Sexually dimorphic distribution of orbitofrontal sulcogyral pattern in schizophrenia

Authors

  • Kumi Uehara-Aoyama MD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    2. Department of Clinical Research, Kanagawa Psychiatric Center, Kanagawa, Japan
    Search for more papers by this author
  • Motoaki Nakamura MD, PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    2. Department of Clinical Research, Kanagawa Psychiatric Center, Kanagawa, Japan
    Search for more papers by this author
  • Takeshi Asami MD, PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    2. Psychiatry Neuroimaging Laboratory, Department of Psychiatry, Brigham and Women's Hospital, Harvard Medical School
    3. Clinical Neuroscience Division, Laboratory of Neuroscience, VA Boston Healthcare System, Brockton and Harvard Medical School, Massachusetts, USA
    Search for more papers by this author
  • Takeshi Yoshida MD, PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    Search for more papers by this author
  • Fumi Hayano PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    Search for more papers by this author
  • Tomohide Roppongi MD, PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    2. Psychiatry Neuroimaging Laboratory, Department of Psychiatry, Brigham and Women's Hospital, Harvard Medical School
    3. Clinical Neuroscience Division, Laboratory of Neuroscience, VA Boston Healthcare System, Brockton and Harvard Medical School, Massachusetts, USA
    Search for more papers by this author
  • Akiko Fujiwara MD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    Search for more papers by this author
  • Tomio Inoue MD, PhD,

    1. Department of Psychiatry, Yokohama City University, Yokohama
    Search for more papers by this author
  • Martha E. Shenton PhD,

    1. Psychiatry Neuroimaging Laboratory, Department of Psychiatry, Brigham and Women's Hospital, Harvard Medical School
    2. Clinical Neuroscience Division, Laboratory of Neuroscience, VA Boston Healthcare System, Brockton and Harvard Medical School, Massachusetts, USA
    Search for more papers by this author
  • Yoshio Hirayasu MD, PhD

    Corresponding author
    1. Department of Psychiatry, Yokohama City University, Yokohama
    Search for more papers by this author

Yoshio Hirayasu, MD, PhD, Department of Psychiatry, Yokohama City University, Yokohama 236-0004, Japan. Email: hirayasu@yokohama-cu.ac.jp

Abstract

Aim:  The sulcogyral pattern of the orbitofrontal cortex (OFC) is characterized by a remarkable inter-individual variability that likely reflects neurobehavioral traits and genetic aspects of neurodevelopment. The aim of the present study was to evaluate the OFC sulcogyral pattern of patients with schizophrenia (SZ) and healthy controls (HC) to determine group differences in OFC sulcogyral pattern as well as gender differences between groups.

Methods:  Forty-seven SZ patients (M/F, 23/24) and forty-seven HC (M/F, 17/30), matched on age and gender, were analyzed using magnetic resonance imaging. The sulcogyral pattern was classified into type I, II, or III based on the guidelines set by Chiavaras and Petrides in a previous paper. Chi-squared analysis was used to investigate group and gender differences in the sulcogyral pattern distribution, and categorical regression was used to explore clinical correlations.

Results:  The distribution of OFC sulcogyral pattern in HC replicated the results found in the previous study (left, χ2 = 0.02, P = 0.989; right, χ2 = 0.97, P = 0.616), in that there were no gender differences. Moreover, the distribution in SZ-M was in accordance with that in the previous study (left, χ2 = 1.59, P = 0.451; right, χ2 = 0.14, P = 0.933). Additionally, within SZ-M, patients with the type III pattern had a higher total positive and negative syndrome scale score (β = 0.902, F = 14.75, P = 0.001). In contrast, the distribution in the right hemisphere in the SZ-F group differed significantly from that observed in SZ-M (χ2 = 6.017, P = 0.046), but did not differ from HC (χ2 = 2.557, P = 0.110).

Conclusion:  OFC sulcogyral pattern is altered in SZ-M but not in SZ-F, possibly reflecting gender differences in early neurodevelopment.

THERE ARE A number of reports emphasizing gender differences in brain structure as well as their neurobehavioral function in healthy subjects and in patients with schizophrenia (SZ). Concerning brain morphometry in healthy subjects, gender differences are observed already in the neonatal brain1 and these differences remain throughout life within the brain structure, function and chemistry.2 This includes the sulcogyral pattern, which is the characteristic morphological pattern of sulcus and gyrus in the brain structure of healthy adults.3,4

Various studies have illustrated gender differences in SZ,5 emphasizing that male subjects are likely to develop SZ at an earlier age, and that male patients present more externalized behavioral problems and lower social function. Estrogen has been hypothesized to be a neuroprotective agent,6–9 which may play a role in gender differences.

