• Autonomic symptoms;
  • Children;
  • Temporal and extratemporal lobe epilepsy;
  • Lateralizing signs


  1. Top of page
  2. Abstract
  5. Acknowledgments

Summary: Purpose: To analyze systematically the occurrence and age dependence as well as the localizing and lateralizing value of ictal autonomic symptoms (ASs) during childhood partial epilepsies and to compare our results with those of earlier adult studies.

Methods: Five hundred fourteen video-recorded seizures of 100 consecutive children 12 years or younger with partial epilepsy and seizure-free postoperative outcome were retrospectively analyzed.

Results: Sixty patients produced at least one AS; 43 (70%) of 61 with temporal and 17 (44%) of 39 with extratemporal lobe epilepsy (p = 0.012). Apnea/bradypnea occurred more frequently in younger children (p < 0.01), whereas the presence of other ASs was neither age nor gender related. Postictal coughing (p < 0.01) and epigastric aura (p < 0.05) localized to the temporal lobe, whereas no ASs lateralized to the seizure-onset zone.

Conclusions: Our study shows that ASs are common in childhood focal epilepsies, appearing in infants and young children, too. As in adults, childhood central autonomic networks might have a close connection to temporal lobe structures but do not lateralize the seizure-onset zone. To our knowledge, this is the first study comprehensively assessing ASs in childhood epilepsy.

Autonomic symptoms (ASs) during epileptic seizures are of both theoretical and clinical interest for different reasons. Periictal ASs can help us better understand the central representation of the autonomic nervous system (1), and some of them can add further information to the assessment of the epileptogenic focus, which is especially useful in presurgical evaluation of intractable epilepsy (2). Additionally, autonomic alterations, either in the periictal period or as a result of presumed interictal autonomic dysfunction, may play a role in the increasing incidence of sudden unexpected death in epilepsy patients (3). A variety of ASs have been described in adults, including cardiovascular, respiratory, gastrointestinal, cutaneous, pupillary, and urinary manifestations (3–8). Although such symptoms have unique importance in the evaluation of young and preverbal patients with epilepsy, no clinical study assessed ASs in children.

We performed this investigation, by using retrospective videotaped seizure analysis, to describe systematically the frequency and age dependence as well as the localizing and lateralizing value of childhood ASs and to compare our results with those of earlier adult studies.


  1. Top of page
  2. Abstract
  5. Acknowledgments


The study included 100 consecutive patients 12 years or younger who underwent presurgical evaluation at the Bethel Epilepsy Center (Bielefeld, Germany) between January 1990 and January 2005 and became seizure free after frontal, temporal, or posterior cortex resection. Patients with mixed temporal and extratemporal operation (e.g., frontotemporal or temporoparietal) were excluded.

Data collection of autonomic symptoms

Time-labeled ictal video recordings were retrospectively reviewed by two investigators (A.F. and J.J.) blinded to the patients' clinical data except for epilepsy. Seizures in which the recording quality did not permit analysis of the complete seizure from onset to end were excluded (∼5%). Finally, we reviewed 514 seizures of 100 children (median, five; range, two to 15 seizures per child). Each seizure was analyzed by one of the reviewers with regard to the autonomic signs during the preictal, ictal, and postictal periods. Data were documented on an especially designed data sheet listing the following ASs described in adults (3–8): respiratory (hyperventilation, apnea, bradypnea, dyspnea, cyanosis, postictal coughing); gastrointestinal (epigastric aura, nausea, vomiting, spitting, hypersalivation, belch, hiccup); cutaneous (flushing, pallor, piloerection); pupillary (mydriasis, miosis); and urinary (ictal urinary urge) manifestations. We planned to assess symptoms of partial seizures; therefore no ASs during secondarily generalized seizures were marked. Respiratory symptoms were assessed visually. During many seizures—especially if the seizure started in sleep or with hypomotor components—it was not difficult to observe an increase or decrease in breathing frequency or even an arrest of respiration. These visual observations on ictal changes of breathing frequency were defined as hyperventilation, bradypnea, or apnea. Dyspnea was defined as any visible difficulty of breathing (e.g., a stridor during tonic seizure components). Because <10% of children had continuous electrocardiogram (ECG) or pulse monitoring, cardiovascular manifestations were not assessed in our study.

Besides seizure review, each patient's medical charts were reviewed to collect data on age, age at onset, as well as epilepsy surgery. The localization and lateralization of the seizure-onset zone were defined by the localization of surgery resulting in seizure freedom.

