Search Results

The FlowSaver Blood Return System is used with Inari Medical catheters and sheaths for autologous blood transfusion.

The FlowSaver Blood Return System accessory allows for autologous injection of aspirated blood from Inari Medical catheters and sheaths during embolectomy procedures by dual layer 40 µm/200 µm filtration to minimize intraprocedural blood loss.

The provided text describes the 510(k) summary for the Inari Medical FlowSaver Blood Return System. This document is a premarket notification for a medical device and therefore does not contain information about an AI/ML device that requires traditional acceptance criteria and performance studies for diagnostic accuracy.

The FlowSaver Blood Return System is an autotransfusion apparatus. The submission is for an expansion of indications for an already cleared predicate device (K221483), allowing its use with Inari Medical catheters and sheaths for autologous blood transfusion and with a 30 cc large bore syringe.

Therefore, many of the requested details such as acceptance criteria for diagnostic performance, sample size for test sets, expert qualifications, adjudication methods, MRMC studies, standalone performance, training set details, and ground truth establishment for an AI/ML system are not applicable to this type of device and submission.

Instead, the submission focuses on demonstrating that the modified device remains substantially equivalent to its predicate through non-clinical testing.

Here's the information that can be extracted or stated as not applicable based on the provided text:

1. A table of acceptance criteria and the reported device performance:

The document does not specify quantitative acceptance criteria in a typical "table" format for diagnostic performance. Instead, it states that "Mechanical hemolysis testing was performed to demonstrate substantial equivalence to the predicate." It also lists various leveraged non-clinical tests without specific, quantitative acceptance criteria numbers in this summary. The implicit acceptance criterion is that the device performs comparably to the predicate and meets safety/performance standards for its intended use.

2. Sample size used for the test set and the data provenance (e.g., country of origin of the data, retrospective or prospective):

Not applicable for diagnostic accuracy as this is not an AI/ML diagnostic device. The document mentions "Design verification testing was not required to support substantial equivalence for the expanded indications" and lists leveraged tests from previous submissions (K210176 and K221483). These are likely engineering and performance tests on physical devices rather than data from patient studies.

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, as this is not an AI/ML diagnostic device requiring expert interpretation for ground truth.

4. Adjudication method (e.g. 2+1, 3+1, none) for the test set:

Not applicable, as this is not an AI/ML diagnostic device requiring adjudication.

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, as this is not an AI/ML device that assists human readers.

6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done:

Not applicable, as this is a physical medical device (autotransfusion apparatus), not an algorithm.

7. The type of ground truth used (expert consensus, pathology, outcomes data, etc.):

Not applicable, as this is not an AI/ML diagnostic device requiring ground truth for diagnostic accuracy. The "ground truth" for this device would be established engineering and performance specifications and biological compatibility (e.g., hemolysis levels).

8. The sample size for the training set:

Not applicable, as this is not an AI/ML device.

9. How the ground truth for the training set was established:

Not applicable, as this is not an AI/ML device.

Summary of Device Performance (Based on provided text, not in typical AI/ML format):

The device is deemed to meet acceptance criteria through a demonstration of substantial equivalence to an already cleared predicate device (K221483). This is supported by:

• Identical intended use, principles of operation, fundamental scientific technology, and technological characteristics compared to the predicate device.
• Non-clinical testing leveraging previous submissions (K210176 and K221483):
  • Visual Inspection
  • Dimensional Inspection
  • Engagement & Disengagement Force Testing
  • Flow Rate Testing
  • Media Integrity testing
  • Leakage Testing
  • Vacuum Testing
  • Clot Burden Filtration Validation
  • Simulated Use and Tensile Testing
  • Simulated Use and Torque Testing
  • Burst Testing
  • Particulate Matter Determination
  • Filtration Efficiency
• Specific mechanical hemolysis testing performed for the current submission to demonstrate substantial equivalence to the predicate.
• Sterilization validation in accordance with ISO 11135:2014/Amd 1:2018 and AAMI TIR 28:2016 to achieve a sterility assurance level (SAL) of 10-6.

