High-frequency ultrasound confirmation of positional plagiocephaly

Transcription

1 J Neurosurg (5 Suppl Pediatrics) 105: , 2006 High-frequency ultrasound confirmation of positional plagiocephaly JAN REGELSBERGER, M.D., GÜNTER DELLING, M.D., MICHAEL TSOKOS, M.D., KNUTH HELMKE, M.D., GERTRUDE KAMMLER, M.D., HEIDI KRÄNZLEIN, M.D., AND MANFRED WESTPHAL, M.D. Department of Neurosurgery, Institute of Osteopathology, Departments of Forensic Medicine and Pediatric Radiology, University Hospital, Hamburg, Germany Object. Positional plagiocephaly has become an increasing problem for pediatricians and craniofacial specialists. Diagnosis is commonly based on history and clinical features, but may be difficult in some cases when characteristic features are missing and radiographic studies seem to be necessary. Near-field high-frequency ultrasound has been used to evaluate the sonographic findings of suture anatomy and confirm the diagnosis of positional plagiocephaly as well as provide information of prognostic value. Methods. The authors report on 100 pediatric patients between the ages of 2 and 13 months, who were admitted to their department since 2004 with an abnormal head shape suggesting nonsynostotic plagiocephaly (NSP). Suture anatomy was examined using a 7.5-MHz linear transducer and a Siemens Elegra ultrasound scanner by two independent investigators. Measurements of suture width and bone thickness were obtained, and the findings were correlated with clinical data as well as sonographic and histopathological findings in both normal and fused cranial sutures. Interobserver variability was assessed by means of paired t-tests. Linear regression analyses were used for correlating patient age with suture width and bone thickness. Patency of lambdoid sutures was confirmed in 99 cases in which the clinical findings suggested NSP. Morphological characteristics of the sutures interosseous hypoechoic areas between hyperechoic bone plates were comparable to those of normal cranial sutures. In one patient, partial synostosis was diagnosed. Overlapping hyperechoic bone plates were found in 51 patients on the affected side of the skull and in 36 patients on the unaffected side. Suture width decreased over time from 6.5 to 2 mm, and thickness of bone in the affected area increased from 0.6 to 1.2 mm until the age of 13 months. The method was found to be limited by age (upper limit 13 months) and anatomical variations but did not show any interobserver variability (p 0.05). Conclusions. High-frequency ultrasound is a relatively inexpensive, safe, and easy-to-use tool for confirming the diagnosis of positional plagiocephaly and excluding true synostosis. Overlapping bone plates may be seen on the affected side of the skull in a majority of plagiocephalic patients, but this finding seems to have no prognostic value regarding early fusion of sutures and therefore should not affect treatment decisions. With its lack of interobserver variability and the advantage of not involving ionizing radiation, sonography has the potential to be a standard modality for investigating plagiocephaly in infants and should be offered in craniofacial outpatient clinics. KEY WORDS plagiocephaly cranial suture lambdoid suture sonography ultrasound pediatric neurosurgery S Abbreviations used in this paper: CT = computed tomography; NSP = nonsynostotic plagiocephaly. INCE international pediatric societies have adopted guidelines advocating placing infants in the supine position to prevent sudden infant death syndrome, a dramatic increase in the incidence of NSP has been observed. 4,14 The case history and characteristic clinical presentation (occipital flattening, ipsilateral migration of the ear anteriorly, and protrusion of the forehead on the side of flattening) allow a reliable diagnosis in most cases, 14 but some patients present with only minor deformities. In these cases, the differential diagnosis of true synostosis has to be considered, and further imaging investigations may be necessary. 14,19 In patients who have undergone CT studies of the head in early childhood, late irradiation effects have been reported to influence cognitive abilities and to confer a 0.5% lifetime risk of fatal cancer. 2,3,9,10,20 In addition, the sedation necessary for CT imaging in infants and young children is also associated with a risk of morbidity. Conventional plain radiographs entail less radiation-associated risk and do not require sedation, but they led to misdiagnosis in more than 13% of the patients in our cohort. Alternative diagnostic methods that are sensitive enough to identify cases of NSP but do not involve ionizing radiation or require sedation would be welcomed by physicians caring for these young patients. On the basis of many years experience using intraopera- 413

