- Email: [email protected]
Cortical gray matter volume deficits in schizophrenia: a replication
SCHIZOPHRENIA RESEARCH ELSEVIER
Schizophrenia Research 20 (1996) 157-164
Cortical gray matter volume deficits in schizophrenia: a replication Kelvin O. Lim a,b,., Edith V. Sullivan a,b, Robert B. Zipursky c, Adolf Pfefferbaum a,b a Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA b Psychiatry Service, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA, USA c Department of Psychiatry and Behavioral Sciences, University of Toronto, Toronto, Canada Received 27 April 1995; revision 27 July 1995; accepted 7 August 1995
Abstract
We sought to replicate an earlier finding of widespread deficit in cortical gray matter in schizophrenia by testing new samples of 22 schizophrenic patients and 27 controls between the ages of 21-46 years. Brain values for both patients and controls were standardized against age and head size norms derived from a larger control group (n = 73) spanning a wider age range (21-70). Compared to the new age-matched controls, the new schizophrenic sample showed a deficit in gray matter volume affecting the cortex as a whole and enlargement of the lateral and third ventricles. Thus, widespread cortical gray matter deficit is a replicable feature of the brain dysmorphology of schizophrenia in young to middle-aged men. Keywords: Schizophrenia; Cortical gray matter; Ventricles; Brain; Replication; MRI
I. Introduction
Ever since the first computerized tomographic (CT) report of lateral ventricular enlargement in schizophrenia (Johnstone et al., 1976), a large number of studies of this phenomenon have been published (for review see (Marsh et al., 1995; Raz and Raz, 1990). While measurement differences, patient and control sampling, and the relatively small samples used in some studies have contributed to some contradictory reports, there is now a general consensus that both ventricular and sulcal enlargement are c o m m o n features of m a n y samples of schizophrenic patients. The advent of * Corresponding author. Psychiatry Service, (116A3), Veterans Affairs Palo Alto Health Care System, 3801 Miranda Ave., Palo Alto, CA 94304, USA. Tel.: 415 858 3914; Fax: 415 493 4901. 0920-9964/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0920-9964(95)00081-X
magnetic resonance imaging ( M R I ) has enabled a more detailed assessment of tissue changes associated with ventricular and sulcal enlargement. Using this technology, a widespread deficit of cortical gray matter but sparing of white matter was first reported in a study comparing 22 males with schizophrenia and 20 age-matched controls (Zipursky et al., 1992). Since this report, generalized cortical gray matter deficits have been reported in chronic (Harvey et al., 1993; Lim et al., 1995b) and first episode (Keshavan et al., 1994; Lira et al., 1994) schizophrenics. However, a lack of (Corey-Bloom et al., 1995) or more regionalized (Schlaepfer et al., 1994; Suddath et al., 1989; Suddath et al., 1990) gray matter deficit has also been reported, as have reductions in frontal white matter volume (Breier et al., 1992). Because of technical difficulties in segmenting gray matter and
158
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 15~164
white matter, many investigators have concentrated on fluid-tissue differentiation, and thus have not determined whether enlargement of CSF-filled spaces occurs at the expense of gray matter or white matter. As a consequence, cortical gray or white matter deficits, especially over broad cortical areas, have not been widely assessed in studies of schizophrenic patients. We here report a study, based on new samples of healthy control men and of men with schizophrenia, designed to replicate our earlier report. We predicted that the schizophrenic group would show a deficit in cortical gray matter volume and enlargement of sulcal and ventricular volumes relative to the control group. Further we sought to replicate the association previously observed between overall gray matter and negative symptoms, and the trend for an association between negative symptoms and frontal gray matter volume (Zipursky et al., 1992).
2. Materials and methods
2.1. Subjects 2.1.1. Patients with schizophrenia Patients (n=22, a g e = 3 5 + 4 years) were recruited for this study from the inpatient psychiatric wards of the DVA Medical Center, Palo Alto, California. All were male veterans of the United States armed services, meeting DSM-III-R Criteria for the Diagnosis of Schizophrenia and gave written informed consent to participate in this study. Recruitment procedures and selection criteria for this new cohort of subjects followed the general procedures employed for the earlier study (Zipursky et al., 1992). All were inpatients on an unlocked voluntary ward at the time of scan. Patients were excluded from the study if they met criteria for DSM-III-R Alcohol or Substance Abuse in the past 3 months, or had ever met criteria for DSM-III-R Alcohol or Substance Dependence. Other exclusion factors were a history of significant medical illness or head injury resulting in loss of consciousness for greater than 30 min. DSM-III-R diagnoses were determined by consensus between a psychiatrist, clinical psycholo-
gist or psychiatric research fellow who conducted a clinical interview and a trained research assistant who administered the Structured Clinical Interview for Diagnosis (SCID) (Spitzer et al., 1992) (n= 19) or the Schedule for Affective Disorders and Schizophrenia (SADS) (Endicott and Spitzer, 1978) (n=3). Of the 22 patients, 20 had met criteria for more than 2 years. Sub-type diagnostic breakdown for these patients was as follows: undifferentiated, 13; paranoid, 7; disorganized, 2. All but two patients were being treated with antipsychotic medications at the time of scan. Their clinical condition was evaluated using the 18-item Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1988) administered by two raters with established reliability. The average score of BPRS ratings performed closest to the MRI (mean interval = 3 days, range, 0-9 days) was used for this analysis. A negative symptom score was derived by summing scores for the following BPRS items: blunted affect, emotional withdrawal, and motor retardation. A positive symptom score was derived by summing scores for the following BPRS items: hallucinatory behavior, unusual thought content, and conceptual disorganization (Faustman, 1994). Premorbid intelligence was assessed using the NART (Nelson, 1982; O'Carroll et al., 1992) and handedness using a quantitative measure (Crovitz and Zener, 1962). Unlike the original sample, which included only right-handed patients and controls, this sample included three subjects who were not right handed (i.e., scored over 50 on the handedness test)
2.1.2. Healthy control subjects The control subjects used in this study (n =27, age, 33+7.26 years) were drawn from a larger group of health men (n = 73, age 44.1 + 13.8 years), recruited prospectively from the community to participate in a study of normal aging, as well as to serve as controls for neuropsychiatric studies (Lim et al., 1995a,b; Mathalon et al., 1993a; Pfefferbaum et al., 1992, 1994; Sullivan et al., 1995a; Zipursky et al., 1992). MRI data from a subset of 27 men (aged 24 42), matched in age to the schizophrenic group, and not already included in the previous study (Zipursky et al., 1992), were used for this particular analysis. Demographic and
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
clinical details of the replication samples of patients and controls are presented in Table 1. While controls in the original study were all righthanded veterans, only 11 of the controls in this sample were veterans and five were left handed (i.e., scored over 50 on the handedness test).
