ORIGINAL ARTICLE

Development of the human fetal insular cortex: studyof the gyration from 13 to 28 gestational weeks

A. Afif Æ R. Bouvier Æ A. Buenerd Æ J. Trouillas ÆP. Mertens

Received: 5 April 2007 / Accepted: 14 September 2007 / Published online: 26 October 2007

� Springer-Verlag 2007

Abstract To describe the morphological stages of insular

sulci and gyri development we carried out a macroscopical

study on 21 human fetal brains, showing no anomalies,

from 13 to 28 gestational weeks (GWs). Particular focus

was given to morphological appearance during the devel-

opment of insular and periinsular structures, especially the

gyration and sulcation of the insula, central cerebral region

and opercula, as well as the vascularization of these

regions. The periinsular sulci and the central (insular and

cerebral) sulci were the first macroscopical structures

identified on the lateral surface of the human fetal cerebral

hemisphere with earlier development on the right hemi-

sphere. Here we describe five stages of insular gyral and

sulcal development closely related to gestational age: stage

1: appearance of the first sulcus at 13-17 GWs, stage 2:

development of the periinsular sulci at 18–19 GWs, stage

3: central sulci and opercularization of the insula at 20–22

GWs, stage 4: covering of the posterior insula at 24–26

GWs, stage 5: closure of the sylvian fissure at 27–28 GWs.

We provide evidence that cortical maturation (sulcation

and gyration) and vascularization of the lateral surface of

the brain starts with the insular region, suggesting that this

region is a central area of cortical development.

Keywords Insular cortex development � Morphology �Insular sulci � Insular gyri � Periinsular sulci

Introduction

The human insular lobe was initially described by the

anatomist Reil JC (Reil 1809) and named the Island of

Reil. Several later anatomical studies focused on this fifth

cerebral lobe and its development (Guldberg 1887; Cunn-

ingham 1891; Clark 1896; Retzius 1896). The first studies

on the cytoarchitectonical development of the human cor-

tex report that the insula is the first cortex to differentiate

and develop in the fetus beginning from 6 weeks of the

fetal life in a cortical region that will later become the

limen insula (Streeter 1912; Kodam 1926). At the first

stages of telencephalic development, the insula is localized

on the lateral surface of each hemisphere (Lockard 1948).

Nevertheless, its surface expansion is more limited than for

any other cerebral lobe. After becoming buried in the

depths of the sylvian fissure during fetal growth, the insula

is progressively covered by temporoparietal and then

frontal opercula (Dorovini-Zis and Dolman 1977; Feess-

Higgins and Larroche 1987). Anatomical data have shown

a left–right asymmetry of gyral and sulcal development,

with the precession of the right cerebral hemisphere gyri

and sulci (Cunningham 1891; Chi et al. 1977; Dorovini-Zis

and Dolman 1977).

The use of more recent techniques, including macro-

scopical, ultrasonographical, and magnetic resonance

A. Afif � P. Mertens (&)

Department of Anatomy, Lyon-1 University,

Inserm U 879, 8 avenue Rockefeller, Lyon 69003, France

e-mail: Patrick.Mertens@recherche.univ-lyon1.fr

A. Afif � P. Mertens

Department of Neurosurgery, Neurological Hospital,

Hospices civils de Lyon, Lyon 69003, France

e-mail: afif_acc@hotmail.com

R. Bouvier � A. Buenerd

Department of Pathology, Edouard Herriot Hospital,

Hospices civils de Lyon, Lyon 69003, France

J. Trouillas

Laboratory of Histology and Embryology,

Lyon-1 University, Inserm U 842,

Lyon 69003, France

123

Brain Struct Funct (2007) 212:335–346

DOI 10.1007/s00429-007-0161-1

imaging, has led to interesting findings concerning fetal

brain development. In all cases however, the structural

appearance of the insular cortex and its development have

rarely been investigated with no precise description of

insular sulcal development.

The insula constitutes part of the paralimbic structures.

The transitional insular cortex progressively develops from

an allo- to iso-type cortex. Three cytoarchitectonical areas

can be described in primates and man within the insular

cortex (Mesulam and Mufson 1985): (1) the more anterior

agranular area made up of three layers of cells; (2) the

dysgranular area located behind of the agranular area,

occupying the pericentral region; and (3) the posterior

granular area. The topographic distribution of efferent and

afferent projections between the insular cortex and other

cortical regions (Mesulam and Mufson 1982) are related to

different functional systems within the insula. The pos-

terodorsal part of the insula (the granular area) connects

with the retroinsular area, temporal cortex, supplementary

motor area, and both primary and secondary somatosensery

cortex, and specializes in auditory, somesthetic and skel-

etomotor functions. However, the anteroventral agranular

part connects with the entorhinal cortex, cingulated area and

periamygdaloid cortex, and specializes in olfactory, gusta-

tory and viscero-autonomic functions. Limits between these

three insular architectonic areas have never been identified

with respect to insular sulci and gyri organization.

There are a number of recent functional studies that

suggest an important role of the insular cortex especially in

language production and grammatical processing (McCar-

thy et al. 1993; Isnard et al. 2004; Friederici et al. 2006),

pain processing (Peyron et al. 2004; Brooks et al. 2005;

Schreckenberger et al. 2005), viscerosensitive manifesta-

tions (Ostrowsky et al. 2000) and auditory processing

(Lewis et al. 2000; Bamiou et al. 2003).

