ORIGINAL PAPER

Cranial molding helmet therapy and establishment of practicalcriteria for management in Asian infant positional head deformity

Yasuo Aihara & Kana Komatsu & Hitoshi Dairoku &

Osami Kubo & Tomokatsu Hori & Yoshikazu Okada

Received: 29 April 2014 /Accepted: 12 June 2014# Springer-Verlag Berlin Heidelberg 2014

AbstractBackground The growing number of infants with deforma-tional plagiocephaly (DP) has raised clinical questions aboutwhich children, at what age, and how molding helmet therapy(MHT) should be performed especially in Japan.Methods A total of 1,011 Japanese pediatric head deformityinfants had undergone MHT after being diagnosed with non-synostotic DP. Three ratios of left to right comparison (ante-rior, posterior, and overall) were created and analyzed com-paring age of starting treatment, helmet wearing period, andseverity of skull deformity before with after MHT.Results The averages of head symmetry ratios after treatmentin all groups (for the occipital region) showed apparent im-provement; t(930)=−60.86, p=0.000. (t(932)=−57.8, p=0.000.) In the “severe” deformation group, the earlier thetreatment was started, the higher symmetry ratio recoverywas obtained. Treatment was especially effective when startedin 4-month-old infants. In contrast to the “severe” group, the“mild” deformation group showed that MHTwas most effec-tive if treatment started before 6 months of age. Again, theearlier the treatment was started, the higher symmetry ratiowas achieved, but compared to the “severe” group, it had amodest effect when treatment was started in infants older than8 months.Conclusion This is the first large-scale molding helmet studyreporting the method and efficacy in Japanese infants. Itdemonstrated that despite the structural and physiologicaldifferences from infants of other races, molding helmet

therapy is effective in Asian-born infants, provided that inter-vention timing and recognition conditions are met.

Keywords Positional plagiocephaly . Brachycephaly .

Scaphocephaly . Cranial molding helmet therapy .

Non-synostotic . Japanese infants

Introduction

Deformational plagiocephaly (DP) is a multi-planar deformityof the cranium occurring either pre- or post-natally in infantsand is one of the most frequently used terms among manyothers used for positional head deformity (PHD). This, as themany other “non-closed cranial suture” head deformity de-scriptions and reported entities resulting from prolonged in-tentional infant head positioning, if left untreated, may lead tosignificant cosmetic and functional–neurological and psycho-logical consequences. For example, the pressure exerted onthe intraorbital muscles and nerves, among others, can resultin sensory and motor disturbances [1]. As a consequence,infants with head deformities attempt to compensate for thehead’s abnormal orientation in space which can result inocular and vestibular impairment [2]. Given that the skullundergoes 85 % of its postnatal growth [3] within the firstyear of life, early recognition and treatment of PHDwithin thissmall window of opportunity is paramount [4]. However,relevant information on the cranial shape of Japanese infantsin relation to definition, recognition, and treatment of cranialdeformities has not been reported until now [5]. After the“back to sleep” campaign in the USA [2], the number ofPHD cases worldwide increased, and for the first time inJapan, we have evaluated and treated 1,011 patients. Docu-mentation of the effectiveness of deformity treatment requiresfeasible and reproducible methods to quantify head shape andany existing asymmetry. This study aimed to demonstrate that

Y. Aihara (*) :K. Komatsu :O. Kubo : T. Hori :Y. OkadaDepartment of Neurosurgery, Tokyo Women’s Medical University,8-1 Kawada-cho Shinjuku-ku, Tokyo 162-8666, Japane-mail: yaihara@nij.twmu.ac.jp

H. DairokuFaculty of Human Sciences, University of Tsukuba, Ibaraki, Japan

Childs Nerv SystDOI 10.1007/s00381-014-2471-y

despite structural and physiological differences, molding hel-met is effective in Asian-born babies, and this non-surgicalintervention is successful in improving cases provided thattiming and recognition conditions are met.

Materials and methods

Three-dimensional head shape capturing and quantifying wasperformed in a clinical setting using the STAR scanner(Orthomerica, Orlando, FL) laser data acquisition system[6]. The software programs provided with the STAR scannerwere used to capture three-dimensional images, quantify andcompare head shape changes, and then design and produce theindividual cranial orthosis in all infants with positional headdeformity.

The three-dimensional image was divided into 12 equallevels or cross sections (Fig. 1). Level 0 represented a crosssection running through the sellion and right and left tragions.The software reconstructed 10 equal cross sections of the skullsuperior to level 0 and 2 inferior to it. The height of each crosssection or level was determined by dividing the overall heightof the child’s head above the 0 plane into those 10 equallevels, and two levels with the same height were added below.The software used the image of these 12 levels to evaluategrowth on each scan comparing cross sections over time andby that quantified the dynamics of growth. Each cross-sectionwas divided into four quadrants (quadrant 1, anterior left;quadrant 2, anterior right; quadrant 3, posterior right; quadrant4, posterior left), which was created by the software, defininga y-axis from the midpoint between the two tragion landmarks(on the x-axis) and sellion. Based on these x- and y-axes werebuilt the two planes rising above the skull base and dividingthe skull into four quadrant volumes. Q1, Q2, Q3, Q4 defaultvolumes were based on level volumes from 2 to 8 addedtogether (Fig. 2). This method allowed clinicians to assesssymmetry at multiple cross sections and as a total for thequadrants of each group. Quadrant volumes would changerelative to changes in growth, symmetry, and proportion.Quadrant volumes were also used to attempt defining quanti-tatively symmetry by specific ratios, as anterior symmetryratio (ratio of Q1 volume to Q2 volume or vice versa, which-ever is less than one), posterior symmetry ratio (the ratio of Q3volume to Q4 volume or vice versa, whichever is less thanone), and overall symmetry ratio (Fig. 3). This last parameterresulted from averaging the anterior (left and right) and pos-terior (left and right) volume compartments using specificproprietary software method of volume definition. The clini-cal significance of these parameters might be obvious for theanterior and posterior symmetry ratios, where the bossing areahas bigger volume than the opposite side, and in the process ofsuccessful treatment and achieving symmetry, left and rightvolumes will come closer to each other (approaching 1.0). As

it might be expected, the changes and corrections in theseratios will be more significant in plagiocephaly and less inbrachi- and scaphocephaly. We have defined the goal oftreatment at 0.9 or higher ratio values. The overall symmetryratio as a global symmetry formula including all quadrants isless specific in expressing particular deformities and theircorrections, and can indicate the anterior to posterior propor-tion in general, that is influenced by the prolonged face-uphead position.

Molding helmet treatment method

In mild plagiocephaly, the helmet was built according to theinitial diagram, aiming contact with the whole head surfaceexcept the quadrant of reduced volume, while in moderate andsevere cases, the helmet contacted the two bossing areas. Inmild to moderate brachycephaly, contact was maintained in thefrontal and anterior parietal areas. Severe brachycephaly re-quired similar attitude with special attention to the parietal area,where additional buildup should take place (Figs. 4 and 5).

Patient population

A total of 1,011 infants diagnosed by a pediatrician, neuro-surgeon, or craniofacial plastic surgeon over a 4-year period(2007–2011) with moderate to severe craniofacial positionaldeformity were referred to Tokyo Women’s Medical Univer-sity (TWMU) for a cranial remolding orthosis. Themale/female ratio was 721:290. Plagiocephaly was diagnosedin 964, brachycephaly in 44, and scaphocephaly in 3 patients.The age range was from 3 to 12 months. We excluded 80patients lost to follow-up (74 with plagiocephaly, 5 withbrachycephaly, and 1 with scaphocephaly) (Table 1).

All patients had molding helmet therapy for approximately3–6 months. We divided the patients according to the severityof deformity into severe deformity group (SDG)—both over-all symmetry ratio under 85.5 % and posterior symmetry ratiounder 80.5% and mild deformity group (MDG)—above thesevalues.

The patients were divided into groups according to the agein months at presentation (3, 4, 5, 6, 7, 8, and 9-month-oldgroups), showing the distribution as in Table 2.

On the other hand, we analyzed the results according to theage when treatment for plagiocephaly was started, obtainingthe following groups: at 4 months 184, at 5 months 198, at6 months 140, at 7 months 109, at 8 months 88, and older than8 months 125.

The 3 months of age group was excluded at statisticians’recommendation due to its lower number of patients and theirdistribution, compared to the other groups, which would havecompromised the results of the analysis of variance.

Childs Nerv Syst

Analysis

We evaluated the overall and posterior symmetry ratios forsignificant improvement after treatment. We also evaluatedthe role of age when treatment was started relative to outcome.To compensate for the deformation ratio factor before treat-ment, considering the likelihood of a ceiling and floor effect ineach age group, we performed analysis of variance of threefactors (overall/posterior head deformity ratio before and aftertreatment, age inmonths when treatment started (4, 5, 6, 7, 8, orolder), and degree of deformity (mild or severe) (Tables 3 and4), applying a level of significance 0.01. We used the standardWindows software SPSS ver. 18.00 to perform the analysis.

Results

Treatment effect on overall symmetry ratio

(1) Paired sample t test results before and after treat-ment for the overall symmetry ration in allchildren.

Averages of overall symmetry ratio values for all partici-pants, comparing before (M=86.3 %, SD=1.14) and aftertreatment (M=92.0, SD=1.71), showed a statistically signifi-cant difference—t(930)=−60.86, p=0.000 and indicated im-provement after treatment.

Fig. 1 Cross-section software divides the head into 12 cross-sectionallevels and defines the 0 cutting plane by driving a plane through thesellion (se) and tragion landmarks. The height of each cross section orlevel is determined by dividing the overall height of the child’s headabove the 0 cutting plane into equal levels. Ten levels are above the 0

level, and two levels are below the 0 level. This provides a way tocompare cross sections over time and addresses the dynamic of growthwhen comparing two scans. Level 3 is used as the default level for thesummary report. Z-axis is perpendicular to the base plane

Fig. 2 The software creates the y-axis by finding the midpoint ororigin between the two tragionlandmarks and driving a planethrough the sellion and origin.Q1, Q2, Q3, Q4 default volumesare based on levels 2–8 addedtogether. Quadrant 1: Anterior leftquadrant, Quadrant 2: Anteriorright quadrant, Quadrant 3:Posterior right quadrant, Quadrant4: Posterior left quadrant

Childs Nerv Syst

(2) Analysis of variance (ANOVA) for three variables of theoverall symmetry ratio in plagiocephaly children.

We analyzed symmetry ratios before and after treatmentregarding the following variables: age in months when

treatment started (4, 5, 6, 7, 8, 9, or older) and degree ofdeformity (mild or severe), and the results are shown in Table 3.

Comparing pre- and post-treatment, overall symmetry ratioresults in these children showed statistically significant differ-ence (F[1, 832]=4743.60, p<0.0001), indicating a higher

Fig. 3 Asymmetry grading.Mild—one posterior quadrantinvolvement is flat with minimalear shift. Mild-moderate—oneposterior quadrant is flat withmoderate ear shift on same side asposterior flattening. Moderate—one posterior quadrant, andcontralateral anterior quadrant isinvolved with moderate ear shift.Facial asymmetry may be present.Two quadrant involvement andear shift. Severe—All fourquadrants are either flat or bossedwith severe ear shift, forehead,and orbital asymmetry, andasymmetry of the cheek and jaw

Fig. 4 The mechanism of the Helmet therapy which improves the headshape of babies with deformational plagiocephaly and brachycephaly. (a)The flattened areas are built up with plaster in the posterior-lateralquadrant to obtain symmetry. The flattened frontal area is also built upwith plaster to obtain symmetry. Contact will be maintained over the

prominent or bossed areas to deter growth in those areas. (b) Primarybuildup on the positive mold occurs across the central occipital region toobtain improved proportions of the head. Contact is maintained over thefrontal and parietal regions to deter growth

Childs Nerv Syst

symmetry ratio after treatment. The role of age of initiatingtreatment (F[5, 832]=21.76, p<0.0001) and degree of defor-mity (F[1, 832]=943.88, p<0.0001) were also significant.The three-way interaction was also significant (F[5, 832]=4.01, p=0.001) While the effect of age when initiating treat-ment was not significant before treatment for both severitygroups (severe—F[5, 832]=1.584, p=0.162; mild—F[5,832]=2.97, p=0.011), it was after treatment for both ofthem (severe—F[5, 832]=29.25, p<0.0001; mild—F[5,832]=27.74, p<0.0001).

The application of the multiple comparison of Bonferronimethod to post-treatment data showed that for severe defor-mity group, symmetry ratio of 4-month olds was significantlyhigher than that of other age groups (4–5, 4–6, 4–7, 4–8, 4–older than 8—MSe=10.53, p<0.01 for each), symmetry ratioof 5-month-olds was also significantly higher than that of 7-month-olds and those older than 8 months (5–7, 5–older than8—MSe=10.53, p<0.01 for each), and symmetry ratio of 6-

month-olds was significantly higher than that of older than8 months group (6–older than 8 month—MSe=10.53,p<0.01). These results indicated that the earlier the treatmentstarted, the higher the symmetry ratio rises in the severedeformity group, and especially most effective when startedin 4-month-olds (Fig. 6a).

