Agyria-Pachygyria: Clinical, Neuroimaging,
and Neurophysiologic Correlations
Jao-Shwann Liang, MD*, Wang-Tso Lee, MD, PhD*, Chainllie Young, MD, PhD*,
Steven Shinnforng Peng, MD†, and Yu-Zen Shen, MD, PhD*
Agyria-pachygyria complex is a disorder of neuronal
migration and organization. Patients suffer either mo-
tor or intellectual retardation. We report our experi-
ences of 10 patients with agyria-pachygyria complex
and evaluate their clinical features, electroencephalog-
raphy, and evoked potentials. Of nine electroencepha-
lography examinations, five patients demonstrated
characteristically high-amplitude fast activity. One of
nine patients had an abnormal brainstem auditory-
evoked potential. Three of seven patients had abnor-
mal goggled visual-evoked potential. Six patients re-
ceived somatosensory-evoked potential examinations,
and five of these were abnormal, including four with
prolonged central conduction times. Of the 10 patients,
eight survived with variable intellectual and motor
retardation; two died of sepsis. Patients with grades 1-4
agyria-pachygyria had high incidences of somatosen-
sory-evoked potential abnormalities and also suffered
worse neurologic outcomes. Normal brainstem audito-
ry-evoked potential but abnormal cortical somatosen-
sory-evoked potential components and prolonged cen-
tral conduction time in these patients indicate that
agyria-pachygyria is a supratentorial disease. We con-
clude that somatosensory-evoked potential examina-
tion is supplemental to neuroimaging in predicting the
neurologic prognosis of patients with agyria-pachygyria.
© 2002 by Elsevier Science Inc. All rights reserved.
Liang J-S, Lee W-T, Young C, Peng SS, Shen Y-Z.
Agyria-pachygyria: Clinical, neuroimaging, and neuro-
physiologic correlations. Pediatr Neurol 2002;27:171-176.
Introduction
Agyria-pachygyria is a disease of neuronal migration
characterized by the total or partial absence of sulci and a
From the Departments of *Pediatrics and †Radiology; National Taiwan
University Hospital; Taipei, Taiwan
© 2002 by Elsevier Science Inc. All rights reserved.
PII S0887-8994(02)00401-0 ● 0887-8994/02/$—see front matter
thickened cortical mantle [1,2]. Pachygyria is considered
the less severe form and exhibits a broader spectrum of
clinical features than agyria. In many cases, agyria and
pachygyria coexist in the same patient [3]. They represent
degrees of the same fundamental disorders in cell migra-
tion and cortical organization [4]. More and more patients
have been documented as agyria-pachygyria since the
advent of computed tomography (CT) and magnetic res-
onance imaging (MRI). Although some of these disorders
have a genetic basis, the cause of the migrational arrest is
still unknown.
Patients with agyria-pachygyria are always retarded
in their motor or intellectual function, and seizures are
common early in life [5]. Some unusual electroen-
cephalography (EEG) patterns have been proposed to
suggest a diagnosis of agyria-pachygyria [5-7]. However,
there are limited studies focusing on the changes of
evoked potentials in these patients. We report our experi-
ences of 10 patients with various types of agyria-
pachygyria and evaluate their clinical features, EEG, and
evoked potentials. Additionally, we try to clarify the
relationship between clinical presentations, neuroimaging,
and evoked potential findings in these patients.
Patients and Methods
We reviewed the medical records of patients with agyria-pachygyria at
the National Taiwan University Hospital from February 1992 to February
2001. Agyria-pachygyria was diagnosed based on clinical and neuroim-
aging findings. To correlate the extent of agyria-pachygyria with clinical
findings and evoked potential changes, we modified the classification of
agyria-pachygyria complex into six degrees of severity to include focal
pachygyria, according to radiologic findings from Andel’s classification
[8] (Table 1). Neurophysiologic studies, including EEG and evoked
potentials studies, were performed in these patients except in those
patients who did not cooperate. A 14-channel Nihon-Kohden EEG
machine was used in the EEG studies. Electrodes were placed according
to the 10-20 system method. Special attention was given to high-
amplitude (more than 50 ␮V) fast activities. The multimodality evoked
Communications should be addressed to:
Dr. Lee; Department of Pediatrics; National Taiwan University