This study aimed to describe the morphological stages

of insular sulcal and gyral formation by direct anatomical

observation of the human fetal brain between 13 and 28

gestational weeks (GWs). The present study expected to

demonstrate an earlier development of insular sulci than

had previously been suggested by modern imaging tech-

niques (Cohen-Sacher et al. 2006; Ruiz et al. 2006; Garel

et al. 2001). We studied the dynamics of development of

the insular lobe in relation to the opercularization process

and to the development of its arterial supply. Special

interest was devoted to the relationship between the

development of the central cerebral sulcus (CCS) and the

central insular sulcus (CIS).

As embryology enables a better understanding of gross

anatomy, this study also aimed to provide a better com-

prehension of insular morphometry and development. The

results of this study could be useful in: (1) estimating fetal

development at different stages of pregnancy and (2)

describing the sulcal pattern of the insula, thereby helping

to understand the cytoarchitectonical and functional orga-

nization of the gyri and sulci.

Materials and methods

Macroscopical examinations were performed under bin-

ocular vision with low magnification (94) and each

specimen was photographed. Twenty-one human fetal

brains were studied: 42 (30 males and 12 females) for-

malin-fixed hemispheres at different GWs (2–8

hemispheres per GW) ranging from 13 to 28 GWs, except

the 23rd GW. GW in this study was estimated according to

the mother’s last menstrual period and later corrected by

Echography data. All fetal brains came from spontaneous

miscarriage or rarely from medical interruption of preg-

nancy. Fetuses with cerebral or chromosomic anomalies

were excluded. In either case parental consent was obtained

for the study of the fetus.

The first appearance of each of the periinsular and the

insular sulcus was macroscopically identified, as well as

sulci in the central cerebral region. The relationship between

both central (cerebral and insular) sulci was observed. The

dynamics of opercular development was described in the

different stages. The following measurements were taken

(Fig. 1): (1) the length of each periinsular sulcus: the ante-

rior periinsular sulcus separating the insula from the fronto-

orbital operculum, the superior periinsular sulcus separating

the insula from the frontoparietal operculum, and the pos-

teroinferior periinsular sulcus separating the insula from the

Heschl gyrus and the temporal operculum. The length of the

CIS separating the insula in two parts anteroinferior to

posterosuperior, was also studied, (2) the distance between

the projections of the superior extremity of CIS and the

inferior extremity of CCS, (3) the two distances (A and B)

between the superior extremity of CIS and both insular

angles: anterior (the angle between the superior periinsular

sulcus and the anterior periinsular sulcus), and posterior (the

angle between the superior periinsular sulcus and the pos-

teroinferior periinsular sulcus).

We described the chronological development of: (1) the

lateral aspect of the brain hemisphere and the periinsular

sulci, (2) the temporo-parieto-frontal opercular develop-

ment, (3) the insular gyration and sulcation, (4) the sulci in

the central cerebral region (Rolandic), and (5) the devel-

opment of branches of the middle cerebral artery (MCA).

Results

The insular region and the Sylvian fossa were first mac-

roscopically identifiable at 18 GWs by the appearance of

336 Brain Struct Funct (2007) 212:335–346

123

periinsular shallow depressions or periinsular sulci, and the

development of adjacent structures. This development was

always earlier on the right side.

We recorded the mean increase in length of periinsular

sulci and CIS between stage 2 (18–19 GWs) and stage 5

(27–28 GWs) (Table 1): At stage 2, the mean length of the

anterior periinsular sulcus was 8.35 mm (95% confidence

intervals (CI): 7.74–8.96) becoming 14.8 mm (95% CI:

12.10–17.49) at stage 5, with a mean increase equal to

77.24%. The mean length of the superior periinsular sulcus

increased from 16.1 mm (95% CI: 15.27–16.92) to

27.2 mm (95% CI: 22.28–32.12) with a mean increase

equal to 68.9%. The mean length of postero-inferior peri-

insular sulcus increased from 13.3 mm (95% CI: 12.36–

14.24) to 21.1 mm (95% CI: 18.52–23.67) with a mean

increase equal to 58.64%.

The mean length of CIS increased from 10.16 mm (95%

CI: 9.31–11.02) to 17.6 mm (95% CI: 15.71–19.48) with a

mean increase equal to 60.66%.

From 18 and 28 GWs, the mean distance between the

superior extremity of CIS and the anterior insular angle or

the posterior insular angle increased from 11.75 mm (95%

CI: 11.31–12.18) to 19.6 mm (95% CI: 16.70–22.49) with

a mean increase equal to 66.80%, and from 4.91 mm (95%

CI: 4.07–5.75) to 7.6 mm (95% CI: 5.06–10.13) with a

mean increase equal to 54.78%, respectively.

These data show a greater range of development in the

anterior than in the posterior part of the insula. Data

detailing the different stages of development is presented in

Table 1.

Data obtained from study concerning the development

of the insula, the opercula and the central cerebral region

(Table 2) allowed us to identify five stages in the devel-

opment of insula and periinsular structures. Stages of this

original classification are closely related to gestational age

(Table 3).

Stage 1: appearance of the first sulcus (13–17 GWs)

The surface of the brain is smooth (Fig. 2a, c). It is not

possible at this stage to macroscopically differentiate the

insular cortex from other cortical structures. At the 13th

GW, a linear furrow (groove), almost perpendicular to the

ventral border of the hemisphere is visible on the lateral

surface at the level of the pterion (Fig. 2a). This linear

furrow can be considered as the limit between the temporal

lobe below, and the insula, frontal and parietal lobes above.

