Perigenual Cingulate Gyrus Volume in Patients withSchizophrenia: A Magnetic Resonance Imaging Study

Tsutomu Takahashi, Michio Suzuki, Yasuhiro Kawasaki, Hirofumi Hagino,Ikiko Yamashita, Shigeru Nohara, Kazue Nakamura, Hikaru Seto, andMasayoshi Kurachi

Background: Anterior cingulate gyrus abnormalitieshave been suggested to be involved in the pathophysiologyof schizophrenia; however, little is known about morpho-logic changes in the perigenual cingulate gyrus in schizo-phrenia patients.

Methods: We investigated perigenual cingulate gyrusvolume in 40 schizophrenia patients (20 men, 20 women)and 40 age- and gender-matched normal controls usingmagnetic resonance imaging. Volume of both gray andwhite matter of the perigenual cingulate gyrus was mea-sured on consecutive axial 1-mm slices.

Results: Total (left and right) perigenual cingulate graymatter volume was significantly reduced in female schizo-phrenia patients compared with female controls. Therewas no significant difference in the gray matter volume ofthe perigenual cingulate gyrus between male patients andmale controls. Left perigenual cingulate white mattervolume was significantly reduced in the patient comparedwith the control group. Furthermore, significant genderdifferences were found in the total gray and white mattervolume of the perigenual cingulate gyrus in controlsubjects (women � men), although these gender differ-ences were not significant in the patient group.

Conclusions: Our findings suggests volume reduction ofthe perigenual cingulate gyrus in schizophrenia patients,especially women and that gender differences in peri-genual cingulate morphology among normal subjects are,as has been suggested for other parts of the brain, reducedin schizophrenia patients. Biol Psychiatry 2003;53:593–600 © 2003 Society of Biological Psychiatry

Key Words: Schizophrenia, magnetic resonance imaging,perigenual cingulate gyrus, volume reduction, genderdifferences, brain

Introduction

Both postmortem (Albanese et al 1995; Benes 1993;Squires et al 1993; Wyatt et al 1995) and brain

functional imaging (Andreasen et al 1992; Carter et al1997; Dolan et al 1995; Haznedar et al 1997; Kawasaki etal 1993; Liddle et al 1992; Nohara et al 2000; Suhara et al2002; Tamminga et al 1992) studies have suggested thatanterior cingulate gyrus (ACG) abnormalities are involvedin the pathophysiology of schizophrenia. Morphologicimaging studies of this region have been relatively few,however, and have produced inconsistent results. In aprevious magnetic resonance imaging (MRI) study, wefound right ACG gray matter volume to be significantlyreduced in female patients with schizophrenia comparedwith female control subjects (Takahashi et al 2002); incontrast, Noga et al (1995) and Szeszko et al (1999) didnot find a significant difference in ACG volume amongpatients and control subjects. The inconsistencies amongthe previous MRI studies may be due in part to differentdefinitions of anatomic boundaries in measurements of theACG. Cytoarchitectural, connectional, and functional dif-ferences have been reported between the rostral (“affectivedivision”) and the more caudal (“cognitive division”)portions of the ACG, as reviewed by Devinsky et al (1995)and Vogt et al (1995). Interestingly, a recent automated,voxel-based analysis using three-dimensional (3D) MRIsuggested that the rostral portion of the ACG (i.e., peri-genual cingulate gyrus) has more severe morphologicchanges compared with the caudal portion in patients withschizophrenia (Suzuki et al 2002); however, we investi-gated primarily the caudal part in a previous MRI study,and other MRI studies (Noga et al 1995; Szeszko et al1999) did not differentiate between the rostral and caudalportions of the ACG. Thus, to our knowledge, no volu-metric MRI studies have attempted a detailed examinationof the perigenual cingulate gyrus volume in patients withschizophrenia.

Normal gender differences in brain anatomy have beenwell established (reviewed by Pearlson and Marsh 1999).With regard to the cingulate gyrus, Paus et al (1996)

From the Department of Neuropsychiatry (TT, MS, YK, HH, IY, SN, KN, MK) andRadiology (HS), Toyama Medical and Pharmaceutical University, Toyama,Japan

Address reprint requests to Tsutomu Takahashi, M.D., Department of Neuropsy-chiatry, Toyama Medical and Pharmaceutical University, 2630 Sugitani,Toyama 930-0194 Japan.

Received April 1, 2002; revised June 11, 2002; accepted June 20, 2002.

© 2003 Society of Biological Psychiatry 0006-3223/03/$30.00doi:10.1016/S0006-3223(03)01483-X

investigated intrasulcal gray matter volume of the cingu-late sulcus in healthy subjects using MRI and found asignificant gender difference (larger in female than malesubjects); however, it is unclear whether the perigenualcingulate gyrus also shows the normal gender differencefavoring women. On the other hand, given that the brainmorphologic abnormalities in patients represent a neuro-developmental origin of schizophrenia (Weinberger 1987),it is likely that schizophrenic brains show a lack of normalstructural gender differences, which may be regulatedduring fetal development (De Lacoste et al 1986, 1991;Goldstein et al 2001).

