cortical reflex myoclonus in rett syndrome

Cortical Reflex Myoclonus in Rett Syndrome Renzo Guerrini, MD,* Paolo Bonanni, MD,* Lucio Parmeggiani, MD,* Margherita Santucci, MD,t

Antonia Parmeggiani, MD,t and Ferdinand0 Sartucci, MD$

Rett syndrome (RS) is one of the most frequent causes of mental retardation in females. As there are no known biochemical, genetic, or morphological markers, diagnosis is based on clinical phenotype including severe dementia, autism, truncal ataxidapraxia, loss of purposeful hand movements, breathing abnormalities, stereotypies, seizures, and extrapyramidal signs. Myoclonus, although reported in some series, has never been characterized. We studied 10 RS patients, age 3 to 20 years, and observed myoclonus in 9. Severity of myoclonus did not correlate with that of the other symptoms or with age. Multifocal, arrhythmic, and asynchronous jerks mainly involved distal limbs. Electromyographic bursts lasted 48 f 12 msec. Bunst-locked electroencephalographic averaging generated a contralateral centroparietal premyoclonus transient preceding the burst by 34 & 7.2 msec. Motor evoked potentials showed normal latencies, indicating integrity of the corticospinal pathway. Somatosensory evoked potentials were enlarged. The C-reflex was hyperexcitable and markedly prolonged (62 f 4.3 msec), mainly due to increase in cortical relay time (28.4 & 4.5 msec). We conclude that RS patients show a distinctive pattern of cortical reflex myoclonus with prolonged intracortical delay of the long-loop reflex.

Guerrini R, Bonanni P, Parmeggiani L, Santucci M, Parmeggiani A, Sartucci F. Cortical reflex myoclonus in Rett syndrome. Ann Neurol 1998;43:472-479

Rett syndrome (RS) is a neurological disorder affecting females with clinical and neuropathological findings indicating early developmental arrest. 1-5 An X-linked dominant inheritance, lethal in males, has been hypothesized, although an X-autosomal mechanism and non-Mendelian inheritance are also possible.‘ With a prevalence of 1 per 10,000 to 15,000 girls, RS could be the second most frequent specific cause of mental retardation in females, next to Down ~yndrome.~ Di- agnosis of RS is based only on the clinical phen~ type ,~ because, to date, there are no known biochemical, genetic, or morphological markers.

The clinical presentation and course of classic RS show a characteristic pattern indicating extensive and progressive involvement of the nervous system, with definable staging.’,’ After normal development to the age of 7 to 18 months, stagnation occurs, followed by a “destructive” stage, leading to severe dementia within 1.5 years from onset. This, in turn, is followed by a stable course for several decades. Deterioration of higher brain functions is accompanied by autism, truncal ataxia/apraxia, loss of purposeful hand movements, hyperventilation, stereotypic behavior, seizures, and ac- quired microcephaly. Extrapyramidal signs are present in most patients; although a hyperkinetic movement disorder is observed in younger girls, older patients tend to be bradykineti~.’,’~

From the *Institute of Child Neurology and Psychiatry, University of Pisa, and Institute for Medical Research Stella Maris Foundation, and $Department of Neurosciences, Institute of Neurology, Univer- sity of Pisa, Pisa; and $Department of Child Neurology and Psy- chiatry, Institute of Neurology, University of Bologna, Bologna, Italy.

Myoclonus has been estimated to occur in about 50% of patients older than 4 years.” However, in spite of its high frequency, neurophysiological analysis of myoclonus in RS has so far never been addressed.

We used repeated long-term video-electroencephalographic (video-EEG) and simultaneous electromyographic (EMG) recordings to study 10 girls with stage 2 to 4 RS, detecting myoclonus in 9 of them. Jerk- locked back-averaging of EEG activity, somatosensory evoked potentials (SEPs), and long-loop evoked responses (C-reflex) demonstrated a cortical reflex origin of myoclonus. Although myoclonus in itself did not represent a dominant clinical feature in this patient population, it appeared relatively early and showed pe- culiar characteristics that could be useful to provide support for diagnosis.

Patients and Methods Patients Ten girls with classic RS, referred consecutively to two cen- ters (ICNP-Pisa and DCNP-Bologna), were included in the study. Criteria used for clinical diagnosis and staging were those proposed by the Rett Syndrome Diagnostic Criteria Work Group.’ Age at the time of study ranged between 3 and 20 years (mean, 9 years; median, 8 years). Progression of the disease had been followed clinically and with serial EEGs for 2 to 17 years (mean, 7 years). To ensure comparability of

Received Aug 12, 1997, and in revised form Nov 3. Accepted for publication Nov 7, 1997.

