Methods to Quantify Soft Tissue–Based Cranial Growth and Treatment Outcomes in Children

12 Pages·2014·0.46 MB·English

Methods to Quantify Soft Tissue–Based Cranial Growth
and Treatment Outcomes in Children: A Systematic
Review
SanderBrons1*,MachteldE.vanBeusichem1,EwaldM.Bronkhorst2,JosM.Draaisma3,StefaanJ.Berge´4,
Jan G. Schols1, Anne Marie Kuijpers-Jagtman1
1DepartmentofOrthodonticsandCraniofacialBiology,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands,2DepartmentofPreventiveand
CurativeDentistry,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands,3DepartmentofPediatrics,RadboudUniversityNijmegenMedicalCentre,
Nijmegen,TheNetherlands,4DepartmentofOralandMaxillofacialSurgery,RadboudUniversityNijmegenMedicalCentre,Nijmegen,TheNetherlands
Abstract
Context:Longitudinal assessment of cranial dimensions of growing children provides healthcare professionals with
information aboutnormaland deviating growth aswellastreatment outcome.
Objective:Togiveanoverviewofsofttissue–basedmethodsforquantitativelongitudinalassessmentofcranialdimensions
inchildren untilage 6years andto assess the reliability ofthese methodsinstudieswith goodmethodological quality.
Datasource:PubMed,EMBASE,CochraneLibrary,WebofScience,Scopus,andCINAHLweresearched.Amanualsearchwas
performed tocheck foradditional relevantstudies.
Studyselection:Primarypublicationsonfacialgrowthandtreatmentoutcomesinchildrenyoungerthanage6yearswere
included.
Dataextraction:Independentdataextractionwasperformedbytwoobservers.Aqualityassessmentinstrumentwasused
to determine methodological quality. Methods used in studies with good methodological quality were assessed for
reliability expressed as the magnitude of the measurement error and the correlation coefficient between repeated
measurements.
Results:Intotal,165studieswereincluded,formingthreegroupsofmethods:headcircumferenceanthropometry,direct
anthropometry, and 2D photography and 3D imaging techniques (surface laser scanning and stereophotogrammetry). In
general,the measurementerrorwasbelow 2 mm,andcorrelation coefficients were very good.
Conclusion:Various methods for measuring cranial dimensions have shown to be reliable. Stereophotogrammetry is the
most versatile method for quantitative longitudinal assessment of cranial dimensions and shapes in children. However,
direct anthropometry continues to be the best method for routine clinical assessments of linear cranial dimensions in
growingchildren until age6years.
Citation:BronsS,vanBeusichemME,BronkhorstEM,DraaismaJM,Berge´SJ,etal.(2014)MethodstoQuantifySoftTissue–BasedCranialGrowthandTreatment
OutcomesinChildren:ASystematicReview.PLoSONE9(2):e89602.doi:10.1371/journal.pone.0089602
Editor:SamuelJLin,HarvardMedicalSchool,UnitedStatesofAmerica
ReceivedJune4,2013;AcceptedJanuary23,2014;PublishedFebruary27,2014
Copyright: (cid:2) 2014 Brons et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits
unrestricteduse,distribution,andreproductioninanymedium,providedtheoriginalauthorandsourcearecredited.
Funding:ThisworkwassupportedbyagrantfromtheDutchTechnologyFoundationtoSTW(grantnumber10315).Thefundershadnoroleinstudydesign,
datacollectionandanalysis,decisiontopublish,orpreparationofthemanuscript.
CompetingInterests:Theauthorshavedeclaredthatnocompetinginterestsexist.
*E-mail:[email protected]
Introduction Various methods for quantitative evaluation of craniofacial
dimensions have been described for a variety of purposes. The
Longitudinal assessment of cranial dimensions of growing standard technique is direct anthropometry, which has been
childrenprovideshealthcareprofessionalswithinformationabout extensivelyusedforthestudyofcraniofacialdimensionsinthepast
normal and deviating growth as well as treatment outcome, for century[5].Two-dimensional(2D)x-raycephalometry[6–8]and
example in cases of deformational plagiocephaly and craniosyn- photography [9,10] also have been applied for decades and even
ostosis [1,2]. Accurate quantitative evaluation of cranial dimen- todayarethemostcommonlyusedrecordsfordento-skeletaland
sions by comparison of an individual patient to normative values facial diagnosis. Recent technological advancements have led
canprovideinsightintoanunderlyingpathologicprocessorcreate craniofacial researchers and clinicians into the era of three
a basis for treatment planning, as in cases of autism and dimensional (3D) digital imaging. Techniques like cone beam
hydrocephalus [3,4]. computedtomography[11,12],surfacelaserscanning[13,14],and
PLOSONE | www.plosone.org 1 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
stereophotogrammetry [15–17] have become available for de- Study selection
scribing and comparing craniofacial dimensions and shapes, First,studieswereindependentlyscreenedontitleandabstract
makingadiagnosisorplanningtreatment,andevaluatinggrowth by two reviewers (SB and MB) in a standardized manner. In an
and treatment outcomes. additionalstep,disagreementsbetweenreviewerswereresolvedby
In an earlier systematic review, we described various methods discussion and consensus. Second, full-text assessments for
for quantitative evaluation of facial dimensions in children up to eligibility were independently performed by the same two
age6yearsforavarietyofpurposes[18].Thisstudydescribesthe reviewersinastandardizedmanner.Again,inanadditionalstep,
methodsforquantitativeevaluationofcranialdimensions.Itsaims disagreementswereresolvedbydiscussionandconsensus.Results
are to 1) give an overview of soft tissue–based methods for of both first and second independently performed assessments of
quantitative longitudinal assessment of cranial dimensions in eligibilitywereanalyzedtoassessinter-raterreliability.Third,the
childrenuptoage6years;2)assessthemethodologicalqualityof
firstauthorperformedamanualsearchofthereferencelistsofthe
the studies using such approaches; and 3) assess the reliability of
included studies. Finally, all included studies were categorized as
thesemethodsappliedinstudieswithgoodmethodologicalquality.
describing facial or cranial evaluation of growth and treatment
outcome.Theplaneconnectingtheglabellawiththeleftandright
Methods
euryon arbitrarily separates thecranium fromtheface. Measure-
ments on or above this plane were considered to be cranial and
Protocol and registration
those below this plane to be facial. Studies describing cranial
Inclusion criteria and methods of analysis were specified in
evaluation ofgrowth andtreatment are included inthisreview.
advance and documented ina protocol. A registration number is
not available for this review since PROSPERO [19] was still in
Quality assessment
development whenitwas performed.