In this study, we used high-resolution 3D MRI tomeasure the gray and white matter volume of the peri-genual cingulate gyrus in male and female patients withschizophrenia and age- and gender-matched normal con-trol subjects. The purposes of this study were to investi-gate gender differences of the perigenual cingulate gyrusin normal control subjects and to test the hypotheses 1)that perigenual cingulate gyrus volume would be reducedin patients with schizophrenia compared with normalcontrol subjects and 2) that the normal gender differencesof the perigenual cingulate gyrus volume, if present,would be reduced in patients with schizophrenia. More-over, we examined whether these volumetric measure-ments were related to clinical symptoms of schizophrenia.

Methods and Materials

SubjectsForty right-handed patients with schizophrenia (20 men and 20women) were recruited from the inpatient and outpatient clinicsof the Department of Neuropsychiatry, Toyama Medical andPharmaceutical University Hospital. All patients fulfilled ICD-10diagnostic criteria for research on schizophrenia (World HealthOrganization 1993). The patients primarily had paranoid orundifferentiated subtypes, and male and female patients werematched in their clinical subtypes. Their mean age was 26.1 �5.0 (SD) years (range, 15–36). All but two of the patients wereon neuroleptic medication (mean haloperidol equivalent dose �9.3 mg/day, SD � 9.4), with a mean duration of medication of3.7 years (SD � 3.5). All patients were physically healthy at thetime of the study, and none had a lifetime history of serious headtrauma, neurologic illness, serious medical or surgical illness, orsubstance abuse. At the time of the MRI study, their mean scoreson the Scale for the Assessment of Negative Symptoms (SANS;Andreasen 1984a) and the Scale for the Assessment of PositiveSymptoms (SAPS; Andreasen 1984b) were 44.2 (SD � 21.0)and 23.5 (SD � 18.4), respectively.

The control subjects consisted of 40 right-handed healthyvolunteers (20 men and 20 women) recruited from amonghospital staff and medical and pharmaceutical students. Theirmean age was 25.1 � 5.8 (SD) years (range, 18–38). Candidateswere excluded if they had any personal or family history ofpsychiatric illness; a lifetime history of serious head trauma,

neurologic illness, serious medical or surgical illness; or sub-stance abuse. The Minnesota Multiphasic Personality Inventory(MMPI) was administered to all the control candidates, and theywere excluded if any t score for the validity or the clinical scalesexceeded 70.

The demographic and clinical characteristics of the male andfemale patients with schizophrenia and control subjects havebeen reported previously (Takahashi et al 2002). There were nosignificant differences among the four groups in age or parentaleducation; however, there were significant differences in educa-tion across the four groups [male patients, 14.2 � 2.1 years;female patients, 13.1 � 2.1 years; male control subjects, 17.2 �2.6 years; female control subjects, 14.5 � 1.2 years; analysis ofvariance (ANOVA), F � 14.10(3,76), p � .001]. The post hocScheffe test showed the male control subjects to have a higherlevel of education than the female control subjects (p � .002),the female patients (p � .001), and the male patients (p � .001).There was no significant difference between the patients and thecontrol subjects in height, although the men were significantlytaller than the women [men, 171.8 � 4.8 cm; women, 159.1 �3.8 cm; ANOVA, F � 174.72(1,78), p � .001]. There were nosignificant differences between male and female patients in ageat onset, duration of illness, or dosage or duration of neurolepticmedication. There were no significant differences between maleand female patients in the total score or the subscale scores forSAPS and SANS.

All subjects participated in the study after providing writteninformed consent. This study was approved by the Committee onMedical Ethics of Toyama Medical and Pharmaceutical Univer-sity. When subjects were younger than 18 years, informedconsent was also obtained from their parents.

Magnetic Resonance Imaging ProceduresMagnetic resonance images were obtained using a 1.5-T Mag-netom Vision (Siemens Medical System, Erlangen, Germany)with a 3D gradient-echo sequence FLASH (fast low-angle shots)yielding 160–180 contiguous T1-weighted 1-mm-thick slices inthe sagittal plane. Imaging parameters were as follows: repetitiontime � 24 msec; echo time � 10 msec; flip angle � 40°; field ofview � 256 mm; and matrix size � 256 � 256 pixels. The voxelsize was 1.0 � 1.0 � 1.0 mm3. Magnetic field inhomogeneitiesin our scanner were monitored by weekly phantom scanning anddaily basic quality control, and they were stable over the MRacquisition time for this study.

The images were transferred to a Unix workstation (SiliconGraphics, Mountain View, CA). The data were coded randomlyand analyzed with the software package Dr View 5.2 (AsahiKasei Joho System Co., Tokyo, Japan) blind to subjects’ genderand diagnosis. This software allows the rater to view the imagevoxels in three mutually orthogonal planes simultaneously, anyone of which may be stipulated by the rater. Brains wererealigned in three dimensions on a workstation to standardize fordifferences in head tilt during image acquisition. Scans were thenreconstructed into entire contiguous 1-mm-thick axial images,parallel to the anterior commissure–posterior commissure (AC–PC) line. The signal-intensity histogram distributions from theT1-weighted images across the whole brain for each subject were

594 T. Takahashi et alBIOL PSYCHIATRY2003;53:593–600

used to segment the voxels semiautomatically into gray matter,white matter, and cerebrospinal fluid (CSF) according to theAlpert algorithm (Alpert et al 1996). The histogram of graylevels was computed and used to select minimal intensity pointsbetween the gray matter and CSF peaks (lower intensity thresh-old) and between the gray and white matter peaks (upperintensity threshold). After separating the CSF from the tissue bythe lower intensity threshold, the resulting tissue was separatedinto gray and white matter using the upper intensity threshold.Although the images were not corrected for the magnetic fieldinhomogeneities, no visible effect on quality of segmentationwas detected in any of the cases. Before volumetric analysis,masks were semiautomatically created to demarcate the outerextent of the intracranial contents, with the skull, scalp, and necktissue removed. Minimal manual editing of the masks wasrequired.

