(925 days)
INTENDED USE: LIDCO SYSTEM for CARDIAC OUTPUT MONITORING
Indications for Use - Diagnostic aid (cardiac output determination). The LiDCO System is intended for the monitoring of cardiac output (blood flow in litres per minute) in patients of greater than 88 lbs (40 kg) in weight.
The LIDCo System revisits and improves the indicator dilution technique of cardiac output determination. It avoids the complications of pulmonary artery catheterisation. The LIDCo System will have clinical utility in patients with pre-placed arterial and venous lines, where the determination of cardiac output is required minimally invasively without the insertion of a pulmonary artery catheter.
The lithium cation sensor is biocompatible, thermally compensated and unaffected by changes in arterial blood pressure.
Despite the potential complication of secondary marker recirculation, the LIDCo System software has been demonstrated in a number of bench and clinical studies to calculate the area under the primary indicator dilution curve.
Flow cell moulded from polycarbonate.
Lithium ion selective electrode made from PVC membrane cast onto polyurethane.
Blood contact material PVC - blood not returned to patient.
Monitor CM-31/32: Power supply is by + 5VDC AC to DC Power Supply Unit. Recorder and digital outputs by optical circuits and transformer.
Blood Withdrawal Pump CM-33: Connection with sensor interface is optically isolated. Power supply is by battery operation (6v DC). Low battery warning. Fail safe against arterial blood pressure. "One way" only insertion of pump tubing.
Dimensions & Weight: 320 mm x 260 mm x 40 mm,
Here's an analysis of the acceptance criteria and study proving device performance for the LiDCo System, based on the provided text:
1. Table of Acceptance Criteria and Reported Device Performance:
The document doesn't explicitly state "acceptance criteria" in a quantified manner. Instead, it presents "clinical investigations" to demonstrate close correlation with existing standard methods. The implicit acceptance criteria appear to be a strong correlation coefficient, low bias, and acceptable precision when compared to established methods.
| Acceptance Criteria (Implicit) | Reported Device Performance (LiDCo System) |
|---|---|
| High Correlation with Pulmonary Artery Catheter (PAC) | 0.963 |
| Low Bias compared to Pulmonary Artery Catheter (PAC) | 0.19 L/Min |
| Acceptable Precision compared to Pulmonary Artery Catheter (PAC) | 0.38 L/Min |
| High Correlation with Continuous Cardiac Output (CCO) | 0.949 |
| Low Bias compared to Continuous Cardiac Output (CCO) | 0.01 L/Min |
| Acceptable Precision compared to Continuous Cardiac Output (CCO) | 0.55 L/Min |
2. Sample Size Used for the Test Set and Data Provenance:
- Test Set (PAC Comparison): 51 patients
- Test Set (CCO Comparison): 11 patients
- Data Provenance: The document does not explicitly state the country of origin or whether the data was retrospective or prospective. It refers to "Clinical investigations" and "clinical studies," suggesting these were conducted in a clinical setting. Given the company is based in London, UK, it's likely the studies were conducted there. The timing implies these were conducted prior to the submission date of the 510(k) in June 1996, likely making them prospective studies designed to support the regulatory submission.
3. Number of Experts Used to Establish Ground Truth for the Test Set and Their Qualifications:
The document implicitly uses existing, established clinical methods (Pulmonary Artery Catheter and Continuous Cardiac Output monitoring) as the "ground truth." It does not mention a panel of human experts explicitly establishing ground truth for the test set.
4. Adjudication Method for the Test Set:
Not applicable. The ground truth was established by direct comparison with measurements obtained from established medical devices (Pulmonary Artery Catheter and Continuous Cardiac Output), rather than by expert consensus requiring adjudication.
5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study:
No, an MRMC comparative effectiveness study was not done. The LiDCo System is a medical device for measuring cardiac output, not an AI-assisted diagnostic tool that would typically involve human readers. Therefore, the concept of improving human reader performance with AI assistance is not relevant here.
6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study:
Yes, the performance presented for the LiDCo System in comparison to predicate devices is a standalone performance. The LiDCo System itself calculates the cardiac output based on its indicator dilution technique and software, without continuous human intervention or interpretation that would typically define a "human-in-the-loop" scenario in AI/diagnostic device contexts. The human operator injects the indicator and initiates the measurement, but the calculation is performed by the device's software.
7. Type of Ground Truth Used:
The ground truth used was comparison with established medical device measurements. Specifically:
- Pulmonary Artery Catheter (a widely accepted invasive method for cardiac output measurement).
- Continuous Cardiac Output (another established method, likely derived from PAC or similar continuous monitoring).
8. Sample Size for the Training Set:
The document does not explicitly mention a "training set" in the context of machine learning or AI development. The "software" is mentioned as having been "demonstrated in a number of bench and clinical studies to calculate the area under the primary indicator dilution curve." This implies iterative development and testing, but not a distinct 'training set' as understood in current AI contexts. The clinical investigation data (51 and 11 patients) would be considered the validation data for regulatory submission.
9. How the Ground Truth for the Training Set Was Established:
Since there's no explicit "training set" described as in modern AI development, this question is not fully applicable. However, the "bench and clinical studies" that informed the software's development would have used similar methodologies to establish what a correct cardiac output measurement should be (e.g., using known flow rates in bench models or measurements from reference devices in clinical settings) to refine the software's calculation algorithms.
§ 870.1435 Single-function, preprogrammed diagnostic computer.
(a)
Identification. A single-function, preprogrammed diagnostic computer is a hard-wired computer that calculates a specific physiological or blood-flow parameter based on information obtained from one or more electrodes, transducers, or measuring devices.(b)
Classification. Class II (performance standards).