The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform discharges during long-term video-EEG monitoring

The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform discharges during long-term video-EEG monitoring

Accepted Manuscript Title: The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform discharges during long-term video-EEG monit...

513KB Sizes 0 Downloads 54 Views

Accepted Manuscript Title: The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform discharges during long-term video-EEG monitoring Authors: Xi Liu, Naoum P. Issa, Sandra Rose, Shasha Wu, Taixin Sun, Leo V. Towle, Peter C. Warnke, James X. Tao PII: DOI: Reference:

S1059-1311(18)30506-5 https://doi.org/10.1016/j.seizure.2018.10.015 YSEIZ 3314

To appear in:

Seizure

Received date: Revised date: Accepted date:

11-8-2018 20-10-2018 25-10-2018

Please cite this article as: Xi L, Issa NP, Rose S, Shasha W, Sun T, Towle LV, Warnke PC, Tao JX, The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform discharges during long-term video-EEG monitoring, Seizure: European Journal of Epilepsy (2018), https://doi.org/10.1016/j.seizure.2018.10.015 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The first-hour-of-the-day sleep EEG reliably identifies interictal epileptiform

SC RI PT

discharges during long-term video-EEG monitoring

Xi Liu, MD, PhD1, Naoum P. Issa, MD, PhD2, Sandra Rose, MD2, Shasha Wu, MD, PhD2, Taixin Sun, MD3, Leo V. Towle, PhD2, Peter C. Warnke, MD, FRCS4, and James X. Tao, MD, PhD2

of Neurology, Zhongnan hospital , Wuhan University, Wuhan, Hubei province, P. R. China; 2Department of Neurology, University of Chicago, Chicago, IL 60637; 3Department of Neurology, Beijing Electric Power Hospital, Beijing, P. R. China; 4Department of Neurosurgery, University of Chicago, Chicago, IL. USA

U

1Department

*Correspondent

A

N

Author: Xi Liu, MD. PhD, Department of Neurology, Zhongnan hospital , Wuhan University, Wuhan, Hubei province, 169 Donghu Road, Wuhan, Hubei, 430071 email: [email protected]. Phone: +86(027)67813410

M

Running title: The first-hour sleep EEG recording

TE

D

Highlights

Sleep EEG is an effective activation procedure for the identification of IEDs.



The first hour sleep EEG reliably identifies IEDs during long-term video-EEG study.



A three-day video-EEG recording is a reasonable duration for IED identification.

CC

EP



A

ABSTRACT

Purpose: To determine the usefulness of the first-hour sleep EEG recording in identifying interictal epileptiform discharges (IEDs) during long-term video-EEG monitoring. Method: We retrospectively reviewed 255 consecutive patients who underwent continuous long-term video-EEG monitoring in the adult epilepsy monitoring unit (EMU) at the University of Chicago. The complete video-EEG recording was reviewed, and the occurrence of IEDs was determined for each 1

patient. We compared the occurrence of IEDs observed during the first-hour sleep EEG recordings with the occurrence of IEDs observed during the complete video-EEG recordings. Results: Overall, IEDs were observed in 134 (53%) of 255 patients during the full long-term video-EEG recording with a mean duration of 4 days. IEDs were identified in the first-hour sleep EEG in 125 (49%) of 225 patients. Comparing to reviewing full records, the first hour sleep EEG identified IEDs in 125

SC RI PT

(93%) of 134 patients. Of the IED subtypes, the first-hour sleep EEG identified 92 (94%) of 98 patients with temporal lobe IEDs, 11 (92%) of 12 patients with frontal lobe IEDs, 3 (100%) of 3 patients with parietal lobe IEDs, 1(50%) of the 2 patients with occipital lobe IEDs, 16 (94%) of 17 patients with generalized IEDs, and 2 (100%) 2 patients with multi-focal IEDs.

Conclusions: The first-hour sleep EEG reliably predicts the occurrence of IEDs during the long-term

video-EEG recording, and therefore can be a time-efficient tool for identifying patients with IEDs during

N

U

long-term video-EEG recording in the adult epilepsy monitoring unit.

M

A

Keywords: Video-EEG monitoring; Interictal epileptiform discharges; Sleep; Seizures; epilepsy monitoring unit.

