between the regression lines at target levels for pH, $\mathrm{pO_2}$ , and $\mathrm{pCO_2}$ , and (3) comparing the measurement bias to the significant interference criteria (see table above).

| Target Levels | | |
| --- | --- | --- |
| pH | pO2 | pCO2 |
| 7.00 | 10 | 10 |
| 7.20 | 30 | 35 |
| 7.30 | 45 | 50 |
| 7.35 | 60 | 70 |
| 7.45 | 200 | 100 |
| 7.70 | 450 | 125 |

The concentration of each sample was assigned based on the following:

| Analyte | Reference Method |
| --- | --- |
| pH | GEM Premier 3000 |
| pCO2 | Tonometry |
| pO2 | Tonometry |

Three replicates of heparinized whole blood were run per analyte level on the GEM Premier 5000 and the reference methods at each site. The results are as follows:

{15}

| pH | | | |
| --- | --- | --- | --- |
| Target | Low Altitude result | High Altitude result | High Altitude Bias at the target level |
| 7.00 | 7.01 | 7.00 | -0.01 |
| 7.20 | 7.20 | 7.20 | 0.00 |
| 7.30 | 7.30 | 7.30 | 0.00 |
| 7.35 | 7.34 | 7.35 | 0.00 |
| 7.45 | 7.44 | 7.44 | 0.00 |
| 7.70 | 7.68 | 7.69 | 0.01 |
| pCO2 | | | |
| --- | --- | --- | --- |
| Target (mmHg) | Low Altitude result | High Altitude result | High Altitude Bias at the target level |
| 10 | 12.3 | 10.9 | -1.4 |
| 35 | 36.2 | 36.5 | 0.3 |
| 50 | 50.5 | 51.8 | 1.3 |
| 70 | 69.6 | 72.3 | 3.9% |
| 100 | 98.2 | 103.0 | 4.8% |
| 125 | 122.0 | 128.5 | 5.2% |
| pO2 | | | |
| --- | --- | --- | --- |
| Target(mmHg) | Low Altitude result | High Altitude result | High Altitude Bias at the target level |
| 10 | 10.8 | 14.7 | 3.9 |
| 30 | 30.5 | 34.9 | 4.4 |
| 45 | 45.3 | 50.1 | 4.8 |
| 60 | 60.2 | 65.3 | 5.1 |
| 200 | 198.5 | 207.1 | 4.3% |
| 450 | 445.4 | 460.3 | 3.3% |

Based on the altitude study results, the sponsor concluded that the candidate device operates up to 10,000 feet altitude for pH, $\mathrm{pCO_2}$ , and $\mathrm{pO_2}$ .

f. Assay cut-off:

Not applicable.

2. Comparison studies:

a. Method comparison with predicate device:

Method comparison studies were performed following CLSI EP9-A3.

{16}

#1. Point-of-Care Method Comparison for pH, pO₂, and pCO₂ using normal syringe mode

A POC method comparison study was performed to establish the accuracy performance of the GEM Premier 5000 for pH, pO₂, and pCO₂ at three external point-of-care (POC) settings.

A minimum of one operator at each POC site participated in sample analysis. Each site used one GEM Premier 5000 analyzer and one GEM Premier 4000 analyzer. The samples were leftover native whole blood samples from the routine hospital population. These samples comprised both venous and arterial collections.

A subset of about 12% of the total samples was adjusted to cover medical decision levels that could not be sourced by native specimens. These results were combined with each POC sites results.

Samples were tested in the normal syringe mode (150 μL). Patient samples were run in singlicate on one GEM Premier 5000 and the results were compared to those obtained on the predicate device (GEM Premier 4000) by the same operator. The linear regression analysis from the combined POC sites is presented below:

| Analyte | N | Slope | 95% CI for Slope | Intercept | 95% CI for Intercept | r² | Sample range tested |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pH | 479 | 0.941 | 0.927, 0.955 | 0.427 | 0.325, 0.530 | 0.991 | 7.01 to 7.92 |
| pO₂ | 506 | 0.992 | 0.984, 1.000 | 5.093 | 4.000, 5.717 | 0.996 | 6 to 685 |
| pCO₂ | 492 | 0.955 | 0.941, 0.971 | 3.545 | 2.743, 4.176 | 0.991 | 11 to 117 |

#2. Point-of-Care Method Comparison for pH and pO₂ using normal capillary mode with capillary specimens

Note: Capillary not claimed for pCO₂ on the GEM Premier 5000.

