Eur J Pediatr DOI 10.1007/s00431-015-2520-x
ORIGINAL ARTICLE
Preventing deformational plagiocephaly through parent guidance: a randomized, controlled trial Henri Aarnivala 1,3,4 & Ville Vuollo 2,4 & Virpi Harila 2,4 & Tuomo Heikkinen 2,4 & Pertti Pirttiniemi 2,4 & A. Marita Valkama 1,3,4
Received: 26 January 2015 / Revised: 3 March 2015 / Accepted: 10 March 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Deformational plagiocephaly (DP) occurs frequently in otherwise healthy infants. Many infants with DP undergo physiotherapy or helmet therapy, and ample treatment-related research is available. However, the possibility of preventing DP has been left with little attention. We sought to evaluate the effectiveness of intervention in the newborn’s environment, positioning, and handling on the prevalence of DP at 3 months and to investigate the causal relationship between DP and cervical imbalance. We carried out a randomized controlled trial, with healthy newborns randomized into two groups at birth. All families received standard positioning instructions to prevent SIDS. Additionally, the intervention group received detailed instructions regarding the infant’s environment, positioning,
Communicated by Beat Steinmann * Henri Aarnivala [email protected] Ville Vuollo [email protected] Virpi Harila [email protected] Tuomo Heikkinen [email protected] Pertti Pirttiniemi [email protected]
and handling, with the goal of creating a nonrestrictive environment that promotes spontaneous physical movement and symmetrical motor development. Two- and three-dimensional photogrammetry served to assess cranial shape and goniometry to measure cervical motion. At 3 months, the prevalence of DP was lower in the intervention group in both 2D (11 vs 31 %) and 3D analyses (15 vs 33 %), and the asymmetry was milder in the intervention group. Infants with DP at follow-up had also developed more torticollis. Conclusion: An early educational intervention reduces the prevalence and severity of DP at 3 months. What is Known: •Deformational plagiocephaly, often with associated torticollis, is common in healthy infants. •Parental education is frequently recommended for preventing deformational plagiocephaly, although information regarding the effectiveness of preventive strategies is scarce. What is New: •Early parent guidance effectively reduces the prevalence and severity of DP and improves the cervical range of motion at three months. •Educating both parents and professionals about proper infant positioning on a national scale could help minimize public healthcare costs.
Keywords Plagiocephaly . Nonsynostotic plagiocephaly . Cranial deformations . Torticollis . Prevention . Motor development
A. Marita Valkama [email protected] 1
Department of Children and Adolescence, Oulu University Hospital, Oulu, Finland
2
Department of Oral Development and Orthodontics, Oulu University Hospital, Oulu, Finland
3
PEDEGO Research Group, University of Oulu, Oulu, Finland
4
Medical Research Center Oulu, University of Oulu, Oulu, Finland
Abbreviations ACAI Anterior cranial asymmetry index CI Cephalic index CMT Congenital muscular torticollis DP Deformational plagiocephaly
Eur J Pediatr
enL enR Mp OCLR-2D OCLR-3D PCAI TrL TrR
Endocanthion left Endocanthion right Midpoint Oblique cranial length ratio from 2D image Oblique cranial length ratio from 3D image Posterior cranial asymmetry index Tragion left Tragion right
hospitalization. The primary objective was to determine whether such measures significantly impact cranial shape. Another subject of interest was the role of cervical imbalance in the development of cranial deformations. To minimize possible bias and diagnostic inaccuracy resulting from the subjective assessment of cranial shape, we used two- and threedimensional photogrammetric methods in a blinded setting.
