(81 days)
The ABL90 FLEX analyzer is a portable, automated analyzer that measures pH, blood gases, electrolytes, glucose, lactate, and oximetry in heparinised whole blood. The ABL90 FLEX analyzer is intended for use by trained technologists, nurses, physicians and therapists. It is intended for use in a laboratory environment, near patient or point-of-care setting. These tests are only performed under a physician's order.
Indications for use:
pH, pO2 and pCO2: pH, pCO2 and pO2 measurements are used in the diagnosis and treatment of life-threatening acid-base disturbances.
Potassium (cK*): potassium measurements are used to monitor electrolyte balance in the diagnosis and treatment of disease conditions characterized by low or high blood potassium levels.
Sodium (cNa*): sodium measurements are used in the diagnosis and treatment of aldosteronism, diabetes insipidus, adrenal hypertension, Addison's disease, dehydration, inappropriate antidiuretic secretion, or other diseases involving electrolyte imbalance.
Calcium (cCa2+): calcium measurements are used in the diagnosis and treatment of parathyroid disease, a variety of bone diseases, chronic renal disease and tetany.
Chloride (cCl¯): chloride measurements are used in the diagnosis and treatment of electrolyte and metabolic disorders such a cystic fibrosis and diabetic acidosis.
Glucose (cGlu): glucose measurements are used in the diagnosis and treatment of carbohydrate metabolism disorders including diabetes mellitus and idiopathic hypoglycemia, and of pancreatic islet cell carcinoma.
Lactate (cLac): The lactate measurements measure the concentration of lactate in plasma. Lactate measurements are used to evaluate the acid-base status and are used in the diagnosis and treatment of lactic acidosis (abnormally high acidity of the blood.)
Total Hemoglobin (ctHb): total hemoglobin measurements are used to measure the hemoglobin content of whole blood for the detection of anemia.
sO2: oxygen saturation, more specifically the ratio between the concentration of oxyhemoglobin and oxyhemoqlobin plus reduced hemoglobin.
FO2Hb: oxyhemoqlobin as a fraction of total hemoglobin.
FCOHb: carboxyhemoqlobin measurements are used to determine the carboxyhemoglobin content of human blood as an aid in the diagnosis of carbon monoxide poisoning.
FMetHb: methemoqlobin as a fraction of total hemoglobin.
FHHb: reduced hemoglobin as a fraction of total hemoglobin.
Fraction of Fetal Hemoglobin (FHbF): FHbF indicates the amount of fetal hemoglobin. FHbF is seldom used clinically.
The ABL90 FLEX is a portable, automated system intended for in vitro testing of samples of whole blood for the parameters pH, pO2, pCO2, potassium, sodium, chloride, glucose, lactate, and co-oximetry parameters (total hemoqlobin, oxygen saturation, and the hemoglobin fractions FO2Hb, FCOHb, FMetHb, FHHb and FHbF).
This document describes modifications to the ABL90 FLEX device, specifically software changes to suppress glucose results under certain low pO2 conditions. The following is a summary of the acceptance criteria and the study that proves the device meets them:
1. Table of Acceptance Criteria and Reported Device Performance
| Hazard | Validation/Verification Activity | Pre-determined Acceptance Criteria | Testing Results Summary | Met Acceptance Criteria? |
|---|---|---|---|---|
| 41/Too low Glucose result in the upper reportable range obtained from samples with low pO2 level | Interference study at different pO2 levels and at different glucose levels covering the reportable range of the analyzer using fresh heparinized whole blood samples. | Bias: ≤10% for glucose when pO2 is > 10 mmHg when compared to the control | 6 different pO2 levels (+ pO2 ≥90 mmHg as control), 7 different glucose levels, 3 analyzers, 6 tests of each sample on each analyzer, 2 runs. Total of 1512 measurements. |
The results are valid under the conditions that:
- all glucose results are suppressed when the pO2 level of the sample is below 10 mmHg.
- all glucose results are suppressed when pO2 level of the sample is between 10 mmHg and 25 mmHg and the glucose level is above 270 mg/dL.
All acceptance criteria are met: Bias 90 mmHg | 18.1 | 0.1 | 0.7 | 4.1 | 240 |
| Glu Mid $98.7 \pm 9 mg/dL$ | 10 mmHg | 101.7 | 1.1 | 3.8 | 3.7 | 240 |
| | 30 mmHg | 101.0 | 0.7 | 3.3 | 3.3 | 240 |
| | >90 mmHg | 101.2 | 0.5 | 3.3 | 3.2 | 240 |
| Glu High $270 \pm 6 mg/dL$ | 10 mmHg | 254.1 | 1.6 | 10.8 | 4.2 | 240 |
| | 30 mmHg | 262.3 | 1.1 | 8.8 | 3.4 | 240 |
| | >90 mmHg | 271.9 | 1.7 | 7.4 | 2.7 | 240 | | Passed |
| 48/Unacceptable bias on Glucose results obtained from samples with pO2 levels above 25 mmHg | (Covered by Interference Study) | (Covered by Interference Study) | (Covered by Interference Study) | Passed |
| (General Performance) Method Comparison | Method comparison study versus a comparative analyzer (ABL735). | Slope: 0.9 - 1.1
Intercept: 0
Correlation Coefficient: ≥ 0.95 | Linear regression of the pooled data gives a slope of 0.9206, intercept of 0.084 and an R² ≥ 0.95. | Passed |
2. Sample Sizes and Data Provenance
Interference Study (Test Set):
- Sample Size:
- 6 different pO2 levels (including a control at ≥90 mmHg)
- 7 different glucose levels
- 3 analyzers
- 6 tests of each sample on each analyzer
- 2 runs
- Total: 1512 measurements
- Data Provenance: Fresh heparinized whole blood samples. The document does not specify the country of origin but implies an in-house study ("Radiometer Medical ApS"). The samples are likely prospective as they are "fresh heparinized whole blood samples" and "untreated donor samples in combination with spiked donor blood."