At 15 GWs this limit becomes oblique from anteroinferior

to posterosuperior (Fig. 2b) due to the important develop-

ment of the anterosuperior part of the brain hemisphere.

This first groove, visible on the lateral surface of the

hemisphere at 13 GWs, will later become the posteroin-

ferior periinsular sulcus. It is important to note that all later

development of the insular cortex is above of this groove

(Fig. 2b, d). At 16–17 GWs the inferior and posterosupe-

rior limits of the insula and sylvian fossa are

macroscopically identifiable above of the temporal lobe

(Fig. 2d). The MCA divides into two branches (superior

and inferior), with the predominant superior branche sup-

plying the insular cortex. The MCA and its proximal M1

and M2 branches are present as early as in the 13th GW

(Fig. 2a), at which stage, they are more developed on the

right side than on the left side.

Fig. 1 Insular schematic

showing the different gyri and

sulci. Gyri: I Anterior short

gyrus; II middle short gyrus; IIIprecentral gyrus; IV postcentral

gyrus; V posterior long gyrus;

VI insular pole. Sulci: APSanterior periinsular sulcus; SPSsuperior periinsular sulcus;

PIPS posteroinferior periinsular

sulcus; 1 Anterior insular

sulcus; 2 precentral sulcus; 3central insular sulcus (CIS); 4postcentral sulcus; A distance

between the anterior insular

angle and the superior extremity

of the CIS; B distance between

the posterosuperior insular

angle and the superior extremity

of the CIS

Brain Struct Funct (2007) 212:335–346 337

123

Stage 2: development of the periinsular sulci (18–19

GWs)

The cerebral cortex is still smooth. At this stage, it is

possible to identify the insular region due to an important

development in the surrounding cortex allowing periinsular

sulci formation (Fig. 3). Both the posteroinferior periin-

sular sulcus, taking the place of the linear furrow seen in

stage 1, and the superior periinsular sulcus are well dis-

tinguished shallow grooves. The APS can only be

distinguished in the superior half, as the smaller sulcus

between periinsular sulci, whereas the superior periinsular

Table 1 Mean lengths with standard deviation and confidence intervals of the insular and periinsular sulci from 18 to 28 gestational weeks

(GWs)

Stage Column title APS SPS PIPS CIS A B

Stage 2 (18–19 GWs) Mean length (mm) 8.35 16.1 13.3 10.167 11.75 4.91

Standard deviation (SD) 0.85 1.15 1.31 0.81 0.41 0.80

Sample size (N) 10 10 10 6 6 6

Confidence intervals

Lower 95% conf. limit 7.74 15.27 12.35 9.31 11.31 4.07

Upper 95% conf. limit 8.95 16.92 14.24 11.02 12.18 5.75

Stage 3 (20–22 GWs) Mean length (mm) 10.09 18.86 15.45 10.59 12.86 6

Standard deviation (SD) 0.91 1.58 1.23 0.66 1.20 0.70

Sample size (N) 11 11 11 11 11 11

Confidence intervals

Lower 95% conf. limit 9.47 17.80 14.62 10.14 12.05 5.52

Upper 95% conf. limit 10.70 19.92 16.28 11.03 13.67 6.47

Stage 4 (24–26 GWs) Mean length (mm) 12.16 24.5 20 13.83 16.66 7.83

Standard deviation (SD) 0.40 1.04 0.89 0.75 1.03 0.40

Sample size (N) 6 6 6 6 6 6

Confidence intervals

Lower 95% conf. Limit 11.73 23.39 19.06 13.04 15.58 7.40

Upper 95% conf. limit 12.59 25.60 20.93 14.62 17.75 8.26

Stage 5 (27–28 GWs) Mean length (mm) 14.8 27.2 21.1 17.6 19.6 7.6

Standard deviation (SD) 2.16 3.96 2.07 1.51 2.32 2.04

Sample size (N) 5 5 5 5 5 5

Confidence intervals

Lower 95% conf. limit 12.10 22.28 18.52 15.71 16.70 5.06

Upper 95% conf. limit 17.49 32.11 23.67 19.48 22.49 10.13

APS anterior periinsular sulcus, mm millimeter, SPS superior periinsular sulcus, PIPS postero-inferior periinsular sulcus, CIS central insular

sulcus, A the distance between the anterior insular angle and the superior extremity of the CIS, B the distance between the posterior insular angle

and the superior extremity of the CIS

Table 2 First observation of

insular and central cerebral sulci

and gyri in the fetus

GW gestational week, Ins.

insular, inf. inferior, Ant.anterior, post. posterior

Adjacent ins. structure GW Insular structures GW

Postero-inf. periinsular sulcus 13 Insula 18

Periinsular sulci 18 Central insular sulcus 18–21

Central cerebral sulcus 18–21 Ant. insula and post. insula 18–21

Sylvian fossa 18 Pre-central Ins. sulcus 22

Opercula 19 Post-central Ins. sulcus 24

Pre-central cerebral sulcus 22 Pre-central Ins. gyrus 22

Post-central cerebral sulcus 24 Post-central Ins. gyrus 24

Pre-central cerebral gyrus 22 Ant. insular sulcus 27–28

Post-central cerebral gyrus 24 Middle Ins. gyrus 27–28

Closure of post. sylvian fossa 24 Hidden post. insula 24

Closure of sylvian fossa 28 Hidden total of insula 27

338 Brain Struct Funct (2007) 212:335–346

123

sulcus is longer and consistently bigger on the left side in

both sexes. At this stage, opercularization has not yet

begun. Anterior and posterior insular regions are separated

by a very fine notch at the surface that will become the CIS.