Methods Introduction Deformational plagiocephaly (DP) is capturing ever more attention from parents and professionals alike. The past two decades have seen the prevalence of DP rise, generating interest not only in studying the nature and evolution of DP but even more so in identifying ways to effectively treat the condition. Many researchers have suggested that the rising prevalence of DP may stem from the now standard supine sleeping position [1], although recent studies have challenged this perception [6, 19, 22, 28, 43, 44]. DP results from prolonged, unevenly distributed external forces on an infant’s head that direct the growth of the cranium towards an asymmetrical shape. In its most classical form, DP consists of unilateral occipital flattening, ipsilateral frontal bossing, and anterior shifting of the ipsilateral ear and cheek [34]. Congenital muscular torticollis (CMT) was often considered the main predisposing factor for DP [15, 36], but most recent studies suggest that cranial shape is more often determined by postnatal factors than pre- and perinatal factors and that most concomitant cervical imbalance—positional torticollis—develops postnatally along with DP [11]. One-sided positioning of the infant, infant positional preference, and slower motor development have been associated with a higher risk for DP [5, 9, 21, 32, 44]. A rapidly growing head is malleable and susceptible to deformation, especially during the first months of life, and the prevalence of DP peaks between 2 and 4 months, declining slowly thereafter [2, 19, 28, 44]. Although DP is generally considered an aesthetic problem, many infants with DP must undergo physical therapy or even orthotic helmet therapy, which can not only be difficult for the parents but also cause side effects for the infant, including skin irritation, sweating, and pain [13, 14, 27, 31, 35, 43, 45–47]. Given the high prevalence of the condition, surprisingly little research has explored its prevention. To our knowledge, the only prospective study to do so thus far was a multicentric, non-randomized trial conducted by Cavalier et al., who tested the effectiveness of early intervention in the newborn’s environment and reported promising results [6]. We conducted a single-center randomized controlled trial to compare the effect of early intervention in the newborn’s environment, positioning, and handling, with that of the current advice given to the parents of newborns during birth
Design and randomization This study was a populationbased, partly blinded, two-armed randomized controlled trial (Fig. 1). The subjects were recruited at Oulu University Hospital on preselected dates between February 2012 and December 2013. The recruitment dates were spread evenly throughout the year to minimize any possible effect of seasonal variation. Infants born on these dates were considered eligible if they were born after ≥35 weeks of gestation, healthy enough to manage without intensive care, resided within a 30-min driving distance from Oulu University Hospital, and had no cheilopalatoschisis, craniosynostosis, or dysmorphic features. We collected participants’ background data regarding gestation and delivery from maternal and infant medical records. During their birth hospitalization, all participants underwent an initial physical examination 36–72 h after birth, after which they were randomized in a 1:1 ratio to either the intervention group or the control group. Allocation was performed according to a computer-generated randomization plan in permuted blocks of four. At approximately 3 months of age, the participants attended a follow-up visit, where they underwent another detailed physical examination. Examination protocol During the initial examination, the newborns were screened for clavicular fractures, sternocleidomastoideal masses, and any other malformations or deformations that could account for the supposed cervical imbalances. The cervical range of motion was measured with a Scheppach® digital goniometer, with the infant lying supine on an examination table with the shoulders at the edge. The head was rotated (chin past shoulder) passively by supporting the head from below and allowing it to turn freely as far as possible. Lateral flexion (ear to shoulder) was assessed by applying very gentle pressure until resistance was noted. All measurements were performed three times with a digital goniometer, and torticollis was defined as a ≥15° difference in lateral flexion or rotation between sides from the highest values [7]. Finally, a digital photograph of the vertex view was taken with a Canon® EOS 600D DSLR camera, with the infant still lying on the edge of the examination table. Head position was standardized for the photograph by extending the head slightly until an imaginary line through the otobasion superius and exocanthion was perpendicular to the examination table. The facial midline was parallel to a straight line in
Eur J Pediatr Assessed for eligibility (n=270)
Excluded (n=159) Not meeting inclusion criteria* (n=5) Declined to participate in the followup (n=154)
Randomized (n=111)
Allocated to intervention group (n=55)
Allocated to control group (n=56)
Lost to follow-upa (n=6)
Lost to follow-upa (n=5)
Discontinued interventionb (n=2)
Analyzed (n=45) Excluded from analysisc (n=2)
Analyzed (n=51)
Fig. 1 Study flow. *2 syndromic infants, 2 with cleft lip and palate and 1 craniosynostosis; a2 infants in the intervention group moved out of the study area, 9 discontinued due to personal reasons; bDuring the guidance session, parents of 2 infants reported that following the instructions for the
intervention group would be overwhelmingly difficult and thus were excluded from the study; cAs only one set of twins participated in the study, they were excluded from the final analysis, as twins are known to be at a greater risk for DP
the floor, and the camera was situated 80 cm from the head on the level of the examination table.
