ADVERTISEMENT

Issue Navigator

Volume 09 No. 01
Earn CME
Accepted Papers
Classifieds







Scientific Investigations

Manual Quantitative Assessment of Amplitude and Sleep Stage Distribution of Excessive Fragmentary Myoclonus

http://dx.doi.org/10.5664/jcsm.2336

Romy Hoque, M.D.; David E. McCarty, M.D.; Andrew L. Chesson, M.D., F.A.A.S.M.
Department of Neurology, Division of Sleep Medicine, Louisiana State University School of Medicine, Shreveport, LA

ABSTRACT

Introduction:

Excessive fragmentary myoclonus (EFM) consists of brief, asynchronous, twitch-like movements appearing asymmetrically in sleep. The new AASM Manual for the Scoring of Sleep and Associated Events identifies some EFM scoring criteria but does not provide amplitude criteria for scoring EFM. Older observational series have used 50 μVs. We report data from various amplitude criteria using blinded comparisons.

Methods:

EFMs were analyzed on the polysomnograms of 8 patients (7 men and 1 woman, mean age 57 years, range: 47-79) using a standardized protocol for sensitivity, tonus threshold, impedance, amplitude measurements, and sleep stage. The first 20 minutes each of wake, Stage 1-2, SWS, and REM were analyzed. EFMs ≥ 25, ≥ 40, and ≥ 50 microvolts (μVs) in negative deflection above the baseline were counted in tibialis anterior muscle electromyography (EMG) channels bilaterally.

Results:

The mean EFM Index per minute for wake, regardless of impedance, was: 7.19 ± 5.90 for ≥ 25 μV amplitude; 2.43 ± 2.02 for ≥ 40 μVs; and 2.08 ± 2.23 for ≥ 50 μVs. For sleep stages, the EFM index by stage and amplitude criteria used for measurements were: Stage 1-2: 7.38 ± 5.79 for ≥ 25 μVs; 3.13 ± 3.33 for ≥ 40 μVs; and 2.36 ± 2.66 for ≥ 50 μVs; SWS: 10.05 ± 8.04 for ≥ 25 μVs; 2.71 ± 3.13 for ≥ 40 μVs; and 1.38 ± 1.92 for ≥ 50 μVs; Total REM: 15.96 ± 11.32 for ≥ 25 μVs; 6.32 ± 4.25 for ≥ 40 μVs; and 3.94 ± 3.73 for ≥ 50 μVs; Phasic REM: 19.69 ± 15.45 for ≥ 25 μVs; 8.63 ± 7.06 for ≥ 40 μVs; and 5.52 ± 6.44 for ≥ 50 μVs; Non-phasic REM: 13.93 ± 11.31 for ≥ 25 μVs; 5.16 ± 3.57 for ≥ 40 μVs; and 3.20 ± 2.92 for ≥ 50 μVs.

Conclusion:

EFM rates increase with SWS and total REM with the highest EFM rates occurring during phasic REM. EFM rates were increased across all sleep stages when impedance was > 30 KΩ.

Citation:

Hoque R; McCarty DE; Chesson Jr AL. Manual quantitative assessment of amplitude and sleep stage distribution of excessive fragmentary Myoclonus. J Clin Sleep Med 2013;9(1):39-45.


Excessive fragmentary myoclonus (EFM) was first described by Dagnino et al. in an observational study of 18 patients.1 In the 2007 American Academy of Sleep Medicine (AASM) Scoring Manual for Sleep and Associated Events, EFM is described as a benign movement phenomenon.2 The characteristic electromyogram (EMG) pattern for recommended scoring is burst duration of 150 msec, presence in at least 20 minutes of NREM sleep, and a rate of 5 EMG EFM potentials per minute. The AASM scoring manual does not address sleep stage distribution of EFM or define amplitude criteria for scoring. Older observational series used 50 microvolts (μV).3,4

This observational study was to assess functionality of use and identify potential enhancements to the AASM scoring manual EFM criteria that may aid sleep medicine clinicians, sleep medicine technologists, and clinical electrophysiologic research. Previous studies of EFMs have not used scoring manual criteria. In this study using the AASM scoring guidelines, we manually analyzed the polysomnograms (PSGs) of 8 consecutive patients with obvious EFM to assess EFM amplitude and sleep stage distribution. We also analyzed the effect of electrode impedance on EFM frequency.

