Neurobiology of Aging 32 (2011) 22992307

A lifespan MRI evaluation of ventricular enlargement in

normal aging mice
Chiao-Chi V. Chen a,b , Yu-Ying Tung a,b , Chen Chang a,b,
b

a Institute of Biomedical Sciences, Academic Sinica, Taipei, Taiwan, ROC

Functional and Micro-magnetic Resonance Imaging Core Facility, Academic Sinica, Taipei, Taiwan, ROC

Received 18 May 2009; received in revised form 11 January 2010; accepted 18 January 2010
Available online 5 February 2010

Abstract
Ventricular enlargement has been proposed as a structural biomarker for the progression of Alzheimers disease (AD). This biomarker,
established in human patients, needs to be translated to animals to facilitate drug development for the disease. However, ventricular enlargement
is not exclusive to AD, since the ventricle size increases during normal aging. A longitudinal characterization of ventricular enlargement in
normal aging in mice is therefore crucial before further evaluations of mouse models or neurodegenerative diseases associated to brain atrophy.
To this end, ventricular enlargement in normal aging mice was characterized over the lifespan (i.e., 2 years). The results showed that the
overall ventricle size increased with age, with the expansion beginning during the early life stages and continuing to old age. The reported data
represent a biomarker benchmark for normal aging mice under unmodified conditions. This provides a foundation for evaluating the validity
of AD mouse models or the effects of potential drugs. The considerable physiological ventricular enlargement during normal aging must be
considered in related experiments.
2010 Elsevier Inc. All rights reserved.
Keywords: Biomarker; Transgenic; Knockout; Volumetry; Age; Parkinson; Schizophrenia; Multiple sclerosis; Ventricular expansion; Cross-sectional; Ventricle
size; Lateral ventricle; Third ventricle; Fourth ventricle; Cerebral aqueduct

1. Introduction
The intense worldwide interest in searching for biomarkers of the progression of Alzheimers disease (AD) has at least
three important goals: facilitating early diagnosis, longitudinally assessing the severity, and promoting drug development
(Hampel et al., 2008). Toward these goals, homologous
biomarkers for humans and animals are highly desirable since
they can be applied in both preclinical and clinical trials. Neuroimaging modalities such as magnetic resonance imaging
(MRI) are ideal methods for this due to their wide applications in both animals and humans (Bradley et al., 2002; de
Leon et al., 2004, 2007; Fox et al., 2000; Jack et al., 2004;

Corresponding author at: N123, Institute of Biomedical Sciences,

Academia Sinica, 128 Section 2, Academia Rd. Nankang, Taipei 11529,
Taiwan. Tel.: +886 2 27899027; fax: +886 2 27887641.
E-mail address: bmcchen@ibms.sinica.edu.tw (C. Chang).

0197-4580/$ see front matter 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.neurobiolaging.2010.01.013

Maheswaran et al., 2009; Ridha et al., 2008; Schott et al.,

2005; Wang et al., 2002).
Ventricular enlargement is an obvious MRI-based structural biomarker that characterizes the neuropathological
changes of AD (Bradley et al., 2002; Jack et al., 2004, 2005;
Luxenberg et al., 1987; Nestor et al., 2008; Ridha et al., 2008;
Schott et al., 2005). It has been reported that the ventricle
size increased by 5.7% over a 6-month tracking period in AD
patients aged from 55 to 90 years, whereas the increase was
only 1.5% in normal elderly subjects (Nestor et al., 2008).
Such human studies set the stage for the use of ventricular enlargement as a biomarker for AD progression. The next
goal should be translating this biomarker to preclinical use for
facilitating the development of treatments for AD. This will
standardize evaluations on the validity of AD animal models
and the efficacy of developed drugs (Frisoni and Delacourte,
2009).
Ventricular enlargement is not exclusive to AD, since the
ventricle size increases during normal aging (Foundas et al.,

2300

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

1998; Kalpouzos et al., 2009; Mu et al., 1999; Resnick et al.,

2000; Scahill et al., 2003; Walhovd et al., 2005). A longitudinal characterization of ventricular enlargement in normal
aging is therefore crucial to differentiating the causes of
ventricular enlargement prior to using this biomarker in preclinical settings. This was achieved in the current study by
characterizing ventricle enlargement in normal aging mice
over the lifespan (i.e., 2 years). The investigations elucidated (1) how the individual ventricular chambers vary with
age, (2) when the enlargements begin, and (3) the lifetime
enlargement pattern. Neonatal mice were repeatedly scanned
by 3D T2-weighted imaging (T2WI) from week 3 to week
100 after birth. T2WI detects the water content in brain tissues, with water-enriched areas appearing as high signals
on the obtained images. Since the ventricles are filled with
cerebrospinal fluid (CSF), T2WI is particularly suitable for
identifying the ventricles. A cross-sectional experiment was
carried out alongside the longitudinal experiment, allowing
a comparison of the two approaches.

