Objective. The management of infants with posterior plagiocephaly has been controversial both because of widely differing estimates in the literature of the relative frequencies of true lambdoidal synostosis vs positional molding and because of divergent approaches to treating this problem in different institutions. Based on our experience, we hypothesized that the vast majority of children with posterior plagiocephaly did not have true synostosis and that the cosmetic impairment in such patients could be effectively treated with nonsurgical modalities.
Methods. Between 1992 and 1995, we prospectively applied in 71 infants a consistent management philosophy for these malformations that has incorporated a detailed evaluation of sutural anatomy as the basis for a physiologic approach to treatment. This approach has been directed at distinguishing true synostosis from deformational plagiocephaly and at avoiding surgery for patients with deformational abnormalities by using a combination of nonsurgical modalities to restore normal cranial growth dynamics. All children first underwent skull radiographs to determine whether the lambdoidal sutures were patent. In equivocal cases, computed tomography was also performed. Patients without true synostosis were enrolled on a course of positional therapy. In patients that did not improve after 2 to 3 months, a custom-fitted orthoplastic molding helmet was applied to facilitate passive skull recontouring.
Results. Forty children had patent sutures based on skull radiographs, and 29 others, in whom the radiographs were equivocal, had open sutures based on computed tomography, thus establishing the diagnosis of deformational plagiocephaly in 69. Predisposing factors for this deformity included a strong positioning preference during early infancy (n = 67), torticollis (n = 10), prematurity (n = 6), and developmental delay (n = 2). Only two patients had true lambdoidal synostosis; in each case, this was associated with synostosis of the posterior sagittal suture and was managed effectively with cranial reconstructive surgery. Thirty-five patients with deformational plagiocephaly had a dramatic improvement in their cranial contour with positional therapy alone; 34 patients failed to improve and were treated with molding helmets. All but five children, each of whom was more than 6 months old at initial intervention (P < .025), developed a normal or nearly normal head shape with these measures.
Conclusion. The vast majority of children with posterior plagiocephaly do not have true synostosis and can be effectively managed by nonsurgical means. The impact of positional preference on the development of this process is discussed.
- CT =
- computed tomography
The management of patients with posterior plagiocephaly has long presented a challenge both for the primary care physician involved in making the initial diagnosis and the craniofacial team involved in prescribing a treatment plan. This deformity may result from or be exacerbated by a variety of factors, such as a restrictive intrauterine environment, torticollis, a strong preference for sleeping position in early infancy, and neurologic impairment.1-6 The approaches that have been used for managing this deformity have varied widely in different centers, which has complicated efforts to formulate reliable guidelines for diagnostic evaluation and treatment. In particular, the issue of which children require operation has often rested as much on institutional bias as on objective factors. For example, in some centers, moderate to severe posterior plagiocephaly has been treated with limited craniectomy techniques7 or posterior calvarial reconstruction8-11 performed early in infancy in the hope of optimizing cranial appearance. However, pathologic studies have indicated that the majority of patients who undergo operation for this deformity have narrowing rather than true synostosis of the affected lambdoidal suture. This observation has encouraged several groups to attempt to treat the deformity using nonsurgical means.1,4-6,12,13
Advances in imaging technology coupled with refinements in surgical as well as nonsurgical modalities have facilitated efforts at our institution to standardize the management of young children with synostosis and posterior plagiocephaly. We prospectively applied a consistent management philosophy in a series of 71 children who were treated between 1992 and 1995. The overall goal of this approach was to avoid subjecting the child to operation for deformational (nonsynostotic) deformities and, instead, to rely heavily on nonsurgical means to restore normal cranial growth dynamics. Patients were initially treated with physical therapy and positional therapy to encourage normal neck mobility and to discourage continued dependency on the flattened area. Patients who failed to improve with these approaches were then treated with a molding orthoplastic helmet to encourage the growing skull to adopt a more normal contour. The rationale for using such an approach is based on several factors. First, because 85% of postnatal skull growth occurs during the first year of life, the helmet, if fitted appropriately, can harness this property to gradually mold the developing skull into a more normal configuration. In many ways, this technique resembles the practice of cranial recontouring that was used in a variety of cultures to achieve a distinctive cranial shape.4,13-17 The power of such an approach to gradually remodel the calvarial appearance is illustrated strikingly in anthropologic and paleontologic studies. Second, the use of such a conservative strategy does not preclude the performance of surgery later in infancy or early childhood, but if successful clearly diminishes the potential morbidity involved in treating this problem. The results of this study validate the efficacy of this approach and provide insights into possible ways of minimizing the incidence of posterior plagiocephaly.