2.2. Procedure 2.2.1. M R l acquisition All subjects were scanned using 1.5T General Electric Signa MRI scanners and the same acquisition parameters and procedures previously described in detail and used in the earlier study (Lim and Pfefferbaum, 1989; Pfefferbaum et al., 1994; Zipursky et al., 1992). Briefly, axial MRI sections (5 mm thick, 2.5 mm skip) were acquired on an oblique plane, perpendicular to the sagittal plane, and passing through the anterior and posterior commissures, using a spin-echo sequence with a field of view of 24 cm and a 256 x 256 matrix, TR of > 2400 ms, TE values of 20 and 80 ms. For each data set, the most inferior section above the level of the orbits, where the anterior horns of the lateral ventricles could be seen bilaterally, was identified as an index section. Seven consecutive sections, beginning at index section and proceeding superiorly, were analyzed for each subject. The third ventricle was measured at either the index section or the section below it, wherever it was
159
larger. This was the case for nine of the schizophrenic patients and 14 of the normal controls. Head size was estimated by modeling the intracranial volume as a sphere, the diameter of which was the head height, measured from a coronal section. Full details of this technique and its reliability are provided in an earlier report (Mathalon et al., 1993a). Each of the MRI sections was segmented into cerebrospinal fluid (CSF), gray matter and white matter compartments, using a semi-automated image analysis technique (Lim and Pfefferbaum, 1989). Each section was also divided into an inner 55% region (to facilitate quantification of central CSF, in particular the lateral and third ventricles) and an outer 45% (to facilitate quantification of the cortical tissue volumes and sulcal CSF) (Pfefferbaum et al., 1986). Each section was further subdivided into four quadrants. These quadrants, which were determined according to internal anatomical landmarks and a priori geometric rules, enabled reliable estimation of volume in six geometrically defined regions of interest (ROIs), which corresponded roughly to the cortical regions after which they are named (Zipursky et al., 1992) (Fig. 1).
2.3. Statistical analysis Pixel counts for gray matter, white matter, and CSF were transformed into cubic centimeters (cc)
Table 1 Schizophrenic and normal control subjects demographic and clinical features
Age at scan (years) Education (years) NART-IQ SZ = 16, NC=23 Handedness score SZ = 20, NC = 23 Lifetime alcohol consumption (kg) SZ=21, NC=27 BPRS total BPRS negative symptoms BPRS positive symptoms Length of illness (years) Age of onset (years)
Schizophrenics, mean___SD (range)
Normal controls, mean + SD (range)
t-test (2-tailed)
35.0+4.0 (24-45) 13.2+1.6 (10 16) 104+9.4 (89-124)
32.9+7.3 (21-46) 16.4+3.0 (11.5 23) 113+5.9 (100-123)
t = 1.1, ns t=4.49, p=0.0001 t=3.64, p=0.0004
25+ 18 (14-17)
32.5__ 17.5 (14-68)
t = 1.34, ns
59___78 (0-309)
28.9+38 (0-172.32)
t = 1.73, p=0.05
---
--
42.4+6.5 (27.5-52)
7.8+2.2 (3-11.5) 8.6___3.3 (3-17) 12.5+6.4 (1-22) 23+5.6 (15-38)
160
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
:$ :$ !$ Fig. 1. Seven axial MRI sections segmented into CSF (black), gray matter (dark gray) and white matter (light gray). White lines mark the geometric division of each section into inner 55% and outer 45% and delineate the four quadrants. The six cortical ROIs are defined as follows: a, prefrontal; b, frontal; c, temporal-parietal; d, frontal-temporal; e, parietal; and f, parietal-occipital.
to provide estimates of the absolute volume encompassed by cortical gray matter, cortical white matter, sulcal CSF and the lateral and third ventricles. These 'raw' scores were corrected for normal variation associated with differences in head size and age using a two-step regression analysis based on values from the 73 normal comparison subjects. This two-step process, described previously (Mathalon et al., 1993a; Pfefferbaum et al., 1992), which takes into account variation attributable to age and head size in normal controls, yields a Zscore. For the total group of 73 control subjects, the expected mean Z-score is 0 with an S.D. of 1. It is expected that mean Z-scores will likely be close to 0 for subgroups of the entire sample. For patients, Z-scores will deviate from 0 to the extent that the volume of their brain structures deviate from that seen in the normal control subjects of similar head size and age (Mathalon et al., 1993b). This analysis differed from the analysis used in the original paper in several ways. First, in the present paper, we estimated head size by modeling the intracranial volume as a sphere (Mathalon et al., 1993a), rather than using an estimate of intracranial volume calculated by summing pixels in a limited number of axial sections. Second, we used regression analysis to correct CSF and gray matter volumes for normal variations in head size rather than expressing these volumes as a percentage of estimated head size. Finally, we modeled age-related changes in brain volumes with a sample of 73 control men spanning most of the adult age range, rather than using a small sample (n=20) with a restricted age range to establish normative values (Pfefferbaum et al., 1994).
Based on our previous report (Zipursky et al., 1992), and reports of other investigators (Harvey et al., 1993; Keshavan et al., 1994; Lim et al., 1995a; Schlaepfer et al., 1994), we predicted that the schizophrenic group would have smaller cortical gray matter volumes in addition to the more widely reported increases in ventricular volumes than the control group. We made no predictions concerning volumes of cortical white matter, for which no consistent pattern of difference has been described in the literature. Predicted relationships between clinical and demographic variables and brain variables among the schizophrenic patients were tested using Pearson product-moment correlations.
3. Results 3.1. Brain M R I measures
For measures of tissue, lower Z-scores signify less volume and higher Z-scores more volume than would be expected for a given head size and age. Thus, low Z-scores for gray matter and high Zscores for CSF measures signify predicted volume abnormalities. Figs 2 and 3 plot the Z-scores for volumes of cortical gray matter, cortical white matter, sulcal CSF and the lateral and third ventricles for each patient and control. Table 2 lists corresponding mean Z-scores for these global measures. The results of the group comparisons supported the study predictions. Specifically, relative to the control group, the schizophrenic group had a
161
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
3.
Cortical Gray Matter
21
3
o o
Corllcal While Matter
5
Cortical Sulcai CSF
2 O
~o .11 .21 .31 -4'
~2
rk. 1 O
rk o -1
-2
NC>SZ,p<.001
1
-3
Controls
Schizophrenics
-4
!
Schizophrenics -3
c~mmls
Fig. 2. Head size and age-corrected Z-scores for cortical gray matter, cortical white matter, and cortical CSF for 22 schizophrenic patiems and 27 control subjects. Lateral Ventricles 3 m
.o2 8
4 0 o
o
Third Ventricle
3 to
o
E
82
U)
-I-
o o
!
-2
8 controls
8 o
II
-1
8
o
O3
e
O
.1 ¸
NC
@ -2
Schlz:phrenlcs
~
~,.~1
Controls
Schlzoplmmlcs
Fig. 3. Head size and age-corrected Z-scores for lateral ventricles and third ventricles for 22 schizophrenic patients and 27 control subjects. Table 2 CSF, gray matter and white matter Z-scores of schizophrenic patients and age-matched control subjects Regions of interest
Total cortical Prefrontal Frontal Frontal-temporal Temporal-parietal Parietal Parietal occipital Lateral ventricles Third ventricles
CSF (mean _+SD)
Gray matter (mean -+SD)
White matter (mean + SD)
Schizophrenic patients
Controls
Schizophrenic patients
Controls
Schizophrenic Controls patients
+0.60+1.2" +0.83+1.30"* +0.78+1.00"* 1.10+1.13"** +0.63+1.27" +0.44+0.99 -0.05_+1.12 +0.84_+0.85*** +0.97-+1.05"**
-0.05+1.02 -0.10+0.92 -0.05+1.02 -0.07+0.90 -0.12+1.20 -0.07+0.94 +0.04_+1.01 -0.12_+0.97 -0.16+1.00
-1.18+0.93"** -1.15+0.87"** -1.16+1.33"* -1.40+1.65"* -1.18+1.31"* -0.80+1.21" -0.46+0.84
-0.11+1.03 -0.05+1.10 -0.14+1.10 -0.09+1.13 -0.21+1.04 -0.10+1.03 -0.11-+0.97
+0.56+0.96* +0.08+1.13 +0.15+1.21 -0.15+1.01 +0.12+1.03 +0.12+1.21 +0.71+0.97"
.
.
.