The same method was applied to the post-treatment dataand showed that for the mild deformation group, symmetryratio of 4-month-olds was significantly higher than that of 7-month-olds, 8-month-olds, and older than 8 months (4–7, 4–8,4–older than 8—MSe=10.53, p<0.01 for each), symmetryratio of 5-month-olds was also significantly higher than that of7-month-olds, 8-month-olds, and those older than 8 months(5–7, 5–8, 5–older than 8—MSe=10.53, p<0.01 for each),symmetry ratio of 6-month-olds was also significantly higherthan that of 8-month-olds and older than 8 months (6–8, 6–older than 8—MSe=10.53, p<0.01 for each), and symmetryratio of 7-month-olds was significantly higher than that of thegroup older than 8 months (7–older than 8 months—MSe=10.53, p<0.01). Again, these results indicate that the earlierthe treatment is started, the higher the symmetry ratio rises inboth deformity group, but in the mild deformity group, it wasmost effective if started even before the age of 6 months(Fig. 6b).

Treatment effect on occipital symmetry

(3) Evaluation of results before and after treatment in theoccipital region of all children.

Fig. 5 Analyzing measurementsheet of plagiocephaly pre- (redshape) and post-treatment (blueshape) head shape. This providesa way to compare cross sectionsover time and addresses thedynamics of growth whencomparing the two scans. (a)Three-dimensional skull viewpre- (b) and post-molding (c)helmet therapy for brachycephaly.The part of blue color shows thepart of remodeling

Table 1 Excluded cases in each skull deformity type

Exclusion

Whole head Occipital region

Plagiocephaly 74 72

Branchycephaly 5 5

Scaphocephaly 1 1

Total 80 78

Childs Nerv Syst

Paired sample t test

The comparison of the posterior symmetry ratio before andafter treatment for all participants resulted in a significantdifference in the averages of this ratio before (M=81.8 %,SD=1.29) and after treatment (M=91.3, SD=2.72); t(932)=−57.8, p=0.000. These results indicated that after treatment,the occipital region in all participants showed apparentimprovement.

(4) Analysis of variance (ANOVA) for three variables in theoccipital region of plagiocephaly children.

Symmetry ratio data before and after treatment were ana-lyzed by a three-way analysis of variance (ANOVA) with theage of starting treatment (4, 5, 6, 7, 8, or older) and degree ofdeformity (mild or severe) in the posterior quadrants in chil-dren with plagiocephaly. The results of pre-post treatment areshown in Table 4.

The change of the posterior symmetry ratio comparedbefore and after treatment was higher after treatment (F[1,833]=3716.8, p<0.0001), being significant. The effect of agewhen initiated treatment (F[5, 833]=20.2, p<0.0001), thedeformation severity (F[1, 833]=251793.8, p<0.0001), andthe three-way interaction were also significant (F[5, 833]=

3.18, p=0.008). As in the case of the overall symmetry ratio,the effect of age was not significant before treatment for bothseverity groups (severe—F[5, 833]=1.81, p=0.108; mild—F[5, 833]=1.08, p=0.369); however, it was significant after it(severe—F[5, 833]=25.5, p<0.0001; mild—F[5, 833]=20.8,p<0.0001).

The multiple comparison of Bonferroni for post-treatmentdata showed that in the severe deformation group, symmetryratio of 4-month-olds was significantly higher than that of 7, 8,and older than 8 month of age (4–7, 4–8, 4–older than 8—MSe=28.63, p<0.01 for each), symmetry ratio of 5-month-olds was also significantly higher than that of 7, 8, and olderthan 8-month-olds (5–7, 5–8, 5–older than 8—MSe=28.63,p<0.01 for each), and symmetry ratio of 6-month-olds wassignificantly higher than that of 7, 8, and older than the 8-month-old group (6–7, 6–8, 6–older than 8 months—MSe=28.63, p<0.01). These results indicated that the earlier thetreatment started, the higher the symmetry ratio rises in thesevere deformation group, and it is most effective if startedbefore the age of 6 months (Fig. 6c).

The use of the same multiple comparison of Bonferronimethod for post-treatment data also showed that in the milddeformity group, posterior symmetry ratio of 4-month-oldswas significantly higher than that of 7, 8, and older than

Table 2 Age grouping by sexand case attributes in the study Age (month) Number Sex (male/female) Plagiocephaly Brachycephaly Scaphocephaly

3 50 (34:16) 48 2 0

4 205 (145:60) 196 9 0

5 223 (167:56) 212 10 1

6 159 (116:43) 151 8 0

7 127 (81:46) 120 7 0

8 99 (70:29) 95 3 1

9∼ 148 (108:40) 142 5 1

Total 1,011 (721:290) 964 44 3

Table 3 Mean value (according to the initial severity) of the overallsymmetry ratio after helmet therapy in each starting age

Overall ratio

Mild group Severe group

Age(month) Pre-M (SD) Post-M (SD) Pre-M (SD) Post-M (SD)

4 89.2 (±2.7) 95.5 (±2.4) 81.9 (±2.9) 92.6 (±2.3)

5 88.9 (±2.7) 95.0 (±2.8) 81.9 (±2.5) 90.9 (±2.0)

6 89.7 (±2.5) 94.6 (±2.8) 81.9 (±3.0) 90.2 (±2.3)

7 89.1 (±3.1) 93.6 (±2.8) 81.7 (±2.7) 88.2 (±2.3)

8 88.5 (±1.7) 92.8 (±2.5) 83.3 (±2.2) 89.0 (±1.8)

9∼ 88.6 (±2.2) 92.1 (±2.4) 82.5 (±2.2) 87.3 (±2.4)

M molding

Table 4 Mean value (according to the initial severity) of right and leftoccipital heteromorphic rates and Posterior symmetry Ratios pre and posthelmet therapy by starting age

Occipital heteromorphic rates

Mild group Severe group

Age(month) Pre-M (SD) Post-M (SD) Pre-M (SD) Post-M (SD)

4 87.5 (±5.0) 96.7 (±3.3) 73.6 (±5.0) 91.3 (±2.7)

5 87.0 (±5.1) 96.6 (±3.4) 73.4 (±4.3) 89.2 (±2.5)

6 87.5 (±5.0) 95.7 (±4.3) 75.0 (±4.3) 89.6 (±3.1)

7 87.2 (±6.1) 94.7 (±4.7) 73.9 (±4.2) 85.5 (±3.5)

8 86.9 (±4.8) 93.8 (±5.1) 74.7 (±4.7) 85.3 (±3.8)

9∼ 86.4 (±5.0) 91.9 (±5.2) 73.8 (±4.1) 82.6 (±3.9)

M molding

Childs Nerv Syst

8 months of age (4–7, 4–8, 4–older than 8—MSe=28.63,p<0.01 for each), symmetry ratio of 5-month-olds was alsosignificantly higher than that of 7, 8, and older than 8 monthsof age (5–7, 5–8, 5–older than 8—MSe=28.63, p<0.01 foreach), and symmetry ratio of 6, 7, and 8-month-olds were alsosignificantly higher than that of older than 8 months (6–olderthan 8, 7–older than 8, 8–older than 8—MSe=28.63,p<0.01). Again, the meaning of these results is that the earlierthe treatment started, the higher the symmetry ratio rises inboth deformity groups, but the effect of helmet treatment fell

markedly in older than 8 months of the mild deformity group(Fig. 6d).

Discussion

“Positional” plagiocephaly, among the other PHDs, is themost common condition and is defined by specific pathophys-iology and biomechanical factors and not determined only by

Fig. 6 (a–d). a In the severe deformity group, the overall deformity ratioimproved with a statistically significant difference comparing before andafter therapy. Mean values after treatment indicated strong correctiveeffect for children with plagiocephaly, being the highest from multiplecomparisons after the treatment for those started at 4 months of age. b Inplagiocephaly cases, the overall symmetry ratio improved with a statisti-cally significant difference comparing before with after therapy in themild deformity group. From the multiple comparisons after treatment, itwas shown that the earlier the age of starting therapy, the higher the meanof symmetry ratio; however, maximal correction effect was obtained iftreatment started by the 6 months after birth. c Posterior symmetry ratio

improved with a statistically significant difference comparing before andafter therapy for the severe deformity plagiocephaly group. If moldinghelmet therapy could be started before 6 months of age, its efficacy wassignificantly higher. d Posterior symmetry ratio improved with a statisti-cally significant difference comparing before and after therapy in the milddeformity plagiocephaly group. Multiple comparison analysis after thetreatment indicated that the overall symmetry ratio after the helmettherapy was higher in the group where the start of therapy was early. Itwas shown in particular that the correction effect significantly loweredwhen treatment started after the age of 9 months

Childs Nerv Syst

deformity description as a result of prolonged supine headposition. It has specific management related to its pathophys-iology and biomechanics, and its incidence demands the es-tablishment of principles and goals of treatment. As the childstarts free head movement in the early months of life, partic-ularly after 4 months of age (when able to roll), the applicationof “molding” forces on the cranium gradually becomes im-possible, and treatment might become difficult [7], [8]. There-fore, an oversimplified concept of being a temporary cosmeticeffect on the head shape because of prolonged head position-ing is not acceptable. Analysis of the factors at the backgroundof this condition and those related to the effective treatmentcan bring proper solution to this problem. [9]

Predisposition to positional plagiocephaly

Many factors—both prenatal and postnatal—have been relat-ed to the development of positional plagiocephaly. It is in-creased in firstborns, twins, and other multiples, even whencesarean delivery is performed. This was demonstrated to bethe more frequent scaphocephalic head shape in these infants.That is also more often seen in breech positions and prolongedvaginal deliveries, exerting overload to the otherwise naturalhead deformability in newborn. If a difficult labor, particularlyafter forceps or vacuum extraction delivery, leads to birthtrauma, the resulting motor deficit can create conditions forpoor or restricted head movements and increase the positionalinconveniences. Premature born infants have also higher de-formity rates for the same reason.

In general, the supine sleeping position exacerbates defor-mation process; however, positional plagiocephaly is only onecomponent of this deformation process. Often, other medicalfactors accompany this deformity. Factors accompanying di-rect deformation include facial deformation, mandibularasymmetry, congenital and/or acquired muscular torticollis,abnormal eye placement, external ear deformity and misalign-ment, orbital asymmetry resulting in strabismus and otherocular problems, and epicanthal fold on the side of flatness.[10] The indirect factors (often diagnosed in infants withpositional plagiocephaly) are positional foot deformities, de-velopmental hip dysplasia, middle ear infections, and mi-graine headaches [5, 11].

Cosmetic vs. reconstructive procedures

In plagiocephaly, the deformation of one element leads tocompensatory deformation and displacement of all other con-nected elements of the system, and as a result, other cranialasymmetries are commonly identified as abnormal cranialheight, abnormal cranial width/breadth, and occipital flatten-ing with ipsilateral forehead bossing, significant ear misalign-ment, abnormal alignment and asymmetry of the orbits,among others. This compensation for the head’s abnormal

orientation in space results in ocular and vestibular impair-ment and distortion of the orbits with pressure on theextraocular muscles and nerves, resulting in sensorimotordisturbances [11].

A cranial remolding orthosis has a direct effect on frontal,parietal, sphenoid, temporal, and one of the occipital bones ofthe neurocranium. Indirectly, it affects the entire facial align-ment (i.e., viscerocranium) due to the direct transfer of forcesthrough the neurocranial structures. Throughout the orthotictreatment program, measurable changes in the cranial base,cranial vault, orbitotragial depth, and cephalic index havebeen documented. By returning the cranial and facial bonesto a normal alignment, long-term dysfunction to hearing,vision, and mandibular mechanics could likely be avoided[10] together with other medical conditions and developmentdifficulties in cognition.

Criteria for management

The presence of anthropometric data was verified for moder-ate to severe plagiocephaly [7]. Several studies have alreadypointed out on the maximum efficacy of orthotic treatmentduring the window of rapid head growth and its proportionaldecrease with the increase of cranial rigidity concomitant withage. The pace of re-formation relates to the rate of braingrowth, which is much more rapid during the first 6 monthsthan later in infancy. In a prospective study of 114 infants,those managed with helmet (51 infants) did significantlybetter than those with only head positioning in the crib (63infants) Therefore, timing, recognition/grading, and earlytreatment were the main criteria for success. Our results indi-cate that the same tendencies of effective “time window”combined with importance of proper attitude of the medicalcommunity exist with Japanese infants. Clarren et al. docu-mented the safety and efficacy of cranial remolding orthosisfor positional plagiocephaly. Rekate et al. reported that cranialremolding orthosis should be prescribed for all infants withpositional plagiocephaly when asymmetry persists after repo-sitioning attempts and also prior to the consideration of surgi-cal intervention for infants less than 12 months of age [12].Littlefield et al. reported that significant correction of cranialdeformation was achieved through the use of cranialremolding orthosis and was maintained after the discontinua-tion of the orthotic treatment program. Kelly et al. also iden-tified the need for early intervention. Statistically significantincreases in cranial growth were associated with concomitantreductions of the cranial asymmetries in deformationalplagiocephaly. Joganic et al. documented the use of cranialremolding orthosis to facilitate post-surgical outcomes. Gra-ham found that early intervention relates to both the lengthand success of the orthotic treatment program. All thesestudies are very similar to our findings, indicating that Asianinfants respond in a similar way to the molding helmet

Childs Nerv Syst

therapy. But definite Asian infants’ skull bone deformity datahas not been available until now [5].

