10393_Neurodevelopmental Risks Of Non-Syndromic Craniosynostosis

Yale University
EliScholar – A Digital Platform for Scholarly Publishing at Yale
Yale Medicine Thesis Digital Library
School of Medicine
January 2019
Neurodevelopmental Risks Of Non-Syndromic
Craniosynostosis
Robin T. Wu
Follow this and additional works at: https://elischolar.library.yale.edu/ymtdl
This Open Access Thesis is brought to you for free and open access by the School of Medicine at EliScholar – A Digital Platform for Scholarly
Publishing at Yale. It has been accepted for inclusion in Yale Medicine Thesis Digital Library by an authorized administrator of EliScholar – A Digital
Platform for Scholarly Publishing at Yale. For more information, please contact elischolar@yale.edu.
Recommended Citation
Wu, Robin T., “Neurodevelopmental Risks Of Non-Syndromic Craniosynostosis” (2019). Yale Medicine Thesis Digital Library. 3541.
https://elischolar.library.yale.edu/ymtdl/3541

0

Neurodevelopmental Risks of Non-syndromic Craniosynostosis

A Thesis Submitted to the
Yale University School of Medicine
In Partial Fulfillment of the Requirements for the
Degree of Doctor of Medicine

By
Robin T. Wu
2019

Neurodevelopmental Risks of Non-syndromic Craniosynostosis
Robin T. Wu, Kyle S. Gabrick, Andrew T. Timberlake, Anusha Singh, Paul F. Abraham,
James Nie, Taylor Halligan, Raysa Cabrejo, Derek M. Steinbacher, Michael Alperovich,
John A. Persing, Yale University, School of Medicine, New Haven, CT.
Purpose: Nonsyndromic craniosynostosis may manifest with complex cognitive,
language, behavioral, and emotional sequelae, depending on the suture fusion involved.
De-novo or rare transmitted mutations in the SMAD6 gene affect midline synostosis in
7% of patients. Current standards of assessment, such as the Bayley Scales of Infant
Development (BSID), may not predictive of long-term development, paving the way for
newer assessments such as functional magnetic resonance imaging (fMRI) and the event
related potentials (ERP), which measures passive neurological responses to speech
sounds.

Methods: Cranially-mature, post-operative unilateral coronal, metopic, midline SMAD6
mutated and age/race/gender/synostosis/operation matched non-SMAD6 controls from
the Yale Craniofacial Clinic and the Children’s Hospital of Philadelphia (CHOP)
completed a double-blinded neurodevelopmental assessment, which included the
Wechsler Fundamentals, Wechsler Abbreviated Scale of Intelligence, and Beery-
Buktenica Developmental Test. Unilateral coronal (ULC) or metopic synostosis were
age/gender/handedness matched to controls and participated in a GoNoGo task under
fMRI. Craniosynostosis infants were given the BSID and ERP testing at two points (pre
and post operatively), and after they reached >6 years of age, patients completed the
Wechsler Abbreviated Scale of Intelligence and Wechsler Fundamentals to measure 5
language functional domains.
Results: ULC patients had a mean verbal IQ of 117.3 and performance IQ of 106.4,
performed above average on academic achievements except for numerical, but below
average on all visual-motor tests. Right ULC had improved spelling compared to left ULC,
controlled for exogenous influences (p=0.033). Metopic patients with mild phenotype
(endocranial bifrontal angle <124) performed better in word reading (p=0.035) and reading composite (p=0.014) than patients with severe stenosis (>124). After controlling for
exogenous factors, midline synostosis patients with SMAD6 mutations performed worse
on numerical operations(p=0.046), performance IQ(p=0.018), full IQ(p=0.010), and motor
coordination(p=0.043) than those without the mutation. Among seven ULC and six
metopic patients that participated in fMRI, metopic patients had decreased blood-
oxygenation-level-dependent signal in the posterior cingulate(p=0.017) and middle

