Address correspondence to Fabrice Bartolomei, Pr Fabrice Bartolomei, Service de Neurophysiologie Clinique, CHU Timone-264 Rue st Pierre, F-13005-Marseille, France. E-mail: firstname.lastname@example.org
We aimed at determining the lateralizing and localizing values of “hemiballic-like” ictal movements observed in some partial seizures. Among 20 patients disclosing ictal hyperkinetic features and explored by stereotactic-EEG (SEEG), this sign was observed in four patients. In these cases, hemiballic movement was ipsilateral to the ictal-onset zone and was associated with contralateral ictal dystonia. Noninvasive and subsequent invasive recording revealed seizure origin in the inferior parietal lobule or the parietal operculum in three patients and in the inferior prefrontal cortex in one.
Ictal hyperkinetic motor behavior involves large amplitude, and rapid and often explosive movements of the body axis (such as body rocking, pelvic thrusting) and limbs, sometimes associated with vocalization, emotional symptoms (particularly fear), and/or facial grimacing. Seizures giving rise to such behaviors have been reported as originating from diverse brain regions, most notably frontal lobe (Williamson et al., 1985; Rheims et al., 2008), but also temporo-perisylvian cortices (Ryvlin et al., 2006; Vaugier et al., 2009) and posterior regions (Bartolomei et al., 2011).
In such ictal hyperkinetic motor behavior, purposeless movements of the upper limbs are common, generally bilateral, and may be associated with tonic posturing. In this context we recently observed the occurrence of particular repetitive unilateral rotational arm movements during seizures, reminiscent of hemiballismus due to the proximal rotational quality of the movements and their large amplitude. However, unlike classical hemiballismus the movements were typically highly repetitive and stereotyped. Highly stereotyped upper limb movements have been previously reported as a very rare semiologic feature of supplementary motor area seizures (Barba et al., 2005). We aimed to determine the lateralizing and localizing value of this clinical sign occurring as a feature of ictal hyperkinetic motor behavior in patients investigated by intracerebral recordings.
We assessed the features of rotational proximal upper limb stereotyped movements, occurring during partial seizures including hyperkinetic features, in a population of patients with epilepsy who were undergoing presurgical evaluation and admitted to the adult epilepsy unit of Timone hospital (Marseille, France) between 2001 and 2009. Depth electrodes recording using stereoelectroencephalography (SEEG) was performed according to Talairach’s stereotactic method (Vaugier et al., 2009).
Among 20 patients disclosing ictal hyperkinetic features, four (two men) presented ictal rotational proximal upper limb (“pseudo-ballic”) stereotyped movements. Their clinical and electrophysiologic features are summarized in Table 1. This movement occurred at the beginning of the hyperkinetic signs, several seconds (mean 10 s, SD ±6) after seizure onset. Most of the studied seizures occurred during sleep. In all four cases, the characteristic movement was ipsilateral to the ictal-onset zone and was associated with simultaneous contralateral dystonia (see Fig. 1). We did not observe any contralateral movement before the beginning of the ipsilateral abnormal movement. This semiology was present in all or most of the seizures in three patients (Patients 1, 2, and 3). Noninvasive and subsequently invasive recording revealed that the seizure origin was located in the inferior parietal lobule or the parietal operculum in three patients (Patients 1, 2, and 4) and in the inferior prefrontal cortex in one (Patient 3) (see Table 1; Fig. 2).
Table 1. Clinical and electrophysiologic features of the four patients with ictal rotational proximal upper limb stereotyped movement
MCD, malformation of cortical development; FCD, focal cortical dysplasia; F, female; M, male; LL, lower limbs; UL, upper limbs.
aMean value for three studied seizures.
bSMA and premotor cortex was not explored in this case.