Intervention and control groups Before their discharge from the maternity ward, the parents of the newborns in the intervention group received detailed recommendations regarding their infant’s environment, positioning, and handling. An experienced neonatologist (AMV) presented the information to each family in a private 15-min guidance session and as well as in print (Appendix). Following the AAP recommendations [24], the parents were instructed to put their infant to sleep on the back, alternating the position of the head evenly between left and right. Complementary advice was adapted from previous studies and articles on DP and its prevention [6, 24, 35]. The parents were encouraged to prepare the infant’s environment so as not to restrict the infant’s spontaneous movement. Because the infant should receive stimuli equally from all directions, toys and other interesting objects should be spread out on the floor evenly and, if needed, their positions changed. Similarly, the bed/cot should be placed ideally with the infant’s head or feet towards the window or other major source of light. If this could not be arranged, parents were instructed to alternate the infant’s sleeping position regularly in relation to the source of light. Parents should regularly alternate sides when handling the
During the follow-up visit, a clinical examination was performed, and a standard digital photograph was taken in the same way as at birth. We assessed active rotation by encouraging the infant to rotate his head as far as possible in both directions using either a musical toy or a parent. Passive lateral flexion was assessed in the same way as at birth. We assessed motor development with the Griffiths Development Scale (sections A, D, and E) with the sum of calculated sub-quotients converted into Z-scores according to the formula B(individual score−average score)/standard deviation,^ using the average score and standard deviation from our sample. Finally, a three-dimensional image of the head was generated using a stereophotogrammetric method in a standardized setting. First, a tight nylon cap was fitted on the infant’s head to avoid hair artifacts. Next, the infant was seated on a chair, centered in the 3D scanner (3dMD® 5-pod system, 3dMD®, Atlanta, GA) and lured to look through a small window in the panel in front of him or her. We then used five synchronized cameras to capture a 360° image of the head.
Eur J Pediatr
infant (e.g., right or left hand when supporting the infant’s head). The infant should spend time in the prone position (i.e., tummy time) when awake and under supervision, with minimal time spent in car seats and bouncers. If the infant began to show a positional preference, the parents were instructed to place interesting objects predominantly on the opposite side as well as to favor the opposite side in handling, bottle-feeding, sleeping position, and other activities. In addition, parents were informed about stretching exercises for the cervical muscles, which could be used if the infant showed or began to show signs of cervical imbalance [46]. The parents of infants in the control group received the standard guidance on infant positioning given before their discharge from Oulu University Hospital; it is worth noting that although Finnish national guidelines state that infants should explicitly sleep on their backs to prevent SIDS, the instructions given in the maternity ward also allow infants to sleep on their side. Quantifying cranial shape From the two-dimensional digital photographs, we measured the Oblique Cranial Length Ratio (OCLR-2D) and Cephalic Index (CI) on the plane of maximum head circumference using a semiautomated, computer-based method [2]. OCLR-2D is the ratio between the longest and shortest oblique transcranial diameter×100 %, where the diameters are measured in a 40° angle to the sagittal midline. CI is the ratio between skull width/length. We processed and analyzed the three-dimensional images with Rapidform® 2006 3D software using custom macros written with Visual Basic for Applications (VBA). First, we used ready-made software tools to remove shoulders and other excess data and to level out possible bumps caused by cap seams. Next, we standardized the position of the head using a coordinate system based on the craniofacial landmarks endocanthion left (EnL), endocanthion right (EnR), tragion left (TrL), tragion right (TrR), and the