Address correspondence to Dr Guerrini, Institute of Child Neurol- ogy and Psychiatry, University of Pisa, Via dei Giacinti 2, 56018 Calambrone, Pisa, Italy.

472 Copyright 0 1998 by the American Neurological Association

neurophysiological parameters between patients and age- matched controls, we calculated mean values by dividing both patients and controls into two subgroups, ie, age 9 years and younger and age 10 years and older.

Clinical and Video-Electrophysio Logical Studies All patients were evaluated using long-term video-EEG and simultaneous EMG monitoring with bipolar and referential montages, using silver/silver chloride surface cup electrodes. All neurophysiological recordings were performed in the same center (ICNP-Pisa). Scalp electrode placement was performed according to the International 10-20 system. EMG activity was recorded by using pairs of electrodes applied 3 cm apart over the masseter, orbicularis oris, deltoid, biceps, finger flexors and extensors, abductor pollicis brevis (APB), and quadriceps femori muscles. EEG/EMG activity was recorded by computer for later analysis. Back-averaging of EEG activity related to the EMG bursts was performed in the 9 patients with clinically significant myoclonus (Patients 1-8 and 10). The EEG signal was filtered, using a bandpass of 1 to 100 Hz, and digitized at the sampling rate of 1,024 Hz. The average of 100 to 120 consecutive 500-msec artifact-free EEG epochs centered at the onset of the EMG burst (burst-locked EEG averages) was computed. At least two averages were generated for each patient to ensure reli- ability between trials. Averages of 100 to 120 consecutive 500-msec artifact-free EEG epochs related to EMG silent periods (SPs) (silence-locked EEG averages) were generated from the same EEG data. Consistency of the responses was verified by using the coefficient of variability (CV), ‘ in terms of the ratio between standard deviation of each sample of peak latency measurements and relative mean [CV = (SD/ mean) X 1001. A threshold of 5% was set to define the measurements as statistically reproducible. Back-averaged EEG was displayed as topographic voltage maps to identify electrical field distribution via a computerized EEG system.

To define myoclonus severity, a score of 0 to 4 was assigned. The score was determined by the number of spontaneous jerks per minute occurring in the EMG of wrist flexors and extensors bilaterally at rest (quantification of jerks was unreliable during movement). The jerks were quantified during 15 minutes of awake and drowsy recordings. The patient showing the highest number of jerks per minute was assigned a score of 4 (loo%), and the other patients were subsequently assigned scores of 3 (<75%), 2 (<50%), 1 (<25%), and 0 (<5%).

Somatosensory Evoked Potentials and C-ReJex SEPs were recorded in all patients from centroparietal (C3’ and C4’ = 2 cm behind the International 10-20 system C3 and C4) regions, at the level of the seventh cervical vertebra (C7), and from Erb’s point (EP), using several referential montages. Scalp electrodes were referenced to the homolo- gous contralateral area and contralateral mastoid, C7 to the anterior neck (thyroid cartilage), and EP to the contralateral equivalent location. The C-reflex at rest was sought simultaneously by recording EMG activity from the following muscles: masseter, orbicularis oris, finger extensors and flexors, APB, and quadriceps femori. The median nerve was stimulated electrically at the wrist, using an intensity strong

enough to produce a visible twitch of the thenar muscle, with a frequency of 1 Hz and a duration of 0.2 msec. The EEG signal was filtered with a bandpass of 1 to 2,000 Hz. Blocks of 50 to 250 consecutive artifact-free responses were averaged. Trials were replicated to ensure reproducibility of the responses. Peak latency and amplitude were measured for each recognizable component at the C3’ and C4’ electrodes. The amplitude was measured from the preceding peak of the opposite polarity. The nomenclature of each wave was established based on the work of Ikeda and colleagues.” SEPs were considered to be giant when the amplitude was greater than 2.5 SD above the mean normal established in 10 normal subjects, of whom 5 were age 3 to 9 years (6.4 5 1.5 years; mean 2 SD) and 5 were age 10 to 20 years (16 t 2.3 years). C-reflex with muscular facilitation was sought simultaneously in the same control population. Statistical comparisons were performed by using Student’s t test for unpaired data. In selected patients, multiple-channel SEPs were collected by using the average reference and displayed as topographic voltage maps. Taps to fingers were also used, to produce myoclonic jerks.

