Accepted Manuscript

Title: The Evans’ Index revisited: New cut-off levels for use in
radiological assessment of ventricular enlargement in the
elderly.

Authors: Maiken K. Brix, Eric Westman, Andrew Simmons,

Geir Andre Ringstad, Per Kristian Eide, Kari Wagner-Larsen,
Christian M. Page, Valeria Vitelli, Mona K. Beyer

PII: S0720-048X(17)30297-8
DOI: http://dx.doi.org/doi:10.1016/j.ejrad.2017.07.013
Reference: EURR 7902

To appear in: European Journal of Radiology

Received date: 20-4-2017

Revised date: 12-7-2017
Accepted date: 17-7-2017

Please cite this article as: Brix Maiken K, Westman Eric, Simmons Andrew, Ringstad
Geir Andre, Eide Per Kristian, Wagner-Larsen Kari, Page Christian M, Vitelli Valeria,
Beyer Mona K.The Evans’ Index revisited: New cut-off levels for use in radiological
assessment of ventricular enlargement in the elderly.European Journal of Radiology
http://dx.doi.org/10.1016/j.ejrad.2017.07.013

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Title:
The Evans’ Index revisited: New cut-off levels for use in radiological assessment of
ventricular enlargement in the elderly.

Authors:
Maiken K Brix1,2*, Eric Westman3,4, Andrew Simmons3,4,5,6, Geir Andre Ringstad7,8. Per
Christian Eide9,10, Kari Wagner-Larsen1, Christian M Page11,12, Valeria Vitelli13, Mona K
Beyer8,14

1
Department of Radiology, Haukeland University Hospital, Bergen, Norway
2
Department of Clinical Medicine (K1), University of Bergen, Bergen, Norway
3
Division of Clinical Geriatrics, Center for Alzheimer Research, Department of
Neurobiology, Care Sciences and Society, Karolinska Institutet, Stockholm, Sweden
4
Institute of Psychiatry, King’s College London, UK
5
NIHR Biomedical Research Centre for Mental Health, London, UK
6
NIHR Biomedical Research Unit for Dementia, London, UK
7
Department of Radiology and Nuclear Medicine, Institute of Clinical Medicine, Faculty of
Medicine, University of Oslo, Oslo, Norway
8
Department of Radiology and Nuclear Medicine, Oslo University Hospital
9
Department of Neurosurgery, Institute of Clinical Medicine, Faculty of Medicine, University
of Oslo, Oslo, Norway
10
Department of Neurosurgery, Oslo University Hospital
11
Department of Neurology, Institute of Clinical Medicine, Faculty of Medicine, University of
Oslo, Oslo, Norway
12
Department of Neurology, Division of Surgery and Clinical Neuroscience, Oslo University
hospital, Oslo, Norway13Oslo Center for Biostatistics and Epidemiology, Department of
Biostatistics, University of Oslo, Oslo, Norway14Department of Life Sciences and Health,
Faculty of Health Sciences, Oslo and Akershus University College of Applied Sciences, Oslo,
Norway

Maiken Kirkegaard Brix, MD,

Phone: +47 936 28 389e-mail: [email protected]/ maiken.kirkegaard.brix@helse-

bergen.no Resident in Radiology at Haukeland University Hospital, Bergen Norway.

