This article describes the workflow for LM and PLMS analysis in Noxturnal. Users will learn how to use the auto-analysis for LM and PLMS scoring and methods to score these events manually. 

Scoring LM and PLMS in Noxturnal

Although the AASM rules outline two events to score, LM and PLMs, Noxturnal features three events to aid in scoring and visual review of limb movements. The three events are: 

1. LM - Limb movements
   1. Noxturnal considers these events candidates for periodic limb movements in Sleep (PLMS). 
2. PLMS - Periodic Limb Movement in Sleep: 
   1. LM events that qualify as PLM are converted to PLMS events (see auto-analysis section for more details). 
3. PLM - Periodic Limb Movement: 
   1. Shows a connecting event marker between a series of PLMS events to aid the visual representation of PLMS events. The PLM event is automatically scored on the Leg leg signal. 

Tip - it can be easier to visually differentiate between event types if you go into the event settings and change the color of the events. 

Methods for Auto-Analyzing LM and PLMS in Noxturnal

Noxturnal has two tools that aid users in scoring and updating LM and PLMS events. These are the PLM Auto-Analysis and the PLM Update tool. 

The Periodic Limb Movement (PLM) Analysis automatically assesses the limb signals for LM and PLMS events. The sequence of the auto-analysis process is:

1. Limb movements (LM) are scored on both leg EMG signals per the LM definition in the PLM Analysis settings. 
2. Then, the analysis works through the scored LMs and converts suitable events to PLMS if they meet the definition set in the PLM Analysis settings. 
3. The PLM Analysis then adds a new PLM event that links PLMS events to show the PLMS series in Noxturnal visually.  
4. The final step is the association of LM and PLMS events with arousals. This manual process requires the use of the Arousal Tool in Noxturnal. 

The second is the PLM Update tool. This tool runs through manually scored or edited auto-scored LM events and updates them based on PLM settings. This tool will be considered in the Manual Scoring of LM and PLMs in Noxturnal chapter. 

Noxturnal PLM Auto-Analysis

To use the PLM Analysis, go to Analysis and PLM. 

PLM Analysis Settings

The PLM analysis has several settings that can be adjusted. The image below shows the General Properties settings for the PLM Analysis in Noxturnal. It can be divided into three parts: 

1. Limb Movements (LM)
2. Periodic Limb Movements in Sleep (PLMS) 
3. Excluded LM from PLMS 

Limb Movement Settings

Limb Movement (LM) settings allow users to define limb movement (LM) events scored by the analysis. It includes various criteria for scoring LM events. These will be considered below with workflow examples. 

Users can define the LM event duration with the Minimum (1) and Maximum Duration (2) settings. It defaults to the AASM recommendations of a minimum of 0.5 seconds and a maximum of 10 seconds. 

The minimum inter-movement duration sets the minimum period between LM events (4). By default, Noxturnal requires a minimum 0.5-second interval between the end of one LM and the scoring of a new LM event. The auto-analysis considers this setting for each limb signal individually - it does not compare the inter-movement interval across both limbs. 

In the example below, the analysis does not mark the second LM event (2) on the Left leg signal because it is within 0.5s (0.317s in this illustrative example) of the initial LM (1). Because Noxturnal only considers the individual leg signals when scoring LM, the overlapping LM event is scored on the Right leg signal (3). Overlapping events will be discussed further in the section on combined leg scoring. 

The amplitude criteria are determined by the threshold type (3), the onset threshold (5), and the offset threshold (6) of an individual LM event. The Threshold Type (3) allows users to choose between relative or absolute threshold values for determining what qualifies as an LM event. The default analysis settings in Noxturnal use Relative Threshold, but users can adjust this by editing them in Admin Mode or by creating a custom analysis. The two Threshold Types are outlined below. 

Relative Threshold (Noxturnal Default):

A limb movement (LM) is detected if the amplitude of the leg signal exceeds 5x the background noise present during periods of rest (baseline signal amplitude). However, the LM event is defined as when it exceeds 3x the background noise to the point where it goes below that limit again.  

Absolute Threshold (AASM default): 

This setting works the same way as the relative threshold, except the thresholds are now the exact number of µV that the signal has to rise above the background noise.

