Search Results

The iv Watch Model 400 is indicated for the detection of subcutaneous infiltrations and extravasations of 10 cc or less of optically clear infusates, as an adjunctive device to the clinical evaluation in the healthcare setting of adults and pediatrics with peripherally-inserted catheters (PIVs). The device is indicated to assess patients for the subcutaneous infiltrations and extravasations but should not serve as a substitute for regular clinician assessment of the PIV site. The ivWatch Model 400 is intended for use by healthcare practitioners who have been trained in the use of the device.

The ivWatch Model 400 is a medical device that provides continuous, non-invasive monitoring of human tissue adjacent to peripheral intravenous (PIV) insertion sites to aid in the early detection of infiltration and extravasation events. The ivWatch Model 400 consists of the ivWatch Patient Monitor (IPM), a reusable optical sensor cable and a single-use sensor receptacle.

The ivWatch Model 400 uses visible and near-infrared light to measure changes in the optical properties of the tissue near a PIV insertion site. The IPM contains an optical system that generates visible and near-infrared light signals that are sent through the optical sensor cable to the patient's skin. Simultaneously, the IPM measures the light reflected back through the optical sensor cable from the patient's skin. Measured changes between the emitted and reflected signals are processed by the ivWatch signal processing algorithm to determine if an infiltration event may have occurred. If changes in the optical properties of the tissue near the PIV insertion site are consistent with an infusate pooling in the subcutaneous tissue, the IPM emits audible and visual notifications intended to prompt the clinician to inspect the PIV site for a possible infiltration event.

Here's a breakdown of the acceptance criteria and study information for the ivWatch Model 400, based on the provided document:

1. Acceptance Criteria and Reported Device Performance

The core of this submission is to expand the device's indications for use to include pediatric patients, demonstrating that the device is "just as safe and effective" in this population as it was in the adult population (the predicate device's cleared indication). The document focuses on demonstrating non-inferiority for the pediatric population compared to the predicate device's performance in adults.

Note: The FDA letter explicitly notes this is a 510(k) submission, meaning the acceptance criteria are generally to demonstrate substantial equivalence to a legally marketed predicate device rather than meeting specific performance thresholds for a novel device. The study aims to show that extending the indication to pediatrics does not negatively impact the device's safety and effectiveness compared to its performance in adults.

2. Sample Size for the Test Set and Data Provenance

• Sample Size: Not explicitly stated as a single number but derived from the description: "Safety and efficacy of the Model 400 has been demonstrated in accepted case studies representing an age distribution from 2 weeks to 17 years, patient weight measurements from 3.3 to 196.7 kg (mean, 34.5 kg), distributed skin pigmentations and similar proportions between male and female subjects."
  • The "95% CI" values for sensitivity (56.5% to 92.5%) for the pediatric population suggests a smaller sample size compared to the predicate's sensitivity (91.4% to 98.7%). While precise numbers aren't given, the width of the confidence interval usually correlates inversely with sample size.
• Data Provenance: The study was a "pediatric IRB approved clinical study," indicating it was prospective clinical data from a real-world healthcare setting. The country of origin is not explicitly stated, but typically FDA submissions for US market clearance involve studies conducted in the US or in compliance with international guidelines acceptable to the FDA.

3. Number of Experts Used to Establish the Ground Truth for the Test Set and Qualifications

The document refers to "clinician detection" as the benchmark for comparison.

• Number of Experts: Not explicitly stated as a fixed number for ground truth establishment. Instead, the "clinician" is the reference point for independent diagnosis.
• Qualifications: "Clinician" typically refers to trained medical professionals (e.g., nurses, doctors) responsible for patient care and PIV site assessment. No specific expert qualifications (e.g., "radiologist with 10 years of experience") are provided for establishing the ground truth of infiltration events. The normal standard of care for PIV assessment by healthcare professionals appears to be the implied ground truth.

4. Adjudication Method for the Test Set

Not specified. The phrasing "prior to the clinician diagnosing the infiltration" suggests that individual clinicians' assessments were taken as the reference point for "ground truth" rather than a consensus or adjudicated panel. The "blinded non-alarming group" indicates that clinician detection was independent of the device's alarm for a portion of the study.

5. Multi-Reader Multi-Case (MRMC) Comparative Effectiveness Study

No, an MRMC comparative effectiveness study was not explicitly mentioned or presented in the context of comparing human readers with and without AI assistance. The study compares the device's detection against "clinician detection," implying a focus on the device's standalone performance or its ability to assist the clinician by alarming earlier. The study design doesn't appear to be set up to measure improvement in human reader performance directly with AI assistance.

6. Standalone (Algorithm Only Without Human-in-the-Loop Performance) Study

Yes, implicitly. The performance metrics such as "ivWatch detected 80.0% of infiltrations... before detection by a clinician" and "Sensitivity: 78.3% of infiltrations (95% CI [56.5 to 92.5%]) were detected prior to the clinician diagnosing the infiltration in the blinded non-alarming group" are measures of the algorithm's ability to detect events independently of, and often before, explicit human recognition, even if the device's ultimate intended use is as an adjunctive tool. The "blinded non-alarming group" specifically evaluates the device's detection capability when its alerts are not influencing the clinician.

7. Type of Ground Truth Used

The ground truth used was clinician diagnosis of an infiltration event, based on standard clinical assessment. This is implicitly the "gold standard" against which the device's detection capability is measured. There is no mention of pathology, imaging, or specific objective outcomes data being used as the primary ground truth.

8. Sample Size for the Training Set

The document does not provide specific details on the sample size for the training set. It describes the device's general principle of operation (optical changes in tissue processed by an algorithm) but does not delineate between training and testing data for the algorithm development itself. This submission refers to clinical performance testing for validation of the device in a new population, not the initial development of the algorithm.

9. How the Ground Truth for the Training Set Was Established

Similarly, the document does not describe how the ground truth for any potential training set was established. The focus of this specific 510(k) is on the clinical validation of the already developed ivWatch Model 400 for an expanded indication (pediatrics), building upon its predicate clearance. Information about algorithm training and its associated ground truth would typically be part of an earlier submission or internal documentation.

Ask a specific question about this device