Is It Possible to Differentiate the Impact of Pediatric Monophasic Demyelinating Disorders and Multiple Sclerosis After a First Episode of Demyelination? - Longitudinal Registration and Shape Modeling - page 35

Is It Possible to Differentiate the Impact of Pediatric
Monophasic Demyelinating Disorders and Multiple
Sclerosis After a First Episode of Demyelination?
Bérengère Aubert-Broche 1(  ) , Vladimir Fonov 1 , Katrin Weier 1 , Sridar Narayanan 1 ,
Douglas L. Arnold 1 , Brenda Banwell 2,3 , and D. Louis Collins 1
1 McConnell Brain Imaging Center, Montreal Neurological Institute,
McGill University, Montreal, Canada
[e-mail]
2 The Hospital for Sick Children, University of Toronto, Toronto, Canada
3 Children's Hospital of Philadelphia, University of Pennsylvania, Philadelphia, USA
Abstract. A first episode of acute demyelination of the central nervous system
may be a monophasic transient illness or represent the first attack of multiple
sclerosis (MS). This study investigates if it is possible to distinguish these
two groups of patients retrospectively at the time of the first episode, in a pedi-
atric population. For each patient, the method consists in fitting an individual
brain growth curve using multiple follow-up time-points, and using this curve
to predict 4 metrics at the first attack: brain volume, brain growth rate, thalamus
volume normalized by the brain volume (called normalized thalamus) and nor-
malized thalamus growth rate. These metrics were compared to age-and-sex
matched healthy controls by computing z-scores.
In this study, 85 patients were scanned up to 8 years after the first attack.
During this follow-up period, 23 patients were subsequently diagnosed with MS
( MS group). Among the 62 patients with a transient illness, 9 suffered from mo-
nophasic acute disseminated encephalomyelitis ( ADEM group). The 53 remain-
ing formed the non-ADEM monophasic (MONO) group.
The normalized thalamus growth rate was the only metric that distinguished
patient groups: the z-scores were significantly smaller for MS than for the
MONO group (p<0.01). Whereas 93% of monophasic subjects were correctly
classified with a linear discriminant analysis, only 13% of the MS subjects were
correctly classified, due to a large inter-individual variability in this group.
1 Introduction
Establishing a model for a neurological disease evolution is particularly challenging
in children as brain growth and development occurs simultaneously. Brain changes
due to disease can be detected only if the individual change trajectories are interpreted
in relationship to a normal population growth model.
In cross-sectional studies [1,2] analyzing MRI scans acquired a few years after onset
of multiple sclerosis (MS) of pediatric-onset MS patients (onset prior to age 18 years)
have been shown to have reduced thalamus and brain volumes relative to age- and
sex-matched healthy controls. A recent longitudinal study provides evidence that the