Perigenual Cingulate Gyrus Measurements

As illustrated in Figure 1, the perigenual cingulate gyrus wasbounded anteriorly by the cingulate sulcus and posteriorly by theplane that was perpendicular to the AC–PC line and passingthrough the anterior margin of the genu of the corpus callosum.The left and right perigenual cingulate gyri were separatelytraced in consecutive axial 1-mm slices from ventral to dorsal,beginning with the plane showing the appearance of the cingulatesulcus and ending dorsally with that showing disappearance ofthe corpus callosum, following the methods of Hazneder et al(1997).

There are distinct differences in the connectivities and func-tions of the rostral (perigenual) versus caudal parts of the ACG,as reviewed by Devinsky et al (1995) and Vogt et al (1995),although these parts cannot be anatomically bounded by theirown structures. Therefore, we defined the previously mentionedplane as the posterior boundary of the perigenual cingulate gyrusnot to include the caudal part of the ACG on the basis of theproportions from the diagram of Vogt et al (1995) and fromTalairach and Tournoux (1988).

All measurements were carried out by one rater (TT), who wasblinded to subjects’ identity, gender, and diagnosis. To determinethe reliability of the measurements, five subjects were randomlyselected. The perigenual cingulate gyrus in a subset of these fivesubjects were measured independently by two raters (TT, YK),and intraclass correlation coefficients (ICCs) were calculated.The interrater ICCs of the gray and white matter of the peri-genual cingulate gyrus were greater than .96. Each volume wasthen remeasured after at least 4 months by the first rater; theintrarater ICCs of the gray and white matter of the perigenualcingulate gyrus were greater than .92.

Statistical Analysis

Statistical analysis was carried out using the software packageSTATISTICA 4.1J for Macintosh (StatSoft, Tulsa, OK). Relativeperigenual cingulate gyrus volume, used to control for differ-ences in whole brain volume, was obtained by dividing theabsolute volume of the perigenual cingulate gyrus by whole brainvolume and multiplying the result by 100. The whole brainvolume of these subjects was reported elsewhere (Takahashi et al2002). There was no significant difference between the patientsand control subjects in the whole brain volume (see Table 1). Therelative volumes of gray and white matter of the perigenualcingulate gyrus were analyzed by repeated-measures multivariateanalysis of variance with age as a covariate (MANCOVA), group(control subjects, patients with schizophrenia), and gender (male,female) as between-subject factors and hemisphere (left, right) asa within-subject variable. Post hoc Tukey tests were conducted tofollow up the significant main effects or interactions. Correlationsbetween the relative perigenual cingulate gyrus volume and age atonset, duration of illness, medication dosage and duration ofneutoleptics medication were analyzed using Kendall’s � correlationcoefficients. To analyze the volume changes in relation to theclinical symptoms, Kendall’s � correlation with Bonferroni correc-tion was also calculated between the relative perigenual cingulategyrus gray or white matter volume and scores for subscales of SAPSand SANS. Statistical significance was defined as p � .05.

Figure 1. Boundaries for the rostral (a) and the caudal(b) parts of the anterior cingulate gyrus (ACG) areillustrated. The perigenual cingulate gyrus (a) wastraced bilaterally in consecutive axial 1-mm slicesparallel to the anterior commissure–posterior commis-sure (AC–PC) line. The most dorsal axial plane show-ing the corpus callosum (1) and the most ventral axialplane showing the cingulate sulcus (2) were chosen asthe superior and inferior boundaries, respectively. Oneach axial slice, the perigenual cingulate gyrus wasbounded anteriorly by the cingulate sulcus (3), andposteriorly by the plane that was perpendicular to theAC–PC line and passing through the anterior margin ofthe genu of the corpus callosum (4). The caudal part ofthe ACG (b) has been measured in a previous magneticresonance imaging study (Takahashi et al 2002).

Perigenual Cingulate Gyrus Volume in Schizophrenia 595BIOL PSYCHIATRY2003;53:593–600

Results

The relative volumes of the perigenual cingulate gyrus inpatients with schizophrenia and control subjects are shownin Table 1. Repeated-measures MANCOVA of the graymatter of the perigenual cingulate gyrus revealed a signif-icant main effect for group [F � 4.02(1,75); p � .048] andgender [F � 7.69(1,75); p � .007] and a nonsignificanttrend in main effect for hemisphere [F � 3.31(1,76); p �.073]. There was a significant group � gender interaction[F � 4.33(1,75); p � .041]. There was no group �hemisphere [F � 1.43(1,76); p � .230], gender � hemi-sphere [F � 0.00(1,76); p � .952], or group � gender �hemisphere [F � 0.02(1,76); p � .883] interaction. Re-peated-measures MANCOVA of the white matter revealedhemisphere to have a significant main effect [F �9.31(1,76); p � .003]. There was no significant maineffect for group [F � 1.19(1,75); p � .278] or gender [F� 2.10(1,75); p � .151]. There were significant group �gender [F � 5.49(1,75); p � .022] and group � hemi-sphere [F � 6.61(1,76); p � .012] interactions. There wasno gender � hemisphere [F � 0.69(1,76); p � .408] orgroup � gender � hemisphere [F � 0.05(1,76); p � .827]interaction.