Studyqualitywasassessedbythequalityassessmentinstrument
(QAI)forclinicaltrialsusedbyGordonetal.(Table2)[21].This
Eligibility criteria
instrumentincludesanassessmentofstudybias.Acheckmarkwas
Eligibleforinclusionwereprimarypublicationsreportingon1)
made when a criterion was fulfilled. Depending on study design,
softtissue–basedevaluationoftheheadandface;2)childrenunder
qualityassessmentwasperformedonamaximumof15criteria.In
age6yearsatthestartofthestudy;3)quantitativechanges;and4)
case criteria were not applicable to a certain study design, fewer
longitudinal studies.
than 15 criteria were scored. Study quality is expressed as the
Excluded were publications describing 1) skeletal changes, 2)
percentage of criteria fulfilled in relation to the total number of
fetal growth, 3) animal studies, or 4) cross-sectional studies or
applicable criteria.
featuring 5) case reports, reviews, and letters. No restrictions for
Thescoreperstudyiscalculatedasapercentagebydividingthe
language, publication date, andpublication status wereimposed.
number of checkmarks by the number of applicable criteria and
multiplying by 100. Studies were grouped according to similarity
Information resources
of methods for measurement of cranial growth or treatment
Studies were identified by searching electronic databases. The
outcome. A mean quality score for each group of methods was
search was applied to PubMed (from 1948), EMBASE Excerpta
calculated. Arbitrarily, a cut-off of 60% or higher was graded as
Medica (from 1980), Cochrane Library (from 1993), Web of
goodqualityandbelow60%aspoorquality.Toassesstheinter-
Science (from 1945), Scopus (from 2004), and CINAHL (from
rater reliability of the assessment of study quality, 19 randomly
1982). The last search was run on October 1, 2012. In addition,
selected studies werescored bytworeviewers (SBandAK).
we manually searched the reference lists of included studies for
potentially eligible studies. Digital full-text publications were
Data extraction
retrieved from licensed digital publishers, and paper publications
were retrieved from the library. In cases in which the full-text Methodsusedinstudieswithgoodmethodologicalqualitywere
publication could not be retrieved, authors were requested by e- assessed for reliability expressed as the magnitude of the
mailtoprovide thearticle.Thegray literature wasnot searched. measurement error and the correlation coefficient between
repeated measurements.
Search strategy
Thesearchstrategywasdevelopedanddatabasesselectedwith Statistics
the help of a senior librarian specialized in health sciences. Cohen’s kappa statistics were used to assess the inter-rater
Databases selected were PubMed, EMBASE, Excerpta Medica, agreementfortheprocessofstudyselectionandforeachcriterion
Cochrane Library, Web of Science, Scopus, and CINAHL. oftheQAI.AccordingtoLandisandKoch,thelevelofinter-rater
Medical Subject Headings and free-text words were used for the agreementisverygoodifthevalueofKis0.81–1.00,goodifKis
searchstrategyofPubMed(Table1).Thesearchstrategiesforthe 0.61–0.80,moderateifKis0.41–0.60,fairifKis0.21–0.40,and
other databases were directly derived from the former. The last poor ifKis#0.20[22].
search was performed onOctober 1,2012. Analysis of variance and non-parametric Kruskal-Wallis tests
The search strategy focused on four categories of terms, as were performed to evaluate differences in mean scores between
follows: (1) terms to search for the population of interest (i.e., groups of methods.SPSS version21.0 wasused foranalysis.
babies, infants, and pre-school children), for which a selection of
theappropriatetermsfromthe‘Child’searchstrategywasmadeto Results
sortoutcitationsnotreportingonchildrenbetween0and6years
of age [20]; (2) terms to search for growth and methods for Study selection
quantitativeevaluation(i.e.,growth,anthropometrics,andimaging The inter-rater kappa for screening on title and abstract was
techniques);(3)termstosearchfortheanatomicregionofinterest 0.76.Forfull-textassessmentofeligibility,thekappawas0.69.The
(i.e., face and head); and (4) terms to search for the longitudinal reliabilityofbothstepsintheprocessofstudyselectionisqualified
aspect (i.e., cohortandfollow-up studies). as good[22].
PLOSONE | www.plosone.org 2 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Table1. SearchstrategyPubMed.