A POC method comparison study was performed to establish the accuracy performance of the GEM Premier 5000 for pH and pO₂ at an external POC site in a hospital and an internal laboratory which simulated a POC site.

The samples were prospectively collected whole blood capillary specimens. On each patient, two skin puncture specimens were collected consecutively into two capillary tubes containing lithium heparin. A minimum of 300 μL whole blood per patient was collected. The internal site tested healthy donors and the external site used samples

{17}

from a Cardiac Care Unit and a Progressive Cardiac Care Unit. There were 99 males and 55 females in the study. The age range was 18 to 84 years.

Each site used one GEM Premier 5000 analyzer and one GEM Premier 4000 analyzer. There were two POC operators at the external site and three POC operators at the internal site. The operators were hospital staff with no formal laboratory training.

After collection, the samples were immediately tested in normal capillary mode on the GEM Premier 5000 and GEM Premier 4000 (predicate) by the same POC operator. The results based on native capillary samples at Medical Decision Level are summarized below.

| Analyte | Range Min | Range Max | Medical decision level (MDL) | Bias at MDL | 95% CI of Bias at MDL |
| --- | --- | --- | --- | --- | --- |
| pH (N=171) | 7.36 | 7.59 | 7.30 | 0.002 | -0.015 to 0.020 |
| | | | 7.35 | 0.005 | -0.006 to 0.016 |
| | | | 7.45 | 0.010 | 0.005 to 0.014 |
| pO2 (N=167) | 52 | 105 | 30 | 6.1 | 1.3 to 11.6 |
| | | | 45 | 5.1 | 2.0 to 8.6 |
| | | | 60 | 4.1 | 2.0 to 5.7 |

The data from this POC method comparison study with native capillary samples was combined with contrived samples and tested internally in order to cover the claimed measuring range.

Linear regression analysis from the combined data is presented below:

| Analyte | N | Slope | 95% CI for Slope | Intercept | 95% CI for Intercept | r2 | Range |
| --- | --- | --- | --- | --- | --- | --- | --- |
| pH | 189 | 0.935 | 0.899, 0.970 | 0.494 | 0.228, 0.760 | 0.975 | 7.07 to 7.89 |
| pO2 | 218 | 1.008 | 0.994, 1.040 | 2.545 | 0.858, 4.000 | 0.996 | 6 to 676 |

b. Matrix comparison:

Not applicable. Sponsor stated that only lithium heparin anticoagulant whole blood is the acceptable sample type to be used for their device.

3. Clinical studies:

{18}

a. Clinical Sensitivity:

Not applicable.

b. Clinical specificity:

Not applicable.

c. Other clinical supportive data (when a. and b. are not applicable):

Not applicable.

4. Clinical cut-off:

Not applicable.

5. Expected values/Reference range:

The following are the reference ranges for pH, pCO₂, and pO₂ taken from published literature.

pH 7.35 to 7.45, arterial adult blood
7.32 to 7.42, venous blood

pCO₂ 35 to 48 mmHg, adult male arterial blood
32 to 35 mmHg, adult female arterial blood
6 to 7 mmHg, venous blood (right atrium), reported as an increase over arterial blood

pO₂ 83 to 108 mmHg, arterial adult blood

The sponsor recommends that each laboratory establish their own reference ranges.

Reference: Burtis, Carl and David Bruns, Tietz Textbook of Clinical Chemistry and Molecular Diagnostics, Elsevier Saunders, 7th edition, 2015, pp 952-982.

N. Proposed Labeling:

The labeling is sufficient and it satisfies the requirements of 21 CFR Part 809.10.

O. Conclusion:

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