2 J. Regelsberger, et al. tive ultrasound in neurosurgery, we have established an ultrasound protocol for analyzing the anatomy of normal and pathological sutures in infants with craniofacial abnormalities. In this study, we report on 100 cases in which NSP was investigated by means of near-field high-frequency ultrasound. Clinical Material and Methods Our study group consisted of 100 consecutive children, 2 to 13 months of age, who were admitted to our department with an abnormal head shape and/or findings on plain radiographs suggestive of plagiocephaly. Clinical findings of posterior plagiocephaly were categorized according to the classification system of Argenta, 1 which allows simple and accurate determinations of the degree of deformity. Nearfield high-frequency ultrasonography (Elegra; Siemens AG, Erlangen, Germany) was routinely performed immediately after the physical examination. A 7.5-MHz linear transducer was used with the following parameters: field of view 4 cm; depth of focus 5 mm, gain 20 db; and gel used as contact medium. The ultrasound probe was held perpendicular to the plane of the long axis of the suture to obtain a precise coronal view of the suture and sufficient adjacent bone for assessment. Sonographic measurements of the lambdoid suture (suture width and thickness of adjacent bone) were obtained in the pupillary line by two independent investigators. The shortest distance between the bone plates was recorded. Distance was measured perpendicular to the suture. According to the criteria developed by Soboleski, et al., 16 and Sze, et al., 18 sutures were judged to be patent if no bridging was found and an anechoic gap was present throughout their entire length. 11 The results were correlated with the clinical data as well as with anatomical characteristics of normal and fused sutures. Statistical Analysis Interobserver variability was assessed by performing pairwise t-tests. Linear regression analyses were used to evaluate correlations between patient age and 1) suture width, and 2) bone thickness. Mean values are presented with standard deviations. Results Characteristic clinical signs of NSP, with flattening of the occipital area, were seen in 91 patients. Posterior cranial asymmetry with forward displacement of the ipsilateral ear (Argenta Type II) was seen in 48 of 91 patients. Additional frontal bone protrusion was clinically diagnosed in another 31 patients (Argenta Type III), and 12 patients showed malar deformity to some extent (Argenta Type IV). There were 55 cases of right-sided and 33 cases of left-sided NSP. Eleven patients had bilateral flattening of the occiput with (eight patients) or without (three patients) widening of the posterior skull. In nine cases, the patients had only a minor degree of head deformity (for instance, bitemporal protrusion or unilateral bossing of the parietal bone), and the clinical diagnosis remained unclear. None of the 100 children presented with the typical clinical signs of true lambdoid synostosis: downward displacement of the ear, visible when the head is observed from behind, or trapezoidal shape of the head, visible when viewed from above. In 99 cases, cranial sutures were found to be patent, manifesting as hypoechoic gaps between hyperechoic osseous plates, excluding the diagnosis of true synostosis (Fig. 1). In one patient with a minor head deformity, partial synostosis was found in a small segment of the lambdoid suture. An interosseous gap could not be identified sonographically in this area, and a degree of echogenicity characteristic of bone was found without evidence of ridging. Overlapping of the bone plates of the lambdoid sutures was seen on the affected (flattened) side in 51 patients and on the unaffected side in 36 patients. The bones were not fused but overlapped by less than 3 mm without a beveled clinical appearance. In 79% of these cases, the overlapping occurred close to the skull base. Lambdoid sutures at the skull base posterior and superior to the ear were identified with a wide-open interosseous gap. The periosteum and, less frequently, the dura mater were visible as thin hyperechoic layers in the area of patent sutures. In 91 patients, the mean width of the lambdoid suture on the affected side ( mm) did not differ significantly from that on the nonaffected side ( mm). Linear trend analysis showed increasing bone thickness (from 0.6 to 1.2 mm) and decreasing suture width (from 6.5 to 2 mm) until the age of 13 months (Fig. 2). Analysis of suture width and bone thickness measurements revealed no statistically significant interobserver variability (p 0.05). Ultrasound findings of patent sutures confirmed the clinical findings of NSP in 99 of 100 cases. In one case, we identified a partially fused suture that was not suspected on the basis of the clinical examination. None of the patients underwent surgery. In 57 cases, the infants were treated with an orthotic helmet. At the 3- to 6-month follow-up examination, cranial vault asymmetry was found to have diminished in all patients. Cranial base asymmetry had not diminished as much, even in children who had worn a helmet, and displacement of the ipsilateral ear was found to be unchanged in 36% of cases. Anatomical variations included early closure of the anterior fontanelle (one case) and varying width of the posterior fontanelle, unrelated to age (26 cases). In one patient with microcephalus and a head circumference smaller than the 3rd percentile, suture width and bone thickness remained unremarkable compared with the overall group. All sutures in this patient were found to be patent. Discussion Patients in this study were admitted to our department due to radiographic findings that were falsely thought to represent synostosis. The children were referred by pediatri- FIG. 1. Ultrasound image showing patent lambdoid suture. The interosseous suture appears as a hypoechoic gap between hyperechoic bone plates. The periosteum is visible as thin layers of soft tissue at the top of the image. 414