SearchstrategyPubMed
(‘‘Face’’[Mesh:noexp]ORface[TiAb]ORfacial[TiAb]ORcraniofacial[TiAb]ORORORborn*
craniomaxillofacial[TiAB]ORmaxillofacial[TiAb]ORdentofacial[TiAb]OR‘‘Facies’’[Mesh]
facies[TiAb]OR‘‘Head’’[Mesh:noexp]ORhead[TiAb])AND(‘‘Growthand
Development’’[Mesh:noexp]OR‘‘Growth’’[Mesh:noexp]OR‘‘growthanddevelopment’’[Sh]
growth[TiAb]OR‘‘Anthropometry’’[Mesh:noexp]ORanthropometr*[TiAb]OR
‘‘cephalometry’’[Mesh]ORcephalometr*[TiAb]OR‘‘imaging,three-dimensional’’[MeSH
Terms]OR‘‘three-dimensionalimaging’’[TiAb]OR‘‘3dimaging’’[TiAb]OR
‘‘Photogrammetry’’[Mesh]ORphotogrammetry[TiAb]OR‘‘Tomography,XRay
Computed’’[Mesh]OR‘‘Tomography,XRayComputed’’[TiAb]OR‘‘Lasers’’[Mesh:noexp]OR
laser[TiAb]OR‘‘MagneticResonanceImaging’’[Mesh:noexp]OR‘‘magneticresonance
imaging’’[TiAb]ORMRI[TiAb])AND(infantORinfan*ORnewbornORnewborn*ORnew
ORbabyORbaby*ORbabiesORneonat*ORperinat*ORpostnat*ORtoddler*OR
kindergar*ORpreschool*ORpreschool)AND(‘‘CohortStudies’’[Mesh]OR((cohort[TiAb]
ORlongitudinal[TiAb]ORfollowup[TiAb]ORfollowup*[TiAb])AND(study[TiAb]OR
studies[TiAb])))
doi:10.1371/journal.pone.0089602.t001
The search of PubMed, EMBASE, Cochrane Library, Web of identified212studiesthatmettheinclusioncriteria.Thelaststep
Science,Scopus,andCINAHLprovidedatotalof7027citations, in the inclusion process divided the studies into those on facial
andthemanualsearchprovided198citations.Afteradjustingfor evaluation (n=47) and studies on cranial evaluation (n=165). A
duplicates, 5599 citations remained for screening of title and total of 188 studies originated from the electronic databases; the
abstract. Of these, 4490 studies were discarded because they did remaining24studiesoriginatedfromtheadditionalmanualsearch
not meet the eligibility criteria so that a total of 1109 studies of the references of the included studies. Figure 1 shows the
remained for full-text assessment. Of these, 897 studies were PRISMA flow diagram, and Checklist S1 shows the PRISMA
excludedfordifferentreasons,and195werediscardedbecausethe checklist[23].Thecurrentsystematicreviewisrestrictedtostudies
full-text publication could not be retrieved. We ultimately on cranial evaluation of growth and treatment outcomes in
Table2. Qualityassessment instrument [21].
I.Studydesign(7 )
A.Objective—objectiveclearlyformulated( )
B.Samplesize—consideredadequate( )
C.Samplesize—estimatedbeforecollectionofdata( )
D.Selectioncriteria—clearlydescribed( )
E.Baselinecharacteristics—similarbaselinecharacteristics( )
F.Timing—prospective( )
G.Randomization—stated( )
II.Studymeasurements(3 )
H.Measurementmethod—appropriatetotheobjective( )
I.Blindmeasurement—blinding( )
J.Reliability—adequatelevelofagreement( )
III.Statisticalanalysis(5 )
K.Dropouts—dropoutsincludedindataanalysis( )
L.Statisticalanalysis—appropriatefordata( )
M.Confounders—confoundersincludedinanalysis( )
N.Statisticalsignificancelevel—pvaluestated( )
O.Confidenceintervalsprovided( )
Maximumnumberof s =15
doi:10.1371/journal.pone.0089602.t002
PLOSONE | www.plosone.org 3 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Figure1.PRISMAflowdiagramofthestudyselectionprocess.
doi:10.1371/journal.pone.0089602.g001
children; a systematic review of studies on facial evaluation of Study quality assessment
growth and treatment outcomes in children is described in a Inter-rater reliability values for all 15 criteria of the QAI were
separate publication [17]; betweenkappa0.19and1.00;11outof15criteriahadakappaof
Of the 165 included studies, 136 studies used direct anthro- 0.50 or higher. Inter-rater agreement on criteria E (similar
pometry for head circumference, 19 studies used other direct baseline characteristics), I (blind measurement), and K (dropouts
cranial anthropometry approaches, 3 studies used 2D photogra- included indata analysis) wasbelow 0.20.
phy, and 7 studies used 3D imaging techniques (5 stereophoto- Assessment of methodological quality of all reviewed studies
grammetry and2 surface laserscanning). resulted in scores ranging from 20% to 100%. A total of 118
studiesqualifiedasgoodbasedonamethodologicalqualityscore
equal to or above 60%. Score summaries of studies with good
methodological quality using direct anthropometry for head
PLOSONE | www.plosone.org 4 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Table3. Methodologicalquality scores of studiesusingdirectanthropometry forhead circumference reporting on softtissue–
basedquantitativelongitudinalassessmentofcranialdimensionsinchildrenuntilage6yearswithascoreequaltoorabove60%
(n=95) [24–118].
Firstauthor Year Design Measure Statistics Score
A B C D E F G H I J K L M N O
Headcircumferenceanthropometry
Belfort 2012 o . o . . o . o 64%
Bendersky 1998 o . o . . o 75%
Berry 1997 o . . . o . o 64%
Bhalla 1993 o . o . . . . o o 60%
Binns 1996 o . o . . . . o 70%
Bouthoorn 2012 o . . . o o 75%
Bracewell 2007 o . o . o o . 67%
Butz 2005 o . o o o 69%
Cardoso 2007 o . . . o o o 67%
Carlson 1996 o o o 80%
Chaudhari 2012 o . . o o o 69%
DeBruin 1998 o . o o o o 64%
DeReignier 1996 . . o o o o 69%
Desmyttere 2009 . o o o 79%
Desmyttere 2009 . o o o o 71%
Donma 1997 o o o . o o 64%
D’Souza 1986 o . . o o o o 62%
Durmus 2011 o . . o 85%
Ekblad 2010 o . . o o 77%
Elwood 1987 o . o . . . o o 64%
Erasmus 2002 o o o 80%
Eregie 2001 o . . . o o . 73%
Ernst 1990 o . o . . o . o 64%
Farooqi 2006 o . . o o 77%
Ford 2000 o . . o 85%
Ford 1986 o . . . o o . o 64%
Friel 1993 o o o o 73%
Fukumoto 2008 o . o . . o . . o 60%
Gale 2006 o o . . o 77%
Gale 2004 o . o . . o . . 70%
Georgieff 1995 o . . o o o o 62%
Georgieff 1989 o . o . o o . . o 60%
Gross 1983 o . . o o o o 62%
Gross 1983 o . . . o o o 75%
Guo 1988 o . . . . . o o 60%
Hagelberg 1990 o o o o o 67%
Hansen-pupp 2011 o . . . o o o 67%
Herrmann 2010 o . o . . o . . o 60%
Ishikawa 1987 o . . . o o . . 69%
Jaffe 1992 o o . o . . . . o 60%
Jaldin 2011 . . . o o . o 73%
Jaruratanasirikul 1999 o . . o o o o 62%
Kan 2007 . o . o . 77%
Karatza 2003 o . . o o 77%
Kiran 2007 . o o o 79%
PLOSONE | www.plosone.org 5 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Table3. Cont.