-0.09+1.03 +0.002+1.02 -0.15+1.13 -0.15_+0.91 -0.07_+1.08 -0.11 _+1.08 -0.001-+0.90
.
Group differences are based on 2-tailed t-tests: *p < 0.05; **p < 0.01; ***p< 0.001. significant cortical gray m a t t e r v o l u m e deficit (t(47) = 3.77; p < 0.0005) a n d e n l a r g e m e n t o f lateral ventricles ( t ( 4 7 ) = 3.64; p < 0 . 0 0 0 7 ) , a n d third ventricle (t(47) = 3.85; p < 0.0004). Cortical white matter, for which n o significant difference was observed in the earlier study, was unexpectedly
greater in schizophrenics t h a n controls ( t ( 4 7 ) = 2.26, p < 0 . 0 5 , two-tailed). The cortical gray matter, lateral a n d third ventricle differences are at the 0.01 level needed for significance after applying a B o n f e r r o n i correction for this family of five global measures.
162
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 157-164
This pattern of results, based on age and head size corrected Z-scores, persisted even when raw volume measures (cc) of cortical gray matter and lateral and third ventricles were analyzed. Raw cortical white matter volume, however, did not differentiate the groups (Table 3). Gray matter deficit was also assessed in each of the six cortical ROIs (Table 2). Predicted group differences were found in gray matter and CSF Zscores for each cortical ROI, and were statistically significant for four of the six ROIs. The exceptions were the posterior (parietal and parietal-occipital regions). Gray matter volume differences for the four anterior ROIs are at the 0.01 level needed for significance after applying a Bonferroni correction for this group of 18 measures (6 ROIs× 3 (two tissue types and CSF)). 3.2. C l i n i c a l m e a s u r e s
No significant correlations were present between brain Z-scores and general demographic (age of onset, length of illness, years of education, lifetime alcohol consumption, handedness, NART, or overall clinical (BPRS total)) features of the patients. We sought to replicate our original finding that negative symptoms were associated with overall gray matter deficit, and to confirm the trend linking negative symptoms with gray matter deficit in the frontal cortex. We found that negative symptoms were not significantly associated with gray matter deficit in the cortex as a whole (r = 0.13, ns) or in any individual ROI. No correlation reached significance. Of the three correlational trends for worse negative symptoms to be associated with greater gray matter deficits (frontal-temporal (p < 0.06); temporal-parietal (p < 0.06); and
parietal (p<0.08)), only one involved a frontal region. As before, positive symptoms were not significantly associated with gray matter volume deficit in any individual cortical region, or in the cortex as a whole (p values ranged from 0.14 to 0.79).
4. Discussion Our prior finding of a widespread gray matter deficit in male veterans with schizophrenia has been replicated using independent samples of veterans with schizophrenia and normal controls. For this analysis we used a broader age range (21-70 years) and larger number ( n = 7 3 ) of controls to derive age norms than used initially (20 men aged 23-45 years). To the extent that age and head size norms were derived from a sample containing 17 of the 20 original controls, this was not a completely independent replication of the original study. Group differences in global and regional cortical gray matter were significant, however, even without using age and head size norms to which the overlapping controls contributed. While this and the original study tested only veterans with schizophrenia, similar results have been obtained using the same technique with schizophrenic patients in a state hospital The lack of association between cortical gray matter deficit and either age of onset or length of illness is also consistent with our prior study of veteran patients (Zipursky et al., 1992), and also with more recent studies of state hospital patients (Lim et al., 1995a, b). Demonstrating associations between static brain morphology and less stable measures of clinical symptoms has proved difficult and conceptually
Table 3 Schizophrenicand normal control subjects (global brain volumes in cc)
Headsize Sulcal fluid Cortical gray matter Cortical white matter Lateral ventricles Third ventricle
Schizophrenics(mean± SD)
Normal controls (mean+__SD)
t-test (2-tailed)
1333+ 97 44.66+ 13.62 116.17+ 13.07 74.43+ 9.45 23.81 + 6.50 0.45+ 0.16
1363_+111 83.12_ 10.52 130.27± 14.68 72.69_ 12.70 17.77__+6.8 0.29___0.14
t= 1.19,p=0.24 t= 1.90,p <0.06 t= 3.51, p < 0.001 t = 0.84, p =0.41 t=3.16, p <0.003 t= 3.55, p<0.001
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 15~164
problematic (Marsh et al., 1995; Roth and Pfefferbaum, 1992). In this sample, we obtained only a trend for an association between overall cortical gray matter volume and negative symptoms. This association is consistent with the associations found in an overlapping sample of these subjects between cortical gray matter volume and four composite neuropsychological test scores representing executive function, short-term memory and production, declarative memory, and motor ability (Sullivan et al., 1995b). The finding of an increase in cortical white matter volume, localized to the parietal-occipital region, remains to be further evaluated using a measurement approach in which the white matter boundaries are determined anatomically rather than geometrically (Pfefferbaum et al., 1995). However, a similar increase in white matter, also localized to the parietal-occipital regions was found in the original study. Neurodevelopmental displacement of neurons in white matter bands in schizophrenics (Akbarian et al., 1993a; Akbarian et al., 1993b) may provide one model to account for this apparent white matter surplus, though in those studies only frontal and temporal lobe data were reported. While ventricular and sulcal enlargement are nonspecific findings in many neuropsychiatric disorders, our observations of a widespread deficit in cortical gray matter volume in schizophrenia may have greater specificity, since two recent reports (Harvey et al., 1994; Zipursky et al., 1994) fail to find a gray matter deficit in bipolar patients. Further work is currently underway to address the extent to which this gray matter deficit of schizophrenia is present at first break, is static or progressive over the course of the illness, whether it manifests a characteristic pattern of regional involvement, how it may be associated with illness severity and symptom profile, whether it is also found in women with schizophrenia, or in family members at risk. Proton spectroscopic measures of the concentration of N-acetylaspartate (NAA), an amino acid thought to be present almost exclusively in neurons and their dendritic and axonal extensions (Miller, 1991), could next be used to characterize the composition of cortical gray matter in schizophrenics, and assess the patho-
163
physiology of this widespread and replicable deficit.
Acknowledgment This work was supported by grants from the National Institute of Mental Health (NIMH 30854) and the Medical Research Service of the Department of Veterans Affairs.
References Akbarian, S., Bunney, W.E., Potkin, S.G., Wigal, S.B., Hagman, J.O., Sandman, C. and Jones, E.G. (1993a) Altered distribution of nicotinamid-adenine dinucleotide phosphate diaphorase cells in frontal lobe of schizophrenics implies disturbances of cortical development. Arch. Gen. Psychiatry 50, 169-177. Akbarian, S., Vinuela, A., Ki, J.J., Potkin, S.G., Bunney, W.E. and Jones, E.G. (1993b) Distorted distribution of nicotinamid-adenine dinucleotide phosphate diaphorase neurons in temporal lobe of schizophrenics implies anomalous cortical development. Arch. Gen. Psychiatry 50, 178 187. Breier, A., Buchanan, R.W., Elkashef, A., Munson, R.C., Kirkpatrick, B. and Gellad, F. (1992) Brain morphology and schizophrenia - a magnetic resonance imaging study of limbic, prefrontal cortex, and caudate structures. Arch. Gen. Psychiatry 49, 921-926. Corey-Bloom, J., Jernigan, T., Archibald, S., Harris, M.J. and Jeste, D.V. (1995) Quantitative magnetic resonance imaging of the brain in late-life schizophrenia. Am. J. Psychiatry 152, 447-449. Crovitz, H.F. and Zener, K.A. (1962) Group test for assessing hand and eye dominance. Am. J. Psychol. 75, 271-276. Endicott, J. and Spitzer, R.L. (1978) A diagnostic interview: The Schedule for Affective Disorders and Schizophrenia. Arch. Gen. Psychiatry 35, 837 844. Faustman, W.O. (1994). Brief psychiatric rating scale. In: M. Maruish (Ed.), The Use of Psychological Testing For Treatment, Planning and Outcome Assessment. Lawrence Erlbaum, Hillsdale, NJ, pp. 371-401. Harvey, I., Persaud, R., Ron, M.A., Baker, G. and Murray, R.M. (1994) Volumetric MRI measurements in bipolars compared with schizophrenics and healthy controls. Psychol. Med. 24, 689 699. Harvey, I., Ron, M., du Bouley, G., Wicks, D., Lewis, S. and Murray, R.M. (1993) Reduction of cortical volume in schizophrenia on magnetic resonance imaging. Psychol. Med. 23, 591-604. Johnstone, E.C., Crow, T.J., Frith, D.C., Husband, J. and Kreel, L. (1976) Cerebral ventricular size and cognitive impairment in schizophrenia. Lancet ii, 924 926.