Post hoc analyses showed the gray matter volume of theperigenual cingulate gyrus to be significantly reduced infemale patients compared with female control subjects (p� .020). The gray matter volume of female controlsubjects was significantly larger than that of male controlsubjects (p � .005), whereas this gender difference wasnot significant in the patients group (p � .951). There wasno significant difference in the gray matter volume of theperigenual cingulate gyrus between male patients andmale control subjects. As for white matter, the white

matter volume of the perigenual cingulate gyrus wassignificantly larger in the right than in the left hemisphere(p � .003). The left white matter volume was significantlyreduced in patients compared with normal control subjects(p � .033). The white matter findings were similar tothose of the gray matter in that the volume was signifi-cantly larger in female control subjects than in malecontrol subjects (p � .045), and this gender difference wasreduced in the patient group (p � .919).

There were no significant correlations between therelative perigenual cingulate gyrus volume and age atonset, duration of illness, or dosage or duration of neuro-leptic medication. There were no significant correlationsbetween the perigenual cingulate gyrus volume and thescores for subscales of SAPS and SANS.

Discussion

In this study, we found that total (left and right) gray andwhite matter volume of the perigenual cingulate gyruswere significantly larger in female control subjects than inmale control subjects, but these normal gender differenceswere not significant in patients with schizophrenia. Fur-thermore, we found the volume of the total perigenualcingulate gray matter to be significantly reduced in femalepatients with schizophrenia compared with female controlsubjects. Left perigenual cingulate white matter volumewas also reduced in patients group compared with controlsubjects. To our knowledge, this is the first volumetricMRI study that reports such morphologic changes in theperigenual cingulate gyrus in patients with schizophrenia.

The rostral and caudal parts of the ACG have beenreported to have cytoarchitectural, connectional, and func-

Table 1. Absolute Whole Brain Volume and Relative Volume of Gray and White Matter of the Perigenual Cingulate Gyrus inPatients with Schizophrenia and in Control Subjects

Brain Region

Patients with Schizophrenia Control Subjects

Male (n � 20) Female (n � 20) Male (n � 20) Female (n � 20)

Whole Brain (cm3) 1124.6 � 103.3 1008.8 � 87.4 1166.5 � 84.0 1058.7 � 76.0Gray Mattera

Left .153 � .051 .161 � .062 .164 � .039 .207b � .072Right .179 � .040 .184 � .049 .168 � .051 .213b � .055

White Matterc

Left .016 � .010 .015 � .008 .017d � .008 .028d,e � .015Rightf .027 � .021 .024 � .012 .020 � .013 .027e � .012

Values represent means � SDs. Relative perigenual cingulate gyrus volumes were calculated as follows: (absolute perigenual cingulate gyrus volume/whole brainvolume) � 100.

aSignificant main effect (repeated-measures multivariate analysis of covariance [MANCOVA]) for group [F � 4.02 (1,75), p � .048] and gender [F � 7.69 (1,75), p� .007] and a significant group � gender interaction [F � 4.33 (1,75), p � .041].

bSignificantly different from gray matter in male control subjects (Tukey test, p � .005) and gray matter in female patients with schizophrenia (Tukey test, p � .020).cSignificant main effect for hemisphere [repeated-measures MANCOVA, F � 9.31 (1,76), p � .003] and significant group � gender [repeated-measures MANCOVA,

F � 5.49 (1,75), p � .022] and group � hemisphere [repeated-measures MANCOVA, F � 6.61 (1,76), p � .012] interaction.dSignificantly different from left white matter in patients with schizophrenia (Tukey test, p � .033).eSignificantly different from white matter in male control subjects (Tukey test, p � .045).fSignificantly different from left white matter (Tukey test, p � .003).