2. Materials and methods

2.1. Subjects
Forty-eight male C57BL/6J mice purchased from the
National Laboratory Animal Center of Taiwan were used
in the study. The mice were group-housed in cages with
free access to food and water. The housing environment had
a 12-h/12-h light/dark cycle with controlled humidity and
temperature. The mice were kept in a specific-pathogen-free
environment throughout the study. All experimental procedures were approved by the Institute of Animal Care and
Utilization Committee at Academia Sinica, Taipei, Taiwan.
In the longitudinal experiment, five animals were scanned
for MRI at weeks 3, 6, 12, 24, 48, and 75, and four at week
100 because one mouse died at week 100. The cross-sectional
experiment used 11, 12, 6, 6, and 8 animals for the 5 age
groups of weeks 3, 6, 18, 32, and 44, respectively. Mice
are generally considered to be young adults when they are
89 weeks old, at which time the growth of most biological processes and structures is slower than during the earlier
developmental stages. Mice are considered to be old when
they reach approximately 70 weeks, at which time major
senescence changes begin to appear. The average lifespan
of C57BL/6J mice is 125 weeks (Kunstyr and Leuenberger,
1975). The time points investigated in the current study
were chosen to cover the life stages as completely as
possible.
2.2. MRI acquisition
On the experiment day, each mouse was anesthetized by
2% isoflurane in O2 at a flow rate of 1 l/min. The breathing rate was maintained between 60 and 70 breaths/minute.
The anesthetized mouse was fixed in a customized head

holder by two ear bars and an incisor fixer. The holder was
then horizontally inserted into a 7-T scanner (PharmaScan
70/16, Bruker, Germany) with an active shielding gradient
of 300 mT/m in 80 s. The scanner used a 38-mm volume
coil for signal transmission and reception. A mid-sagittal
T2-weighted image (repetition time (TR) = 2000 ms, effective echo time (TEeff ) = 88.3 ms, field of view (FOV) = 3 cm,
matrix = 256 128 zero-filled to 256 256, in-plane resolution = 117 m 117 m, slice thickness = 1.5 mm) was
acquired and used to measure the length of the brain. An
axial T2-weighted 3D RARE (rapid acquisition relaxationenhanced) image was acquired (TR = 4000 ms, TEeff = 80 ms,
FOV = 2 2 1.5 cm, matrix = 256 128 64 zero-filled
to 256 256 64, in-plane resolution = 78 m 78 m,
number of excitation = 2, RARE factor = 8). The total scan
time was 85 minutes per animal.
2.3. Data analysis
All images from the longitudinal and cross-sectional
experiments were processed using the manual tracing tool and
edge editing function provided by ANALYZE (Biomedical
Imaging Resource, Mayo Foundation, Rochester, Minnesota). Two imaging analysts who were blind to the test
subjects manually delineated the regions of interest for the
right lateral ventricle, the left lateral ventricle, the third ventricle, the cerebral aqueduct, and the fourth ventricle. The size
of each ventricular compartment was determined by the total
voxel volumes of the ventricle from multiple slices. The sum
of the volumes of all the chambers determined the total ventricular volume. The whole brain volume was also measured
as a reference, covering slices from the rostral end of the forebrain to the caudal end of the cerebellum. The cerebrum on
each slice was manually traced and the sizes from all slices
were summed to obtain the whole brain volume. In order to
compare the variations of ventricles with other brain regions,
the size of the hippocampus was also measured using procedures similar to those described above. The hippocampus
was distinguished from the image because the hippocampus
is bordered by its input (the fimbria) and output (the subiculum), which had a lower signal intensity on T2WI as opposed
to the hippocampus due to the two structures being white matter. The two sets of data acquired by the two analysts were
examined for interrater reliability using Pearsons correlation
tests. The agreement of the data was confirmed by the correlation coefficient, which was 0.92 0.04 (mean standard
deviation), and p < 0.0001 for all tests. Thus, the values of the two sets were averaged for subsequent data
processing.
To prevent measurements of the ventricle size being confounded by changes in the total brain size (Herbert et al.,
2003; Mathalon et al., 1993; OBrien et al., 2006), the volumes of the ventricles were normalized according to the
brain volume or length to yield two sets of ratios: the ventricular volume/brain volume (VV) ratio and the ventricular
volume/brain length (VL) ratio. The brain length was defined

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

2301

as the distance between the forebrain and the cerebellum on

the midsagittal slice. The two volume measures (VV and VL
ratios) represent fractions of the brain size or ratios to the
length, but for simplicity they are referred to here as ventricular volume measures. The hippocampal volume was also
normalized to yield two sets of ratios for comparison with
the ventricular measures: the hippocampal volume/brain volume (HV) ratio and the hippocampal volume/brain length
(HL) ratio.
2.4. Statistical analysis
Repeated-measures ANOVAs were applied to the longitudinal data, whereas one-way ANOVAs were applied to the
cross-sectional data followed by Fishers post hoc tests. The
level of significance was set at p < 0.05. Logarithmic regression analyses were performed to assess the growth pattern of
the ventricular system: y = b1 ln(x) + b0 , where y represents
the volume and x represents the age. The value of R2 derived
from the analysis denotes the degree to which the data can be
explained by the regression model.