temporal
gyrus(MTG;p=0.042).
ULC
had
decreased
signal
in
the
posterior
cingulate(p=0.023), MTG(p=0.027), and thalamus(p=0.033), but increased signal in the
cuneus(p=0.009) and cerebellum(p=0.009). Among 10 craniosynostoses patients who
received ERP/BSID testing in infancy followed by school-age neurocognitive testin, the
left frontal ERP cluster strongly correlated with word reading (R 0.713, p=0.031), reading
comprehension (R 0.745, p=0.021), and language composite scores (R=0.771, p=0.015).
Correlations for BSID cognitive, expressive language, and language composite scores had
no predictive value (R<0.5, p>0.05) for neurocognitive scores.
Conclusions: Post-operative cranially mature ULC patients have higher verbal IQ
scores, but worse mathematical and visual-motor achievement. Left-sided ULC patients
may perform worse in spelling. The severity of orbito-frontal dysmorphology in
metopic synostosis significantly impacts long-term cognitive function and academic
achievement. Neuropsychiatric development may be in whole or in part under genetic
control. SMAD6 mutations led to poorer performance on mathematics, performance-IQ,
full-IQ, and motor coordination, even after controlling for exogenous factors. ULC
patients may have emotional dyregulation in response to frustration while metopic
patients may have attenuated emotional reactions. ERP assessment in nonsyndromic
craniosynostosis patients has significantly better predictive value for future
neurocognitive assessment than the standard BSID test. Use of ERP assessment may
help tailor treatment for language deficits earlier in development.

Acknowledgements
I would like to thank my co-authors for their tireless contributions to this work and my
faculty mentors for their belief in me. I am so lucky to have gotten the chance to
participate in research with Dr. John Persing on work for which he has pioneered and
changed the field of craniosynostosis.
I am grateful for my wonderful friends and future colleagues in medical school Alyssa
Zupon, Rebecca Fine, Elliot Morse, Matthew Swallow, Tejas Sathe, and Brandon
Sumpio, for surviving all the long rotations and late nights with me.
A very special thank you to Pranavi Vemuri, Yohan Perera, Jessica Shepis, and Jeffrey
Chen for their lifelong friendship.
Finally, I give all my love to my mom and dad, my life coaches and perpetual
cheerleaders.

This work was completed with monetary contributions from the Plastic Surgery Foundation
(Award Number: 513938).

Table of Contents

Introduction
…………………………………………………………………………………………………….. 1
Non-syndromic Craniosynostosis ………………………………………………………………………………. 1
Surgical Correction of Craniosynostosis………………………………………………………………………. 2
Long-Term Neurodevelopmental Outcomes ……………………………………………………………….. 3
Predictors of Neurodevelopmental Performance
…………………………………………………………. 6
Genetics in Craniosynostosis …………………………………………………………………………………….. 6
Functional Magnetic Resonance Imaging in Craniosynostosis ………………………………………… 7
Event Related Potentials in Craniosynostosis
………………………………………………………………. 8
Purpose
…………………………………………………………………………………………………………. 11
Methods ……………………………………………………………………………………………………….. 13
Patient Selection and Individualized Testing Parameters
…………………………………………….. 13
Unilateral Coronal Craniosynostosis Categorization …………………………………………………. 13
Metopic Craniosynostosis Categorization
……………………………………………………………….. 14
SMAD6 Comparison ……………………………………………………………………………………………. 14
Functional MRI Analysis ………………………………………………………………………………………. 15
Event Related Potential Analysis …………………………………………………………………………… 15
Neuropsychiatric Testing Battery ………………………………………………………………………….. 16
Neurocognitive Tests ………………………………………………………………………………………….. 16
Parental/Guardian Surveys ………………………………………………………………………………….. 18
Quality of Life Survey ………………………………………………………………………………………….. 19
Computed Tomographic Scan Analysis
……………………………………………………………………… 19
Direct Neuroimaging and Genetic Analysis ……………………………………………………………….. 20
Functional MRI ………………………………………………………………………………………………….. 20
Event Related Potentials ……………………………………………………………………………………… 22
Genetic Analysis…………………………………………………………………………………………………. 23
Statistical Analysis
…………………………………………………………………………………………………. 24
Unilateral Coronal Craniosynostosis Neuropsychiatric Outcomes
……………………………….. 24
Metopic Craniosynostosis Neurocognitive Comparison to Severity …………………………….. 25
SMAD6 Comparison to non-SMAD6 Neurocognitive Outcomes
………………………………….. 25
fMRI Comparison ……………………………………………………………………………………………….. 26
ERP and BSID Comparison with Neurocognitive Outcomes………………………………………… 26
Results ………………………………………………………………………………………………………….. 27
Unilateral Coronal Craniosynostosis Neurodevelopmental Outcomes
…………………………… 27
Subjects ……………………………………………………………………………………………………………. 27
Neurocognitive Test Performance
…………………………………………………………………………. 29
Behavioral Survey Performance ……………………………………………………………………………. 30
Impact of Patient Factors on Neurocognitive Performance ……………………………………….. 31
Post-hoc power …………………………………………………………………………………………………. 33
Metopic Craniosynostosis Neurocognitive Outcomes …………………………………………………. 34
Subjects ……………………………………………………………………………………………………………. 34
Neurocognitive Test Performance
…………………………………………………………………………. 35