Transcranial Magnetic Stimulation A Novametrix Magstim 200 magnetic device was used to elicit motor evoked potentials (MEPs) in 6 patients (Table), using a flat, single round coil (inner diameter, 9.5 cm) placed on the scalp. The EMG surface-recording electrodes were placed in the APB. The effects of peripheral nerve, cervical, and transcranial magnetic stimulation (TMS) on voluntary motor contraction were also analyzed. Single TMSs were delivered to the contralateral side of the scalp. Stimulus intensity was 100% of the stimulator’s output (1.5 T). Cervical stimulation was delivered with the coil centered over the C7. Trials were replicated at least four times to ensure reproducibility of responses. Control data for TMS, including total motor conduction time (TMCT), peripheral direct and indirect (F wave) motor conduction times, direct and indirect (F wave) central motor conduction times (CMCTs), amplitude of responses, and post-MEP SP were obtained from 10 healthy subjects, age 4 to 10 years (7 -+ 2 years), and from 20 healthy subjects, age 15 to 30 years (25.6 ? 2.3 years). Statistical comparisons were performed by using Student’s t test for unpaired data.

Neuroimaging Eight patients underwent brain magnetic resonance imaging, using 0.5- or 1.5-T instrumentation. The remaining 2 patients underwent brain computed tomographic scans.

Efects o f Piracetam on Myoclonus All patients were treated with various antiepileptic drugs because of epileptic seizures or severe epileptiform EEG abnormalities. To assess rhe effects of piracetam on myoclonus, 400 mg/kg of the drug was administered intravenously over 30 minutes in 3 patients (Patients 4, 5, and 10). Intravenous testing was chosen in order to assess drug efficacy acutely, thereby allowing an immediate decision on the suitability of long-term oral treatment without an additional video-EEG recording. The number of myoclonic jerks recorded synchro- nously in the EMG of wrist extensors and flexors from both

Guerrini et al: Cortical Reflex Myoclonus in Rett Syndrome 473

Table. Neurophysiological Data on 10 Patients with Rett Syndrome

SEPs (measured at C4' and C3') Back-Averaged EEG

Disease N20- P30- Stage1 Latency Positivity Burst N20 P30 N35 P30 N35 Myoclonus (Wrist Flex) Duration Duration Latency Latency Latency Interval Interval Severity (msec) CV (msec) (msec) (msec) (msec) (msec) (msec) (msec)

Patient no./age (yr) 1 I3 216 317 417 518 618 Mean (3-9 yr) SD 7110 8/10 91 1 bb 10120 Mean (210 yr) SD

Control subject age Mean (3-9 yr) SD Mean (2 10 yr) SD

2-31 1 311 31 1 314 313 312

31 1 312

3-410 3-41 1

32 40 39 22 35 34 33.7

6.5 39 23.3 NE 42.1 34.8 10.1

NE NE NE NE

3.1 2.6 5 4.7 0.4 4.6 3.4 1.8 1.7 2.7 NE 0.8 1.7 1

NE NE NE NE

65 125 79 88

133 154 107.3 35 91 84

NE 93 89.3 4.7

NE NE NE NE

44.9 38.4 44.4 46 58 67 49.8 10.6 50 52 NE 58 53.3 4.2

NE NE NE NE

18.6 28 45 9.4 17 20.2 29 42 8.8 13 14.4 26 NE 11.6 NE 16.2 29.8 37.6 13.6 7.8 15.8 31 52 15.2 21 17 29 46.6 12 17.6 17 28.8' 44.6' 11.8' 15.3' 2.1 1.7 5.4 2.4 5.1

21.8 31 52 9.2 21 15.8 30.6 37.8 14.8 7.2 22.1 31 38 8.9 7 21.8 32 43 10.2 11 20.4 31.2' 44.3d 11.4d 13.1 3.1 0.6 7.2 3 7.1

16.6 20.3 26.2 3.8 5.8 0.7 1.7 4.3 1.4 3.8

18.8 23.3 28 4.8 4.3 0.9 2.6 5 2.5 2.6

"C-reflex latency - (N20 latency + TMCT). bThis patient was excluded from means calculation because of lack of myoclonus. 'p < 0.005; d0.005 < p < 0.05 (t test).

EEG = electroencephalogram; SEPs = somatosensory evoked potentials; MEPs = motor evoked potentials; CV = coefficient of variability; TMCT = total motor conduction time; CMCT = central motor conduction time; SP = silent period; Flex = flexors; APB = abductor pollicis brevis; ND = not done; NE = not elicitable; NC = not calculated.

sides after the end of the infusion was then calculated from the first 15 minutes of wakeful rest within 1 hour from drug administration. Resdts were then compared with the number of jerks occurring in similar circumstances and over iden- tical time samples in the EMG preceding the infusion. Sta- tistical comparisons were performed by using Student's t test for unpaired data. In addition, the effect of piracetam administration on the motor pattern was monitored clinically, comparing videotape recordings made during an hour of ob- servation before the infusion and during the 2 subsequent hours. Further similar assessment of myoclonus was performed 3 months later, after the 3 patients had been taking piracetam orally (300 mgfkglday).