mirror face. We constructed the sagittal reference plane (yz) with the mirror face method [48]. In short, the original facial shell and a mirror shell were registered together using the best fit technique, and the symmetry plane of the resulting structure was treated as the sagittal plane for the original face. The midpoint (Mp) was set at the intersection of the tragus connection line (between TrL and TrR) and the sagittal plane. We then defined the transverse reference plane (xz) to run through the point right in the middle of the two endocanthions (midendocanthion) and Mp perpendicular to the sagittal plane, and the coronal reference plane (xy) perpendicular to the other two planes, also through Mp. The transverse plane served as the base plane. At this point, the x-, y-, and z-axes have also been defined. After the aligning process, we defined two reference planes for analysis. The plane for two-dimensional measurements is the plane parallel to the base plane at the maximum posterior curvature in the occipital region (maximum head circumference). Finally, to prevent the ears from interfering with the
volumetric analysis, the plane for three-dimensional measurements was defined as running parallel to the base plane, immediately above the highest part of the helix of the higher set ear. To quantify cranial symmetry, we calculated the twodimensional variables diagonal difference and ear offset and the three-dimensional variables anterior cranial asymmetry index (ACAI) and posterior cranial asymmetry index (PCAI) according to Meyer-Marcotty et al. (Figs. 2 and 3) [30]. We also calculated the two-dimensional variable Oblique Cranial Length Ratio from the 3D image (OCLR-3D) in the same way as OCLR-2D from the 2D photographs, in order to assess the presence of DP as a discrete variable, as no cut-off points for plagiocephaly have been established for the previously mentioned variables [10]. We set the cut-off point for DP at OCLR-3D≥104 %. Cut-offs for moderate and severe DP were set at OCLR-3D≥108 % and OCLR-3D≥112 %, respectively [45]. We used the same cut-offs for OCLR-2D in the 2D analysis. Outcome measures The primary outcome was the presence of DP at follow-up in either the 2D or 3D analysis. Another main outcome was the severity of DP in affected infants. We assessed the rest of the variables from the 3D analysis, the cephalic index from the 2D analysis, the cervical range of motion and cervical imbalance at follow-up, and motor development as secondary outcomes. Identifying possible risk factors for cranial deformations and cervical imbalance was another secondary objective. Blinding Both researchers and parents were blinded to group allocation until after the neonatal examination. The
Fig. 2 Diagonal difference [cm]=difference between the transcranial diagonals (a, b). OCLR-3D [%]=ratio between the longest and shortest oblique transcranial diagonal (a, b)×100 %. Ear offset [cm]=distance between the tragion landmarks (TrL, TrR) on the z-axis (c)
Eur J Pediatr Fig. 3 3D measurement plane, sagittal plane, and coronal plane, dividing the cranium into four quadrants. ACAI [%]=ratio of the anterior cuboid volumes and PCAI [%]=ratio of the posterior cuboid volumes using the formula: (larger cuboid volume− smaller cuboid volume)/smaller cuboid volume×100 %
researchers responsible for obtaining and analyzing the 2D and 3D images (HA and VV) were blinded to group allocation until after the measurements. Parents in the intervention group were urged not to discuss the instructions with other parents in the maternity ward or elsewhere until after the study ended. Therapy compliance At the follow-up visit, parents completed a questionnaire about their infant’s environment and care, which served to determine their compliance with the instructions given. The questionnaire inquired about the type of feeding (breast or bottle) and number of hours the infant used a pacifier. The number of sleeping hours, position when asleep, and the type of bedding were recorded, as was time spent on the floor (e.g., on a play mat), in carriers, bouncers, and car seats. The positioning of interesting objects and toys was recorded, and any positional preference reported by the parents was noted. We calculated a mobility score (hours per day on the floor when) and an immobility score (hours per day in a carrier, car seat, or bouncer when awake and asleep) from the responses according to Cavalier et al. [6]. Statistics We calculated a sample size of 86 (43 in each group) using a 5 % significance level, a power of 80 % and a fall in the prevalence of DP from 31 to 8 % (one quarter). We compared the main outcome variables between the groups with the χ2 test, Fisher’s exact test, and the Mann-Whitney U test, as appropriate. We used the χ2 test, Fisher’s exact test, the independent samples t test, and the Mann-Whitney U test to compare secondary outcome variables, environmental