164
Kelvin O. Lira et al./Schizophrenia Research 20 (1996~ 15~164
Keshavan, M.S., Pettegrew, J.W., Bagwell, W.W., Haas, G.L., Sweeney, J. and Schooler, N.R. (1994) Gray matter volume deficits in first-episode schizophrenia (Abstr.). Biol. Psychiatry 35, 713. Lim, K.O., Beal, M., Harvey, R.L. Jr., Myers, T., Lane, B., Sullivan, E.V., Faustman, W.O. and Pfefferbaum, A. (1995a) Brain dysmorphology in adults with congenital rubella plus schizophrenia-like symptoms. Biol. Psychiatry 37, 764 776. Lim, K.O., Harris, D., Beal, M., Minn, K., Hoff, A., Csernansky, J., Faustman, W., Marsh, L., Sullivan, E.V. and Pfefferbaum, A. (1995b) Gray matter volume deficits in young onset schizophrenia are independent of age of onset. Biol. Psychiatry in press, Lim, K.O. and Pfefferbaum, A. (1989) Segmentation of MR brain images into cerebrospinal fluid spaces, white and gray matter. J. Comput. Assist. Tomogr. 13, 588-593. Lim, K.O., Tew, W., Kushner, M., Chow, K., Matsumoto, B. and DeLisi, L.E. (1994) First episode schizophrenics have a deficit in cortical gray matter volume (Abstr.). Am. Coll. Neuropharmacol. Abstracts of Panels and Posters, 205. Marsh, L., Lauriello, J., Sullivan, E. and Pfefferbaum, A. (1995). Neuroimaging in neuropsychiatric disorders. In: E. Bigler (Ed.), Handbook of Human Brain Function: Neuroimaging. Plenum Press, New York, in press. Mathalon, D.H., Sullivan, E.V., Rawles, J.M. and Pfefferbaum, A. (1993a) Correction for head size in brain imaging measurements. Psychiatry Res. Neuroimaging 50, 121 139. Mathalon, D.H., Sullivan, E.V., Rawles, J.M. and Pfefferbaum, A. (1993b) Correction for head size in brain imaging measurements: Letter to editor. Psychiatry Res. Neuroimaging 50, 284. Miller, B.L. (1991) A review of chemical issues in ~H NMR spectroscopy: N-acetyl-L-aspartate, creatine and choline. NMR Biomed. 4, 47-52. Nelson, H.E. (1982) National Adult Reading Test, NART. Nelson Publishing Company, Windsor. O'Carroll, R.A., Walker, M., Dunan, J., Murray, C., Blackwood, D., Ebmeier, K. and Goodwin, G. (1992) Selecting controls for schizophrenia research studies: The use of the national adult reading test (NART) as a measure of pre-morbid ability. Schizophrenia Res. 8, 137 141. Overall, J.E. and Gorham, D.R. (1988) The Brief Psychiatric Rating Scale (BPRS): Recent developments in ascertainment and scaling. Psychopharmacol. Bull. 24, 97 99. Pfefferbaum, A., Lim, K.O., Zipursky, R.B., Mathalon, D.H., Lane, B., Ha, C.N., Rosenbloom, M.J. and Sullivan, E.V. (1992) Brain gray and white matter volume loss accelerates with aging in chronic alcoholics: A quantitative MRI study. Alcoholism Clin. Exp. Res. 16, 1078-1089.
Pfefferbaum, A., Mathalon, D.H., Sullivan, E.V., Rawles, J.M., Zipursky, R.B. and Lim, K.O. (1994) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood. Arch. Neurol. 51, 874 887. Pfefferbaum, A., Sullivan, E.V., Mathalon, D.H., Shear, P.K., Rosenbloom, M.J. and Lim, K.O. (1995) Longitudinal changes in MRI brain volumes in abstinent and relapsed alcoholics. Alcoholism Clin. Exp. Res. in press, Pfefl'erbaum, A., Zatz, L. and Jernigan, T.L. (1986) Computerinteractive method for quantifying cerebrospinal fluid and tissue in brain CT scans: effects of aging. J. Comput. Assist. Tomogr. 10, 571-578. Raz, S. and Raz, N. (1990) Structural brain abnormalities in the major psychoses: A quantitative review of the evidence from computerized imaging. Psychol. Bull. 108, 93-108. Roth, W.T. and Pfefferbaum, A. (1992) Abnormalities of the left temporal lobe in schizophrenia (letter). New Engl. J. Med. 327, 1689. Schlaepfer, T.E., Harris, G.J., Tien, A.Y., Peng, L.W., Lee, S., Federman, E.B., Chase, G.A., Barta, P.E. and Pearlson, G.D. (1994) Decreased regional cortical gray matter volume in schizophrenia. Am. J. Psychiatry 151,842-848. Spitzer, R.L., Williams, J.B.W., Gibbon, M. and First, M.B. (1992) The structured clinical interview for DSM-III-R (SCID). 1. History, rationale, and description. Arch. Gen. Psychiatry 49, 624 629. Suddath, R., Casanova, M.F., Goldberg, T.E., Daniel, D.G., Kelsoe, J.R. and Weinberger, D.R. (1989) Temporal lobe pathology in schizophrenia. Am. J. Psychiatry 146, 464-472. Suddath, R.L., Christison, G.W., Fuller Torrey, E., Casanova, M.F. and Weinberg, D.R. (1990) Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia. New Engl. J. Med. 322, 789 794. Sullivan, E.V., Marsh, L., Mathalon, D.H., Lim, K.O. and Pfefferbaum, A. (1995a) Age-related decline in MRI volumes of temporal lobe gray matter but not hippocampus. Neurobiol. Aging in press, Sullivan, E.V., Shear, P.K., Lim, K.O., Zipursky, R.B. and Pfefferbaum, A. (1995b) Cognitive and motor impairments are related to gray matter volume deficits in schizophrenia. Biol. Psychiatry submitted for publication. Zipursky, R.B., Bury, A., Langevin, R. and Seeman, M.V. (1994) Gray and white matter volumes in schizophrenia and bipolar disorders (Abstr.). Biol. Psychiatry 35, 719-720. Zipursky, R.B., Lim, K.O., Sullivan, E.V., Brown, B.W. and Pfefferbaum, A. (1992) Widespread cerebral gray matter volume deficits in schizophrenia. Arch. Gen. Psychiatry 49, 195 205.