Firstauthor Year Design Measure Statistics Score
A B C D E F G H I J K L M N O
Headcircumferenceanthropometry
Kitchen 1992 o . o . o o . 67%
Koo 2006 o o 87%
Lainhart 1997 o . . . o . o 75%
Lasekan 2011 o o o o 73%
Lasekan 2006 o o 87%
Lira 2009 o . o . . o 75%
Lucas 2001 o o 87%
Maguire 2008 o o 87%
Makrides 2000 o 93%
Makrides 1999 o o 87%
Mamabolo 2004 o . o . . o . . o 60%
Marks 1979 o . o . . o . . o 60%
Maserei 2007 o o 87%
Mathur 2009 o . . . o o o 62%
McCowan 1999 o . o . . o . . 70%
McLeod 2011 o . . . o o . o 64%
Mercuri 2000 o . o . o . . o 64%
Meyer-Marcotty 2012 o . . o o o o 62%
Moore 1995 o . . . o o o 62%
Moye 1993 o . . o o o 69%
Nelson 1997 o . . . o o o 62%
Ochiai 2008 o . . . o o o 73%
Olivan 2003 o . . . o o . o 64%
Oliveira 2007 o . . . o o o 67%
Padilla 2010 . . o o 85%
Paul 2008 . . . o o . o 72%
Peng 2005 o . o o o 71%
Piemontese 2011 o o 87%
Polberger 1999 o o o o 73%
Rijken 2007 o . . . o 77%
Roberfroid 2012 o o 87%
Roche 1987 o o . o . . o . o 60%
RodriguezGarcia 2003 o . . . o . . o o 60%
Ross 2012 o . . . o o o 67%
Rothenberg 1999 o . o . . . 82%
Saliba 1990 o . . o o o 69%
Sawada 2010 o . . . o o . o 64%
Schaefer 1994 o . . . o o o 62%
Sharma 2011 o o . . . o o . 64%
Shaw 1999 o o o o 76%
Sheth 1995 o . o . . o . o o 60%
Shortland 1998 o o o o o 76%
Tan 2008 o o o o 73%
Tinoco 2009 . o . . o . . o 70%
Touwslager 2008 o . o . . o . . o 60%
Vaidya 2008 100%
VanDaalen 2007 o . o . . . . o 70%
PLOSONE | www.plosone.org 6 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Table3. Cont.
Firstauthor Year Design Measure Statistics Score
A B C D E F G H I J K L M N O
Headcircumferenceanthropometry
Whitehouse 2010 o . . . o o o 67%
Wood 2003 o . . . o o 75%
Zabaneh 2011 o . . . o o o 67%
= fulfilledthemethodologicalcriteriasatisfactorily.
o = didnotfulfillthemethodologicalcriteria.
. = notapplicable.
doi:10.1371/journal.pone.0089602.t003
circumference [24–118] are shown in Table 3 (n=95); those for a correlation coefficient between repeated measurements
other direct anthropometry [119–132] are shown in Table 4 [9,143,144] and represents the ability of observers to make a
(n=14); and those for indirect 2D and 3D imaging techniques consistent analysis. In this systematic review, reliability in studies
[133–141] are shownin Table5 (n=9). with good methodological quality was assessed and expressed by
Analysisofvariance(p=0.14)andKruskal-Wallistest(p=0.16) duplicatemeasurementerrorsandcorrelationcoefficientsbetween
revealed no statistically significant difference for methodological repeated measurements. Direct measurement of head circumfer-
quality, expressed as a percentage between groupsof methods. enceisthemostoftenusedmethodforsofttissue–basedevaluation
of cranial growth and treatment outcomes. The use of growth
Reliability charts in the clinical assessment of growing infants and children
Scores for reliability of methods for soft tissue–based quantita- and in pediatric nutritional screening and epidemiologic assess-
tivelongitudinalassessmentareshowninTable6.Only12ofthe ments has already been recommended for decades [27]. For this
118 studies with good methodological quality report data for the purpose, length and weight also are recorded in many countries
reliability of themethod toquantifycranial dimensions;8 studies frombirthonwards.Directanthropometryforheadcircumference
used direct anthropometry for head circumference seems to be a generally accepted method for most researchers
[25,27,35,58,63,71,100,103], two used other kinds of direct becauseonly8outof95studieswithgoodmethodologicalquality
anthropometry [126,127], one study used 3D laser scanning reportedonitsreliability.Measurementerrorsvariedfrom0.2to
[136],and oneused3Dstereophotogrammetry [139]. 1.7 mm,andcorrelationcoefficientswereverygood.Otherkinds
Regarding direct anthropometry for head circumference, 5 of direct anthropometry yielded a measurement error of 1mm.
studieswithgoodmethodologicalqualityreportedameasurement Normalgrowthofheadcircumferenceshowsanincreaseformean
error equal to or below 1.7mm. Two studies with good head circumference from 34–36cm at birth to 51–52 by age 6
methodological quality using direct anthropometry reported years [145]. The measurement errors for head circumference
correlation coefficients between repeated measurements of 0.87 anthropometry presented in this review are within 1% of the
and 0.99 and were both qualified as very good. Regarding other values of normal growth, which seems to be negligible. Direct
kinds of direct anthropometry, two studies with good methodo- anthropometry is a reliable and cheap method to study linear
logical quality reported a measurement error of1 mm. dimensionsandhasbeenregardedasthegoldstandardformany
Studies of good methodological quality using 2D photography decades, but it requires patient cooperation and precludes
and reporting the measurement error or correlation coefficients archiving [146].
were not identified among the included studies. One study with Photographic techniques, on the other hand, can capture
good methodological quality using 3D laser scanning reported a imagesfordatastorage. Themostcommonimaging techniqueis
measurement error of 0.5mm, and another using 3D stereo- 2Dphotography,whichhastheadvantagesofbeingsafe,relatively
photogrammetry reported a measurement errorof 0.02–4.3mm. cheap,anduser friendly.However, because noneof theincluded
studiesreportedameasurementerrororcorrelationcoefficient,its
Discussion reliabilityforevaluationofcranialgrowthandtreatmentoutcome
isuncertain.