3. Results
3.1. Views of the ventricular system
Fig. 1 shows various structures constituting the ventricular
system: the fourth ventricle of the brainstem (Fig. 1A), the
cerebral aqueduct and the fourth ventricle at another brainstem level (Fig. 1B), the third and lateral ventricles of the
midbrain (Fig. 1C), and the third and lateral ventricles of the
forebrain (Fig. 1D). A 3D view of the entire ventricular system obtained by reconstructing the 3D images is shown in
Fig. 1E.

Fig. 2. Changes in total brain size with age: total brain volume (A) and
length (B).

3.2. The volume and length of the whole brain increases

with age
The age-related changes in the volume and length of the
whole brain are shown in Fig. 2, with the raw data summarized in Table 1A. Both brain volume and length increased
monotonically with age, as indicated by repeated-measures
ANOVAs (brain volume: (F(3, 6) = 43.20, p < 0.0001)); brain
length: (F(3, 6) = 67.57, p < 0.0001)).
3.3. The overall volume of the ventricular system
increases with age
The age-related changes in the overall volume of the
entire ventricular system are shown in Fig. 3A and C, and

Fig. 1. Example 2D T2WI images showing the fourth ventricle of the brainstem (A), the fourth ventricle and cerebral aqueduct of the midbrain (B), the third
and lateral ventricles of the midbrain (C), and the third and lateral ventricles of the forebrain (D). The 3D view of the entire ventricular system (E) depicts the
left lateral ventricle in dark blue, the right lateral ventricle in light blue, the third ventricle in red, the cerebral aqueduct in yellow, and the fourth ventricle in
green. The green cursors indicate the covered regions of interest. (For interpretation of the references to color in this figure legend, the reader is referred to the
web version of the article.)

2302

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

Table 1
Raw volume of the brain, ventricular system, and hippocampus. CA, cerebral aqueduct.
Week 3
A. Whole brain size changes with age
Length (mm)
11.34 0.05
Volume (mm3 )
423.79 2.33

Week 6

Week 12

Week 24

Week 48

Week 75

Week 100

11.86 0.07
451.58 2.39

12.22 0.10
462.17 2.73

12.32 0.06
469.66 1.78

12.46 0.04
467.56 2.83

12.98 0.08
471.92 1.52

12.8 0.04
478.54 1.50

B. Ventricular volume changes with age (mm3 )

4th
1.89 0.07
2.28
CA
0.91 0.03
1.10
3rd
3.38 0.10
3.90
Left
2.76 0.25
3.86
Right
3.19 0.27
4.18
Total
12.14 0.45
15.31

0.08
0.05
0.18
0.30
0.33
0.82

C. Hipppcampal volume changes with age (mm3 )

16.85 0.21
18.46 0.76

2.41
1.12
3.85
4.58
4.67
16.62

0.06
0.03
0.14
0.38
0.35
0.87

19.57 0.56

listed in Table 1B. Repeated-measures ANOVAs indicated

that age significantly affected the VV ratio (F(4, 24) = 47.07,
p < 0.0001)) and the VL ratio (F(4, 24) = 34.31, p < 0.0001)).
Fishers post hoc tests used to compare all pairs from the
seven time points indicated that the volume increased significantly with age except for the following consecutive age
pairs: week 12 vs. week 24, week 48 vs. week 75, and week
75 vs. week 100.
Additional ANOVAs were performed with week 3 and
week 6 data removed. The results were similar. The
trend of increasing ventricular size remained clear at
later ages even when young ages were not taken into
account.

2.35
1.03
4.20
4.89
4.96
17.43

0.05
0.04
0.19
0.34
0.42
0.87

20.66 0.41

2.43
1.06
4.43
5.13
5.55
18.61

0.08
0.08
0.22
0.22
0.40
0.78

21.78 0.47

2.38
1.08
4.84
5.70
5.95
19.95

0.10
0.07
0.17
0.27
0.39
0.77

22.56 0.36

2.09
1.16
4.84
5.97
6.24
20.30

0.18
0.11
0.20
0.28
0.45
0.82

23.48 0.81

3.4. The effects of age are disproportional across

ventricular compartments
The changes in the volumes of the individual ventricular compartments are shown in Fig. 3B and D, and listed in
Table 2. ANOVAs revealed that the sizes of the left and right
lateral ventricles increased over time: F(4, 24) = 40.15 for the
VV ratio and F(4, 24) = 39.83 for the VL ratio (p < 0.0001)
in the left lateral ventricle; and F(4, 24) = 47.01 for the VV
ratio and F(4, 24) = 36.8 for the VL ratio (p < 0.0001) in the
right lateral ventricle. Fishers post hoc tests indicated that
the volume increased with age, except for the consecutive
age pair of week 12 vs. week 24.

Fig. 3. Changes in VV (A, B) and VL (C, D) ratios in the longitudinal experiment. (A) VV ratio changes with age of the total ventricular system, (B) VV ratio
changes with age of the individual ventricle compartments, (C) VL ratio changes with age of the total ventricular system, and (D) VL ratio changes with age
of the individual ventricle compartments. CA, cerebral aqueduct.

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

2303

Table 2
Percentage changes in three ventricular volume measures (volume, VV ratio, and VL ratio) relative to week 3. CA, cerebral aqueduct.
Week 6

Week 12

Week 24

Week 48

Week 75

W eek 100

Volume
4th
CA
3rd
Left
Right
Total
Hippo.