TE

D

INTRODUCTION

Long-term video-EEG monitoring in the epilepsy monitoring unit (EMU) is the gold standard for the diagnosis, classification and localization of epileptic seizures.1 Specific EEG epileptiform discharges

EP

allow better informed decisions regarding medical and surgical treatments and more accurate prediction of seizure control and ultimate remission. An epileptiform discharge is predictive of seizure recurrence,

CC

particularly in patients with idiopathic epilepsy.2 EMU study has resulted in establishing a definitive diagnosis in 76% to 88% of patients and changing in diagnosis or treatment in up to 79% of patients.3, 4 Identification of interictal and ictal epileptiform discharges is essential for achieving these clinical

A

objectives. While the identification of ictal EEG is relatively straightforward, reviewing daily EEG recordings for the identification of IEDs by visual analysis are both labor intense and time consuming. Automatic EEG spike detection software is helpful, but has a low specificity for detecting IEDs and has been used largely as a complementary tool to visual analysis.5 As such, new methods to improve the efficiency of spike detection during the long-term video-EEG monitoring are desirable, particularly in patients with infrequent IEDs. 2

Sleep modulates IEDs in patients with focal and primary generalized epilepsies.6, 7 Synchronous EEG oscillations including tonic background slow waves and phasic sleep EEG transients (i.e. sleep spindles, k-complexes, ripples) activate IEDs during drowsiness and NREM sleep,8-11 whereas asynchronous neuronal oscillations during wakefulness and REM sleep suppress IEDs.11, 12Sleep is the most effective activation procedure as compared to other activation procedures such as photic stimulation and

SC RI PT

hyperventilation and shows the highest yield of IEDs.13 An EEG after sleep deprivation improves the detection of IEDs in 13% to 35% of patients whose standard EEG findings were normal.2 Sleep

recordings have been used as an activation procedure to increase the yield of spike detection during

routine EEG studies. However, the usefulness of sleep EEG in the detection of IEDs during long-term

video-EEG monitoring has not been assessed. The aim of this study is to determine whether the first-hour sleep EEG can reliably predict the occurrence of IEDs, and therefore could be used as a time-efficient tool for the detection of IEDs during long-term video-EEG monitoring.

U

METHODS

N

Patient population

A

We retrospectively included consecutive patients who underwent long-term video-EEG monitoring at

M

the University of Chicago adult epilepsy center between April, 2016 and October, 2017. Patients were referred for long-term video-EEG monitoring by physicians in our epilepsy center and local community hospitals. Indications for video-EEG monitoring included: 1) differential diagnosis of epileptic vs. non-

D

epileptic spells; 2) pre-surgical evaluation for epilepsy surgery; 3) classification of seizures; 4) change

TE

and toxicity of anti-epileptic drugs (AEDs). The enrolled patients had a minimum of 24 hours of videoEEG recording. Exclusion criteria included: 1) age <18 years; 2) video-EEG recording <24 hours; 3) intracranial EEG monitoring. The study was approved by the institutional review board (IRB) and

EP

individual patient’s written consent was not required by the IRB.

CC

Extraction of clinical data

Clinical data were collected from review of medical records including clinical notes, EMU

A

reports and hospital discharge summaries. Seizure type (focal vs. generalized) was determined based on clinical history, ictal semiology, interictal and ictal EEG discharges. Focal seizures were further categorized according to seizure onset location (temporal, frontal, parietal or occipital). The occurrence of IEDs during the complete course of video-EEG recording was determined by reviewing the EMU reports, which was then compared to the occurrence of IEDs during the first-hour of sleep for each patient. Two investigators were blinded to whether IEDs were documented in the EMU reports prior to reviewing the archived first-hour sleep EEG recording. 3

Long-term video-EEG monitoring Long-term video-EEG was recorded using the Xltek EEG system (Natus Medical Incorporated, Pleasanton, CA, USA) according to the international 10-20 system with 6 additional sub-temporal electrodes (F9, T9, M1, F10, T10, and M2). EEGs were sampled at 512 Hz, filtered at 1-70 Hz and reviewed usually at 10µv/mm and 30mm/s. At the discretion of attending physicians, partial or complete

SC RI PT

withdrawal of anti-epileptic drugs (AEDs) was performed as early as the day of admission to increase the likelihood of interictal and ictal recordings. Activation procedures including hyperventilation, photic stimulation, and sleep deprivation were performed as clinically warranted. AEDs were commonly restarted one day before hospital discharge. Identification of IEDs

Video-EEG recordings were reviewed during routine clinical care by experienced EEG technologists,

U

clinical neurophysiology fellows, and a board-certified attending physician. Interictal and ictal EEG

N

abnormalities were routinely segmented and archived on a computer server on a daily basis. EMU reports were generated during the course of hospital admission, describing the presence and localization of

A

interictal, ictal EEG abnormalities and non-epileptic events on a daily basis. The first-hour sleep EEG

M

recording was also routinely segmented and archived on a daily basis. The occurrence of IEDs during the first-hour sleep EEG recording of the enrolled patients was retrospectively and independently reviewed

D

by two investigators (XL and JXT). The interrater reliability between the two reviewers was excellent with kappa at 0.91 for the identification of IEDs. Differences between the two reviewers were resolved by

TE

consensus. IEDs were defined as spike, sharp wave, polyspike, or spike/wave discharges. They are typically <200 ms in duration and distinguished by their morphology and/or amplitude from normal EEG

EP

background. Normal EEG variants such as small sharp spikes and wicket spikes were not considered IEDs in this study.