Results The Table summarizes the stage of the disease, the severity of myoclonus, and the neurophysiological findings in the 10 patients.

Clinical and Video-Electrophysiologicaogical Analysis of Myoclonm Nine patients exhibited multifocal myoclonus, varying greatly in severity. By assigning a score of 100% to the patient with the highest number of jerks per minute at rest (Patient 4) , we found that the only patient with no

clinically observable myoclonus (Patient 9) had a score of 3%.

Myoclonus occurred at rest in isolated and arrhythmic multifocal jerks and was enhanced by voluntary movement. At rest, jerks were predominately distal but also involved one entire limb or facial muscles. Syn- chronous EMG bursting of agonist and antagonist muscles had a mean duration of 48 % 12 msec. Bilat- eral or generalized myoclonus was never recorded. Myo- clonus at rest was exacerbated during drowsiness and attenuated, to the point of disappearance, during deep sleep. In all patients, EEG showed slow background activity (3-6 Hz) with the subcontinuous 4- to 6-Hz rhythmic sinusoidal pattern typical of RS.'* Superim- posed multifocal epileptiform abnormalities were present, clearly predominating over the frontocentral areas.

O n visual inspection of the polygraphic traces, myoclonus was only occasionally time-locked with spike EEG activity. However, in all 9 patients with myoclonus, jerk-locked back-averaging of EEG activity generated a reproducible premyoclonic negative-positive- negative or positive-negative spike in the centroparietal region of the hemisphere contralateral to the jerking

474 Annals of Neurology Vol 43 No 4 April 1998

Myoclonic Latency SEP- (C-Reflex) Myoclonus MEPs

SEPs (measured at C4' and C3') Median Taps Cortical P30-

N20-N35 N20-P30 P30-N35 (Wrist (Wrist Relay Myoclonus TMCT Interval Amplitude Amplitude Flex) Flex) Time" (Wrist (APB) ( m s 4 (FV (PW (msec) (msec) (msec) Flex) (msec) (msec)

-

26.4 10 21.8 12

NE 20.5 21.4 30 36.2 12 29.6 14.4 27.1' 16.5'

6.1 7.6 30.2 25.4 22 7.3 15.9 3.2 21.2 12.6 24.5" 15.1d

5 9.3

9.6 3.9 4.3 1.9 8.2 3.2 4.8 1.1

6 59 7 66.5 4 63.6

20 55 6 63 6.8 60 8.3d 61.2' 5.8 4

26 67.2 5 NE 3 NE 4 NE

11.7 67.2 12.4 NC

2.4 42.1 1.4 2.1 1.4 45.9 0.9 1.2

68 ND 75 ND 70 ND 71

3.6 70 58.8 NE 85.4 71.4 13.3

ND ND ND ND

26.8 ND 35.2 23.6 30.4 25.8 28.4' 4.5

ND ND ND ND NC NC

10.3 1.2 8.5 1.3

31 37.5 37.6 25.2 32 31 32.4 4.7

36.2 NE NE NE 36.2 NC

NE NE NE NE

13.6 ND 14 15.2 16.8 17.2 15.4 1.6

ND ND ND 17.4 17.4 NC

16.2 1

18.6 1.6

MEP CMCT Duration SP (APB) (APB) (APB) (msec) (msec) (msec)

5.2 ND 6 5.2 4.8 6

0.5 ND ND ND 7 7 NC

6 0.72 7.5 1.5

5.4d

muscle (Fig 1). Premyoclonic positivity had a mean duration of 101 ? 29 msec; peak latency was 34 ? 7.2 msec when EMG bursts were recorded from the belly of the wrist extensors. The Table shows the level of reproducibility of latency values for each patient.

Fig 1. Patient 8. Back-averaged electroencephalographic activity (n = 100; average refeYence; rectified electromyogram) in relation to spontaneous jerks involving lefi wrist extensor (L WE) and flexor (L WF) muscles. A positive-negative potential, recognimble over the P4 (maximal amplitude) and C4 electrodes, precedes the jerk by 23 mec.

F4 6

p3 a LWE

LWF A- ~

30 ms

13 57.7 ND ND 10 57 11.6 60 16.4 108 11.2 65 12.4 69.5d 2.5 21.7

ND ND ND ND ND ND 29.4 60 29.4 60 NC NC

22.9 181 8.9 76

27.8 135 7.5 37

EEG mapping of premyoclonic spikes showed maximal positive distribution over the P3 or P 4 electrode and a concomitant maximal negative distribution over frontal leads (Fig 2A and B). Silence-locked EEG averages did not generate a potential related to the EMG silence in any of the 9 patients.