Neurodevelopmental aspects of brain morphology could be crucial for such gender differences in SZ. Fissurization of the cortex is considered to be a product of gross mechanical processes related to cortical growth and local cytoarchitectural characteristics.3 From a human ontogeny perspective, 95% of gyrification occurs during the second–third trimester of gestation.10 Sulcogyral variability is also significantly related to genetic factors11 and neurodevelopment, including neuronal migration, local neuronal connection, synaptic development, lamination and the formation of cytoarchitecture.12,13 Thus, it is not likely that environmental effects change sulcogyral variability, where the latter may affect volume changes in brain tissue.14

The orbitofrontal cortex (OFC) is a ventral surface component of the prefrontal cortex. Chiavaras and Petrides reported on the structural characteristics of human OFC that divide the orbitofrontal sulci into four divisions, including olfactory, medial, lateral and transverse orbital sulci (MOS, LOS and TOS), and five gyri, which were identified on magnetic resonance imaging (MRI).15 These investigators further classified the OFC sulcus into three types (I, II and III) by identifying mutual continuity among MOS, LOS and TOS (Fig. 1).

Figure 1.

Classification of the orbitofrontal cortex sulcogyral pattern on magnetic resonance imaging: (a) type I; (b) type II; (c) type III.

The OFC is extensively connected to various parts of the brain. The orbital network, involving most of the areas on the orbital surface, receives and integrates sensory information from several modalities. More specifically, this network is associated with sensation and with the appreciation and anticipation of reward, the latter as evidenced from anatomical and physiological data.12 The OFC, together with the medial frontal cortex, has also been suggested as a viscero-motor and emoto-motor system that could modulate visceral activity in response to affective stimuli.16 By the various connections to other areas, the OFC is also known to play an important role in decision making by integrating sensory and reward information through receiving information from all sensory modalities;17,18 it also involves human social functions.19

Recent neurobiological studies indicate that SZ may be a neurodevelopmental and progressive disorder with multiple biochemical abnormalities.20,21 Nakamura et al. reported that the prevalence of type III sulcogyral pattern of the OFC, the rarest type, was increased in SZ patients, and that the prevalence of type I, the most frequent type, was decreased in patients relative to healthy subjects.22 Patients with type III expression in any hemisphere also had poorer socioeconomic status (SES), poorer cognitive function and more severe symptoms and impulsivity, while patients with type I had better cognitive function. This finding was interpreted as indicating that the type III expression could be a part of a systematic neurodevelopmental alteration. The majority of the SZ subjects in the Nakamura et al. study, however, were male. Because there are various gender differences in SZ, there is a need to investigate gender differences in the OFC sulcogyral pattern.

The aim of the present study was to evaluate OFC sulcogyral pattern between male and female SZ patients, and male and female healthy controls, to determine whether or not there are gender-related morphological differences in OFC.

METHODS

Subjects

Forty-seven patients with chronic SZ were recruited from the Department of Psychiatry at Yokohama City University Hospital. Forty-seven group-, age- and gender-matched healthy control subjects (HC) participated in the study. Table 1 lists the subject demographic data.

Table 1.  Subject demographic data
 Healthy subjects (n = 47) (Mean ± SD)Schizophrenia patients (n = 47) (Mean ± SD)d.f.StatisticsP
  • t-test;

  • χ2 test.

  • GAF, Global Assessment of Functioning.

Age (years)34.7 ± 10.134.2 ± 10.11,920.20.839
M/F17/3023/2411.570.211
Socioeconomic status     
Subject's own (n = 38/40)1.9 ± 0.83.5 ± 1.11,72−6.69<0.001
Parental (n = 37/39)2.1 ± 0.72.5 ± 0.71,74−2.29<0.05
GAF (n = 33/44)88.7 ± 6.253 ± 171,5712.87<0.001

All patients were diagnosed with SZ based on DSM-IV criteria, using information obtained from the Structured Clinical Interview for DSM-IV Axis I Disorders, Clinician Version, fourth edition (SCID-I)23 by two trained mental health professionals (Y.H. and T.A.). All patients were receiving antipsychotic medication, with a mean daily dose equivalent to 463.3 mg of chlorpromazine (typical antipsychotic, 39.4%; atypical antipsychotic, 90.9%; both, 30.3%). All subjects met the following criteria: age 19–55 years, right-handed, no history of seizures, no head trauma with loss of consciousness, no neurological disorders, and no lifetime history of substance abuse. The positive and negative symptom scale (PANSS) was used to evaluate patient symptoms.

