Accepted Manuscript Post-Anoxic Reticular Reflex Myoclonus in a Child & Proposed Classification of PostAnoxic Myoclonus Min Tsui Ong, MBChB, MRCPCH, Ptolemaios Georgios Sarrigiannis, MD, FRCP, Peter Stuart Baxter, MA MD FRCPE DCHFRCPCH PII: S0887-8994(16)30883-9 DOI: 10.1016/j.pediatrneurol.2016.12.014 Reference: PNU 9046 To appear in: Pediatric Neurology Received Date: 26 October 2016 Revised Date: 27 December 2016 Accepted Date: 30 December 2016 Please cite this article as: Ong MT, Sarrigiannis PG, Baxter PS, Post-Anoxic Reticular Reflex Myoclonus in a Child & Proposed Classification of Post-Anoxic Myoclonus, Pediatric Neurology (2017), doi: 10.1016/j.pediatrneurol.2016.12.014. 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. Page 1 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Post-Anoxic Reticular Reflex Myoclonus in a Child & Proposed Classification of Post-Anoxic Myoclonus Authors: Min Tsui Ong, MBChB, MRCPCH; Ptolemaios Georgios Sarrigiannis, MD, FRCP; Peter Stuart Baxter, MA MD FRCPE DCH FRCPCH. RI PT Min Tsui Ong, Department of Paediatric Neurology, Sheffield Children's Hospital Ptolemaios Sarrigiannis, Department of Neurophysiology, Royal Hallamshire Hospital Peter Baxter, Department of Paediatric Neurology, Sheffield Children's Hospital Title character count: 91 Supplemental Data: Video 1 TE D Corresponding Author; Min Tsui Ong Paediatric Neurology Sheffield Children's Hospital Phone: +44 114 271 7000 Fax: Not available mintsui.ong@gmail.com M AN U Word count abstract: 155 Word count paper: 1142 SC Number of references: 15 Number of tables: 1 Number of figures: 5 Number of videos: 1 EP Min Tsui Ong (mintsui.ong@gmail.com) Ptolemaios Georgios Sarrigiannis (ptolemaios.sarrigiannis@sth.nhs.uk) Peter Stuart Baxter (peter.baxter@sch.nhs.uk) AC C Statistical Analysis: Not applicable Search Terms: post-anoxic myoclonus; reticular reflex myoclonus; Lance-Adams syndrome; myoclonus Author Contributions: Min Tsui Ong, wrote up findings and manuscript, prepared images and revised manuscript. Ptolemaios Sarrigiannis, performed and analysed neurophysiology tests, prepared videos and images, revised manuscript. Peter Baxter, critical revision of manuscript for intellectual content Acknowledgements: Neurophysiology department, Sheffield Teaching Hospitals Page 2 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT AC C EP TE D M AN U SC RI PT Author Disclosures: Min Tsui Ong -- Reports no disclosures Ptolemaios Sarrigiannis -- Reports no disclosures Peter Baxter -- Reports no disclosures Page 3 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Abstract Objective: and discuss the classification of post-anoxic myoclonus. Results: RI PT To describe a childhood case of post-anoxic myoclonus of the reticular reflex type SC A 9 year old boy with severe hypoxic-ischemic encephalopathy due to submersion developed early epileptic spasms followed by stimulus sensitive multifocal M AN U generalised myoclonus and later dystonia. Video EMG (electromyography) polygraphy performed before treatment showed the discharges associated with the myoclonus lasted less than 50ms. Cortical myoclonus was excluded by jerk-locked averaging (JLA) using arm muscles, which showed no cortical correlates. The TE D recruitment order on EMG polygraphy was consistent with a brainstem generator for the myoclonus, characteristic of reticular reflex myoclonus. Both myoclonus and state. AC C Conclusion: EP dystonia responded to Clonazepam. He has remained in a persistent vegetative Reticular reflex myoclonus can be demonstrated by detailed neurophysiological assessment in children as in adults, and has a similar poor prognosis. Post-anoxic myoclonus can have several differing mechanisms and should not be considered synonymous with Lance-Adams myoclonus. Page 4 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Introduction Survivors of anoxic brain injury can develop myoclonus. The incidence of postanoxic myoclonus (PAM) is unknown but is thought to be rare.1 Lance and Adam first RI PT described four cases in 1963 of intention myoclonus with post-anoxic encephalopathy.2 In 2000 Hallet classified PAM into acute and chronic forms with Lance-Adam’s type as chronic form.1 Cases reported have largely been in the adult SC age range and different myoclonus type can co-exist in a patient at the same time.3 A very rare form of PAM is reticular reflex myoclonus (RRM). Only 4 cases of post- M AN U anoxic RRM has been illustrated in detail before, all adults.3-6 A neonate has also been reported to have RRM secondary to presumed prenatal hypoxic insult7 but the authors did not provide in their work the electrophysiological evidence required to demonstrate