Schizophrenia Research 20 (1996) 157-164
Cortical gray matter volume deficits in schizophrenia: a replication Kelvin O. Lim a,b,., Edith V. Sullivan a,b, Robert B. Zipursky c, Adolf Pfefferbaum a,b a Department of Psychiatry and Behavioral Sciences, Stanford University School of Medicine, Stanford, CA, USA b Psychiatry Service, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA, USA c Department of Psychiatry and Behavioral Sciences, University of Toronto, Toronto, Canada Received 27 April 1995; revision 27 July 1995; accepted 7 August 1995
Abstract
We sought to replicate an earlier finding of widespread deficit in cortical gray matter in schizophrenia by testing new samples of 22 schizophrenic patients and 27 controls between the ages of 21-46 years. Brain values for both patients and controls were standardized against age and head size norms derived from a larger control group (n = 73) spanning a wider age range (21-70). Compared to the new age-matched controls, the new schizophrenic sample showed a deficit in gray matter volume affecting the cortex as a whole and enlargement of the lateral and third ventricles. Thus, widespread cortical gray matter deficit is a replicable feature of the brain dysmorphology of schizophrenia in young to middle-aged men. Keywords: Schizophrenia; Cortical gray matter; Ventricles; Brain; Replication; MRI
I. Introduction
Ever since the first computerized tomographic (CT) report of lateral ventricular enlargement in schizophrenia (Johnstone et al., 1976), a large number of studies of this phenomenon have been published (for review see (Marsh et al., 1995; Raz and Raz, 1990). While measurement differences, patient and control sampling, and the relatively small samples used in some studies have contributed to some contradictory reports, there is now a general consensus that both ventricular and sulcal enlargement are c o m m o n features of m a n y samples of schizophrenic patients. The advent of * Corresponding author. Psychiatry Service, (116A3), Veterans Affairs Palo Alto Health Care System, 3801 Miranda Ave., Palo Alto, CA 94304, USA. Tel.: 415 858 3914; Fax: 415 493 4901. 0920-9964/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved SSDI 0920-9964(95)00081-X
magnetic resonance imaging ( M R I ) has enabled a more detailed assessment of tissue changes associated with ventricular and sulcal enlargement. Using this technology, a widespread deficit of cortical gray matter but sparing of white matter was first reported in a study comparing 22 males with schizophrenia and 20 age-matched controls (Zipursky et al., 1992). Since this report, generalized cortical gray matter deficits have been reported in chronic (Harvey et al., 1993; Lim et al., 1995b) and first episode (Keshavan et al., 1994; Lira et al., 1994) schizophrenics. However, a lack of (Corey-Bloom et al., 1995) or more regionalized (Schlaepfer et al., 1994; Suddath et al., 1989; Suddath et al., 1990) gray matter deficit has also been reported, as have reductions in frontal white matter volume (Breier et al., 1992). Because of technical difficulties in segmenting gray matter and
158
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 15~164
white matter, many investigators have concentrated on fluid-tissue differentiation, and thus have not determined whether enlargement of CSF-filled spaces occurs at the expense of gray matter or white matter. As a consequence, cortical gray or white matter deficits, especially over broad cortical areas, have not been widely assessed in studies of schizophrenic patients. We here report a study, based on new samples of healthy control men and of men with schizophrenia, designed to replicate our earlier report. We predicted that the schizophrenic group would show a deficit in cortical gray matter volume and enlargement of sulcal and ventricular volumes relative to the control group. Further we sought to replicate the association previously observed between overall gray matter and negative symptoms, and the trend for an association between negative symptoms and frontal gray matter volume (Zipursky et al., 1992).
2. Materials and methods
2.1. Subjects 2.1.1. Patients with schizophrenia Patients (n=22, a g e = 3 5 + 4 years) were recruited for this study from the inpatient psychiatric wards of the DVA Medical Center, Palo Alto, California. All were male veterans of the United States armed services, meeting DSM-III-R Criteria for the Diagnosis of Schizophrenia and gave written informed consent to participate in this study. Recruitment procedures and selection criteria for this new cohort of subjects followed the general procedures employed for the earlier study (Zipursky et al., 1992). All were inpatients on an unlocked voluntary ward at the time of scan. Patients were excluded from the study if they met criteria for DSM-III-R Alcohol or Substance Abuse in the past 3 months, or had ever met criteria for DSM-III-R Alcohol or Substance Dependence. Other exclusion factors were a history of significant medical illness or head injury resulting in loss of consciousness for greater than 30 min. DSM-III-R diagnoses were determined by consensus between a psychiatrist, clinical psycholo-
gist or psychiatric research fellow who conducted a clinical interview and a trained research assistant who administered the Structured Clinical Interview for Diagnosis (SCID) (Spitzer et al., 1992) (n= 19) or the Schedule for Affective Disorders and Schizophrenia (SADS) (Endicott and Spitzer, 1978) (n=3). Of the 22 patients, 20 had met criteria for more than 2 years. Sub-type diagnostic breakdown for these patients was as follows: undifferentiated, 13; paranoid, 7; disorganized, 2. All but two patients were being treated with antipsychotic medications at the time of scan. Their clinical condition was evaluated using the 18-item Brief Psychiatric Rating Scale (BPRS) (Overall and Gorham, 1988) administered by two raters with established reliability. The average score of BPRS ratings performed closest to the MRI (mean interval = 3 days, range, 0-9 days) was used for this analysis. A negative symptom score was derived by summing scores for the following BPRS items: blunted affect, emotional withdrawal, and motor retardation. A positive symptom score was derived by summing scores for the following BPRS items: hallucinatory behavior, unusual thought content, and conceptual disorganization (Faustman, 1994). Premorbid intelligence was assessed using the NART (Nelson, 1982; O'Carroll et al., 1992) and handedness using a quantitative measure (Crovitz and Zener, 1962). Unlike the original sample, which included only right-handed patients and controls, this sample included three subjects who were not right handed (i.e., scored over 50 on the handedness test)
2.1.2. Healthy control subjects The control subjects used in this study (n =27, age, 33+7.26 years) were drawn from a larger group of health men (n = 73, age 44.1 + 13.8 years), recruited prospectively from the community to participate in a study of normal aging, as well as to serve as controls for neuropsychiatric studies (Lim et al., 1995a,b; Mathalon et al., 1993a; Pfefferbaum et al., 1992, 1994; Sullivan et al., 1995a; Zipursky et al., 1992). MRI data from a subset of 27 men (aged 24 42), matched in age to the schizophrenic group, and not already included in the previous study (Zipursky et al., 1992), were used for this particular analysis. Demographic and
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
clinical details of the replication samples of patients and controls are presented in Table 1. While controls in the original study were all righthanded veterans, only 11 of the controls in this sample were veterans and five were left handed (i.e., scored over 50 on the handedness test).
2.2. Procedure 2.2.1. M R l acquisition All subjects were scanned using 1.5T General Electric Signa MRI scanners and the same acquisition parameters and procedures previously described in detail and used in the earlier study (Lim and Pfefferbaum, 1989; Pfefferbaum et al., 1994; Zipursky et al., 1992). Briefly, axial MRI sections (5 mm thick, 2.5 mm skip) were acquired on an oblique plane, perpendicular to the sagittal plane, and passing through the anterior and posterior commissures, using a spin-echo sequence with a field of view of 24 cm and a 256 x 256 matrix, TR of > 2400 ms, TE values of 20 and 80 ms. For each data set, the most inferior section above the level of the orbits, where the anterior horns of the lateral ventricles could be seen bilaterally, was identified as an index section. Seven consecutive sections, beginning at index section and proceeding superiorly, were analyzed for each subject. The third ventricle was measured at either the index section or the section below it, wherever it was
159
larger. This was the case for nine of the schizophrenic patients and 14 of the normal controls. Head size was estimated by modeling the intracranial volume as a sphere, the diameter of which was the head height, measured from a coronal section. Full details of this technique and its reliability are provided in an earlier report (Mathalon et al., 1993a). Each of the MRI sections was segmented into cerebrospinal fluid (CSF), gray matter and white matter compartments, using a semi-automated image analysis technique (Lim and Pfefferbaum, 1989). Each section was also divided into an inner 55% region (to facilitate quantification of central CSF, in particular the lateral and third ventricles) and an outer 45% (to facilitate quantification of the cortical tissue volumes and sulcal CSF) (Pfefferbaum et al., 1986). Each section was further subdivided into four quadrants. These quadrants, which were determined according to internal anatomical landmarks and a priori geometric rules, enabled reliable estimation of volume in six geometrically defined regions of interest (ROIs), which corresponded roughly to the cortical regions after which they are named (Zipursky et al., 1992) (Fig. 1).