This notch, better identified in its inferior part (Fig. 3b, d),

crosses the insula in an oblique axis, anteroinferior to

posterosuperior, to join the superior periinsular sulcus.

Dorsally the CCS becomes macroscopically identifiable,

especially in its inferior part, limiting the posterior part of

the frontal cortex (Fig. 3b, d). The inferior extremity of

CCS is localized 3–4 mm anterior to the superior extremity

of the CIS. Both CIS and CCS are consistently better

identified on the right side.

Middle cerebral artery branch development over the

surface of the insula is more important compared with other

regions of the lateral cerebral cortex. Anastomosis between

the MCA and anterior cerebral artery can be observed in the

parasagittal region of the lateral cortex and is more devel-

oped within the peri-central (peri-Rolandic) branches. The

diameter of the vessels associated with the central insular

notch progressively increases, with more rapidity on the

right side, representing the rise of the vascular supply to the

future pericentral cerebral region (Fig. 3a, c).

Stage 3: central sulci and opercularization of insula

(20–22 GWs)

The sylvian region is like a fossa that becomes narrower,

especially in its posterosuperior part, due to greater

Table 3 filled triangle: The identification of insular and periinsular sulci in relationships with the five stages

Insular structures Stages

1 2 3 4 5

(13–17 GWs) (18–19 GWs) (20–22 GWs) (24–26 GWs) (27–28 GWs)

PIPS filled triangle

SPS and APS filled triangle

CIS filled triangle

Pre-CIS filled triangle

Post-CIS filled triangle

AIS filled triangle

PIPS postero-inferior periinsular sulcus, SPS superior periinsular sulcus, APS anterior periinsular sulcus, CIS central insular sulcus, Pre-CISprecentral insular sulcus, Post-CIS postcentral insular sulcus, AIS anterior insular sulcus, GWs gestational weeks

Fig. 2 Photographs of the

lateral view of the fetal brain at

13, 15, 16 and 17 GWs showing

the posteroinferior periinsular

sulcus (black arrows) in (a), (b)

and (c); 1 posteroinferior

periinsular sulcus in (d); 2posterior part of the superior

periinsular sulcus (white arrow)

Brain Struct Funct (2007) 212:335–346 339

123

development of the temporal and parietal opercula com-

pared with the frontal opercula (Fig. 4). Moreover,

opercularization takes place more rapidly on the right side.

The periinsular sulci (especially posteroinferior periinsular

sulcus and superior periinsular sulcus) and the majority of

the posterior insula are, at this stage, hidden by operculi

(Fig. 4c, d). The inferior part of the anterior periinsular

sulcus becomes macroscopically identifiable. Both CCS

and CIS become more visible ‘‘shallow grooves’’. Projec-

tion of the lower extremity of the CCS is localized 1–2 mm

anterior to the superior extremity of CIS. At this stage of

development, it becomes possible to identify the notches of

precentral sulci (insular and cerebral) by the 20th GW, and

postcentral sulci (insular and cerebral) by the 22nd GW. It

is therefore possible to distinguish the pre- and postcentral

(insular and cerebral) gyri (Fig. 7).

Fig. 3 Photographs of the

lateral view of the fetal brain aat 18 GWs showing the middle

cerebral artery and its branches

‘‘M2, M3, M4 and M5’’; A the

artery associated with the

central insular notch

representing the future

pericentral region arteries (Barrows). b At 18 GWs showing

the insular region (blackarrows) due to development of

adjacent insular structures; 1central cerebral sulcus; 2 central

insular sulcus. c At 19 GWs

showing the vascularization of

insular region; A central insular

artery and the central cerebral

‘‘Rolandic’’ artery (B). d At 19

GWs showing the central

insular sulcus (2 arrowheads);

and the central cerebral sulcus

(2 arrows). Both sulci are better

identified in the inferior part

Fig. 4 Photographs of the

lateral view of the fetal brain aat 21 GWs showing the sylvian

fossa which has become narrow

at its posterior part following

temporal and parietal opercula

development (arrows). b At 21

GWs showing the central

cerebral sulcus (1) and the

central insular sulcus (2). c At

22 GWs showing the artery

which corresponds to the central

insular sulcus (‘a’ arrow) giving

the central cerebral artery (‘b’

arrow). A large area of the

posterior insula is hidden by the

temporal and parietal opercula

(white arrows). d At 22 GWs

showing the inferior extremity

of the central cerebral sulcus

(1), the superior extremity of the

central insular sulcus (2) and the

precentral cerebral sulcus (3)

340 Brain Struct Funct (2007) 212:335–346

123

The diameter of MCA and its branches M2 and M3 are

increased, in particular the vessels which correspond to the

CIS. The superior division and its branches supply the

anterior part of the insula. The central cerebral artery arises

from the central insular artery.