1
Abstract:
Background and purpose:
Assessment of ventricular enlargement is subjective and based on the radiologist’s
experience. Linear indices, such as the Evans Index (EI), have been proposed as markers of
ventricular volume with an EI ≥ 0.3 indicating pathologic ventricular enlargement in any
subject. However, normal range for EI measured on magnetic resonance imaging (MRI) scans
are lacking in healthy elderly according to age and sex. We propose new age and sex specific
cut-off values for ventricular enlargement in the elderly population.
Materials and methods:
534 participants (53% women) aged 65-84 years; 226 patients with Alzheimer’s disease
(AD), and 308 healthy elderly controls (CTR) from the AddNeuroMed and ADNI studies
were included. The cut-off for pathological ventricular enlargement was estimated from
healthy elderly categorized into age groups of 5 years range and defined as EI 97,5 percentile
(mean + 2SD). Cut-off values were tested on patients with Alzheimer’s disease and a small
sample of patients with probable idiopathic normal pressure hydrocephalus (iNPH) to assess
the sensitivity.
Results:
The range of the EI in healthy elderly is wide and 29% of the CTR had an EI of 0.3 or greater.
The EI increases with age in both CTR and AD, and the overall EI for women were lower
than for men (p<0.001). New EI cut off values for male/female: 65-69 years 0.34/0.32, 70-74
years 0.36/0.33, 75-79 years 0.37/0.34 and 80-84 years 0.37/0.36. When applying the
proposed cut-offs for EI in men and women aged 65 to 84, they differentiated between iNPH
and CTR with a sensitivity of 80% and for different age and sex categories of AD and CTR
with a sensitivity and specificity of 0 - 27% and 91-98%, respectively.
Conclusion:
The range of the EI measurements in healthy elderly is wide, and a cut-off value of 0.3 cannot
be used to differentiate between normal and enlarged ventricles in individual cases. The
proposed EI thresholds from the present study show good sensitivity for the iNPH diagnosis.

Abbreviations:
EI = Evans Index; CTR = Healthy elderly controls; AD = Alzheimer’s disease; VV =
Ventricular volume; ADNI = Alzheimer’s disease Neuroimaging Initiative; iNPH = idiopathic
normal pressure hydrocephalus

2
Keywords: Evans Index, Ventricular volume, Magnetic Resonance Imaging, Alzheimer’s
disease, idiopathic normal pressure hydrocephalus, Healthy elderly

Introduction

Tools for visual evaluation of the aging brain are published and widely used to evaluate
subjects with memory impairment or other symptoms of dementia. There are scales for the
evaluation of global atrophy [1] medial temporal atrophy [2] and for the classification of
white matter lesions [3]. For these scales, there are also published cut-off threshold values
according to age [4]. Visual inspection of the size and shape of the brain ventricles is a
standard procedure in radiologic evaluation of diagnostic computer tomography (CT) and
magnetic resonance imaging (MRI). Several studies have suggested that ventricular
enlargement may be an objective and sensitive measure of neuropathological change
associated with brain atrophy in mild cognitive impairment (MCI) and Alzheimer’s disease
(AD) at group level [5]. In the 2005 guidelines for diagnosis of idiopathic normal pressure
hydrocephalus, an EI of 0.3 or greater combined with gait dysfunction plus either urinary or
cognitive dysfunction is required prior to consideration of treatment with ventriculo-
periteoneal shunt [6].

However, assessment of ventricular enlargement is in most cases performed by subjective

means and is typically based on the radiologist’s experience.

The Evans Index (EI) [7] is an indirect linear measurement of ventricular size initially used on
pneumoencephalography in paediatric patients. In 1987, Sherman et al stated that the EI
normally is smaller than 0.3 in adults and that the index can be an objective form to diagnose
hydrocephalus [8]. Currently, the EI is applied as an indirect, surrogate marker of ventricular
volume (VV) in CT [9] and MRI [6] in adults.

The relationship between linear indices and ratios and true VV has been evaluated in several
studies using CT [10, 11] and MRI [12-14]. Two recent studies in adults have questioned the
reliability of the EI for assessment of ventricular size, and volumetric analyses of VV have
been suggested instead [11, 13]. Volumetric ventricular analyses are time consuming, require
specialized software, and are not available in every hospital. A quick and reliable
measurement of the ventricular size, to separate normal from pathologically enlarged

3
ventricles, is lacking in routine radiological evaluation. The EI is well known by radiologists
and easy to perform, with previously reported high reliability [15].

Several studies have proposed a normal range for EI in the elderly population, but many were
limited by small samples, combinations of sex and age ranges, and did not specify the method
of EI measurement [10, 12, 13].

The aim of this study is to define the upper normal value of EI (97,5 percentile) in healthy
elderly controls (CTR) according to sex and age. The cut off values are applied to compare EI
as a marker of pathological ventricular enlargement in iNPH and AD groups, to test the
method’s performance for the everyday routine radiological practice. The hypotheses are: 1)
the EI is not different in men and women, and 2) the EI increases with age, and differs in
patients with iNPH and AD compared to healthy elderly controls (CTR).

