Cerebral volume measurements and subcortical white matter lesions and short-term
treatment response in late life depression.
Joost Janssen1,2,A, Hilleke E. Hulshoff Pol1, Hugo G. Schnack1,
Rob M. Kok2, Indrag K. Lampe3, Frank-Erik de Leeuw4,
Rene S. Kahn1, Thea J. Heeren1,5.
1Department of Psychiatry, Rudolf Magnus Institute of Neuroscience, University Medical Centre Utrecht,
Utrecht, The Netherlands
2Department of Old Age Psychiatry, Altrecht, Zeist, The Netherlands
3Department of Psychiatry, University Medical Centre St Radboud Nijmegen, Nijmegen, The Netherlands
4Department of Neurology, University Medical Centre St Radboud Nijmegen, Nijmegen, The Netherlands
5Symfora Group, Centres of mental health care, Amersfoort, The Netherlands
Background: Late-life depression is associated with volumetric reductions of gray
matter and increased prevalence of subcortical white matter lesions. Previous studies
have shown a poorer treatment outcome in those with more severe structural brain
abnormalities. In this study, quantitative and semi-quantitative magnetic resonance
imaging (MRI) measures were studied in relation to response to a 12-week controlled
antidepressant monotherapy trial.
Methods: MRI (1.5T) brain scans of 42 elderly inpatients with major depression, of
which 23 were non-responder to a controlled 12-week antidepressant monotherapy trial,
were acquired. In addition, clinical outcome was assessed after a one year period.
Measures were volumes of global cerebral and subcortical structures. White matter
lesions were measured volumetrically and semi-quantitatively.
Results: After controlling for age, intracranial volume and sex, no differences were
found between non-responders and responders after 12 weeks and after one year in
volumes of cerebral gray and white matter, orbitofrontal cortex, hippocampus and white
Conclusions: Structural brain measures associated with late-life depression may not be
related to short-term treatment response.
Structural Magnetic Resonance Imaging (MRI) studies report that depression is
associated with volumetric decreases in the frontal and orbitofrontal cortex and
hippocampus in adult and older subjects.1-7 In addition, many found an increased
prevalence of white matter lesions in late-life depressed subjects compared to controls.8-11
In late-life depression, the relation of decreased cerebral volume and increased white
matter lesion prevalence with treatment response has been investigated. In studies that
looked at long-term outcome (≥ 2 years), reduced hippocampal gray matter density and
increased white matter lesion load at baseline were associated with a chronic course of
illness.12-17 Others have shown that progression of lesions over time, rather than static
baseline lesion severity, is an important predictor for long-term outcome.18
Evidence for an association of poor short-term antidepressant treatment response with
decreased cerebral volumes and increased subcortical white matter lesion prevalence is
inconclusive.19-24 For example, in older severely depressed inpatients with a long history
of treatment resistance an association between increased subcortical white matter lesion
severity, especially in the frontal white, basal ganglia, and the pontine reticular formation,
and poorer acute treatment response was reported.19-21 However, a recent study among in-
and outpatients suggest no such relation.24 Most of the studies investigating the
relationship between structural brain measures and short-term antidepressant treatment
response have used an uncontrolled naturalistic treatment protocol or they have focused
exclusively on white matter lesion load as a potential predictor for short-term treatment
response. We therefore wanted to investigate the difference in baseline global cerebral
and subcortical brain measures, including white matter lesions, associated with late-life
depression between responders and non-responders who participated in a controlled 12-
week antidepressant monotherapy trial.