BRIEF SUMMARY

Current Knowledge/Study Rationale: This observational study was to assess functionality of use and identify potential enhancements to the AASM scoring manual EFM criteria. We manually analyzed the PSGs of 8 consecutive patients with obvious EFM to assess: EFM amplitude; EFM sleep stage distribution; and impedance effect on EFM frequency.

Study Impact: Our proposed additions to the standardized AASM definition of EFM may allow for more extensive research into the physiologic and clinical associations of this phenomenon in the future.

METHODS

PSGs were acquired using the Alice Sleepware version 2.5.11 platform (Respironics, Murrysville, PA) according to the following the AASM scoring manual guidelines: left and right frontal, central and occipital electroencephalogram (EEG) leads referenced to the opposite ear; bilateral oculogram, submental EMG; bilateral tibialis anterior muscle electromyography (EMG); and cardiorespiratory recordings consisting of nasal pressure monitoring, nasal-oral thermistors, abdominal, and chest effort; pulse oximetry from the digit; and electrocardiogram. In accordance with AASM scoring manual technical specifications, surface EMG electrodes were placed 2-3 cm apart longitudinally and symmetrically on the middle of the tibialis anterior muscles bilaterally. Both legs were monitored on separate channels.

This was an observational study of 8 consecutive patients collected over a 6-month period meeting scoring manual criteria for EFM on PSG. Scoring manual criteria for EFM requires ≥ 20 min of NREM sleep with recorded EFMs at a rate of ≥ 5 EMG EFM potentials per minute. In the scoring manual criteria for periodic limb movements in sleep (PLMS), a stable resting EMG in relaxed tibialis anterior muscle is defined as a deflection above baseline < 5 μV. There are no such explicitly stated criteria for EFM in the scoring manual. The inclusion criterion for our study was that obvious EFMs were noted throughout the course of the night and the EMG rest amplitude at the start of the recording was < 2 μV. In all 8 recordings evaluated, the end-of-night calibrations showed EMG rest amplitudes < 2 μV. Leg movements clearly associated with sleep disordered breathing were excluded.

The scoring manual definition for PLMS defines the end of a LM event as a period lasting 0.5 seconds during which the EMG does not exceed 2 μV. There are no such explicitly stated criteria for EFM in the scoring manual. Minimal duration between individual EFM potential is also not defined in the AASM scoring manual. Scorers used a subjective determination of visual distinctiveness in determining individual EFM potentials.

Reflecting requirement of ≥ 20 min of NREM sleep with EFM in the scoring manual definition of EFM, the first scored 20 min each of wake, stage 1-2, slow wave sleep (SWS), and REM sleep were analyzed for each patient. EMG potentials of 150 msec were included in the manual analysis. The study designer and primary scorer (RH) is board certified in sleep medicine and was trained using the AASM scoring manual criteria by one of the manual's authors (AC). A similarly trained board certified secondary scorer's (DM) repeat scoring of three patient PSGs were assessed to determine reproducibility of the scoring method in polysomnograms with all 4 stages of sleep (N1, N2, SWS, and REM). The κ statistic calculations were used to assess statistically significant differences between the 2 scorers. For the κ statistic, the p-value calculated was for testing the null hypothesis of no agreement; a p-value < 0.05 rejects the null hypothesis and indicates significant agreement between the 2 scorers.

