FDA 510(k) Search - By Innolitics (SaMD and AI Experts)

The Randox CSF Controls Levels 2 and Level 3 are intended for in vitro diagnostic use as assayed quality control material to monitor the precision of laboratory testing procedures for Clinical Chemistry systems. This device is intended for prescription use only.

Device Description

The Randox CSF controls consist of a buffered aqueous solution with biological materials; they are supplied at 2 levels, level 2 and 3. Target values and ranges are supplied for the following analytes at both levels: Chloride, Glucose, IgG, Lactic Acid, Sodium, Total Protein (Pyrogallol Red), and % Total Protein quoted for Electrophoresis regions at both levels: Albumin, Alpha-1-Globulin, Alpha-2-Globulin, Beta Globulin and Gamma Globulin.

AI/ML Overview

The Randox CSF Controls Levels 2 and 3 are in vitro diagnostic devices. The acceptance criteria and the study proving the device meets these criteria are detailed below, primarily focusing on stability and value assignment studies.

1. Table of Acceptance Criteria and Reported Device Performance

Study/Parameter	Acceptance Criteria	Reported Device Performance
Open Vial Stability	Percentage deviation of 14-day reconstituted vial to fresh vial should be ± 3% (for Chloride, Glucose, Lactate, Sodium, and Total Protein).	Control Level 2 (Lot 059CF): Stable for 14 days reconstituted and stored at +2-8°C.
Control Level 3 (Lot 060CF): Stable for 14 days reconstituted and stored at +2-8°C.
(Specific deviation percentages per analyte are not provided in the summary but it's stated the data "demonstrates" stability within the criteria.)
Shelf Life (Real-time Stability)	Control recovery for routinely stored samples compared to ultra-frozen samples should be within ± 15% deviation (for Total Protein, and implicitly for other analytes calculated as a percentage of Total Protein).	Control Level 2 (Lot 038CF): Tested and passed for 37 months.
Control Level 3 (Lot 035CF): Tested and passed for 37 months.
This exceeds the claimed shelf life of 3 years by 1 month.
Value Assignment (Level 2 - Electrophoresis analytes)	For Albumin: Percentage Range Applied (%) 10% (53.9 - 65.9).
For Alpha-1-globulin, Alpha-2-globulin, Beta-globulin, Gamma-globulin: Percentage Range Applied (%) 25%.	Target values and acceptable ranges are provided for each analyte. The study implies that target values are successfully assigned and fall within these specified ranges.
Albumin: Target 59.9, Range 53.9 - 65.9
Alpha-1-globulin: Target 2.0, Range 1.5 – 2.5
Alpha-2-globulin: Target 4.6, Range 3.5 - 5.8
Beta-globulin: Target 5.3, Range 4.0 - 6.6
Gamma-globulin: Target 28.3, Range 21.2 - 35.4
Value Assignment (Level 2 - Other analytes)	For Chloride, Glucose (Oxidase/Hexokinase), Lactate, Sodium: %CV 3%, % Recovery error 5%, with specific percentage ranges applied (15% for Glucose/Lactate, 5% for Sodium, 10% for Chloride).
For Immunoglobulin G: %CV 10%, % Recovery error 10%, 25% percentage range applied.
For Protein Total: %CV 5%, % Recovery error 5%, 20% percentage range applied.	Target values and acceptable ranges are provided. The study implies results for precision (%CV) and recovery error were within these criteria during value assignment.
Chloride: Target 86.8, Range 78.1 - 95.5
Glucose Oxidase: Target 3.11, Range 2.64 - 3.58
Glucose Hexokinase: Target 2.85, Range 2.42 - 3.28
Immunoglobulin G: Target 25.3, Range 19.0 - 31.6
Lactate: Target 0.38, Range 0.32 - 0.44
Protein Total: Target 0.235, Range 0.188 - 0.282
Sodium: Target 127, Range 121 - 133
Value Assignment (Level 3 - Electrophoresis analytes)	For Albumin: Percentage Range Applied (%) 10% (53.1-64.9).
For Alpha-1-globulin, Alpha-2-globulin, Beta-globulin, Gamma-globulin: Percentage Range Applied (%) 25%.	Target values and acceptable ranges are provided for each analyte. The study implies that target values are successfully assigned and fall within these specified ranges.
Albumin: Target 59.0, Range 53.1-64.9
Alpha-1-globulin: Target 2.4, Range 1.8-3.0
Alpha-2-globulin: Target 7.4, Range 5.5-9.3
Beta-globulin: Target 6.9, Range 5.2-8.6
Gamma-globulin: Target 24.3, Range 18.23-30.4
Value Assignment (Level 3 - Other analytes)	For Chloride, Glucose (Oxidase/Hexokinase), Lactate, Sodium: %CV 3%, % Recovery error 5%, with specific percentage ranges applied (15% for Glucose/Lactate, 5% for Sodium, 10% for Chloride).
For Immunoglobulin G: %CV 10%, % Recovery error 10%, 25% percentage range applied.
For Protein Total: %CV 5%, % Recovery error 5%, 20% percentage range applied.	Target values and acceptable ranges are provided. The study implies results for precision (%CV) and recovery error were within these criteria during value assignment.
Chloride: Target 110, Range 99-121
Glucose Oxidase: Target 5.81, Range 4.94-6.68
Glucose Hexokinase: Target 5.53, Range 4.70-6.64
Immunoglobulin G: Target 70.3, Range 52.7-87.9
Lactate: Target 3.93, Range 3.34-4.64
Protein Total: Target 0.489, Range 0.391-0.587
Sodium: Target 169, Range 161-177

