(176 days)
The ACIST Rapid Exchange (RXi) System is indicated for obtaining intravascular pressure measurements for use in the diagnosis and treatment of coronary and peripheral artery disease. The Navvus Catheter is intended for use with the ACIST RXi System.
The Rapid Exchange (RXi) System and Navvus Catheter are intended to provide hemodynamic measurement information for use in the diagnosis and treatment of coronary or peripheral artery disease. When used in catheterization and related cardiovascular specialty laboratories, the Rapid Exchange (RXi) System and Navvus Catheter will compute and display physiological parameters based on the output of an optically-based pressure measuring sensor.
The Rapid Exchange (RXi) System and Navvus Catheter consist of a single patient use, catheter-based sensing device, a hardware system containing embedded software, a user interface touchscreen, and electronics for converting measured pressure signals into Fractional Flow Reserve (FFR) measurements. The intravascular blood pressure sensor is optically-based and adhered to the end of a fiber optic. The sensor and fiber optic are integrated into a monorail catheter, which connects to the system to deliver the measured pressure data.
Here's a summary of the acceptance criteria and the study that proves the device meets them, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance
| Acceptance Criteria | Reported Device Performance (ACIST Rapid Exchange (RXi) System and Navvus Catheter) |
|---|---|
| Primary Objectives: | |
| Valid FFR measurement by Navvus Catheter when Radi takes valid FFR | 100% successful in taking valid FFR measurements in all lesions where Radi took valid FFR measurements. |
| Secondary Criteria (from Bland-Altman, Kolmogorov-Smirnov, Passing-Bablok analysis): | |
| Bias (Bland-Altman) | -0.01 |
| 95% Limits of Agreement (Bland-Altman) | -0.13 to 0.10 |
| Linearity (Kolmogorov-Smirnov test, p-value > 0.1) | Cumulative p-value of 0.72 (met success criteria) |
| Intercept (Passing-Bablok linear regression, 95% CI containing 0) | 95% CI of -0.15 to 0.19 (met success criteria) |
| Slope (Passing-Bablok linear regression, 95% CI containing 1) | 95% CI of 0.76 to 1.17 (met success criteria) |
| AUC (ROC, vs. first Radi FFR) | 87.4% |
| Rate of drift (mean) | 0.02 ± 0.02 |
| Rate of clinically relevant drift | 13% |
| Safety Profile | Very good safety profile, no serious adverse events or adverse events related to the RXi system. |
2. Sample Size Used for the Test Set and Data Provenance
- Sample Size: 58 subjects.
- Data Provenance: The study was a prospective, observational, multi-center study. The text does not explicitly state the country of origin, but given the submitter's address (Eden Prairie, MN) and the FDA filing, it's highly likely to be a US-based study.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of Those Experts
The ground truth was established by comparison to a predicate device, the Radi Analyzer® and PressureWire® Sensor, within the same subjects. Therefore, the "ground truth" for the test set was not established by a panel of human experts in the traditional sense, but rather through paired measurements against an already legally marketed and accepted device. The study collected FFR measurements directly, so the "experts" were the interventional cardiologists performing the procedures and interpreting the real-time FFR data from both devices. Their qualifications are implicitly that they are clinicians capable of performing these procedures.
4. Adjudication Method for the Test Set
There was no explicit adjudication method described for the test set in the context of expert review. The study design involved directly comparing the Navvus Catheter's FFR measurements to those obtained concurrently from the Radi PressureWire Sensor within the same subject. The performance measures (Bland-Altman, Passing-Bablok, etc.) are statistical comparisons between these two measurements, not an adjudication of discrepancy by a separate expert panel.
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
No, a Multi-Reader Multi-Case (MRMC) comparative effectiveness study was not done. This device is a measurement tool (a catheter and system for obtaining FFR), not an AI-powered diagnostic imaging tool that assists human readers. Therefore, the concept of "human readers improving with AI vs. without AI assistance" does not apply here.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) Was Done
The study described is essentially a standalone performance assessment of the device (RXi System and Navvus Catheter) in generating FFR measurements compared to a predicate device. While a human (the physician) is involved in positioning the catheter and viewing the output, the FFR measurement itself is derived by the device's embedded software. The clinical study directly evaluated the performance of this system against the predicate, which can be considered its standalone performance relative to the established gold standard (the predicate device for FFR measurement).
7. The Type of Ground Truth Used
The ground truth used was the Fractional Flow Reserve (FFR) measurements obtained from the Radi Analyzer® and PressureWire® Sensor (the predicate device). This is a comparative effectiveness study where the predicate device acts as the reference standard for FFR measurement.
8. The Sample Size for the Training Set
The document does not mention a specific training set size. This appears to be a 510(k) submission for a medical device that measures physiological parameters. Training sets are typically relevant for machine learning algorithms. While the device contains "embedded software" for converting signals to FFR, the submission focuses on demonstrating substantial equivalence to a predicate device through non-clinical and clinical testing, rather than an AI/ML development and validation process that would typically involve distinct training and test sets as primary components of the submission. The "training" for such a system would typically involve engineering calibration and validation against known physical standards.
9. How the Ground Truth for the Training Set Was Established
As no specific training set for a machine learning algorithm is detailed, the concept of establishing ground truth for a training set in this context is not directly applicable. If the embedded software involved parameters that were "trained," this would likely have been done through internal engineering testing and calibration against physical pressure standards and potentially in vivo or ex vivo models, verified against known true values. The documentation provided focuses on the clinical and non-clinical validation against the predicate device.
§ 870.2870 Catheter tip pressure transducer.
(a)
Identification. A catheter tip pressure transducer is a device incorporated into the distal end of a catheter. When placed in the bloodstream, its mechanical or electrical properties change in relation to changes in blood pressure. These changes are transmitted to accessory equipment for processing.(b)
Classification. Class II (performance standards).