Developmental positional plagiocephaly

A common concept on positional plagiocephaly is that theinfant’s head “will round out on its own as the child becomesmore active, begins to roll over, and learns to sit up.” Thismisconception is based in part on outdated scales of motordevelopment and a lack of understanding on the effect ofsupine sleep positioning. The pattern of early motor develop-ment is affected by sleep position. On average, supine sleepersattain common motor milestones later than prone sleepers[13]. Prior to 1992, infants’ heads often corrected in the firstfew months of life because infants that were placed prone tosleep were generally in a variety of positions during the day,thus avoiding prolonged time in one position. Now that supineis the position of choice and there is a 4–6-week delay in theacquisition of head and trunk control, infants’ heads often donot “round out” as they did previously. As noted previously,torticollis may be the cause or effect of positionalplagiocephaly. Binder et al. (1987) studied the long-termeffects of torticollis and found significant “persistent function-al asymmetry of the involved body side despite mild ormoderate severity, early diagnosis, and complete resolutionof the torticollis,” leading to long-term complications [14].

Treatment regimens

Optimal beginning for positional helmet therapy was reportedto be at the age of 5 to 6 months in the European countries [1].The evidence obtained from Japanese infant clinical datashows clearly decreased efficacy of molding helmet therapyjust after the age of 6 months [15]. The weight of the helmet,particularly that made in the USA, created a problem with thestill unstable neck at an earlier age and could not be applied inour studies. A lighter helmet could bring down the treatmentinitiation age. [16] A further improvement might be to designdifferent helmets, proportional to the physical characteristicsof Caucasian, African American, Asian, and so on infants.The different groupsmay have also different starting treatmentage.

Current situation in Japan and the international community

Currently, there is no study estimating the prevalence ofpositional skull deformity in infants in Japan as comparedwith American and European countries. [17, 18] At the sametime, the interest of the Japanese medical community onpositional plagiocephaly, particularly the pediatricians on gen-eral practice, has been very low and not based on any guide-lines, leaving the problemwithout established criteria betweenpediatric plastic surgeons and pediatric neurosurgeons [9].

That often leads to the recommendation of only observationon a huge number of infants with deformities. As a result, wecurrently do not have any data on patients suitable for com-parison or control group.

van Wijk et al. reported a recent randomized trial on helmettherapy which showed that no effect of helmet therapy can beshown in infants with moderate to severe positional skulldeformation [19]. Regarding our cases, even though skulldeformation was moderate to severe, the earlier the treatmentwas started, the higher symmetry ratio recovery was obtained.Treatment was especially effective when started in 4-month-oldinfants. In contrast to the “severe” group, the “mild” deforma-tion group showed that MHT was most effective if treatmentstarted before 6 months of age. Again, the earlier the treatmentwas started, the higher symmetry ratio was achieved, butcompared to the “severe” group, it had a modest effect whentreatment was started in infants older than 8 months.

In our study, treatment was started since the initial diagno-sis of each case as soon as possible and in a significant portionof patients before the age of 5 months. As a result, the time ofconsultation (diagnosis establishment) and start of therapywere correlating.

The difference of results between our study and theHEADS could be related to the younger infant age at diagno-sis. For those reasons, earlier diagnosis and start of treatmentwas related to different quantitative expression of the defor-mity because of the early age, a major degree of parents’apprehension and disapproval regarding their child’s headshape and all that leading to early entry into the study.

Last, but not least, the patients population is different, asconsidering Asian patients, compared to European ones.Some cultural influences, as head shape perception and par-ents’ considerations of importance if minor deformities con-tinue to exist, might have influenced results. Another point ofdifference with the HEADS study is the rate of complicationsof the helmet group, which are in very significant proportions.Being a “pragmatic study,” that factor might have influencedthe compliance, which has not been recorded in details, some-thing that a future study should address.

On correcting the cranial deformity of infants, pediatriciansshould be aware of the critical time period within which theinitiation of orthotic treatment is able to correct effectively thecranial deformity as our data indicate. In a case of severe skulldeformity when we have to undertake immediately correctiveorthotic management, consenting with the parents can beextremely important. In addition, awareness of positionaldeformities and their management should be extended to theobstetricians, pediatricians, midwives, and nursing staff [18].

Mental distress

Skull deformities are well known for inducing inferioritycomplex in childhood [20]. The parents of more than 80 %

Childs Nerv Syst

of our treated skull deformity infants had some degree ofinferiority complex because of skull deformity from whichboth their fathers and mothers had difficulties in wearingglasses and auricular asymmetries, hair style problems,temporo-mandibular joint asymmetries, and teeth alignmentproblems, even occasionally have been unable to wear motor-bike helmets. Their own experience was a motivating factorfor the management of their child.

Future grading system

While similar reports exist for American and European infantpopulations, this is the first one able to demonstrate efficacy ofthe helmet orthotic treatment for Japanese infantile skull de-formity. However, some problems in the evaluation method-ology remain. Plagiocephaly can be diagnosed and gradedbased on the existing asymmetry or as we have applied thesymmetry ratios. Brachycephaly diagnosis and grading how-ever remain more difficult as different criteria using the ratioof occipito-frontal length and skull breadth or the overallsymmetry ratio averaging anterior and posterior quadrant withthe purpose of comparison have been applied. The existingdifferences in measuring methodology might have influencedinterpretations among different institutions [21–23]. The ac-cumulation of Japanese infant skull data form a database usinga single standard method and creating reference values be-came an urgent task [2, 24]. While at the moment only height,weight, and head circumference are routinely recorded inJapan for infant development evaluation, considering Ameri-can and European sources, we are intending to promote thatthe index of the Japanese infantile skull deformity rate beroutinely included in the future [25, 26].

Conclusion

We introduced a Japanese infant positional skull deformityevaluation system and described the efficacy of helmet thera-py for its treatment, discussing the critical periods for inter-vention. Its application showed that similar to European andAmerican PHD infants’ management, it is an effective treat-ment according to specifically established criteria. Analysis ofresults showed the optimal time and deformity criteria valuesfor obtaining adequate treatment. The application of this ap-proach traces the possibility of a wider application of defor-mity evaluation and management for the whole Japaneseinfant population. In conjunction with data from other geo-graphical regions, it may provide more precise guidelines forcountries with existing racial diversity in their infantpopulations.

Acknowledgments We thank Dr. Kostadin Karagiozov for his adviceand manuscript review and David Huang for his guidance, and wegratefully acknowledge the radiological technologists, nurses, and staffof the Departments of Neurosurgery, Tokyo Women’s Medical Universi-ty, in preparing this paper.

Disclose financial relationships for all authors The authors have nofinancial relationships relevant to this article to disclose.

Conflicts of interest The authors have no conflicts of interest relevantto this article to disclose.

Funding None.

Contributors’ statement page Yasuo Aihara—Dr. Aihara conceptual-ized and designed the study and drafted the initial manuscript.

Kana Komatsu Osami Kubo, Tomokatsu Hori—Ms. Komatsu and Dr.Kubo and Dr. Hori carried out the initial analyses and reviewed themanuscript.

Hitoshi Dairoku Yoshikazu Okada—Mr Dairoku designed the datacollection instruments and coordinated and supervised data collection. Dr.Okada critically reviewed the manuscript.

All authors approved the final manuscript as submitted and agree to beaccountable for all aspects of the work.

References

1. Kluba S, Kraut W, Reinert S, Krimmel M (2011) What is the optimaltime to start helmet therapy in positional plagiocephaly? PlastReconstr Surg 128:492–498

2. Lipira AB, Gordon S, Darvann TA, Hermann NV, Van Pelt AE,Naidoo SD, Govier D, Kane AA (2010) Helmet versus active repo-sitioning for plagiocephaly: a three-dimensional analysis. Pediatrics126:e936–e945

3. Biggs WS (2003) Diagnosis and management of positional headdeformity. Am Fam Physician 67:1953–1956

4. Mortenson P, Steinbok P, Smith D (2012) Deformationalplagiocephaly and orthotic treatment: indications and limitations.Childs Nerv Syst 28:1407–1412

5. Yoo HS, Rah DK, Kim YO (2012) Outcome analysis of cranialmolding therapy in nonsynostotic plagiocephaly. Arch Plast Surg39:338–344

6. Plank LH, Giavedoni B, Lombardo JR, Geil MD, Reisner A (2006)Comparison of infant head shape changes in deformationalplagiocephaly following treatment with a cranial remolding orthosisusing a noninvasive laser shape digitizer. J Craniofac Surg 17:1084–1091

7. Wilbrand JF, Seidl M, Wilbrand M, Streckbein P, Bottger S, Pons-Kuehnemann J, Hahn A, Howaldt HP (2013) A prospective random-ized trial on preventative methods for positional head deformity:physiotherapy versus a positioning pillow. J Pediatr 162:1216–1221, 1221 e1211

8. Paquereau J (2013) Non-surgical management of posterior positionalplagiocephaly: orthotics versus repositioning. Ann Phys RehabilMed 56:231–249

9. Pogliani L, Mameli C, Fabiano V, Zuccotti GV (2011) Positionalplagiocephaly: what the pediatrician needs to know. A review. ChildsNerv Syst 27:1867–1876

10. Kluba S, Schreiber R, Kraut W, Meisner C, Reinert S, Krimmel M(2012) Does helmet therapy influence the ear shift in positionalplagiocephaly? J Craniofac Surg 23:1301–1305

Childs Nerv Syst

11. Shamji MF, Fric-Shamji EC, Merchant P, Vassilyadi M (2012)Cosmetic and cognitive outcomes of positional plagiocephaly treat-ment. Clin Investig Med 35:E266

12. Rekate HL (1998) Occipital plagiocephaly: a critical review of theliterature. J Neurosurg 89:24–30

13. Davis BE, Moon RY, Sachs HC, Ottolini MC (1998) Effects ofsleep position on infant motor development. Pediatrics 102:1135–1140

14. Binder H, Eng GD, Gaiser JF, Koch B (1987) Congenital musculartorticollis: results of conservative management with long-termfollow-up in 85 cases. Arch Phys Med Rehabil 68:222–225

15. Seruya M, Oh AK, Taylor JH, Sauerhammer TM, Rogers GF (2013)Helmet treatment of deformational plagiocephaly: the relationshipbetween age at initiation and rate of correction. Plast Reconstr Surg131:55e–61e

16. Wilbrand JF, Wilbrand M, Malik CY, Howaldt HP, Streckbein P,Schaaf H, Kerkmann H (2012) Complications in helmet therapy. JCraniomaxillofac Surg 40:341–346

17. Ifflaender S, Rudiger M, Konstantelos D, Wahls K, Burkhardt W(2013) Prevalence of head deformities in preterm infants at termequivalent age. Early Hum Dev 89:1041–1047

18. Mawji A, Vollman AR, Hatfield J, McNeil DA, Sauve R (2013) Theincidence of positional plagiocephaly: a cohort study. Pediatrics 132:298–304

19. van Wijk RM, van Vlimmeren LA, Groothuis-Oudshoorn CG, Vander Ploeg CP, Ijzerman MJ, Boere-Boonekamp MM (2014) Helmettherapy in infants with positional skull deformation: randomisedcontrolled trial. BMJ 348:g2741

20. Roby BB, Finkelstein M, Tibesar RJ, Sidman JD (2012) Prevalenceof positional plagiocephaly in teens born after the “Back to Sleep”campaign. Otolaryngol Head Neck Surg 146:823–828

21. Meyer-Marcotty P, BohmH,LinzC,Kunz F,KeilN, Stellzig-EisenhauerA, Schweitzer T (2012) Head orthesis therapy in infants with unilateralpositional plagiocephaly: an interdisciplinary approach to broadening therange of orthodontic treatment. J Orofac Orthop 73:151–165

22. Schweitzer T, Bohm H, Linz C, Jager B, Gerstl L, Kunz F, Stellzig-Eisenhauer A, Ernestus RI, Krauss J, Meyer-Marcotty P (2013)Three-dimensional analysis of positional plagiocephaly before andafter molding helmet therapy in comparison to normal head growth.Childs Nerv Syst

23. Schaaf H, Wilbrand JF, Boedeker RH, Howaldt HP (2010) Accuracyof photographic assessment compared with standard anthropometricmeasurements in nonsynostotic cranial deformities. Cleft PalateCraniofac J 47:447–453

24. Schaaf H, Malik CY, Streckbein P, Pons-Kuehnemann J, HowaldtHP, Wilbrand JF (2010) Three-dimensional photographic analysis ofoutcome after helmet treatment of a nonsynostotic cranial deformity.J Craniofac Surg 21:1677–1682

25. Wilbrand JF, Wilbrand M, Pons-Kuehnemann J, Blecher JC,Christophis P, Howaldt HP, Schaaf H (2011) Value and reliabilityof anthropometric measurements of cranial deformity in early child-hood. J Craniomaxillofac Surg 39:24–29

26. Wilbrand JF, Schmidtberg K, Bierther U, Streckbein P, Pons-Kuehnemann J, Christophis P, Hahn A, Schaaf H, Howaldt HP(2012) Clinical classification of infant nonsynostotic cranial defor-mity. J Pediatr 161:1120–1125

Childs Nerv Syst

Effectiveness of Cranial Remolding Orthoses in Reducing Asymmetry inPatients with Plagiocephaly in Relation to Start Age and Severity

Tiffany Graham CPO, LPO, MSPO, Terran Gregory MPO, Jichele Harris MPO, Mary Walsh MPOUT Southwestern Prosthetics-Orthotics Program

This study focuses on how well CranialRemolding Orthosis (CRO) treatment reducescranial asymmetry in infants with deformationalplagiocephaly based on initial severity and startage of treatment. The researchers propose thatyounger start ages will show greater rates ofcorrection and more severe cases will requirelonger treatment times to obtain correction.