Age at onset (years)
FCD in the left parietal region (parietal operculum)
MCD in the right hemisphere
FCD in the right inferior parietal lobule
Type of UL motor automatisms
Dystonic posture of the right UL associated with ballic-like movement of the left UL Present in all seizures
Dystonic posture of the left UL associated with ballic-like movement of the right UL and LL Present in most of the seizures
Tonic elevation and dystonic posture of the right UL associated with a tonic movement immediately followed by a ballic-like movement Present in most of the seizures
Tonic posture of the left UL, and right UL ballic-like movement Present only in three seizures ( of 15)
Delay of appearance of ballic-like movement compared to onset of seizure discharge (seconds)a
Loss of contact, LL motor automatisms, hypersalivation, respiratory arrest
Loss of contact, right ocular and cephalic deviation, groaning, vocalization, LL motor automatisms
Loss of contact, facial contraction (left cephalic deviation, rocking, LL motor automatisms, shrieks or grunts
Loss of contact, motor automatisms such as rocking and pedaling, shriek, axial body movements
Epileptogenic zone (regions involved at onset)
Left parietal operculum
Right parietal (inferior parietal lobule) and first temporal gyrus
Left lateral inferior prefrontal cortex
Right parietal (inferior parietal lobule)
Involved structure at the onset of ballic-like movement
Parietal operculum Inferior parietal lobuleb
Parietal operculum First temporal gyrus Inferior parietal lobule SMA not involved
SMA Orbitofrontal cortex
SMA Posterior cingulate Parietal operculum
Taylor-type IIb FCD
Taylor-type IIb FCD
Follow-up after surgery (years/Engel)
Our results suggest that ictal rotational proximal upper limb stereotyped movement is an infrequent but possibly underestimated phenomenon in seizures with hyperkinetic motor behavior. When present, it appears to be a useful clue to lateralize the ictal-onset zone. Indeed, the four reported patients had this abnormal movement in the upper limb ipsilateral to the ictal-onset zone. In all four cases it was associated with contralateral ictal dystonia. In a previous study, Geier et al. (1977) also reported that in epilepsy originating from parietal regions, ipsilateral ictal automatisms and contralateral tonic postural modifications could be observed. However, a similar pattern of contralateral dystonia and ipsilateral automatic movements commonly occurs in temporal lobe seizures (TLS) (Dupont et al., 1998). Of note, another type of automatic movement (“nonmanipulative proximal upper extremity automatisms”) has been described in TLS. These movements were described as rhythmic or semirhythmic, nonmanipulative movements with a frequent circular component, affecting mainly the proximal part of the extremity (Kelemen et al., 2010). These were found to have high lateralizing value to the contralateral side, a finding contrasting our observations. In our cases, it seems likely that contralateral dystonia could reflect a specific ictal basal ganglia implication (Dupont et al., 1998). In contrast to a previous report of two similar cases (Barba et al., 2005), we did not find an origin within the mesial premotor cortex. Indeed, in one case the origin of seizures was prefrontal ([although supplementary motor area (SMA) cortex was involved at the time of appearance of abnormal movements] and in the other three cases the epileptogenic zone was located in posterior cortex including inferior parietal region or the operculoinsular cortex. This characteristic sign has not, however, been observed in any other frontal lobe epilepsies explored in our center. This sign could therefore have some localizing value orienting toward a posterior origin of seizures producing hyperkinetic motor behavior.
Outside the context of epileptic seizures, the movement disorder of true hemiballismus is due to a lesion of the subthalamic nucleus. Although the movements we describe here clearly differ from the classical description of hemiballismus (which is multidirectional and wholly unpredictable as opposed to stereotyped), we may nevertheless hypothesize that the ictal association of simultaneous contralateral dystonia and ipsilateral abnormal movement is secondary to the spread of the ictal discharge to subcortical regions, possibly implicating the striatum and the subthalamic nucleus. This hypothesis is supported by known anatomofunctional connections between temporoparietal cortex and basal ganglia (Uddin et al., 2010). The explanation of why the abnormal upper limb ictal movement should be ipsilateral is unclear. Activation of motor system ipsilateral to the seizure onset could be associated with a contralateral inhibition of subcortical structures. Some authors have demonstrated that in voluntary actions during choice reaction time tasks, activation of the motor structures involved in the required response was associated with a contralateral inhibition (Burle et al., 2002).
We thank Professor Jean Regis for stereotactic placement of electrodes and Professor JC Peragut for surgical procedure of selected patients.
None of the authors has any conflict of interest to disclose. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.