This study was designed as an adjunct to a double-blind, randomized 12 week parallel-
group trial of venlafaxine and nortryptiline in elderly depressed inpatients. All subjects
were inpatients from the depression unit at the Altrecht old age psychiatry department in
Zeist, The Netherlands. Inclusion criteria required that patients were 60 years of age or
older, all patients had to meet the Diagnostic and Statistical Manual of Mental Disorders,
Fourth Edition (DSM-IV) criteria for major depression (single episode or recurrent) as
assessed by an old age psychiatrist and confirmed by the International Diagnostic Check
List and a total score of at least 20 on the Montgomery Åsberg Depression Rating Scale
(MADRS).25,26 Patients in the treatment trial were excluded if the current episode had
been treated unsuccessfully with venlafaxine or nortriptyline, if there was a medical
contra-indication to the study medication, if they met DSM-IV diagnostic criteria for
dementia or had a Mini Mental Status Examination score (MMSE) < 15, or if they met
DSM-IV diagnostic criteria for alcohol or drug abuse within the last year.27 At baseline
complete psychiatric and medical histories were taken, and a thorough physical
examination was performed prior to study entry. Physical illnesses were recorded
according to Burvill.28
This trial was conducted in accordance with the Declaration of Helsinki (1964), as
amended in South Africa (1996) and Scotland (2000) and has been approved by the ethics
committee of the University Medical Center Utrecht. Written informed consent was
obtained from all patients or their legal representatives before study entry.
Efficacy was evaluated with the MADRS. Remission was defined as a final score of 10 or
less on the MADRS. Good response was a reduction of at least 50% of score on the
MADRS. Partial response was defined as a 25-49% reduction in score on MADRS. Poor
response was defined as less then 25% reduction in score on MADRS or drop-out due to
insufficient response. Non-response was defined as partial- or poor response; response
was defined as good response or remission.
Functional limitations were measured by the Barthel index of Activities of Daily Living
(ADL) with a score of 0-20 and functional status was assessed at baseline using the 100-
point Global Assessment of Functioning (GAF) scale, on which a score of 100 represent
best possible functioning (American Psychiatric Association 1994).29 Vital signs (hart
frequency and blood pressure) were measured weekly. Orthostatic hypotension was
defined as a fall of ≥ 20 mg Hg in systolic blood pressure or ≥ 10 mg Hg in diastolic
blood pressure, within 3 minutes of standing. All baseline and endpoint assessments were
made by an old age psychiatrist (RMK).
Description short-term trial sample
Of the 50 subjects who received an MRI scan, eight did not finish the 12-week
antidepressant trial due serious adverse events according to the patient (n=2), refusal of
the patient for other reasons (n=2), protocol violation (n=1), and withdrawal for medical
reasons (n=3). These patients were excluded from the study. Of the remaining 42
patients, five patients did not finish the trial due to insufficient response (all had at least
seven weeks of treatment) and these subjects were analyzed as non-responders in the
completers group. This left a final sample of 23 non-responders (55%) and 19 responders.
There were no significant psychiatric, cognitive, physical and treatment response
differences between patients having a scan and those not (analyses not shown).
Description of follow-up sample
Patients that finished the short-term treatment trial were asked for assessment of their
clinical outcome after a one year period. To investigate whether clinical response,
independent of treatment type, was related to the neuroimaging variables we classified
patients into poor outcome and good outcome using the baseline and one year follow-up
MADRS score. Classification-criteria were identical to the division in non-response and
response. At the one year mark, eight patients were lost due to missing MADRS data. Of
the remaining 34 patients, 24 (71%) were responders.
Scans were acquired as soon as possible after initiation of treatment. In a number of
cases, patients consented to the MRI acquisition only after the treatment trial had ended.