CC

RESULTS

A total of 255 patients were included in the study. Mean age was 39 ± 14; range 18-76. Ninety seven

A

patients were male and 158 were female. Mean duration of long-term video-EEG recordings was 4±1 days; range 1-7. The indications for long term video-EEG monitoring include differential diagnosis of spells, classification of seizures, presurgical evaluation and adjustment of anti-epileptic medications (Table 1). In 14 patients the first-hour sleep EEG recording was not available for review on one of the admission days. Interictal epileptiform discharges were identified in 134 (53%) of 255 patients during the complete long-term video-EEG recordings as documented in EMU reports. IEDs were identified in the first-hour 4

sleep EEG in 125 (49%) of 225 patients. Comparing to reviewing full records, the first hour sleep EEG identified IEDs in 125 (93%) of 134 patients. The frequency of IED detection was well correlated over the course of video-EEG monitoring between the complete EEG recording and the first- hour sleep EEG recording (Figure 1). Of the IED subtypes, the first-hour sleep EEG identified 92 (94%) of 98 patients with temporal lobe IEDs, 11 (92%) of 12 patients with frontal lobe IEDs, 3 (100%) of 3 patients with

SC RI PT

parietal lobe IEDs, 1 (50%) of the 2 patients with occipital lobe IEDs, 16 (94%) of 17 patients with generalized IEDs, and 2 (100%) 2 patients with multi-focal IEDs. IEDs were not identified in the firsthour sleep recording in 9 (7%) of 134 patients. Of the 255 patients, 99 (39%) had definitive (epileptic

seizures recorded) epilepsy throughout the continuous video-EEG recordings. IEDs were not identified in 13(13%) of the 99 patients after 72 h and in 11(11%) throughout their stay in EMU.

The cumulative frequency of IED detection over the course of video-EEG monitoring was

assessed. With complete video-EEG recording, 80% of patients with IEDs were identified after 1 day,

U

92% after 2 days, 97% after 3 days, and 100% after 4 days. With the first-hour sleep EEG recording, 75%

N

of patients with IEDs were identified after 1 day, 84% after 2 days, 90% after 3 days, and 93% after 4 days. The frequency of IEDs in different IED subtypes was also well correlated between reviewing the

A

complete study and the first-hour EEG (Table 2). The frequency of IED detection appeared to plateau

M

after 3 days of video-EEG recording. DISCUSSION

D

Interictal epileptiform discharges reflect the hyper-excitability and functional impairment of the brain and are EEG markers of epilepsy.14 However, routine EEG studies have a low sensitivity for

TE

identifying IEDs. IEDs are detected only in 13-50% of the first routine EEG studies in patients with new onset seizures.15, 16 The sensitivity of IED detection can be improved 84% after third EEG and up to 92%

EP

after the fourth routine EEG studies in patients with epilepsy. 15Long term video-EEG monitoring has been an efficient method for the detection of IEDs in patients with epilepsy. With the withdrawal of AEDs, IEDs can be identified in the 75% of epilepsy patients within 24 hours of EEG recording, and 96%

CC

within 72 hours,17 although IEDs were not identified in 8-12% of definitive epilepsy patients after 72 hours of continuous EEG recordings.18-20 Our study is consistent with these findings. With the review of

A

the complete video-EEG recording, IEDs were identified in 80% of patients in the first 24 hours and in 97% within 72 hours. IEDs were not identified in 13 (13%) of 99 patients with definitive epilepsy after 72 hours and 11 (11%) throughout their stay in EMU. Therefore, a 3-day EMU admission might be a reasonable duration for the identification of IEDs during long-term video-EEG monitoring. Interictal epileptiform discharges are strongly activated during NREM sleep, 7, 8, 21 although it has remained controversial whether IEDs are activated more significantly during NREM stage 1 and 2, or 5

during slow-wave sleep.22 IEDs are more effectively enhanced during sleep in patients with focal epilepsy than generalized epilepsy as compared with wakefulness.23, 24 Using intracranial EEG recording, IEDs were increased in all brain regions during NREM sleep compared with wakefulness. NREM sleep strongly activated interictal spikes in the seizure onset zone and promotes their propagation to other cortical areas.8 Larger scalp voltage field potentials of IEDs were also detected in patients with temporal

SC RI PT

lobe epilepsy during sleep than wakefulness.10 Additionally, generalized spike and wave discharges are activated by drowsiness and NREM sleep, and they may be entirely absent during wakefulness in some patients.