measures, and Z-scores for motor development. We compared the groups on an intention-to-treat basis. The level of significance was set at p
ORIGINAL ARTICLE
Preventing deformational plagiocephaly through parent guidance: a randomized, controlled trial Henri Aarnivala 1,3,4 & Ville Vuollo 2,4 & Virpi Harila 2,4 & Tuomo Heikkinen 2,4 & Pertti Pirttiniemi 2,4 & A. Marita Valkama 1,3,4
Received: 26 January 2015 / Revised: 3 March 2015 / Accepted: 10 March 2015 # Springer-Verlag Berlin Heidelberg 2015
Abstract Deformational plagiocephaly (DP) occurs frequently in otherwise healthy infants. Many infants with DP undergo physiotherapy or helmet therapy, and ample treatment-related research is available. However, the possibility of preventing DP has been left with little attention. We sought to evaluate the effectiveness of intervention in the newborn’s environment, positioning, and handling on the prevalence of DP at 3 months and to investigate the causal relationship between DP and cervical imbalance. We carried out a randomized controlled trial, with healthy newborns randomized into two groups at birth. All families received standard positioning instructions to prevent SIDS. Additionally, the intervention group received detailed instructions regarding the infant’s environment, positioning,
Communicated by Beat Steinmann * Henri Aarnivala [email protected] Ville Vuollo [email protected] Virpi Harila [email protected] Tuomo Heikkinen [email protected] Pertti Pirttiniemi [email protected]
and handling, with the goal of creating a nonrestrictive environment that promotes spontaneous physical movement and symmetrical motor development. Two- and three-dimensional photogrammetry served to assess cranial shape and goniometry to measure cervical motion. At 3 months, the prevalence of DP was lower in the intervention group in both 2D (11 vs 31 %) and 3D analyses (15 vs 33 %), and the asymmetry was milder in the intervention group. Infants with DP at follow-up had also developed more torticollis. Conclusion: An early educational intervention reduces the prevalence and severity of DP at 3 months. What is Known: •Deformational plagiocephaly, often with associated torticollis, is common in healthy infants. •Parental education is frequently recommended for preventing deformational plagiocephaly, although information regarding the effectiveness of preventive strategies is scarce. What is New: •Early parent guidance effectively reduces the prevalence and severity of DP and improves the cervical range of motion at three months. •Educating both parents and professionals about proper infant positioning on a national scale could help minimize public healthcare costs.
Keywords Plagiocephaly . Nonsynostotic plagiocephaly . Cranial deformations . Torticollis . Prevention . Motor development
A. Marita Valkama [email protected] 1
Department of Children and Adolescence, Oulu University Hospital, Oulu, Finland
2
Department of Oral Development and Orthodontics, Oulu University Hospital, Oulu, Finland
3
PEDEGO Research Group, University of Oulu, Oulu, Finland
4
Medical Research Center Oulu, University of Oulu, Oulu, Finland
Abbreviations ACAI Anterior cranial asymmetry index CI Cephalic index CMT Congenital muscular torticollis DP Deformational plagiocephaly
Eur J Pediatr
enL enR Mp OCLR-2D OCLR-3D PCAI TrL TrR
Endocanthion left Endocanthion right Midpoint Oblique cranial length ratio from 2D image Oblique cranial length ratio from 3D image Posterior cranial asymmetry index Tragion left Tragion right
hospitalization. The primary objective was to determine whether such measures significantly impact cranial shape. Another subject of interest was the role of cervical imbalance in the development of cranial deformations. To minimize possible bias and diagnostic inaccuracy resulting from the subjective assessment of cranial shape, we used two- and threedimensional photogrammetric methods in a blinded setting.