Methods

Subjects

AD and CTR

A total of 534 participants aged 65-84 years were included; 226 patients with AD (57 %
women) and 308 CTR (51 % women), see Table 1. Age did not differ between the two groups
(men/women p=0.081/0.295). Data were obtained from the AddNeuroMed (33% of the
participants) and the ADNI database (adni.loni.usc.edu). Participant recruitment and
eligibility criteria were similar in the two cohorts [16, 17]. AddNeuroMed is part of the
Innovative Medicines in Europe (InnoMed) European Union Sixth Framework program [18].
The ADNI was launched in 2003 as a public-private partnership, led by Principal Investigator
Michael W. Weiner, MD. The primary goal of ADNI has been to test whether serial MRI,
PET, other biological markers, and clinical and neuropsychological assessment can be
combined to measure the progression of MCI and early AD.

Briefly, a diagnosis of AD was based on the NINCDS-ADRDA and DSM-IV criteria for
probable AD, as well as a total Clinical Dementia Rating score [19] of ≥0.5. The inclusion
criteria for CTR were an MMSE score of between 24 and 30, a total Clinical Dementia Rating
score of 0 and a Geriatric Depression Scale score ≤5. For the CTR and AD groups, exclusion

4
criteria included significant neurological or psychiatric illness, significant unstable systemic
illness or organ failure and history of alcohol or substance abuse or dependence.

Both the Alzheimer’s disease Neuroimaging Initiative (ADNI) and the AddNeuroMed studies
had ethical approval obtained from each institution involved and the data were anonymized
before being shared.

iNPH

The Regional Ethics Committee (REK South-East; S-07237) and Institutional Review Board
(07/5869) approved the iNPH study. Inclusion was by written and oral informed consent. 21
patients (range 56-84 years, 11 female) referred to a tertiary hospital for pre-surgical work-up
of probable iNPH were included. Severity of symptoms was graded using a NPH grading
scale, which assesses the combined severity of gait disturbance, urinary incontinence, and
dementia [20]. Each component is graded from 1 to 5, giving a possible total score of 3
(worst) to 15 (best).

MRI Acquisition

The ADNI data was acquired at 55 sites in North America from 2004-2009, while the
AddNeuroMed data was acquired at six sites across Europe from 2006-2009. The
AddNeuroMed study was designed to be comparable to the ADNI study, so although data are
drawn from two studies, they are homogeneous and have previously been used in other
publications as combined cohorts [21]. A high-resolution sagittal 3-dimensional (3D) T1-
weighted MPRAGE sequence was acquired in the AddNeuroMed and ADNI studies (voxel
size1.1 x 1.1 x 1.2 mm3). Full brain and skull coverage was required for the MRI datasets and
detailed quality control was carried out on all MR images according to previously published
criteria [16]. For the iNPH group, ultrafast gradient echo T1-weighted imaging was obtained
with a Philips 3 T Achieva system (Philips Medical Systems, Best, The Netherlands) with
sequence parameters including TR/TE = 8.6ms/2.3ms, matrix 256 x 256 x 192 and voxel size
1.0/1.0/1.0 mm3.

The EI

The EI [7] was calculated on the individual T1-weighted axial images reconstructed from the
MR images. The width was measured on three consecutive axial slices and the slice with the

5
largest diameter at the maximal width of the frontal horns was selected. In the same slice the
largest internal diameter of the cranium (see Figure 1) was measured. If the measured
maximal width of the frontal horns was identical in two or more slices, the slice with the
largest internal diameter of the cranium was selected. The frontal horn diameter was divided
by the largest diameter of the cranium, thus correcting for different head size in each subject.
Mutually blinded operators measured the EI for AD and CTR for inter- and intra-rater
reliability measurements. Indices were measured at two time points; thereafter one single
operator measured all indices for all the participants.

Determination of upper normal cut-off values

EI cut-off thresholds for normal ventricular sizes were determined for each sex and 5-year age
ranges in which we had a good number of subjects in each age range. The cut-off’s were
found by calculating the mean  2SD [22], which is the same as the 95% reference interval.
Only the upper limit is interesting in this context as we are looking for upper normal levels of
EI.