Statistical methods

For the analysis of the categoric data, χ2 or Fisher's exact tests were carried out. For the continuous variable (age dependence), the Mann–Whitney test was used. Two-tailed error probabilities of p < 0.05 were considered to be significant. Very rare ASs (observed in ≤5% of patients) were not assessed with statistical methods. All statistical analyses were carried out with the SPSS 11.5 statistical package for WINDOWS (SPSS Inc., Chicago, IL, U.S.A.).


  1. Top of page
  2. Abstract
  5. Acknowledgments

Clinical data

The study group consisted of 100 patients (52 girls) aged 10 months to 12 years (mean, 6.1 ± 3.5 years; Fig. 1). According to the seizure-free results of lobar resections, 61 children had temporal lobe epilepsy (TLE), and 39 had extratemporal (21 frontal, 12 occipital, and six parietooccipital) seizure-onset zones [postoperative follow-up was between 6 months and 5.5 (mean, 2.0) years]. Surgery was left-sided in 54 and right-sided in 46 cases. The underlying etiology of epilepsy was malformation of cortical development in 49, dysgenetic tumors (ganglioglioma, hamartoma, and dysembryoplastic neuroepithelial tumor) in 32, hippocampal sclerosis in nine, and other pathology (tumor, tuberous sclerosis, stroke, Sturge–Weber syndrome, cavernoma, or abscess) in 10 children.


Figure 1. Age distribution of the study population (n = 100).

Download figure to PowerPoint

Autonomic symptoms

According to seizure analysis, 60 patients produced at least one AS during their seizures: 43 (70%) of 61 with TLE and 17 (44%) of 39 with extratemporal epilepsy (p = 0.012). Each child showed no to three (median, one) different ASs. The most frequent ASs were ictal flushing (19), postictal coughing (16), epigastric aura (12), apnea/bradypnea (12), and hyperventilation (in 11 cases); other ASs appeared in <10% of children (Table 1.). Apnea/bradypnea happened more frequently in younger children (p < 0.01; Fig. 2), whereas the presence of other ASs was neither age nor gender related. Postictal coughing (p < 0.01) and epigastric aura (p < 0.05) localized to the temporal lobe, whereas no ASs lateralized the seizure-onset zone.

Table 1. Frequency, localization, lateralization, and age dependence of autonomic signs observed in 100 patients 12 years or younger with partial epilepsy
Autonomic symptomsTotal (n = 100)LocalizationLateralizationAge dependence (p value)
Temporal (n = 61)Extratemporal (n = 39)p ValueLeft (n = 54)Right (n = 46)p Value
  1. Very rare autonomic symptoms (observed in ≤5% of patients) were not assessed with statistical methods.

  2. NS, no significant correlation (p ≥ 0.05)

  3. aIctal apnea/bradypnea happened more frequently among younger patients (p < 0.001).

Flushing19118NS712 NSNS
Postictal coughing161510.00410 6NSNS
Apnea, bradypnea12 84NS75NS<0.001a
Epigastric aura121110.02657NSNS
Hyperventilation11 83NS65NSNS
Dyspnea 9 63NS36NSNS
Hypersalivation 5 41NS23NSNS
Vomiting 5 4141
Nausea 3 3003
Spitting 2 2002
Miosis 1 1001
Hiccup 1 1010
Belch 1 0101
Total604317 0.01229 31 NSNS

Figure 2. Ictal apnea/bradypnea happened more frequently in younger children (Mann–Whitney test; p < 0.001). Presence of other autonomic signs, however, was not age dependent adding important clinical information to the lateralization and localization of seizure onset in infants and children of all ages. (Box plots show range, standard deviation, and median of each population. *An outlier.)

Download figure to PowerPoint

Among the ASs described in adults, cyanosis, pallor, pilomotor seizure, mydriasis, and urinary urge were not observed during the 514 seizures of these children.


Our study shows that ASs are common in childhood focal epilepsies appearing already in infants and young children. Except for cardiovascular changes (which require objective ictal measuring), we assessed changes in all human systems innervated by the autonomic nervous system.

Respiratory manifestations

Hyperventilation is a frequently reported symptom of partial seizures. It can be found in both frontal and temporal lobe epilepsies, occurring more frequently in mesial compared with neocortical temporal lobe seizures (9). Our group of children showed hyperventilation more frequently in TLE; however, this difference was not significant. Both apnea and cyanosis are typical signs of generalized tonic–clonic seizures in adults. Isolated apnea or bradypnea during partial seizures occurs in children, especially in neonates and infants with hypomotor seizures (10). Similarly, apnea or bradypnea happened as age-specific phenomena in our children, occurring more frequently in younger patients. The 12% prevalence is very high, considering that ictal apnea might be life threatening (11). Dyspnea is not a well-defined ictal autonomic entity in the literature; however, our children had with this symptom relatively frequently (9%); mostly (six of nine cases) during the tonic phase of epileptic seizures.