2.3. Statistical analysis Pixel counts for gray matter, white matter, and CSF were transformed into cubic centimeters (cc)
Table 1 Schizophrenic and normal control subjects demographic and clinical features
Age at scan (years) Education (years) NART-IQ SZ = 16, NC=23 Handedness score SZ = 20, NC = 23 Lifetime alcohol consumption (kg) SZ=21, NC=27 BPRS total BPRS negative symptoms BPRS positive symptoms Length of illness (years) Age of onset (years)
Schizophrenics, mean___SD (range)
Normal controls, mean + SD (range)
t-test (2-tailed)
35.0+4.0 (24-45) 13.2+1.6 (10 16) 104+9.4 (89-124)
32.9+7.3 (21-46) 16.4+3.0 (11.5 23) 113+5.9 (100-123)
t = 1.1, ns t=4.49, p=0.0001 t=3.64, p=0.0004
25+ 18 (14-17)
32.5__ 17.5 (14-68)
t = 1.34, ns
59___78 (0-309)
28.9+38 (0-172.32)
t = 1.73, p=0.05
---
--
42.4+6.5 (27.5-52)
7.8+2.2 (3-11.5) 8.6___3.3 (3-17) 12.5+6.4 (1-22) 23+5.6 (15-38)
160
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
:$ :$ !$ Fig. 1. Seven axial MRI sections segmented into CSF (black), gray matter (dark gray) and white matter (light gray). White lines mark the geometric division of each section into inner 55% and outer 45% and delineate the four quadrants. The six cortical ROIs are defined as follows: a, prefrontal; b, frontal; c, temporal-parietal; d, frontal-temporal; e, parietal; and f, parietal-occipital.
to provide estimates of the absolute volume encompassed by cortical gray matter, cortical white matter, sulcal CSF and the lateral and third ventricles. These 'raw' scores were corrected for normal variation associated with differences in head size and age using a two-step regression analysis based on values from the 73 normal comparison subjects. This two-step process, described previously (Mathalon et al., 1993a; Pfefferbaum et al., 1992), which takes into account variation attributable to age and head size in normal controls, yields a Zscore. For the total group of 73 control subjects, the expected mean Z-score is 0 with an S.D. of 1. It is expected that mean Z-scores will likely be close to 0 for subgroups of the entire sample. For patients, Z-scores will deviate from 0 to the extent that the volume of their brain structures deviate from that seen in the normal control subjects of similar head size and age (Mathalon et al., 1993b). This analysis differed from the analysis used in the original paper in several ways. First, in the present paper, we estimated head size by modeling the intracranial volume as a sphere (Mathalon et al., 1993a), rather than using an estimate of intracranial volume calculated by summing pixels in a limited number of axial sections. Second, we used regression analysis to correct CSF and gray matter volumes for normal variations in head size rather than expressing these volumes as a percentage of estimated head size. Finally, we modeled age-related changes in brain volumes with a sample of 73 control men spanning most of the adult age range, rather than using a small sample (n=20) with a restricted age range to establish normative values (Pfefferbaum et al., 1994).
Based on our previous report (Zipursky et al., 1992), and reports of other investigators (Harvey et al., 1993; Keshavan et al., 1994; Lim et al., 1995a; Schlaepfer et al., 1994), we predicted that the schizophrenic group would have smaller cortical gray matter volumes in addition to the more widely reported increases in ventricular volumes than the control group. We made no predictions concerning volumes of cortical white matter, for which no consistent pattern of difference has been described in the literature. Predicted relationships between clinical and demographic variables and brain variables among the schizophrenic patients were tested using Pearson product-moment correlations.
3. Results 3.1. Brain M R I measures
For measures of tissue, lower Z-scores signify less volume and higher Z-scores more volume than would be expected for a given head size and age. Thus, low Z-scores for gray matter and high Zscores for CSF measures signify predicted volume abnormalities. Figs 2 and 3 plot the Z-scores for volumes of cortical gray matter, cortical white matter, sulcal CSF and the lateral and third ventricles for each patient and control. Table 2 lists corresponding mean Z-scores for these global measures. The results of the group comparisons supported the study predictions. Specifically, relative to the control group, the schizophrenic group had a
161
Kelvin O. Lim et al./Schizophrenia Research 20 (1996) 157-164
3.
Cortical Gray Matter
21
3
o o
Corllcal While Matter
5
Cortical Sulcai CSF
2 O
~o .11 .21 .31 -4'
~2
rk. 1 O
rk o -1
-2
NC>SZ,p<.001
1
-3
Controls
Schizophrenics
-4
!
Schizophrenics -3
c~mmls
Fig. 2. Head size and age-corrected Z-scores for cortical gray matter, cortical white matter, and cortical CSF for 22 schizophrenic patiems and 27 control subjects. Lateral Ventricles 3 m
.o2 8
4 0 o
o
Third Ventricle
3 to
o
E
82
U)
-I-
o o
!
-2
8 controls
8 o
II
-1
8
o
O3
e
O
.1 ¸
NC
@ -2
Schlz:phrenlcs
~
~,.~1
Controls
Schlzoplmmlcs
Fig. 3. Head size and age-corrected Z-scores for lateral ventricles and third ventricles for 22 schizophrenic patients and 27 control subjects. Table 2 CSF, gray matter and white matter Z-scores of schizophrenic patients and age-matched control subjects Regions of interest
Total cortical Prefrontal Frontal Frontal-temporal Temporal-parietal Parietal Parietal occipital Lateral ventricles Third ventricles
CSF (mean _+SD)
Gray matter (mean -+SD)
White matter (mean + SD)
Schizophrenic patients
Controls
Schizophrenic patients
Controls
Schizophrenic Controls patients
+0.60+1.2" +0.83+1.30"* +0.78+1.00"* 1.10+1.13"** +0.63+1.27" +0.44+0.99 -0.05_+1.12 +0.84_+0.85*** +0.97-+1.05"**
-0.05+1.02 -0.10+0.92 -0.05+1.02 -0.07+0.90 -0.12+1.20 -0.07+0.94 +0.04_+1.01 -0.12_+0.97 -0.16+1.00
-1.18+0.93"** -1.15+0.87"** -1.16+1.33"* -1.40+1.65"* -1.18+1.31"* -0.80+1.21" -0.46+0.84
-0.11+1.03 -0.05+1.10 -0.14+1.10 -0.09+1.13 -0.21+1.04 -0.10+1.03 -0.11-+0.97
+0.56+0.96* +0.08+1.13 +0.15+1.21 -0.15+1.01 +0.12+1.03 +0.12+1.21 +0.71+0.97"
.
.
.
-0.09+1.03 +0.002+1.02 -0.15+1.13 -0.15_+0.91 -0.07_+1.08 -0.11 _+1.08 -0.001-+0.90
.