21.93
20.39
15.51
41.27
31.87
26.30
9.54

19.22
14.86
15.07
18.15
18.41
15.98
13.94

28.35
22.65
14.19
68.95
47.68
37.06
16.17

17.76
12.30
14.56
116.29
19.85
15.99
12.79

25.37
13.58
24.28
79.97
56.13
43.73
22.63

17.87
16.33
14.21
111.83
19.30
15.91
11.88

29.40
16.63
31.15
91.05
75.39
53.79
29.41

17.27
110.04
16.00
116.09
19.92
16.95
16.63

27.10
19.43
43.34
111.20
88.29
64.73
34.02

19.52
19.98
14.80
114.80
110.54
15.86
12.80

12.25
31.47
41.02
118.02
94.34
65.77
40.19

19.89
116.80
15.87
118.90
111.74
15.65
16.72

VV ratios
4th
CA
3rd
Left
Right
Total
Hippo.

14.46
12.99
8.46
32.66
23.85
18.59
2.80

18.78
14.65
15.02
17.99
18.24
15.91
13.68

17.72
12.45
4.70
55.10
35.37
25.70
6.51

17.22
11.89
14.15
115.52
18.85
15.63
12.49

13.24
2.55
12.17
62.41
41.03
29.77
10.65

17.63
15.96
14.03
110.70
19.08
15.83
11.68

17.48
5.68
19.05
73.01
59.21
39.50
17.28

17.45
18.96
16.37
114.05
110.07
16.91
13.08

14.35
7.24
28.86
89.75
69.35
48.10
20.30

19.43
18.93
15.27
113.55
110.66
16.37
11.90

0.30
16.33
25.26
93.59
72.63
47.22
24.30

19.21
114.11
16.31
117.26
111.66
16.31
15.49

VL ratios
4th
CA
3rd
Left
Right
Total
Hippo.

16.53
15.09
10.52
35.25
26.13
20.82
4.69

18.58
14.51
15.23
18.52
18.10
15.92
13.34

19.07
13.77
6.03
57.27
37.12
27.31
7.83

16.96
11.06
14.53
116.45
19.33
16.08
12.83

15.32
4.55
14.38
65.64
43.64
32.26
12.88

16.87
15.87
13.75
110.87
18.20
15.16
11.82

17.77
6.04
19.34
73.99
59.55
39.95
17.75

16.58
18.82
15.29
115.02
18.66
16.25
13.00

10.97
4.15
25.32
84.87
64.53
43.99
17.10

18.05
18.22
14.74
114.09
19.27
15.51
12.55

0.72
16.26
24.80
93.18
71.89
46.70
24.09

18.69
114.77
15.47
117.79
110.10
15.30
16.34

ANOVAs indicated that the third ventricle also enlarged

with age: F(4, 24) = 13.15 for the VV ratio and F(4,
24) = 11.13 for the VL ratio (p < 0.0001). Fishers post
hoc tests used to compare all pairs from the seven time
points indicated that the volume increased significantly
with age except for the following age pairs: week 3 vs.
week 12, week 6 vs. week 12, week 6 vs. week 24, week
24 vs. week 48, week 24 vs. week 100, week 48 vs.
week 75, week 48 vs. week 100, and week 75 vs. week
100 (p > 0.05). ANOVAs indicated that the volume of the
fourth ventricle varied with age: F(4, 24) = 3.8 for the VV
ratio and F(4, 24) = 4.03 for the VL ratio (p < 0.01). The
effect of age on the fourth ventricle expansion was mainly
restricted to middle age, with Fishers post hoc tests revealing that the volumes at weeks 3 and 100 were smaller than
those at other ages (p < 0.05). The cerebral aqueduct was
the only compartment whose size did not vary with age
(p > 0.05).

3.6. Ventricular enlargement described as a logarithmic

function of age

3.5. The size of the hippocampus increases with age

In the cross-sectional experiment, one-way ANOVAs indicated that the total ventricle size differed among the five age
groups irrespective of whether the VV ratio (F(4, 38) = 16.2,
p < 0.001) or the VL ratio (F(4, 38) = 17.6, p < 0.0001) was
used. Fishers post hoc tests showed that the whole ventricle
was significantly smaller at week 3 than at the other ages
(all p < 0.05), and significantly smaller at week 6 than at
weeks 18, 32, and 44 (all p < 0.05), with its size not differing
between weeks 18, 32, and 44. The statistical results for the

In order to compare the variations of ventricles with other

brain regions, the size of the hippocampus was also measured. The age-related changes in the hippocampal volume
are summarized in Table 1C. The volume of the hippocampus increased monotonically with age. The normalized HV
and HL ratios also exhibited age-related increases in the hippocampal size.