596 T. Takahashi et alBIOL PSYCHIATRY2003;53:593–600

tional differences (Devinsky et al 1995; Vogt et al 1995).The rostral part of the ACG (i.e., perigenual cingulategyrus) approximately corresponds to Brodmann’s area 24,which has extensive connections with the amygdala andperiaqueductal gray matter. This part is referred to as theaffective division, which is activated in response to emo-tional manipulations, as demonstrated in several neuroim-aging studies among healthy subjects (George et al 1993,1995; Ploghaus et al 2001; Whalen et al 1998). In addition,previous functional neuroimaging studies have implicatedthe perigenual cingulate gyrus abnormalities in patientswith depression (Kennedy et al 2001; Mayberg et al 1997)or obsessive–compulsive disorder (Breiter et al 1996;Rauch et al 1994). Significant volume reduction of theperigenual cingulate gyrus in schizophrenia, as observedin our study, is potentially of interest because patients withschizophrenia generally show abnormalities in affect,often accompanied by obsessive–compulsive symptoms(Fenton and McGlashan 1986; Samuel et al 1993). Incontrast to the rostral portion, the more caudal part of theACG, which approximately corresponds to Brodmann’sarea 24� has extensive connections with the parietal cortexand is considered to be the cognitive division (reviewed byDevinsky et al 1995). In a previous MRI study, weexamined the volume of the caudal cognitive division ofthe ACG in the same subjects discussed here and found thegray matter volume to be significantly reduced in the rightbut not the left hemisphere, in the female patients withschizophrenia compared with female control subjects (Ta-kahashi et al 2002). On the other hand, in the presentstudy, we found that the gray matter volume of theperigenual cingulate gyrus was reduced bilaterally infemale patients compared with female control subjects.With regard to the white matter, left perigenual cingulategyrus white matter volume was significantly reduced inpatients group in the study reported here, although thewhite matter volume of the caudal part of the ACGshowed no significant changes in the patients (Takahashiet al 2002). Thus, it is suggested that the rostral and thecaudal parts of the ACG have at least partially differentpatterns of structural abnormalities in schizophrenia, pos-sibly related to varying involvement of affective versuscognitive divisions of the ACG in the pathophysiology ofschizophrenia.

The relative perigenual cingulate gyrus volume did notcorrelate with clinical symptoms of schizophrenia in thisstudy. The reason for our failure to find significantcorrelations is unclear, but it may be related to the timingof clinical assessment of the patients. Clinical symptomsof schizophrenia can be reversibly altered with neurolepticmedication, but brain morphologic changes are consideredmore static. An important confound in this study is that weassessed clinical symptoms of schizophrenia at the time of

MRI study. To clarify the relationships between peri-genual cingulate morphology and clinical symptoms, first-episode or unmedicated patients with schizophreniashould be investigated in future study.

In this study, we found a normal gender difference,favoring female control subjects, in the total gray mattervolume of the perigenual cingulate gyrus. The total whitematter volume was also larger in the female controlsubjects than in the male control subjects; however, thesegender differences were reduced in patients with schizo-phrenia. As described earlier, perigenual cingulate gyrus isinvolved in affect; healthy women subjects were reportedto have relatively higher glucose metabolism during rest-ing states than their male counterparts (Gur et al 1995). Ina previous MRI study, Paus et al (1996) reported theintrasulcus gray matter volume of the anterior part of thecingulate sulcus to be significantly larger in female than inmale control subjects. Our finding of normal genderdifferences in the perigenual cingulate gyrus volume is inagreement with these previous observations. In healthysubjects, gender differences that occur during fetal devel-opment, such as differences in shape of the corpuscallosum (De Lacoste et al 1986) or in cortical asymme-tries (De Lacoste et al 1991), have been reported. Giventhat the gender differences of the perigenual cingulategyrus are also regulated prenatally, our finding of a lack ofnormal gender differences in patients may suggest aprocess involving abnormal neurodevelopment in schizo-phrenic brains. Lack of the normal structural genderdifferences in schizophrenia has also been reported inother brain regions. For example, a left-greater-than-rightasymmetry in the planum temporale is seen more often inhealthy male subjects than in their females counterparts(reviewed by Pearlson and Marsh 1999). In schizophrenia,several MRI (Barta et al 1997; Kulynych et al 1996) andpostmortem (Falkai et al 1995) studies have reported alack of the normal structural asymmetry of the planumtemporale, especially in male patients. From the stand-point of gender differences, this finding could be inter-preted as a disruption of the normal patterns of genderdifferences of the planum temporale in schizophrenia.Interestingly, Goldstein et al (2002) conducted a compre-hensive assessment of all the brain regions using asophisticated novel method based on MRI; these research-ers also reported that the normal patterns of sexualdimorphisms, particularly in the cortex, was disrupted inpatients with schizophrenia. These previous studies, aswell as the study we report here, suggest that disruption ofthe normal patterns of gender differences in the brainmight be a common feature in schizophrenia.

Male and female patients in our study did not differ onsymptom ratings, suggesting that they were similar in theseverity of their illness; however, significant volume

Perigenual Cingulate Gyrus Volume in Schizophrenia 597BIOL PSYCHIATRY2003;53:593–600

reduction of the gray matter of the perigenual cingulategyrus was observed only in the female patients. In schizo-phrenia, structural brain abnormalities such as ventricularenlargement or volume reductions of the temporal lobestructures have been suggested to be greater in male thanin female patients (Lawrie and Abukmeil 1998; Pearlsonand Marsh 1999). On the contrary, gray matter reductionsin the frontal areas (Nasrallah et al 1990; Suzuki et al2002) and the caudal part of the ACG (Takahashi et al2002) have been observed predominantly in female pa-tients. It thus may be assumed that male and femalepatients with schizophrenia have, at least in part, differentpatterns of structural brain abnormalities.