Age at onset of myoclonus was between 2 and 7 years (mean, 5 years; median, 5 years 6 months) but could have been earlier because mild myoclonus may have been unrecognized. There was no correlation of severity of myoclonus with that of the other symptoms or with age.

Somatosensoty Evoked Potentials and C-Rejex Responses recorded at EP (N9), C7 (N13), and the N9-Nl3 interpeak time were within normal values in all patients. Central conduction time (CCT = latency difference between the cervical N13 and the scalp N20) was significantly increased (>3 SD) in 7 patients (Patients 1, 2, 4, 6, 7, 9, and 10) and borderline in the remaining 3 (Patients 3, 5, and 8). The N20-P30- N35 complex appeared broad in morphology (Fig 3; see Fig 2C). In particular, in some patients, the N35 wave was ill defined in the parietal area, being better represented in more anterior leads. N20-P30 and N20-N35 intervals were significantly delayed in all patients (+6.7 SD and +3.5 SD, respectively). Mean baseline peak amplitude of the N20 component was in the normal range. In the 9 patients with myocionus


Fig 2. Patient 4. (A) Back-averaged electroencephalographic (EEG) activity (n = 100; average reference; rectijed electromyogram) in relation to spontaneous myoclonic jerks of the lefi wrist extensor (L WE) muscle. A negative-positive-negative complex with higher amplitude over the P4 electrode precedes the jerk by 22 msec (latency from the positive peak to the jerk onset). Only activity from the P4 lead is shown, to show molphological similarities of the spontaneous premyoclonic complex and the somatosensory evoked potential (SEP) shown in C. (B) Topographic map analysis of the initial positive wave of the premyoclonus complex shown in A (at arrow). (C) Cortical SEP to electrical stimulation o f the lefi median nerve at the wrist; mastoid reference, average of 50 shocks. An enlarged N20-P30-N35 complex is followed, 28 msec after the positive P30 peak, by a C-reflex in the left abductor pollicis brevis (LAPB). The shock-Greflex latency is 58 msec. (0) Topographic map analysis of the initial positive peak o f SEP as shown in C (at arrow). The electvicaljeld distribution of the premyoclonic complex and o f the SEP are overlapping.

(Patients 1-8 and lo), a significant increase was observed in interpeak amplitudes of the N20-P30 and P30-N35 complexes (+7 SD and + 3 SD, respectively). The C-reflex was hyperexcitable at rest in all of them (see Fig ZC). The response was recorded in the ipsilateral muscles after electrical stimulation in 7 patients (Patients 1-7) and after mechanical stimulation in 6 (Patients 1, 3, 5, 7, 8, and 10). In no patient was a bilateral response observed. The latency of onset of reflex myoclonus after electrical stimulation was 62 5 4.3 msec in wrist extensors and flexors and 65 2 5 msec in the APB. The reflex response to mechanical stimulation had a latency of 71.2 ? 8.7 msec in wrist

extensors and flexors. The SEP (P30 wave)-myoclonus latency and that of the premyoclonus spike were similar (about 32.9 msec). Mapping of the multichannel SEP (P30 wave) showed the same field distribution as that of the premyoclonic spike (see Fig 2C and D).

Transcranial Magnetic Stimulation MEPs showed TMCT within normal limits in all patients. CMCTs, either direct or indirect, were reduced significantly compared with controls, especially in younger patients. In addition, all patients showed a shortening of mean MEP duration after scalp stimula-


P3

0 1

I I stim.

10 ms

Fig 3 . Patient 4. Cortical somatosensory evoked potential to electrical stimulation of the right median nerve at the wrist; average reference, average of 500 shocks. Note the relatively small size of the P25 wave (C3) and the enlarged P30 (P3), N30 (F3), and N35 (C3) waves.

tion. The post-MEP SP was significantly shortened compared with controls (see Table).

Neuroimaging Neuroimaging was considered normal in 8 patients. Cavum vergae was observed in 1 patient and a small arachnoid cyst in another.

Treatment o f Myoclonus with Piracetam The 3 patients treated with intravenous piracetam during video-EEG-polygraphic monitoring showed a significant reduction in myoclonus ( p < O.Ol), with the drug being well tolerated. Clinical and neurophysiological tests performed 3 months after the beginning of oral treatment confirmed persistence of the antimyoclonic effect. However, reduction of myoclonus did not translate into substantial clinical benefit, as patients’ motor pattern remained purposeless and ataxic, although less jerky. The drug was therefore withdrawn.