Considering that the interval of the examined time points

is not the same (3, 6, 12, 24, 27, 25 weeks between the time
points week 3, 6, 12, 24, 48, 75, and 100), the rate of ventricular enlargement was higher in young mice, and lower in
old mice. Regression analysis indicated that the expansion
of the total ventricular system with age can be described
by a logarithmic model (Fig. 4A for unadjusted volume:
R2 = 0.71, p < 0.0001 for b1 and b0 ; Fig. 4B for the VV ratio:
R2 = 0.63, p < 0.0001 for b1 and b0 ; and Fig. 4C for the VL
ratio: R2 = 0.58, p < 0.0001 values for b1 and b0 ). Note that,
due to differences in the units, the b0 and b1 values for the
three regression analysis can not be compared.
3.7. The ventricular size changes evaluated by the
cross-sectional experiment

2304

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

Fig. 4. Enlargement of the ventricular system fitted by a logarithmic curve

with the unadjusted volume (A), VV ratio (B), and VL ratio (C).

left and right lateral ventricles were similar to those for the
whole ventricle. For the fourth, third, and cerebral aqueduct,
ANOVAs followed by Fishers post hoc tests did not indicate
incremental size increases with age. The data obtained in the
cross-sectional experiment are shown in Fig. 5.

4. Discussion
The current study documented longitudinal changes in
the ventricle volume in normal aging mice over their lifespan. The results showed that the overall ventricle volume
increased with age, with the expansion beginning during the
early life stages and continuing to old age. The enlargement was faster in the early ages, which was supposed to
be development related effects. The size increase was slower
in the later times, which should be aging-related changes.
The enlargement in ventricular spaces was disproportional in
the various ventricular chambers, with it being largest in the
lateral ventricles. The reported data represent a biomarker
benchmark for normal aging mice under unmodified conditions. This provides a foundation for evaluating the validity

of AD mouse models or the effects of potential drugs. The

considerable physiological ventricular enlargement in normal
aging must be carefully differentiated from the enlargement
induced in AD animal models.
The size of the mouse brain increases with age throughout the lifetime (Maheswaran et al., 2009), which probably
indicates that the rodent cranium is capable of continual
expansion (unlike the human cranium). Such developmental cranial expansion makes extreme caution necessary when
using rodent volume measures to assess AD animal models
or drug effects.
Measuring the hippocampus helped to better characterize ventricular enlargement with age. We found that (1) the
hippocampus was overall 35 mm3 larger than the whole
ventricle and (2) the hippocampus enlarged throughout the
lifetime even when the data were normalized as HV and
HL ratios. This hippocampal size enlargement may be due
to hypertrophy. Nevertheless, it is evident that the ventricular system expanded at a higher rate than the hippocampus
because the hippocampus was 4.7 mm3 larger than the whole
ventricle at week 3, but this difference had reduced to 3.2 mm3
at week 100 (Table 1). In terms of percentage changes, the
hippocampus enlarged by 40.2% in volume, 24.3% in HV
ratio, and 24.1% in HL ratio at week 100 relative to week
3, with the corresponding changes being 65.8%, 47.2%, and
46.7% for the whole ventricle.
In humans, atrophy of hippocampus has been referred to
as a cause of ventricular enlargement during aging. However,
our results indicate that, in mice, the mechanisms underlying ventricular expansion may be different. We suspect
that it is either caused by brain atrophy of other brain areas
than the hippocampus or possibly enhanced CSF releasing rates. The secretion of CSF tends to be upregulated
when pathological conditions are present (Redzic et al.,
2005).
Ventricular expansion has been assessed in at least three
AD mouse models: TASTPM, APP/presenilin 1 (PS1), and
APP/PS2. In the TASTPM model, the expansion rate of the
whole ventricle from month 6 to month 11 was estimated
to be 17% in the transgenic mice and 10% in the wild type
(Maheswaran et al., 2009). The ventricular expansion rate in
the wild type was slightly higher than we found in our normal aging mice (approximately 7% from month 6 to month
12). The aged APP/PS1 mice showed dilated ventricles at
the midbrain level but not in the lateral ventricles (Delatour
et al., 2006). The alterations were different from human AD
features, which might indicate that the APP/PS1 model only
partially resembles human AD pathologies. The APP/PS2
mice had a slightly larger ventricle, but the expansion rate
did not differ from that of normal aging mice (von Kienlin et
al., 2005). The change in ventricle size has not been fully
examined in the PDAPP (APP under control of plateletderived growth factor promoter) model, but a few studies
have demonstrated that these mice show brain atrophy that is
caused by the lack of development of white matter tracts in
young animals (Gonzalez-Lima et al., 2001; Redwine et al.,

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

2305

Fig. 5. Changes in VV and VL ratios in the cross-sectional experiment. (A) VV ratio changes with age of the total ventricular system, (B) VV ratio changes
with age of the individual ventricle compartments, (C) VL ratio changes with age of the total ventricular system, and (D) VL ratio changes with age of the
individual ventricle compartments. CA, cerebral aqueduct.