Hospital; No 7, Chung-Shan South Road; Taipei 100, Taiwan.
Received September 11, 2001; accepted February 21, 2002.
Liang et al: Analysis of Agyria-Pachygyria Complex 171
Table 1. Classification of agyria-pachygyria according to
neuroimaging findings
Grade 1 Complete agyria
Grade 2 Agyria with some sulci
Grade 3 A mixture of approximately 50% agyria and 50%
pachygyria
Grade 4 Complete pachygyria
Grade 5 Focal pachygyria more than 50% of cortex
Grade 6 Focal pachygyria less than 50% of cortex
Modified from Andel et al. [8].
potentials were recorded with a Bravo EP machine (Nicolet, Madison,
WI) and included goggled visual-evoked potential, brainstem auditory-
evoked potential, and somatosensory-evoked potential to the median
nerve stimulation.
Visual-evoked potentials using light-emitting diode goggles were
performed. Light-emitting diode stimulus was a red flash presented
monocularly using Nicolet 1015 goggles delivered at 1 flash/second.
Silver/silver chloride electrodes were positioned using the International
Electrode system at Oz. All active sites were referred to Cz, and the
patient was grounded from the FPz.
Monaural rarefaction click auditory brainstem responses (ABRs) were
elicited at an 21.1 clicks/second with the contralateral ear masked.
Recording electrodes were placed on the vertex (Cz) and referred to the
ears ipsilateral and contralateral to the side of stimulation. Waves I, III,
and V were identified in the resulting ABR tracing, and the latencies of
these peaks were determined, as well as the I-III, III-V, and I-V interpeak
intervals.
In somatosensory-evoked potential studies, the median nerves were
stimulated with electric pulses of 0.1-ms duration at a stimulus frequen-
cies of 2.7/seconds. Surface electrodes were placed on the scalp 2 cm
posterior to C3 and C4 and the cervical spine C7. The reference elec-
trode was placed at Fpz. Latencies of the sensory-evoked potential
components at the points of C7 (N13) and the cortex (N1) were especially
reviewed. Central conduction time (N1-N13) was compared with the
normative data for the median nerve sensory-evoked potentials of Taylor
and Fagan [9].
Results
Ten patients, including six females and four males, with
agyria-pachygyria were enrolled (mean age ⫽ 5.4 years;
age range ⫽ 1 year-14 years and 8 months). The initial
clinical presentations of these patients are presented in
Table 2. Four patients were admitted for seizures, and
three were admitted for developmental delays. Of the 10
patients, nine received MRI examinations, and only Pa-
tient 1 received CT examination. Seven patients had
agyria-pachygyria with neuroimaging grading 1-4 (Fig
1A), and three were grade 5-6 (Fig 1B). All of our grade
1-4 patients were classic type I lissencephaly. Six patients
have received chromosome studies, and only Patient 2 was
proven to have Miller-Dieker syndrome. Of the seven
patients with grade 1-4 agyria-pachygyria, five patients
had infantile spasms and had seizure onset at less than 1
year of age. In contrast, only Patient 10, with grade 6
agyria-pachygyria, developed seizures at 10 years of age.
Neurophysiologic Evaluations
Electroencephalography was performed in nine patients,
seven of whom were grade 1-4 and two of whom were
grade 5-6. Four patients had a neuroimaging grading of
1-4, and only one patient was a grade 6 with high-
amplitude fast activity (Fig 2). The frequencies of the fast
activities ranged from 8 to 16 Hz with an amplitude higher
than 50 ␮V. The high-voltage tracings were dominant at
areas corresponding to the pachygyria, as demonstrated in
Patient 10 (Fig 1B).
Eight of the nine patients had a normal brainstem
auditory-evoked potential, with the exception of Patient 7
who underwent prolonged ventilator usage because of
septic shock and cardiopulmonary dysfunction. Goggled
visual-evoked potential was performed in seven patients.
Three of these findings were abnormal, including Patient 9
who had congenital anomalies of optic nerve hypoplasia,
iris heterochromia, and agenesis of corpus callosum (Table
3). Five of the six patients with somatosensory-evoked
potential examinations had abnormal N1 cortical compo-
nents in somatosensory-evoked potentials (Table 4). Four
of these patients also had prolonged central conduction
times (Fig 3). The N1 abnormalities included poor wave-
forms or near absence of cortical components. In contrast,
latencies of the sensory-evoked potential component at the