Increased upper-airway secretion during seizures might result in early postictal reactive coughing (12,13). Our results showed temporal dominance in postictal coughing but no right-hemisphere predominance as described in adults (7,14). Its high frequency and age-independent appearance make this easily observable phenomenon a clinically important AS of childhood partial seizures.

Gastrointestinal manifestations

Twelve children reported epigastric aura; the youngest of them was 5.3 years old at video-EEG monitoring. Although this type of aura may consist of a wide variety of complex sensations in adults (e.g., tenseness, squeezing, tingling, vibrating, warmth, burning) (15), our children simply reported abdominal pain or discomfort as their epigastric aura. Epigastric auras are most frequently encountered in TLE (7,16); this tendency was observed in our childhood study as well. Electrical-stimulation studies showed that a sensation similar to epigastric aura can be elicited from the hippocampus, amygdala, and insula, as well as the basal ganglia and thalamus (15). Similarly, all of our 11 children with epigastric aura and TLE had a lesion in the mesial temporal structures: nine with pure mesial lesion and two patients with a lesion involving both mesial and neocortical regions.

Some authors stated that epigastric aura lateralizes to the nondominant temporal lobe (17), whereas most studies found no hemispheric differences (13,16). Similar to the latter results, epigastric auras were reported in our children with both left- and right-sided epilepsies. In two children, only isolated epigastric auras were observed without further complex partial seizures. This special phenomenon—generally referred to as abdominal epilepsy—was found more commonly in children than in adults (18). Both of our patients with abdominal epilepsy—a 6-year-old girl and a 9-year-old boy—had right mesial TLE due to dysgenetic tumors.

Three children had isolated ictal nausea, and five patients had ictal vomiting in our study. This frequency is slightly higher than the incidence of vomiting reported by adult studies (19,20). According to subdural grid electrode (19) and ictal single-photon emission computed tomography (SPECT) studies (21), ictal vomiting was associated with medial and lateral temporal lobe seizures and with spread to the superior lateral temporal region. Although medial and lateral temporal lobe seizure semiology shows no important differences in young children (22), it is important to note that none of our children showed a seizure consisting both epigastric aura (a usually medial TLE symptom) and ictal vomiting (rather temporal lateral sign).

Autonomic manifestations of other systems

Ictal flushing was observed in 19 children and showed neither lateralizing nor localizing value. It was mostly facial and happened not only during motor (eight cases) but also during psychomotor seizures (in 11 children), proving that this phenomenon is caused not exclusively by skin hyperperfusion during motor activities but is also under control of the central nervous system. One reason for ictal flushing of the latter cases could be psychic stress during certain psychomotor seizures.

We observed ictal miosis—a rare seizure-associated AS (23)—in a 3-year-old girl with right-sided mesiotemporal ganglioglioma. Her seizures consisted of pure miosis series without other motor signs, ASs, or automatism. Ictal miosis in our case was not accompanied with the ophthalmoplegia reported earlier in a patient with left temporooccipital epilepsy (24).

Conclusions and limitations of the study

Periictal ASs were frequent in our children and showed a strong relation to the temporal lobe and its functional networks. Although the human autonomic network includes several parts of the brain (hypothalamus, insula, medial prefrontal cortex, amygdala, periaqueductal gray matter, parabrachial complex, nucleus tractus solitarius, and the ventrolateral medulla) (1), the mechanism by which partial seizures cause ASs is thought to be spread by an ictal discharge from cortex to the richly connected hypothalamus (6). Thus the richness of limbic–hypothalamic connections also might account for the prominence of ASs of temporal lobe seizures in children.

To our knowledge, this is the first study comprehensively describing the hemispheric lateralization of different ASs in human epilepsy. Our study's limitation is its retrospective method, possibly influencing the report of subtle ASs by the variable diligence and skills of the EEG technicians. Missing some very delicate ASs (e.g., piloerection or mydriasis) might not mean their total lack in childhood, but that even a very modern diagnostic method (video-EEG monitoring) can be unable to record them. Conversely, efficient data of such a selected patient population (12 years or younger and seizure-free postoperative outcome) were archived during 15 years, making a prospective study very time consuming, even in a specialized epilepsy center. Nevertheless, analyzing epileptic seizures seems to be a useful method for assessing functions of the human autonomic nervous system; therefore we suggest prospectively verifying our results in a large population of children and adults with partial epilepsy.