Stage 4: covering of the posterior insula (24–26 GWs)

The surrounding cortex overlaps the posterior part of the

sylvian region at this stage, with only partial covering of

the anterior tip of the anterior insula and the anterior

periinsular sulcus (Fig. 5c, d). The superior part of the

anterior insula gradually becomes hidden along with the

development of the frontoorbital opercula. The lower

extremity of CCS is orientated in the same axis to the

superior extremity of the CIS (Fig. 5b). The superior

extremity of CIS separates the superior periinsular sulcus

into two interior thirds and one posterior third (Table 1). At

this stage, the postCIS can clearly be seen, and the insular

gyri (posterior, postcentral, and precentral) are macro-

scopically identifiable (Fig. 5c, d).

The distal branches (M4 and M5) of the MCA and their

anastomoses with the branches of the anterior cerebral

artery are more easily identifiable. The central cerebral

artery arises from the central insular artery (collateral of

the superior division of the MCA) (Fig. 5a).

Stage 5: closure of sylvian fissure (27–28 GWs)

At 27 GWs, the sylvian region has become a real fissure

that is already closed in the right side with complete cov-

erage of the right insula by the operculum (temporo-

parieto-frontal) (Fig. 6b). On the left side, the anterior tip

of the anterior insula is only partially covered (Fig. 6a)

with complete coverage only observed at 28 GWs, when

the anterior insular sulcus also becomes visible and all

insular sulci and gyri are in place. The lateral shape of the

insula becomes similar to its adult trapezoid-like form. At

this stage, the insula is separated into two parts: (1) anterior

consisting of three short gyri; and (2) posterior consisting

of two long gyri. The superior periinsular sulcus is con-

sistently the longest, predominating in the left side

(Table 1). CIS is the deeper of the insular sulci with CCS

deeper than the pre- and post-central cerebral sulci. At this

stage, the vessels are organized as in the final vasculari-

zation pattern. No arteries run along the superior

periinsular sulcal axis. At this sulcus, the insular arteries

become the opercular and supraopercular arteries.

Fig. 5 Photographs of the lateral view of the fetal brain a at 25 GWs

showing: 1 central cerebral sulcus; 2 precentral cerebral sulcus; 3postcentral cerebral sulcus; A central insular artery; and B central

cerebral artery. b At 25 GWs without the pia mater showing that the

projection of the inferior extremity of central cerebral sulcus

corresponds to the superior extremity of the central insular sulcus. 1

The inferior extremity of the central cerebral sulcus; a the superior

extremity of the central insular sulcus; b precentral insular sulcus; and

c postcentral insular sulcus. c and d at 26 GWs showing that all the

posterior insula is now hidden by the temporal and parietal opercula

(arrowheads); 1 central cerebral sulcus; 2 precentral cerebral sulcus;

3 postcentral cerebral sulcus

Brain Struct Funct (2007) 212:335–346 341

123

Finally, we observed a stable structural organization of

the fetal hemispheres. Only four hemispheres (9.5%)

from two fetuses presented some variability concerning

the CIS, as in these cases this sulcus was not present as

early as observed at 18th GWs for all the other cases.

There are no variations detected concerning other insular

and periinsular sulci in the fetal population observed in

this study.

Discussion

The main findings here concern the development of insular

sulci and gyri, and the formation of the opercular and

central cerebral regions in relation to gestational age. In

addition, we suggest a chronological classification of

insular sulcation and gyration in the human fetus.

Insular cortex and the opercularization

According to Streeter (1912), the insular cortex is the first

cortex to undergo development and differentiation in the

fetus. The present study demonstrates that the formation of

insular cortex begins by its inferior region that will later

become the limen insula. This cortical development pro-

gresses in parallel with that of the MCA and its branches in

a superior and posterosuperior axis. In this study, consistent

with findings of Cunningham (1891) and Chi et al. (1977),

the first description of all periinsular sulci was at 18 GWs

(Table 2). Discrepancies are however found with MR

imaging and ultrasound studies, where periinsular sulci

formation was observed 4 and 6 weeks later respectively

(Garel et al. 2001; Govaert et al. 2004), (Table 4).

Streeter (1912) and Retzius (1896) observed the first

appearance of the sylvian fissure at the third and fourth

month respectively, whereas Chi et al. (1977) reported this

fissure at 14 GWs. Our study, observing original data,

indicates that the earlier linear furrow on the surface of the

lateral hemispheric region at 13 GWs will become the

posteroinferior periinsular sulcus at 16 to 17 GWs (Fig. 2

b, d) and not the sylvian fissure as previously suggested.

The sylvian fossa appears parallel to the periinsular sulci

formation, beginning at 16–17 GWs for its posterior part,

forming a margined indentation at 18 GWs.

Consistent with the findings of Cunningham (1891),

Retzius (1896) and Chi et al. (1977), we found that oper-

cularization seen at 20 GWs (Table 2) is the result of

important development of the surrounding insular cortex.

Moreover, opercularization begins at the posterior half then

progresses to the anterior half of the superior edge of the

temporal lobe and the anterior part of the inferior edge of

the parietal lobe. Secondarily, this development involves

the inferior edge of the frontal lobe and finally, the pos-

terior edge of the orbital cortex (stage 4). Progressive

development over the entire surface of the insular cortex

(stage 5) encloses the sylvian fissure along a posterosupe-

rior to anteroinferior axis. Sonographic and MR imaging

studies report the first appearance of the opercularization

3–5 weeks later than our data show here (Monteagudo and

Timor-Tritsch 1997; Govaert et al. 2004; Garel et al. 2001;

Toi et al. 2004) (Table 4). This is most probably due to two

reasons: one reason for this may be the limited imaging

resolution of these techniques and an inability to detect

small range modifications of the cortical surface. A second

explanation could be the difficulty in imaging the con-

vexity and the surface of the lateral cerebral hemisphere

using these techniques.