Analysis of Severity …………………………………………………………………………………………….. 36
Sagittal and Metopic SMAD6 Neurocognitive Outcomes …………………………………………….. 37
Subjects ……………………………………………………………………………………………………………. 37
Head-to-head T-test comparison between SMAD6 and non-SMAD6 controls
……………….. 39
Correlation Analysis ……………………………………………………………………………………………. 40
Controlling for significant patient factors ……………………………………………………………….. 41
Parental Surveys ………………………………………………………………………………………………… 41
Power Analysis…………………………………………………………………………………………………… 41
fMRI Analysis ……………………………………………………………………………………………………….. 42
Demographics
……………………………………………………………………………………………………. 42
Behavioral/Functional Scores
……………………………………………………………………………….. 43
GoNoGo Performance ……………………………………………………………………………………………. 44
fMRI Whole-Brain T-Test and Region of Interest Analysis ………………………………………….. 44
BOLD Signal Analysis …………………………………………………………………………………………… 47
ERP and BSID Analysis ……………………………………………………………………………………………. 52
Patient Demographics
…………………………………………………………………………………………. 52
Neurocognitive Correlation with Infant ERP/BSID Testing …………………………………………. 53
Controlling for Demographic Confounders ……………………………………………………………… 55
ERP Comparison between Subtypes of Craniosynostosis …………………………………………… 55
Discussion ……………………………………………………………………………………………………… 57
Unilateral Coronal Craniosynostosis Neurodevelopmental Outcomes
…………………………… 57
Metopic Craniosynostosis Neurocognitive Outcomes …………………………………………………. 59
Sagittal and Metopic SMAD6 Neurocognitive Outcomes …………………………………………….. 61
fMRI Analysis ……………………………………………………………………………………………………….. 64
ERP and BSID Analysis ……………………………………………………………………………………………. 67
Citations
………………………………………………………………………………………………………… 70

1
Introduction

Non-syndromic Craniosynostosis

Cranial growth is governed by complex interactions between the brain, dura mater,
cartilaginous sutures, and bony plates.1 Patent calvarial sutures permit the skull to
accommodate rapid expansion of the underlying brain in early infancy. Physiologic
closure follows a conserved sequence; the posterior fontanelle obliterates between 1-3
months, followed by the metopic suture between 3-8 months, the anterior fontanelle
between 9-18 months, and the remainder of sutures in adulthood.2
Premature fusion of calvarial sutures restricts skull growth perpendicular to the affected
suture3. This pathology, known as non-syndromic craniosynostosis, affects 1 in every
2000 to 2500 births4. Presentations are varied based on suture type but yield reliable
phenotypes.
Ossification of midline calvarial sutures, metopic or sagittal nonsyndromic
craniosynostosis, predicates abnormal skull growth in the anteroposterior direction and
comprise the vast majority of cases.5-7 Sagittal synostosis patients have stereotypical
scaphocephaly, resulting in compensatory growth in the frontal/occipital regions and
limited anteroposterior width.8 Metopic synostosis is characterized by trigonocephaly,
bitemporal narrowing, and orbital hypotelorism. The orbito-frontal dysmorphology
includes symmetric supra-orbital retrusion with a keel-shaped deformity in the

midline.9-12 Unilateral coronal craniosynostosis (ULC) is the next most common, with a
prevalence of 66 per million children born.13-15 Unilateral coronal synostosis (ULC) limits
the frontal cranium asymmetrically and is characterized by ipsilateral forehead
flattening, a shallow orbit, and a recessed supraorbital rim, often with contralateral
frontoparietal bossing.16,17 The rarest form of craniosynostosis is lambdoid fusion,
comprising only 1-5% of craniosynostoses. Lambdoid synostosis results in ipsilateral
occipital flattening and mastoid bossing.18
Surgical Correction of Craniosynostosis

