(72 days)
The Pacific Hemostasis ThromboScreen® 400C is a photo-optical instrument used for the performance of in-vitro diagnostic coagulation testing of citrated plasma samples in the clinical laboratory. The ThromboScreen® 400C has both clot and chromogenic testing capabilities. Assays performed on the instrument include routine clotting tests such as Prothrombin Time (PT), Activated Partial Thromboplastin Time (APTT), Fibrinogen (Clauss and Derived methods), and PT and APTT-based factor assays. Chromogenic tests include assays such as Antithrombin III, Protein C and Heparin Xa.
The ThromboScreen® 400C (TS400C) is a photo-optical instrument used for the performance of in-vitro diagnostic clotting and chromogenic procedures in the clinical laboratory. The instrument utilizes photo-optical principles for both clotting and chromogenic assays. The ThromboScreen® 400C light source is provided by a halogen lamp. The incubator block is temperature regulated to 36.5 - 37.5℃ and contains four measuring positions, five reagent and 24 cuvette prewarming positions.
Here's an analysis of the provided text regarding the ThromboScreen® 400C, structured to address your request:
Acceptance Criteria and Device Performance Study
The ThromboScreen® 400C (TS400C) is a manual coagulation instrument using photo-optical principles for in-vitro diagnostic clotting and chromogenic procedures. The study aims to demonstrate its substantial equivalence to the MLA-900C and MLA-1000C. The acceptance criteria are implicitly defined by the performance of the predicate devices (MLA-900C and MLA-1000C) in precision and method comparison studies. The TS400C's performance is deemed acceptable if its results are comparable to those of the predicate devices.
1. Table of Acceptance Criteria (as implied by predicate devices) and Reported Device Performance (TS400C)
| Test | Metric (Implied Acceptance) | TS400C Reported Performance |
|---|---|---|
| Within-run Precision (%CV) | ||
| PT (Low/Normal/High) | Comparable to MLA-900C/1000C | Site 1: 1.9-2.5; Site 2: 4.3-5.2 |
| APTT (Low/Normal/High) | Comparable to MLA-900C/1000C | Site 1: 3.1-5.4; Site 2: 2.4-4.1 |
| Clauss Fibrinogen (Low/Normal/High) | Comparable to MLA-900C/1000C | 6.4-7.2 |
| Derived Fibrinogen (Low/Normal/High) | Comparable to MLA-900C/1000C | 2.3-7.5 |
| Factor V (Low/Normal) | Comparable to MLA-900C/1000C | 4.0-4.5 |
| Factor VIII (Low/Normal) | Comparable to MLA-900C/1000C | 9.7-9.9 |
| Chromogenic ATIII (Low/Normal) | Comparable to MLA-900C/1000C | 4.9-6.9 |
| Between-run Precision (%CV) | ||
| PT (Normal/High) | Comparable to MLA-1000C | 3.4-6.0 |
| APTT (Normal/High) | Comparable to MLA-1000C | 1.8-2.5 |
| Clauss Fibrinogen (Low/Normal/High) | Comparable to MLA-1000C | 2.9-4.1 |
| ATIII (Low/Normal) | Comparable to MLA-1000C | 4.5-9.4 |
| Method Comparison (Correlation Coefficient, r) | ||
| Prothrombin Time (seconds) | High correlation (e.g., >0.95) | 0.99 (both sites) |
| Prothrombin Time (INR) | High correlation (e.g., >0.95) | 0.98, 0.99 |
| Derived Fibrinogen | High correlation (e.g., >0.95) | 0.98 |
| Activated Partial Thromboplastin Time | High correlation (e.g., >0.95) | 0.98, 0.96 |
| Clauss Fibrinogen | High correlation (e.g., >0.95) | 0.98 (both sites) |
| Factor VIII | High correlation (e.g., >0.95) | 0.97 |
| Factor V | High correlation (e.g., >0.95) | 0.97 |
| Chromogenic ATIII | Moderate-High correlation (e.g., >0.85) | 0.88 |
Summary of Device Performance:
- Precision: The TS400C demonstrated within-run and between-run precision comparable to the predicate MLA instruments, as detailed in Tables 1 and 2. For instance, PT within-run %CV for TS400C ranged from 1.9-5.2%, while the MLA ranged from 1.1-3.8% (Site 1) and 1.5-2.0% (Site 2). APTT within-run %CV for TS400C ranged from 2.4-5.4%, while the MLA ranged from 0.8-0.9% (Site 1) and 2.2-3.3% (Site 2). While not identical, the presented data is used to support substantial equivalence.
