(273 days)
The ABL90 FLEX PLUS System is an in vitro diagnostic, portable, automated analyzer that quantitatively measures electrolytes (cK+, cNa+, cCa2+), glucose, and lactate in heparinized arterial and venous whole blood.
The ABL90 FLEX PLUS System 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.
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, delydration, 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.
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 measure the concentration of lactate. Lactate measurements are used to evaluate the acid-base status and are used in the diagnosis and treatment of lactic acidity of the blood).
The ABL90 FLEX PLUS System consists of the ABL90 FLEX PLUS analyzer, sensor cassette and solution pack consumables, and related accessories for the analyzers. The ABL90 FLEX PLUS is a portable, automated system intended for in vitro testing of samples of balanced heparinized whole blood for electrolytes (cK+, cNa*, cCa²), glucose, and lactate. The ABL90 FLEX PLUS System has an automated sample inlet mechanism, which can collect blood through two different measuring modes: the S65 syringe mode and the SP65 short probe mode.
The provided text is a 510(k) Summary for the ABL90 FLEX PLUS System, an in vitro diagnostic device. This document focuses on demonstrating substantial equivalence to a legally marketed predicate device (ABL90 FLEX) rather than proving the device meets specific acceptance criteria as might be defined for a novel AI/ML device.
Therefore, much of the requested information regarding acceptance criteria for AI/ML performance, study design (test set, ground truth establishment, expert adjudication, MRMC studies, standalone performance, training set details) is not applicable to this type of device and its regulatory submission.
The document primarily proves the analytical performance of the new device is comparable to the predicate device through various analytical studies.
Here's a breakdown of the applicable information based on the provided text, and an explanation of why other requested information is not present:
1. A table of acceptance criteria and the reported device performance
The document does not explicitly present "acceptance criteria" in a pass/fail table for each performance metric in the way it might for a novel AI/ML device. Instead, it presents analytical performance data (linearity, precision, detection, method comparison, interference) which is implicitly compared against pre-defined internal specifications or what is considered acceptable for the similar predicate device. The goal is to show the new device performs equivalently to the predicate.
Below is a summary of the reported device performance from the tables in the document. The "Acceptance Criteria" column cannot be fully populated as precise numerical thresholds are not explicitly stated as "acceptance criteria" in this 510(k) summary, but are rather implied by the successful demonstration of performance often within CLSI guidelines and comparable to the predicate.
| Parameter (Unit) | Test Category | Reported Performance (Range / Values) | Implicit Acceptance Criteria (based on predicate equivalence and CLSI) |
|---|---|---|---|
| cCa2+ (mg/dL) | Linearity | Slope: 0.883, Intercept: 0.445, R^2: 1.000 | R^2 near 1.0, slope near 1.0, intercept near 0, demonstrating linearity over the reportable range. |
| LoQ | 1.26 | Established lower limit of reliable quantitation. | |
| Precision (QC) | Repeatability SD: 0.003-0.014, CV%: 0.1-0.3 | Low SD and CV%, demonstrating consistent results. | |
| Precision (Blood) | Repeatability SD: 0.003-0.022, CV%: 0.06-0.45 | Low SD and CV%, demonstrating consistent results within biological samples. | |
| Method Comp. (Bias at MD) | S65: 0.001-0.003, SP65: 0.003-0.009 | Low bias compared to the predicate device, indicating equivalent measurements. | |
| cK+ (mEq/L) | Linearity | Slope: 1.001, Intercept: 0.027, R^2: 1.000 | R^2 near 1.0, slope near 1.0, intercept near 0, demonstrating linearity over the reportable range. |
| LoQ | 1.6 | Established lower limit of reliable quantitation. | |
| Precision (QC) | Repeatability SD: 0.00-0.01, CV%: 0.1-0.2 | Low SD and CV%, demonstrating consistent results. | |
| Precision (Blood) | Repeatability SD: 0.007-0.026, CV%: 0.14-0.96 | Low SD and CV%, demonstrating consistent results within biological samples. | |
| Method Comp. (Bias at MD) | S65: 0.002-0.004, SP65: 0.004-0.008 | Low bias compared to the predicate device, indicating equivalent measurements. | |
| cNa+ (mEq/L) | Linearity | Slope: 1.001, Intercept: -0.642, R^2: 1.000 | R^2 near 1.0, slope near 1.0, intercept near 0, demonstrating linearity over the reportable range. |
| LoQ | 99 | Established lower limit of reliable quantitation. | |
| Precision (QC) | Repeatability SD: 0.1-0.2, CV%: 0.1 | Low SD and CV%, demonstrating consistent results. | |
| Precision (Blood) | Repeatability SD: 0.061-0.194, CV%: 0.05-0.14 | Low SD and CV%, demonstrating consistent results within biological samples. | |
| Method Comp. (Bias at MD) | S65: 0.265-0.290, SP65: 0.221-0.259 | Low bias compared to the predicate device, indicating equivalent measurements. | |
| cGlu (mg/dL) | Linearity | Slope: 1.032, Intercept: -1.073, R^2: 1.000 | R^2 near 1.0, slope near 1.0, intercept near 0, demonstrating linearity over the reportable range. |
| LoD/LoQ | LoD: 5, LoQ: 5 | Established lower limits of detection and reliable quantitation. | |
| Precision (QC) | Repeatability SD: 0.3-1.3, CV%: 0.5-1.1 | Low SD and CV%, demonstrating consistent results. | |
| Precision (Blood) | Repeatability SD: 0.207-2.221, CV%: 0.35-0.85 | Low SD and CV%, demonstrating consistent results within biological samples. | |
| Method Comp. (Bias at MD) | S65: -0.460 to -2.028, SP65: -0.663 to -2.045 | Low bias compared to the predicate device, indicating equivalent measurements. | |
| cLac (mg/dL) | Linearity | Slope: 0.971, Intercept: -0.433, R^2: 1.000 | R^2 near 1.0, slope near 1.0, intercept near 0, demonstrating linearity over the reportable range. |
| LoD/LoQ | LoD: -0.3, LoQ: 2 | Established lower limits of detection and reliable quantitation. (Note: Negative LoD likely a calculation artifact near zero) | |
| Precision (QC) | Repeatability SD: 0.2-0.3, CV%: 0.3-1.1 | Low SD and CV%, demonstrating consistent results. | |
| Precision (Blood) | Repeatability SD: 0.177-0.379, CV%: 0.75-2.25 | Low SD and CV%, demonstrating consistent results within biological samples. | |
| Method Comp. (Bias at MD) | S65: -0.116 to 0.013, SP65: -0.156 to -0.169 | Low bias compared to the predicate device, indicating equivalent measurements. |
2. Sample size used for the test set and the data provenance (e.g. country of origin of the data, retrospective or prospective)
- Test Set (for performance validation):
- Linearity: The specific number of samples tested for linearity is not explicitly stated as 'N' values in Table 1 but ranges presented (e.g., 1.896-11.146 for cCa2+) imply a sufficient number of points across the range were used.