3 High-frequency ultrasound confirmation of plagiocephaly diagnosis Of the patients in our study, 91 presented with the typical clinical signs of NSP. In most cases, clinical examination alone is sufficiently sensitive to confirm the diagnosis. In the coronal view, occipital flattening, forward distorsion of the ipsilateral ear, and bossing of the frontal bone are visible in NSP, whereas a trapezoidal head is typical in lambdoid synostosis. In the posterior view, the ears will be at the same level in NSP. In true lambdoid synostosis a downward displacement of the ear is seen on the fused side. The other nine patients in our study were found to have temporoparietal or frontal deformities. Imaging studies are not necessary in most patients who present with signs of NSP. 13 Moreover, plain anteroposterior radiographs may be misleading because the lambdoid suture on the flattened side may appear blurred as a result of the left and right sutures occupying different planes (Fig. 3). This phenomenon was observed in four of the patients in our series who were admitted with the diagnosis of synostosis and who were therefore believed to need surgical correction prior to our investigations. The potential for misdiagnosis and exposure to ionizing radiation led us to exclude plain radiographs as a means of confirming diagnoses in our study. Sonographic characteristics of lambdoid sutures in patients with NSP appear to be comparable to the histopathological and sonographic characteristics of normal cranial sutures in autopsy specimens (Fig. 4). There is a welldefined demarcation between the calcified tissue, which appears hyperechoic on ultrasound images, and collagen suture tissue, which appears as hypoechoic gaps. This pattern was found in 91 patients with clinical findings indica- FIG. 2. Graphs showing measurements of suture width (upper) and bone thickness (lower) decreasing and increasing, respectively, with increasing age. A broad spectrum of individual measurements is seen in the first months of life. (Open circles indicate left-side measurements, filled circles indicate right-side measurements.) cians not familiar with positional plagiocephaly or brought by parents concerned about the extent of their child s head deformity and the possibility of developmental delay. This phenomenon is striking, because the incidence of NSP has dramatically increased since the beginning of the 1990s and it would be expected that information on diagnosis and treatment would have reached a wider audience, especially among clinicians who care for these children. Clinical aspects of NSP have been extensively described and may be summarized as a screw-like rotation to one side with a parallelogram-shaped head in contrast to the more trapezoidal head shape typically seen in true lambdoid synostosis. 14 The incidence rate of lambdoid synostosis is estimated to be as low as 3 in 100,000 births (0.003%). 14 The incidence rate of NSP in infancy is not known, but is expected to be much higher than the 1 in 300 births reported by Dunn in and not as high as the 48% of healthy newborns reported by Bruneteau and Mulliken in ,13 Nevertheless, it may be concluded that clinicians are more likely to see an infant with NSP than one with lambdoid synostosis. FIG. 3. Plain anteroposterior radiograph showing sharp detail of the right lambdoid suture. The left suture is blurred and was misdiagnosed as a fused suture. Arrows indicate lambdoid sutures. 415

4 J. Regelsberger, et al. FIG. 4. Ultrasound image (upper) and photomicrograph (lower) of a comparable section of a normal cranial suture. Identifiable morphological features include the periosteum above and collagenous tissue in between calcified, hyperechoic bone edges. The patent suture presents as a hypoechoic gap between hyperechoic bone plates. Toluidine blue, original magnification 200 (lower). tive of NSP as well as in the remaining nine patients in whom the characteristic clinical findings of NSP were not seen. In contrast, in sonographic images, fused sutures appear as hyperechoic bridging. This manifestation was demonstrated in a previous series of 26 patients with true craniosynostosis: there were no anechoic or hypoechoic gaps, and bridging of bone plates presented with a characteristic ridging. The histopathological sections of the resected specimens in these cases were compared with corresponding ultrasound images, thus facilitating identification of the distinguishing sonographic characteristics of patent and fused sutures (Fig. 5). In patients with NSP, overlapping of bone plates may be seen more often on the affected side of the skull due to mechanical factors, but it may also be seen in unaffected lambdoid sutures. Overlapping per se is not a pathological sign, and it may be observed in all transverse sutures (coronal and lambdoid). 12,13 Because overlapping is thought to represent stabilization of the suture or a first step of fusion, the higher percentage of overlapping bone plates seen on the affected side in patients with NSP may have prognostic value. We expect that long-term sonographic follow up of patients with untreated NSP will add important morphological information for use in cases of NSP and partial synostosis that are managed with or without surgical treatment. In patients undergoing craniosynostosis surgery, the length of suture closure can be defined precisely using ultrasound and may be marked on the skin for adapting the extent of the procedure. There was extensive interindividual variation in suture width and bone thickness among our patients, especially during the first weeks of life. This is probably due to individual developmental periods in which osseous growth is mediated through varying factors and at varying rates. 5 7 Sutures do not have intrinsic growth potential. 