• Onset Threshold - 8uV from baseline background - determines when LM onset
• Offset Threshold - return to 2uV from baseline (background) - determines when LM ends 

The final setting to discuss is the maximum amplitude (7). The maximum amplitude is set to 300uV to minimize the impact of signal artifacts on scoring. Users can adjust this setting when signal artifact impacts the automatic scoring of LM events. 

PLMS Settings

PLMS settings allow users to set the criteria for determining which LM events are converted to PLMS events. As per the AASM requirement, a minimum of 4 LM events are required to form a PLMS series. The minimum and maximum interval between LM onsets is set at 5 seconds to a maximum of 90 seconds. 

Combine LM events on separate channels 

The AASM Scoring Manual outlines that LM events on the different leg signals must be separated by more than 5s or scored as one event. This means the onset-to-onset period between the two events should be greater than 5 seconds (as shown in the image below).

In Noxturnal, the automatic analysis does not combine overlapping LM events into one event as outlined by the AASM rules. Noxturnal focuses on PLMS events as the output of the analysis. As a result, the “Combine LM events from all channels into one PLMS scoring” setting applies to PLMS events, not LM events. This results in one event being converted to a PLMS event. However, the other LM event remains in place and is visible to the user. 

The example below highlights how this works in practice. 

• Events 1 and 2 were marked as LM by the analysis. When the PLMS part of the analysis works through the LM events, it determines that 1 and 2 overlap, resulting in only event 1 being converted to a PLMS event. LM event 2 is not converted to a PLMS event and is not reported as a PLMS event but is still visible and reportable as an LM event. 
• Event 3 comes before the LM event on the other leg signal (5). Therefore, event 3 is converted to a PLMS event, and the LM event (5) remains visible on the Right leg signal. 
• The PLM event (4) is added to show the connected series of PLMS events. This is always auto-added to the Left leg signal, irrespective of what signal the PLMS events are scored on. Note that this shows the connected PLMS events and not non-candidate LM events, despite which events are on the Left leg signal. 

In rare situations, this can impact PLMS scoring, as the timing of the LM events can impact the exclusion criteria. In the example below, these LM should be combined and excluded from PLMS consideration, but because the events are not combined, the first event is excluded by the respiratory overlap rule. However, the second LM event is not excluded as it is >0.5s away from the respiratory event.

LM Exclusion Settings for PLMS Conversion 

The Exclusion settings allow users to define which LM events are excluded from conversion to PLMS events. It’s important to understand that these settings do not exclude LM from being scored but prevent LM events from being converted to PLMS. The scored LM events will still be visible in the study. 

It is important to understand that these settings will not exclude LM events from being scored in wake epochs, during movement, or near respiratory events. Instead, these LM events will be prevented from contributing to PLMS events. 

LM events scored in wake epochs

When starting in a wake epoch, Noxturnal ignores LM events within wake epochs for conversion to PLMS events. However, these LM can still be viewed in Noxturnal. However, the AASM rules outline situations where the PLMS series can “jump” over LM in wake epochs if the LM and PLMS scoring criteria are maintained as outlined below: 

1. Events 2 and 3 are within 90 seconds of onset 
2. The two red-circled LMs are excluded because they are in wake epochs 
3. Events 1-4 make the PLMS series and are all within sleep epochs 

In this example below, the auto-analysis correctly “jumps” over the LM in the wake epoch as it continues with the PLMS series with the next event as LM 2 and 3 are within 90 seconds of onset to onset. 

LM events scored during movement events

LM events are excluded as candidates for PLMS events if they start during a movement event. Movement events are automatically scored on the Activity signal. More details on the Activity analysis can be found here. 

For PLMS scoring, LM-induced movement events can prevent LM conversion to PLMS. In this example, the LM events at 1 and 2 are excluded from conversion to PLMS events because they occur during a movement event (see the automatically scored movement event on the Activity signal). If these movement events are manually removed and the Update PLM tool is run, the LM events will be converted to PLMS events, as seen at points 3 and 4. 

Users who notice movement events preventing LMs from converting to PLMS events have two options. They can turn off the movement exclusion in the Update PLM tool (see below). Another workflow option is to delete the movement events impacting the correct PLMS scoring and run the PLM Update Tool to correctly convert LM events to PLMS events. More details on the Update PLM Tool can be found in the Manual Scoring of LM and PLMs in Noxturnal section below. 