The conclusion states that "The verification testing results demonstrate that the proposed FlowSaver Blood Return System is substantially equivalent to the predicate device." This implies the device performed acceptably across all these tests, meeting the safety and performance standards equivalent to the predicate.

Ask a specific question about this device

The FlowSaver Blood Return System is used with Triever Catheters for autologous blood transfusion.

The FlowSaver Blood Return System accessory allows for autologous injection of aspirated blood from the FlowTriever Retrieval/Aspiration System embolectomy procedure by dual layer 40 µ/200 µ filtration to minimize intraprocedural blood loss.

The provided document is an FDA 510(k) clearance letter for the FlowSaver Blood Return System. It primarily focuses on regulatory approval based on demonstrating substantial equivalence to a predicate device. It briefly mentions "Non-Clinical Testing" and "Clinical Testing" as part of the evidence presented for substantial equivalence.

However, the document does not contain detailed information about acceptance criteria or the specific study design, performance metrics, sample sizes, expert qualifications, or ground truth establishment methods for a new AI/ML-driven device. The request asks for details typically found in the clinical study section of a 510(k) submission or a peer-reviewed publication for an AI/ML medical device.

The "Clinical Testing" section states: "Post-market clinical data from PEERLESS (NCT05111613), FLAME (NCT04795167), and FLASH (NCT03761173) studies demonstrated the FlowSaver's filtration efficiency and safety without use of the second filter in support of substantial equivalence." This refers to clinical studies on the device's performance in a clinical setting, but it does not describe a study to prove the device meets acceptance criteria in the context of an AI/ML algorithm or a new diagnostic tool where performance metrics like sensitivity, specificity, or AUC are typically evaluated. The "acceptance criteria" mentioned in the request, in the context of AI, usually refers to pre-defined thresholds for these performance metrics.

Therefore, many of the requested details cannot be extracted from this document, as it outlines the regulatory approval process for a physical medical device with a labeling change, not a software-as-a-medical-device (SaMD) or AI-enabled diagnostic tool requiring specific AI model validation studies.

Here's a breakdown of what can be extracted and what cannot:

Information Present in the Document:

• 1. A table of acceptance criteria and the reported device performance:
  • The document implies acceptance criteria were met for "Filtration Efficiency" in non-clinical testing.
  • For clinical studies, it states the studies "demonstrated the FlowSaver's filtration efficiency and safety without use of the second filter."
  • However, no specific numerical acceptance criteria (e.g., minimum filtration percentage) or detailed performance results tables are provided.
• 2. Sample sizes used for the test set and the data provenance:
  • The document mentions clinical studies: PEERLESS (NCT05111613), FLAME (NCT04795167), and FLASH (NCT03761173).
  • Sample sizes (NCT numbers are provided, which could be looked up, but not directly stated in the document for the purpose of this submission).
  • Data provenance (country of origin, retrospective/prospective) is not explicitly stated in this document but would be detailed in the full clinical study protocols.
• 7. The type of ground truth used:
  • For filtration efficiency, the ground truth would be the measurement of particles/cellular components filtered out.
  • For safety, it would be adverse event reporting and clinical outcomes.
  • Not explicitly defined as "ground truth" in the AI/ML sense, but rather as measured performance and safety endpoints.

Information NOT Present in the Document (and likely irrelevant for this type of device clearance):

• 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): This is for expert-labeled ground truth, typically for image or signal interpretation. Not applicable here.
• 4. Adjudication method (e.g., 2+1, 3+1, none) for the test set: Not applicable as it's not an AI diagnostic study needing expert adjudication.
• 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: This applies to AI-assisted human reading. Not applicable here.
• 6. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable, as this is a physical device, not an algorithm.
• 8. The sample size for the training set: Not applicable, as no AI model is being "trained" in this context.
• 9. How the ground truth for the training set was established: Not applicable.