Magnetic resonance images were acquired using a scanner (Philips Gyroscan; Philips
Medical Systems, Best, the Netherlands) operating at 1.5 T in all subjects. T1-weighted,
3-dimensional, fast field echo scans with 160 to 180 1.2-mm contiguous coronal slices
(echo time [TE], 4.6 milliseconds; repetition time [TR], 30 milliseconds; flip angle, 30°;
field of view [FOV], 256 mm; and in-plane voxel sizes, 1 x 1 mm2) and T2-weighted,
dual echo turbo spin echo scans with 90 2.1-mm contiguous coronal slices (TE1, 14
milliseconds; TE2, 90 milliseconds; TR, 4530 milliseconds; flip angle, 90°; FOV, 256
mm; and in-plane voxel sizes, 1 x 1 mm2) of the whole head were used for quantitative
measurements. In addition, FLAIR-weighted scans with 19 axial 5-mm slices and a 1.2-
mm gap (TE1, 100 milliseconds, TE2, 90 milliseconds, flip angle, 90°, FOV 256 mm,
and in-plane voxel sizes, 0.89 x 0.89 mm2) and T2-weighted, dual echo turbo spin echo
scans with 19 axial 5-mm slices and a 1.2-mm gap (TE1, 30 milliseconds; TE2, 90
milliseconds; TR, 2377 milliseconds, flip angle, 90°; FOV, 256 mm; and in-plane voxel
sizes, 0.89 x 0.89 mm2) were used forclinical neurodiagnostic evaluation and all 19 axial
5-mm slices (1.2 mm gap) of both scans were used for manual segmentation of white
matter lesions and inspected for white matter lesion ratings.
Before quantitative assessments and white matter lesion rating, 10 MRI scans were
randomly chosen and cloned for intrarater and reliability determined by the intraclass
correlation coefficient (ICC) for volumetric assessments and weighted kappas for white
matter lesion rating.30 All MRI scans were coded to ensure masking for subject
identification and diagnosis. The MRI data sets were transformed (no scaling) to the
Talairach frame with software developed in house.31 The transformation used information
gathered from the placement of a midline in coronal and axial views and the marking of
the superior edge of the anterior commissure and the inferior edge of the posterior
commissure in the sagittal view. In addition, scans were correctedfor inhomogeneities in
the magnetic field.32 Intracranial, total brain, gray and white matter volumes of the
cerebrum (total brain excluding cerebellum and stem), and ventricular volume were
measured automatically by using histogram analysis algorithms and a series of
mathematical morphology operators to connect all voxels of interest.33,34 Intracranial
volume was segmented on the DTSE scans, with the foramen magnum being used as
inferior boundary. Total brain volumes were segmented on the 3D-FFE (T1-weighted)
scans and contained gray and whitematter tissue only. All images were checked after the
measurements and corrected manually if necessary.
Quantitative measurements of the hippocampus, parahippocampal gyrus, orbitofrontal
cortex, and white matter lesions were done with the software package DISPLAY
developed at the Montreal Neurological Institute. This program allows simultaneous
viewing in coronal, sagittal and axial sections. Neuroanatomic borders of the
hippocampus and parahippocampal gyrus have been previously described.6 Segmentation
of the hippocampus started in the coronal slice in which the mamillary bodies were
visible and stopped when the fornix was visible as a continuous tract. Parahippocampal
gyrus segmentation began in the coronal slice in which the optic tract is situated above
the amygdala. The posterior commissure was its posterior border.35 The orbitofrontal
cortex was manually segmented within the total brain mask in the coronal plane using a
geometrical method.6 To summarize this method briefly, the posterior border was
determined by the tip of the genu of corpus callosum located in the sagittal plane. The
anterior border was the first slice where brain tissue could be identified. The superior
limit was divided in two parts. In the subgenual regions, the superior boundary was
represented by the inferior border of the anterior cingulate corresponding to a midpoint at
the interhemispheric fissure about five slices below the intercommissural line. More
anteriorly, the superior limit was represented by a midpoint placed on the
intercommissural line. For the lateral borders, horizontal and vertical crosshairs were
placed as tangent lines at the inferior and lateral surfaces of the frontal lobes in all slices.
The intersection of these two lines generated two lateral points that were connected to the
superior limit point, composing the lateral boundaries of the segment. Anterior and
inferior borders were crossed, as segmentations in these areas were larger then necessary.
This manual segmentation was multiplied with the binary mask of the total brain
resulting in an orbitofrontal cortex segmentation without the superfluous segmentation.
White matter lesions were manually segmented on the FLAIR scan by the first author
under supervision of an expert (FEL) who identified and marked the lesions using both
dual echo and FLAIR scans. Finally, the expert inspected the segmentations for accuracy.