From July 1992 to March 1995, 71 children younger than 1 year were referred to our institution for evaluation and treatment of posterior plagiocephaly, presumed to be secondary to lambdoidal synostosis. Thirty-seven were younger than 6 months at initial referral; 34 were between 6 months and 1 year. All children underwent plain radiographs of the skull to define the sutural anatomy. In patients with sutures that all were clearly patent, no further evaluation was performed. However, in those children with sutures that were visualized poorly or apparently closed, computed tomography (CT) scans of the brain, calvarial vault, and cranial base, supplemented by three-dimensional reconstructions, were routinely performed. For children with plagiocephaly in the absence of synostosis (deformational plagiocephaly), a careful assessment was made of the following factors: 1) the timing at which the deformity was detected (eg, at birth or postnatally) and at which therapy was initiated; 2) the association of various predisposing factors, such as prematurity, torticollis, and developmental delay; and 3) the relation between positioning preferences of the parent and/or infant and the cranial deformity.
Our overall management philosophy was based on the premise that children with deformational molding (eg, those with significant skull deformity but without synostosis), if detected early in infancy, should generally be managed without surgery. Thus, children with posterior plagiocephaly in the absence of true synostosis were initially managed with positional therapy. Parents were instructed to keep the child from lying on the flattened occipital region by using a combination of sandbags or pillows to keep the child from rolling onto the deformed side. In patients with associated torticollis, physical therapy was also initiated to encourage gradual stretching of the neck musculature and increased range of neck movements. Particular attention was focused on keeping the child from continuing to lie on the flattened occipital region. These therapies were continued for a 2- to 3-month period.
For patients in whom the above measures were ineffective in correcting the cranial deformity, a molding orthoplastic helmet was fashioned to facilitate skull growth in the areas of flattening (Fig1, A and B). The cost of the helmet was approximately $400.00. This was constructed by first making a plaster mold of the patient's skull. The mold was used to create a plastic model slightly larger than the patient's calvarium, which was custom-fitted with inserts to provide gentle pressure to the areas of maximal protuberance. Areas of skull flattening were maintained largely free of compression to encourage passive expansion of the calvarium in these areas during the course of subsequent cranial growth. The helmet was worn for approximately 22 hours per day to provide a consistent stimulus to calvarial growth. The scalp was checked several times daily for evidence of irritation or abrasions. The patient was reexamined every 4 to 6 weeks after beginning helmet therapy to determine whether the helmet needed to be refitted or recontoured internally to adjust to the child's continued skull growth. Helmet therapy was discontinued after a symmetrical calvarial contour had been established.
Surgical intervention was recommended for patients with objective evidence of lambdoidal suture synostosis or for those with severe deformational plagiocephaly that did not respond to conservative measures. This consisted of resection of the involved suture and perisutural bone with reconstruction of the surrounding calvarium to achieve an appealing cranial contour.
To evaluate the results of our approach and to identify patients who had unacceptable outcomes, we assessed outcome in two ways. First, the parents of each of the affected children were questioned closely regarding their satisfaction with the child's appearance, particularly whether they believed the child's craniofacial features were 1) normal or nearly normal, 2) abnormal but acceptable, or 3) unacceptable and in need of further intervention. Second, members of the craniofacial team (I.F.P. or H.W.L.) assessed outcome using a similar scale. In cases involving a discrepancy between the opinions of the craniofacial team and the patient's family regarding the cosmetic outcome, the least favorable response was recorded as the final outcome.
In 40 of the 71 patients referred for evaluation, skull radiographs clearly demonstrated patency of all the cranial sutures. In 31 patients, the skull radiographs were inconclusive and CT was performed; this demonstrated patent lambdoidal sutures in 29 children, 4 of whom had sclerosis and thickening of the bone adjacent to the suture without obvious closure of the suture. Only two children had radiographic evidence of true lambdoidal synostosis. In both cases, this was bilateral, although asymmetrical in extent, and associated with clear-cut posterior sagittal synostosis.
