Summary

Nox Medical has identified an issue where the cannula Flow signal may appear asynchronous relative to other recorded signals from Nox Recorders in Noxturnal. When this occurs, the cannula Flow channel may not align correctly with other signals. 

Over the past period, we’ve taken a close look this and made it one of our highest priorities. Our focus has been on finding the root cause of the asynchronous flow behaviour. Investigation has identified contributing factors, including hardware-related behavior under certain conditions.

To maintain transparency and support our customers, we are providing guidance on the currently recommended workaround for scoring.

Observed Behavior

On rare occurrences, in certain recordings, the following may be observed:

• The cannula Flow signal appears offset or asynchronous compared to other signals.
• Respiratory events may be difficult to score using the cannula Flow signal.
• Other signals (e.g., RIP belts - including cRIP Flow and RIP Flow, oximetry) remain synchronized.

Recommended Workaround

Until a released fix,  and in cases where users are noticing the signals are not in sync, users should:

Use the RIP flow or the cRIP flow (respiratory inductance plethysmography) derived from the respiratory belts as the primary reference for respiratory scoring.

The RIP signals remains correctly synchronized with the rest of the signals and can be used reliably for respiratory event scoring.

Material on the Backup Signal 

While we are not aware of articles that directly compare the accuracy of cRIP-based respiratory event detection to other airflow measures, we have strong evidence supporting its reliability and utility as a backup signal:

• Feasibility and Utility: A recent article (Hansen, et al. 2023, https://doi.org/10.1111/apa.16964) on home polygraphy with Nox T3/T3s showed that using the cRIP flow as a backup to the cannula allowed 26 more sleep studies to be scored. The mean signal quality for the thoracic and abdominal RIP signals was very high (~98-99%).
• High Signal Quality (Pediatrics): Multiple studies in pediatric populations have demonstrated exceptional RIP signal quality, often outperforming the nasal cannula. For example, Olmo Arroyo, et al. 2021 (https://doi.org/10.1111/jsr.13241) reported RIP signals present for 95-99% of recording time, compared to 64% for the cannula. Ioan et al. (2020 & 2023) similarly showed RIP signals were artefact-free for 81-100% of Total Sleep Time in all children studied.
• Internal Reliability Study: We also have an internal study on the reliability of the RIP belts: DOI: https://pubmed.ncbi.nlm.nih.gov/33411184/

Additionally, over 20 published studies using Nox devices with RIP belts in pediatric populations report issues with nasal cannula and/or oximeter signals, but none report concerns with RIP belt signal quality or tolerability.

Overall, the available evidence supports RIP belts as a stable, high-quality, and reliable alternative when primary airflow measurements are unavailable.

AASM Scoring Guidance

The use of RIP flow (calibrated or uncalibrated) is consistent with the recommendations in the:

American Academy of Sleep Medicine Version 3 (AASM) Manual for the Scoring of Sleep and Associated Events – Rules for Adults

Specifically, RIP flow is included as a recommended signal under Section VIII – Respiratory Rules for Adults.

Effects On Safety

The offset in time is large enough to be clearly identifiable by a trained professional when the cannula Flow signal is compared with other signals recorded from the same device. 

Based on intended use, mandatory manual review, visibility of the desync, and availability of RIP/cRIP as an alternate respiratory signal, the residual patient risk is low. This  behavior maps to  existing risk-controls present in the Risk Management File for the Product. The conclusion is the residual risk is acceptable.

When Noxturnal is used as intended, the residual risk is considered low and acceptable due to mandatory manual verification, detectability of the desynchronization during review, and availability of RIP/cRIP as an alternative respiratory signal.

Hardware Status

Investigation has identified contributing factors, including hardware-related behavior under certain conditions. For devices experiencing the desynchronization issue, they can be returned through the RMA process for replacement.

Purpose of This Notice

This communication is intended to provide:

• Transparency regarding a known issue
• Clear guidance on the recommended workaround
• A standardized message that can be shared with customers and support teams

Sincerely,

Nox Medical