The first hour sleep EEG recording in this study was composed of NREM stage 1 and 2 sleep as well as slow wave sleep. Our study demonstrated that the first-hour sleep EEG recording was nearly as

informative as the complete video-EEG recording. IEDs were identified in 75% of the patients in the first night’s sleep, 90% after 3 days, and plateaued at 93% after 4 days. The first-hour sleep EEG identified

U

93% of IEDs detected by reviewing complete video-EEG recording. Nevertheless, our study is limited by

N

its retrospective nature, affecting the type and diversity of the epilepsy population included. Interrater reliability could not be assessed among the EEG interpreters who reviewed complete video-EEG

M

interpretation of IEDs cannot be ruled out.

A

recording during the routine clinical care, so the possibility of false negative or false positive

CONCLUSION

D

In summary, the first-hour sleep EEG reliably predicts the occurrence of IEDs during long-term video-EEG recording as compared to the full review of EEG recording. It can be used as a time-efficient

TE

tool for identifying patients with IEDs and mitigating the burden of IED identification in the adult epilepsy monitoring unit during long-term video-EEG recording, similar to the function of spike detection

EP

computer software. The first-hour sleep EEG recording for the identification of IEDs in patients with epilepsy during long-term video-EEG monitoring can be useful for the diagnosis and classification of

CC

epilepsy, although it is less useful for seizure localization.

A

Conflict of interest statement We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our manuscript entitled “The first-hour sleep EEG reliably identifies interictal epileptiform discharges during video-EEG monitoring”. None of the authors had any conflict of interest to report.

6

Disclosure Authors have no funding or conflict of interest to disclose

REFERENCES

A

CC

EP

TE

D

M

A

N

U

SC RI PT

1. Shih JJ, Fountain NB, Herman ST, et al. Indications and methodology for videoelectroencephalographic studies in the epilepsy monitoring unit. Epilepsia 2018;59:27-36. 2. Wirrell EC. Prognostic significance of interictal epileptiform discharges in newly diagnosed seizure disorders. J Clin Neurophysiol 2010;27:239-248. 3. Lobello K, Morgenlander JC, Radtke RA, Bushnell CD. Video/EEG monitoring in the evaluation of paroxysmal behavioral events: duration, effectiveness, and limitations. Epilepsy & behavior : E&B 2006;8:261-266. 4. Smolowitz JL, Hopkins SC, Perrine T, Eck KE, Hirsch LJ, O'Neil Mundinger M. Diagnostic utility of an epilepsy monitoring unit. Am J Med Qual 2007;22:117-122. 5. Halford JJ, Shiau DS, Kern RT, Stroman CA, Kelly KM, Sackellares JC. Seizure Detection Software Used to Complement the Visual Screening Process for Long-Term EEG Monitoring. Am J Electroneurodiagnostic Technol 2010;50:133-147. 6. Ujma PP, Halasz P, Kelemen A, Fabo D, Eross L. Epileptic interictal discharges are more frequent during NREM slow wave downstates. Neurosci Lett 2017;658:37-42. 7. Singh S, Shukla G, Goyal V, et al. Impact of sleep on the localizing value of video EEG in patients with refractory focal seizures - a prospective video-EEG with EOG and submental EMG study. Clin Neurophysiol 2014;125:2337-2343. 8. Lambert I, Roehri N, Giusiano B, et al. Brain regions and epileptogenicity influence epileptic interictal spike production and propagation during NREM sleep in comparison with wakefulness. Epilepsia 2018;59:235-243. 9. Nobili L, Baglietto MG, Beelke M, et al. Spindles-inducing mechanism modulates sleep activation of interictal epileptiform discharges in the Landau-Kleffner syndrome. Epilepsia 2000;41:201-206. 10. Del Felice A, Storti SF, Manganotti P. Sleep affects cortical source modularity in temporal lobe epilepsy: A high-density EEG study. Clin Neurophysiol 2015;126:1677-1683. 11. Shouse MN, Farber PR, Staba RJ. Physiological basis: how NREM sleep components can promote and REM sleep components can suppress seizure discharge propagation. Clin Neurophysiol 2000;111 Suppl 2:S9-S18. 12. Campana C, Zubler F, Gibbs S, et al. Suppression of interictal spikes during phasic rapid eye movement sleep: a quantitative stereo-electroencephalography study. J Sleep Res 2017;26:606-613. 13. Elisa Baldin WAH, Jeffery R. Buchhalter, Dale C. Hesodorffer and Ruth Ottman. Utility of EEG activation procedures in epilepsy: a population- based study. J Clin Neurophysiol 2017;34:512-519. 14. Janszky J HM, Clemens Z, et al. Spike frequency is dependent on epilepsy duration and seizure frequency in temporal lobe epilepsy. Epileptic Disord 2005;7:355-359. 15. Salinsky M, Kanter R, Dasheiff RM. Effectiveness of multiple EEGs in supporting the diagnosis of epilepsy: an operational curve. Epilepsia 1987;28:331-334. 16. Goodin DS, Aminoff MJ. Does the interictal EEG have a role in the diagnosis of epilepsy? Lancet 1984;1:837-839. 17. Werhahn KJ, Hartl E, Hamann K, Breimhorst M, Noachtar S. Latency of interictal epileptiform discharges in long-term EEG recordings in epilepsy patients. Seizure 2015;29:20-25. 7