Subjects: Out of 1177 charts reviewed, 218patients with a diagnosis of nonsynostoticplagiocephaly met the inclusion criteria of thestudy.

Procedures: Researchers collected retrospectivedata from Level-4 P&O’s Texas clinics using anexisting patient database.

Data Analysis: Unbalanced two-way ANOVAsand Excel analysis were utilized.

Figure 1. Average CVAI improvement based onstarting age and severity group.

Results suggest orthotists should start CROtreatment as early as possible, particularly beforenine months of age for optimum results.Additionally, very severe head shapes should betreated for at least six months. The presence oftorticollis or prematurity does not significantlyaffect treatment outcomes.

CHOA Scale Hanger Clinic.Federal Register. Federal Register. Vol 63, Num 146, pp. 40650 -

40652. 1998Fish, D., & Lima, D. JPO, Vol 15, Num 2, pp. 37-47. 2003Freudlsperger, C., et al. Journal of Cranio-Max Surg., 44(2), 110-

115. 2016Grigsby, K. JPO, Vol. 21, Num. 1, pp. 55-63. 2009Kluba, S., et al. P&RS, 128(2), 492-498. 2011Lima, D. &. Journal of Proceedings. 2002Seruya, M., et al. P&RS, 131(1). 2013Steinberg, J., et al. P&RS 135, 833-842. 2015van Wijk, Renske M., et al. BMJ 348, g2741. 2014March 2017

HYPOTHESIS

METHODS

RESULTS

DISCUSSION

CONCLUSION

REFERENCES

The results of this study support the hypothesisand the findings of previous studies that cranialremolding orthoses are effective treatment ofnonsynostotic plagiocephaly. Furthermore, thedata suggests that subjects that started treatmentbefore nine months of age obtained significantlymore correction than those that started at ninemonths of age or later. This should transfer toclinical practice in the collaboration of insuranceproviders, the referring physician, and theorthotist to start treatment before nine months ofage for the greatest correction. . Data alsosuggests that more severe deformation requireslonger treatment time to obtain a fully correctedhead shape, and severe cases obtain significantly

Of the 218 included patients in the preliminaryanalysis, 104 patients were in the earlyintervention group, 88 in the middle interventiongroup, and 26 in the late intervention group.Severity classifications based on the CHOAscale yielded the following distribution: 67 mild,

Average Treatment Times (months)Mild Moderate Severe Very Severe

3.57m 3.79m 4.23m 5.11m

Table 1. Differences in average treatment timesamong the four severity groups.

0

2

4

6

8

10

12

14

CVAI

Improvement by CRO Treatment

Starting Age vs Severity CRO Correction

Start Age 2-5 Months Start age 6-8 months Start Age 9-17 months

76 moderate, 47 severe, and 28 very severecases. Overall treatment times were groupedinto 102 short term, 99 average, and 17extended treatment times.

Unbalanced ANOVAs comparing the datarevealed that children with mild head shapesand minimal time in the CRO had similarpositive outcomes as those with mild shapeswho spent extended time in the CRO. Furtheranalysis revealed that time plays a significantfactor in the overall treatment based onpresenting severity of the infant (p=0.018).There was not a significant difference foundwhen comparing severity and the age group ofeach participant, indicating that presentingseverity was independent from age of initiationof treatment in the data set. The analysiscomparing starting severity to the presence oftorticollis and prematurity gave no indicationthat either of the latter factors were related tothe overall severity of the patient at the start oftreatment.

significantly more correction if treated longerthan six months.

This adds to the literature that supports treatmentand weakens the few studies that have been usedby insurance companies to deny coverage. Basedon the positive treatment results in this study andmany others, CRO treatment is an effective wayto treat deformational plagocephaly, regardless ofstarting severity. Data trends suggest youngerinfants have shorter CRO treatment times andbetter correction outcomes.

CLINICAL ARTICLEJ Neurosurg Pediatr 19:684–689, 2017

Sagittal strip craniectomy is an established and ef-fective treatment for sagittal synostosis. Unlike early strip craniectomy procedures,8,9 the techniques em-

ployed today utilize a variety of modifications to augment the changes in head shape achieved after surgery. These modifications include minimally invasive approaches, extending the width of the strip craniectomy, adding wedge ostectomies or barrel staves, and/or inserting cra-nial springs to facilitate lateral movement of the parietal bones.1,4–6, 12–15,17 The most common postoperative adjunct

is the wearing of a helmet to harness the rapid growth of the brain to augment the “passive” changes achieved in the shape of the cranial vault.3,5,6,10

Some groups have reported 3D analysis of cranial mor-phology and changes after open cranial vault treatment of sagittal synostosis;7,11,16 however, the outcome of sagittal strip procedures has primarily been reported utilizing the cephalic index (CI).1,3,6,10,13,15 We retrospectively reviewed the outcomes of extended sagittal strip craniectomy to-gether with wedge ostectomies and postoperative helmet

ABBREVIATIONS CI = cephalic index.SUBMITTED November 28, 2016. ACCEPTED January 30, 2017.INCLUDE WHEN CITING Published online March 31, 2017; DOI: 10.3171/2017.1.PEDS16660.

Three-dimensional changes in head shape after extended sagittal strip craniectomy with wedge ostectomies and helmet therapyPang-Yun Chou, MD,1,2 Rami R. Hallac, PhD,1,3 Shitel Patel, MD,1 Min-Jeong Cho, MD,1 Neil Stewart, MS,1 James M. Smartt, MD,1 James R. Seaward, MD,1 Alex A. Kane, MD,1,3 and Christopher A. Derderian, MD1

1Department of Plastic Surgery, UT Southwestern; 3Analytical Imaging and Modeling Center, Children’s Medical Center, Dallas, Texas; and 2Department of Plastic and Reconstructive Surgery, Chang Gung Memorial Hospital, Taoyuan, Taiwan

OBJECTIVE Outcome studies for sagittal strip craniectomy have largely relied on the 2D measure of the cephalic in-dex (CI) as the primary indicator of head shape. The goal of this study was to measure the 2D and 3D changes in head shape that occur after sagittal strip craniectomy and postoperative helmet therapy.METHODS The authors performed a retrospective review of patients treated with sagittal strip craniectomy at their insti-tution between January 2012 and October 2015. Inclusion criteria were as follows: 1) isolated sagittal synostosis; 2) age at surgery < 200 days; and 3) helmet management by a single orthotist. The CI was calculated from 3D images. Color maps and dot maps were generated from 3D images to demonstrate the regional differences in the magnitude of change in head shape over time.RESULTS Twenty-one patients met the study inclusion criteria. The mean CI was 71.9 (range 63.0–77.9) preoperatively and 81.1 (range 73.0–89.8) at the end of treatment. The mean time to stabilization of the CI after surgery was 57.2 ± 32.7 days. The mean maximum distances between the surfaces of the preoperative and 1-week postoperative and between the surfaces of the preoperative and end-of-treatment 3D images were 13.0 ± 4.1 mm and 24.71 ± 6.83 mm, respective-ly. The zone of maximum change was distributed equally in the transverse and vertical dimensions of the posterior vault.CONCLUSIONS The CI normalizes rapidly after sagittal strip craniectomy (57.2 days), with equal distribution of the change in CI occurring before and during helmet therapy. Three-dimensional analysis revealed significant vertical and transverse expansion of the posterior cranial vault. Further studies are needed to assess the 3D changes that occur after other sagittal strip craniectomy techniques. https://thejns.org/doi/abs/10.3171/2017.1.PEDS16660KEY WORDS sagittal craniosynostosis; surgical outcome; molding helmet; wedge ostectomies; 3D imaging; strip craniectomy; craniofacial

©AANS, 2017J Neurosurg Pediatr Volume 19 • June 2017684

Postoperative head shape change in sagittal craniosynostosis

J Neurosurg Pediatr Volume 19 • June 2017 685

therapy by using 3D analysis to assess the distribution and speed of the change in head shape that occurs after sur-gery.

MethodsThe UT Southwestern Medical Center and Children’s

Medical Center institutional review boards approved this retrospective study of all patients who had been treated with extended sagittal strip craniectomy at the Children’s Medical Center in Dallas, Texas, between January 2012 and October 2015. Inclusion criteria were as follows: a diagnosis of nonsyndromic, isolated sagittal synosto-sis; treatment with sagittal strip craniectomy in patients younger than 200 days, age-corrected for preterm deliv-ery; postoperative helmet management by a single or-thotist; and completion of the prescribed course of hel-met therapy until the age of at least 12 months. Collected demographic data included patient age at surgery, race, and sex. The surgical procedure for extended strip crani-ectomy with wedge ostectomies was modified from the original description by Jimenez et al.6 The technique is performed using two 3- to 4-cm incisions placed at the an-terior and posterior limits of the synostotic sagittal suture. As described by others,15,17 we find that an endoscope is not necessary and we perform our procedure under direct visualization using lighted retractors and a head light. A 5-cm-wide strip craniectomy is centered on and incor-porates the closed sagittal suture. One-centimeter-wide wedge ostectomies are performed under direct visualiza-tion just posterior to the coronal sutures and anterior to the lambdoid sutures, all directed toward the squamosal suture. The anterior wedge ostectomies extend up to but not beyond the temporalis muscle. The posterior wedge ostectomies extend up to or close to the squamosal suture.

Three-dimensional images were captured throughout treatment by using a laser scanner (STARscanner, Ortho-merica). The mean time before surgery for the preopera-tive scan was 23 days (range 1–71 days). The 1-week post-operative scan was captured for helmet production. The end-of-treatment scan was captured at the last orthotist visit. Patient age at the end of helmet therapy ranged be-tween 12 and 18 months depending on the provider.

The CI was calculated from measurements obtained from 3D images captured at all time points. A paired t-test was performed to compare the preoperative and end-of-treatment cranial indices. A p value < 0.05 was significant. MATLAB (MathWorks Inc.) was used to curve fit expo-nential regressions from time point measurements to iden-tify the time point of head shape stability. This time point was identified when the curve hit 99.99% of the maximum change. The mean of all patients’ time points to a stable CI was used as the time of head shape stability. End-of-treat-ment results with regard to CI were classified as excellent (CI > 80), good (CI 75–80), or poor (CI < 75).

Head scans at 3 different time points were used for 3D analysis: preoperatively, prehelmet (1 week postsurgery), and 1 year postoperatively (end of treatment). To generate a composite image from all subjects at each time point, a 3D template with known polygonal mesh connectiv-ity was created from a 3D scan from a healthy subject.

For each 3D image, registration was performed through rigid translation and rotation to match the template us-ing 25 landmarks placed at recognizable anatomical loca-tions. In addition, anisotropic scaling of the template was performed to match each individual 3D image. Then, the template was deformed to each 3D scan using a thin-plate spline algorithm and closest point deformation. In this way, the polygonal indexing of all the 3D images matches that of the template, allowing for the creation of a mean (composite) head from the deformed templates by point-wise averaging. Thus, a composite head was created by combining the 3D deformed templates of all patients at each time point.

Color maps were generated to show the distance be-tween the surfaces of the composite 3D images to show the 3D changes of the head and their magnitude over time. Color maps were designed to show increased distance in warm colors and decreased distance in cool colors. Analy-sis was performed using proprietary programs written in MATLAB.

To determine the location of maximal change for each subject at the prehelmet and end-of-treatment time points, a dot map in blue and red was created on the composite image.

ResultsTwenty-one patients met the study inclusion criteria.