- Method Comparison: The TS400C showed high correlation coefficients (r-values ranging from 0.88 to 0.99) when compared to the MLA-900C/1000C for various coagulation tests (PT, APTT, Fibrinogen, Factor V, Factor VIII, Chromogenic ATIII), as shown in Table 3. This indicates a strong agreement between the results obtained from the TS400C and the predicate devices. The regression equations also suggest a close linear relationship.
2. Sample Size Used for the Test Set and Data Provenance
- Test Set Sample Sizes:
- Within-run Precision (Table 1): Unspecified number of samples for each "Low," "Normal," and "High" category for each test. The table implies at least three distinct sample types (Low, Normal, High) for most tests.
- Between-run Precision (Table 2): Unspecified number of samples for each "Low," "Normal," and "High" category for each test.
- Method Comparison (Table 3):
- Prothrombin Time (seconds & INR): 94 samples (Site #1), 139 samples (Site #2)
- Derived Fibrinogen: 47 samples (Site #1)
- Activated Partial Thromboplastin Time: 93 samples (Site #1), 117 samples (Site #2)
- Clauss Fibrinogen: 50 samples (Site #1), 20 samples (Site #2)
- Factor VIII: 49 samples (Site #1)
- Factor V: 50 samples (Site #1)
- Chromogenic ATIII: 58 samples (Site #1)
- Data Provenance: The study was conducted at two sites, referred to as "Site 1" and "Site 2." Site 1 used the MLA-1000C as the predicate, and Site 2 used the MLA-900C. The data is from "specimens... from apparently healthy individuals and from patients with different pathological conditions." This indicates prospective and/or retrospective clinical samples, collected with a mix of healthy and diseased states, to ensure a wide range of values. The country of origin is not explicitly stated but implies testing within a clinical laboratory setting, likely in the US given the FDA submission.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Qualifications
The concept of "experts" and "ground truth" as typically applied to image analysis or diagnostic interpretation by human readers is not directly applicable here. For a coagulation instrument, the "ground truth" is typically established by:
- The reference method (here, the predicate MLA-900C and MLA-1000C instruments).
- Clinically validated reagents and controls.
- Adherence to established laboratory assay protocols.
Therefore, no information on the number or qualifications of "experts" (e.g., radiologists) in this context is provided or expected. The expertise implicitly lies in the design and validation of the predicate devices and the laboratory personnel performing the assays.
4. Adjudication Method for the Test Set
Not applicable. As this is a comparison between two instruments measuring quantifiable analytes, there is no human adjudication process involved in establishing the "correct" measurement for each sample. The comparison is statistical (correlation, regression, precision metrics).
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study
No. This is not an MRMC study. MRMC studies are typically used to assess reader performance (e.g., radiologists interpreting images) with and without AI assistance. This study involves comparing the performance of a new instrument against established predicate instruments for quantitative laboratory tests. Therefore, there is no "effect size of how much human readers improve with AI vs without AI assistance" to report.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study
Yes, this is essentially a standalone (algorithm only) performance study of the TS400C. The instrument, while "manual" in that samples are pipetted by a user, performs the measurement and calculation independently of human interpretation of the results themselves. The data presented demonstrates the instrument's intrinsic accuracy and precision when performing specified assays, without a human interpretation step that the AI assists. The comparison is between the standalone TS400C and the standalone predicate MLA instruments.
7. Type of Ground Truth Used
The "ground truth" for the test set is established by the measurements obtained from the legally marketed predicate devices, the MLA-900C and MLA-1000C. These devices serve as the reference standard against which the new device's performance is compared. The use of "clinically significant ranges" and "patients with different pathological conditions" ensures that the comparison covers the relevant analytical range encountered in clinical practice.
8. Sample Size for the Training Set
The document does not describe a separate "training set" in the context of machine learning or AI models with distinct training and test phases. This is a traditional medical device validation study where the instrument's performance is evaluated. The "training" in this context would have occurred during the development and calibration of the TS400C by the manufacturer (Pacific Hemostasis). No specific sample size for such internal development is provided.
9. How the Ground Truth for the Training Set Was Established
As explained above, the concept of a separate "ground truth" for a training set (as in AI/ML) is not explicitly detailed. The TS400C instrument's calibration and optimization would have been performed by the manufacturer, likely using internal standards, reference materials, and comparing against established laboratory methods, much like the predicate devices themselves would have been developed. This "ground truth" for development would involve the known values of calibrators and controls used to ensure the instrument's accuracy and linearity across its measuring range.
§ 864.5400 Coagulation instrument.
(a)
Identification. A coagulation instrument is an automated or semiautomated device used to determine the onset of clot formation for in vitro coagulation studies.(b)
Classification. Class II (special controls). A fibrometer or coagulation timer intended for use with a coagulation instrument is exempt from the premarket notification procedures in subpart E of part 807 of this chapter subject to the limitations in § 864.9.