Methods Introduction Deformational plagiocephaly (DP) is capturing ever more attention from parents and professionals alike. The past two decades have seen the prevalence of DP rise, generating interest not only in studying the nature and evolution of DP but even more so in identifying ways to effectively treat the condition. Many researchers have suggested that the rising prevalence of DP may stem from the now standard supine sleeping position [1], although recent studies have challenged this perception [6, 19, 22, 28, 43, 44]. DP results from prolonged, unevenly distributed external forces on an infant’s head that direct the growth of the cranium towards an asymmetrical shape. In its most classical form, DP consists of unilateral occipital flattening, ipsilateral frontal bossing, and anterior shifting of the ipsilateral ear and cheek [34]. Congenital muscular torticollis (CMT) was often considered the main predisposing factor for DP [15, 36], but most recent studies suggest that cranial shape is more often determined by postnatal factors than pre- and perinatal factors and that most concomitant cervical imbalance—positional torticollis—develops postnatally along with DP [11]. One-sided positioning of the infant, infant positional preference, and slower motor development have been associated with a higher risk for DP [5, 9, 21, 32, 44]. A rapidly growing head is malleable and susceptible to deformation, especially during the first months of life, and the prevalence of DP peaks between 2 and 4 months, declining slowly thereafter [2, 19, 28, 44]. Although DP is generally considered an aesthetic problem, many infants with DP must undergo physical therapy or even orthotic helmet therapy, which can not only be difficult for the parents but also cause side effects for the infant, including skin irritation, sweating, and pain [13, 14, 27, 31, 35, 43, 45–47]. Given the high prevalence of the condition, surprisingly little research has explored its prevention. To our knowledge, the only prospective study to do so thus far was a multicentric, non-randomized trial conducted by Cavalier et al., who tested the effectiveness of early intervention in the newborn’s environment and reported promising results [6]. We conducted a single-center randomized controlled trial to compare the effect of early intervention in the newborn’s environment, positioning, and handling, with that of the current advice given to the parents of newborns during birth
Design and randomization This study was a populationbased, partly blinded, two-armed randomized controlled trial (Fig. 1). The subjects were recruited at Oulu University Hospital on preselected dates between February 2012 and December 2013. The recruitment dates were spread evenly throughout the year to minimize any possible effect of seasonal variation. Infants born on these dates were considered eligible if they were born after ≥35 weeks of gestation, healthy enough to manage without intensive care, resided within a 30-min driving distance from Oulu University Hospital, and had no cheilopalatoschisis, craniosynostosis, or dysmorphic features. We collected participants’ background data regarding gestation and delivery from maternal and infant medical records. During their birth hospitalization, all participants underwent an initial physical examination 36–72 h after birth, after which they were randomized in a 1:1 ratio to either the intervention group or the control group. Allocation was performed according to a computer-generated randomization plan in permuted blocks of four. At approximately 3 months of age, the participants attended a follow-up visit, where they underwent another detailed physical examination. Examination protocol During the initial examination, the newborns were screened for clavicular fractures, sternocleidomastoideal masses, and any other malformations or deformations that could account for the supposed cervical imbalances. The cervical range of motion was measured with a Scheppach® digital goniometer, with the infant lying supine on an examination table with the shoulders at the edge. The head was rotated (chin past shoulder) passively by supporting the head from below and allowing it to turn freely as far as possible. Lateral flexion (ear to shoulder) was assessed by applying very gentle pressure until resistance was noted. All measurements were performed three times with a digital goniometer, and torticollis was defined as a ≥15° difference in lateral flexion or rotation between sides from the highest values [7]. Finally, a digital photograph of the vertex view was taken with a Canon® EOS 600D DSLR camera, with the infant still lying on the edge of the examination table. Head position was standardized for the photograph by extending the head slightly until an imaginary line through the otobasion superius and exocanthion was perpendicular to the examination table. The facial midline was parallel to a straight line in
Eur J Pediatr Assessed for eligibility (n=270)
Excluded (n=159) Not meeting inclusion criteria* (n=5) Declined to participate in the followup (n=154)
Randomized (n=111)
Allocated to intervention group (n=55)
Allocated to control group (n=56)
Lost to follow-upa (n=6)
Lost to follow-upa (n=5)
Discontinued interventionb (n=2)
Analyzed (n=45) Excluded from analysisc (n=2)
Analyzed (n=51)
Fig. 1 Study flow. *2 syndromic infants, 2 with cleft lip and palate and 1 craniosynostosis; a2 infants in the intervention group moved out of the study area, 9 discontinued due to personal reasons; bDuring the guidance session, parents of 2 infants reported that following the instructions for the
intervention group would be overwhelmingly difficult and thus were excluded from the study; cAs only one set of twins participated in the study, they were excluded from the final analysis, as twins are known to be at a greater risk for DP
the floor, and the camera was situated 80 cm from the head on the level of the examination table.