Several limitations of this study need to be taken intoaccount before any conclusion can be drawn. First, mostpatients with schizophrenia in this study were chronicallymedicated. In schizophrenia, the relation between brainmorphologic features and certain clinical factors, such asduration of illness (DeLisi et al 1997; Gur et al 1998) andneuroleptics (Chakos et al 1995; Gur et al 1998; Keshavanet al 1994, 1998), have been reported. We did not excludethese factors, although perigenual cingulate gyrus volumewas not correlated with duration of illness, neurolepticdosage, or duration of neuroleptic medication. A secondlimitation is that the control subjects in our study were notselected to be educationally equivalent to the patients withschizophrenia, and there was a gender difference ineducational level among control subjects (higher in menthan in women); however, we optimally matched theparental education among the male and female patientsand control subjects in our study according to the notionthat matching on the basis of educational level of parentsmay reduce confounding factors in selection of controlgroups when brain measures are studied (Andreasen et al1990). A third limitation of our study is that we used thecorpus callosum as an anatomic landmark for definition ofthe perigenual cingulate gyrus. The corpus callosum is oneof several brain regions suggested to be abnormal inschizophrenic brains (Downhill et al 2000; Tibbo et al1998; Woodruff et al 1993, 1995). We are not excludingthe possibility that differences in size or shape of thecorpus callosum between the patients with schizophreniaand control subjects biased the results. In addition to theselimitations, the anterior end of the cingulate sulcus wasoften difficult to identify because of several variationsamong individuals in the course of the sulcus; however,the variation generally did not affect the accuracy of themeasurement because the anterior end is positioned pos-teriorly to the anterior margin of the corpus callosum inmost cases (Ono et al 1990).

In conclusion, we found a volume reduction of theperigenual cingulate gyrus in patients with schizophrenia,especially in female patients. As a result, gender differ-

ences in the perigenual cingulate gyrus volume in normalcontrol subjects (larger in female than in male subjects)were reduced in patients with schizophrenia. This lack ofnormal gender difference in schizophrenic brains mayrepresent a neurodevelopmental origin of the disease. It isalso suggested that rostral and caudal portions of the ACGhave a partly different pattern of structural brain abnor-malities in schizophrenia. Patients with other psychiatricdisorders, such as affective disorders or obsessive–com-pulsive disorder, and patients with schizophrenia whohave obsessive–compulsive symptoms, should be in-cluded in further studies to clarify the diagnostic specific-ity of our findings and to assess whether perigenualcingulate morphology is related to affective or obsessive–compulsive symptoms of schizophrenia.

This study was supported in part by a Research Grant (11–3) for Nervousand Mental Disorders from the Ministry of Health and Welfare (Japan).We are grateful to all attending doctors who collected clinical data ontheir patients.

ReferencesAlbanese AM, Merlo AB, Mascitti TA, Tornese EB, Gomez EE,

Konopka V, et al (1995): Inversion of the hemisphericlaterality of the anterior cingulate gyrus in schizophrenics.Biol Psychiatry 38:13–21.

Alpert NM, Berdichevsky D, Levin Z, Morris ED, Fischman AJ(1996): Improved methods for image registration. Neuroim-age 3:10–18.

Andreasen NC (1984a): The Scale for the Assessment of NegativeSymptoms (SANS). Iowa City, IA: University of Iowa.

Andreasen NC (1984b): The Scale for the Assessment of PositiveSymptoms (SAPS). Iowa City, IA: University of Iowa.

Andreasen NC, Ehrhardt JC, Swayze VW II, Alliger RJ, YuhWT, Cohen G, et al (1990): Magnetic resonance imaging ofthe brain in schizophrenia: The pathophysiologic significanceof structural abnormalities. Arch Gen Psychiatry 47:35–44.

Andreasen NC, Rezai K, Alliger R, Swayze II VW, Flaum M,Kirchner P, et al (1992): Hypofrontality in neuroleptic-naivepatients and in patients with chronic schizophrenia: Assess-ment with xenon 133 single-photon emission computedtomography and the Tower of London. Arch Gen Psychiatry49:943–958.

Barta PE, Pearlson GD, Brill LB II, Royall R, McGilchrist IK,Pulver AE, et al (1997): Planum temporale asymmetryreversal in schizophrenia: Replication and relationship to graymatter abnormalities. Am J Psychiatry 154:661–667.

Benes FM (1993): Neurobiological investigations in cingulatecortex of schizophrenic brain. Schizophr Bull 19:537–549.

Breiter HC, Rauch SL, Kwong KK, Baker JR, Weisskoff RM,Kennedy DN, et al (1996): Functional magnetic resonanceimaging of symptom provocation in obsessive-compulsivedisorder. Arch Gen Psychiatry 53:595–606.

Carter CS, Mintun M, Nichols T, Cohen JD (1997): Anteriorcingulate gyrus dysfunction and selective attention deficits in

598 T. Takahashi et alBIOL PSYCHIATRY2003;53:593–600

schizophrenia: [15O]H2O PET study during single-trialStroop task performance. Am J Psychiatry 154:1670–1675.

Chakos MH, Lieberman JA, Alvir J, Bilder R, Ashtari M (1995):Caudate nuclei volumes in schizophrenic patients treated withtypical antipsychotics and clozapine. Lancet 345:456–457.

De Lacoste MC, Holloway RL, Woodward DJ (1986): Sexdifferences in the fetal human corpus callosum. Hum Neuro-biol 5:93–96.

De Lacoste MC, Horvath DS, Woodward DJ (1991): Possiblesex differences in the developing human fetal brain. J ClinExp Neuropsychol 13:831–846.