For the HC, the SCID (Edition for Non-Patients), and the Mini-International Neuropsychiatric Interview were used to ascertain that the subjects had no axis I disorders. In addition, none of their first-degree relatives had axis I disorders. The SES of all subjects and their parents was assessed using the method developed by Hollingshead.24 For cognitive associations, Wechsler Adult Intelligence Scale–Revised (WAIS-R; full-scale IQ) was also evaluated.

Male schizophrenia patients (SZ-M) and female schizophrenia patients (SZ-F) were group matched for age (mean, M/F, 32.3/36.1 years, P = 0.195), subject's own SES (M/F, 3.7/3.3, P = 0.279), parental SES (M/F, 2.6/2.5, P = 0.573), Global Assessment of Functioning (GAF; M/F, 52.4/53.5, P = 0.835), total IQ of WAIS-R (M/F, 92.6/83.3, P = 0.144), PANSS (M/F, 68.0/61.4, P = 0.29), and age of onset (M/F, 22.3/24.8 years, P = 0.197). Healthy male controls (HC-M) and healthy female controls (HC-F) were matched in age (mean, M/F, 33.2/35.5 years, P = 0.453) and GAF (M/F, 89.1/88.5, P = 0.79). Subject's own SES (M/F, 1.3/2.2, P < 0.005), parental SES (M/F, 2/2.2, P = 0.432) and total IQ (M/F, 123.5/104.8, P < 0.005) were significantly higher in HC-M than in HC-F.

The present study was approved by the Institutional Review Board and Ethics Committee of Yokohama City University, and was performed after obtaining written informed consent from all subjects.

Magnetic resonance imaging

Magnetic resonance imaging was done with a 1.5-T Magnetom Symphony (Siemens Medical System, Erlangen, Germany) at Yokohama City University Hospital. A series of 128 contiguous T1-weighted slices in the sagittal plane were acquired using a Turbo FLASH sequence with the following parameters: echo time, 3.93 ms; repetition time, 1960 ms; inversion time, 1100 ms; flip angle, 15°; field of view, 24 cm; matrix, 256 × 256 × 128; voxel dimensions, 0.9375 × 0.9375 × 1.5 mm. Based on an evaluation by a clinical neuroradiologist, no gross abnormalities were found in the scans. In order to provide reliable classification of the OFC sulcogyral pattern, all images were realigned consistently using the 3-D slicer (http://www.slicer.org/).

Sulcogyral pattern

The sulcogyral pattern was analyzed using the same aforementioned software package for medical image analysis (3D-slicer) to classify the OFC. The MOS, LOS and TOS were used to classify the H-shaped pattern into three types as Chiavaras and Petrides described in 2000.15 All brain hemispheres were analyzed using the axial image from the ventral end of OFC in order to assign the three types (I–III). Type I was defined as rostral and caudal portion of MOS interrupted in continuity, and the rostral and caudal portion of LOS continued at the level of horizontally oriented transverse orbitofrontal sulcus (Fig. 1a). Type II was defined as rostral and caudal MOS and LOS connected to each other (Fig. 1b). Type III was defined as rostral and caudal portion of both MOS and LOS interrupted (Fig. 1c). The analysis is described in detail by Nakamura et al.22 The sulcogyral pattern in each hemisphere of the 94 subjects was determined by K.U. and M.N., who were blinded to subject group. For assessment of interrater reliability, they independently evaluated the sulcogyral pattern for 10 random cases. The intraclass correlation coefficients (Cronbach's α) were 0.902 for left hemisphere and 0.947 for right hemisphere. For assessment of intra-rater reliability, five subjects were rated twice and blinded to diagnosis. Intra-rater reliability was 1.00 for both right and left hemisphere.

Statistical analysis

Independent sample t-tests were performed to assess demographic differences between the groups, including age, subjects' own SES, parental SES and WAIS-R.

The χ2 goodness-of-fit test was applied to compare the OFC sulcogyral pattern distribution with the previous studies in each group (SZ/HC, 47/47) and in each gender (SZ, MF, 23/24; HC, M/F, 17/30). Alpha level was set at 0.05 for both χ2 tests and categorical regression analyses in the present study. Data from the original paper by Nakamura et al.22 were used for the expected distribution of sulcogyral pattern for HC, and in SZ, respectively. A χ2 test was used to evaluate independence on left–right distribution within each group and for each gender. Post-hoc χ2 analysis was also carried out to identify which sulcus type was responsible for significant difference between groups.