Fig. 6 a Photograph of the lateral view of the fetal left cerebral

hemisphere at 27 GWs showing that the anterior tip of the anterior

insula is only partially covered by the edges of the lateral sylvian

fissure (arrows); b Photograph of the lateral view of the fetal right

cerebral hemisphere at 27 GWs showing that the insula is completely

hidden due to total closure of the lateral sylvian fissure (arrows); 1central cerebral sulcus; 2 precentral cerebral sulcus; 3 postcentral

cerebral sulcus

342 Brain Struct Funct (2007) 212:335–346

123

Insular and central cerebral region sulcation and

gyration

Central sulci

In agreement with studies by Cunningham (1891) and

Retzius (1896), we first observed both CCS and CIS at 18

GWs (stage 2), with the right side appearing before the left.

In the present study, the initial formation of the CCS was

identified in the inferior part (Fig. 7) and not in the para-

sagittal part as suggested by Chi et al. (1977). No data

concerning the maturation of CIS exists from imaging

technique studies, which report the appearance of CCS 6–8

weeks later (Cohen-Sacher et al. 2006; Ruiz et al. 2006;

Garel et al. 2001) (Table 4).

The CCS and the CIS were found to develop inde-

pendently suggesting that the CCS is not strictly

continuous with the CIS. Original data from our study

show that at early development stages, the inferior

extremity of CCS is located anterior to the superior

extremity of CIS. The two extremities correspond at stage

4. This evolution may be due to two reasons: (1) the

posterior displacement of CCS, especially in its inferior

part, related to the development of the frontal lobe (2) the

anterior displacement of the superior portion of CIS, due

to postero-anterior growth of the upper posterior area of

the insular cortex. The CIS, straight in the early stages 2

and 3, becomes curved in its superior part at stage 4 of

development.

Pre- and postcentral sulci

Consistent with a study by Cunningham (1891), we

observed the precentral (insular and cerebral) sulci at 20–

22 GWs, and the postcentral (insular and cerebral) sulci at

22–24 GWs. Chi et al. (1977) observed the insular sulci

later at 32–35 GWs. To our knowledge no radiological

studies have reported the first appearance of the insular

sulci. Under MRI, the pre- and postcentral cerebral sulci

appeared only 6–7 weeks later (Garel et al. 2001, 2003)

(see Table 4). According to a geometric study by Toro and

Table 4 Chronology of sulcation and gyration in the present study in comparison with anatomical, RM imaging, and ultrasound studies

Structure Anatomical studies (GWs) RMI studies (GWs) Ultrasound studies (GWs) Present study (GWs)

Periinsular sulci 16–19 (1) After 17 (5) 17–18

CCS 20–23 (1) 24–25 (2) 26 (3) 18–19

Sylvian fissure 14 (1) 19 (6) 18 (4) 18

Opercula 16–19 (1) 22–23 (2) After 17 (5) 20

Pre-CCS 24 (1) 26 (2) 27 (3) 22

Post-CCS 25 (1) 27 (2) 28 (3) 24

Pre-central cerebral Gy. 24 (1) 26 (2) 27 (3) 22

Post-central cerebral Gy. 25 (1) 27 (2) 28 (3) 24

Closure post. part of Syl. 26–30 (1) After 23 (2) 25 (3) 24

Closure total of Syl. After 30 (1) 34 (3) 28

Insula 18 (1) 22–23 (2) After 17 (5) 18

CIS 18–19

Ant. insula and post. Ins. 18–9

Pre-CIS 5.5–6 GMs (7) 22

Post-CIS 5.5–6 GMs (7) 24

AIS 27–28

Pre-central Ins. Gy. 22

Post-central Ins. Gy. 24

Middle Ins. Gy. 27–28

Hidden post. Ins. 26–30 (1) After 23 (2) 25 (3) 24

Hidden total of Ins. 32 (3) 27–28

Total Ins. sulci and gyri 32–35 (1) 33–34 (2) 33–34 (3) 28

GWs gestational weeks, GMs gestational months, (1) Chi, (2) Garel, (3) Ruiz, (4) Monteagudo, (5) Toi, (6) Abe, (7) Cunningham, CCS central

cerebral sulcus, Pre-CCS pre-central cerebral sulcus, Post-CCS post-central cerebral sulcus, Gy. gyrus, Syl. sylvian, Ant. anterior, Post posterior,

CIS central insular sulcus, Pre-CIS precentral insular sulcus, Post-CIS postcentral insular sulcus, AIS anterior insular sulcus, Ins. insular

Brain Struct Funct (2007) 212:335–346 343

123

Burnod (2003) using anatomical MR images, the primary

sulci (CCS and CIS) were deeper than adjacent secondary

pre- and postcentral sulci.