Statistical methods

Age, MMSE scores and EI between in males and females were compared between CTR and
AD. Normally distributed data were analysed with either a t-test or ANOVA. When the data
were not normally distributed, a log-transformation was performed, and if the data were still
not normally distributed, the nonparametric Mann-Whitney U test was used. For analysis of
data regarding sex, a Chi square test was used. SPSS (version 22, IBM Corp. USA) statistical
analysis software package was used for all group comparison, with P-values smaller than 0.05
considered statistically significant. Neither group comparison between CTR and iNPH nor
AD and iNPH were performed, as the iNPH group was small and only used to test the
performance of the proposed cut-off’s. Sensitivity was calculated.

In addition EI was estimated as a function of age by using non-linear regression, in particular

a smoothing splines with 5 degrees of freedom and all observations as knots was used [23].
Separate estimates were determined for sex/diagnosis groups. Non-linear confidence bands
were estimated at 95% level with the following procedure: First, patients were stratified into
age groups including 10% of the observations each. Then, for each age group, the 2.5% and
97.5% percentiles were computed. Finally, the lower and upper bounds were respectively

6
estimated as a function of age by a non-linear regression of the 2.5% and 97.5% percentiles
with respect to the midpoint of the age intervals, using cubic smoothing splines with the same
settings. All analyses were performed in the R statistical analysis software package (v 3.1.3,
2015-03-09).

Inter- and intra-rater reliability was calculated with intraclass correlation coefficients in SPSS
(version 22, IBM Corp. USA) statistical analysis software package. The operator measuring
EI for the iNPH subjects were blinded to the cut off values, but not to the diagnosis.

Results

Age and sex

There were no significant age differences between the CTR and the AD group, mean age
being 75 years. As expected, the MMSE scores were different between CTR and AD (p
<0.001), with lower values for AD versus CTR for both males and females. The EI increased
with age in both CTRs and AD in males and females. This is shown in Figure 2. A general
result is the broad 95% confidence bands in all groups in both males and females. A
confidence band represents the uncertainty in an estimate of a curve based on limited or noisy
data. Confidence bands are a good numerical approximation to the confidence intervals for
the curves. The confidence bands were calculated as a smoothed bin-wise standard deviation,
with at least five observations in each bin.

Upper bound cut-off values for the EI in CTR at different age ranges are made for both men
and women, see Table 1. Values above the cut-off indicate enlarged ventricles.

Regarding normal reference values for healthy controls we found that the EI values in males
varied between 0.22 – 0.43 and females varied between 0.19 and 0.38 for the age group 65 –
85 years. Overall there were smaller EI for women than men (p<0.001) for all subjects.

iNPH

Median (with range) for NPH-score was 10 (4-13).

After MRI, 17/21 patients were treated with a shunt procedure, where 16 (94 %) responded
clinically defined by an improvement of at least two points at the NPH-score.

7
Diagnosis

In males, the mean EI was 0.29 in CTR and 0.31 in the AD group, whereas in females EI was
0.28 in CTR group and 0.29 in AD group. These group differences were significant in both
males and females (p<0.001), see Table 1. The mean for EI in the iNPH group was
(males/females) 0.37/0.37.

Sensitivity and specificity

Using the cut-off values presented in Table 2 the sensitivity and specificity of the cut-offs for
separating AD and iNPH from CTR were tested. For AD, we found a sensitivity of 0 – 27 %
depending on sex and age range examined, with the lowest sensitivity for males 75-79 years
and females 80-84years and the highest sensitivity for females 65-69 years, and a specificity
of 91 - 98%, The sensitivity for the iNPH group was 80%.

Inter-/intra-rater reliability

All reliability analyses were found to be excellent with a ICC >0.9 both for inter-rater and
intra-rater analyses.

Discussion

The main finding of our study is that there is a wide range of EI in healthy elderly. Cut-off
values based on healthy controls aged 65-85 shows that for age groups 65-69, 70-74, 75-79
and 80-84, the upper bound for EI was higher than 0.3 in both men and women.

The new EI cut-off values suggested here, cannot be applied to separate AD from CTR due to
low sensitivity, but may separate CTR from iNPH with high sensitivity.