DeLisi LE, Sakuma M, Tew W, Kushner M, Hoff AL, GrimsonR (1997): Schizophrenia as a chronic active brain process: Astudy of progressive brain structural change subsequent to theonset of schizophrenia. Psychiatry Res Neuroimaging74:129–140.

Devinsky O, Morrell MJ, Vogt BA (1995): Contributions ofanterior cingulate cortex to behavior. Brain 118:279–306.

Dolan RJ, Fletcher P, Frith CD, Friston KJ, Frackowiak RSJ,Grasby PM (1995): Dopaminergic modulation of impairedcognitive activation in the anterior cingulate cortex in schizo-phrenia. Nature 378:180–182.

Downhill JE Jr, Buchsbaum MS, Wei T, Spiegel-Cohen J,Hazlett EA, Haznedar MM, et al (2000): Shape and size of thecorpus callosum in schizophrenia and schizotypal personalitydisorder. Schizophr Res 42:193–208.

Falkai P, Bogerts B, Schneider T, Greve B, Pfeiffer U, Pilz K, etal (1995): Disturbed planum temporale asymmetry in schizo-phrenia. A quantitative post-mortem study. Schizophr Res14:161–176.

Fenton WS, McGlashan TH (1986): The prognostic significanceof obsessive-compulsive symptoms in schizophrenia. Am JPsychiatry 143:437–441.

George MS, Ketter TA, Gill DS, Haxby JV, Ungerleider LG,Herscovitch P, et al (1993): Brain regions involved inrecognizing facial emotion or identity: An oxygen-15 PETstudy. J Neuropsychol Clin Neurosci 5:384–394.

George MS, Ketter TA, Parekh PI, Horwitz B, Herscovitch P,Post RM (1995): Brain activity during transient sadness andhappiness in healthy women. Am J Psychiatry 152:341–351.

Goldstein JM, Seidman LJ, Horton NJ, Makris N, Kennedy DN,Caviness VS Jr, et al (2001): Normal sexual dimorphism ofthe adult human brain assessed by in vivo magnetic resonanceimaging. Cereb Cortex 11:490–497.

Goldstein JM, Seidman LJ, O’Brien LM, Horton NJ, KennedyDN, Makris N, et al (2002): Impact of normal sexualdimorphisms on sex differences in structural brain abnormal-ities in schizophrenia assessed by magnetic resonance imag-ing. Arch Gen Psychiatry 59:154–164.

Gur RE, Cowell P, Turetsky BI, Gallacher F, Cannon T, BilkerW, et al (1998): A follow-up magnetic resonance imagingstudy of schizophrenia. Arch Gen Psychiatry 55:145–152.

Gur RC, Mozley LH, Mozley PD, Resnick SM, Karp JS, AlaviA, et al (1995): Sex differences in regional cerebral glucosemetabolism during a resting state. Science 267:528–531.

Haznedar MM, Buchsbaum MS, Metzger M, Solimando A,Spiegel-Cohen J, Hollander E (1997): Anterior cingulategyrus volume and glucose metabolism in autistic disorder.Am J Psychiatry 154:1047–1050.

Kawasaki Y, Maeda Y, Suzuki M, Urata K, Higashima M, KibaK, et al (1993): SPECT analysis of regional cerebral bloodflow changes in patients with schizophrenia during theWisconsin Card Sorting Test. Schizophr Res 10:109–116.

Kennedy SH, Evans KR, Kruger S, Mayberg HS, Meyer JH,McCann S, et al (2001): Changes in regional brain glucosemetabolism measured with positron emission tomographyafter paroxetine treatment of major depression. Am J Psychi-atry 158:899–905.

Keshavan MS, Bagwell WW, Haas GL, Sweeney JA, SchoolerNR, Pettegrew JW (1994): Changes in caudate volume withneuroleptic treatment. Lancet 344:1434.

Keshavan MS, Haas GL, Kahn CE, Aguilar E, Dick EL, SchoolerNR, et al (1998): Superior temporal gyrus and the course ofearly schizophrenia: Progressive, static, or reversible? J Psy-chiatr Res 32:161–167.

Kulynych JJ, Vladar K, Jones DW, Weinberger DR (1996):Superior temporal gyrus volume in schizophrenia: A studyusing MRI morphometry assisted by surface rendering. Am JPsychiatry 153:50–56.

Lawrie SM, Abukmeil SS (1998): Brain abnormality in schizo-phrenia: A systematic and quantitative review of volumetricmagnetic resonance imaging studies. Br J Psychiatry172:110–120.

Liddle PF, Friston KJ, Frith CD, Hirsch SR, Jones T, FrackowiakRSJ (1992): Patterns of cerebral blood flow in schizophrenia.Br J Psychiatry 160:179–186.

Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, BrickmanJS, Tekell JL, et al (1997): Cingulate function in depression:A potential predictor of treatment response. Neuroreport8:1057–1061.

Nasrallah HA, Schwarzkopf SB, Olson SC, Coffman JA (1990):Gender differences in schizophrenia on MRI brain scans.Schizophr Bull 16:205–210.

Noga JT, Aylward E, Barta PE, Pearlson GD (1995): Cingulategyrus in schizophrenic patients and normal volunteers. Psy-chiatry Res Neuroimaging 61:201–208.