The number of EMG potentials with amplitude above the baseline ≥ 25 μV, 40 μV, and 50 μV were counted for the first 20 min each of wake, NREM stage 1-2, SWS, and REM. Superimposed reference lines placed at 25 μV, 40 μV, and 50 μV were used to make these manual measurements (Figure 1). REM epochs were further analyzed by breaking down each epoch into phasic and non-phasic REM segments. Extraocular motion (EOM) was noted in the oculogram channels. The phasic-REM period was defined as starting with the first EOM in the series and ended 5 sec after the last EOM. REM that did not meet phasic criteria was designated as non-phasic REM.

Overlaid reference lines placed at 25 μV and 50 μV were used to make manual measurements of excessive fragmentary myoclonus potential amplitudes

Similar reference lines were used at 40 μV.

jcsm.9.1.39a.jpg

jcsm.9.1.39a.jpg
Figure 1

Overlaid reference lines placed at 25 μV and 50 μV were used to make manual measurements of excessive fragmentary myoclonus potential amplitudesSimilar reference lines were used at 40 μV.

(more ...)

Impedance was also assessed. The software platform used had an impedance cutoff for the leg EMG channels of 30 KΩ. Each epoch was labeled as either having leg channel impedance > 30 KΩ or < 30 KΩ. Epoch impedance was considered > 30 KΩ if ≥ 15 sec of the epoch had an impedance value ≥ 30 KΩ.

EFM rates were calculated using the following equation:

jcsm.9.1.39a-e1.jpg

EFM rates were calculated for wake, NREM stage 1-2, SWS, total REM, phasic REM, and non-phasic REM.

RESULTS

Demographics

Records of 8 consecutive patients with EFMs were collected over a 6-month period, reviewed to confirm the diagnosis of EFM by the scoring manual criteria, and evaluated (Table 1). Mean age was 57 years (range: 47–79); 7 patients were men. Wake and stage 1-2 were evaluated for 8/8 patients. SWS and/or REM were present in 7 of the 8 patients. Three recordings were successful positive airway pressure titration studies. Two recordings showed mild obstructive sleep apnea (OSA) with total sleep time (TST) respiratory disturbances indices (RDI) of 6.5 and 9, respectively. One recording showed moderate OSA with a TST RDI of 19.

Demographics, comorbidity, medication, and polysomnography data

jcsm.9.1.39.t01.jpg

table icon
Table 1

Demographics, comorbidity, medication, and polysomnography data

(more ...)

Bilateral EFMs Regardless of Impedance

EFM rate for potentials ≥ 25 μV increased from wake into stage 1-2, SWS, and REM (Table 2). EFM rates progressively decreased in all stages with the utilization of higher amplitude criteria. Total REM, phasic REM, and non-phasic REM all had higher rates across all amplitude criteria. Phasic REM had a higher EFM rate across all amplitude criteria than either non-phasic REM or total REM.

Excessive fragmentary myoclonus (EFM) rate data

jcsm.9.1.39.t02.jpg

table icon
Table 2

Excessive fragmentary myoclonus (EFM) rate data

(more ...)

Interestingly, with the 40 μV and the 50 μV amplitude criteria, EFMS occurred less often in SWS than wake, stages 1-2, and total REM.

Impedance Effects

Regardless of amplitude criteria, the mean EFM frequencies in wake and stage 1-2 and were higher in PSG epochs with impedance > 30 KΩ versus PSG epochs with impedance < 30 KΩ. In REM, the mean EFM frequencies were similar for amplitude criteria of 25 μV and 40 μV. EFM rates were higher in wake and stages 1-2 utilizing all amplitude criteria when impedance was > 30 KΩ versus epochs with impedance < 30 KΩ.

Reproducibility of Scoring Method

The rescoring of the EFMs of 3 patients by 2 independent scorers (RH, DM) showed that though this method is not perfectly reproducible, the results were very similar from scorer to scorer (Table 3). There are only 3 cases for which there was no agreement between the 2 readers (p-value > 0.05 for the κ statistic): readings utilizing the 25 μV amplitude criterion in Patient 2 during SWS, readings utilizing the 25 μV in Patient 8 during REM; and readings utilizing the 50 μV amplitude criteria in Patient 8 during wake. For these 3 cases, the 95% confidence interval for the κ statistic included zero, the hypothesized value.