In order to examine the relationship between sulcogyral pattern and clinical features, categorical regression analyses were applied. Subjects were classified according to sulcogyral type (e.g. subjects with type I in either or both hemispheres vs subjects without type I in both hemispheres), and these three nominal variables (types I–III) were entered as independent variables in a single model of categorical regression with each of the clinical measures entered as a dependent variable within each group.

Alpha level was set at 0.05 for both χ2 tests and categorical regression analyses in the present study. It should be noted that contributions of all three sulcogyral patterns to the variance in each dependent variable (ordinal or interval variable) were tested in a single model of categorical regression, rather than multiple unvaried comparisons, in order to reduce the risk of false positives.

RESULTS

Age and gender were statistically matched between HC and SZ patients. The average scores of SES and GAF were significantly lower in the SZ groups. The average score of SES for SZ patients, and the average score for SZ parental group was also significantly different (P = 0.025) from that of HC.

The OFC sulcogyral pattern distribution in the HC (M/F, 17/30) for types I : II : III was 22:16:9 (46.8%:34.0%:19.1%) in the left hemisphere, and 26:16:5 (55.3%:34.0%:10.6%) in the right hemisphere. Of note, even given differences in terms of racial background, age range, and gender frequency, the distribution was not significantly different from that of the previous study conducted by Chiavaras and Petrides (48%:34%:18% in the left, 64%:26%:10% in the right; left, χ2 = 0.05, P = 0.976; right, χ2 = 1.74, P = 0.419), or from the Nakamura et al. study (46%:36%:18% in the left, 62%:28%:10% in the right; left, χ2 = 0.02, P = 0.989; right, χ2 = 0.97, P = 0.616).15,22 From the χ2 test of independence, significant asymmetry was not observed for right–left (male, χ2 = 1.074, P = 0.585; female, χ2 = 1.216, P = 0.544) or male–female in HC (χ2 = 1.32, P = 0.517).

The distribution of OFC sulcogyral pattern in SZ (M/F, 23/24) for types I:II:III was 25:15:7 in the left hemisphere and 28:12:7 in the right hemisphere The distribution in the present study was not significantly different in the left hemisphere, but was significantly different in the right hemisphere as compared to the Nakamura et al. study (40%:34%:26% in the left, 42%:34%:24% in the right; left, χ2 = 4.33, P = 0.114; right, χ2 = 6.07, P = 0.048).22 Because the present sample had a different gender distribution to that of the Nakamura et al. sample (M/F, 45/5), we further analyzed the distribution within each gender.

As shown in Fig. 2, the distribution of OFC sulcogyral pattern in 23 SZ-M for type I : II : III was 12:7:4 in the left hemisphere and 10:7:6 in the right hemisphere, which was not statistically different from the distribution in the Nakamura et al. study (left, χ2 = 1.59, P = 0.451; right, χ2 = 0.14, P = 0.933).22 The distribution of OFC sulcogyral pattern in 24 SZ-F for type I : II : III was 13:8:3 in the left hemisphere and 18:5:1 in the right hemisphere. The distribution in the right hemisphere was significantly different (χ2 = 11.38, P = 0.003) from that previously reported by Nakamura et al.22 In the present data set, goodness of fit was applied and the distribution in SZ-F for the right hemisphere differed significantly from that observed in SZ-M (χ2 = 6.017, P = 0.046), but did not differ from HC (χ2 = 2.557, P = 0.110). A post-hoc test was conducted and the prevalence of type I in SZ-F was significantly higher (χ2 = 10.73, P = 0.001) and that of type III was significantly lower than that in the Nakamura et al. report (χ2 = 5.18, P = 0.023).

Figure 2.

Distribution of orbitofrontal cortex sulcogyral pattern.

Within the SZ-M group, type I or II sulcogyral pattern in any hemisphere was associated with milder clinical symptom evaluated by total PANSS score (type I, β = −0.775, F = 11.24, P = 0.004; type II, β = −0.631, F = 13.22, P = 0.002), type III sulcogyral pattern was associated with more severe clinical symptoms (β = 0.902, F = 14.75, P = 0.001). Within the SZ-F group, there was no association between sulcogyral pattern and total PANSS score.

DISCUSSION

In the present study the distribution of the OFC sulcogyral pattern in Japanese HC was almost identical to a previous study reported by Chiavaras and Petrides,15 and Nakamura et al.22 Specifically, consistent with the previous reports, the present HC showed no gender differences in either the left or right hemisphere regarding the OFC sulcogyral pattern distribution.