A recent study by Missori et al concluded that an EI > 0.3 reflects an underlying neurological
condition in every individual [24]. According to our study, this would imply that 29% of all
healthy elderly would wrongfully be suspected to have an underlying neurological condition.
However, there is one important flaw in the statistics of the Missori paper, in which only the
average EI for each age group was presented, thus ignoring the normal variation in subjects of
the same age.

8
The EI values measured in this study are in line with results from previous MRI studies in
healthy elderly subjects. If we combine the CTR female and male values from all age groups
in our study, mean EI  SD is 0.28  0.03 (NO: 308, M/F: 150/158, age: 75  7). Ishii et al
[12] found EI to be 0.26  0.003 (NO: 34, M/F: 21/13, age 75  5) and Ambarki et al [13]
found EI to be 0.28  0.03 (NO: 46, M/F 18/28, age 71  6). Nonetheless, these are not
suitable as cut-off values for application in single subjects, as the merely represent averages
of EI values for both sexes and any age group combined.

Several MRI studies examining the gender effects on brain structures, including VV in
healthy ageing, have been published. Some studies found a different age-related regional
pattern across different brain structures in men and women [25], while others found no gender
differences [26]. In our study, we found that women at all ages, and in both the CTR and AD
groups, have lower EI and LVV than men, and, in addition, the shape of the curves for EI are
different in men and women implying that VV changes differently in males and females as we
grow older. This has also been shown in a previous study of healthy elderly subjects [27].
This is in contrast to another study which showed that while men have larger brain volume,
CSF volume and lateral ventricles, compared to women, VVs and intracranial areas corrected
for differences in cranial size do not vary between sexes [28]. The sensitivity for detecting
AD in different age groups using the cut-offs from this study was low, ranging from 0 – 21%,
owing to the highly overlapping EI values between CTR and ADs. EI cut-off can therefore
not be regarded as a well-suited method for diagnosing AD on MRI scans. There seems to be
a smooth transition between normal and pathological values. Still it may be a valuable tool in
the evaluation of brain scans from elderly patients in addition to other visual scales.

EI ≥0.3 is a suboptimal indicator when diagnosing iNPH, and diagnosing iNPH is challenging
as a whole [29]. In this study, 94 % of patients treated surgically with a shunt procedure after
MRI improved clinically, indicating a high proportion of what may be considered “true”
iNPH. The sensitivity of detecting iNPH using the new cut off values is high (80%), and
might be applied in the clinical setting. Our results need to be confirmed in larger groups of
iNPH patients and unselected cases referred to the neurosurgical departments to test if the
sensitivity is reproducible.

A limitation of working with databases such as AddNeuroMed and ADNI is that their
population represents a clinical trial population and not an epidemiologically selected

9
population based study. Findings such as low educational level and increased prevalence of
depression have for example been mentioned as possible confounders in the AddNeuromed
study [30] and this may limit to which extent the results from this study can be generalized to
the entire population. The results should be confirmed with data from other large population
based cohorts.

Advantages of our study is the relatively large cohort of patients and controls compared to
previous studies with EI measurements, high intra – and inter-observer reliability in the EI
measurements and good clinical information about their health and cognition, which is
lacking in many other studies. Cut-offs have also been tested on patients with a clinically
probable iNPH diagnosis and a high shunt response rate.

In conclusion, we report normal EI ranges and propose new age-dependent cut-off values for
men and women based on our healthy controls for use in routine radiological practice. EI in
women are significantly lower than for men for all age intervals examined. We found that the
sensitivity for discriminating AD from normal is very low in individual cases, but high in
iNPH. Future studies are needed to confirm the utility of the proposed cut-off values to
identify iNPH.