Discrepancies are found to exist between our study and

the literature concerning the appearance of the sulci. These

discrepancies may be due to: (1) different criteria for the

gestational age for instance where post-conception bleed-

ing cannot be excluded when considering the data obtained

by Chi et al. (1977) and McArdle et al. (1987); (2) neu-

ropathologic studies not focused on the beginning of

insular sulci formation and setting a limit to observe only

the presence of all insular sulci at later stages of develop-

ment Chi et al. (1977); (3) the difficulties in imaging the

convexity and the lateral surface of the brain hemisphere;

or (4) difficulties in observing the insular sulcation by

imaging techniques, either at the beginning of sulci for-

mation or after sylvian fissure closing.

Vascularization

The MCA and its branches (M2, M3 and M4) are distin-

guishable from the 13th GW (stage 1), first appearing in the

right side, and supply the insula and adjacent structures.

The insular arteries come from both superior and inferior

divisions of the MCA, predominating the precentral and

postcentral part of the insula respectively. These insular

arteries cross the surface of the insula and superior peri-

insular sulcus towards the lateral (frontoparietal) cortex.

The terminal branches of MCA anastomose with terminal

branches of the anterior cerebral artery at the upper portion

of the lateral cortex (parasagittal) starting at 18 GWs. The

central insular artery has the larger diameter of all other

insular branches and was consistently prolongated by the

central cerebral (Rolandic) artery in all 42 cases.

The MCA is the first artery to progress on the lateral

surface of the telencephalon suggesting that the early

development of insular gyration could be in parallel to this

vascular progression over its surface.

The anterior part of the insular cortex, and in particular

the anterior and middle gyri, is exclusively supplied by the

superior branch of the MCA. The central and posterior

insular regions are supplied by both superior and inferior

MCA branches. Moreover, we have observed that the

superficial cortical anastomosis system between the MCA

and the anterior cerebral artery is more developed within

the peri-central (peri-Rolandic) branches arising from the

arteries of the central insular region. The distribution of the

insular supply and the cortical anastomosis system, both

more developed in the central and the posterior parts of the

insula may account for the lesser involvement of these

parts in vascular infarction than the anterior part in adult

insular stroke.

Chronological classification of macroscopical

development

We propose an original 5-stage macroscopic classification

of maturation in the insular and central cerebral regions of

normal cerebral fetuses from 13 to 28 GWs. This classifi-

cation organizes the sulcal and gyral formation of these

regions with respect to GW. It is the first gestational age-

dependent classification based on macroscopical data taken

from the human fetal brain. Stages are based on first ma-

croscopical sightings of sulci and gyri in insular and central

cerebral regions of the lateral cortex. This classification

reports the grade and the mechanism of insular, operculum

and central cerebral development.

Previous suggestions have been made to classify human

fetal cerebral maturation based on MR imaging in the

second and third trimesters ‘‘trimenon’’ of pregnancy.

Fig. 7 Photograph of the lateral view of the fetal cerebral hemisphere

at 22 GWs. a Complete but without pia mater and b without the

opercula, showing that both CIS and CCS (arrows) are better

identified in the inferior part

344 Brain Struct Funct (2007) 212:335–346

123

McArdle et al. (1987) proposed a classification of five

stages. This classification focuses on frontal and occipital

sulcation and gyration, observing the folds of the insular

cortex at the stage 4 only (37–39 GWs). The development

of the cerebral cortex was later classified into eight stages

using MR images in the Ac-Pc plane, with special attention

to the frontal and temporal lobes (Abe et al. 2003).

Comparing previous classifications using MR images

with the classification proposed in this study shows that

differences found could explain the discrepancies seen

between this and previous studies. Firstly, neither of the

MRI classifications observed complete development of the

insular cortex, and sulci appearance was described 4–8

weeks later than first seen in this study. The reason could

be that the MRI studies used thicker slices (6–8 mm slice

thickness). Secondly, the MRI studies are not tangential to

the convexity of the surface of the lateral cerebral cortex

and it is difficult to obtain true axial, sagittal, and coronal

planes, so they are not able to detect small ranges (the

linear furrows) of cortical surface modification.

Conclusion

To our knowledge this is the first study focusing on the

morphological and chronological development of the

human insular cortex and adjacent structures. We have

described an earlier appearance of insular and periinsular

sulci than previously described by neuroimaging studies.

This study suggests that the insular lobe is the first

cerebral lobe to be identified by macroscopic techniques

with earlier right side development. Periinsular sulci, CCS

and CIS are the first sulci to develop in the lateral surface

of the human fetal cerebral hemisphere. We have proposed

a five stage classification describing the precise chronology

of fetal insular cortex gyration in relation to fetal gesta-

tional age.

Acknowledgment We thank Ms Emily Witty and Mr Michel

Magnin for editorial review of the text.

References

Abe S, Takagi K, Yamamoto T, Okuhata Y, Kato T (2003)

Assessment of cortical gyrus and sulcus formation using MR

images in normal fetuses. Prenat Diagn 23:225–231

Bamiou DE, Musiek FE, Luxon LM (2003) The insula (Island of Reil)

and its role in auditory processing literature review. Brain Res Rev

Brooks JCW, Zambreanu L, Godinez A, Craig AD, Tracey I (2005)

Somatotopic organization of the human insula to painful heat

studied with high resolution functional imaging. Neuroimage

27:201–209

Chi JG, Dooling EC, Gilles FH (1977) Gyral development of the

human brain. Ann Neurol 1:86–93

Clark TE (1896) The comparative anatomy of the insula. J Comp

Neurol 6:59–100

Cohen-Sacher B, Lerman-Sagie T, Lev D, Malinger G (2006)