Table 2. Clinical manifestations and neuroimaging findings in 11 patients with agyria-pachygyria

Seizure Imaging
Case Initial presentations Age Gender Patterns Imaging Findings Grading

1 Seizure 14 yr 8 mo F G-T-C Complete agyria Grade 1

2 SGA and ventriculomegaly 6 yr F IS Complete agyria Grade 1
3 SGA and DD 1 yr 2 mo F IS Complete agyria Grade 1
4 Seizure 9 yr 5 mo F IS Agyria-pachygyria Grade 2
5 Seizure 3 yr 6 mo M IS Agyria-pachygyria Grade 2
6 DD 3 yr 2 mo F IS Complete pachygyria Grade 4
7 Congenital anomaly with facial dysmorphysm 1 yr 10 mo F G-T-C Complete pachygyria Grade 4
8 DD 2 yr 6 mo M No seizure Focal pachygyria Grade 5
9 SGA and congenital anomaly 1 yr F No seizure Focal pachygyria Grade 5
10 Seizure 10 yr 8 mo M G-T-C Focal pachygyria Grade 6

Abbreviations:
DD ⫽ Developmental delay IS ⫽ Infantile spasms
G-T-C ⫽ Generalized tonic clonic seizure SGA ⫽ Small for gestational age

172 PEDIATRIC NEUROLOGY Vol. 27 No. 3

 Figure 2. Electroencephalogram of Patient 10, recorded at 10 years of
age, exhibits the dominant high-voltage tracings at bilateral frontal
areas corresponding to the pachygyric areas demonstrated in Figure 1B.

Clinical Outcome

In all, eight of 10 patients survived and had variable

Figure 1. (A) T2-weighted magnetic resonance imaging of Patient 2
demonstrates smooth brain surface, lack of insular operculation, thick-
ened cerebral cortex, and enlarged lateral ventricles (TR ⫽ 2,500 ms, TE
⫽ 90 ms). (B) T2-weighted magnetic resonance imaging of Patient 10
displays decreased sulcation and thickened cortex at bilateral middle and
superior frontal areas (TR ⫽ 5,552 ms, TE ⫽ 112 ms).
intellectual and motor retardation. Of the seven patients
with grade 1-4 agyria-pachygyria, two died because of
sepsis, and five patients had profound intellectual and
motor retardation with intractable seizures during the
9-96-month follow-up period (Table 5). The intellectual
and motor outcome of the children who were grade 5-6
was much better than that of children who were grade 1-4

Table 3. Summary of goggled visual-evoked potential findings in

seven patients with agyria-pachygyria

Imaging
neck point were normal in these patients. Of these five Case Grading Visual-Evoked Potential Findings
patients with abnormal somatosensory-evoked potentials,
four had grade 1-4 agyria-pachygyria, and one patient 2 Grade 1 Poor waveform of goggled visual-evoked potential
(Patient 10) was grade 6. Patient 10, who developed 3 Grade 1 Normal
5 Grade 2 Absence of goggled visual-evoked potential
seizures at 10 years of age, had only mild prolongation of 6 Grade 4 Normal
central conduction time at the left side. Somatosensory- 8 Grade 5 Normal
evoked potentials could not be elicited in Patient 5, who 9 Grade 5 Absence of goggled visual-evoked potential
10 Grade 6 Normal
had seizures at 3 months of age and died of sepsis.