Most of the patients were male (71%). Fifty-seven per-cent of the patients were white, 38% were Hispanic, and 5% were African American. The mean age at surgery was 113.6 ± 27.9 days (mean ± standard deviation, range 83–202 days). The mean CI before surgery was 71.9 ± 3.8 (range 63.0–77.9) and at the end of treatment was 81.1 ± 4.1 (range 73.0–89.8; Fig. 1). The average duration of post-operative helmet therapy was 327.5 ± 27.9 days (range 132–514 days).

To determine the change in CI from the surgical pro-cedure alone (prehelmet) and the change in CI during hel-met therapy, we measured the CI on 3D images obtained preoperatively, 1 week postoperatively (prehelmet), and at subsequent visits during helmet therapy. Figure 1 shows that the mean change in the CI at the end of treatment was 9.2% (range 1.5%–18.3%, p < 0.05). The mean change in the CI from the preoperative image to the 1-week postop-erative image (prehelmet scan) was 4.6. The mean change in the CI during helmet therapy was also 4.6. The average time from surgery to stabilization of the CI was 57.2 ± 32.7 days (Fig. 2). Representative before and after photo-graphs of a patient with a starting CI of 71 and an end-of-treatment CI of 80 are featured in Fig. 3. Treatment out-comes were categorized based on the CI for each patient at the end of treatment. Sixty-two percent of the patients achieved excellent results (CI > 80), 33% good results (CI 75–80), and 5% (1 patient) poor results (CI < 75).

To further assess global changes in head shape after surgery, composite 3D images were generated for the pre-operative, prehelmet, and posthelmet time points. Figure 4 depicts overlays of the preoperative, prehelmet, and end-of-treatment composite images to demonstrate the change in head shape occurring from the preoperative to the

P. Y. Chou et al.

J Neurosurg Pediatr Volume 19 • June 2017686

1-week postoperative (prehelmet) to the end-of-treatment time points. Color maps were generated to visually quan-tify the distance between the surfaces of the preoperative and end-of-treatment composite 3D images (Fig. 5).

The average maximum distance between the surfaces of the preoperative and 1-week postoperative 3D images was 13.0 ± 4.1 mm (range 8.1–25.2 mm). The average maximum distance between the surfaces of the preopera-tive and end-of-treatment 3D images was 24.7 ± 6.8 mm (range 3.5–33.7 mm). Figure 6 shows a dot plot of the loca-tion of the maximum change in head shape for each sub-ject between the preoperative 3D image and the 1-week postoperative 3D image (blue) and between the preopera-tive 3D image and the end-of-treatment 3D image (red).

DiscussionThe purpose of this study was to analyze the change

in head shape occurring after extended sagittal strip cra-niectomy with wedge ostectomies and postoperative hel-met therapy using CI and 3D analysis. Cephalic index was used as a benchmark because of its long-standing use in previous studies for grading outcomes of treatment for sagittal synostosis. Three-dimensional analysis was em-ployed to assess global changes in head shape occurring after surgery.

We found that CI changes rapidly after the surgical procedure, with 50% of the mean total change in CI (4.6) occurring by 7 days after surgery, before helmet therapy is initiated (Fig. 1). The mean time to stabilization of the CI was 57.2 days after surgery (Fig. 2). We found that 95% of the patients achieved a CI ≥ 75. Figure 3 features repre-sentative photographs of a patient before and after surgery with a change in CI equal to the mean CI change of 9.2.

Variability in the ratio of change in the CI was ob-served between the immediate postoperative period (be-fore helmet therapy) and during helmet therapy (Fig. 2 upper). The possible reasons for this difference may in-

clude minor variations in the surgical technique between surgeons, variability in preoperative head shape, and/or variability in helmet design and management. Our study

FIG. 1. Cranial index change for each subject. The starting CI is shown in blue, the change in the CI between surgery and the start of postoperative helmet therapy appears in green, and the change in the CI while wearing the helmet appears in orange. The mean changes are displayed on the far right of the figure. PT = patient. Figure is available in color online only.

FIG. 2. Upper: The change in the CI over time for each subject. Lower: The mean change in the CI over time, with stabilization of the CI at 57 days after surgery. Figure is available in color online only.

Postoperative head shape change in sagittal craniosynostosis

J Neurosurg Pediatr Volume 19 • June 2017 687

is not sufficiently powered to determine the significance of these individual variables with regard to their impact on the speed of change in CI. Statistical analysis did not show any correlation between the severity of preoperative CI and the rate or magnitude of change in CI.

The mean CI stabilized at 57.2 days after surgery (Fig. 2 lower), results consistent with those in a previous report by Jimenez et al.6 This time point may reflect a homeo-stasis being achieved between the opposing forces of the scalp and remodeling bone and the growing brain. This rapid change and stabilization in CI suggests that a shorter period of helmet therapy should be considered.

The duration of helmet therapy varies significantly among institutions. Jimenez et al. prescribe helmet thera-py until 18 months of age and the use of 3 helmets during this period.6 Proctor prescribes the helmet for an aver-age of 7–8 months and typically uses a single helmet.13 We currently prescribe helmet therapy until an age of 12–15 months old, utilizing 2–3 helmets, believing that sufficient growth is complete by this age to retain the cor-rected head shape. A randomized prospective study is warranted to explore whether shortening the duration of helmet therapy negatively affects head shape outcomes. Incrementally decreasing the duration of helmet thera-py in serial cohorts would allow a threshold of change in head shape outcomes to be detected while minimizing the number of patients with compromised outcomes from surgery.

Cephalic index is a useful barometer for grading the severity of scaphocephaly in sagittal synostosis; howev-er, it provides little objective data about head shape and whether that shape is perceived as normal. Dvoracek et al.

FIG. 3. Preoperative (upper) and end-of-helmet-therapy (lower) images obtained in a patient treated with extended sagittal strip craniectomy with wedge ostectomies. The patient’s preoperative CI was 0.71 and postoperative CI was 80. Figure is available in color online only.

FIG. 4. Composite images were created for the preoperative (red), 1-week postoperative (green), and end-of-treatment (blue) time points. Note the significant increase in the vertical dimension of the posterior vault in both the immediate and long-term postoperative periods, with normalization of the vertex position. Figure is available in color online only.

P. Y. Chou et al.

J Neurosurg Pediatr Volume 19 • June 2017688

recently reported that improved CIs were achieved after sagittal strip craniectomy, but that the maximal transverse dimension of the skull (euryon) remained in an abnormal location after surgery.2 Thus, CI can normalize while the shape of the cranial vault remains abnormal. We sought to further define changes in head shape in our patients after sagittal strip craniectomy using 3D imaging techniques.

Composite images were generated from the 3D images of the 21 subjects at the preoperative, prehelmet (1-week postoperative), and end-of-treatment time points. Figure 4 shows the overlay of the preoperative, prehelmet, and end-of-treatment composite images, demonstrating the signifi-cant change in head shape that occurs during treatment. Over time, the vertex assumes a more normal posterior position due to significant vertical expansion and remodel-ing of the posterior cranial vault.

The color maps in Fig. 5 show the areas of greatest change between the preoperative and end-of-treatment images in red. There is a relatively even distribution in the magnitude of maximal change in the posterior cranial vault in the transverse and vertical dimensions. The dot plot in Fig. 6 demonstrates an even distribution of maxi-mum change in the vertical and transverse vectors in the posterior cranial vault for both the 1-week postoperative and end-of-treatment time points.

The purpose of the wedge ostectomies is to facilitate lateral movement of the parietal bone flaps. After the bone cuts are made, there is an immediate transverse expansion of the posterior cranial vault that is appreciable before the incisions are closed; thus, the increased transverse dimen-sion of the cranial vault in the immediate postoperative period was an expected finding. However, the increased vertical dimension of the posterior vault at the immediate postoperative (prehelmet) time point was unexpected.

In the initial postoperative period, the patients continue to lie with their head positioned to one side because of the persistently narrow occiput of the scaphocephalic head. The vertical expansion observed before the helmet is ap-plied is presumably attributable to the passive expansion of the brain and dura mater through the aperture created by the wide strip craniectomy.

Once applied, the helmet facilitates bilateral transverse expansion by transferring the weight of the head from the parietal bone flaps to the occiput. Figures 4–6 demon-strate the occipital remodeling and transverse and vertical expansion that occur during postoperative helmet thera-py. The continued transverse hinging of the parietal bone flaps further increases the transverse dimension of the cra-niectomy defect, thus increasing the size of the aperture through which the dura can vertically expand. The vertical expansion itself can push the mobile parietal bone flaps laterally; thus, the vertical and transverse expansion and growth of the brain may synergistically drive the remodel-ing observed in this study.

A side-by-side comparison of subjects treated with and without wedge ostectomies was not performed; therefore, the direct impact of the wedge ostectomies alone cannot be determined in this study. It would be interesting to compare the vertical changes that accompany procedures that utilize narrower strip craniectomies with or without wedge ostectomies to see if comparable changes occur in the vertical dimension. Long-term 3D studies are needed

FIG. 5. Color maps were created to demonstrate the distance between the surfaces of the preoperative and end-of-treatment composite im-ages. Scale shows the color correlation to distance. As shown in red, the greatest distances between the images occur in the vertical and transverse directions in the parietal regions.

FIG. 6. A dot plot demonstrating the maximum change for each patient and the location of that change. Blue dots represent the maximum change between the preoperative and 1-week postoperative 3D images. Red dots indicate the location of maximum change between the preop-erative and end-of-treatment images. Figure is available in color online only.

Postoperative head shape change in sagittal craniosynostosis

J Neurosurg Pediatr Volume 19 • June 2017 689

to assess the durability of the head shape changes with continued growth.

One limitation of the current study is its sample size. Another lies in our inclusion criterion of patients treated by a single orthotist. We included only patients treated by the orthotist who manages the majority of our patients to control for any variability between multiple orthotists and their approach to and experience with postoperative hel-met therapy.

ConclusionsAfter extended sagittal strip craniectomy with wedge

ostectomies and postoperative helmet therapy, the CI nor-malizes rapidly (57.2 days) with an equal distribution of change in the CI occurring in the 1st week after surgery (prehelmet) and during helmet therapy. This finding may indicate a role for shortening the duration of helmet ther-apy. Three-dimensional analysis demonstrated significant expansion of the posterior cranial vault in the vertical and transverse dimensions. Further 3D studies are needed to determine how the width of the strip craniectomy and the presence or absence of wedge ostectomies affect 3D head shape outcomes.

References 1. David LR, Proffer P, Hurst WJ, Glazier S, Argenta LC:

Spring-mediated cranial reshaping for craniosynostosis. J Craniofac Surg 15:810–818, 2004

2. Dvoracek LA, Skolnick GB, Nguyen DC, Naidoo SD, Smyth MD, Woo AS, et al: Comparison of traditional versus nor-mative cephalic index in patients with sagittal synostosis: measure of scaphocephaly and postoperative outcome. Plast Reconstr Surg 136:541–548, 2015

3. Gociman B, Marengo J, Ying J, Kestle JR, Siddiqi F: Mini-mally invasive strip craniectomy for sagittal synostosis. J Craniofac Surg 23:825–828, 2012

4. Guimarães-Ferreira J, Gewalli F, David L, Olsson R, Friede H, Lauritzen CG: Spring-mediated cranioplasty compared with the modified pi-plasty for sagittal synostosis. Scand J Plast Reconstr Surg Hand Surg 37:208–215, 2003

5. Jimenez DF, Barone CM: Endoscopy-assisted wide-vertex craniectomy, “barrel-stave” osteotomies, and postoperative helmet molding therapy in the early management of sagittal suture craniosynostosis. Neurosurg Focus 9(3):e2, 2000

6. Jimenez DF, Barone CM, McGee ME, Cartwright CC, Baker CL: Endoscopy-assisted wide-vertex craniectomy, barrel stave osteotomies, and postoperative helmet molding thera-py in the management of sagittal suture craniosynostosis. J Neurosurg 100 (5 Suppl Pediatrics):407–417, 2004

7. Khechoyan D, Schook C, Birgfeld CB, Khosla RK, Saltzman B, Teng CC, et al: Changes in frontal morphology after sin-gle-stage open posterior-middle vault expansion for sagittal craniosynostosis. Plast Reconstr Surg 129:504–516, 2012

8. Lane L: Pioneer craniectomy for relief of mental imbecility due to premature sutural closure and microcephalus. J Am Med Assoc 18:49–50, 1892

9. Lannelongue M: De la craniectomie dans la microcephalie. Compt Rend Seances Acad Sci 50:1382–1385, 1890

10. Le MB, Patel K, Skolnick G, Naidoo S, Smyth M, Kane A, et al: Assessing long-term outcomes of open and endoscopic sagittal synostosis reconstruction using three-dimensional photography. J Craniofac Surg 25:573–576, 2014

11. Marcus JR, Domeshek LF, Loyd AM, Schoenleber JM, Das RR, Nightingale RW, et al: Use of a three-dimensional, normative database of pediatric craniofacial morphology for modern anthropometric analysis. Plast Reconstr Surg 124:2076–2084, 2009

12. Proctor MR: Endoscopic craniosynostosis repair. Transl Pediatr 3:247–258, 2014

13. Ridgway EB, Berry-Candelario J, Grondin RT, Rogers GF, Proctor MR: The management of sagittal synostosis using endoscopic suturectomy and postoperative helmet therapy. J Neurosurg Pediatr 7:620–626, 2011

14. Shah MN, Kane AA, Petersen JD, Woo AS, Naidoo SD, Smyth MD: Endoscopically assisted versus open repair of sagittal craniosynostosis: the St. Louis Children’s Hospital experience. J Neurosurg Pediatr 8:165–170, 2011

15. Taylor JA, Maugans TA: Comparison of spring-mediated cranioplasty to minimally invasive strip craniectomy and barrel staving for early treatment of sagittal craniosynostosis. J Craniofac Surg 22:1225–1229, 2011

16. Toma R, Greensmith AL, Meara JG, Da Costa AC, Ellis LA, Willams SK, et al: Quantitative morphometric outcomes fol-lowing the Melbourne method of total vault remodeling for scaphocephaly. J Craniofac Surg 21:637–643, 2010

17. van Veelen MLC, Mathijssen IM: Spring-assisted correction of sagittal suture synostosis. Childs Nerv Syst 28:1347–1351, 2012

DisclosuresThe authors report no conflict of interest concerning the materi-als or methods used in this study or the findings specified in this paper.

