The Vigilance CCO/SvO2 Monitor is a microprocessor-based instrument which. when connected to a Baxter thermodilution catheter, measures mixed venous oxygen saturation and cardiac output both continuously (CCO) and by the intermittent bolus (injectate) method (ICO). The Vigilance monitor measures cardiac output continuously by injecting small pulses of electrical power into the blood and recording the corresponding blood temperature changes via the catheter. The software based CCO algorithm within the monitor converts these power and blood temperature measurements into an estimate of cardiac output.

AI/ML Overview

Here's an analysis of the provided text regarding the acceptance criteria and study for the Baxter Vigilance CCO/SvO2 Monitor:

No acceptance criteria or study results are explicitly detailed in the provided text in the format requested. The document is a 510(k) summary, which focuses on demonstrating substantial equivalence to a predicate device rather than presenting a formal study with detailed acceptance criteria and performance metrics for a novel device.

The document states: "Bench testing was performed to show that the software requirements were met and that the modified Vigilance CCO/SvO2 Monitor performs comparably to the unmodified (predicate) device." and "The results of the bench testing demonstrate that the modified Baxter Vigilance CCO/SvO2 Monitor meets the performance requirements and its performance is comparable to the predicate device." However, the specific performance requirements or acceptance criteria (e.g., "accuracy within X%," "reproducibility less than Y standard deviation") are not listed. The "reported device performance" is summarized as being "comparable to the predicate device" and "meets the performance requirements," without providing quantitative data.

Therefore, many of the requested fields cannot be directly extracted from the provided text because it does not contain the detailed study design, acceptance criteria, and quantitative performance results typically found in a comprehensive clinical or validation study report.

Below is an attempt to populate the table and answer the questions based on the inference and limited information available in the text, highlighting where information is missing.

1. Table of Acceptance Criteria and Reported Device Performance

Acceptance Criteria (Quantitative/Qualitative)	Reported (Modified Device) Performance
Bench Testing:
Reproducibility of CCO software (No specific quantitative criterion provided)	Evaluated in a hydro flow model system and compared against the previous software version (predicate device). Results "demonstrate that the modified Baxter Vigilance CCO/SvO2 Monitor meets the performance requirements and its performance is comparable to the predicate device."
Response time of CCO software (No specific quantitative criterion provided)	Evaluated in a hydro flow model system and compared against the previous software version (predicate device). Results "demonstrate that [it] meets the performance requirements and its performance is comparable to the predicate device."
Ability of monitor to adapt to external noise (No specific quantitative criterion provided)	Evaluated in a hydro flow model system and compared against the previous software version (predicate device). Results "demonstrate that [it] meets the performance requirements and its performance is comparable to the predicate device."
Animal Testing (In Vivo Verification for CCO algorithm):
STAT and trend CCO estimates perform as required (No specific quantitative criterion provided, e.g., agreement with reference method)	Testing using two sheep "demonstrate that the STAT and trend CCO estimates perform as required." (Performance details, e.g., accuracy, bias, precision, compared to a gold standard, are not provided).
Overall Clinical/In Vivo Equivalence:
Modified algorithm performs as required and is substantially equivalent to the unmodified/predicate device. (No specific quantitative criteria provided)	"These evaluations [bench and in vivo] demonstrated that the modified algorithm performs as required and is substantially equivalent to the unmodified/predicate device." (No specific quantitative metrics for equivalence (e.g., mean difference, bland-altman limits of agreement) are provided in this summary).

A full, detailed study would include specific performance metrics like:

Accuracy: Mean difference (bias) and limits of agreement with a reference method (e.g., intermittent bolus thermodilution or Fick principle).
Precision/Reproducibility: Standard deviation or coefficient of variation of repeated measurements.
Response Time: Time to detect and display changes in cardiac output within a certain percentage.
Drift: Deviation over time.

None of these quantitative criteria or results are present in the summary.

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

Test Set Sample Size:
- Bench Testing: Not specified (e.g., number of test conditions, runs, data points within the "hydro flow model system").
- Animal Testing: Two sheep.
Data Provenance: Not specified (e.g., country of origin). Both bench and animal testing appear to be internal evaluations.
Retrospective/Prospective: Not specified, but animal testing would be considered prospective for the data collection in those animals. Bench testing is generally controlled experimental setup.