Independent amplitude analysis of EFMs in three patients with N1, N2, REM, and SWS

jcsm.9.1.39.t3a.jpg

table icon
Table 3A

Independent amplitude analysis of EFMs in three patients with N1, N2, REM, and SWS

(more ...)

Cohen κ statistic analysis of independent scoring of EFMs

jcsm.9.1.39.t3b.jpg

table icon
Table 3B

Cohen κ statistic analysis of independent scoring of EFMs

(more ...)

DISCUSSION

EFM rates increased with deeper stages of sleep (i.e., SWS and REM; Figure 2). EFM frequency was highest in phasic REM. This finding is consistent with increased muscle activity during phasic REM. It is not clear why EFMs were more prominent in SWS and total REM than stages 1-2 and wake.

Excessive fragmentary myoclonus (EFM) rates from this study compared to the literature

Regardless of amplitude criteria employed (≥ 25 μV, ≥ 40 μV, or ≥ 50 μV) our study showed greater EFM rates in wakefulness and across all sleep stages. Hoque et al (2012) combined Stages 1 and 2, with the data displayed as Stage 1. Hoque et al. (2012) and Mizuma et al. (1997) combined stages 3 and 4 into SWS, with the data displayed as SWS. REM, rapid eye movement sleep; SWS, slow wave sleep.

jcsm.9.1.39b.jpg

jcsm.9.1.39b.jpg
Figure 2

Excessive fragmentary myoclonus (EFM) rates from this study compared to the literatureRegardless of amplitude criteria employed (≥ 25 μV, ≥ 40 μV, or ≥ 50 μV) our study showed greater EFM rates in wakefulness and across all sleep stages. Hoque et al (2012) combined Stages 1 and 2, with the...

(more ...)

Prior studies of EFM used varied terminology and varied equations to define EFM rates (Table 4). Historical terminology for EFM-like events include such terms as excessive fragmentary hypnic myoclonus, fragmentary myoclonus, hypnic myoclonus, physiologic hypnic myoclonus, and sleep related fragmentary pathological multifocal myoclonus. The AASM scoring manual has provided a useful definition for EFM, but does not discuss amplitude or sleep stage distribution due to lack of prior validated data. In comparison to prior studies of EFM, our small study shows higher EFM rates across all sleep stages regardless of amplitude criteria employed.

Literature on excessive fragmentary myoclonus (EFM)

jcsm.9.1.39.t04.jpg

table icon
Table 4

Literature on excessive fragmentary myoclonus (EFM)

(more ...)

A limitation of our impedance analysis was that none of the 8 patients had EFMs in SWS with PSG impedances greater than 30 KΩ. A second limitation was that only one patient had EFMs in REM with PSG impedances greater than 30 KΩ.

Many PSG software platforms recommend an impedance cutoff of 30 KΩ. Utilizing impedance criteria of ≤ 30 KΩ resulted in smaller EFM rates across all sleep stages regardless of amplitude criteria. This rate reduction implies a possible artifactual origin to some measure of the EFMs noted. Unfortunately it is not possible to determine what percentage of EFMs is purely artifactual in origin vis-á-vis the percentage of EFMs that represent a true electrophysiologic manifestation of the tibialis anterior muscles recorded bilaterally. Despite this limitation, the use of clearly defined impedance standards in documentation of EFM potentials may: allow for electrophysiologic evaluation less clouded by artifact; aid sleep laboratory technicians intervention during the recording when the values are problematic; and help scorers in either clinical or research settings to standardize data gathering.

Neither of the impedance categorical values we used (i.e., <30 KΩ or > 30 KΩ) are gold standard values from the literature. They are common settings for commercially available equipment guidelines. However, these same leads are used in PSG for monitoring periodic limb movements of sleep (PLMS), and by scoring manual criteria should have impedances less than 10 KΩ. We recommend for consistency and dual use of the leads that 10 KΩ be adopted to facilitate consistency and future research.