2003). Thus, the usefulness of measuring ventricular size in

the PDAPP model as an AD biomarker needs to be carefully
evaluated.
The expansion of the ventricular system with age
occurs disproportionally across individual compartments.
Our results indicate that the lateral ventricle expands faster
than the overall average whereas the third ventricle expands
in a more gradual, stepwise manner with age. The fourth
ventricle showed decreases in size at older ages whereas the
size of the cerebral aqueduct remained constant across the
lifetime. Data on changes in compartment sizes in the ventricular system are rare, but our findings are consistent with
those from a human study (Walhovd et al., 2005) in that the
size increase with age was largest for lateral ventricles followed by the third ventricle, with no significant size change
for the fourth ventricle.
In normal aging mice, the lateral ventricles enlarged by
approximately 7293% at week 100 relative to week 3, giving a weekly expansion rate of 0.740.96%. In normal aging
Rhesus monkeys, the CSF volume increased by 0.28% per
year from age 5 years to approximately 25 years (Andersen et
al., 1999). In mouse lemur primates, the CSF volume doubled
within 12 years in many animals, regardless of age (Dhenain
et al., 2000). Even though the examined periods have differed
among studies, the data appear to indicate that the expansion rates also differ across species. The essential differences
among species in ventricle sizes suggest that extreme care

must be taken when extrapolating data from mouse models

to human patients.
The lifetime ventricular expansion pattern of normal aging
mice is continual, indicating (1) that the expansion process
occurs continually throughout the lifetime, (2) the expansion
begins as early as 3 weeks after birth, and (3) there is no apparent acceleration of expansion at older ages. Due to the lack
of directly comparable studies, it remains to be determined
whether these findings also apply to humans. There have been
reports showing that ventricular expansion in humans accelerates at older ages (Walhovd et al., 2005) but other authors
report otherwise (Foundas et al., 1998). We speculate that
the longevity of the human species is responsible for these
inconsistencies, since any expansion of the ventricle over the
human lifespan (i.e., over several decades) would be more
variable than a change over the 2-year lifespan of the mouse.
There is little information on when ventricular expansion
begins. Our results suggest that ventricular expansion begins
in very young animals and lasts throughout their lifetime. In
humans, due to the difficulties of collecting lifetime longitudinal data, the earliest time points included in time frames
have been from ages of 2040 years (Foundas et al., 1998;
Scahill et al., 2003; Walhovd et al., 2005), and so investigations of younger subjects such as the children are needed
to extrapolate our findings to humans. However, our results
suggest that it is always necessary to control for age when
measuring ventricular volumes, irrespective of the age range

2306

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

chosen for brain characterization. This view differs from the

general assumption that controlling for age is only necessary
during the early developmental and aging stages. Instead, it
appears that age affects ventricle sizes even during physiologically stable stages such as adulthood.
Adjustment for brain volumes is deemed necessary
to determine the real effects of an independent variable
(Mathalon et al., 1993; OBrien et al., 2006), although it
is possible that reliability is decreased when ratios are used
(Arndt et al., 1991). In our study we controlled for the influence of variation in the total brain size on ventricle sizes by
normalizing the ventricular volumes into ratios relative to the
total brain volume (i.e., VV ratio) or the total brain length
(i.e., VL ratio). When the volume was not corrected, the
age-related effects revealed by ANOVA had a slightly higher
power/significance, whereas the adjusted volume ratios had
lower power/significance. Despite this, due to the significance
of each volumetry analysis reaching the ceiling of p < 0.0001,
the overall ventricle change with age was almost identical
for the two ratio volume measures and the unadjusted volume measurement (correlation coefficients for VV ratio vs.
VL ratio, VV ratio vs. unadjusted volume, and VL ratio vs.
unadjusted volume were all >0.98 with a significance level
p < 0.0001). This suggests that the effects of age on the ventricular volume are substantial irrespective of whether the
influence of the total brain size is considered.
The statistical results were very similar for the two volume measures used in the study. However, the percentage
increases in the ratios as listed in Table 2 were overall smaller
than the unadjusted volume measurement, which could be
explained by the age-related modification of brain volume or
brain length. As evident in Table 2, the VV and VL ratios
were approximately 20% larger at week 6 and 45% larger at
week 100 than at week 3. In contrast, the unadjusted total
ventricle volume was 26% larger at week 6 and 65% larger
at week 100 than at week 3.
The longitudinal and cross-sectional data were compared
for the period during which the two sets of data overlapped. There were two major differences in the results:
(1) the standard errors were overall larger in the crosssectional experiment than in the longitudinal experiment,
and (2) the ventricle expansion was more continual in the
longitudinal experiment than in the cross-sectional experiment. The difficulty of tracking age for a long time in
human studies has made cross-sectional studies more popular
than longitudinal ones. However, a longitudinal experimental
design can be advantageous (Scahill et al., 2003). Although
the cross-sectional approach is inevitable in human studies, strategies that adopt a combined cross-sectional and
longitudinal approach might be preferable over a purely
cross-sectional approach (Scahill et al., 2003).
Despite the inherent differences in brain growth patterns
across species, ventricular enlargement remains a useful
structural biomarker for assessing the progression of AD.
Translating this biomarker to animal research will greatly
assist the development of therapies for AD, such as estab-

lishing more reliable AD animal models. The longitudinal

characterization of changes in ventricle size in the current
study constitutes a building block for the translation. It is
hoped that the data reported here will be useful to researchers
that apply this biomarker to validate AD models or the effects
of drugs, and also serve as a quantitative basis to compare
brain volume measurements of other mouse strains and models of brain disorders.

Conict of interest
There were no actual or potential conflicts of interest associated with the work.