Sonographic developmental milestones of the fetal cerebral

cortex: a longitudinal study. Ultrasound Obstet Gynecol 27:494–

502

Cunningham DJ (1891) The development of the gyri and sulci on the

surface of the island of Reil of the brain. J Anat Physiol 25:338–

348

Dorovini-Zis K, Dolman CL (1977) Gestational development of

brain. Arch Pathol Lab Med 101:192–195

Feess-Higgins A, Larroche J-C (1987) Le developpement du cerveau

foetal humain. Atlas Anatomique. Masson, Paris (In French)

Friederici AD, Bahlmann J, Heim S, Schubotz RI, Anwander A

(2006) The brain differentiates human and non-human gram-

mars: functional localization and structural connectivity. Proc

Natl Acad Sci USA 14 103(7):2458–2463

Garel C, Chantrel E, Brisse H, Elmaleh M, Luton D, Oury JF, Sebag

G, Hassan M (2001) Fetal cerebral cortex: normal gestational

landmarks identified using prenatal MR imaging. AJNR Am J

Neuroradiol 22:184–189

Garel C, Chantrel E, Elmaleh M, Brisse H, Sebag G (2003) Fetal

MRI: normal gestational landmarks for cerebral biometry,

gyration and myelination. Childs Nerv Syst 19:422–425

Govaert P, Swarte R, De Vos A, Lequin M (2004) Sonographic

appearance of the normal and abnormal insula of Reil. Dev Med

Child Neurol 46(9):610–616

Guldberg A (1887) —Zur morphologie der insula Reillis. Anat Anz

2:659–665

Isnard J, Guenot M, Sindou M, Mauguiere F (2004) Clinical

manifestations of insular lobe seizures: a stereo-electroenceph-

alographic study. Epilepsia 45(9):1079–1090

Kodam S (1926) Uber die sogenannten Basalganglien, Morphoge-

netische und pathologisch-anatomische Untersuchunger.

Schweiz Arch Neurol Psychiatr 18:179–246

Lewis JM, Beauchamp MS, De Yoe EA (2000) A comparison of

visual and auditory motion processing in human cerebral cortex.

Cereb Cortex 10(9):873–888

Lockard I (1948) Certain developmental relations and fiber connec-

tions of the triangular gyrus in primates. J Comp Neurol 89:349–

386

McArdle CB, Joan Richardson C, Nicholas DA, Mirfakhraee M,

Keith Hayden C, Amparo EG (1987) Developmental features of

the neonatal brain: MR imaging. Part I. gray white matter

differentiation and myelination. Radiology 162:223–229

McCarthy G, Blamier AM, Rothman DL, Gruelter R, Shulman RG

(1993) Echo-planar magnetic resonance imaging studies of

frontal cortex activation during word generation in humans. Proc

Natl Acad Sci USA 90:4952–4956

Mesulam MM, Mufson EJ (1985) The Insula of Reil in man and

monkey. In: Jones EG, Peters AA (eds) Cerebral cortex. Plenum

Press, New York, pp 179–226

Mesulam MM, Mufson EJ (1982) Insula of the old world monkey. I.

Architectonics in the insulo-orbito-temporal component of the

paralimbic brain. J Comp Neurol 212:1–22

Monteagudo A, Timor-Tritsch IE (1997) Development of fetal gyri,

and fissures: a transvaginal sonographic study. Ultrasound Obstet

Gynecol 9:222–228

Ostrowsky K, Isnard J, Ryvlin P, Guenot M, Fischer C, Mauguiere F

(2000) Functional mapping of the insular cortex: clinical

implication in temporal lobe epilepsy. Epilepsia 41(6):681–686

Peyron R, Schneider F, Faillenot I, Convers P, Barral FG, Garcia-

Larrea L, Laurent B (2004) An fMRI study of cortical

representation of mechanical allodynia in patients with neuro-

pathic pain. Neurology 63(10):1838–1846

Brain Struct Funct (2007) 212:335–346 345

123

Reil JC (1809) Unterfuchungen uber den Bau des grofsen Gehirns im

Menfchen: Vierte Fortsetzung VIII. Arch Physiol Halle 9:136–146

Retzius G (1896) Das menschenhirn studien in der makroskopischen

morphologie. Morstedt, Stockholm

Ruiz A, Sembely-Taveau C, Paillet C, Sirinelli D (2006) Sonographic

cerebral sulcal pattern in normal fetuses. Radiologie 87:49–55

Schreckenberger M, Siessmeier T, Viertmann A, Landvogt C,

Buchholz HG, Rolke R, Treede RD, Bartenstein P, Birklein F

(2005) The unpleasantness of tonic pain is encoded by the

insular cortex. Neurology 64(7):1175–1183

Streeter GL (1912) The development of the nervous system. In:

Keibel F, Mall FP (eds) Manual of human embryology, vol II,

chapter XIV. Lippincott, Philadelphia

Toi A, Lister WS, Fong KW (2004) How early are fetal cerebral sulci

visible at prenatal ultrasound and what is the normal pattern of

early fetal sulcale development? Ultrasound Obstet Gynecol

24:706–715

Toro R, Burnod Y (2003) Geometric atlas: modeling the cortex as an

organized surface. Neuroimage 20:1468–1484

346 Brain Struct Funct (2007) 212:335–346

123