- Detection (LoB, LoD, LoQ): Not explicitly stated as 'N' values in Table 2.
- Precision (using stable, aqueous ampoule-based QC material): Varies per parameter/level, but generally 243-244 replicates (N) per parameter/level.
- Precision (using blood): Varies per parameter/mode/interval, ranging from 2 to 202 replicates (N).
- Method Comparison:
- Arterial blood (S65 mode): 221-225 samples (N) across parameters.
- Arterial blood (SP65 mode): 214-218 samples (N) across parameters.
- Venous blood (S65 mode): 231-234 samples (N) across parameters.
- Venous blood (SP65 mode): 219-225 samples (N) across parameters.
- Combined (S65 mode): 436-441 samples (N) for combined arterial/venous.
- Combined (SP65 mode): 420-425 samples (N) for combined arterial/venous.
- Interference: "Large panel of likely interferents" for paired-difference study; dose-response studies for significant interferents. Specific sample sizes for each interferent are not detailed in the summary.
- Data Provenance: The document states that precision studies using QC material were conducted at "three external sites." Method comparison and precision studies using blood were conducted using both arterial and venous blood, and in both sample collection modes. The country of origin for the data (patients or samples) is not specified in this summary. The studies are described as "analytical performance testing," implying they are prospective or controlled laboratory studies rather than retrospective analysis of existing clinical data.
3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts (e.g. radiologist with 10 years of experience)
- Not Applicable: This device is an in vitro diagnostic (IVD) analyzer that quantitatively measures analytes. Its performance is evaluated against reference measurement procedures or highly controlled materials, not by expert interpretation of images or clinical cases requiring expert consensus or qualifications. Ground truth is established by the reference method itself or the known concentration of QC materials.
4. Adjudication method (e.g. 2+1, 3+1, none) for the test set
- Not Applicable: As this is an IVD device measuring quantitative analytes, there is no expert adjudication process in this context, unlike an AI/ML device interpreting medical images. Performance is determined by comparison to reference methods or statistical analysis against known values.
5. If a multi reader multi case (MRMC) comparative effectiveness study was done, If so, what was the effect size of how much human readers improve with AI vs without AI assistance
- Not Applicable: This is an IVD analyzer, not an AI/ML device that assists human readers. Therefore, an MRMC study is not relevant to its regulatory approval process.
6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done
- Partially Applicable (in a different sense): The ABL90 FLEX PLUS System is a standalone automated analyzer. Its performance is measured directly (algorithm only, if you consider the device's internal measurement algorithm) against reference methods or known concentrations, without a human-in-the-loop interpretation being the primary output that's being evaluated for accuracy. The results presented (linearity, precision, method comparison) are representative of its standalone performance.
7. The type of ground truth used (expert concensus, pathology, outcomes data, etc)
- Quantitative Reference Methods / Known Concentrations:
- Linearity/Detection: Ground truth is established by preparing samples with known, precise concentrations across the measurement range, or by the inherent properties of the measurement system for LoB/LoD/LoQ.
- Precision: Ground truth is the expected value of the quality control (QC) materials or the prepared blood samples, or simply the reproducibility of measurements on the same sample.
- Method Comparison: Ground truth is the measurement from the legally marketed predicate device (ABL90 FLEX, specifically "ABL90 FLEX PLUS analyzer as it was designed at the time of the clearance of K160153") that the new device is being compared against. This device itself serves as the "reference method" for substantial equivalence.
- Interference: Ground truth is the expected measurement of known samples, with and without the interferent, using a reference method, to identify if the interferent causes a clinically significant deviation.
8. The sample size for the training set
- Not Applicable (in the AI/ML sense): This document describes the analytical validation of a traditional IVD device, not an AI/ML algorithm. There is no "training set" in the machine learning sense for this type of submission. The device is a physical instrument with established chemical/electrochemical measurement principles.
9. How the ground truth for the training set was established
- Not Applicable: As there is no "training set" in the AI/ML context, this question is not relevant. The device's internal parameters and calibration would be established through a manufacturing and calibration process, not through a "training" phase with a ground truth dataset in the way an AI model is trained.
§ 862.1600 Potassium test system.
(a)
Identification. A potassium test system is a device intended to measure potassium in serum, plasma, and urine. Measurements obtained by this device are used to monitor electrolyte balance in the diagnosis and treatment of diseases conditions characterized by low or high blood potassium levels.(b)
Classification. Class II.