Summary of evidence Various 3Dimaging techniques have beenrecently introduced
The objectives of this systematic review were to 1) give an andwereappliedin6outof165studiesincludedinthissystematic
overviewofsofttissue–basedmethodsforquantitativelongitudinal review. Only two studies with good methodological quality
assessment of cranial dimensions in children until age 6 years; 2) reported the measurement error (Table 6). The measurement
assess the methodological quality of the studies using such error in one study using 3D laser scanning was 0.5 mm; in one
approaches; and 3) assess reliability of these quantitative study using 3D stereophotogrammetry, it was 0.02 mm for head
measurement methods used in studies with good methodological width, 0.04 mm for head circumference, and 4.3mm for vertex
quality. height. Therefore, 3D imaging might be a reliable method to
In the literature, various terms to describe measurement error quantifycranialdimensions.Reliabilityofmeasurementsfrom3D
exist. Some studies use accuracy to describe landmark identifica- imaging seemstobemorerelated totheexactanatomical region
tion error, which in turn may consist of operator error, capture of interest than tothemethoditself.
error, and registration error [142]. More often in the literature, Whenreviewingtheliteratureforthisstudy,wefoundonlysix
reliability is used to describe landmark identification error of a included studies with good methodological quality using 3D
method.Reliabilitycanbeexpressedbythemeasurementerroror imagingtoquantifysofttissue–basedcranialgrowthandtreatment
PLOSONE | www.plosone.org 7 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
Table4. Methodologicalquality scores of studiesusingdirectanthropometry reporting onsoft tissue–based quantitative
longitudinalassessmentofcranialdimensionsinchildrenuntilage6yearswithascoreequaltoorabove60%(n=14)[119–132].
Firstauthor Year Design Measure Statistics Score
A B C D E F G H I J K L M N O
Anthropometry
Agrawal 2006 o . o . . o . . 70%
Chatterjee 2009 o . o . . o . . o 60%
Fearon 2009 o . o . . o . . o 60%
Fearon 2006 o . o . . o o . o 60%
Kelly 1999 o . o . . o . o 64%
Lee 2006 o . o . . o . . 70%
Lee 2008 . . o o . 77%
Littlefield 1998 o . o . . . o o 64%
Mulliken 1999 o . . o . o o 62%
Pedroso 2008 o . o . . o . o 64%
Teichgraeber 2004 o . o . o . . 62%
Teichgraeber 2002 o . . . o o . o 64%
VanVlimmeren 2008 o o 78%
VanVlimmeren 2007 o . . o o o 69%
= fulfilledthemethodologicalcriteriasatisfactorily.
o = didnotfulfillthemethodologicalcriteria.
. = notapplicable.
doi:10.1371/journal.pone.0089602.t004
outcome. The explanation is that only recently have techniques will be published. Advantages of these 3D techniques are
become available to capture the full 360u image needed to study millisecond fast image capture, archival capabilities, a good-
cranialdimensions.Moststudiesusing3Dimagingconcernfacial resolution color representation, and no exposure to ionizing
growth and treatment outcome because this technique has been radiation. Furthermore, assessment of linear dimensions and
availablefortwodecades.Therefore,itisexpectedthatwithinthe cranialsizeandshapecanbemadethree-dimensionally.Theseare
nextdecade,morestudiesusing3Dimagingofcranialdimensions majoradvantagescomparedtomoresimplisticanalysesperformed
Table5. Methodologicalquality scores of studiesusingindirect 2D and3D imaging techniques reporting on softtissue–based
quantitativelongitudinalassessmentofcranialdimensionsinchildrenuntilage6yearswithascoreequaltoorabove60%(n=9)
[133–141].
Firstauthor Year Design Measure Statistics Score
A B C D E F G H I J K L M N O
2Dphotography
Hutchinson 2010 o o o o 73%
Hutchinson 2004 o . . . o o 75%
Hutchison 2011 o . . . o o o 67%
3Dsurfacelaserscanning
Plank 2006 o . o o o . o o 62%
3Dstereophotogrammetry
Collet 2012 o . . 92%
Lipira 2010 o o . o o . o 62%
Meyer-Marcotty 2012 o . . o o o o 62%
Schaaf 2010 o . . . o o . o 64%
Toma 2010 o . o . . o . o 64%
= fulfilledthemethodologicalcriteriasatisfactorily.
o = didnotfulfillthemethodologicalcriteria.
. = notapplicable.
doi:10.1371/journal.pone.0089602.t005
PLOSONE | www.plosone.org 8 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
with direct anthropometry. A shortcoming of 3D imaging is its acceptedreasonsforbias,despitealackofvalidationoftheQAI.
inabilitytocapturesurfacemorphologyinthepresenceofcranial Themajorityofinter-raterdisagreementsaroseintheassessment
hair.Thereliabilityof3Dimagingtechniquesforsofttissue–based of applicability of criteria E, I, and K to certain studies (similar
evaluationofcranialgrowthandtreatmentoutcomesneedstobe baselinecharacteristics,blindmeasurement,anddropoutsinclud-
investigated further. ed in data analysis, respectively). This greater frequency can be
explained by the absence of adequate instructions of this QAI
Limitations together with the presence of a wide variety of study designs.
Failuretoidentifyallrelevantreportsforasystematicreviewis Therefore, raters should test this QAI thoroughly and obtain
consensus before scoring. In the literature, no single tool is an
likelytoresultinbias[147].Forthisreason,highlysensitivesearch
obviouscandidateforassessmentofmethodologicalqualityofnon-
strategies were developed with the help of a senior librarian
randomized studies [150]. Attempts to validate QAIs like the
specialized in health sciences for a combination of both narrow
Newcastle-Ottowa [151] scale or the Jadad scale [152] produce
and broadhealth sciencedatabases.
highly arbitrary resultsandcannot demonstrate significanteffects
Theprocessofstudyselectionwasperformedinanindependent
on quality scores [153,154]. There is a need for a validated QAI
standardized manner by two reviewers to prevent unjustified
that ispreferably applicable toa wide range of studydesigns.
exclusion of eligible studies. The manual search of the reference
lists of the included studies was performed by only one reviewer.