Group differences are based on 2-tailed t-tests: *p < 0.05; **p < 0.01; ***p< 0.001. significant cortical gray m a t t e r v o l u m e deficit (t(47) = 3.77; p < 0.0005) a n d e n l a r g e m e n t o f lateral ventricles ( t ( 4 7 ) = 3.64; p < 0 . 0 0 0 7 ) , a n d third ventricle (t(47) = 3.85; p < 0.0004). Cortical white matter, for which n o significant difference was observed in the earlier study, was unexpectedly
greater in schizophrenics t h a n controls ( t ( 4 7 ) = 2.26, p < 0 . 0 5 , two-tailed). The cortical gray matter, lateral a n d third ventricle differences are at the 0.01 level needed for significance after applying a B o n f e r r o n i correction for this family of five global measures.
162
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 157-164
This pattern of results, based on age and head size corrected Z-scores, persisted even when raw volume measures (cc) of cortical gray matter and lateral and third ventricles were analyzed. Raw cortical white matter volume, however, did not differentiate the groups (Table 3). Gray matter deficit was also assessed in each of the six cortical ROIs (Table 2). Predicted group differences were found in gray matter and CSF Zscores for each cortical ROI, and were statistically significant for four of the six ROIs. The exceptions were the posterior (parietal and parietal-occipital regions). Gray matter volume differences for the four anterior ROIs are at the 0.01 level needed for significance after applying a Bonferroni correction for this group of 18 measures (6 ROIs× 3 (two tissue types and CSF)). 3.2. C l i n i c a l m e a s u r e s
No significant correlations were present between brain Z-scores and general demographic (age of onset, length of illness, years of education, lifetime alcohol consumption, handedness, NART, or overall clinical (BPRS total)) features of the patients. We sought to replicate our original finding that negative symptoms were associated with overall gray matter deficit, and to confirm the trend linking negative symptoms with gray matter deficit in the frontal cortex. We found that negative symptoms were not significantly associated with gray matter deficit in the cortex as a whole (r = 0.13, ns) or in any individual ROI. No correlation reached significance. Of the three correlational trends for worse negative symptoms to be associated with greater gray matter deficits (frontal-temporal (p < 0.06); temporal-parietal (p < 0.06); and
parietal (p<0.08)), only one involved a frontal region. As before, positive symptoms were not significantly associated with gray matter volume deficit in any individual cortical region, or in the cortex as a whole (p values ranged from 0.14 to 0.79).
4. Discussion Our prior finding of a widespread gray matter deficit in male veterans with schizophrenia has been replicated using independent samples of veterans with schizophrenia and normal controls. For this analysis we used a broader age range (21-70 years) and larger number ( n = 7 3 ) of controls to derive age norms than used initially (20 men aged 23-45 years). To the extent that age and head size norms were derived from a sample containing 17 of the 20 original controls, this was not a completely independent replication of the original study. Group differences in global and regional cortical gray matter were significant, however, even without using age and head size norms to which the overlapping controls contributed. While this and the original study tested only veterans with schizophrenia, similar results have been obtained using the same technique with schizophrenic patients in a state hospital The lack of association between cortical gray matter deficit and either age of onset or length of illness is also consistent with our prior study of veteran patients (Zipursky et al., 1992), and also with more recent studies of state hospital patients (Lim et al., 1995a, b). Demonstrating associations between static brain morphology and less stable measures of clinical symptoms has proved difficult and conceptually
Table 3 Schizophrenicand normal control subjects (global brain volumes in cc)
Headsize Sulcal fluid Cortical gray matter Cortical white matter Lateral ventricles Third ventricle
Schizophrenics(mean± SD)
Normal controls (mean+__SD)
t-test (2-tailed)
1333+ 97 44.66+ 13.62 116.17+ 13.07 74.43+ 9.45 23.81 + 6.50 0.45+ 0.16
1363_+111 83.12_ 10.52 130.27± 14.68 72.69_ 12.70 17.77__+6.8 0.29___0.14
t= 1.19,p=0.24 t= 1.90,p <0.06 t= 3.51, p < 0.001 t = 0.84, p =0.41 t=3.16, p <0.003 t= 3.55, p<0.001
Kelvin O. Lira et al./Schizophrenia Research 20 (1996) 15~164
problematic (Marsh et al., 1995; Roth and Pfefferbaum, 1992). In this sample, we obtained only a trend for an association between overall cortical gray matter volume and negative symptoms. This association is consistent with the associations found in an overlapping sample of these subjects between cortical gray matter volume and four composite neuropsychological test scores representing executive function, short-term memory and production, declarative memory, and motor ability (Sullivan et al., 1995b). The finding of an increase in cortical white matter volume, localized to the parietal-occipital region, remains to be further evaluated using a measurement approach in which the white matter boundaries are determined anatomically rather than geometrically (Pfefferbaum et al., 1995). However, a similar increase in white matter, also localized to the parietal-occipital regions was found in the original study. Neurodevelopmental displacement of neurons in white matter bands in schizophrenics (Akbarian et al., 1993a; Akbarian et al., 1993b) may provide one model to account for this apparent white matter surplus, though in those studies only frontal and temporal lobe data were reported. While ventricular and sulcal enlargement are nonspecific findings in many neuropsychiatric disorders, our observations of a widespread deficit in cortical gray matter volume in schizophrenia may have greater specificity, since two recent reports (Harvey et al., 1994; Zipursky et al., 1994) fail to find a gray matter deficit in bipolar patients. Further work is currently underway to address the extent to which this gray matter deficit of schizophrenia is present at first break, is static or progressive over the course of the illness, whether it manifests a characteristic pattern of regional involvement, how it may be associated with illness severity and symptom profile, whether it is also found in women with schizophrenia, or in family members at risk. Proton spectroscopic measures of the concentration of N-acetylaspartate (NAA), an amino acid thought to be present almost exclusively in neurons and their dendritic and axonal extensions (Miller, 1991), could next be used to characterize the composition of cortical gray matter in schizophrenics, and assess the patho-
163
physiology of this widespread and replicable deficit.
Acknowledgment This work was supported by grants from the National Institute of Mental Health (NIMH 30854) and the Medical Research Service of the Department of Veterans Affairs.
References Akbarian, S., Bunney, W.E., Potkin, S.G., Wigal, S.B., Hagman, J.O., Sandman, C. and Jones, E.G. (1993a) Altered distribution of nicotinamid-adenine dinucleotide phosphate diaphorase cells in frontal lobe of schizophrenics implies disturbances of cortical development. Arch. Gen. Psychiatry 50, 169-177. Akbarian, S., Vinuela, A., Ki, J.J., Potkin, S.G., Bunney, W.E. and Jones, E.G. (1993b) Distorted distribution of nicotinamid-adenine dinucleotide phosphate diaphorase neurons in temporal lobe of schizophrenics implies anomalous cortical development. Arch. Gen. Psychiatry 50, 178 187. Breier, A., Buchanan, R.W., Elkashef, A., Munson, R.C., Kirkpatrick, B. and Gellad, F. (1992) Brain morphology and schizophrenia - a magnetic resonance imaging study of limbic, prefrontal cortex, and caudate structures. Arch. Gen. Psychiatry 49, 921-926. Corey-Bloom, J., Jernigan, T., Archibald, S., Harris, M.J. and Jeste, D.V. (1995) Quantitative magnetic resonance imaging of the brain in late-life schizophrenia. Am. J. Psychiatry 152, 447-449. Crovitz, H.F. and Zener, K.A. (1962) Group test for assessing hand and eye dominance. Am. J. Psychol. 75, 271-276. Endicott, J. and Spitzer, R.L. (1978) A diagnostic interview: The Schedule for Affective Disorders and Schizophrenia. Arch. Gen. Psychiatry 35, 837 844. Faustman, W.O. (1994). Brief psychiatric rating scale. In: M. Maruish (Ed.), The Use of Psychological Testing For Treatment, Planning and Outcome Assessment. Lawrence Erlbaum, Hillsdale, NJ, pp. 371-401. Harvey, I., Persaud, R., Ron, M.A., Baker, G. and Murray, R.M. (1994) Volumetric MRI measurements in bipolars compared with schizophrenics and healthy controls. Psychol. Med. 24, 689 699. Harvey, I., Ron, M., du Bouley, G., Wicks, D., Lewis, S. and Murray, R.M. (1993) Reduction of cortical volume in schizophrenia on magnetic resonance imaging. Psychol. Med. 23, 591-604. Johnstone, E.C., Crow, T.J., Frith, D.C., Husband, J. and Kreel, L. (1976) Cerebral ventricular size and cognitive impairment in schizophrenia. Lancet ii, 924 926.