Acknowledgments

Data collection and sharing for this project was funded by the ADNI (National Institutes of
Health Grant U01 AG024904) and DOD ADNI (Department of Defense award number
W81XWH-12-2-0012). ADNI is funded by the National Institute on Aging, the National
Institute of Biomedical Imaging and Bioengineering, and through generous contributions
from the following: AbbVie, Alzheimer’s Association; Alzheimer’s Drug Discovery
Foundation; Araclon Biotech; BioClinica, Inc.; Biogen; Bristol-Myers Squibb Company;
CereSpir, Inc.; Eisai Inc.; Elan Pharmaceuticals, Inc.; Eli Lilly and Company; EuroImmun; F.
Hoffmann-La Roche Ltd and its affiliated company Genentech, Inc.; Fujirebio; GE
Healthcare; IXICO Ltd.; Janssen Alzheimer Immunotherapy Research & Development, LLC.;
Johnson & Johnson Pharmaceutical Research & Development LLC.; Lumosity; Lundbeck;
Merck & Co., Inc.; Meso Scale Diagnostics, LLC.; NeuroRx Research; Neurotrack
Technologies; Novartis Pharmaceuticals Corporation; Pfizer Inc.; Piramal Imaging; Servier;
Takeda Pharmaceutical Company; and Transition Therapeutics. The Canadian Institutes of
Health Research is providing funds to support ADNI clinical sites in Canada. Private sector

10
contributions are facilitated by the Foundation for the National Institutes of Health
(www.fnih.org). The grantee organization is the Northern California Institute for Research
and Education, and the study is coordinated by the Alzheimer's Disease Cooperative Study at
the University of California, San Diego. ADNI data are disseminated by the Laboratory for
Neuro Imaging at the University of Southern California.

AddNeuroMed is funded through the EU FP6 program as part of InnoMed and with
additional support from the Alzheimer’s Research Trust and from the NIHR Specialist
Biomedical Research Centre for Mental Health at the South London and Maudsley NHS
Foundation Trust and King’s College London, Institute of Psychiatry, London, United
Kingdom. The first author received financial support from Helse Vest grant no 911948.

11
References
[1] P. Scheltens, F. Pasquier, J.G. Weerts, F. Barkhof, D. Leys, Qualitative assessment of
cerebral atrophy on MRI: inter- and intra-observer reproducibility in dementia and
normal aging, Eur. Neurol. 37(2) (1997) 95-9.
[2] P. Scheltens, D. Leys, F. Barkhof, D. Huglo, H.C. Weinstein, P. Vermersch, M. Kuiper, M.
Steinling, E.C. Wolters, J. Valk, Atrophy of medial temporal lobes on MRI in "probable"
Alzheimer's disease and normal ageing: diagnostic value and neuropsychological
correlates, J. Neurol. Neurosurg. Psychiatry 55(10) (1992) 967-72.
[3] F. Fazekas, J.B. Chawluk, A. Alavi, H.I. Hurtig, R.A. Zimmerman, MR signal
abnormalities at 1.5 T in Alzheimer's dementia and normal aging, AJR Am. J. Roentgenol.
149(2) (1987) 351-6.
[4] D. Ferreira, L. Cavallin, E.M. Larsson, J.S. Muehlboeck, P. Mecocci, B. Vellas, M. Tsolaki,
I. Kloszewska, H. Soininen, S. Lovestone, A. Simmons, L.O. Wahlund, E. Westman, c.
AddNeuroMed, I. the Alzheimer's Disease Neuroimaging, Practical cut-offs for visual
rating scales of medial temporal, frontal and posterior atrophy in Alzheimer's disease
and mild cognitive impairment, J. Intern. Med. 278(3) (2015) 277-90.
[5] C.R. Jack, Jr., M.M. Shiung, S.D. Weigand, P.C. O'Brien, J.L. Gunter, B.F. Boeve, D.S.
Knopman, G.E. Smith, R.J. Ivnik, E.G. Tangalos, R.C. Petersen, Brain atrophy rates predict
subsequent clinical conversion in normal elderly and amnestic MCI, Neurology 65(8)
(2005) 1227-31.
[6] N. Relkin, A. Marmarou, P. Klinge, M. Bergsneider, P.M. Black, Diagnosing idiopathic
normal-pressure hydrocephalus, Neurosurgery 57(3 Suppl) (2005) S4-16; discussion ii-
v.
[7] W.A. Evans, An encephalographic ratio for estimating ventricular enlargement, Arch.
Neurol. Psychiatry 47(6) (1942) 931-937.
[8] J.L. Sherman, C.M. Citrin, R.E. Gangarosa, B.J. Bowen, The MR appearance of CSF flow
in patients with ventriculomegaly, AJR Am. J. Roentgenol. 148(1) (1987) 193-9.
[9] V. Synek, J.R. Reuben, G.H. Du Boulay, Comparing Evans' index and computerized
axial tomography in assessing relationship of ventricular size to brain size, Neurology
26(3) (1976) 231-3.