The 69 patients with nonsynostotic (deformational) plagiocephaly all exhibited a characteristic craniofacial appearance. The ear on the affected side was typically displaced away from the site of maximal occipital flattening and thus was positioned anteriorly, inferiorly, and laterally in comparison to the contralateral side. The contralateral occipital region was characteristically protuberant, and in most cases, the ipsilateral frontal and temporal regions were also prominent and the contralateral frontal region was somewhat recessed, which gave the head a parallelogram18 shape (Fig2, A and B). In some cases, the face was also affected, with prominence ipsilateral to the side of occipital flattening and canting of the nose toward the contralateral side.
In 24 of the 69 infants with deformational plagiocephaly, the calvarial asymmetry was apparent at birth. Thirty-five children were noted to develop asymmetrical occipital flattening during the first 2 months of life. In 10 patients, the parents had not been aware of the deformity until it was noted by a health care provider or other nonfamily member, and thus were not certain of the time of onset of the deformity. In 35 children, the deformity was reported to clearly have worsened between the time it was noted initially and the time of referral. Thirty-five children were younger than 6 months at the time of referral and initiation of therapy; 34 were between 6 months and 1 year.
Ten patients had associated torticollis that had been noted shortly after birth. Eight patients had developmental delay as a result of prematurity (n = 6) or Down syndrome (n = 2). Sixty-seven of the families reported that the child had predominantly been positioned supine since birth; two children had been positioned prone. In each of the children, a strong side-to-side positional preference was reported, such that the child was noted to lie virtually always on the flattened region of the skull (right = 49; left = 20).
Thirty-five children were managed strictly with positional therapy and physical therapy and developed a normal or nearly normal appearance after 2 to 3 months of treatment, as assessed by both the family and the surgeon. Thirty-four patients with significant residual deformity after the aforementioned approach were placed in a molding orthoplastic helmet. All but five achieved a normal or nearly normal appearance. Each of these five children was older than 6 months at initial referral and initiation of therapy (P < .025, Fisher's exact test). In addition, two of the five children with significant residual deformity had Down syndrome. In both cases, the head shape was improved by the helmet, but was clearly abnormal as judged by both the family and the surgeon, and might potentially have benefitted from operative intervention; however, the surgeon (and families) did not favor cranial reconstruction because these children each had a severe associated cardiac anomaly. Two other patients with suboptimal responses had severe torticollis in conjunction with profound flattening of the left posterolateral hemicranium. In both patients, the appearance was improved but not normalized by physical therapy and helmet therapy. Finally, one patient with perisutural sclerosis showed only a modest improvement with helmet therapy. In each of the above three patients, the family was satisfied with the degree of correction achieved with the helmet, and viewed the ultimate cranial shape as abnormal but acceptable. Accordingly, they did not wish to pursue surgical therapy despite the fact that in two children, the surgeon believed that the result was suboptimal and might have been improved by operative intervention.
In the other 29 patients who underwent helmet therapy, the cosmetic results with respect to calvarial symmetry and contour were uniformly normal or nearly normal with no cosmetically significant deformity (Fig3). The only residual deformity that was commonly apparent on close inspection was subtle ear position asymmetry, which was visible when looking down at the top of the patient's head, but not apparent on other projections.
As noted previously, only two patients in this series were found to have radiographic evidence of true lambdoidal synostosis. The cranial appearance in these children differed from the characteristic features of children with deformational plagiocephaly in that the skull was somewhat brachycephalic with prominent ridging over the lambdoidal sutures. In addition, in contrast to the situation with deformational plagiocephaly, the ear position in these two patients was displaced toward rather than away from the site of maximal occipital flattening. Because both patients had synostosis of not only the lambdoidal but also the sagittal sutures, these children were treated by posterior calvarial reconstruction. First, strip craniectomies were performed that encompassed the involved sutures. After completion of the bilateral lambdoid and sagittal craniectomies, the occipital bone plate was freely mobile. A series of relaxing cuts was made in the remaining occipital bone, which allowed in situ remodeling of the posterior calvarial vault. Postoperatively, both patients have had excellent cosmetic results with no residual deformity and normal cranial growth.