Conclusions
Possiblyeligiblestudiescouldhavebeenmissedinthisstageofthe
selectionprocess.However,becauseonlyapproximatelyoneoutof
Directanthropometricalmeasurementofheadcircumferencein
tenstudieswasretrievedbythemanualsearch,thisomissionmight growing children below age 6 years is a reliable method for
be negligible. Furthermore, every effort was made to contact the assessing cranial dimensions. The non-invasive 3D surface laser
authorsbyemailincaseswhereonlineaccesswasnotpermittedor scanning and 3D-stereophogrammetry techniques can assess size
thejournalwasnotavailableinthelibrary.Nevertheless,failureto and shape three-dimensionally. However, their reliability for
retrieve full-text publications of possibly eligible studies (n=195) assessing cranialdimensions needstobeinvestigated further.
was inevitable.
The instrument used to assess methodological quality was Supporting Information
adapted from Lagrave`re et al. [148] who developed a methodo-
logical quality checklist to assess study design, study measure- ChecklistS1 PRISMA checklist.
ments, and statistical analysis in clinical trials. Since the (DOCX)
introductionthechecklisthasbeenmodifiedandusedbyGordon
et al. [21] and Van Vlijmen et al. [149]. Scientific use of the Acknowledgments
checklisthasbeenacceptedbecausethecriteriacheckforgenerally
WethankseniorhealthscienceslibrarianElmiePetersforsupportingthe
developmentofsearchstrategies.
Table6. Reliabilityofmethods forsofttissue–based
quantitativelongitudinalassessmentofcranialdimensionsin Author Contributions
childrenuntil age6 yearsinstudies withgood
Conceived and designed the experiments: S. Brons AK. Performed
methodologicalquality (methodological quality scoreequal the experiments: S. Brons MvB AK. Analyzed the data: S. Brons MvB.
toor above60%). Contributed reagents/materials/analysistools:S. Brons EB. Wrote
thepaper:S. Brons.Intellectual contribution:JDJSS. Berge´.
Correlation References
Firstauthor Year Measurementerror coefficient
1. McGarryA,DixonMT,GreigRJ,HamiltonDR,SextonS,etal.(2008)Head
Headcircumference,directanthropometry shapemeasurementstandardsandcranialorthosesinthetreatmentofinfants
withdeformationalplagiocephaly.DevMedChildNeurol50:568–576.
Bendersky 1998 . 0.99
2. ChanFC,KawamotoHK,FedericoC,BradleyJP(2013)Soft-tissuevolumetric
Bhalla 1993 0–0.2mm . changesfollowingmonoblocdistractionprocedure:analysisusingdigitalthree-
dimensionalphotogrammetrysystem(3dMD).JCraniofacSurg24:416–420.
DeBruin 1998 ns .
3. MorhardtDR,BarrowW,JaworskiM,AccardoPJ(2013)HeadCircumfer-
Guo 1988 1.6mm . ence in Young Children With Autism: The Impact of Different Head
CircumferenceCharts.JChildNeurol.
Jaffe 1992 1–2mm .
4. Boros CA, Spence D, Blaser S, Silverman ED (2007) Hydrocephalus and
Lainhart 1997 . 0.87 macrocephaly:newmanifestationsofneonatallupuserythematosus.Arthritis
Rheum57:261–266.
Roche 1987 0.9mm .
5. FarkasLG(1994)Anthropometryoftheheadandface;kj,editor.NewYork:
Rothenberg 1999 0.8–1.7mm . RavenPress.405p.
6. BartzelaTN,KatsarosC,BronkhorstEM,RizellS,HalazonetisD,etal.(2011)
Directanthropometry
Atwo-centrestudyonfacialmorphologyinpatientswithcompletebilateral
Littlefield 1998 1mm . cleftlipandpalateatnineyearsofage.IntJOralMaxillofacSurgInpress.
7. Broadbent BH (1931) A new X-ray technique and its application to
Mulliken 1999 1mm . orthodontia.AngleOrthod1:45–66.
3Dlaserscanning 8. NolletPJ,KatsarosC,HuyskensRW, Borstlap WA,Bronkhorst EM,etal.
(2008) Cephalometric evaluation of long-term craniofacial development in
Plank 2006 0.5mm . unilateralcleftlipandpalatepatientstreatedwithdelayedhardpalateclosure.
3Dstereophotogrammetry IntJOralMaxillofacSurg37:123–130.
9. FarkasLG,BrysonW,KlotzJ(1980)Isphotogrammetryofthefacereliable?
Meyer-Marcotty 2012 0.02–4.3mm . PlastReconstrSurg66:346–355.
10. DavisJP,ValentineT,DavisRE(2010)Computerassistedphoto-anthropo-
. = notreported. metricanalysesoffull-faceandprofilefacialimages.ForensicSciInt200:165–
ns = notsignificant. 176.
doi:10.1371/journal.pone.0089602.t006
PLOSONE | www.plosone.org 9 February2014 | Volume 9 | Issue 2 | e89602
MethodsforCranialGrowthinChildren:AReview
11. Cevidanes LH, Alhadidi A, Paniagua B, Styner M, Ludlow J, et al. (2011) embryobiopsyappliedinpreimplantationgeneticdiagnosisorpreimplantation
Three-dimensional quantification of mandibular asymmetry through cone- geneticscreeningHumReprod24470–476.
beam computerized tomography. Oral Surg Oral Med Oral Pathol Oral 39. Donma MM, Donma O (1997) The influence of feeding patterns on head
RadiolEndodInpress. circumferenceamongTurkishinfantsduringthefirst6monthsoflife.Brain
12. Nada RM, Maal TJ, Breuning KH, Berge SJ, Mostafa YA, et al. (2011) Dev19:393–397.
Accuracy and reproducibility of voxel based superimposition of cone beam 40. D’Souza SW, Gowland M, Richards B, Cadman J, Mellor V, et al. (1986)
computedtomographymodelsontheanteriorcranialbaseandthezygomatic Headsize,braingrowth,andlateralventriclesinverylowbirthweightinfants
arches.PLoSOne6:e16520. ArchDisChild611090–1095.