Liang et al: Analysis of Agyria-Pachygyria Complex 173

Table 4. Summary of the somatosensory-evoked potentials to median nerve stimulation in six patients with agyria-pachygyria

Site of Stimulation
Imaging Age at Right (ms) Left (ms) Upper Limit of N1-N13
Case Grading Recording N13 N1 N1-N13 N13 N1 N1-N13 for the Age (ms)

2 Grade 1 5 yr 6 mo 9.8 29.4 19.6 10 30.6 20.6 7.91

3 Grade 1 10 mo 9 31.6 22.6 9 32.2 23.2 10.31
5 Grade 2 3 mo — — — — — —
6 Grade 4 2 yr 9 mo 8.6 31.8 23.2 8.2 34.4 26.2 9.7
8 Grade 5 4 yr 1 mo 8 15 7 8.2 15.2 7 7.91
10 Grade 6 10 yr 4 mo 11 17.6 6.6 10.8 18 7.2 6.92

Note: Bold characters are components with abnormalities.

—, Somatosensory-evoked potential cannot be elicited.

Abbreviation:
CCT ⫽ Central conduction time

agyria-pachygyria. In addition to neuroimaging, neuro-
physiologic studies provide another tool for predicting the
neurologic outcome of patients with agyria-pachygyria.
Although Patient 10 had grade 6 agyria-pachygyria, he had
mild abnormalities in somatosensory-evoked potentials
and developed seizures at 10 years of age. In contrast,
somatosensory-evoked potentials could not be elicited in
Patient 5, who had intractable seizure onset as early as 3
months of age and died of sepsis at 3 years and 6 months
of age. He had a worse clinical outcome than those

Figure 3. Somatosensory-evoked potentials of Patient 2, recorded at 5 years and 6 months of age, demonstrates prolonged and poor waveform of scalp
sensory-evoked potentials with normal latencies of cervical component.