Our proposed amendments to the AASM scoring manual EFM criteria definition are put forth in Table 5. A priori benefits of using lower amplitude criteria (i.e., ≥ 25 μV) compared to the one employed by other authors previously (i.e., ≥ 50 μV) are the allowance for a more sensitive assessment of EFM across all sleep stages and a more robust correlation of EFMs with other polysomnographic data such as sleep disordered breathing indices. While studying the significance of EFMs in the future, inclusion of all likely EFM events seems reasonable. Correlation of individual EFM potentials with cortical arousals may still prove to be difficult given the short duration of EFM potentials and their high frequency when present. Data regarding correlation of EFMs with cortical arousals were not collected in this study.

Hoque et al.'s excessive fragmentary myoclonus (EFM) proposed criteria to standardize EFM evaluation

jcsm.9.1.39.t05.jpg

table icon
Table 5

Hoque et al.'s excessive fragmentary myoclonus (EFM) proposed criteria to standardize EFM evaluation

(more ...)

The use of the 50 μV criterion in the past may have been an attempt to adjust for impedance related skin artifact. However, with newer software and better skin preparation techniques this may be no longer necessary. Also a manual assessment of EFM, as performed in this study, will no longer be necessary, as the process can be automated with more sophisticated software. Since the maximum allowable impedance limits for the EMG electrodes during a PSG are not defined by the AASM scoring manual, we recommend explicitly defined the maximum EMG electrode impedance cut off at 10 KΩ to fit with the AASM scorning manual requirements for periodic limb movements of sleep monitoring. Our proposed additions to the standardized AASM definition of EFM may allow for more extensive research into the physiologic and clinical associations of this phenomenon in the future.

DISCLOSURE STATEMENT

This was not an industry sponsored study. The authors have indicated no financial conflicts of interest.

REFERENCES

1 

Dagnino N, Loeb C, Massazza G, Sacco G, authors. Hypnic physiological myoclonias in man: an EEG-EMG study in normals and neurological patients. Eur Neurol. 1969;2:47–58. [PubMed]

2 

Iber C, Ancoli-Israel S, Chesson A, authors; and Quan SF for the American Academy of Sleep Medicine.. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. 2007. 1st ed. Westchester, Illinois: American Academy of Sleep Medicine;

3 

Broughton R, Tolentino MA, authors. Fragmentary pathological myoclonus in NREM sleep. Electroencephalogr Clin Neurophysiol. 1984;57:303–9. [PubMed]

4 

Lins O, Castonguay M, Dunham W, Nevsimalova S, Broughton R, authors. Excessive fragmentary myoclonus: time of night and sleep stage distributions. Can J Neurol Sci. 1993;20:142–6. [PubMed]

5 

Broughton R, Tolentino MA, Krelina M, authors. Excessive fragmentary myoclonus in NREM sleep: a report of 38 cases. Electroencephalogr Clin Neurophysiol . 1985;61:123–33. [PubMed]

6 

Montagna P, Liguori R, Zucconi M, et al., authors. Physiological hypnic myoclonus. Electroencephalogr Clin Neurophysiol. 1988;70:172–6. [PubMed]

7 

Mizuma H, Sakamoto T, authors. Excessive twitch movements in rapid eye movement sleep with daytime sleepiness. Psychiatry Clin Neurosci. 1997;51:393–6. [PubMed]

8 

Vetrugno R, Plazzi G, Provini F, Liguori R, Lugaresi E, Montagna P, authors. Excessive fragmentary hypnic myoclonus: clinical and neurophysiological findings. Sleep Med. 2002;3:73–6. [PubMed]

9 

Pincherle A, Mantoani L, Villani F, Confalonieri P, Erbetta A, authors. Excessive fragmentary hypnic myoclonus in a patient affected by a mitochondrial encephalomyopathy. Sleep Med. 2006;7:663[PubMed]