the generator and the EMG characteristics (i.e. very short durations) of TE D this extremely rare form of myoclonus. The presentation, response to treatment and EP prognosis differs between types of PAM.8,9 We describe a child with post-anoxic RRM whose myoclonus responded initially with AC C Clonazepam but had an overall poor prognosis. We demonstrate with electrophysiological evidence that our case had RRM. Page 5 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Case Report Clinical details A previously healthy 9 year old boy developed severe hypoxic-ischaemic RI PT encephalopathy and transient multi-organ injury following 10 minutes of cold water immersion and subsequent resuscitation for 66 minutes before return of spontaneous circulation. He then received full intensive care, with therapeutic SC hypothermina for 24 hours. His initial GCS was 3. On day 3 he developed spasms which were treated with Phenytoin and Midazolam infusion. His M AN U electroencephalogram (EEG) showed bursts of spikes/polyspikes and sharp waves interspersed with periods of relative suppression. His magnetic resonance imaging (MRI) showed focal signal change involving corpus striatum bilaterally, hippocampi, and cortex of the parietal and occipital lobes bilaterally (Figure 1). On day 12 when TE D he was transferred to our hospital he showed spontaneous eye opening and variable tone. He had multifocal or generalised myoclonus, present from day 7, involving his limbs, trunk, abdomen, face, neck, mouth, tongue and eyes. This was stimulus EP sensitive to sound but mainly to tendon and muscle stretch (Video 1). Startle reflex was absent on nose tap. Repeat MR brain on day 20 post-anoxia showed volume AC C loss with gliosis in the previously involved areas (Figure 1). Page 6 of 19 Min Tsui Ong M AN U SC RI PT ACCEPTED MANUSCRIPT Figure 1. Axial T2 weighted brain MRI images. Top row. Imaging at presentation showing focal high signal change at presentation involving corpus striatum bilaterally TE D and parafalcine parietal lobes bilaterally. Bottom row. Day 20 imaging showing EP volume loss with gliosis in the affected areas. On day 17 he developed dystonic spasms, which evolved into episodes of status AC C dystonicus treated with Chloral Hydrate, Baclofen, Diazepam, Clonazepam, Clonidine, Botulinium toxin injection, and Midazolam. Myoclonus stopped by day 29, 6 days after the introduction of Clonazepam, which was then weaned on day 35 as he was over-sedated. By day 90 the spasms had resolved and the other medications had been weaned. Stimulus sensitive myoclonus unfortunately recurred at 5 months. Between days 16-23 and days 51-54 he was reported to respond appropriately although not consistently to familiar voices suggesting a minimally conscious state. Page 7 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Subsequent JFK Coma Recovery Scale (Revised) scores were more compatible with AC C EP TE D M AN U SC RI PT a persistent vegetative state.10 Page 8 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Neurophysiological results Recordings were performed with the Natus and the Xltek128 headboxes (Optima RI PT Medical Ltd) sampling at 512Hz. The myoclonus was investigated with EEG and limited EMG leads on day 14 (Figure 2 and Video 1). This showed that tendon or muscle stretch in all limbs, spontaneous movement, sternal tapping, and forehead SC tapping often induced a generalised myoclonic response. The EEG was very abnormal with generalised sharpened theta and delta slow wave activity M AN U interspersed with occasional, mainly frontocentral, sharp waves and spikes and brief periods of relative EEG attenuation but no consistent electro-clinical associations. The possibility of cortical myoclonus was further assessed in Spike2 version 8 software (CED Ltd) with jerk locked averaging (JLA) using arm muscles as trigger TE D points for the myoclonus. This showed no evidence of a cortical generator (Figure 3). Jerk lock averaging is a technique used to assess the possibility of cortically driven myoclonus. Numerous sweeps, commonly more than 100, of time locked EMG and EP EEG recordings of the myoclonic jerks are mathematically averaged into a single epoch. This achieves a very significant increase of the signal to noise ratio on the AC C EEG (i.e. random noise included in the EEG signals cancels out on the averaged data). In this way, EEG spikes preceding the onset of the myoclonic jerks, otherwise buried in noise and hence, undetectable on the raw EEG recordings can be revealed. Page 9 of 19 Min Tsui Ong M AN U SC RI PT ACCEPTED MANUSCRIPT Figure 2. Still shot of video 1 (in supplemental section). Video 1 shows day 14 video EEG with