Patients who undergo treatment prior to three months of age may be offered strip
craniectomy by some centers with selective use of postoperative cranial orthoses.1,19-21 At
this vulnerable age, emphasis is placed on limiting blood loss and operative time.20 Strip
craniectomy relies on subsequent brain growth to yield skull expansion and improved
cranial shape. After six months, the cranium begins to ossify and skull bones lose
malleability. In these older patients, with some institutional exceptions, cranial vault
remodeling is generally preferred for more predictable outcomes.22
Choice in surgical technique involves an array of variables including type of fused
suture, clinical severity, patient age and comorbidities, and perspectives regarding
neurologic development.23,24 Controversy exists regarding the timing of surgical repair
and indications for cranial vault remodeling versus strip craniectomy. Strip craniectomy
is less invasive but cranial vault remodeling (CVR) carries the advantage of more

complete correction of the deformity and release of brain compression post-operatively,
which may have a positive influence on brain development.25,26

Long-Term Neurodevelopmental Outcomes

Premature fusion of calvarial sutures, or nonsyndromic craniosynostosis has direct
sequelae on abnormal skull growth and deformation of underlying brain structures.5-7
While study results are varied, current literature has suggested that long-term
neurodevelopmental sequelae may exist in up to 50% of nonsyndromic craniosynostosis
patients.7,26-29 Treatment goals for nonsyndromic craniosynostosis are two-fold:
normocephaly of skull shape and improved long-term functional neurocognitive
outcomes.25,27,30 Surgical treatment can improve global cognitive development and IQ,
however, recent scrutiny has revealed persistence of subtle learning deficits.7,26-28
Children born with craniosynostosis typically have normal global intelligence, but have
speech and or language impairments. Magge et al. tested 16 children aged 6 to 16 years
with surgically corrected sagittal synostosis, and found despite normal intelligence
scores, 50% were diagnosed with at least one language related learning disorder.7
Similarly, Shipster et al. tested 75 children aged 9 months to 15 years with sagittal
synostosis and found no global cognitive impairment.31 However, 37% had speech
and/or language impairment, with expressive language being most frequently affected.
Naran et al. reported a series of 101 patients, aged 2-18 years, in which a majority had

metopic pathology.32 Abnormal language development was identified in 1 in 1.7 patients
and speech therapy was necessary in 1 in 3.4 subjects. Chieffo et al. studied 65 children,
9 to 16 years of age, and found 30% of unicoronal synostoses patients to be comorbid
with speech delays.28
Different sutures govern particular patterns of brain restriction. Thus, neurocognitive
outcomes may vary based on suture fusion. The metopic and coronal sutures, in
particular, are positioned in the anterior cranium. The adjacent frontal brain region is
tasked with executive function, impulse inhibition, and personality.33 Lesions are
classically associated with emotional dysregulation such as depression, anxiety,
aggression, and social inappropriateness.34 Of particular interest, the limbic system leads
emotional processing, comprising areas such as the cingulate cortex involved in stress
processing.35 A plausible hypothesis, then, would implicate metopic and coronal
synostosis with frontal lobe associated behavioral deficits. Indeed, abnormally low
frontal lobe volume and corpus callosum abnormalities in metopic patients has been
hypothesized to predispose for cognitive, motor, verbal, attention, and visuospacial
deficits.26,36,37 Another study reported 30% of ULC patients demonstrated processing and
planning speech delays.28
Shillito and Matson reported mental retardation rates of 2.6% in 66 ULC patients in
1968.38 In 1977, Hunter and Rudd published up to a 10% retardation and 11% borderline
personality rate in 52 patients with ULC.38,39 Becker et al. documented 61% of right and
52% of left ULC had speech-language, cognitive, and/or behavioral aberrations, without