Acknowledgements
We acknowledge technical support from the Functional
and Micro-Magnetic Resonance Imaging Center supported
by the National Research Program for Genomic Medicine,
National Science Council, Taiwan (NSC 97-3112-B-001009). We also thank Chao-Zi Hao and Zi-Jun Lin for data
analysis.
All authors have reviewed the contents of the manuscript
being submitted, approve of its contents and validate the
accuracy of the data.

References
Andersen, A.H., Zhang, Z., Zhang, M., Gash, D.M., Avison, M.J., 1999.
Age-associated changes in rhesus CNS composition identified by MRI.
Brain Res. 829, 9098.
Arndt, S., Cohen, G., Alliger, R.J., Swayze, V.W., Andreasen, N.C., 1991.
Problems with ratio and proportion measures of imaged cerebral structures. Psychiatry Res. 40, 7989.
Bradley, K.M., Bydder, G.M., Budge, M.M., Hajnal, J.V., White, S.J.,
Ripley, B.D., Smith, A.D., 2002. Serial brain MRI at 3-6 month intervals as a surrogate marker for Alzheimers disease. Br. J. Radiol. 75,
506513.
de Leon, M.J., DeSanti, S., Zinkowski, R., Mehta, P.D., Pratico, D., Segal,
S., Clark, C., Kerkman, D., DeBernardis, J., Li, J., Lair, L., Reisberg, B.,
Tsui, W., Rusinek, H., 2004. MRI and CSF studies in the early diagnosis
of Alzheimers disease. J. Intern. Med. 256, 205223.
de Leon, M.J., Mosconi, L., Blennow, K., DeSanti, S., Zinkowski, R., Mehta,
P.D., Pratico, D., Tsui, W., Saint Louis, L.A., Sobanska, L., Brys, M.,
Li, Y., Rich, K., Rinne, J., Rusinek, H., 2007. Imaging and CSF studies
in the preclinical diagnosis of Alzheimers disease. Ann. N Y Acad. Sci.
1097, 114145.
Delatour, B., Guegan, M., Volk, A., Dhenain, M., 2006. In vivo MRI and
histological evaluation of brain atrophy in APP/PS1 transgenic mice.
Neurobiol. Aging 27, 835847.
Dhenain, M., Michot, J.L., Privat, N., Picq, J.L., Boller, F., Duyckaerts, C.,
Volk, A., 2000. MRI description of cerebral atrophy in mouse lemur
primates. Neurobiol. Aging 21, 8188.
Foundas, A.L., Zipin, D., Browning, C.A., 1998. Age-related changes of
the insular cortex and lateral ventricles: conventional MRI volumetric
measures. J. Neuroimaging 8, 216221.
Fox, N.C., Cousens, S., Scahill, R., Harvey, R.J., Rossor, M.N., 2000. Using
serial registered brain magnetic resonance imaging to measure disease

C.-C.V. Chen et al. / Neurobiology of Aging 32 (2011) 22992307

progression in Alzheimer disease: power calculations and estimates of
sample size to detect treatment effects. Arch. Neurol. 57, 339344.
Frisoni, G.B., Delacourte, A., 2009. Neuroimaging outcomes in clinical trials
in Alzheimers disease. J. Nutr. Health. Aging 13, 209212.
Gonzalez-Lima, F., Berndt, J.D., Valla, J.E., Games, D., Reiman, E.M., 2001.
Reduced corpus callosum, fornix and hippocampus in PDAPP transgenic
mouse model of Alzheimers disease. Neuroreport 12, 23752379.
Hampel, H., Burger, K., Teipel, S.J., Bokde, A.L., Zetterberg, H., Blennow,
K., 2008. Core candidate neurochemical and imaging biomarkers of
Alzheimers disease. Alzheimers Dement. 4, 3848.
Herbert, M.R., Ziegler, D.A., Deutsch, C.K., OBrien, L.M., Lange, N.,
Bakardjiev, A., Hodgson, J., Adrien, K.T., Steele, S., Makris, N.,
Kennedy, D., Harris, G.J., Caviness, V.S., 2003. Dissociations of cerebral
cortex, subcortical and cerebral white matter volumes in autistic boys.
Brain 126, 11821192.
Jack, C.R., Shiung, M.M., Gunter, J.L., OBrien, P.C., Weigand, S.D., Knopman, D.S., Boeve, B.F., Ivnik, R.J., Smith, G.E., Cha, R.H., Tangalos,
E.G., Petersen, R.C., 2004. Comparison of different MRI brain atrophy
rate measures with clinical disease progression in AD. Neurology 62,
591600.
Jack, C.R., Shiung, M.M., Weigand, S.D., OBrien, P.C., Gunter, J.L., Boeve,
B.F., Knopman, D.S., Smith, G.E., Ivnik, R.J., Tangalos, E.G., Petersen,
R.C., 2005. Brain atrophy rates predict subsequent clinical conversion
in normal elderly and amnestic MCI. Neurology 65, 12271231.
Kalpouzos, G., Chetelat, G., Baron, J.C., Landeau, B., Mevel, K., Godeau,
C., Barre, L., Constans, J.M., Viader, F., Eustache, F., Desgranges, B.,
2009. Voxel-based mapping of brain gray matter volume and glucose
metabolism profiles in normal aging. Neurobiol. Aging 30, 112124.
Kunstyr, I., Leuenberger, H.G., 1975. Gerontological data of C57BL/6J
mice. I. Sex differences in survival curves. J. Gerontol. 30, 157162.
Luxenberg, J.S., Haxby, J.V., Creasey, H., Sundaram, M., Rapoport, S.I.,
1987. Rate of ventricular enlargement in dementia of the Alzheimer type
correlates with rate of neuropsychological deterioration. Neurology 37,
11351140.
Maheswaran, S., Barjat, H., Rueckert, D., Bate, S.T., Howlett, D.R., Tilling,
L., Smart, S.C., Pohlmann, A., Richardson, J.C., Hartkens, T., Hill, D.L.,
Upton, N., Hajnal, J.V., James, M.F., 2009. Longitudinal regional brain
volume changes quantified in normal aging and Alzheimers APP x PS1
mice using MRI. Brain Res. 1270, 1932.
Mathalon, D.H., Sullivan, E.V., Rawles, J.M., Pfefferbaum, A., 1993. Correction for head size in brain-imaging measurements. Psychiatry Res.
50, 121139.
Mu, Q., Xie, J., Wen, Z., Weng, Y., Shuyun, Z., 1999. A quantitative MR
study of the hippocampal formation, the amygdala, and the temporal
horn of the lateral ventricle in healthy subjects 40 to 90 years of age.
AJNR Am. J. Neuroradiol. 20, 207211.