Intervention and control groups Before their discharge from the maternity ward, the parents of the newborns in the intervention group received detailed recommendations regarding their infant’s environment, positioning, and handling. An experienced neonatologist (AMV) presented the information to each family in a private 15-min guidance session and as well as in print (Appendix). Following the AAP recommendations [24], the parents were instructed to put their infant to sleep on the back, alternating the position of the head evenly between left and right. Complementary advice was adapted from previous studies and articles on DP and its prevention [6, 24, 35]. The parents were encouraged to prepare the infant’s environment so as not to restrict the infant’s spontaneous movement. Because the infant should receive stimuli equally from all directions, toys and other interesting objects should be spread out on the floor evenly and, if needed, their positions changed. Similarly, the bed/cot should be placed ideally with the infant’s head or feet towards the window or other major source of light. If this could not be arranged, parents were instructed to alternate the infant’s sleeping position regularly in relation to the source of light. Parents should regularly alternate sides when handling the
During the follow-up visit, a clinical examination was performed, and a standard digital photograph was taken in the same way as at birth. We assessed active rotation by encouraging the infant to rotate his head as far as possible in both directions using either a musical toy or a parent. Passive lateral flexion was assessed in the same way as at birth. We assessed motor development with the Griffiths Development Scale (sections A, D, and E) with the sum of calculated sub-quotients converted into Z-scores according to the formula B(individual score−average score)/standard deviation,^ using the average score and standard deviation from our sample. Finally, a three-dimensional image of the head was generated using a stereophotogrammetric method in a standardized setting. First, a tight nylon cap was fitted on the infant’s head to avoid hair artifacts. Next, the infant was seated on a chair, centered in the 3D scanner (3dMD® 5-pod system, 3dMD®, Atlanta, GA) and lured to look through a small window in the panel in front of him or her. We then used five synchronized cameras to capture a 360° image of the head.
Eur J Pediatr
infant (e.g., right or left hand when supporting the infant’s head). The infant should spend time in the prone position (i.e., tummy time) when awake and under supervision, with minimal time spent in car seats and bouncers. If the infant began to show a positional preference, the parents were instructed to place interesting objects predominantly on the opposite side as well as to favor the opposite side in handling, bottle-feeding, sleeping position, and other activities. In addition, parents were informed about stretching exercises for the cervical muscles, which could be used if the infant showed or began to show signs of cervical imbalance [46]. The parents of infants in the control group received the standard guidance on infant positioning given before their discharge from Oulu University Hospital; it is worth noting that although Finnish national guidelines state that infants should explicitly sleep on their backs to prevent SIDS, the instructions given in the maternity ward also allow infants to sleep on their side. Quantifying cranial shape From the two-dimensional digital photographs, we measured the Oblique Cranial Length Ratio (OCLR-2D) and Cephalic Index (CI) on the plane of maximum head circumference using a semiautomated, computer-based method [2]. OCLR-2D is the ratio between the longest and shortest oblique transcranial diameter×100 %, where the diameters are measured in a 40° angle to the sagittal midline. CI is the ratio between skull width/length. We processed and analyzed the three-dimensional images with Rapidform® 2006 3D software using custom macros written with Visual Basic for Applications (VBA). First, we used ready-made software tools to remove shoulders and other excess data and to level out possible bumps caused by cap seams. Next, we standardized the position of the head using a coordinate system based on the craniofacial landmarks endocanthion left (EnL), endocanthion right (EnR), tragion left (TrL), tragion right (TrR), and the mirror face. We constructed the sagittal reference plane (yz) with the mirror face method [48]. In short, the original facial shell and a mirror shell were registered together using the best fit technique, and the symmetry plane of the resulting structure was treated as the sagittal plane for the original face. The