The generally favorable results achieved in the aforementioned unselected series of infants with posterior plagiocephaly largely validate the efficacy of the management schema proposed. This approach achieved normal or nearly normal results in all but 5 of 69 patients with deformational plagiocephaly and a gratifying outcome in the 2 patients with true lambdoidal synostosis.
A particularly important observation of this report was that the vast majority of patients with posterior plagiocephaly have deformational changes of the calvarial vault in the absence of underlying synostosis. The routine use of CT, supplemented if necessary by three-dimensional reconstructions, for cases in whom the patency of the lambdoidal suture was in question on plain radiographs facilitated a more realistic assessment of the true frequency of lambdoidal synostosis in this population of patients. The rarity of actual synostosis among patients with posterior plagiocephaly contrasts with the situation for anterior plagiocephaly, in which true coronal synostosis is present in a substantial percentage of patients.19 A review of our recent records indicates that fewer than 20% of children with anterior plagiocephaly have a deformational rather than synostotic basis for their calvarial deformity. The implication of this observation is that the occipital region may be inherently more susceptible to pressure-induced deformity than the fronto-orbital region.
In this context, it is interesting that virtually all of the children who developed posterior plagiocephaly after birth spent the bulk of their time lying on their occiput with the head angled onto the flattened side. Parents consistently reported positioning their children supine, but in most cases the child tended to rest with the head turned slightly to one side or the other. In some children, who exhibited clear-cut plagiocephaly at birth, the directional preference resulted from the tendency of the head to roll onto the flattened area. A similar situation was observed in children with torticollis. In general, parents did not attempt to counteract this positioning tendency (until they were specifically instructed to do so), which often exacerbated the initial deformity. In other children who did not exhibit flattening at birth or torticollis, the basis for the directional preference was often idiosyncratic. Many parents reported that they consistently positioned the baby on one side without realizing that this might lead to gradual deformation of the skull. Once the flattening developed, the child tended always to lie on the flattened area unless measures were taken to prevent this. In view of the recent trend toward supine positioning for infants to minimize the risk of sudden infant death syndrome,20,21 it is apparent that parents must be carefully instructed to alternate positioning from one side to the other to avoid progressive calvarial deformity. This applies particularly strongly to children with congenital torticollis, in whom the neck position preference can rapidly translate into a severe calvarial deformity.
Because most children with posterior plagiocephaly have open cranial sutures and, in theory, preservation of relatively normal cranial growth potential, there is a strong rationale to apply nonsurgical means as an initial step to treat the calvarial deformity. Moreover, because of the compensatory contralateral occipital and ipsilateral frontal growth that occurs with deformational occipital plagiocephaly, this process does not produce cranial volume restriction to inhibit brain growth, and thus the major reason for intervention is to optimize the calvarial contour to achieve an acceptable cosmetic result. Our results indicate that the majority of infants with deformational plagiocephaly will have a favorable outcome with positional therapy and, if needed, physical therapy to combat any underlying torticollis and reestablish a full range of neck mobility. Patients who do not respond to these measures will generally show a gratifying response to a course of therapy with a molding orthoplastic helmet. Although five of our patients did have significant residual deformity after a course of helmet therapy, they each nonetheless showed a noticeable improvement such that the families did not wish to pursue additional intervention.
The value of molding helmets for selected children with calvarial vault abnormalities has been previously noted by several groups.1,4-6,12,13 An important caveat in the use of these helmets for children with severe plagiocephaly is that the efficacy of the helmet depends heavily on the intrinsic growth capacity of the infant skull, which enlarges substantially during the first year of life and can be recontoured relatively easily. As the calvarium increases in size and thickness and the bones of the cranial base become severely deformed by the chronic effects of positional molding, the chances for achieving an adequate correction of the deformity without the need for extensive surgery become small. In this regard, each of the five infants in this series with suboptimal results after conservative therapy were older than 6 months when intervention was initiated. Moreover, we have noted almost no benefit from either positional therapy or helmet therapy in four other patients (not included in this series), who because of delayed referral began treatment after 1 year of age. Thus, the earlier that intervention is initiated, the more easily an acceptable cosmetic result can be achieved. With expeditious treatment, the need for surgery can generally be avoided.4-6,12,13 In some ways, these nonoperative results parallel the outcome obtained after surgical correction of sagittal synostosis. In children younger than 4 months of age, any one of a number of procedures that remove the offending suture will allow relatively normal cranial growth and ensure an acceptable cosmetic result.22,23 In contrast, in older children, much more extensive and potentially dangerous operations are required to achieve a similarly acceptable cosmetic correction.24,25These results highlight the importance of early diagnosis and prompt intervention for the aforementioned deformities to ensure a favorable outcome.