LM and Respiratory Events

According to the AASM, LM events should not be scored if they occur during or 0.5s before or after a respiratory event. The respiratory events that preclude LM scoring are: 

• Apnea 
• Hypopnea 
• RERA 

The “starting close to or in an apnea, hypopnea, or RERA event” setting in the PLM Analysis correctly ignores LM events for conversion to PLMS events if the LM event starts during or 0.5s before an apnea or hypopnea. However, if the respiratory occurs after the LM event, the LM will not be excluded from PLMS conversion.

Another consideration is that the respiratory event analysis in Noxturnal does not score RERAs, so LM and PLMS events could be overestimated in the case of patients with frequent upper airway flow limitation. Manually scored RERAs will exclude LM events per the settings outlined using the PLM Update tool. 

Manual Scoring of LM and PLMs in Noxturnal 

Manual scoring of LM and PLMS in Noxturnal can be achieved with and without help from the Update PLM Tool. Users can go through and score all of the LM events and use the PLM Update Tool to convert LM that qualify into PLMS. The Update PLM tool can be found in the Noxturnal menu (shown below). 

The Update PLM tool works off the assumption that all manually scored LM events are correct. The Update PLM tool will review all scored LM, convert those that meet the defined criteria into PLMS events, and add a PLM connecting event between the individual PLMS events of a series. 

Noxturnal does not automatically convert manually scored LM events into PLMS. Users can score PLMS events, which will be used to calculate the PLMS Index. Note that all manually scored events are assumed to be correct by Noxturnal for reporting purposes. If users score PLMS events, the PLM Update Tool will not change any of these events, even if they meet exclusion criteria. 

Scoring Tips

An example PSG scoring workflow by an experienced sleep scientist in the field utilising the auto-analysis in Noxturnal is outlined below: 

• Adjust Analysis Start and End time based on lights on and off time 
• Select Analysis in Period Tab (prevent auto-analysis from scoring in the area outside lights on and off) 
• Run the Sleep Staging and Arousal Analysis (users will have to create a custom analysis separate to the standard PSG Analysis in Noxturnal). 
• User performs a manual review of sleep staging and arousals analysis output 
• Run Respiratory analysis (custom analysis)
• User performs manual review of respiratory analysis output  
• Run PLM analysis (custom analysis) 
• User performs a manual review of the output of the PLM analysis
• If the user makes changes, run the Update PLM tool 
• Run the Arousal Association tool
• Generate report 

LM and PLMS Arousals

PLMS and arousal events are not automatically associated when manually or auto-scored in Noxturnal. Users need to use the Arousal Tool to associate arousal from sleep with other events. Arousals will be categorized into subtypes based on the below priority:

• Respiratory Arousal
• PLM Arousal
• LM Arousal 
• Spontaneous Arousal 

The arousal association tool can be accessed via the Tools menu in Noxturnal. 

The arousal association tool features several settings that the user can adjust. According to the AASM, arousal from sleep should be associated with a PLM event when they coincide, overlap, or <0.5 seconds between the end of one event and the onset of the other event, regardless of which is first. 

As noted in the respiratory events section, the respiratory event analysis in Noxturnal does not score RERAs, so LM and PLM Arousals could be overestimated in the case of patients with frequent upper airway flow limitation. Manually scored RERAs will be prioritized for arousal categorization over PLMS and LM events. 

LM and PLMS Reporting in Noxturnal

Overview:

• Noxturnal 6 and later primarily reports PLMS events and PLM series for limb movements.
• There is no standardized reporting for LM events.
• Users are required to create custom reports for LM events.

Noxturnal default reporting table for PLM / PLMS events:

Limitations Impacting LM Reporting:

• LM events on separate signals are not combined, resulting in multiple events.
• Overlap with converted PLMS events will be counted as LM events.
• LM events in wake epochs are reported. They are excluded from PLMS conversion but remain reportable LM events.
• LM events overlapping with respiratory events will also be reported, as they are only prevented from conversion to PLMS events.

More details on LM and PLMS reporting in Noxturnal can be found here.