174 PEDIATRIC NEUROLOGY Vol. 27 No. 3

Table 5. Correlation of imaging grading, age of seizure onset, and neurologic outcomes
Imaging Age of
Case Grading Seizure Onset
1 Grade 1 12 yr
2 Grade 1 8 mo
3 Grade 1 8 mo
4 Grade 2 2 mo
5 Grade 2 3 mo
6 Grade 4 9 mo
7 Grade 4 10 mo
8 Grade 5 No seizure
9 Grade 5 No seizure
10 Grade 5 10 yr
Abbreviations:
ADL ⫽ Activities of daily living MR ⫽ Mental retardation
patients with grade 1-2 agyria-pachygyria. Therefore the
somatosensory-evoked potential evaluation was supple-
mentary to neuroimaging in predicting the clinical out-
come of these patients.
Discussion
Agyria-pachygyria complex is a spectrum of cortical
malformations and a relatively severe form among migra-
tion disorders. Although the exact etiologies remain un-
clear, agyria-pachygyria complex may arise from genetic
factors or more often may be secondary to acquired injury
of the glial guides that direct the migration of neuroblasts
[10]. As reported in our present study, seizures and
delayed development are often the presenting symptoms in
patients with agyria-pachygyria. Seizure attacks, espe-
cially infantile spasms, are common early in life for those
with complete agyria. Although all of these patients were
retarded either in motor or intellectual function, the
clinical features and neurodevelopmental dysfunction
were diverse and suggest dependence on the extent,
distribution, and detailed structure of agyria-pachygyria
[10]. In previous studies, those patients with predomi-
nance of the pachygyria cortex had a lesser degree of
psychomotor disability [3]. However, in our studies the
severity of psychomotor retardation may not completely
correspond to the degree of image abnormalities and may
partially be related to the age at seizure onset. Patients
with agyria and those who had earlier onset of seizures had
the worse outcome in our study.
The advent of CT and MRI has made the diagnosis of
agyria-pachygyria during life possible in recent years. In
addition to neuroimaging, EEG abnormalities are also
helpful in the diagnosis of agyria-pachygyria [5-7,11].
These abnormalities include generalized high-amplitude
fast activity, high-amplitude sharp- and slow-wave com-
plexes, and a pattern consisting of bursts of sharp waves
alternating with periods of electrocerebral depression [5].
The most common form is the high-amplitude fast activity.
This unusual pattern of abnormally high-amplitude rhyth-
Duration of
Neurologic Outcome Follow-up (mo)
MR with extensive support to maintain basic ADL 32
Bedridden 72
Bedridden 9
Bedridden 96
Expired 42
MR with extensive support to maintain basic ADL 24
Expired 9
MR with limited support to maintain basic ADL 24
Severe psychomotor retardation 8
MR with intermittent support to maintain basic ADL 5
mic activity, which is abnormally rapid for age, has
diagnostic value for agyria-pachygyria [5,7]. As demon-
strated in Patient 10, there is a correlation between
high-voltage fast activities and the anatomic configura-
tions of agyria-pachygyria in neuroimaging. These high-
amplitude fast activities may be caused by the electric
activity of a relatively large part of the cortex picked up by
a scalp electrode or resulting from denervation supersen-
sitivity caused by deafferentation of cortical neurons [5,7].
Compared with the EEG findings in agyria-pachygyria,
the experience of the role of evoked potentials in evalu-
ating patients with agyria-pachygyria is limited [5,11-14].
Although delayed conduction through the brainstem path-
way in brainstem auditory-evoked potential studies has
been observed [12], our study did not reveal abnormalities
in brainstem auditory-evoked potentials, except for one
patient who required prolonged ventilator usage. Our
result was similar to that reported by Capua [13].
Previous studies of visual-evoked potential abnormali-
ties in agyria-pachygyria are varied [11-12]. Visual-
evoked potentials seem unable to identify or distinguish
specific supratentorial cerebral malformations, including
agyria-pachygyria [12]. Even in patients with holoprosen-
cephaly, callosal agenesis, and colpocephaly, visual-
evoked potentials could be normal [12]. In Coupland’s
study [12], abnormal flash visual-evoked potential wave-
forms and delayed conduction times were observed. How-
ever, normal flash visual-evoked potential results were
evident in Kriss’ study [11]. The normal visual-evoked
potential findings in Kriss’ study lead to the conclusion
that occipital development is better preserved than that of
the brainstem in agyria-pachygyria [11]. However, in our
present study visual-evoked potential abnormalities were
present in three of the seven patients through visual-
evoked potential evaluations, whereas only one of the nine
patients with brainstem auditory-evoked potential exami-
nations was abnormal. These results indicate occipital
development is also involved in agyria-pachygyria, but the
brainstem is much better preserved.