In addition, the expert rated the number and size of subcortical white matter lesions using
a semi-quantitative scale.36
A single operator (JJ) performed the volume measurements of the hippocampus,
parahippocampal gyrus and orbitofrontal cortex. ICCs for the left and right hippocampus
were 0.98 and 0.97, for the parahippocampal gyrus 0.73 and 0.78, for the orbitofrontal
cortex 0.98, for the periventricular lesions 0.94, and for the subcortical lesions 0.98. For
semi-quantitative white matter lesion rating the intrarater study showed good to excellent
agreement, weighted kappas for grading the periventricular and subcortical WML ranged
between 0.90 and 0.95.
Data were examined for outliers, extreme values and the normality of the distribution.
Non-normally distributed data were transformed using natural logarithms and re-
examined for fit to the normal distribution. To assess the relationship between short-term
treatment response, one year outcome and brain volumes we used analysis of covariance
(ANCOVA) with brain volumes as the dependent variables and response or outcome as
the independent variable. Adjustments were made for age, sex and intracranial volume.
Total brain volume was used as a covariate instead of intracranial volume in the analysis
of covariance of orbitofrontal cortex, hippocampal and parahippocampal gyrus volumes.
For continuous and non-continuous variables (demographic, clinical and prevalence of
white matter lesions), independent-samples t-Test and Chi-Square analyses were used to
test for differences.
Short-term treatment trial
Non-responders versus responders
Non-responders were older, and had a trend for more family psychiatric history (see
Table 1). Responders and non-responders did not differ significantly in sex, level of
education, vascular (smoking, hypertension and diabetes) comorbidity, and number of
previous depressive episodes.
Non-responders versus responders: baseline neuroimaging
After controlling for cranial volume and age, total brain volume did not differ between
non-responders and responders (F=1.48, df=1,37, p=0.23, see Table 1). In addition, other
neuroimaging variables, both raw data and logarithmically transformed data, did not
differ between non-responders and responders as well (see Table 1). Leaving out age as a
covariate did not change the results.
One year follow-up
Poor outcome versus good outcome: baseline neuroimaging
No differences in baseline neuroimaging variables between patients with poor outcome
and good outcome after a one year follow-up were found (Table 2).
Response versus outcome
Response to the short-term treatment trial was not associated with good outcome, 11
responders (46%) and 13 non-responders (54%) had good outcome (χ²=0.05, degrees of
freedom (df)=1, p=0.82).
We compared structural brain abnormalities between older non-responders and
responders in a controlled short-term antidepressant monotherapy trial and after a one
year follow-up period. Non-responders and responders did not differ on any of the
quantitative and semi-quantitative structural brain measures.
Our findings in inpatients are in line with a previous short-term controlled trial that
measured white matter lesions in older depressed outpatients and also found no
association with acute treatment response.24 In addition, naturalistic treatment studies in
older in-and outpatients did not report a relationship between total white matter lesion
load and a poorer response to short-term antidepressant monotherapy as well.20,21,37,38
Moreover, the current finding of no differences in neuroimaging variables between
patients that had poor outcome or good outcome after one year compared to baseline
strengthen the evidence that no clear association exists between short-term treatment
response and structural brain volume changes. In older severely depressed inpatients with
a long history of treatment resistance an association between increased total white matter
lesion load and poor acute antidepressant response was found.19 Simpson et al reported a
relationship between regionally specific subcortical lesion load (frontal white, basal
ganglia, and pontine reticular formation) and poorer response to short-term antidepressant
Table 1: Demographical, clinical and neuroimaging data of responders and non-responders to a 12-week antidepressant monotherapy
trial and outcome after 1 year.