limited EMG leads demonstrating muscle stretch in all limbs, sternal TE D tapping, and forehead tapping often induced a generalised myoclonic response. Day 22 video EEG and EMG polygraphy with subsequent change in sweep speed (250ms/div). The EMG polygraphy shows a recruitment order that reveals a EP consistent lower brainstem lead (sternocleidomastoid and trapezius muscles are recruited first) of the myoclonus with subsequent fast rostral and caudal spread. The AC C EMG discharges are of very short duration, <50ms. The constellation of the electrophysiological findings is consistent with RRM. X1-X2 refers to single channel electrocardiogram. Page 10 of 19 Min Tsui Ong M AN U SC RI PT ACCEPTED MANUSCRIPT Figure 3. Jerk-locked averaged EMG/EEG polygraphy of 275 sweeps showing no cortical correlates with myoclonus. C4Cz and CzC3: central scalp EEG derivations. ECG: single channel electrocardiogram, R TR: right triceps EMG, R BB: right biceps TE D brachii EMG, ROOr: right orbicularis oris EMG, ROOc: right orbicularis oculi EMG. Please note that the activity appearing on the central EEG derivations has the same morphology on the averaged data on the single channel ECG and similar on the EP averaged and rectified activity from the orbicularis oris (OOr), suggesting that this is EMG and movement related activity from the myoclonus. The averaged EMG data AC C show co-activation of agonist/antagonist in the right upper arm (biceps and triceps brachii) preceded by activation of facial nerve muscles (orbicularis oculi and oris). The R BB myoclonic EMG discharges were used to trigger events for the technique of averaging to be applied. EMG channels are rectified (i.e. add the EMG signal below the baseline, of positive polarity in this instance, to the signal above the baseline) and high pass filtered at 2.7Hz while EEG channels are high pass filtered at 0.8Hz (sampling rate 512Hz) to produce a clearer onset of the EMG discharges. The vertical cursor marks the onset of the averaged and rectified data from the OOr. Page 11 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT RI PT . EMG polygraphy performed on day 22 (Video 1) to identify the generator of the SC myoclonus showed that the EMG discharges associated with the myoclonus were very short in duration, typically below 50ms (Figure 4), much faster than the activity M AN U commonly seen in adult exaggerated startle responses.11 The recruitment order on EMG polygraphy and the short duration of the EMG discharges was consistent with a brainstem generator for the myoclonus (rostral and caudal spread from a lower brain stem lead) which is characteristic of RRM.1 Our patient's myoclonus came in clusters. It frequently involved only the lower brainstem area and only the more AC C EP TE D prominent jerks spread in more cranial and caudal areas. Page 12 of 19 Min Tsui Ong SC RI PT ACCEPTED MANUSCRIPT M AN U Figure 4. EMG polygraphy during spontaneous myoclonic jerks. Rectified EMG in spike 2 software shows a lead of the trapezius in the recruitment order followed by caudal and rostral spreading (A). The EEG and EMG polygraphy show frequent clusters of spontaneous jerks, some spreading into higher brainstem and spinal TE D regions, like the first shown in the diagram, while others remain confined in the lower AC C EP brainstem region and the upper arm (B). Page 13 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Discussion Confusion arises in the classification of PAM because some authorities label all forms, and others all late onset forms, as 'Lance-Adams'.1,12 However as our patient differing prognostic implications and treatment.3 RI PT shows and as reported in adults more than one type of myoclonus can occur, with SC In general there are several ways to classify myoclonus: one method is to use the site of origin guided by the electrophysiological findings into cortical, cortical- M AN U subcortical, subcortical non-segmental, segmental and peripheral.8,12 Examples of subcortical non-segmental myoclonus are myoclonus-dystonia syndrome, opsoclonus-myoclonus syndrome, exaggerated startle, RRM, and propriospinal AC C EP TE D myoclonus (Figure 5). Page 14 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Cortical-subcortical Exaggerated startle reflex Subcortical, non-segmental Myoclonus RI PT Cortical SC including brainstem Reticular reflex M AN U Segmental TE D Peripheral AC C EP Figure 5. Classification of myoclonus based on the site of origin. Predominant type in PAM coloured red. Page 15 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT The predominant types in the post-anoxic patient are cortical, exaggerated startle and RRM. The difference between the three lies