12
[10] R.J. Jacoby, R. Levy, J.M. Dawson, Computed tomography in the elderly: I. The
normal population, Br. J. Psychiatry 136 (1980) 249-55.
[11] A.K. Toma, E. Holl, N.D. Kitchen, L.D. Watkins, Evans' index revisited: the need for an
alternative in normal pressure hydrocephalus, Neurosurgery 68(4) (2011) 939-44.
[12] K. Ishii, T. Kanda, A. Harada, N. Miyamoto, T. Kawaguchi, K. Shimada, S. Ohkawa, T.
Uemura, T. Yoshikawa, E. Mori, Clinical impact of the callosal angle in the diagnosis of
idiopathic normal pressure hydrocephalus, Eur. Radiol. 18(11) (2008) 2678-83.
[13] K. Ambarki, H. Israelsson, A. Wahlin, R. Birgander, A. Eklund, J. Malm, Brain
ventricular size in healthy elderly: comparison between Evans index and volume
measurement, Neurosurgery 67(1) (2010) 94-9; discussion 99.
[14] S.K. Bourne, A. Conrad, J.S. Neimat, T.L. Davis, Linear measurements of the cerebral
ventricles are correlated with adult ventricular volume, J. Clin. Neurosci. 20(5) (2013)
763-4.
[15] K. Reinard, A. Basheer, S. Phillips, A. Snyder, A. Agarwal, K. Jafari-Khouzani, H.
Soltanian-Zadeh, L. Schultz, T. Aho, J.M. Schwalb, Simple and reproducible linear
measurements to determine ventricular enlargement in adults, Surg. Neurol. Int. 6
(2015) 59.
[16] A. Simmons, E. Westman, S. Muehlboeck, P. Mecocci, B. Vellas, M. Tsolaki, I.
Kloszewska, L.O. Wahlund, H. Soininen, S. Lovestone, A. Evans, C. Spenger, The
AddNeuroMed framework for multi-centre MRI assessment of Alzheimer's disease:
experience from the first 24 months, Int. J. Geriatr. Psychiatry 26(1) (2011) 75-82.
[17] R.C. Petersen, P.S. Aisen, L.A. Beckett, M.C. Donohue, A.C. Gamst, D.J. Harvey, C.R.
Jack, Jr., W.J. Jagust, L.M. Shaw, A.W. Toga, J.Q. Trojanowski, M.W. Weiner, Alzheimer's
Disease Neuroimaging Initiative (ADNI): clinical characterization, Neurology 74(3)
(2010) 201-9.
[18] S. Lovestone, P. Francis, I. Kloszewska, P. Mecocci, A. Simmons, H. Soininen, C.
Spenger, M. Tsolaki, B. Vellas, L.O. Wahlund, M. Ward, C. AddNeuroMed, AddNeuroMed--
the European collaboration for the discovery of novel biomarkers for Alzheimer's
disease, Ann. N. Y. Acad. Sci. 1180 (2009) 36-46.
[19] J.C. Morris, The Clinical Dementia Rating (CDR): current version and scoring rules,
Neurology 43(11) (1993) 2412-4.