Response 12 weeks
Age, years, mean (SD)
Sex, female (%)
Weight, kilograms, mean (SD)a
Level of education, years , mean (SD)
Married, n (%)
Widow, n (%)
Outcome 52 weeks
MADRS score baseline , mean (SD)
MADRS score final, mean (SD)
MMSE score baseline, mean (SD)
Age of onset, years, mean (SD)b
Number of previous episodes, mean (SD)c
Family history of depression, n (%)
Health status at baselined
Smoking, n (%)
Diabetes, n (%)
Hypertension, n (%)
Burvill acute baseline, mean (SD)e
Burvill chronic baseline, mean (SD)
Barthel ADL baseline, mean (SD)
DSM IV GAF score baseline, mean (SD)
Time between start of trial and scan
(weeks), mean (SD)
MADRS, Montgomery Ǻsberg Depression Rating Scale; MMSE, Mini-Mental State Examination; ADL, Assessment of Daily Living;
GAF, Global Assessment of Functioning; MRI, Magnetic Resonance Imaging; Missing data for a10, b3, c3, d6 subjects; eOther Burvill
summary scores were similar.
Table 2: Baseline neuroimaging data of responders and non-responders to a 12-week antidepressant monotherapy trial and outcome
after 1 year.
Response 12 weeks
Cranium, ml, mean (SD)f
Total Brain Volume, ml, mean (SD)
Cerebral Gray Matter Volume, ml, mean (SD)
Cerebral White Matter Volume, ml, mean (SD)
Lateral Ventricle Volume, ml, mean (SD)
Third Ventricle Volume, ml, mean (SD)
Cerebellum Volume, ml, mean (SD)
Outcome 52 weeks
Orbitofrontal Cortex Volume, ml, mean (SD)
Orbitofrontal Gray Volume, ml, mean (SD)
Hippocampus Volume, ml, mean (SD)
Parahippocampus Volume, ml, mean (SD)
Periventricular lesion volume, ml, mean (SD)
Subcortical lesion volume, ml, mean (SD)
White matter lesion Prevalenceg
Frontal, n (%)
Lateral, n (%)
Occipital, n (%)
Small, n (%)
Medium, n (%)
Large, n (%)
Volumes are raw volumes, p-value represents significance after controlling for cranial volume, age and sex; gNumber of people with
periventricular and subcortical white matter lesions. Periventricular lesions were rated per region (adjacent to the frontal horn, the
lateral ventricles and the occipital horn), subcortical lesions were categorized according to their largest diameter in small (<3 mm),
medium (3-10 mm), or large lesions (>10 mm).
1165. 69 (130.92)
treatment in older depressed patients.21 These latter findings suggests that lesion
location rather than total lesion load might be related to short-term antidepressant
treatment response. However, these findings need to be replicated in future studies.
In our study, global cerebral volumes and volumes of the orbitofrontal cortex and
hippocampus were not associated with treatment response. This is congruent with the
negative results from previous short-term and long-term treatment studies in adult and
elderly subjects.23,39,40 Patients with the smallest hippocampal volume had a poor
short-term treatment response in one study but there was no difference in
hippocampal volume when responders and non-responders were compared.41 One
study, using computed tomography instead of MRI, reported an association between
global cerebral atrophy and poorer treatment response.22
In the current sample, we did not find a correlation of increased white matter lesion
prevalence and decreased cerebral volumes with poorer response to short-term
antidepressant treatment in late-life depressed inpatients and with poor outcome after
a one year follow-up period. The majority of previous MRI studies investigating
short-term treatment response (< one year) in older depressed patients have yielded
similar results. Our findings should be interpreted with care due to a relatively small
sample size and the delay of the scan acquisition after the start of treatment in a
number of patients. Nevertheless, all scans were made preceding the one year mark
with a mean period of 32 weeks between the date of the scan and the one year follow-
up observation. This makes it unlikely that prompt post-trial brain changes, such as
the development of white matter lesions, (< 12-52 weeks) have influenced our results.
The previous and current findings do not give substantial evidence for an alteration of
the short-term treatment strategy in the presence of structural cerebral abnormalities.