  1. Top of page
  2. Abstract
  5. Acknowledgments

Acknowledgment:  We are grateful to all neurologists—especially Dr. Hans Holthausen, Dr. Reinhard Schulz, and Dr. Alois Ebner—and EEG technicians who participated in the presurgical evaluation of the examined patients. We also thank Dr. Tom Pieper and Ms. Hedwig Freitag for developing and taking care of the pediatric database of the Bethel Epilepsy Center. A.F. was supported by grant D 048517 from the Hungarian Scientific Research Fund. J.J. was supported by the Humboldt Fellowship (Germany) and by the Bolyai Scholarship (Hungary).


  1. Top of page
  2. Abstract
  5. Acknowledgments
  • 1
    Benarroch EE. The central autonomic network: functional organization, dysfunction, and perspective. Mayo Clin Proc 1993;68: 9881001.
  • 2
    Rosenow F, Luders H. Presurgical evaluation of epilepsy. Brain 2001;124: 16831700.
  • 3
    Reeves AL. Autonomic activity in epilepsy: diagnostic considerations and implications. J Epilepsy 1997;10: 111116.
  • 4
    Chadwick D, Cartlidge N, Bates D. Medical Neurology. London : Churchill Livingstone, 1989: 8097.
  • 5
    Freeman R, Schachter SC. Autonomic epilepsy. Semin Neurol 1995;15: 158166.
  • 6
    Liporace JD, Sperling MR. Simple autonomic seizures. In: EngelJJr, PedleyTA, eds. Epilepsy: A Comprehensive Textbook. Philadelphia : Lippincott-Raven, 1997: 549555.
  • 7
    Baumgartner C, Lurger S, Leutmezer F. Autonomic symptoms during epileptic seizures. Epileptic Disord 2001;3: 103106.
  • 8
    Devinsky O. Effects of seizures on autonomic and cardiovascular function. Epilepsy Curr 2004;4: 4346.
  • 9
    Foldvary N, Lee N, Thwaites G, et al. Clinical and electrographic manifestations of lesional neocortical temporal lobe epilepsy. Neurology 1997;49: 757763.
  • 10
    Watanabe K, Hara K, Hakamada S, et al. Seizures with apnea in children. Pediatrics 1982;70: 8790.
  • 11
    Singh B, Al Shahwan A, Al Deeb SM. Partial seizures presenting as life-threatening apnea. Epilepsia 1993;34: 901903.
  • 12
    Van Ness PC, Marotta J, Kucera A, et al. Postictal cough is a sign of temporal lobe epilepsy. Neurology 1993;43(suppl 2):A273.
  • 13
    Gil Nagel A, Risinger MW. Ictal semiology in hippocampal versus extrahippocampal temporal lobe epilepsy. Brain 1997;120: 183192.
  • 14
    Fakhoury T, Abou-Khalil B, Peguero E. Differentiating clinical features of right and left temporal lobe seizures. Epilepsia 1994;35: 10381044.
  • 15
    Van Buren JM. The abdominal aura: a study of abdominal sensations occurring in epilepsy and produced by depth stimulation. Electroencephalogr Clin Neurophysiol 1963;15: 119.
  • 16
    Palmini A, Gloor P. The localizing value of auras in partial seizures: a prospective and retrospective study. Neurology 1992;42: 801808.
  • 17
    Gupta AK, Jeavons PM, Hughes RC, Covanis A. Aura in temporal lobe epilepsy: clinical and electroencephalographic correlation. J Neurol Neurosurg Psychiatry 1983;46: 10791083.
  • 18
    Livingston S. Abdominal pain as a manifestation of epilepsy (abdominal epilepsy) in children. J Pediatr 1958;38: 687695.
  • 19
    Kramer RE, Lüders H, Goldstick LP, et al. Ictus emeticus: an electroclinical analysis. Neurology 1988;38: 10451052.
  • 20
    Devinsky O, Frasca J, Pacia SV, et al. Ictus emeticus: further evidence of nondominant temporal involvement. Neurology 1995;45: 11581160.
  • 21
    Baumgartner C, Olbrich A, Lindinger G, et al. Regional cerebral blood flow during temporal lobe seizures associated with ictal vomiting: an ictal SPECT study in two patients. Epilepsia 1999;40: 10851091.
  • 22
    Fogarasi A, Jokeit H, Faveret E, Janszky J, Tuxhorn I. The effect of age on seizure semiology in childhood temporal lobe epilepsy. Epilepsia 2002;43: 638643.
  • 23
    Afifi AK, Corbett JJ, Thompson HS, Wells KK. Seizure-induced miosis and ptosis: association with temporal lobe magnetic resonance imaging abnormalities. J Child Neurol 1990;5: 142146.
  • 24
    Rosenberg ML, Jabbari B. Miosis and internal ophthalmoplegia as a manifestation of partial seizures. Neurology 1991;41: 737739.