2307

Nestor, S.M., Rupsingh, R., Borrie, M., Smith, M., Accomazzi, V., Wells,
J.L., Fogarty, J., Bartha, R., 2008. Ventricular enlargement as a possible measure of Alzheimers disease progression validated using
the Alzheimers disease neuroimaging initiative database. Brain 131,
24432454.
OBrien, L.M., Ziegler, D.A., Deutsch, C.K., Kennedy, D.N., Goldstein,
J.M., Seidman, L.J., Hodge, S., Makris, N., Caviness, V., Frazier, J.A.,
Herbert, M.R., 2006. Adjustment for whole brain and cranial size in
volumetric brain studies: a review of common adjustment factors and
statistical methods. Harv. Rev. Psychiatry 14, 141151.
Redwine, J.M., Kosofsky, B., Jacobs, R.E., Games, D., Reilly, J.F., Morrison,
J.H., Young, W.G., Bloom, F.E., 2003. Dentate gyrus volume is reduced
before onset of plaque formation in PDAPP mice: a magnetic resonance
microscopy and stereologic analysis. Proc. Natl. Acad. Sci. U.S.A. 100,
13811386.
Redzic, Z.B., Preston, J.E., Duncan, J.A., Chodobski, A., SzmydyngerChodobska, J., 2005. The choroid plexus-cerebrospinal fluid system:
from development to aging. Curr. Top. Dev. Biol. 71, 152.
Resnick, S.M., Goldszal, A.F., Davatzikos, C., Golski, S., Kraut, M.A., Metter, E.J., Bryan, R.N., Zonderman, A.B., 2000. One-year age changes in
MRI brain volumes in older adults. Cereb. Cortex 10, 464472.
Ridha, B.H., Anderson, V.M., Barnes, J., Boyes, R.G., Price, S.L., Rossor,
M.N., Whitwell, J.L., Jenkins, L., Black, R.S., Grundman, M., Fox,
N.C., 2008. Volumetric MRI and cognitive measures in Alzheimer
disease: comparison of markers of progression. J. Neurol. 255,
567574.
Scahill, R.I., Frost, C., Jenkins, R., Whitwell, J.L., Rossor, M.N., Fox, N.C.,
2003. A longitudinal study of brain volume changes in normal aging
using serial registered magnetic resonance imaging. Arch. Neurol. 60,
989994.
Schott, J.M., Price, S.L., Frost, C., Whitwell, J.L., Rossor, M.N., Fox, N.C.,
2005. Measuring atrophy in Alzheimer disease: a serial MRI study over
6 and 12 months. Neurology 65, 119124.
von Kienlin, M., Kunnecke, B., Metzger, F., Steiner, G., Richards, J.G.,
Ozmen, L., Jacobsen, H., Loetscher, H., 2005. Altered metabolic profile
in the frontal cortex of PS2APP transgenic mice, monitored throughout
their life span. Neurobiol. Dis. 18, 3239.
Walhovd, K.B., Fjell, A.M., Reinvang, I., Lundervold, A., Dale, A.M., Eilertsen, D.E., Quinn, B.T., Salat, D., Makris, N., Fischl, B., 2005. Effects
of age on volumes of cortex, white matter and subcortical structures.
Neurobiol. Aging 26, 12611270.
Wang, D., Chalk, J.B., Rose, S.E., de Zubicaray, G., Cowin, G., Galloway,
G.J., Barnes, D., Spooner, D., Doddrell, D.M., Semple, J., 2002. MR
image-based measurement of rates of change in volumes of brain structures. Part II: application to a study of Alzheimers disease and normal
aging. Magn. Reson. Imaging 20, 4148.