A

CC

EP

TE

D

M

A

N

U

SC RI PT

18. Faulkner HJ AH, Mohamed A. Latency to first interictal epileptiform discharge in epilepsy with outpatient ambulatory EEG. Clinical Neurophysiology 2012;123:1732-1735. 19. Narayanan JT, Labar DR, Schaul N. Latency to first spike in the EEG of epilepsy patients. Seizure 2008;17:34-41. 20. Friedman DE, Hirsch LJ. How long does it take to make an accurate diagnosis in an epilepsy monitoring unit? J Clin Neurophysiol 2009;26:213-217. 21. Klein KM, Knake S, Hamer HM, Ziegler A, Oertel WH, Rosenow F. Sleep but not hyperventilation increases the sensitivity of the EEG in patients with temporal lobe epilepsy. Epilepsy Res 2003;56:43-49. 22. Frauscher B, von Ellenrieder N, Ferrari-Marinho T, Avoli M, Dubeau F, Gotman J. Facilitation of epileptic activity during sleep is mediated by high amplitude slow waves. Brain 2015;138:1629-1641. 23. Delil S, Senel GB, Demiray DY, Yeni N. The role of sleep electroencephalography in patients with new onset epilepsy. Seizure 2015;31:80-83. 24. Malow BA, Kushwaha R, Lin X, Morton KJ, Aldrich MS. Relationship of interictal epileptiform discharges to sleep depth in partial epilepsy. Electroencephalogr Clin Neurophysiol 1997;102:20-26.

8

Figure legend.

A

CC

EP

TE

D

M

A

N

U

SC RI PT

Figure1. Cumulative frequency of IED detected by reviewing the first-hour sleep (blue) and complete EEG recording (red) during the course of video-EEG recording. The frequency of IED detection was well correlated between the first- hour sleep EEG recording and the complete EEG recording.

9

D

TE

EP

CC

A

SC RI PT

U

N

A

M

Figr-1

10

Table 1. Criteria for EMU admission N (%)

Differential diagnosis of spells

156 (61.2)

Presurgical evaluation

54 (21.2)

Classification of seizures

30 (11.8)

SC RI PT

Indication for admission

Medication change/toxicity

15 (5.9) 255

A

CC

EP

TE

D

M

A

N

U

Total

11

Table 2. Cumulative frequency of IED detection in different IED types EEG type

1 Day

2 Days

3 Days

4 Days

5 Days

6 Days

7Days

Temporal

slEEG

75%

85%

90%

94%

94%

94%

94%

cEEG

81%

93%

98%

100%

100%

100%

100%

slEEG

75%

92%

92%

92%

92%

92%

92%

cEEG

75%

100%

100%

100%

100%

100%

100%

slEEG

33%

33%

67%

100%

100%

100%

100%

cEEG

33%

67%

100%

100%

100%

100%

100%

slEEG

50%

50%

50%

50%

50%

50%

50%

cEEG

50%

100%

100%

100%

100%

100%

100%

slEEG

50%

100%

100%

100%

100%

100%

100%

cEEG

50%

100%

100%

100%

100%

100%

100%

slEEG

77%

88%

88%

94%

94%

94%

94%

cEEG

88%

93%

100%

100%

100%

100%

100%

slEEG

75%

84%

90%

93%

93%

93%

93%

cEEG

80%

97%

100%

100%

100%

100%

Generalized

92%

D

Combined

U

Multifocal

N

Occipital

A

Parietal

M

Frontal

SC RI PT

IED type

A

CC

EP

TE

slEEG: the first-hour sleep EEG recording; cEEG: complete EEG recording;

12