Summary of Device Acceptance and Study Information (Based on the Provided Text):

The FlowSaver Blood Return System is an autotransfusion apparatus that performs dual-layer 40 µ/200 µ filtration of aspirated blood during embolectomy procedures to minimize intraprocedural blood loss. The current 510(k) submission (K221483) is for a labeling change to remove the secondary filtration procedure, implying that the single-stage filtration by the FlowSaver alone is sufficient.

The acceptance criteria for the device, as evidenced by the studies, are related to its filtration efficiency and safety.

Table of Acceptance Criteria and Reported Device Performance:

Clinical Testing: Post-market clinical data from PEERLESS (NCT05111613), FLAME (NCT04795167), and FLASH (NCT03761173) studies "demonstrated the FlowSaver's filtration efficiency... without use of the second filter." No specific quantitative metrics (e.g., % of particles removed) are provided in this regulatory letter, but the implication is that the performance was acceptable to the FDA. |
| Safety (without secondary filter) | Clinical Testing: Post-market clinical data from PEERLESS (NCT05111613), FLAME (NCT04795167), and FLASH (NCT03761173) studies "demonstrated the FlowSaver's... safety without use of the second filter." This suggests an acceptable safety profile was observed in these studies, supporting the removal of the secondary filtration step from the instructions for use. |
| Functional Performance | Non-Clinical Testing: The document states that various tests were leveraged from the predicate device (K210176) and "demonstrated that all acceptance criteria were met; therefore, the device conforms to established product specifications." These tests include: Visual Inspection, Dimensional Inspection, Engagement & Disengagement Force Testing, Flow Rate Testing, Media Integrity testing, Leakage Testing, Vacuum Testing, Clot Burden Filtration Validation, Simulated Use and Tensile Testing, Simulated Use and Torque Testing, Burst Testing, Hematocrit Testing, Mechanical Hemolysis Testing. Numerical acceptance criteria and detailed performance of these tests are not provided in this summary. |

Study Information:

1. Sample size used for the test set and the data provenance:

   • Test Set Sample Size: Not explicitly stated as a single "test set" size for the purpose of this submission. The document references post-market clinical data from three studies: PEERLESS (NCT05111613), FLAME (NCT04795167), and FLASH (NCT03761173). The full sample size for these studies would need to be checked in their respective ClinicalTrials.gov entries or publications. The document implies that data from these studies collectively served as the clinical evidence.
   • Data Provenance: Not specified in the provided document (e.g., country of origin, retrospective or prospective). These details would be available in the protocols/publications of the mentioned clinical studies. The mention of "post-market clinical data" suggests the data was collected after the initial marketing of the device, likely prospectively as part of ongoing clinical trials.

2. Number of experts used to establish the ground truth for the test set and the qualifications of those experts: Not applicable. This is not an AI/ML diagnostic device requiring expert labeling of ground truth. The "ground truth" for this device's performance would be objective measurements of filtration efficiency (e.g., particle counts, volume filtered) and clinical outcomes related to safety.

3. Adjudication method (e.g. 2+1, 3+1, none) for the test set: Not applicable. This concept is typically relevant for studies validating AI models against expert interpretation, not for evaluating a physical filtration device.

4. 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 device is not an AI-assisted reading tool.

5. If a standalone (i.e. algorithm only without human-in-the-loop performance) was done: Not applicable. This is a physical medical device, not a standalone AI algorithm.

6. The type of ground truth used:

   • For filtration efficiency: Likely quantitative measurement of filtered components (e.g., blood cell counts, clot burden measurements) before and after filtration, or particle size analysis.
   • For safety: Clinical outcomes data, adverse event reporting, and potentially hematological parameters (e.g., mechanical hemolysis testing mentioned in non-clinical tests).

7. The sample size for the training set: Not applicable. This is not an AI/ML device where a "training set" for an algorithm is used.

8. How the ground truth for the training set was established: Not applicable, for the same reason as above.

Ask a specific question about this device