Discussion Patients with RS have an abnormal, jerky, purposeless, and dystonic motor pattern and are often engaged compulsively in performance of ~tereotypies.~ How- ever, the patients’ severe mental impairment and lack of cooperation preclude unequivocal recognition of the individual components of this abnormal movement. Few clinical studies have reported myoclonus in RS, giving highly variable estimates of its f r e q ~ e n c y . ~ ’ ~ ’ ~ ~ By using long-term video-EEG and polygraphic monitoring, we recorded periods of quiet behavior during which multifocal myoclonus was captured in most patients, although with variable severity.

Demonstration of a cortical reflex origin of myoclo- nus16 was based on the following three findings: ( 1 )

cortical activity preceding myoclonic jerks as demonstrated by jerk-locked back-averaging, (2) enlarged cortical SEPs, and (3) an enhanced C-reflex, at rest, to either electrical or mechanical stimulation. Burst- locked EEG averaging revealed a reproducible potential preceding the EMG burst by an interval appropriate to corticomotoneuronal conduction.” SEP amplitudes were consistent with the definition of “giant” SEPs,” although their values were not as high as in progressive myoclonus epilepsies (PMEs) and their morphology was broader. The C-reflex had unusually long latencies (mean, 62 msec), which were noticeably longer than in PMEs (35-50 msec)’8-20 and showed some similarity to the cortical reflex myoclonus seen in Alzheimer’s disease.’l The latency of the premyoclonus spike was likewise increased (mean, 34 msec), being twice as high as observed in PMEs (-18 msec).22 CMCT was short (mean, 5.44 2 0.54 msec), as previously reported by other investigator^,^^^^^ who considered this finding to indicate integrity and hyperexcitability of corticospinal pathways. The MEP duration was shortened, possibly indicating that only fast conducting descending motor pathways were activated by magnetic stimulation. Also, the post-MEP SP was shortened, confirming cortical hypere~citabil i ty.’~~~~

The time interval between the P30 component and the C-reflex is usually the same as the latency between the premyoclonus spike and the jerk and as the TMCT after TMS.1s327 According to the TMCT we observed, a mean delay of 28.4 2 4.5 msec was necessary from the time when the afferent input reached the postcentral cortex up to the origin of the descendent corticomotoneuronal volley. This intracortical delay is three to four times greater than that observed in PMEs.” The following sequence” could therefore occur in RS patients: mildly delayed conduction of the afferent input up to the sensory cortex via the dorsal columns medial lemniscal and thalamocortical systems (N20 component), slow processing of the afferent input (delayed N20-P30 interval; mean, l l .4 msec), delayed transmission via corticocortical pathways to the precentral cortical neurons corresponding to the stimulated body segment (delayed P30-EMG response; mean, 32.9 msec), and rapid descending volley to the spinal motoneurons. Therefore, the premyoclonus EEG potential (as also the corresponding P30 wave) would appear to represent a discharge in postcentral neurons that would, in turn, slowly drive the motor output through their connections with precentral neurons.’* Another possibility is that the premyoclonus EEG potential directly activates projections from the parietal cortex” to spinal interneurons, which, in turn, recruit motoneurons.28 Regardless of the corticofugal neuronal contingent (precentral or postcentral) whose reflex ac- tivation originates the descendent volley, intracortical conduction time is likely to be markedly delayed due


to the synaptic abnormalities that have been reported in the RS brains.”

Cortical hyperexcitability in RS was also suggested by other investigators, using both SEPs and visual evoked potential^.^"." Our own and previous findings are apparently at variance with the considerable delay of the long-loop reflex and also with the absence of any intrahemispheric and interhemispheric spread of myoclonic cortical activity. Abnormal vertical cortical hyperexcitability and reduced horizontal corticocortical transmission could be hypothesized, resulting in enhanced activity of sensorimotor columns processing input-output volleys from and to the corresponding somatic areas, and in reduced spread of cortical activity. The pathological changes observed in RS could support this model. Selective, nonprogressive involvement of projection neurons in the motor, association, and limbic cortex was reported by Amstrong and col- league~,~ who found pyramidal neurons to show a significant shortening of the basal dendrites in cortical layers I11 and V in the motor and frontal cortex, of the apical dendrites in layer V of the motor cortex, and in the basal dendrites of layer IV of the subiculum. This reduction in dendritic branching is probably responsible for reduction in the total number of synapses and is considered to reflect abnormalities in circuitry and not in neuronal cells themselves.‘

Abnormalities have also been found in the cerebel- lum, where the molecular and the granule cell layers are thinned and hemispheric Purkinje cells are small, pale, and reduced in n ~ m b e r , ~ ~ , ~ ~ showing a truncated dendritic pattern.34 Such structural abnormalities could be responsible for cortical reflex myoclonus, due to loss of cerebellar inhibitory influence on cortical mecha- n i s m ~ . ~ ~ ’ ~ ‘ ; Although several energy metabolism or neu- rotransmitter abnormalities have been reported in RS, they have been contr~versial~~ and their relevance to myoclonus is uncertain.