mainly in neurophysiological parameters (Table 1) and it can be difficult to differentiate them clinically. In RRM there are no EEG consistent time locked correlates, EMG bursts are of very short RI PT duration and the recruitment order is in a rostral and caudal direction from a brain stem generator. Recommended treatment also varies between the myoclonus types AC C EP TE D M AN U SC thus the importance of distinguishing them (Table 1). Page 16 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT Table 1. Clinical and neurophysiological features and treatment of types of PAM. Exaggerated Startle Cortical Focal, multifocal, bilateral Generalized Generalized or generalized RI PT Clinical features Reticular reflex myoclonus Response to stimuli Action & sensory stimulation Sensory, auditory & visual stimuli EEG correlates Yes No EMG duration Bursts typically <75 ms Bursts 50-400 ms shortening with habituation EMG recruitment order Rostral to caudal direction from the cortex Rostral and caudal direction from a brain stem generator Treatment Valproate SC M AN U Clonazepam Action, auditory & sensory stimuli No Bursts 10-50 ms Clonazepam Diazepam Levetiracetam 5-Hydroxytryptophan Piracetam Deep-brain stimulation AC C EP TE D Clonazepam The myoclonus originally described by Lance and Adams is a cortical myoclonus. While the myoclonus does not usually improve, the cognitive outcome is reasonable.2 There have also been reports of an acute or early form of PAM, described as a form of myoclonic epilepsy associated with a burst suppression EEG, and 100% mortality.13,15 In contrast to the former two types of myoclonus, RRM in Page 17 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT PAM may improve spontaneously, as in our patient, but is associated with a poor overall outcome.3,7 RI PT In conclusion PAM can have several differing mechanisms. Multiple types may occur in one patient. We recommend classifying myoclonus based on the site of origin, guided by the electrophysiological findings to aid management and SC provide prognosis. Our case illustrates this very rare form of PAM, the reticular AC C EP TE D M AN U reflex myoclonus, and the value of detailed neurophysiological assessment. Page 18 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT References 1. Hallett M. Physiology of human posthypoxic myoclonus. Mov Disord 2000;15 2. RI PT Suppl 1:8-13. Lance JW, Adams RD. The syndrome of intention or action myoclonus as a sequel to hypoxic encephalopathy. Brain 1963;86(1):111-136. 3. Werhahn KJ, Brown P, Thompson PD, Marsden CD. The clinical features and Hallett M, Chadwick D, Adam J, Marsden CD. Reticular reflex myoclonus: a M AN U 4. SC prognosis of chronic posthypoxic myoclonus. Mov Disord 1997;12(2):216-220. physiological type of human post-hypoxic myoclonus. J Neurol Neurosurg Psychiatry 1977;40(3):253-264. 5. Brown P, Thompson PD, Rothwell JC, Day BL, Marsden CD. A case of postanoxic encephalopathy with cortical action and brainstem reticular reflex 6. TE D myoclonus. Mov Disord 1991;6(2):139-144. Inoue M, Kojima Y, Kinboshi M, Kanda M, Shibasaki H. A case of post-anoxic 7. EP reticular reflex myoclonus. Rinsho Shinkeigaku 2012;52(8):557-560. Kakisaka Y, Haginoya K, Togashi N, Kitamiura T, Uematsu, M, Hino-Fukuyo N, AC C Kure S, Saito J, Kitaoka S, Watanabe S, Yoshikawa H, Nara T, Suzuki Y, Tsuchiya S. Neonatal-onset brainstem reticular reflex myoclonus following a prenatal brain insult: Generalized myoclonic jerk and a brainstem lesion. Tohoku J Exp Med 2007;211(3):303-308. 8. Caviness JN. Treatment of myoclonus. Neurotherapeutics 2014;11(1):188-200. 9. Kojovic M, Cordivari C, Bhatia K. Review: Myoclonic disorders: A practical approach for diagnosis and treatment. Ther Adv Neurol Disord 2011;4(1):47-62. Page 19 of 19 Min Tsui Ong ACCEPTED MANUSCRIPT 10. Giacino JT, Kalmar K, Whyte J. The JFK Coma Recovery Scale-Revised: measurement characteristics and diagnostic utility. Arch Phys Med Rehabil 2004;85(12):2020-2029. 11. Brown P, Rothwell JC, Thompson PD, Britton TC, Day BL, Marsden CD. New (Pt4):1891-1902. RI PT observations on the normal auditory startle reflex in man. Brain 1991;114 Lancet Neurology 2004;3(10):598-607. SC 12. Caviness JN, Brown P, Myoclonus: Current concepts and recent advances. 13. Van Cott AC, Blatt I, Brenner RP. Stimulus-sensitive seizures in postanoxic M AN U coma. Epilepsia 1996;37(9):868-874. 14. Niedermeyer E, Bauer G, Burnite R, Debby R. Selective stimulus-sensitive in acute cerebral anoxia myoclonus report. Arch Neurol 1997;34:365-368. 15. Young GB, Gilbert JJ, Zochodne DW. The significance of myoclonic status AC C EP TE D epilepticus in postanoxic coma. Neurology 1990;40(12):1843-1848.