164
Kelvin O. Lira et al./Schizophrenia Research 20 (1996~ 15~164
Keshavan, M.S., Pettegrew, J.W., Bagwell, W.W., Haas, G.L., Sweeney, J. and Schooler, N.R. (1994) Gray matter volume deficits in first-episode schizophrenia (Abstr.). Biol. Psychiatry 35, 713. Lim, K.O., Beal, M., Harvey, R.L. Jr., Myers, T., Lane, B., Sullivan, E.V., Faustman, W.O. and Pfefferbaum, A. (1995a) Brain dysmorphology in adults with congenital rubella plus schizophrenia-like symptoms. Biol. Psychiatry 37, 764 776. Lim, K.O., Harris, D., Beal, M., Minn, K., Hoff, A., Csernansky, J., Faustman, W., Marsh, L., Sullivan, E.V. and Pfefferbaum, A. (1995b) Gray matter volume deficits in young onset schizophrenia are independent of age of onset. Biol. Psychiatry in press, Lim, K.O. and Pfefferbaum, A. (1989) Segmentation of MR brain images into cerebrospinal fluid spaces, white and gray matter. J. Comput. Assist. Tomogr. 13, 588-593. Lim, K.O., Tew, W., Kushner, M., Chow, K., Matsumoto, B. and DeLisi, L.E. (1994) First episode schizophrenics have a deficit in cortical gray matter volume (Abstr.). Am. Coll. Neuropharmacol. Abstracts of Panels and Posters, 205. Marsh, L., Lauriello, J., Sullivan, E. and Pfefferbaum, A. (1995). Neuroimaging in neuropsychiatric disorders. In: E. Bigler (Ed.), Handbook of Human Brain Function: Neuroimaging. Plenum Press, New York, in press. Mathalon, D.H., Sullivan, E.V., Rawles, J.M. and Pfefferbaum, A. (1993a) Correction for head size in brain imaging measurements. Psychiatry Res. Neuroimaging 50, 121 139. Mathalon, D.H., Sullivan, E.V., Rawles, J.M. and Pfefferbaum, A. (1993b) Correction for head size in brain imaging measurements: Letter to editor. Psychiatry Res. Neuroimaging 50, 284. Miller, B.L. (1991) A review of chemical issues in ~H NMR spectroscopy: N-acetyl-L-aspartate, creatine and choline. NMR Biomed. 4, 47-52. Nelson, H.E. (1982) National Adult Reading Test, NART. Nelson Publishing Company, Windsor. O'Carroll, R.A., Walker, M., Dunan, J., Murray, C., Blackwood, D., Ebmeier, K. and Goodwin, G. (1992) Selecting controls for schizophrenia research studies: The use of the national adult reading test (NART) as a measure of pre-morbid ability. Schizophrenia Res. 8, 137 141. Overall, J.E. and Gorham, D.R. (1988) The Brief Psychiatric Rating Scale (BPRS): Recent developments in ascertainment and scaling. Psychopharmacol. Bull. 24, 97 99. Pfefferbaum, A., Lim, K.O., Zipursky, R.B., Mathalon, D.H., Lane, B., Ha, C.N., Rosenbloom, M.J. and Sullivan, E.V. (1992) Brain gray and white matter volume loss accelerates with aging in chronic alcoholics: A quantitative MRI study. Alcoholism Clin. Exp. Res. 16, 1078-1089.
Pfefferbaum, A., Mathalon, D.H., Sullivan, E.V., Rawles, J.M., Zipursky, R.B. and Lim, K.O. (1994) A quantitative magnetic resonance imaging study of changes in brain morphology from infancy to late adulthood. Arch. Neurol. 51, 874 887. Pfefferbaum, A., Sullivan, E.V., Mathalon, D.H., Shear, P.K., Rosenbloom, M.J. and Lim, K.O. (1995) Longitudinal changes in MRI brain volumes in abstinent and relapsed alcoholics. Alcoholism Clin. Exp. Res. in press, Pfefl'erbaum, A., Zatz, L. and Jernigan, T.L. (1986) Computerinteractive method for quantifying cerebrospinal fluid and tissue in brain CT scans: effects of aging. J. Comput. Assist. Tomogr. 10, 571-578. Raz, S. and Raz, N. (1990) Structural brain abnormalities in the major psychoses: A quantitative review of the evidence from computerized imaging. Psychol. Bull. 108, 93-108. Roth, W.T. and Pfefferbaum, A. (1992) Abnormalities of the left temporal lobe in schizophrenia (letter). New Engl. J. Med. 327, 1689. Schlaepfer, T.E., Harris, G.J., Tien, A.Y., Peng, L.W., Lee, S., Federman, E.B., Chase, G.A., Barta, P.E. and Pearlson, G.D. (1994) Decreased regional cortical gray matter volume in schizophrenia. Am. J. Psychiatry 151,842-848. Spitzer, R.L., Williams, J.B.W., Gibbon, M. and First, M.B. (1992) The structured clinical interview for DSM-III-R (SCID). 1. History, rationale, and description. Arch. Gen. Psychiatry 49, 624 629. Suddath, R., Casanova, M.F., Goldberg, T.E., Daniel, D.G., Kelsoe, J.R. and Weinberger, D.R. (1989) Temporal lobe pathology in schizophrenia. Am. J. Psychiatry 146, 464-472. Suddath, R.L., Christison, G.W., Fuller Torrey, E., Casanova, M.F. and Weinberg, D.R. (1990) Anatomical abnormalities in the brains of monozygotic twins discordant for schizophrenia. New Engl. J. Med. 322, 789 794. Sullivan, E.V., Marsh, L., Mathalon, D.H., Lim, K.O. and Pfefferbaum, A. (1995a) Age-related decline in MRI volumes of temporal lobe gray matter but not hippocampus. Neurobiol. Aging in press, Sullivan, E.V., Shear, P.K., Lim, K.O., Zipursky, R.B. and Pfefferbaum, A. (1995b) Cognitive and motor impairments are related to gray matter volume deficits in schizophrenia. Biol. Psychiatry submitted for publication. Zipursky, R.B., Bury, A., Langevin, R. and Seeman, M.V. (1994) Gray and white matter volumes in schizophrenia and bipolar disorders (Abstr.). Biol. Psychiatry 35, 719-720. Zipursky, R.B., Lim, K.O., Sullivan, E.V., Brown, B.W. and Pfefferbaum, A. (1992) Widespread cerebral gray matter volume deficits in schizophrenia. Arch. Gen. Psychiatry 49, 195 205.