In addition, the benefit of one MRI measurement in short-term treatment trials in late-
life depression is expected to be minimal. Future larger controlled trials that include
longitudinal MRI data in specific groups of older depressed patients (for example
those with a first-episode of late-onset depression) are needed in order to explore the
possibilities for clinical contribution of brain changes over time.
We thank Sjoerd Fluitman, MD, and Ana Sierra Blancas Lopez-Barajas, MD for
DECLARATION OF INTEREST
1 Coffey, C. E., Wilkinson, W. E., Weiner, R. D., Parashos, I. A., Djang, W. T.,
Webb, M. C., Figiel, G. S. and Spritzer, C. E. (1993). Archives of General
Psychiatry 50(1), 7-16.
Lai, T., Payne, M. E., Byrum, C. E., Steffens, D. C. and Krishnan, K. R. (2000).
Biological Psychiatry 48(10), 971-975.
Sheline, Y. I., Sanghavi, M., Mintun, M. A. and Gado, M. H. (1999). Journal of
Neuroscience 19(12), 5034-5043.
Steffens, D. C., Byrum, C. E., McQuoid, D. R., Greenberg, D. L., Payne, M. E.,
Blitchington, T. F., MacFall, J. R. and Krishnan, K. R. (2000). Biological
Psychiatry 48(4), 301-309.
MacQueen, G. M., Campbell, S., McEwen, B. S., Macdonald, K., Amano, S.,
Joffe, R. T., Nahmias, C. and Young, L. T. (2003). Proceedings of the National
Academy of Scencei U S A 100(3), 1387-1392.
Janssen, J., Hulshoff Pol, H. E., Lampe, I. K., Schnack, H. G., de Leeuw, F. E.,
Kahn, R. S. and Heeren, T. J. (2004). Biological Psychiatry 56(11), 825-831.
O'Brien, J. T., Lloyd, A., McKeith, I., Gholkar, A. and Ferrier, N. (2004).
American Journal of Psychiatry 161(11), 2081-2090.
Krishnan, K. R., Goli, V., Ellinwood, E. H., France, R. D., Blazer, D. G. and
Nemeroff, C. B. (1988). Biological Psychiatry 23(5), 519-522.
O'Brien, J., Desmond, P., Ames, D., Schweitzer, I., Harrigan, S. and Tress, B.
(1996). British Journal of Psychiatry 168(4), 477-485.
Steffens, D. C., Helms, M. J., Krishnan, K. R. and Burke, G. L. (1999). Stroke
Kumar, A., Bilker, W., Jin, Z. and Udupa, J. (2000). Neuropsychopharmacology
Hickie, I., Scott, E., Wilhelm, K. and Brodaty, H. (1997). Biological Psychiatry
O'Brien, J., Ames, D., Chiu, E., Schweitzer, I., Desmond, P. and Tress, B.
(1998). British Medical Journal 317(7164), 982-984.
Shah, P. J., Ebmeier, K. P., Glabus, M. F. and Goodwin, G. M. (1998). British
Journal of Psychiatry 172, 527-532.
Lavretsky, H., Lesser, I. M., Wohl, M., Miller, B. L. and Mehringer, C. M.
(1999). American Journal of Geriatric Psychiatry 7(4), 309-316.
Baldwin, R. C., Walker, S., Simpson, S. W., Jackson, A. and Burns, A. (2000).
International Journal of Geriatric Psychiatry 15(12), 1097-1104.
Chen, P. S., McQuoid, D. R., Payne, M. E. and Steffens, D. C. (2006).
International Psychogeriatrics, 1-12.
Taylor, W. D., Steffens, D. C., MacFall, J. R., McQuoid, D. R., Payne, M. E.,
Provenzale, J. M. and Krishnan, K. R. (2003). Archives of General Psychiatry
Hickie, I., Scott, E., Mitchell, P., Wilhelm, K., Austin, M. P. and Bennett, B.
(1995). Biological Psychiatry 37(3), 151-160.
Simpson, S. W., Jackson, A., Baldwin, R. C. and Burns, A. (1997). International
Psychogeriatrics 9(3), 257-275.
