(73 days)
There are no changes to the intended use of the IABP for which a 510(k) was approved in October 1992 [510(k) number K915580]. The balloon pump is an electromechanical system used to drive intra-aortic balloons. It provides temporary support to the left ventricle via the principle of counterpulsation. The intra-aortic balloon is placed in the descending aorta, just distal to the left subclavian artery. Once the balloon is positioned, the pump is adjusted to trigger in synchrony with the ECG or arterial pressure waveform to ensure that inflation and deflation occur at the appropriate points during the cardiac cycle.
Belmont Instrument Corporation intends to revise software used in the Intra-Aortic Balloon Pump (IABP). There are no other changes to the IABP for which a 510(k) was approved in October 1992. The revised software provides a more convenient way for Operators of the pump to set the ECG gain. In particular, the change will accommodate high gain settings by the operator, in which the ECG R-Wave amplitude is set far higher than the triggering threshold. Setting the ECG gain too high can lead to noise affecting triggering. In the existing software, the thresholds for ECG R-Wave detection, pacer spike detection, and noise detection are all fixed values. In the upgraded version, the R-Wave trigger threshold is scaled to a moving average of prior R-Wave amplitudes with the most recent R-Wave amplitude weighted the most heavily. The pacer detection and noise detection threshold are. in turn, scaled to the new R-Wave threshold.
Here's an analysis of the provided text regarding the acceptance criteria and study for the Belmont Portable Balloon Pump's revised software:
This document, K982538, describes a software revision for an Intra-Aortic Balloon Pump (IABP) and focuses on demonstrating its substantial equivalence to the previously approved IABP software. The study presented here is not a complex clinical trial but rather a series of verification and validation tests for the software functionality.
1. Table of Acceptance Criteria and Reported Device Performance
The acceptance criteria are not explicitly listed in a table format in the original text, but they can be inferred from the "Summary of Nonclinical Tests and Results" section. The reported device performance is presented as a confirmation that these criteria were met.
| Acceptance Criteria (Inferred from Section 12) | Reported Device Performance |
|---|---|
| A. Software algorithms satisfied quantitatively | Verification of software algorithms was performed and deemed satisfied. |
| B. Effective triggering over a heart rate range of 30 - 200 BPM | The revised software enables the IABP to trigger effectively over the full specified range of heart rate, 30 - 200 BPM. |
| C. Effective triggering over a wide range of clinical conditions, including arrhythmias and varied ECG morphologies | The revised software enables IABP to trigger effectively over a wide range of clinical conditions as demonstrated by performance with the American Heart Association ECG Database, and by deliberately synthesizing specific arrhythmias, and ECG morphologies, using ECG Arrhythmia Simulators, and clinical tapes. |
| D. Improved performance at very high ECG gain settings and proper function at low gain, including lower amplitude ECG than current software | The revised software demonstrates improved performance at very high ECG gain setting, and still function properly at low gain, including lower amplitude ECG than is possible with the current (unmodified) software. |
| E. Rapid and effective response to sudden changes in ECG gain from Operator | The revised software is designed to function in the present Belmont PBP system, with an ECG gain control set by the Operator. The system must therefore, respond quickly and effectively to sudden changes in ECG gain. |
| F. All safety features function properly. | All safety features function properly. |
| G. System stops operation safely at all identified adverse conditions. | The system will stop operation in a safe manner at all identified conditions which could adversely affect safety or efficacy. |
| H. System passes all applicable tests defined in "Final Functional Test Procedure, PBP-Q108". | The system passes all applicable tests using the "Final Functional Test Procedure, PBP-Q108". |
2. Sample Size Used for the Test Set and Data Provenance
The document does not explicitly state a numerical sample size for the test set in terms of individual patients or ECG recordings. Instead, it refers to:
- American Heart Association ECG Database: This implies a retrospective collection of diverse ECG data.
- Deliberately synthesizing specific arrhythmias and ECG morphologies: This indicates simulated data, likely prospective in its generation for testing purposes.
- Clinical tapes: This suggests retrospective recordings of patient ECGs.
The country of origin for the AHA database and clinical tapes is not specified, but typically the AHA database originates from the US. The 'clinical tapes' could be from various sources.
3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications of those Experts
The document does not mention the use of human experts to establish ground truth for the test set. The focus is on the algorithmic detection of ECG R-waves, pacer spikes, and noise, and comparison against expected behavior given the input ECG signals (from databases or simulators).