statistical difference, but did not review individual tests with more granularity.40 Speltz
et al. cognitively tested 28 ULC infants, mean age 6.5 months, pre-surgically and found
no significant difference among sidedness or compared with other single-sutures
synostosis.41 Starr et al. tested synostosis infants between 17-19 months and similarly
concluded that despite below-average performance among all subtypes, ULC patients
were not distinguished by suture type or laterality.
Neurodevelopmental delays in patients with metopic synostosis range from 15% to as
high as 61% and may be particularly severe.11,42-45 The metopic suture, positioned
exclusively in the anterior cranium, overlays adjacent frontal brain regions tasked with
executive function, impulse inhibition, and personality.33,34 Mendonca et al. found 30% of
metopic synostoses patients had speech and language delays but denied correlation with
craniometrics measurements.46 Conversely, Bottero et al. reported 23% rate of
developmental deficit in mild non-operative trigonocephaly and a 32% rate in more
severe patients requiring surgical intervention.36 With surgical correction, Kunz et al.
claimed that among 40% of metopic children with delays pre-operatively, all either
completely recovered or improved twelve months postoperatively.47 One quantitative
assessment of phenotypic severity measures the endocranial bifrontal angle.12,48 Prior
studies have identified increased cognitive deficits in infants with a more acute
endocranial bifrontal angle using event-related potentials.48

However, neural plasticity and compensatory development complicate such conclusions
and neurobehavioral variations may be subtle. Long-term influences on brain
development and neurocognition require further investigation.26,41,49-51
Predictors of Neurodevelopmental Performance

Early detection and prevention is essential for cognitive remediation in nonsyndromic
craniosynostosis patients. Therefore, there is a need for proper evaluative tools for
predicting development. Younger age at surgical correction and more comprehensive
surgical remodeling have been associated with better overall intelligence, reading skills,
math, and visuomotor integration.25,30
Final volumetric cranial size and brain network fine-tuning are not reached until ages 7-
11, suggesting neurocognitive testing should be performed at the time of cranial
maturity.52-54 Furthermore, neurocognitive testing is more sensitive for deficits at older
ages given the increased neurocognitive demands of the cranially-mature cohort relative
to toddlers.55,56 While neurodevelopmental surveys have come a long way to categorize
the rates of delay and the impact of surgery, these cognitively vulnerable patients may
benefit from further risk stratification based on pre-operative phenotype.
Genetics in Craniosynostosis

Midline non-syndromic craniosynostoses are found to be under genetic influence.
Common variants downstream of the BMP2 gene have been associated with sagittal

synostosis. Recent breakthroughs revealed that de novo or rare transmitted mutations in
the SMAD6 gene, an inhibitor of BMP signaling, cause non-syndromic midline
synostosis in 7% of patients.57 Genetic interactions between SMAD6 mutations and the
common BMP2 risk allele dramatically affect penetrance in these cases.
Bicoronal synostosis patients with FGFR3 mutations trended towards worse
developmental and intellectual outcomes, though the differences did not achieve
statistical significance.58 Genomic analysis of intellectual disability by Lelieveld et al.
identified the SMAD6 gene as a novel locus for intellectual disability, however the
presence or absence of craniosynsotosis was not noted in children with SMAD6
mutations and intellectual disability.59 Questions arise as to the effect of SMAD6
mutations on neurocognitive development in the setting of craniosynostosis, given that
these mutations are the most frequent genetic cause of nonsyndromic craniosynostosis
identified to date. While optimizing surgical interventions and pioneering new-age tests
have proven efficacious in detecting neurocognitive risks in craniosynostosis, genetic
risks are non-modifiable and easily tested.
Functional Magnetic Resonance Imaging in Craniosynostosis

Functional MRI (fMRI) has been efficacious in teasing out delicate brain dynamics. fMRI
studies in craniosynostosis demonstrated altered connectivity in sagittal patients and
resting state group differences among subtypes of synostosis.60 Sagittal synostosis patients
often demonstrate significant changes in the left supramarginal gyrus, which may

correspond to language related learning disorders.61 Metopic patients exhibit more
changes in the dorsolateral pre-frontal cortex which often impacts working memory and
executive function. Unilateral coronal patients often have altered connectivity in the
anterior prefrontal cortex which distort higher level thinking such as multi-tasking. Still,
higher level emotional performances, such as stress and frustration, are more properly
assessed with executive tasks.
Event Related Potentials in Craniosynostosis