13. DjordjevicJ,TomaAM,ZhurovAI,RichmondS(2011)Three-dimensional 41. DurmusB,KruithofCJ,GillmanMH,WillemsenSP,HofmanA,etal.(2011)
quantificationoffacialsymmetryinadolescentsusinglasersurfacescanning. Parental smoking during pregnancy, early growth, and risk of obesity in
EurJOrthod. preschoolchildren:theGenerationRStudy.AmJClinNutr94:164–171.
14. TomaAM,ZhurovA,PlayleR,RichmondS(2008)Athree-dimensionallook 42. Ekblad M, Korkeila J, Parkkola R, Lapinleimu H, Haataja L, et al. (2010)
forfacialdifferencesbetweenmalesandfemalesinaBritish-Caucasiansample Maternalsmokingduringpregnancyandregionalbrainvolumesinpreterm
aged151/2yearsold.OrthodCraniofacRes11:180–185. infantsJPediatr156185–190e181.
15. KauCH,KamelSG,WilsonJ,WongME(2011)Newmethodforanalysisof 43. ElwoodPC,SweetnamPM,GrayOP,DaviesDP,WoodPD(1987)Growthof
facialgrowthinapediatricreconstructedmandible.AmJOrthodDentofacial children from 0-5 years: with special reference to mother’s smoking in
Orthop139:e285–290. pregnancyAnnHumBiol14543–557.
16. Maal TJ, van Loon B, Plooij JM, Rangel F, Ettema AM, et al. (2010) 44. ErasmusHD,Ludwig-AuserHM,PatersonPG,SunD,SankaranK(2002)
Registration of 3-dimensional facial photographs for clinical use. J Oral Enhanced weight gain in preterm infants receiving lactase-treated feeds: a
MaxillofacSurg68:2391–2401. randomized,double-blind,controlledtrialJPediatr141532–537.
17. vanLoonB,vanHeerbeekN,MaalTJ,BorstlapWA,IngelsKJ,etal.(2011) 45. EregieCO(2001)Exclusivebreastfeedingandinfantgrowthstudies:reference
Postoperative volume increase of facial soft tissue after percutaneous versus standards for head circumference, length and mid-arm circumference/head
endonasalosteotomytechniqueinrhinoplastyusing3Dstereophotogramme- circumferenceratioforthefirst6monthsoflifeJTropPediatr47329–334.
try.Rhinology49:121–126. 46. Ernst JA, Bull MJ, Rickard KA, Brady MS, Lemons JA (1990) Growth
18. BronsS,vanBeusichemME,BronkhorstEM,Draaisma J,BergeSJ,etal. outcomeandfeedingpracticesoftheverylowbirthweightinfant(lessthan
(2012) Methods to quantify soft-tissue based facial growth and treatment 1500grams)withinthefirstyearoflifeJPediatr117S156–166.
outcomesinchildren:asystematicreview.PLoSOne7:e41898. 47. FarooqiA,HagglofB,SedinG,GotheforsL,SereniusF(2006)Growthin10-
19. PROSPERO:Internationalprospectiveregisterofsystematicreviews. to 12-year-old children born at 23 to 25 weeks’ gestation in the 1990s: a
20. BoluytN,TjosvoldL,LefebvreC,KlassenTP,OffringaM(2008)Usefulnessof Swedishnationalprospectivefollow-upstudyPediatrics118e1452–1465.
systematicreviewsearchstrategiesinfindingchildhealthsystematicreviewsin 48. Ford G, Rickards A, Kitchen WH, Ryan MM, Lissenden JV (1986)
MEDLINE.ArchPediatrAdolescMed162:111–116. Relationshipofgrowthandpsychoneurologicstatusof2-year-oldchildrenof
birthweight500-999gEarlyHumDev13329–337.
21. GordonJM,RosenblattM,WitmansM,CareyJP,HeoG,etal.(2009)Rapid
49. FordGW,DoyleLW,DavisNM,CallananC(2000)Verylowbirthweightand
palatalexpansioneffectsonnasalairwaydimensionsasmeasuredbyacoustic
growthintoadolescenceArchPediatrAdolescMed154778–784.
rhinometry.Asystematicreview.AngleOrthod79:1000–1007.
50. FrielJK,AndrewsWL,MatthewJD,LongDR,CornelAM,etal.(1993)Zinc
22. Landis JR, Koch GG (1977) The measurement of observer agreement for
Supplementation in Very-Low-Birth-Weight Infants Journal of Pediatric
categorialdata.Biometrics33:159–174.
GastroenterologyandNutrition1797–104.
23. The PRISMA Statement website. Available: http://www.prisma-statement.
51. Fukumoto A, Hashimoto T, Ito H, Nishimura M, Tsuda Y, et al. (2008)
org.Accessed2013May9.
Growthofheadcircumferenceinautisticinfantsduringthefirstyearoflife.
24. BelfortMB,Rifas-ShimanSL,SullivanT,CollinsCT,McPheeAJ,etal.(2011)
JAutismDevDisord38:411–418.
Infantgrowthbeforeandafterterm:effectsonneurodevelopmentinpreterm
52. GaleCR,O’CallaghanFJ,BredowM,MartynCN(2006)Theinfluenceof
infants.Pediatrics128:e899-906.
headgrowthinfetallife,infancy,andchildhoodonintelligenceattheagesof4
25. BenderskyM,KoonsA,LewisM,HegyiT(1998)Developmentalimplications
and8yearsPediatrics1181486–1492.
ofheadgrowthfollowingintracranialhemorrhageClinicalPediatrics37469–
53. GaleCR,O’CallaghanFJ,GodfreyKM,LawCM,MartynCN(2004)Critical
476
periodsofbraingrowthandcognitivefunctioninchildrenBrain127321–329.