Nohara S, Suzuki M, Kurachi M, Yamashita I, Matsui M, Seto H,et al (2000): Neural correlates of memory organizationdeficits in schizophrenia: A single photon emission computedtomography study with 99 m Tc-ethyl-cysteinate dimer duringa verbal learning task. Schizophr Res 42:209–222.

Ono M, Kubik S, Abernathy CD (1990): Atlas of the CerebralSulci. Stuttgart: Thieme Verlag.

Paus T, Otaky N, Caramanos Z, MacDonald D, Zijdenbos A,D’Avirro D, et al (1996): In vivo morphometry of theintrasulcal gray matter in the human cingulate, paracingulate,and superior-rostal sulci: Hemispheric asymmetries, genderdifferences and probability maps. J Comp Neurol 376:664–673.

Pearlson GD, Marsh L (1999): Structural brain imaging inschizophrenia: A selective review. Biol Psychiatry 46:627–649.

Ploghaus A, Narain C, Beckmann CF, Clare S, Bantick S, WiseR, et al (2001): Exacerbation of pain by anxiety is associatedwith activity in a hippocampal network. J Neurosci 21:9896–9903.

Rauch SL, Jenike MA, Alpert NM, Baer L, Breiter HCR, SavageCR, et al (1994): Regional cerebral blood flow measured

Perigenual Cingulate Gyrus Volume in Schizophrenia 599BIOL PSYCHIATRY2003;53:593–600

during symptom provocation in obsessive–compulsive disor-der using 15O-labeled CO2 and positron emission tomogra-phy. Arch Gen Psychiatry 51:62–70.

Samuel J, Nesdadt G, Wolyniec P, Adler L, Liang KY, PulverAE (1993): Obsessive–compulsive symptoms in schizophre-nia. Schzophr Res 9:139.

Squires RF, Lajtha A, Saederup E, Palkovits M (1993): Reduced[3H]flunitrazepam binding in cingulate cortex and hippocam-pus of postmortem schizophrenic brains: Is selective loss ofglutamatergic neurons associated with major psychoses?Neurochem Res 18:219–223.

Suhara T, Okubo Y, Yasuno F, Sudo Y, Inoue M, Ichimiya T, etal (2002): Decreased dopamine D2 receptor binding in theanterior cingulate cortex in schizophrenia. Arch Gen Psychi-atry 59:25–30.

Suzuki M, Nohara S, Hagino H, Kurokawa K, Yotsutsuji T,Kawasaki Y, et al (2002): Regional changes in brain gray andwhite matter in patients with schizophrenia demonstratedwith voxel-based analysis of MRI. Schzophr Res 55:41–54.

Szeszko PR, Bilder RM, Lencz T, Pollack S, Alvir JMJ, AshtariM, et al (1999): Investigation of frontal lobe subregions infirst-episode schizophrenia. Psychiatry Res Neuroimaging90:1–15.

Takahashi T, Kawasaki Y, Kurokawa K, Hagino H, Nohara S,Yamashita I, et al (2002): Lack of normal structural asym-metry of the anterior cingulate gyrus in female patients withschizophrenia: A volumetric magnetic resonance imagingstudy. Schizophr Res 55:69–81.

Talairach J, Tournoux P (1988): Coplanar Stereotaxic Atlas ofthe Human Brain. New York: Thieme Medical.

Tamminga CA, Thaker GK, Buchanan R, Kirkpatrick B, AlphsLD, Chase TN, et al (1992): Limbic system abnormalities

identified in schizophrenia using positron emission tomogra-phy with fluorodeoxyglucose and neocortical alterations withdeficit syndrome. Arch Gen Psychiatry 49:522–530.

Tibbo P, Nopoulos P, Arndt S, Andreasen NC (1998): Corpuscallosum shape and size in male patients with schizophrenia.Biol Psychiatry 44:405–412.

Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR (1995): Humancingulate cortex: Surface features, flat maps, and cytoarchi-tecture. J Comp Neurol 359:490–506.

Weinberger DR (1987): Implications of normal brain develop-ment for the pathogenesis of schizophrenia. Arch Gen Psy-chiatry 44:660–669.

Whalen PJ, Bush G, McNally RJ, Wilhelm S, McInerney SC,Jenike MA, et al (1998): The emotional counting Stroopparadigm: A functional magnetic resonance imaging probe ofthe anterior cingulate affective division. Biol Psychiatry44:1219–1228.

Woodruff PWR, McManus IC, David AS (1995): Meta-analysisof corpus callosum size in schizophrenia. J Neurol NeurosurgPsychiatry 58:457–461.

Woodruff PWR, Pearlson GD, Geer MJ, Barta PE, Chilcoat HD(1993): A computerized magnetic resonance imaging study ofcorpus callosum morphology in schizophrenia. Psycol Med23:45–56.

World Health Organization (1993): The ICD-10 Classification ofMental and Behavioural Disorders: Diagnostic Criteria forResearch. Geneva: World Health Organization.

Wyatt RJ, Karoum F, Casanova MF (1995): Decreased DOPACin the anterior cingulate cortex of individuals with schizo-phrenia. Biol Psychiatry 38:4–12.

600 T. Takahashi et alBIOL PSYCHIATRY2003;53:593–600