Author ContributionsConception and design: Derderian, Chou, Hallac, Kane. Acquisi-tion of data: Chou, Hallac, Patel, Stewart, Smartt, Seaward. Anal-ysis and interpretation of data: Derderian, Chou, Hallac, Patel, Cho. Drafting the article: Derderian, Chou. Critically revising the article: Derderian, Chou. Reviewed submitted version of manu-script: all authors. Approved the final version of the manuscript on behalf of all authors: Derderian. Study supervision: Derderian, Hallac.

Supplemental InformationPrevious PresentationsThis work was presented at the 2016 American Society of Plastic Surgeons Annual Meeting, Los Angeles, California.

CorrespondenceChristopher Derderian, Department of Plastic Surgery, University of Texas Southwestern Medical Center, 1801 Inwood Rd., Dallas, TX 75390-9132. email: christopher.derderian@utsouthwestern.edu.

www.PRSJournal.com 833

The prevalence of deformational plagio-cephaly and deformational brachycephaly has increased significantly since the adop-

tion of the Back to Sleep campaign in 1992.1,2 Although the rate of sudden infant death syn-drome has decreased by as much as 40 percent,2

the incidence of positional cranial deformation is estimated to have increased by as much as 600 percent,3 with a significant increase in the number of case referrals to specialized treatment centers.4 Current prevalence estimates of positional cranial deformities are as high as 47 to 48 percent.5,6

Although various treatments for deformational plagiocephaly/deformational brachycephaly have

Disclosure: The authors have no financial interest in any of the products or devices mentioned in this article. There were no sources of funding for this work.

Copyright © 2014 by the American Society of Plastic Surgeons

DOI: 10.1097/PRS.0000000000000955

Jordan P. Steinberg, M.D., Ph.D.

Roshni Rawlani, B.A.Laura S. Humphries, M.D.

Vinay Rawlani, M.D.Frank A. Vicari, M.D.

Chicago and Park Ridge, Ill.

Background: The authors investigated the effectiveness of conservative (repo-sitioning therapy with or without physical therapy) and helmet therapy, and identified factors associated with treatment failure.Methods: A total of 4378 patients evaluated for deformational plagiocephaly and/or deformational brachycephaly were assigned to conservative (reposi-tioning therapy, n = 383; repositioning therapy plus physical therapy, n = 2998) or helmet therapy (n = 997). Patients were followed until complete correction (diagonal difference <5 mm and/or cranial ratio <0.85) or 18 months. Rates of correction were calculated, and independent risk factors for failure were identified by multivariate analysis.Results: Complete correction was achieved in 77.1 percent of conserva-tive treatment patients; 15.8 percent required transition to helmet therapy (n = 534), and 7.1 percent ultimately had incomplete correction. Risk factors for failure included poor compliance (relative risk, 2.40; p = 0.009), advanced age (relative risk, 1.20 to 2.08; p = 0.008), prolonged torticollis (relative risk, 1.12 to 1.74; p = 0.002), developmental delay (relative risk, 1.44; p = 0.042), and severity of the initial cranial ratio (relative risk, 1.41 to 1.64; p = 0.044) and diagonal difference (relative risk, 1.31 to 1.48; p = 0.027). Complete cor-rection was achieved in 94.4 percent of patients treated with helmet therapy as first-line therapy and in 96.1 percent of infants who received helmets after failed conservative therapy (p = 0.375). Risk factors for helmet failure included poor compliance (relative risk, 2.42; p = 0.025) and advanced age (relative risk, 1.13 to 3.08; p = 0.011).Conclusions: Conservative therapy and helmet therapy are effective for position-al cranial deformation. Treatment may be guided by patient-specific risk factors. In most infants, delaying helmet therapy for a trial of conservative treatment does not preclude complete correction. (Plast. Reconstr. Surg. 135: 833, 2015.)CLINICAL QUESTION/LEVEL OF EVIDENCE: Therapeutic, III.

From the Ann and Robert H. Lurie Children’s Hospital of Chicago; the Division of Plastic Surgery, Northwestern Uni-versity Feinberg School of Medicine; and the Advocate Medi-cal Group, Lutheran General Hospital.Received for publication March 25, 2014; accepted Septem-ber 26, 2014.Presented in part at the 81st Annual Scientific Meeting of the American Society of Plastic Surgeons, in New Orleans, Loui-siana, October 26 through 30, 2012; and at the 92nd Annu-al Meeting of the American Association of Plastic Surgeons, in New Orleans, Louisiana, April 20 through 23, 2013.

Effectiveness of Conservative Therapy and Helmet Therapy for Positional Cranial Deformation

A Video Discussion by Kant Lin, M.D., accom-panies this article. Go to PRSJournal.com and click on “Video Discussions” in the “Videos” tab to watch.

PEDIATRIC/CRANIOFACIAL

834

Plastic and Reconstructive Surgery • March 2015

been well described, including repositioning therapy, physical therapy, and helmet therapy, limited evidence exists to guide treatment.7,8 Pre-vious studies have been limited by small cohorts, inadequate follow-up, lack of reliable objective outcome measures, and influence of commer-cial interests.7–10 Moreover, factors associated with treatment failure have not been identified. The purpose of this study was to (1) analyze the effec-tiveness of conservative therapy (repositioning therapy and repositioning therapy plus physical therapy) and helmet therapy in the treatment of deformational plagiocephaly and deformational brachycephaly and (2) identify independent risk factors for treatment failure in a large patient cohort using objective outcome measures. We hypothesized that although helmet therapy would allow the achievement of corrective endpoints with a high degree of success, conservative treat-ment alone would be effective in a significant pro-portion of infants.

PATIENTS AND METHODSThis retrospective cohort study was approved

by the Institutional Review Board of Children’s Memorial Hospital (presently Ann and Robert H. Lurie Children’s Hospital) in Chicago, Illinois. All patients who underwent treatment of nonsynostotic deformational plagiocephaly and/or deformational brachycephaly by a single pediatric craniofacial sur-geon (F.A.V.) between 2004 and 2011 were included in the analysis. Patients were excluded from final analysis if they (1) had undergone formal reposi-tioning therapy/physical therapy before initial eval-uation, (2) deviated from standardized treatment protocols (e.g., different helmet type because of patient preference), (3) had incomplete STARscan-ner data, or (4) were lost to follow-up.

Evaluation and Treatment AlgorithmAll patients underwent evaluation by the same

multidisciplinary team that included a pediatric craniofacial surgeon, physical therapist, nurse practitioner, and orthotist specifically trained in anthropometric cranial vault analysis. As part of the initial evaluation, parents completed a stan-dard birth history and demographic/socioeco-nomic survey administered by a nurse practitioner. Objective anthropometric measurements of the patient’s cranial vault were obtained using a three-dimensional laser surface scanner as detailed below. Patients also underwent a clinical evalua-tion of their cranial deformity by the surgeon with visual assessment of craniofacial asymmetry from

the posterior, vertex, and anterior views. A trained physical therapist assessed motor development, which included evaluation for the presence or absence of torticollis.

With the aforementioned assessments taken into consideration, patients were assigned nonran-domly to specific treatment modalities, including (1) conservative therapy, which consisted of repo-sitioning therapy with or without formal physical therapy; and (2) passive cranial orthotic molding (helmet therapy) (Fig. 1). Patients undergoing conservative therapy received either repositioning therapy or a combination of repositioning ther-apy and physical therapy, with the need for formal physical therapy based on the presence and sever-ity of head position preference, torticollis, and/or neuromuscular developmental delay at the initial consultation. All patients who received helmets also received repositioning therapy with or with-out physical therapy based on the same criteria. The rationale for concomitant repositioning ther-apy with or without physical therapy relates not only to the potential for relapse of deformational plagiocephaly/deformational brachycephaly with inattention to factors such as torticollis, but also to the prevention or remediation of significant delays in motor development.

Patients underwent cranial measurement with the three-dimensional laser surface scanner before therapy initiation and at follow-up every 3 months or sooner if progress was not grossly apparent on clinical examination. In a similar fashion to the ini-tial assessment, clinical judgment, rather than strict numeric criteria, was used to determine whether the patient should continue with his or her origi-nal treatment protocol or whether a change in therapy was warranted. A subset of patients who failed to achieve correction with initial conserva-tive therapy (i.e., repositioning therapy alone or combined repositioning therapy plus physical therapy) subsequently underwent helmet therapy with continued repositioning therapy with or with-out physical therapy (crossover group). Compli-ance was assessed at each follow-up visit by parental questionnaire for repositioning therapy and hel-met therapy and by clinical records for physical therapy. All patients were followed until complete cranial deformity correction was achieved, defined by a diagonal difference less than 5 mm for defor-mational plagiocephaly, a cranial ratio less than 0.85 for deformational brachycephaly, or until 18 months of age. Failure of treatment was defined as failure to achieve a diagonal difference less than 5 mm and/or a cranial ratio less than 0.85 by 18 months of age in either treatment group.

Volume 135, Number 3 • Positional Craniofacial Deformation

835

Anthropometric Cranial Vault AnalysisCranial vault anthropometrics were obtained

using the STARscanner Laser Data Acquisition System (Orthomerica Products, Inc., Orlando, Fla.) and analyzed using Yeti computer software (Yeti Software LLC, Seattle, Wash.) by a trained orthotist as described previously.11 Briefly, a stocki-net was placed over the infant’s head and mark-ers were placed identifying the superior aspect of the tragus (tragion). The sellion and tragion were used to define the base plane. Parallel to the base plane, 10 virtual two-dimensional sec-tions were constructed thorough the cranium up to the vertex. Section 3, which is one-third of the distance from the base plane to the vertex, was used to calculate anthropometric measurements in all patients. Cranial ratio and diagonal differ-ence measurements were calculated in standard

fashion as detailed previously (Fig. 2).11 Diagonal difference was used instead of the cranial vault asymmetry index because of the tendency for ratio measurements to naturally improve with continued head growth.12

Therapy ProtocolsRepositioning therapy involved parental coun-

seling and training by an in-office physical therapist. Training included discussion of positional prefer-ence, repositioning techniques to stretch tight-ened neck muscles, emphasizing the importance of “tummy time” lasting greater than 50 percent of awake time, carrying techniques to promote inde-pendent neck and truncal muscle development, and limiting the use of infant walking devices. Physical therapy involved an initial home program based on age and specific needs determined at the

Fig. 1. Initial treatment algorithm and assignments.

836

Plastic and Reconstructive Surgery • March 2015

initial visit followed by continued in-office sessions with regimented exercises. These exercises were designed to address deficiencies in range of motion and extension and general weakness, and also to help in the achievement of motor milestones. Any asymmetry in automatic head righting reflexes13 was addressed with active rather than passive exer-cises. The interval for treatment was adjusted at the discretion of the therapist, although children were typically followed until they were walking inde-pendently. Helmet therapy involved 23-hour daily wear of a STARband customized cranial orthotic molding helmet (Orthomerica Products, Inc., Orlando, Fla.). Helmets were precisely fabricated by a orthotic specialist using a subtractive process from idealized and actual STARscanner data.11 Hel-mets were evaluated at each visit and adjusted by the orthotist as needed.

Statistical AnalysisA multivariate logistic regression statistical

analysis using SPSS Version 17.0 (SPSS, Inc., Chi-cago, Ill.) was performed using a set of predefined clinical factors to identify independent risk fac-tors for treatment failure. Risk factors identified

as having a significance of p < 0.10 in univariate analysis were included in the multivariate analysis. The relative risk of treatment failure for a range of values within each identified independent clinical factor was calculated.