2. Sample Size Used for the Test Set and Data Provenance

Open Vial Stability: For open vial stability, a "set" of CSF Controls Levels 2 and 3 was used. On day 14, a "fresh sample" was reconstituted and compared. This implies at least two reconstituted vials (one fresh, one 14-day aged) for comparison for each control level, and "replicates" were performed, though the specific number of replicates is not stated.
Shelf Life (Real-time Stability): For real-time stability, "10 sets of each control" (Level 2 and Level 3) were used. This means 10 vials of Level 2 and 10 vials of Level 3 were stored at ultra-frozen conditions, and another 10 vials of Level 2 and 10 vials of Level 3 were stored at routine temperature (2-8°C).
Value Assignment: "Several replicates" were used for both electrophoresis measurements and for the remaining analytes, but the exact number of replicates is not specified.
Data Provenance: The document does not explicitly state the country of origin for the data or whether the studies were retrospective or prospective. Given that it's a submission for a new device manufactured by Randox Laboratories Limited in the United Kingdom, it's highly probable the studies were conducted prospectively by the manufacturer.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts

The document describes the establishment of target values for the control material (which serves as the "ground truth" for quality control purposes).
For electrophoresis analytes, vials were "analyzed at an external site." The number of experts or the specific qualifications of those performing the analysis are not explicitly stated, but it's implied that these are qualified professionals at a specialized laboratory.
For other analytes, comparison was made against a "master lot." This does not involve human expert interpretation in the same way as, for example, image analysis.

4. Adjudication Method for the Test Set

Adjudication methods like 2+1 or 3+1 typically apply to studies where human interpretation of results is involved and discrepancies need to be resolved.
For laboratory control materials, the "ground truth" (target values and acceptable ranges) is established through analytical processes and statistical analysis of multiple replicates. Discrepancies would be resolved by re-running samples, checking instrument calibration, or investigating analytical issues, rather than through expert adjudication in the human-reader sense. The value assignment methods described involve calculating means of "several replicates" and comparing against established criteria, implicitly handling variability through statistical methods rather than human adjudication.

5. If a Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study Was Done

No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done.
This device is a quality control material for in vitro diagnostic tests, not an algorithm or imaging device that requires human interpretation or comparison of human reader performance. The studies focus on the analytical performance (stability, value assignment) of the control material itself.

6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done

This device is a quality control material, not an algorithm. Therefore, a standalone performance study (in the context of an algorithm's performance independent of human input) is not applicable or described. The performance studies focus on the intrinsic properties and stability of the control material.

7. The Type of Ground Truth Used

The ground truth for this device (Randox CSF Controls) is its assigned target values and acceptable ranges for various analytes.
This ground truth is established through:
- Analytical Testing and Expert Consensus (implicit for electrophoresis): For electrophoresis analytes (Albumin, Alpha-1-globulin, etc.), the mean of "several replicates" analyzed at an external site is taken, and percentages of total protein are calculated. This implies the analytical results from a qualified external lab are considered the definitive "ground truth."
- Comparison to a Master Lot and Analytical Testing: For other analytes (Chloride, Glucose, Immunoglobulin G, Lactate, Sodium, Total Protein), the test controls are assessed against a "master lot" using analytical methods. The mean of several replicates and CV are calculated, and recovery of the master lot is measured against predefined acceptance criteria.

8. The Sample Size for the Training Set

The concept of a "training set" is typically associated with machine learning algorithms.
For this quality control material, there isn't a "training set" in that sense. The manufacturing process and formulation are developed, and then the stability and value assignment studies (described above) act as validation and characterization. The "master lot" mentioned in value assignment method 2 could be considered a reference or calibration standard that helps establish the "ground truth" for new lots, but it's not a training set for an AI model.

9. How the Ground Truth for the Training Set Was Established

As there is no "training set" in the context of an algorithm for this device, this question is not fully applicable.
However, if we interpret "ground truth for the training set" as the basis for establishing the expected values for the control material, it's established through:
- Reference materials and methods: For individual analytes, traceable sources are used, such as "Synthetic Analytical Grade chemical" from commercial suppliers (e.g., SIGMA, EUROPA) or biological materials "derived from Human serum" (e.g., Intergen).
- Comprehensive analytical testing: As described in section 9 "Summary of Value Assignment," extensive laboratory analysis involving "several replicates" and comparison to a master lot or external site analysis is performed to determine the "Target Value" and "Acceptable range" for each analyte.

Summary

Regulation Number and Section

§ 862.1660 Quality control material (assayed and unassayed).

(a)
Identification. A quality control material (assayed and unassayed) for clinical chemistry is a device intended for medical purposes for use in a test system to estimate test precision and to detect systematic analytical deviations that may arise from reagent or analytical instrument variation. A quality control material (assayed and unassayed) may be used for proficiency testing in interlaboratory surveys. This generic type of device includes controls (assayed and unassayed) for blood gases, electrolytes, enzymes, multianalytes (all kinds), single (specified) analytes, or urinalysis controls.(b)
Classification. Class I (general controls). Except when intended for use in donor screening tests, quality control materials (assayed and unassayed) are exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 862.9.