3. Number of experts used to establish the ground truth for the test set and the qualifications of those experts:

Not applicable / Not mentioned. The described testing (bench and animal) does not involve human experts establishing ground truth in the way a diagnostic imaging study would. The comparisons were against the predicate device's performance or expected physical/physiological responses.

4. Adjudication method for the test set:

Not applicable. No human experts were involved in adjudicating the ground truth for the test sets described.

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. This was not an MRMC study. The device is a monitor that provides a measurement; it does not involve human readers interpreting images or data that would be assisted by AI in the context of this 510(k) summary.

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

Yes, implicitly. The bench and animal testing evaluated the performance of the modified CCO algorithm and the monitor's readings independently of a human operator's interpretation or intervention, beyond connecting the device and observing its output. The 510(k) focuses on the algorithm's performance within the monitor.

7. The type of ground truth used:

Bench Testing: The ground truth for bench testing was likely based on known inputs or established reference measurements within the "hydro flow model system" and comparisons to the predicate device's performance.
Animal Testing: The ground truth was the physiological state of the animals, with the comparison being to the predicate device's performance or "as required" for STAT and trend CCO estimates. A true "gold standard" reference method for continuously monitoring cardiac output in vivo is not explicitly mentioned as being used for comparison (e.g., flow probes, direct Fick measurement), but rather the focus is on the modified algorithm performing "as required" and comparably to the predicate.

8. The sample size for the training set:

Not mentioned. The document describes modifications to an existing algorithm. It does not provide details on the development or training of the original or modified CCO algorithm. The focus is on demonstrating the performance of the modified algorithm.

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

Not mentioned. As the sample size for the training set is not provided, neither is how its ground truth was established.

Summary

{0}------------------------------------------------

Food and Drug Administration 510(k) Notification for the Baxter Vigilance CCO/SvO2 Monitor December 21, 1995 Page 24

K955-816

Appendix I

MAY – 6 1997 510(k) Summary

Product (Trade) Name

Vigilance Continuous Cardiac Output/Oximetry (CCO/SvO2) Monitor

VGS Model Number:

Common, Usual or Classification Name

な Cardiac Output/Dual Oximeter/Ejection Fraction Computer
Single-Function, Preprogrammed Diagnostic Computer (21 CFR 870.1435)

Device Classification

This generic device has been classified as Class II by the Circulatory Systems Devices Panel.

Reason for Submission

Baxter intends to market a modified version of the Vigilance CCO/SvO2 Monitor. The software algorithm in the monitor has been modified to enhance the CCO estimation process. No change has been made to other functions of the monitor system or to the hardware.

Predicate Device Identification

The Vigilance CCO/SvO2 Monitor is substantially equivalent to the Vigilance CCO/SvO2 Monitor which was cleared for marketing under premarket notifications K924452 and K940795.

Compliance with 513/514

No performance standards have been established under Section 514 of the Food. Drug and Cosmetic Act.

General Description, Components and Specifications

{1}------------------------------------------------

power and blood temperature measurements into an estimate of cardiac output. The Vigilance instrument displays the cardiac output in two modes:

1. The Standard (or trend) CCO monitoring user interface is an historical trend plot (i.e., cardiac output vs. time) that allows the user to see the trend of the CCO values over a period of time.
1. The STAT mode CCO display is a numeric presentation of the last ten STAT mode cardiac output estimates.

The Vigilance CCO/SvO2 Monitor that is the subject of this premarket notification varies from the predicate device in that it contains modifications to the CCO algorithm. These modifications enhance the CCO estimation process.

ﯿﺎﮞ No change has been made to other functions or features of the monitor system or to the monitor's components or hardware. In addition, the specifications for the software defined in the Operator's Manual have not changed as a result of this software modification.

Comparative Information

Labeling

The labels for the Vigilance CCO/SvQ2 Monitor have not changed as a result of the algorithm modification. However, modifications have been made to the Operator's Manual.