Precision Study (Test Set):
- Sample Size:
- 20 days
- 3 different pO2 levels
- 3 different glucose levels
- 2 tests of each sample each day
- 2 runs
- Total: 2160 measurements
- Data Provenance: Serum pool sample of glucose. The document does not specify the country of origin but implies an in-house study ("Radiometer Medical ApS"). likely prospective.
Method Comparison Study:
- Sample Size: A total of 52 different donors and approximately 500 samples.
- Data Provenance: In-house study using untreated donor samples in combination with spiked donor blood where necessary. Likely prospective.
3. Number of Experts and Qualifications for Ground Truth
The studies described are for an in vitro diagnostic device measuring blood analytes. The ground truth for such devices is typically established through a reference method or comparator device, not clinical expert consensus in the way image analysis or clinical diagnosis algorithms would.
- Interference Study: The ground truth for bias calculation was established by comparing results to a "control" pO2 level (≥90 mmHg). This relies on the established accuracy of the glucose measurement at optimal pO2.
- Precision Study: The ground truth is inherent in the known concentrations of glucose levels in the serum pool samples used.
- Method Comparison Study: The ground truth was established by comparison to results from a "comparative analyzer (ABL735)," which is a predicate device.
Therefore, the concept of "number of experts" and their "qualifications" for establishing ground truth in the context of clinical interpretation (e.g., radiologists for images) is not applicable here. The ground truth is based on laboratory-defined reference measurements or a well-established predicate device.
4. Adjudication Method for the Test Set
Adjudication methods like "2+1" or "3+1" are typically used for subjective assessments where multiple human readers disagree (e.g., interpreting medical images or clinical notes). This document describes performance studies for an in vitro diagnostic device, where results are quantitative measurements. Discrepancies would be resolved through re-testing, calibration, or investigation of instrument malfunction, not clinical adjudication by experts. Therefore, no formal adjudication method of this type is mentioned or expected.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This type of study is relevant for assessing the impact of AI on human reader performance, typically in diagnostic imaging or similar fields where human interpretation is central. This document focuses on the analytical performance of an in vitro diagnostic device, specifically the impact of a software modification on glucose measurement accuracy under certain pO2 conditions. The comparisons are between the device's results and reference methods/predicate devices, not between human readers with and without AI.
6. Standalone Performance
Yes, standalone performance was done. The entire submission describes the standalone analytical performance of the ABL90 FLEX device, particularly its glucose measurement under varying pO2 conditions, after the proposed software modification. The studies (Interference, Precision, Method Comparison) evaluate the device's inherent measurement capabilities and the impact of the software suppression logic without human intervention in the continuous measurement process. The modification itself involves the device automatically suppressing results based on pO2 and glucose levels.
7. Type of Ground Truth Used
- Interference Study: The ground truth for bias calculation was primarily reference measurements at optimal pO2 (≥90 mmHg) and the known concentrations of spiked glucose.
- Precision Study: The ground truth was based on the known concentrations of glucose in the serum pool samples used.
- Method Comparison Study: The ground truth was established through comparison with a predicate device (ABL735) and likely involved reference methods for its initial validation.
In essence, the ground truth for these analytical performance studies is rooted in established reference values, comparator devices, and controlled experimental conditions where "true" concentrations or performance characteristics are either known or determined by a validated reference standard.
8. Sample Size for the Training Set
The document does not explicitly state a separate "training set" sample size. This is common for analytical performance studies of this nature, especially when the modification is primarily a software rule change based on understanding of the underlying chemistry (glucose oxidase co-reaction with oxygen). The "training" in such cases might involve development and initial testing against known samples to define the suppression rules, which is not typically formalized as a distinct "training set" in the context of a 510(k) submission for IVDs. The "test set" described above (1512 measurements for interference, 2160 for precision) serves as the primary validation data.
9. How the Ground Truth for the Training Set Was Established
As noted above, a formal "training set" with ground truth establishment in the machine learning sense is not explicitly described or necessarily applicable here. The software modification (suppression rules) likely emerged from:
- Understanding of the underlying scientific principle: The document states, "The linearity of the glucose is dependent on the oxygen tension of the sample. This dependence is due to the co-reaction of glucose and oxygen by the enzyme glucose oxidase." This foundational knowledge guides the need for suppression.
- Prior internal R&D data/experiments: Radiometer would have conducted extensive internal studies to characterize the glucose sensor's performance across various pO2 and glucose concentrations to determine the thresholds (e.g., 270 mg/dL) at which accuracy is compromised. This data, if it exists, would have informed the development of the suppression logic.
Therefore, the "ground truth" for developing these rules would be based on analytical measurements from controlled experiments using samples with known glucose and pO2 concentrations, helping to define the performance boundaries.
§ 862.1120 Blood gases (P
CO2 , PO2 ) and blood pH test system.(a)
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.(b)
Classification. Class II.