Simpson, S., Baldwin, R. C., Jackson, A. and Burns, A. S. (1998). Psychological
Medicine 28(5), 1015-1026.
Young, R. C., Kalayam, B., Nambudiri, D. E., Kakuma, T. and Alexopoulos, G.
S. (1999). American Journal of Geriatric Psychiatry 7(2), 147-150.
70 Download full-text
23 Simpson, S. W., Baldwin, R. C., Burns, A. and Jackson, A. (2001). International
Journal of Geriatric Psychiatry 16(5), 469-476.
Salloway, S., Boyle, P. A., Correia, S., Malloy, P. F., Cahn-Weiner, D. A.,
Schneider, L., Krishnan, K. R. and Nakra, R. (2002). American Journal of
Geriatric Psychiatry 10(1), 107-111.
Hiller, W., Zaudig, M. and Mombour, W. (1990). Archives of General
Psychiatry 47(8), 782-784.
Montgomery, S. A. and Asberg, M. (1979). British Journal of Psychiatry 134,
Folstein, M. F., Folstein, S. E. and McHugh, P. R. (1975). Journal of Psychiatric
Research 12(3), 189-198.
Burvill, P. W., Mowry, B., Hall,W. D. (1990). International Journal of Geriatric
Psychiatry 5(3), 161-170.
de Haan, R., Limburg, M., Schuling, J., Broeshart, J., Jonkers, L. and van
Zuylen, P. (1993). Nederlands Tijdschrift voor Geneeskunde 137(18), 917-921.
Bartko, J. J. and Carpenter, W. T., Jr. (1976). Journalof Nervous and Mental
Disorders 163(5), 307-317.
Talairach, J., Tournoux, P. (1988). Co-planar stereotaxic atlas of the human
brain. Thieme: NewYork.
Sled, J. G., Zijdenbos, A. P. and Evans, A. C. (1998). IEEE Transactions in
Medical Imaging 17(1), 87-97.
Schnack, H. G., Hulshoff, H. E., Baare, W. F., Viergever, M. A. and Kahn, R. S.
(2001a). Neuroimage 14(1 Pt 1), 95-104.
Schnack, H. G., Hulshoff Pol, H. E., Baare, W. F., Staal, W. G., Viergever, M.
A. and Kahn, R. S. (2001b). Neuroimage 13(1), 230-237.
Watson, C., Andermann, F., Gloor, P., Jones-Gotman, M., Peters, T., Evans, A.,
Olivier, A., Melanson, D. and Leroux, G. (1992). Neurology 42(9), 1743-1750.
de Groot, J. C., de Leeuw, F. E., Oudkerk, M., Hofman, A., Jolles, J. and
Breteler, M. M. (2000). Archives of General Psychiatry 57(11), 1071-1076.
Krishnan, K. R., Hays, J. C., George, L. K. and Blazer, D. G. (1998). Depression
and Anxiety 8(4), 142-146.
Baldwin, R., Jeffries, S., Jackson, A., Sutcliffe, C., Thacker, N., Scott, M. and
Burns, A. (2004). Psychological Medicine 34(1), 125-136.
Frodl, T., Meisenzahl, E. M., Zill, P., Baghai, T., Rujescu, D., Leinsinger, G.,
Bottlender, R., Schule, C., Zwanzger, P., Engel, R. R., Rupprecht, R., Bondy,
B., Reiser, M. and Moller, H. J. (2004). Archives of General Psychiatry 61(2),
Vythilingam, M., Vermetten, E., Anderson, G. M., Luckenbaugh, D., Anderson,
E. R., Snow, J., Staib, L. H., Charney, D. S. and Bremner, J. D. (2004).
Biological Psychiatry 56(2), 101-112.
Hsieh, M. H., McQuoid, D. R., Levy, R. M., Payne, M. E., MacFall, J. R. and
Steffens, D. C. (2002). International Journal of Geriatric Psychiatry 17(6), 519-