Intended Use

The Vigilance CCO/SvO2 Monitor is intended to measure both bolus/injectate and continuous cardiac output in addition to mixed venous oxygen saturation. The system also calculates hemodynamic and oxygenation parameters. The intended use of this monitor has not been changed with the modification to the algorithm that is the subject of this premarket notification and is the same as that of the predicate device.

Physical Characteristics

The physical characteristics/hardware of the Vigilance CCO/SvO2 Monitor have not changed as a result of the algorithm modification. No changes have been made to the circuit components or connectors for the monitors. The CCO specifications provided in the operator's manual have not changed. The changes made to the device are only in the algorithm to enhance CCO processing.

Anatomical Sites

The Vigilance CCO/SvO2 Monitor is connected to a thermodilution catheter which is placed in the pulmonary artery for right heart and pulmonary artery hemodynamic measurements. The site of use is the same as that for the predicate device.

{2}------------------------------------------------

Target Population

The target population for this product includes patients who require hemocynamic monitoring. The target population is the same as that for the predicate device.

Performance Testing

ু

1. Bench Testing
  Bench testing was performed to show that the software requirements were met and that the modified Vigilance CCO/SvO2 Monitor performs comparably to the unmodified (predicate) device. The reproducibility and response time of the modified Vigilance CCO software, as well as the ability of the monitor to adapt to external noise, were evaluated in a hydro flow model system and were compared against the previous software version (predicate device).

The results of the bench testing demonstrate that the modified Baxter Vigilance CCO/SvO2 Monitor meets the performance requirements and its performance is comparable to the predicate device.

1. Animal Testing
  The performance of the modified Vigilance CCO/SvO2 Monitor was demonstrated in the bench testing above. Evaluation of the device using two sheep was conducted to further verify the performance of the CCO algorithm in a simulated clinical environment. The results of the testing demonstrate that the STAT and trend CCO estimates perform as required.
1. Clinical Testing
  The modifications in the software algorithm were evaluated on the bench and in vivo. These evaluations demonstrated that the modified algorithm performs as required and is substantially equivalent to the unmodified/ predicate device. Clinical evaluation of the product is not considered to be necessary for this change.

Safety Characteristics

The testing presented for the modified product demonstrates that the Vigilance CCO/SvO2 Monitor is comparable to the predicate device in safety characteristics.

Software Validation and Verification

Hazard Analysis

Hazard analysis of the software modifications was conducted to evaluate the effect of the changes on the safety of the product. The hazard analysis evaluated the potential hazardous events, their cause, level of concern and method of control/corrective action. All hazards were reviewed and, where appropriate, testing was performed to ensure that the applicable corrective action occurs.

{3}------------------------------------------------

Level of Concern

This device has been determined to have a moderate level of concern based on the intended use and potential hazards of the device. The level of concern for the modified components of the software is identified in the Hazard Analysis. The components were determined to be of moderate or minor concern.

Development Documentation

Development of software at Baxter is conducted in accordance with Baxter's procedure for software development which lists the requirements for the life cvcle of software. All components of the software cycle were documented and

tested appropriately. The protocol and results for testing were documented and ー approved by all appropriate functional departments including Quality Engineering. Design reviews were conducted as required throughout the development cycle. Prior to initiating use of the modified software in production, a Document Change Request will be generated listing the changes made to the device and a justification for the changes. These changes will require the approval of the software R&D engineer and representatives from Quality Engineering, Regulatory Affairs and Manufacturing Engineering in accordance with Baxter procedures to ensure that all requirements have been met for the change.

Summary

The physical characteristics of the modified Vigilance CCO/SvO2 Monitor have not changed. The modifications in the software algorithm were evaluated on the bench and in vivo. These evaluations demonstrated that the modified algorithm performs as required and is substantially equivalent to the unmodified/predicate device.

here Imga Parra
Irene Penza Parker

Irene Ponzoa Parker Manager, Regulatory Affairs and Clinical Programs Edwards Critical-Care Division Baxter Healthcare Corporation

Regulation Number and Section

§ 870.1240 Flow-directed catheter.

(a)
Identification. A flow-directed catheter is a device that incorporates a gas-filled balloon to help direct the catheter to the desired position.(b)
Classification. Class II (performance standards).