In the 3 patients treated with piracetam, this drug confirmed its powerful antimyoclonic proper tie^,",^^ producing a considerable reduction in the frequency and severity of jerks, both on intravenous administration and long-term use. However, piracetam treatment did not translate into a substantial clinical benefit, in that myoclonus was not producing, in itself, significant disability.

9- to 1 3 - H ~ pattern, producing a continuous trembling of hands.

In conclusion, RS patients show a distinctive pattern of multifocal cortical reflex myoclonus, with prolonged intracortical delay of the long-loop reflex and a low propensity for spread of myoclonic activity. These clinical and neurophysiological findings are in agreement with previous neuropathological studies indicating synaptic abnormalities and selective, nonprogressive dendritic alterations of projection neurons of the motor and association c ~ r t e x . ~ ’ ~ ~ In our patients, myoclonus had a highly variable severity, showing no clear correlation with the stage of the disease and never represent- ing a prominent clinical feature.

This study was supported by Italian National Institute of Health research funding (Minister0 della Saniti RC 4/94-3/96).

We are grateful to Dr Hiroshi Shibasaki for helpful criticism and to Patrice Rawlins for technical help.

References 1. Rett A. Uber ein eigenartiges hirnatrophisches syndrom bei hy-

perammoniamie in hndesalter. Wien Med Wochenschr 1966; 116:723-738

2. Rett A. Cerebral atrophy associated with hyperammonaemia. In: Vinken PJ, Bruyn GW, eds. Handbook of clinical neurology, vol 29. Amsterdam: Elsevier Science, 1977:305-329

3. Hagberg B, Aicardi J, Dias K, Ramos 0. A progressive syndrome of autism, dementia, ataxia, and loss of purposeful hand use in girls: Rett’s syndrome: report of 35 cases. Ann Neurol 1983; 14:47 1-479

4. Armstrong D, Kay Dunn J, Antalffy B, Trivedi R. Selective dendritic alterations in the cortex of Rett syndrome. J Neuro- pathol Exp Neurol 1995;54:195-201

5. Hagberg B. Rett syndrome: recent clinical and biological aspects. In: Arzimanoglou A, Goutihres F, eds. Trends in child neurology. Paris: John Libbey Eurotext, 1996: 143-146

6. Percy AK. Meeting report: Second International Rett Syndrome Workshop and Symposium. J Child Neurol 1993;8:97-100

7. Hagberg B, Witt-Engerscrom I. Rett syndrome: a suggested staging system for describing impairment profile with increasing age toward adolescence. Am J Med Genet 1986;24:47-59

8. The Rett Syndrome Diagnostic Criteria Work Group. Diagnos- tic criteria for Rett syndrome. Ann Neurol 1988;23:425-428

9. FitzGerald PM, Jankovic J, Glaze DG, er al. Extrapyramidal involvement in Rett’s syndrome. Neurology 1990;40:293-295

10. FitzGerald PM, Jankovic J, Percy AK. Rett syndrome and associated movement disorders. Mov Disord 1990;5:195-202

11. Ikeda A, Shibasaki H, Nagamine T, et al. Peri-rolandic and fronto-parietal components of scalp-recorded giant SEPs in cortical myoclonus. Eleccroencephalogr Clin Neurophysiol 1995;

Although the behavioral phenotype of classic RS is quite characteristic, some difficulty in differential diagnosis with other disorders may be found, especially in the early stages of the disease8 A condition that may Drove difficult to differentiate clinically from RS is An-

96:300-309 12. Shibasaki H, Yamashira Y, Tsuji S. Somatosensory evoked po-

tentials. Diagnostic criteria and abnormalities in cerebral le- sions. J Neurol Sci 1977;34:427-439

13. Shibasaki H, Yamashita Y, Ryuji N, et al. Pathogenesis of giant somatosensory evoked potentials in progressive myoclonic epi-

gelman’s ~yndrome,~’ in which cortical myoclonus is a

the C-reflex is nor hyperexcitable, SEPs are not en-

lepsy. Brain 1985;108:225-240 14. Niedermeyer E, Rett A, Renner H, et al. Rett syndrome and

the electroencephalogram. Am J Med Genet 1986;24:195-199 15. Rolando S. Rett syndrome: report of eight cases. Brain Dev

constant feature? However, in Angelman’s syndrome,

- larged, and myoclonus shows a characteristic rhythmic 1985;7:290-296


16. Hallett M, Chadwick D, Marsden CD. Cortical reflex myoclonus. Neurology 1979;29:1107-1125

17. Shibasaki H, Kuroiwa Y. Electroencephalographic correlates of myoclonus. Electroencephalogr Clin Neurophysiol 1975;39: 4 5 5- 463