RESULTSBetween 2004 and 2011, 5152 patients with

deformational plagiocephaly and/or deforma-tional brachycephaly were treated with conserva-tive or helmet therapy. A total of 774 patients were excluded: 250 had undergone previous treat-ment, 318 had incomplete STARscanner data, 157 deviated from standardized treatment protocols, and 49 were lost to follow-up. Ultimately, 4378 patients with a diagnosis of deformational pla-giocephaly and/or deformational brachycephaly were included in the study (Fig. 3).

Treatment was initiated with conservative measures in 3381 infants (repositioning therapy, n = 383; repositioning therapy plus physical ther-apy; n = 2998) and with helmet therapy in 997 infants. Baseline characteristics of both groups are shown in Table 1. Starting ages differed

Fig. 2. Cranial measurements obtained with a three-dimensional laser surface scanner. (Above, left) Photograph of child with stockinet and markers. (Center, left) Surface render-ing illustrating the relative location of the base plane and section 3 used for all standard calculations. (Below, left) Two-dimensional schematic illustrating the sellion (Se) and tra-gion (Tr) landmarks for the base plane and the parallel sections, including the standard Section 3. (Right) Computation of cranial ratio as the ratio of biparietal width to anteropos-terior length and diagonal difference as the diagonal transcranial difference at 30 degrees.

Volume 135, Number 3 • Positional Craniofacial Deformation

837

significantly between the groups (7.1 ± 3.8 months for helmet therapy versus 5.1 ± 2.1 months for conservative therapy; p < 0.001), as did diagnoses. Deformational plagiocephaly was more frequent than deformational brachycephaly in those ini-tiated on helmet therapy (41 percent versus 25 percent). Torticollis (49 percent versus 40 per-cent; p < 0.001) and developmental delay (20 per-cent versus 9 percent; p < 0.001) were also more prevalent in the helmet therapy group. Further-more, cranial ratio (0.99 ± 0.28 versus 0.92 ± 0.25; p < 0.001) and diagonal difference (12.8 ± 4.7 ver-sus 9.2 ± 3.8; p < 0.001) were significantly more abnormal in these infants. The presence of multi-ple gestation, method of delivery, and age at deliv-ery did not differ between the groups.

Overall, complete correction was achieved in 4062 of 4378 patients (92.8 percent). A total of 77.1 percent of the 3381 conservatively treated patients (repositioning therapy, n = 329; reposi-tioning therapy plus physical therapy, n = 2279) achieved complete correction with repositioning

therapy with or without physical therapy alone. A subset of patients (15.8 percent of the initial cohort; repositioning therapy, n = 13; reposition-ing therapy plus physical therapy, n = 521) were transitioned to helmets (crossover group) because they failed to improve. The remaining 7.1 per-cent (repositioning therapy, n = 41; repositioning therapy plus physical therapy, n = 198) ultimately failed to achieve complete correction with con-tinued conservative therapy. In total, 1531 infants underwent cranial orthotic molding, including 997 patients who were initially assigned to helmet therapy and the 534 patients transitioned to hel-mets after having failed initial conservative treat-ment. Complete correction was achieved in 95.0 percent of these 1531 total patients who under-went helmet therapy (Fig. 3). There was no dif-ference in outcome between crossover patients who transitioned to helmet therapy after a mean 4.1 ± 1.4 months of conservative therapy and those who received helmet therapy as first-line treatment (96.1 percent versus 94.4 percent; p = 0.375).

Fig. 3. Flow diagram of treatment groups and outcomes. *Crossover versus helmet only, p = 0.17. RT, repositioning therapy; PT, physical therapy.

838

Plastic and Reconstructive Surgery • March 2015

Characteristics of crossover patients were compared with those who received only conser-vative therapy (Table 2). These patients were sig-nificantly older (5.7 ± 2.6 months versus 5.1 ± 1.9 months; p < 0.001) at the start of therapy and had significantly greater deformity, as demonstrated by their increased cranial ratio (0.94 ± 0.35 ver-sus 0.91 ± 0.23; p = 0.012) and diagonal difference (10.3 ± 4.1 versus 9.0 ± 3.8; p < 0.001). Torticol-lis (46 percent versus 39 percent; p = 0.004) and developmental delay (14 percent versus 8 per-cent; p < 0.001) were also more prevalent, as was cesarean delivery (38 percent versus 22 percent; p < 0.001). Compliance was significantly lower in crossover patients (84 percent versus 87 percent; p = 0.043); however, this improved after transition-ing to helmet therapy (84 percent versus 96 per-cent; p < 0.001). Overall compliance for helmet treatment was significantly better than for conser-vative treatment (94 percent versus 87 percent; p = 0.001).

Independent risk factors for conservative and helmet therapy failure were identified using a multivariate logistic regression statistical analysis (Table 3). For this analysis, crossover patients were included as a part of both groups because of their potential to fail conservative therapy and helmet therapy independently. Risk factors for failure of conservative treatment included poor compli-ance (relative risk, 2.4; p = 0.009), advanced age

at the time of therapy initiation (relative risk, 1.76 to 2.08; p = 0.008), the presence of torticol-lis (relative risk, 1.12 to 1.74; p = 0.002), the pres-ence of developmental delay (relative risk, 1.44; p = 0.042), and increased severity of cranial defor-mity at the time of therapy initiation as measured by the cranial ratio (relative risk, 1.08 to 1.11; p = 0.044) and diagonal difference (relative risk, 1.07 to 1.13; p = 0.027). Prematurity and male sex were not risk factors, whereas multiple gestation and vaginal delivery were protective.

Independent risk factors for helmet therapy failure included only advanced age and poor treatment compliance. Older patients at the time of helmet therapy initiation (p = 0.011), particularly age 9 to 12 months and older than 12 months, were 1.93 and 3.08 times more likely to fail helmet therapy than their 3- to 6-month-old counterparts, respectively. Noncompliant patients were 2.4 times more likely to fail helmet therapy (p = 0.025) compared with their com-pliant counterparts. However, cranial deformity severity (diagonal difference and cranial ratio) at therapy initiation, and the presence of torticollis and developmental delay, were not risk factors for helmet therapy failure.

Table 1. Baseline Patient Characteristics

Conservative Therapy (%)

Helmet Therapy (%) p

No. of patients 3381 997Sex 0.365 Male 1860 (55) 565 (57) Female 1521 (45) 432 (43)Age, mo 5.1 ± 2.1 7.1 ± 3.8 <0.001Diagnosis <0.001 Brachycephaly 839 (25) 186 (20) Plagiocephaly 861 (25) 412 (41) Combination 1681 (50) 389 (39)Cranial ratio 0.92 ± 0.25 0.99 ± 0.28 <0.001Diagonal difference 9.2 ± 3.8 12.8 ± 4.7 <0.001Torticollis <0.001 Present 1352 (40) 493 (49) Absent 2029 (60) 504 (51)Developmental delay <0.001 Present 300 (9) 198 (20) Absent 3081 (91) 799 (80)Gestation 0.741 Single 3110 (98) 914 (92) Multiple 271 (2) 83 (8)Method of delivery 0.451 Cesarean 839 (25) 235 (24) Vaginal 2542 (75) 762 (76)Prematurity 0.5325 Premature 465 (14) 145 (15) Full-term 2916 (86) 852 (85)

Table 2. Crossover Patient Characteristics

RT ± PT Only (%)

Crossover (%) p

No. of patients 2847 534Sex 0.4204 Male 1575 (55) 285 (53) Female 1272 (45) 249 (47)Age, mo 5.1 ± 1.9 5.7 ± 2.6 <0.001Diagnosis Brachycephaly 708 (25) 131 (25) 0.946 Plagiocephaly 722 (25) 139 (26) Combination 1417 (50) 264 (49)Cranial ratio 0.91 ± 0.23 0.94 ± 0.35 0.012Diagonal difference 9.0 ± 3.8 10.3 ± 4.1 <0.001Torticollis 0.004 Present 1108 (39) 244 (46) Absent 1739 (61) 290 (54)Developmental delay <0.001 Present 225 (8) 75 (14) Absent 2622 (92) 459 (86)Gestation 0.099 Single 2609 (92) 501 (94) Multiple 238 (8) 33 (6)Method of delivery <0.001 Cesarean 634 (22) 205 (38) Vaginal 2213 (78) 329 (62)Prematurity 0.951 Premature 392 (14) 73 (14) Full-term 2455 (86) 461 (86)Compliance 0.043 Conservative therapy 2488 (87) 449 (84) Helmet therapy — 513 (96)RT, repositioning therapy; PT, physical therapy.

Volume 135, Number 3 • Positional Craniofacial Deformation

839

DISCUSSIONThe results of this study demonstrate that con-

servative treatment (repositioning therapy with or without physical therapy) and helmet therapy are each effective in correcting positional cra-nial deformation. Overall in this study, the larg-est to date, successful treatment endpoints were achieved for 4062 of 4378 infants (92.8 percent). Among those who were initiated on repositioning therapy with or without physical therapy alone, 77.1 percent achieved complete correction. An additional 15.2 percent of these patients achieved complete correction after transitioning to hel-met therapy. Cranial orthotic molding helmets achieved complete correction in 95.0 percent of patients, with no difference in outcomes between patients who received helmet therapy after failed conservative therapy compared with those who received them as initial treatment.

Although many groups have identified general risk factors for the development of deformational plagiocephaly/deformational brachycephaly, no previous reports have focused on risk factors for failure of differing treatment modalities. The

results of the present study show that the risk fac-tors for treatment failure were distinct for con-servative and helmet therapy. For conservative therapy, the severity of deformity (as measured by cranial ratio and diagonal difference), the persis-tence of torticollis beyond 6 months of age, and neuromuscular developmental delay were risk fac-tors in addition to age and compliance. Multiple gestation and vaginal delivery were found to be protective against conservative treatment failure, whereas prematurity and male sex, two previously cited risk factors for deformational plagioceph-aly/deformational brachycephaly,14–17 were not found to be associated with outcomes. By contrast, for helmet therapy, the age at initiation of therapy and patient compliance were the only risk factors for failure.

Our identification of specific independent risk factors for treatment failure provides insight into the pathogenesis and treatment of deformational plagiocephaly/deformational brachycephaly. Current literature suggests the influence of both static and dynamic forces on the cranium that ultimately affect symmetry. Static forces include

Table 3. Analysis of Risk Factors for Treatment Failure

Conservative Therapy* Helmet Therapy†

Failure Rate (%) RR p Failure Rate (%) RR p

Therapy compliance Compliant 19.41 1.00 (Ref) 0.009 4.59 1.00 (Ref) 0.025 Noncompliant 46.58 2.40 11.11 2.42Age <3 mo 17.10 1.00 (Ref) 0.008 — — — 3–6 mo 20.51 1.20 4.22 1.00 (Ref) 0.011 6–9 mo 24.62 1.44 4.77 1.13 9–12 mo 30.09 1.76 8.14 1.93 >12 mo 35.55 2.08 13.00 3.08Torticollis Absent 19.39 1.00 (Ref) 0.002 4.88 1.00 (Ref) 0.587 <6 mo 21.72 1.12 5.22 1.07 >6 mo 33.74 1.74 5.47 1.12Cranial ratio <0.95 17.71 1.00 (Ref) 0.044 4.63 1.00 (Ref) 0.313 0.96–1.05 24.97 1.41 5.01 1.08 >1.05 29.04 1.64 5.13 1.11Diagonal difference <10 18.00 1.00 (Ref) 0.027 4.76 1.00 (Ref) 0.377 11–16 23.58 1.31 5.10 1.07 >16 27.81 1.48 5.38 1.13Developmental delay Absent 22.14 1.00 (Ref) 0.042 4.9 1.00 (Ref) 0.353 Present 31.88 1.44 5.4 1.10Gestation Single 23.66 1.00 (Ref) 0.034 5.01 1.00 (Ref) 0.443 Multiple 15.42 0.65 4.92 0.98Method of delivery Cesarean 26.00 1.00 (Ref) 0.032 4.89 1.00 (Ref) 0.493 Vaginal 22.01 0.85 5.04 1.03RT, repositioning therapy; PT, physical therapy; RR, relative risk; Ref, reference.*RT ± PT only and crossover (n = 3381).†Crossover and helmet only (n = 1531).

840

Plastic and Reconstructive Surgery • March 2015

in utero position and neonatal activities that may influence head turns to a particular side (e.g., the adoption of a preferred breastfeeding posi-tion).14,18,19 Dynamic forces include brain growth and the asymmetric pull from tightened or con-tracted muscles (or lack of pull from weak mus-cles).12,16,17,20 As discussed by Rogers et al.,17 these forces may be interdependent, in that intrauter-ine positioning gives rise to a congenital muscu-lar torticollis, which then continually drives head turning to a particular side. This, in turn, fuels a vicious cycle of asymmetric occipital flattening (deformational plagiocephaly) and a tendency of the infant to rest comfortably on the flat spot. A similar cycle may fuel the development of defor-mational plagiocephaly, in which muscles are sym-metrically weak. Our confirmation of torticollis as an independent risk factor for conservative treat-ment failure is consistent with its hypothesized pathogenic role. Specifically, our findings suggest that if torticollis persists beyond 6 months age, it is more likely to lead to prolonged cranial defor-mation with an increased risk for conservative treatment failure compared with torticollis that improves with normal neck muscular develop-ment at approximately 4 to 6 months of age. Our finding that neuromuscular developmental delay significantly increased the risk of conservative treatment failure is also consistent with the pro-posed etiopathogenesis of deformational plagio-cephaly/deformational brachycephaly. Delays in muscular development may preclude these infants from overcoming deforming forces. Finally, we found both vaginal delivery and multiple gesta-tion to be independently protective against con-servative treatment failure. Both represent static forces from intrauterine life or the birth process that may influence head shape. We hypothesize that when not associated with torticollis develop-ment, these factors may be markers for improved outcomes because the deforming forces cease after birth.