13
[20] P.K. Eide, W. Sorteberg, Diagnostic intracranial pressure monitoring and surgical
management in idiopathic normal pressure hydrocephalus: a 6-year review of 214
patients, Neurosurgery 66(1) (2010) 80-91.
[21] E. Westman, A. Simmons, J.S. Muehlboeck, P. Mecocci, B. Vellas, M. Tsolaki, I.
Kloszewska, H. Soininen, M.W. Weiner, S. Lovestone, C. Spenger, L.O. Wahlund, c.
AddNeuroMed, I. Alzheimer's Disease Neuroimaging, AddNeuroMed and ADNI: similar
patterns of Alzheimer's atrophy and automated MRI classification accuracy in Europe
and North America, Neuroimage 58(3) (2011) 818-28.
[22] M. Bland, An Introduction to Medical Statistics, Third Edition ed., Oxford University
Press, Oxford, 2000.
[23] J. Ramsay, B.W. Silverman, Functional Data Analysis, Springer-Verlag New York,
New York, 2005.
[24] P. Missori, A. Rughetti, S. Peschillo, G. Gualdi, C. Di Biasi, I. Nofroni, L. Marinelli, F.
Fattapposta, A. Curra, In normal aging ventricular system never attains pathological
values of Evans' index, Oncotarget 7(11) (2016) 11860-3.
[25] P.E. Cowell, V.A. Sluming, I.D. Wilkinson, E. Cezayirli, C.A. Romanowski, J.A. Webb,
S.S. Keller, A. Mayes, N. Roberts, Effects of sex and age on regional prefrontal brain
volume in two human cohorts, Eur. J. Neurosci. 25(1) (2007) 307-18.
[26] D.L. Greenberg, D.F. Messer, M.E. Payne, J.R. Macfall, J.M. Provenzale, D.C. Steffens,
R.R. Krishnan, Aging, gender, and the elderly adult brain: an examination of analytical
strategies, Neurobiol. Aging 29(2) (2008) 290-302.
[27] K.B. Walhovd, L.T. Westlye, I. Amlien, T. Espeseth, I. Reinvang, N. Raz, I. Agartz, D.H.
Salat, D.N. Greve, B. Fischl, A.M. Dale, A.M. Fjell, Consistent neuroanatomical age-related
volume differences across multiple samples, Neurobiol. Aging 32(5) (2011) 916-32.
[28] K.P. Cosgrove, C.M. Mazure, J.K. Staley, Evolving knowledge of sex differences in
brain structure, function, and chemistry, Biol. Psychiatry 62(8) (2007) 847-55.
[29] G.L. Gallia, D. Rigamonti, M.A. Williams, The diagnosis and treatment of idiopathic
normal pressure hydrocephalus, Nat. Clin. Pract. Neurol. 2(7) (2006) 375-81.
[30] H. Brodaty, A. Mothakunnel, M. de Vel-Palumbo, D. Ames, K.A. Ellis, S. Reppermund,
N.A. Kochan, G. Savage, J.N. Trollor, J. Crawford, P.S. Sachdev, Influence of population
versus convenience sampling on sample characteristics in studies of cognitive aging,
Ann. Epidemiol. 24(1) (2014) 63-71.

14
Figure headings:

Figure 1: Evans Index. Maximal width of the frontal horns divided by the largest internal
diameter of the cranium in the same slice. Evans Index = A/B

15
Figure 2: The non-linear relationship between EI and age for CTR and AD, men and female.
The top two plots represent men and the bottom two plots represents women. The black curve
in bold represents the mean value for the EI in each group, and the upper and lower black
curves define the upper and lower limits of the 95% confidence bands, respectively.

16
Tables
Table 1: EI and other clinical data of CTR and AD
CTR AD P-value
MALE
Age, years 74.9  4.2 (N=150) 75.9  5.2 (N=98) 0.081

MMSE 29.0  1.2 (N=150) 23.1  3.1 (N=97) < 0.001

EI 0.29  0.035 (N = 150) 0.31  0.039 (N = 98) < 0.001

FEMALE
Age, years 74.4  4.2 (N=158) 75.0  4.9 (N=128) 0.295

MMSE 29.2  0.9 (N=156) 22.3  3.7 (N=124) < 0.001

EI 0.28  0.031 (N = 158) 0.29  0.034 (N=128) 0.04

Results are presented as mean ± standard deviationMMSE = Mini mental state examinationEI
= Evans Index

17
Table 2: EI cut offs values from 308 CTR participants
Mean EI  SD/2SD for men and women and different age range, representing the cut off
values for EI.
Male Female
Mean  2 SD 97,5 percentile N Mean  2 SD 97,5 percentile N

65 – 69 years 0.28  0.06 0.34 11 0.27  0.05 0.32 17

70 – 74 years 0.29  0.07 0.36 63 0.27  0.06 0.33 67

75 – 79 years 0.30  0.07 0.37 55 0.28  0.06 0.34 54

80 – 84 years 0.30  0.07 0.37 21 0.29  0.07 0.36 20

65 – 84 years 0.29  0.07 0.36 150 0.27  0.06 0.33 158

18