26. BerryMA,ConrodH,UsherRH(1997)Growthofveryprematureinfantsfed
54. GeorgieffMK,HoffmanJS,PereiraGR,BernbaumJ,Hoffman-WilliamsonM
intravenoushyperalimentationandcalcium-supplementedformulaPediatrics
(1985) Effect of neonatal caloric deprivation on head growth and 1-year
100647–653.
developmentalstatusinpreterminfants.JPediatr107:581–587.
27. BhallaAK,WaliaBN(1993)Percentilegrowthchartsforheadcircumference
55. GeorgieffMK,MillsMM,ZempelCE,ChangPN(1989)Catch-upgrowth,
inPunjabiinfantsIndianPediatr3041–46.
muscleandfataccretion,andbodyproportionalityofinfantsoneyearafter
28. BinnsHJ,SenturiaYD,LeBaillyS,DonovanM,ChristoffelKK(1996)Growth newbornintensivecareJPediatr114288–292.
ofChicago-areainfants,1985through1987.Notwhatthereferencecurves 56. Gross SJ, Eckerman CO (1983) Normative early head growth in very-low-
predict.PediatricPracticeResearchGroup.ArchPediatrAdolescMed150: birth-weightinfantsJPediatr103946–949.
842–849. 57. GrossSJ,OehlerJM,EckermanCO(1983)Headgrowthanddevelopmental
29. BouthoornSH,vanLentheFJ,Hokken-KoelegaAC,MollHA,TiemeierH,et outcomeinverylow-birth-weightinfantsPediatrics7170–75.
al. (2012) Head circumference of infants born to mothers with different 58. GuoS,RocheAF,MooreWM(1988)Referencedataforheadcircumference
educationallevels;theGenerationRStudy.PLoSOne7:e39798. and1-monthincrementsfrom1to12monthsofageJPediatr113490–494.
30. BracewellMA,HennessyEM,WolkeD,MarlowN(2008)TheEPICurestudy: 59. HagelbergS,LindbladBS,PerssonB(1990)Aminoacidlevelsinthecritically
growthandbloodpressureat6yearsofagefollowingextremelypretermbirth illpreterminfant givenmother’smilkfortified withproteinfromhumanor
ArchDisChildFetalNeonatalEd93F108–114. cow’smilk.ActaPaediatrScand79:1163–1174.
31. ButzAM,PulsiferMB,BelcherHME,LeppertM,DonithanM,etal.(2005) 60. Hansen-PuppI,LofqvistC,PolbergerS,NiklassonA,FellmanV,etal.(2011)
Infantheadgrowthandcognitivestatusat36monthsinchildrenwithIn-Utero Influenceofinsulin-likegrowthfactorIandnutritionduringphasesofpostnatal
drugexposureJournalofChild&AdolescentSubstanceAbuse1415–39. growthinverypreterminfants.PediatrRes69:448–453.
32. CardosoLE,FalcaoMC(2007)Nutritionalassessmentofverylowbirthweight 61. HerrmannKR,HerrmannKR(2010)Earlyparenteralnutritionandsuccessful
infants: relationships between anthropometric and biochemical parameters postnatalgrowthofprematureinfantsNutrClinPract2569–75.
NutrHosp22322–329. 62. IshikawaT,FuruyamaM,IshikawaM,OgawaJ,WadaY(1987)Growthin
33. CarlsonSE,WerkmanSH,TolleyEA(1996)Effectoflong-chainn-3fattyacid headcircumferencefrombirthtofifteenyearsofageinJapanActaPaediatr
supplementation on visual acuity and growth of preterm infants with and Scand76824–828.
withoutbronchopulmonarydysplasiaAmericanJournalofClinicalNutrition 63. Jaffe M, Tal Y, Hadad B, Tirosh E, Tamir A (1992) Variability in head
63687–697. circumferencegrowthrateduringthefirst2yearsoflifePediatrics90190–192.
34. ChaudhariS,OtivM,KhairnarB,PanditA,HogeM,etal.(2012)Punelow 64. JaldinMM,PinheiroFS,DosSantosAM,MunizNC,BritoLMO(2011)Head
birthweightstudy,growthfrombirthtoadulthood.IndianPediatr49:727– circumferencegrowthofexclusivelybreastfedinfantsduringthefirstsixmonths
732. oflife.RevPaulPediatr29:509–514.
35. deBruinNC,DegenhartHJ,GalS,WesterterpKR,StijnenT,etal.(1998) 65. JaruratanasirikulS,ChanvitanP,JanjindamaiW,RitsmitchaiS(1999)Growth
Energyutilizationandgrowthinbreast-fedandformula-fedinfantsmeasured patterns of low-birth-weight infants: 2-year longitudinal study. J Med Assoc
prospectivelyduringthefirstyearoflifeAmJClinNutr67885–896. Thai82:325–331.
36. deRegnierRA,GuilbertTW,MillsMM,GeorgieffMK(1996)Growthfailure 66. KanE,RobertsG,AndersonPJ,DoyleLW(2008)Theassociationofgrowth
andalteredbodycompositionareestablishedbyonemonthofageininfants impairmentwithneurodevelopmentaloutcomeateightyearsofageinvery
withbronchopulmonarydysplasiaJNutr126168–175. pretermchildrenEarlyHumDev84409–416.
37. Desmyttere S, Bonduelle M, Nekkebroeck J, Roelants M, Liebaers I, et al. 67. Karatza AA, Varvarigou A, Beratis NG (2003) Growth up to 2 years in
(2009)Growthandhealthoutcomeof1022-year-oldchildrenconceivedafter relationship to maternal smoking during pregnancy Clin Pediatr (Phila) 42
preimplantationgeneticdiagnosisorscreeningEarlyHumDev85755–759. 533–541.
38. DesmyttereS,DeSchepperJ,NekkebroeckJ,DeVosA,DeRyckeM,etal. 68. KiranPSS,DuttaS,NarangA,BhansaliA,MalhiP(2007)Multiplecoursesof
(2009) Two-year auxological and medical outcome of singletons born after antenatalsteroidsIndianJournalofPediatrics74463–469.
PLOSONE | www.plosone.org 10 February2014 | Volume 9 | Issue 2 | e89602