With regard to risk factors, it is interesting to note that the failure of helmet therapy in this analysis was affected only by compliance and the age at therapy initiation. The severity of the ini-tial cranial deformity and the presence of external deforming forces, such as torticollis and devel-opmental delay, were not found to be associated with treatment failure. Based on these results, one can consider the helmet to provide a passive envi-ronment of fixed shape into which the brain can drive skull growth, isolating out external deform-ing forces. Therefore, although the continued presence of deforming forces may prompt the

initiation of helmet therapy, the forces may be irrelevant to the outcome, provided that patients are compliant and therapy is commenced by the appropriate age. Of note, patients treated with helmets were more compliant in this study than those treated conservatively, likely because com-pliance with helmet therapy (assessed strictly on the basis of helmet wear) demanded less time and work by parents than the intensive home activities and weekly therapist visits required for reposition-ing therapy and physical therapy alone.

The age at which to initiate helmet therapy in infants with positional cranial deformities has been the subject of considerable debate.12,21–27 Although some studies have reported no corre-lation between age and treatment outcomes,24,25 others have shown significantly improved correc-tion with earlier treatment.22,23 In a recently pub-lished report, Kluba et al.23 found that patients whose helmet therapy was initiated earlier than 6 months of age had shorter treatment time and greater improvement in absolute and relative cra-nial asymmetry measurements. Although Seruya et al.12 confirmed a faster and more complete rate of correction in infants treated with helmets from an earlier age, they challenge the concepts espoused by Kluba et al.23 in terms of a specific cutoff age after which it may “too late” to start helmet therapy. They instead suggest that hel-mets can be effective provided that there is still brain growth; in older children, helmets would be required for longer durations because of deceler-ated growth.12 Our results in a significantly larger cohort of infants build on these findings. We have shown that (1) a significant proportion (77.1 per-cent) of patients initially treated with conservative measures alone will achieve complete correc-tion in head shape without the need for helmet therapy, and (2) delaying helmet therapy for a trial of initial conservative management does not increase the risk of helmet failure. We believe this is applicable throughout a “critical period” of infancy during which brain growth is ongoing and consequent volumetric changes in head shape can be directed. The “critical age” after which helmet therapy initiation is likely to be unsuccessful can theoretically be computed for each infant based on standardized volumetric head growth curves constructed from STARscanner population data.28

In comparison with previous reports, our study has several unique strengths. First, the data were derived from the largest cohort of patients here-tofore described in the literature with long-term follow-up. Second, all patients were evaluated and treated by a uniform team, including a single

Volume 135, Number 3 • Positional Craniofacial Deformation

841

pediatric craniofacial plastic surgeon, a cohesive group of trained pediatric physical therapists, and licensed orthotists specifically trained on the STARscanner system. In contrast to other centers that have come to rely on nurse practitioner screen-ing for head shape anomalies,29 all patients at our center were initially evaluated by the surgeon and plans of care were then communicated back to the patient’s referring pediatrician. Finally, rather than relying on imprecise recordings with hand-held calipers, this study used objective, volumetric cranial measurements obtained with a safe, reli-able, and validated instrument (STARscanner).11

An important limitation of our work is that the treatment modalities were not randomized, and the analysis cannot therefore be used to comment on the superiority of one method over another. We reemphasize that the primary pur-pose of this study was not to determine which is more effective. Rather, we aimed to retrospec-tively analyze the overall effectiveness of our treat-ment paradigm for infants with positional cranial deformation. As depicted in Figure 1, our para-digm was based on clinical judgment and social factors, including parent preference. Consistent with this, a majority of patients (3381 of 4378) were initially assigned to a trial of conservative treatment. Older infants and those with more severe cranial deformation (deformational pla-giocephaly with a higher diagonal difference, and deformational brachycephaly with a higher cranial ratio) tended to proceed directly to hel-met therapy. We believe the paradigm to be suc-cessful in that 77.1 percent of infants assigned to conservative therapy alone obtained complete head shape correction. A majority of those who did not (534 of 773) were successfully transi-tioned to helmet therapy (crossover patients), for which the rate of complete correction was 96.1 percent. This proved to be similar to the rate of complete correction of those patients assigned to helmets at the outset (94.4 percent). Moreover, it again supports the contention that delaying helmet therapy for an early trial of repositioning therapy with or without physical therapy does not preclude eventual complete correction.

The effectiveness of orthotic helmets for pas-sive cranial molding is clear. In keeping with our stated purpose, however, the practical import of this work is not to emphasize helmet therapy as the criterion standard because of a higher (i.e., 95 percent) overall correction rate, but instead to clarify to the broader pediatric community when a helmet should be recommended. For patients with minimal risk factors (e.g., age younger than

6 months, cranial ratio <0.95, diagonal difference <10 mm, absence of neuromuscular developmen-tal delay, or persistent torticollis), we strongly favor an initial trial of conservative therapy because of the high potential for success with these techniques alone. For patients with significant risk factor pro-files (e.g., age older than 7 to 8 months, cranial ratio >1.0, diagonal difference >15 mm, presence of developmental delay, or persistent torticollis), we favor counseling families on the increased like-lihood of conservative treatment failure and the option of proceeding straight to helmet therapy. For patients with some combination of the above factors and a more “moderate” risk factor profile, we do not believe any ultimate progress is lost with an initial trial of repositioning therapy with or without physical therapy alone. Future studies will help optimize care recommendations by (1) enabling pediatric practitioners to easily calculate the “critical age” after which helmet therapy can no longer achieve complete correction because of decelerated brain growth, and (2) establishing a more comprehensive normative data set of infants to better define indices of deformational plagio-cephaly and deformational brachycephaly severity.

CONCLUSIONSConservative treatment and helmet therapy

were found to be effective for correcting positional cranial deformation in 92.8 percent of infants. Treat-ment may be guided by patient-specific risk factors. Helmet therapy appears to isolate out external fac-tors that increase the risk of conservative treatment failure and thus may be preferable at the outset when these factors are present. Delaying the initi-ation of helmet therapy for a trial of conservative treatment does not preclude complete correction, provided that the helmet therapy is begun while brain growth is ongoing and patients are compliant.

Frank A. Vicari, M.D. Advocate Medical Group

Lutheran General Hospital1675 Dempster Street, 3rd Floor

Park Ridge, Ill. 60068frank@vicari.com

REFERENCES 1. American Academy of Pediatrics AAP Task Force on Infant

Positioning and SIDS: Positioning and SIDS. Pediatrics 1992;89:1120–1126.

2. Changing concepts of sudden infant death syndrome: Implications for infant sleeping environment and sleep position. American Academy of Pediatrics. Task Force on Infant Sleep Position and Sudden Infant Death Syndrome. Pediatrics 2000;105:650–656.

842

Plastic and Reconstructive Surgery • March 2015

3. Argenta LC, David LR, Wilson JA, Bell WO. An increase in infant cranial deformity with supine sleeping position. J Craniofac Surg. 1996;7:5–11.

4. Kane AA, Mitchell LE, Craven KP, Marsh JL. Observations on a recent increase in plagiocephaly without synostosis. Pediatrics 1996;97:877–885.

5. Rogers GF, Miller J, Mulliken JB. Comparison of a modifi-able cranial cup versus repositioning and cervical stretching for the early correction of deformational posterior plagio-cephaly. Plast Reconstr Surg. 2008;121:941–947.

6. Mawji A, Vollman AR, Hatfield J, McNeil DA, Sauvé R. The incidence of positional plagiocephaly: A cohort study. Pediatrics 2013;132:298–304.

7. Bialocerkowski AE, Vladusic SL, Howell SM. Conservative interventions for positional plagiocephaly: A systematic review. Dev Med Child Neurol. 2005;47:563–570.

8. Robinson S, Proctor M. Diagnosis and management of deformational plagiocephaly. J Neurosurg Pediatr. 2009;3:284–295.

9. McGarry A, Dixon MT, Greig RJ, et al. Head shape mea-surement standards and cranial orthoses in the treatment of infants with deformational plagiocephaly. Dev Med Child Neurol. 2008;50:568–576.

10. Rogers GF. Deformational plagiocephaly, brachycephaly, and scaphocephaly. Part II: Prevention and treatment. J Craniofac Surg. 2011;22:17–23.

11. Plank LH, Giavedoni B, Lombardo JR, Geil MD, Reisner A. Comparison of infant head shape changes in deformational plagiocephaly following treatment with a cranial remolding orthosis using a noninvasive laser shape digitizer. J Craniofac Surg. 2006;17:1084–1091.

12. Seruya M, Oh AK, Taylor JH, Sauerhammer TM, Rogers GF. Helmet treatment of deformational plagiocephaly: The rela-tionship between age at initiation and rate of correction. Plast Reconstr Surg. 2013;131:55e–61e.

13. Nandi R, Luxon LM. Development and assessment of the vestibular system. Int J Audiol. 2008;47:566–577.

14. Hutchison BL, Thompson JM, Mitchell EA. Determinants of nonsynostotic plagiocephaly: A case-control study. Pediatrics 2003;112:e316.

15. McKinney CM, Cunningham ML, Holt VL, Leroux B, Starr JR. A case-control study of infant, maternal and perinatal characteristics associated with deformational plagiocephaly. Paediatr Perinat Epidemiol. 2009;23:332–345.

16. Oh AK, Hoy EA, Rogers GF. Predictors of severity in deformational plagiocephaly. J Craniofac Surg. 2009;20 (Suppl 1):685–689.

17. Rogers GF, Oh AK, Mulliken JB. The role of congenital mus-cular torticollis in the development of deformational plagio-cephaly. Plast Reconstr Surg. 2009;123:643–652.

18. Rogers GF. Deformational plagiocephaly, brachycephaly, and scaphocephaly. Part I: Terminology, diagnosis, and etio-pathogenesis. J Craniofac Surg. 2011;22:9–16.

19. van Vlimmeren LA, van der Graaf Y, Boere-Boonekamp MM, et al. Risk factors for deformational plagiocephaly at birth and at 7 weeks of age: A prospective cohort study. Pediatrics 2007;119:e408–418.

20. Golden KA, Beals SP, Littlefield TR, Pomatto JK. Sternocleidomastoid imbalance versus congenital muscular torticollis: Their relationship to positional plagiocephaly. Cleft Palate Craniofac J. 1999;36:256–261.

21. Graham JM Jr, Gomez M, Halberg A, et al. Management of deformational plagiocephaly: Repositioning versus orthotic therapy. J Pediatr. 2005;146:258–262.

22. Kelly KM, Littlefield TR, Pomatto JK, et al. Importance of early recognition and treatment of deformational plagio-cephaly with orthotic cranioplasty. Cleft Palate Craniofac J. 1999;36:127–130.

23. Kluba S, Kraut W, Reinert S, Krimmel M. What is the opti-mal time to start helmet therapy in positional plagiocephaly? Plast Reconstr Surg. 2011;128:492–498.

24. Mulliken JB, Vander Woude DL, Hansen M, LaBrie RA, Scott RM. Analysis of posterior plagiocephaly: Deformational versus synostotic. Plast Reconstr Surg. 1999;103: 371–380.

25. Teichgraeber JF, Seymour-Dempsey K, Baumgartner JE, et al. Molding helmet therapy in the treatment of brachyceph-aly and plagiocephaly. J Craniofac Surg. 2004;15:118–123.

26. Vles JS, Colla C, Weber JW, et al. Helmet versus nonhelmet treatment in nonsynostotic positional posterior plagioceph-aly. J Craniofac Surg. 2000;11:572–574.

27. Thompson JT, David LR, Wood B, et al. Outcome analysis of hel-met therapy for positional plagiocephaly using a three-dimen-sional surface scanning laser. J Craniofac Surg. 2009;20:362–365.

28. Steinberg JP, Rawlani V, Rawlani R, Dumanian GA, Vicari FA. The “critical age:” Objective, patient-specific timing of hel-met therapy in treatment of deformational plagiocephaly and brachycephaly. Poster presented at: 92nd Annual Meeting of the American Association of Plastic Surgeons; April 20–23, 2013; New Orleans, La.

29. Kuang AA, Bergquist C, Crupi L, Oliverio M, Selden NR. Effectiveness and safety of independent pediatric nurse practitioners in evaluating plagiocephaly. Plast Reconstr Surg. 2013;132:414–418.