Liang et al: Analysis of Agyria-Pachygyria Complex 175
 Somatosensory-evoked potentials may assist in the local-
ization of brain dysfunction and may prove helpful in
formulating a prognostic statement in brain lesions caused by
malformation [15]. Reports on the diagnostic utility of
somatosensory-evoked potentials in agyria-pachygyria are
rare [5,13-14]. The primary somatosensory fields in agyria-
pachygyria patients were restricted to the centroparietal, or
occasionally the frontal areas. The absence of frontal P20 and
parietal N1 suggests a hypoplastic central sulcus or function-
ally undifferentiated subdivision of the somatosensory cortex
[14]. In our studies the configuration of the cortically gener-
ated complexes was abnormal; the latencies at the cortex
(N1) and the central conduction time (N1-N13) were pro-
longed compared with those of normal children. The poor
waveforms or absence of the cortical sensory-evoked poten-
tial components in our results suggests agyria-pachygyria to
be a disease limited to the supratentorial cortex. In the present
study, patients with grade 1-4 agyria-pachygyria or somato-
sensory-evoked potential abnormalities disclosed worse in-
tellectual and motor disability compared with those with
grade 5-6 or without somatosensory-evoked potential abnor-
malities. Therefore somatosensory-evoked potential exami-
nations can be considered a supplementary tool in predicting
the neurologic outcome of patients with agyria-pachygyria.
The timing of neuronal migration arrest in agyria-
pachygyria is at approximately 3-4 months of gestation
[3]. Associated brain abnormalities including enlarged
ventricles, heterotopias, and cerebellar or callosal agenesis
have been located in some agyria-pachygyria patients [3].
Normal brainstem-evoked potential, but abnormal cortical
sensory-evoked potential components, and prolonged cen-
tral conduction time in our patients reflect that agyria-
pachygyria is a supratentorial disease. This result may
presumably be because our patients did not have any
cerebellar and brainstem malformations, which are com-
mon in type II lissencephaly.
In conclusion, although agyria-pachygyria cannot be
diagnosed by using the findings in neurophysiologic stud-
ies, neurophysiologic studies, especially somatosensory-
evoked potentials, may be a supplemental tool in evaluat-
ing the neurologic prognosis of these patients. Combined
with neuroimaging grading, the abnormalities in neuro-
176 PEDIATRIC NEUROLOGY Vol. 27 No. 3
physiologic studies could serve as prognostic factors in
agyria-pachygyria.
References
[1] Dobyns WB, Reiner O, Carrozzo R, Ledbetter DH. Lissenceph-
aly: A human brain malformation associated with deletion of the LIS1
gene located at chromosome 17p13. JAMA 1993;270:2838-42.
[2] Ashwal S. Congenital structural defects. In: Swaiman KF,
Ashwal S, eds. Pediatric neurology, 3rd ed. St Louis: Mosby, 1999:234-
300.
[3] Kuzniecky RI. Magnetic resonance imaging in developmental
disorders of the cerebral cortex. Epilepsia 1994;35(Suppl 6):S44-56.
[4] Menkes JH, Sarnat HB. Neuroembryology, genetic program-
ming, and malformations of the nervous system. In: Menkes JH, Sarnat
HB, eds. Child Neurology, 6th ed. Philadelphia: Lippincott Williams &
Wilkins, 2000:277-400.
[5] de Rijk-van Andel JF, Arts WFM, Weerd AW. EEG and evoked
potentials in a series of 21 patients with lissencephaly type 1. Neurope-
diatrics 1992;23:4-9.
[6] Hakada S, Watanabe K, Hara K, Miyazaki S. The evolution of
electroencephalographic features in lissencephalic syndrome. Brain Dev
1979;1:277-83.
[7] Gastaut H, Pinsard N, Raybaud C, Aicardi J, Zifkin B. Lissen-
cephaly (agyria-pachygyria). Clinical findings and serial EEG studies.
Dev Med Child Neurol 1987;29:167-80.
[8] Andel JFR, Arts WFM, Barth PG, Loonen MCB. Diagnostic
features and clinical signs in 21 patients with lissencephaly type 1. Dev
Med Child Neurol 1990;32:707-17.
[9] Taylor MJ, Fagan ER. SEPs to median nerve stimulation:
Normative data for pediatrics. Electroencephalogr Clin Neurophysiol
1988;71:323-30.
[10] Aicardi J. The agyria-pachygyria complex: A spectrum of
cortical malformations. Brain Dev 1991;13:1-8.
[11] Kriss A, Boyd S, Thompson D, Harris CM. A study of EEG,
electroretinogram, visual evoked potential, and eye movements in clas-
sical lissencephaly. Dev Med Child Neurol 2000;42:48-52.
[12] Coupland SG, Sarnat HB. Visual and auditory evoked potential
correlates of cerebral malformations. Brain Dev 1990;12:466-72.
[13] Capua MD, Vigevano F, Wisniewski K. Somatosensory
evoked potentials in hemimegalencephaly and lissencephaly: Anatomo-
functional correlations. Brain Dev 1993;15:253-7.
[14] Saito Y, Adachi K, Nezu A, Deguchi K, Kimura S. Topograph-
ical features of the sensory-evoked responses in malformed brains.
Pediatr Nerurol 2000;23:243-9.
[15] Scher MS. Pediatric neurophysiologic evaluation. In: Swaiman
KF, Ashwal S, eds. Pediatric neurology, 3rd ed. St Louis: Mosby,
1999:142-81.