In assessing development, the Bayley Scales of Infant and Toddler Development is the
most popular and widely utilized measure of cognitive function in infants aged 1-42
months.52,62 The output variable is a Mental Developmental Index (MDI), which
comprises cognitive, language, motor, social-emotional, and adaptive behavior scales.
Kapp-Simon et al. first began to assess mental development in craniosynostosis infants
with and without treatment with the Bayley Scales of Infant Development.49 They
concluded that cranial reconstruction did not affect mental development, contradictory
to much of the evidence now, which suggests that children often develop deficits in
language and speech development, despite having intelligence scores in the normal
range.7 Recently, the predictive validity of this test has been called into question. Hack et
al. pooled past MDI scores of 344 extremely low birth weight infants and compared
them to the subjects’ current school age cognitive functions; they found a poor positive
predictive value of 0.37 for future IQ, calling into question the utility of this test.53 It is

necessary, then, to develop a better predictor of future function, particularly with
emphasis on language norms.
EEG studies are objective, non-invasive, non-sedative, and thus are considered the best
way to study infant brain activity63. ERPs are convenient as they do not require overt
behavioral/verbal response or even attention from the infant. Most ERP studies to date
aim to elucidate neural networks of healthy infants with a growing field into pathologic
identification of autism spectrum infants. Of the auditory ERPs, the P150/N250
components, two prominent deflections in the EEG waveform, have been extensively
studied. Seery et al. identified atypical lateralization of these ERPs in infants at high risk
for autism spectrum disorder64. Balan et al. also looked at these ERPs in plagiocephaly
infants, finding attenuated P150/N250 amplitudes compared to controls65.
The mismatch negativity (MMN) component of ERP is elicited by having an infant
discriminate a deviant auditory stimulus in the context of repetitive ‘normal’ stimuli66,67,
and has been found to be clinically effective in predicting language acquisition. Infants
are born with the ability to discriminate speech sounds from broad sources68,69. Between
six and twelve months of age, in a process known as perceptual narrowing, infant’s
auditory perceptions specializes towards its native spoken language, virtually
extinguishing non-native verbal phenome recognition70,71. Jansson-Verkasalo et al.
suggested that delayed or atypical perceptual narrowing measured by retained MMN is
longitudinally associated with delayed language skills at one and two years of age,
which has since been verified by other electrophysiologic studies71-73.

Our group was the first to look at ERPs in patients with craniosynostosis. Hashim et al.
reported infants with nonsyndromic craniosynostosis have attenuated P150 waves in
response to speech sounds compared with normal infants.24 Yang et al. found that severe
metopic synostosis, defined by an endocranial bifrontal angle less than 124o, presented
with attenuated P150 waves compared with controls while moderate metopic synostosis
(greater than 124o) had no difference. Recent work, not yet published, by Chuang et al.
has reanalyzed results looking at the MMN waves pre and post-operatively. Preliminary
results found that MMN waves are attenuated preoperatively in sagittal and severe
metopic patients but then improve postoperatively. Thus, validation studies must be
performed to assess the predictive value of ERP on nonsyndromic craniosynostosis
patients.

Purpose

Treatment goals for non-syndromic craniosynostosis are based off of restoring aesthetic
normocephaly and augmenting functional neurocognition. Unfortunately, due to the
early age at intervention and difficulty assessing longitudinal outcomes, the field is
plagued by knowledge gaps as to the long-term results in this patient population. As
such, the purpose of this work was to outline the neurodevelopmental outcomes using
traditional cognitive testing, new-age imaging, and craniometrics analysis. Results may
be critical for predictive outcomes, patient counseling, and understanding the
mechanism of disease.
Aim #1: To present the long-term neurodevelopmental profile of patients with
unilateral coronal craniosynostosis. Among all subtypes of craniosynostosis,
neurocognitive outcomes have not been well established for patients with unilateral
coronal craniosynostosis. Additionally, this study seeks to identify the differential
impact of right verses left sided fusion as well as the influence of exogenous factors in
development.
Aim #2: To compare the long-term neurodevelopmental outcomes between patients
with mild and severe metopic craniosynostosis. Earlier work from our lab has
established an endocranial bifrontal angle of 124 degrees as the cutoff between mild and
severe metopic craniosynostosis. We hypothesize that patients with more severe
synostosis angles will have worse neurocognitive outcomes.