Although, in the past, early surgery was commonly recommended for children with posterior plagiocephaly, the rationale for this approach was generally based on the mistaken assumption that affected children had true synostosis or at least incipient closure of the lambdoid suture. Such a phenomenon certainly occurs in a small percentage of infants with posterior plagiocephaly,26 but in our experience the frequency of this process was much less common than predicted from previous studies. Our results demonstrate that the presence of perisutural sclerosis does not indicate incipient lambdoid synostosis, because three of four patients with this finding had normalization of their cranial shape with positional and/or helmet therapy. In contrast, the two patients with true lambdoid synostosis had unmistakable obliteration and ridging of the lambdoid suture, which was confirmed both by direct inspection intraoperatively and by pathologic examination.
In summary, our results indicate that the vast majority of infants with posterior plagiocephaly have deformational molding rather than true synostosis, and that these two entities exhibit distinct phenotypic features. Good cosmetic results can be achieved in the vast majority of patients with nonsynostotic plagiocephaly using nonoperative modalities. Operative intervention is appropriate for the small percentage of patients with true synostosis and those with deformational plagiocephaly that fails to respond to conservative measures. Because the phenotypic characteristics of deformational plagiocephaly are so distinctive from those of true synostosis and because the response to nonoperative intervention, if instituted early in infancy, is so reliable, we now (in appropriately selected patients) defer obtaining routine skull radiographs at initial evaluation, reserving these studies for the subgroup of infants that fail to respond to a 6-week course of conservative intervention.
The authors thank Susan K. Mahon, OTR/L, Susan Hom, MS, OTR/L, Pamela Judy, OTR/L, and other members of the Occupational Therapy Department at the Children's Hospital of Pittsburgh who assisted with the construction of the molding helmets used in this study.
- Received January 11, 1996.
- Accepted March 18, 1996.
Reprint requests to (I.F.P.) Department of Neurosurgery, Children's Hospital of Pittsburgh, 3705 Fifth Ave., Pittsburgh, PA 15213.
- Moss M
- ↵Pattisapu JV, Walker ML, Myers GG, Cheever J. Use of helmets for positional molding. In: Marlin AE, ed. Concepts Ped Neurosurg. Basel: Karger; 1989;9:178–184
- ↵Kast JM, Duong D, Chadduck WM. The enigma of lambdoid positional molding. Proceedings of the 44th Congress of Neurological Surgeons. Chicago, IL; 1994:70. Abstract
- McComb JG
- ↵Polin RS, Shaffrey ME, Bogaev C, Jane JA. The back of the head: a novel approach to lambdoid synostosis. Proceedings of the 23rd Annual Meeting of the Joint Section of Pediatric Neurosurgery, St Louis, MO. 1994:113. Abstract
- Keating RF,
- Goodrich JT
- ↵Levy M, McComb JG, Wells K, Gans W, Raffel C, Sloan G. Comparison of operative versus non-operative treatment of functional lambdoid synostosis. Proceedings of the 23rd Annual Meeting of the Joint Section of Pediatric Neurosurgery, St Louis, MO. 1994:71. Abstract
- Bjork A,
- Bjork L
- Blackwood B,
- Danby PM
- Rekate HL
- Rainey DY,
- Lawless MR
- ↵Gruss JS, Huang M, Roberts TS, Loeser JD, Clarren S, Mouradian WE, Cunningham MC, Cornell CJ. The differential diagnosis of posterior plagiocephaly. Proceedings of the 24th Annual Meeting of the Section on Pediatric Neurological Surgery of the American Association of Neurological Surgeons, Pasadena, CA. December 6–9, 1995:42. Abstract
- Copyright © 1997 American Academy of Pediatrics