4. Adjudication Method for the Test Set
No adjudication method is described, as the evaluation primarily relies on the physical and electrical characteristics of the ECG signals and the expected behavior of the algorithm, rather than human interpretation.
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 study is focused on the performance of the software itself in processing ECG signals, not on the improvement of human reader performance with or without AI assistance.
6. If a Standalone (i.e., algorithm only without human-in-the-loop performance) was done
Yes, a standalone study of the algorithm's performance was done. The "Summary of Nonclinical Tests and Results" directly describes the testing and verification of the revised software's ability to:
- Satisfy software algorithms quantitatively.
- Trigger effectively over specified heart rates.
- Trigger effectively over a wide range of clinical conditions using databases and simulators.
- Demonstrate improved performance at high ECG gain and proper function at low gain.
- Respond quickly and effectively to sudden changes in ECG gain.
- Ensure safety features function correctly and stop operation safely.
- Pass system tests.
These are all evaluations of the algorithm's intrinsic performance without human interaction during the operational testing itself.
7. The Type of Ground Truth Used
The ground truth for the performance tests appears to be:
- Reference ECG Data: Provided by recognized databases like the American Heart Association ECG Database. These databases typically have annotations or known characteristics of the ECG signals.
- Synthesized Arrhythmias and ECG Morphologies: Created using ECG Arrhythmia Simulators, where the "true" characteristics (e.g., presence and timing of R-waves, pacer spikes, noise) are precisely known by design.
- Clinical Tapes: Real-world ECG recordings, likely with established truth based on the recording conditions and potentially expert annotation from their original use, though this is not explicitly detailed.
8. The Sample Size for the Training Set
The document does not specify a sample size for a training set. This is because the software revision described is primarily an algorithmic change to how thresholds are scaled, not a machine learning model that requires a distinct training and testing phase. The "algorithm" itself is a set of predefined rules and calculations.
9. How the Ground Truth for the Training Set Was Established
Since there is no mention of a separate training set, the concept of establishing ground truth for a training set is not applicable in this context. The core change is a deterministic algorithmic adjustment (scaling R-wave trigger threshold to a moving average, and subsequent scaling of pacer and noise detection thresholds), not a learned model from data. The tests described are validation tests against known or synthesized ECG patterns.
§ 870.3535 Intra-aortic balloon and control system.
(a)
Identification. An intra-aortic balloon and control system is a prescription device that consists of an inflatable balloon, which is placed in the aorta to improve cardiovascular functioning during certain life-threatening emergencies, and a control system for regulating the inflation and deflation of the balloon. The control system, which monitors and is synchronized with the electrocardiogram, provides a means for setting the inflation and deflation of the balloon with the cardiac cycle.(b)
Classification. (1) Class II (special controls) when the device is indicated for acute coronary syndrome, cardiac and non-cardiac surgery, or complications of heart failure. The special controls for this device are:(i) Appropriate analysis and non-clinical testing must be conducted to validate electromagnetic compatibility and electrical safety of the device;
(ii) Software verification, validation, and hazard analysis must be performed;
(iii) The device must be demonstrated to be biocompatible;
(iv) Sterility and shelf-life testing must demonstrate the sterility of patient-contacting components and the shelf life of these components;
(v) Non-clinical performance evaluation of the device must demonstrate mechanical integrity, durability, and reliability to support its intended purpose; and
(vi) Labeling must include a detailed summary of the device- and procedure-related complications pertinent to use of the device.
(2) Class III (premarket approval) when the device is indicated for septic shock and pulsatile flow generation.
(c)
Date premarket approval application (PMA) or notice of completion of product development protocol (PDP) is required. A PMA or notice of completion of a PDP is required to be filed with the Food and Drug Administration on or before March 31, 2014, for any intra-aortic balloon and control system indicated for septic shock or pulsatile flow generation that was in commercial distribution before May 28, 1976, or that has, on or before March 31, 2014, been found to be substantially equivalent to any intra-aortic balloon and control system indicated for septic shock or pulsatile flow generation that was in commercial distribution before May 28, 1976. Any other intra-aortic balloon and control system indicated for septic shock or pulsatile flow generation shall have an approved PMA or declared completed PDP in effect before being placed in commercial distribution.