Aim #3: To compare the long-term neurodevelopmental outcomes between midline
craniosynostosis patients with mutated SMAD6 genotype and those with the wild
type SMAD6 allele. We hypothesize that patients with SMAD6 mutation will have
residual decreased IQ and performance on academic achievement testing compared to
unaffected individuals.
Aim #4: To characterize long-term emotional-response brain activity with the first-
reported use of task-based fMRI analysis in unilateral coronal and metopic
craniosynostosis. We hypothesize that coronal and metopic craniosynostoses will have
different patterns of brain response to emotional stimuli compared to healthy matched
controls.
Aim #5: To validate event related potential EEG testing in infancy with long-term
language performance in craniosynostosis. Our lab began testing infants ten years
earlier and we hypothesize that EEG testing in infancy can predict future language
development.

Methods

Patient Selection and Individualized Testing Parameters

All testing was conducted with parental or legal guardian consent, patient
assent/consent, and Institutional Human Investigations Committee approval. Patients
treated at the Yale School of Medicine consistent with the inclusion and exclusion
criteria specified for each study arm below were collected by the Yale Joint Data
Analytics Team. Patients were excluded if they had any diagnosed
neurological/developmental delay such as cerebral palsy or a Full-Score IQ [FSIQ] < 70. Patients with a documented or suspected syndromic craniosynostosis diagnosis were excluded. Unilateral Coronal Craniosynostosis Categorization Due to the low prevalence of patients with unilateral coronal craniosynostosis, this study was a double-blinded multi-institutional cohort study between patients treated at the Yale School of Medicine and the Children’s Hospital of Pennsylvania (CHOP). Patients who had radiographic confirmation of non-syndromic unilateral coronal craniosynostosis and received cranial vault remodeling in infancy were included in the study. Patients were at an age of cranial maturity, 8.0 years of age or older, at time of testing. All patients were administered the Weschler Fundamentals (WF), Weschler Abbreviated Scale of Intelligence (WASI), Beery VisuoMotor Integration (VMI), Behavior Rating Inventory and Executive Function (BRIEF), Child Behavior Checklist (CBCL), demographic survey, Youth Quality of Life (YQOL), and 3D photograph. Metopic Craniosynostosis Categorization All patients with radiographic confirmation of metopic craniosynostosis and a history of cranial vault remodeling in infancy were recruited from the Yale School of Medicine. All patients were school age, 6.0 years or older, at the time of testing. All patients were administered the WF, WASI, Beery VMI, BRIEF, Behavior Assessment System for Children (BASC), and demographic survey. SMAD6 Comparison This was a prospective double-blinded cohort study conducted at the Yale School of Medicine. Subjects were included if they were diagnosed with midline non-syndromic craniosynostosis and received surgical correction at an earlier age. School age patients currently 6.0 years of age or older were included. Patients with SMAD6 mutations were identified from the index study.57 Non SMAD6 controls who were diagnosed with midline craniosynostosis and underwent whole exome sequencing found to have wild type SMAD6 alleles were included. Controls were matched by current age (within one year), gender, race, synostosis type, and surgery type (whole vault cranioplasty or strip craniectomy). All tests were administered by a single blinded tester between June 2017 – April 2018. Test subjects were blinded as to the testing groups under scrutiny. Subjects in the index study were recruited nationwide via social media. In order to keep a standardized test administrator, subjects who were unable to travel due to geographical constraints were able to participate in virtual webcam testing. Mountable Logitech C615 webcams (Logitech, Lausanne Switzerland) and testing materials were sent to participant homes, which allowed the administrator to interact with and watch participants complete tasks in real time. All patients were administered the WF, WASI, Beery VMI, BRIEF, BASC2, and demographic survey. Functional MRI Analysis Surgically corrected adolescent patients age >9 with isolated nonsyndromic metopic or
unilateral coronal synostosis operated and treated by the senior author were recruited.
Age/gender/handedness non-craniosynostosis healthy matched controls were recruited
from the Yale Child Studies Center.
Prior to fMRI scan, each craniosynostosis subject was administered the Wechsler
Intelligence Scale for Children 3rd edition (WISC-III) and all subject guardians were
given the BRIEF survey.
Event Related Potential Analysis
Craniosynostosis infants were recruited at the Yale Craniofacial Clinic by our lab (Jenny
F. Yang, MD; Roberto Travieso, MD; Joel Beckett, MD) if they were diagnosed with non-
syndromic craniosynostosis and were planned for surgical correction. Patients were
tested pre-operatively with both functional and event-related potentials and then
returned for the same testing battery three months post-operatively.