As compared to the HC, a higher number of patients in the SZ-M group had the type III OFC sulcogyral pattern, and a lower number had type I in the right hemisphere. In contrast, the distribution of the OFC sulcogyral pattern in the SZ-F group was not significantly different from the HC, which suggests that the OFC sulcogyral pattern distribution might be altered in male SZ patients but not in female SZ patients.

Within the SZ group, SZ-M patients with type III also had higher total PANSS score, in comparison to those without type III. With regard to OFC sulcogyral pattern in the SZ-F group, however, there were fewer type III and more type I in comparison to SZ-M, indicating a remarkable gender difference within the group.

Because the distribution of SZ subjects in the Nakamura et al. study was 90% male and 10% female, the results could reflect OFC sulcogyral pattern alteration in male SZ patients, but were inconclusive with respect to female SZ patients. Thus, the gender difference observed in the present study could be a unique finding complementing the previous findings. Nakamura et al. also demonstrated that the type III expression was associated with more severe symptoms, poorer SES, poorer cognitive function, and impulsivity compared to SZ patients without type III expression. In the present study, clinical correlations with PANSS scores in the male SZ patients also indicated that type I and II expressions were associated with milder clinical pictures and that, in contrast, type III expression was associated with a more severe clinical picture, replicating the previous study.22

Schizophrenia is known to involve gender differences in neuroanatomical features25–28 and in clinical characteristics.8,29 Neuroanatomically, male SZ patients have larger ventricles and smaller overall frontal and temporal lobe volumes. Female SZ patients are less likely to show morphological change than male SZ patients, particularly in the prefrontal cortex.26 In the left temporal lobe, female SZ patients showed no significant difference compared to HC while male SZ patients had significant volume reduction compared to HC.25 Also, in comparison to female SZ patients, male SZ patients had a higher incidence of large cavum septi persidi, which is considered to reflect gender difference in brain neurodevelopment.30 Gender differences, such as lower gray matter volume or larger ventricular volume in male SZ patients, may more likely reflect alterations mainly after the onset of illness, while large cavum septi persidi or left-smaller-than-right asymmetry in temporal lobe gray matter volume of male SZ patients may more likely reflect alterations during neurodevelopment. Thus one speculation is that gender differences between SZ patients might be traced back to the neurodevelopmental stage.

Although the alteration of OFC sulcogyral pattern has been reported in SZ patients, gender differences are not well illustrated in the previous study because the majority/all of the subjects were male.22,31 The present study is the first to show that OFC sulcogyral pattern exhibits a gender difference in SZ patients, which may reflect gender difference in neurodevelopmental aberrations.

There is a report indicating that neuroanatomical gender differences are present at the neurodevelopmental stage in healthy populations. Specifically, Yucel et al. identified three patterns of anterior cingulate (AC) sulcus depending on the presence of paracingulate sulcus (PCS) and its anteroposterior extent. They found that male subjects had a hemispheric asymmetric pattern of AC while female subjects had greater hemispheric symmetry in normal volunteers.3 They also reported that male SZ patients had a different distribution of PCS pattern than that of HC.32 To the best of our knowledge, however, the PCS pattern alteration has never been reported in female SZ patients. These facts suggest that sulcogyral alteration in SZ within the frontal paralimbic region might be specific to the male gender.

Why do only male SZ patients show altered distribution in the sulcogyral pattern? One speculation is that the neuroprotective effect of estrogen could prevent female SZ patients from exhibiting an altered distribution of sulcogyral pattern in the frontal paralimbic region during early neurodevelopment. Several reports show that estrogen may cause gender differences in brain morphology, in terms of its neuroprotective effect and differential concentration of estrogen receptors in sexually dimorphic brain regions.29,33 The study indicates that cortical brain regions, including the OFC, that demonstrate gender differences are likely to also have higher estradiol-17β receptor concentrations.

Future studies with larger samples are needed to confirm this gender difference finding for the OFC sulcogyral pattern. In the present subjects, parental SES scored lower in the SZ patient group in comparison to HC. Although the present SZ and HC groups did not demonstrate a significant relation between parental SES and sulcogyral pattern, larger samples of parental SES-matched groups are needed.

The present study investigated the distribution of OFC sulcogyral pattern in each gender in a Japanese population of male and female HC, as well as its possible alteration in a group of male and female SZ patients. The present study is the first to report gender differences in OFC sulcogyral pattern in the SZ patient group. Specifically, the OFC sulcogyral pattern is altered in SZ-M, but not in SZ-F, possibly reflecting gender differences in early neurodevelopment such as the neurotrophic effects of estrogen. The type III OFC sulcogyral pattern, which differentiated SZ-M from HC-M, was also associated with higher total PANSS score in SZ-M.

Ancillary