18. Shibasaki H, Yamashita Y, Kuroiwa Y. Electroencephalographic studies of myoclonus: myoclonus-related cortical spikes and high amplitude somatosensory evoked potentials. Brain 1978; 10 ~447-460

19. Rothwell JC, Obeso JA, Marsden CD. Electrophysiology of somatosensory reflex myoclonus. In: Fahn S, Marsden CD, eds. Advances in neurology, vol 43. Myoclonus. New York: Raven Press, 1986:385-398

20. Thompson I’D, Day BL, Rothwell JC, et al. The myoclonus in corticobasal degeneration: evidence for two forms of cortical reflex myoclonus. Brain 1994;117:1197-1207

21. Wilkins DE, Hallett M, Berardelli A, et al. Physiologic analysis of the myoclonus of Alzheimer’s disease. Neurology 1984;34: 838-903

22. Shibasaki H, Kakigi R, Ikeda A. Scalp topography of giant SEP and pre-myodonus spike in cortical reflex myoclonus. Electro- encephalogr Clin Neurophysiol 199 1;8 1:3 1-37

23. Eyre JA, Kerr AM, Miller S, et al. Neurophysiological observations on corticospinal projections to the upper limb in subjects with Rett syndrome. J Neurol Neurosurg Psychiatry 1990;53: 874-879

24. Heinen F, Korinthenberg R. Does transcranial magnetic stimulation allow early diagnosis of Rett syndrome? Neuropediatrics 1996;27:223-224 (Letter)

25. Uozumi T, Yoichi I, Tsuji S, Murai Y. Inhibitory period following motor potentials evoked by magnetic cortical stimulation. Elecrroencephalogr Clin Neurophysiol 1992;85:273-279

26. Guerrini R, De Lorey T , Bonanni P, et al. Cortical myoclonus in Angelman syndrome. Ann Neurol 1996;39:699-708

27. Barker AT, Jalinous R, Freeston IL. Non-invasive magnetic stimulation of human cortex. Lancet 1985;1:1106-1107

28. Uesaka Y, Terao Y, Ugawa Y, et al. Magnetoencephalographic analysis of cortical myoclonic jerks. Elecrroencephalogr Clin Neurophysiol 1996;99:141-148

29. Belichenko PV, Oldfors A, Hagberg BA, Dahlstrom A. Rett syndrome: 3-D confocal microscopy of cortical pyramidal dendrites and afferena. Neuroreport 1994;5:1509-1513

30. Yoshikawa H, Kaga M, Suzuki H, et al. Giant somatosensory evoked potentials in the Rett syndrome. Brain Dev 1991;13:

31. Yamanouchi H, Kaga M, Arima M. Abnormal cortical excit- ability in Rett syndrome. Pediatr Neurol 1993;9:202-206

32. Oldfors A, Sourander P, Armstrong DL, et al. Rett syndrome: cerebellar pathology. Pediatr Neurol 1990;6:310-314

33. Bauman M, Kemper L, Arin DM. Microscopic observations of the brain in Rett syndrome. Neuropediatrics 1995;26:105-108

34. Cornford ME, Philippart M, Jacobs B, et al. Neuropathology of Rett syndrome: case report with neuronal and mitochondria1 abnormalities in the brain. J Child Neurol 1994;9:424-431

35. Marsden CD, Hallett M, Fahan S. The nosology and patho- physiology of myoclonus. In: Marsden CD, Hallett M, Fahan S, eds. Movement disorders. London: Buttenvorths Scientific, 1982: 196 -249

36. Obeso J, Rothwell JC, Marsden CD. The spectrum of cortical myoclonus: from focal reflex jerks to spontaneous motor epi- lepsy. Brain 1985; 108: 193-224

37. Hanefeld F, Hagberg B, Percy A. Molecular and neurobiology aspects of Rett syndrome. Neuropediatrics 1995;26:60-61

38. Brown P, Steiger PD, Thompson JC, et al. Effectiveness of piracetam in cortical myoclonus. Mov Disord 1993;8:63-68

39. Zori RT, Hendrickson J, Woolven S, et al. Angelman syndrome: clinical profile. J Child Neurol 1992;7:270 -280

36-39


cortical reflex myoclonus in rett syndrome

Documents