Once these same patients were 6.0 years of age or older, they were recruited for follow
up neurocognitive testing. Due to particular sensitivity to reading/language delays, all
patients were administered the WF, WASI, and demographic survey.

Neuropsychiatric Testing Battery
All neuropsychiatric testing for the SMAD6 and event related potential study
arms were performed by the same tester (R.W.). All neurocognitive testing in the
remainder study arms was administered between R.W., K.G., A.S., P.A., and J.N.. All
test administrators were blinded to the clinical variables, synostosis side, and patient
demographics. Surveys were administered to the parent or legal guardian of the
participant to gauge behavior, psychological functioning, and record demographic data.
Neurocognitive Tests
The neurocognitive assessment paradigm utilized is outlined below using previously
published techniques for patients with craniosynostosis.25,30
1. Weschler’s Abbreviated Scale of Intelligence (WASI)
The WASI is an individually administered assessment which is designed to measure
performance, verbal, and full-scale intelligence quotient (IQ). Verbal IQ is determined by
subtests in vocabulary and word similarities which quantify the patient’s word
knowledge and verbal reasoning. Performance IQ is quantified based on subtests in

block design and matrix reasoning which quantify visuospatial reasoning and the ability
to separate figure from ground in visual stimuli.74
2. Weschler’s Fundamentals (WF)
The WF is an individually administered assessment designed to provide a global
assessment of age-based academic achievement. The verbal component consists of
domain scores for word reading, reading comprehension, and reading composite. The
spelling section asks the child to write dictated letters and words. The mathematical
component assesses the patient’s ability to perform multiple arithmetic calculations in a
limited time period.
3. Beery-Buktenica Developmental Test of Visuo-Motor Integration (Beery VMI)
The fifth edition of the Beery VMI is an individually administered assessment which
quantifies the patient’s ability to integrate visual stimuli and motor responses. The child
must by draw geometric forms (VMI), visually distinguishing between similar items
(Visual Perception), and perform fine hand and finger movements (Motor
Coordination).
4. Wechsler Intelligence Scale for Children 3rd edition (WISC-III)
Similar to the WASI, the WISC-III generates a Full Scale IQ that represents a child’s
general intellectual ability. It also provides five primary index scores: Verbal
Comprehension Index, Visual Spatial Index, Fluid Reasoning Index, Working Memory

Index, and Processing Speed Index. These indices represent a child’s abilities in discrete
cognitive domains.

Parental/Guardian Surveys
1. Behavior Rating Inventory of Executive Function (BRIEF)
The BRIEF uses parent questionnaires to assess executive functioning in the home and
school surroundings.7 Results are summarized with eight subcategories: Inhibit, Shift,
Emotional Control, Initiate, Working Memory, Plan/Organize, Organization of
Materials, Monitor, Behavioral Regulation, Metacognition, Global Executive Composite.
2. Behavior Assessment System for Children (BASC), Second Edition
The BASC is administered in a questionnaire format that lists numerous aspects of
behavior and personality functioning. Results are summarized with four subcategories.
3. Child Behavior Checklist (CBCL).
The CBCL is administered in a questionnaire format that assesses behavior and mental
health functioning. Results are summarized with four subcategories: Competence
(Activities + Social + School), Internalizing Problems (Anxious/Depressed +
Withdrawn/Depressed), Externalizing Problems (Rule-breaking + Aggressive behavior),
Total (Internalizing + Externalizing). Category T scores above 70 (98th percentile)
represent behaviors in the range of clinical concern.