midpoint (Mp) was set at the intersection of the tragus connection line (between TrL and TrR) and the sagittal plane. We then defined the transverse reference plane (xz) to run through the point right in the middle of the two endocanthions (midendocanthion) and Mp perpendicular to the sagittal plane, and the coronal reference plane (xy) perpendicular to the other two planes, also through Mp. The transverse plane served as the base plane. At this point, the x-, y-, and z-axes have also been defined. After the aligning process, we defined two reference planes for analysis. The plane for two-dimensional measurements is the plane parallel to the base plane at the maximum posterior curvature in the occipital region (maximum head circumference). Finally, to prevent the ears from interfering with the
volumetric analysis, the plane for three-dimensional measurements was defined as running parallel to the base plane, immediately above the highest part of the helix of the higher set ear. To quantify cranial symmetry, we calculated the twodimensional variables diagonal difference and ear offset and the three-dimensional variables anterior cranial asymmetry index (ACAI) and posterior cranial asymmetry index (PCAI) according to Meyer-Marcotty et al. (Figs. 2 and 3) [30]. We also calculated the two-dimensional variable Oblique Cranial Length Ratio from the 3D image (OCLR-3D) in the same way as OCLR-2D from the 2D photographs, in order to assess the presence of DP as a discrete variable, as no cut-off points for plagiocephaly have been established for the previously mentioned variables [10]. We set the cut-off point for DP at OCLR-3D≥104 %. Cut-offs for moderate and severe DP were set at OCLR-3D≥108 % and OCLR-3D≥112 %, respectively [45]. We used the same cut-offs for OCLR-2D in the 2D analysis. Outcome measures The primary outcome was the presence of DP at follow-up in either the 2D or 3D analysis. Another main outcome was the severity of DP in affected infants. We assessed the rest of the variables from the 3D analysis, the cephalic index from the 2D analysis, the cervical range of motion and cervical imbalance at follow-up, and motor development as secondary outcomes. Identifying possible risk factors for cranial deformations and cervical imbalance was another secondary objective. Blinding Both researchers and parents were blinded to group allocation until after the neonatal examination. The
Fig. 2 Diagonal difference [cm]=difference between the transcranial diagonals (a, b). OCLR-3D [%]=ratio between the longest and shortest oblique transcranial diagonal (a, b)×100 %. Ear offset [cm]=distance between the tragion landmarks (TrL, TrR) on the z-axis (c)
Eur J Pediatr Fig. 3 3D measurement plane, sagittal plane, and coronal plane, dividing the cranium into four quadrants. ACAI [%]=ratio of the anterior cuboid volumes and PCAI [%]=ratio of the posterior cuboid volumes using the formula: (larger cuboid volume− smaller cuboid volume)/smaller cuboid volume×100 %
researchers responsible for obtaining and analyzing the 2D and 3D images (HA and VV) were blinded to group allocation until after the measurements. Parents in the intervention group were urged not to discuss the instructions with other parents in the maternity ward or elsewhere until after the study ended. Therapy compliance At the follow-up visit, parents completed a questionnaire about their infant’s environment and care, which served to determine their compliance with the instructions given. The questionnaire inquired about the type of feeding (breast or bottle) and number of hours the infant used a pacifier. The number of sleeping hours, position when asleep, and the type of bedding were recorded, as was time spent on the floor (e.g., on a play mat), in carriers, bouncers, and car seats. The positioning of interesting objects and toys was recorded, and any positional preference reported by the parents was noted. We calculated a mobility score (hours per day on the floor when) and an immobility score (hours per day in a carrier, car seat, or bouncer when awake and asleep) from the responses according to Cavalier et al. [6]. Statistics We calculated a sample size of 86 (43 in each group) using a 5 % significance level, a power of 80 % and a fall in the prevalence of DP from 31 to 8 % (one quarter). We compared the main outcome variables between the groups with the χ2 test, Fisher’s exact test, and the Mann-Whitney U test, as appropriate. We used the χ2 test, Fisher’s exact test, the independent samples t test, and the Mann-Whitney U test to compare secondary outcome variables, environmental
measures, and Z-scores for motor development. We compared the groups on an intention-to-treat basis. The level of significance was set at p