12 New bone FIG. 5. Ultrasound image (upper) and corresponding photomicrograph (lower) of a specimen of a fused cranial suture. Osseous bridging and the absence of an anechoic gap are clearly demonstrated sonographically, confirming the diagnosis of synostosis. Thickening of bone and narrowing of suture width are the characteristic histopathological features of fused or nearly fused sutures. The remaining suture width in this section was mm. Toluidine blue, original magnification 200 (lower). is produced at the suture edges in response to external stimuli, such as the growing brain. Cranial sutures have therefore been described as intramembranous bone growth sites. 12 With respect to the mechanical implications in patients with NSP, suture width and bone thickness cannot be expected to be linear. However, in our study group of 91 patients with NSP, the increase in bone thickness (from 0.6 mm to 1.2 mm) and decrease in lambdoid suture width (from 6.5 mm to 2.0 mm) followed linear trends. Soboleski and coworkers 16,17 found no correlation between age and suture width or age and bone thickness in 50 infants ranging in age from newborn to 5 months, most likely because of the very young ages of the infants in their cohort. Our data on lambdoid sutures are the first normative sonographic data to be made available on cranial sutures; the database will have to be completed with data for other sutures. Whether greater suture width in younger patients and linear trends of suture closure will be consistent findings in all normal sutures or overlapping of bone plates predicts abnormal early suture fusion in patients with NSP is not clear from the current database. On the basis of previous studies, it seems most likely that critical narrowing of the bone edges accelerates suture fusion. Whether biomechanical forces, as in NSP, or genetic factors, as in suture synostosis, alone may determine this process, is still unclear. We are focusing on these questions in sonographic studies underway as of this writing and 416

5 High-frequency ultrasound confirmation of plagiocephaly diagnosis should have more data on normal and pathological growth of cranial sutures in the near future. Methodological limitations of the current study include patient age and anatomical variations. Patency of sutures was confirmed in 99 patients in this study group, all younger than 13 months of age. In a previous series of 23 patients between the ages of 12 and 18 months, suture width could not be measured in six patients, and patency of sutures remained unclear in eight patients, indicating that the age range of 12 to 13 months may represent a threshold for an ultrasound analysis of sutures. Suture width in these patients was less than 0.5 mm. Widened fontanelles are of major concern in cases in which measurements in the pupillary line cross the lateral part of the fontanelle, because measurements record the lateral edge of the fontanelle rather than the suture itself. Excessive hair growth in early infancy may also be a problem in some children, but is relatively well compensated for by the ultrasound gel, even in patients older than 10 months. Computed tomography has been reported to be helpful for cases in which doubt remains after clinical examination and skull radiography, 8,14 but does not seem to be a legitimate modality for this purpose in our view. Patency is reliably confirmed by ultrasound in sutures wider than 0.5 mm. Infants who present with positional plagiocephaly of varying degrees must therefore initially be given a thorough clinical examination by an experienced physician, and sonographic investigation should be performed after this examination if necessary to resolve doubt concerning the diagnosis. Computed tomography should be reserved for the very few patients who present with complex craniofacial deformities, but the side effects of this modality must be taken into account. Lifetime risk of fatal cancer following CT has been well investigated, but adequate methods for evaluating the long-term effects of radiation exposure in infancy have yet to be developed. The protocols currently in use have been adapted from those developed in adults, and the organ radiation doses that are considered acceptable in children may in fact be too high. 3,10 The use of ultrasound instead of radiography or CT helps to minimize radiation doses, in keeping with the ALARA (As Low As Reasonably Achievable) concept. 15 Ultrasound has the additional advantages of being considerably less expensive than CT and not requiring sedation. Moreover, obliteration of lambdoid sutures and bridging of bone across the suture can be as clearly seen on sonographic images as on CT images. Conclusions In most infants with NSP, the diagnosis may be successfully made on the basis of history and clinical examination. For those in whom the diagnosis remains unclear, near-field high-frequency ultrasound represents a useful alternative to radiographic or CT studies. Patent sutures are reliably identified on ultrasound images without interobserver variability. Ultrasound imaging should therefore be an integral part of the craniofacial specialist s analysis of suture anatomy in infants who present with abnormally shaped heads. References 1. Argenta L: Clinical classification of positioning plagiocephaly. J Craniofac Surg 15: , Benz MG, Benz MW: Reduction of cancer risk associated with pediatric computed tomography by the development of new technologies. Pediatrics 114: , Brenner DJ, Elliston CD, Hall EJ, Berdon WE: Estimated risks of radiation-induced fatal cancer from pediatric CT. AJR Am J Roentgenol 176: , Bruneteau RJ, Mulliken JB: Frontal plagiocephaly: synostotic, compensational or deformational. Plast Reconstr Surg 89:21 31, Dunn PM: Congenital sternomastoid torticollis: an intrauterine postural deformity. Arch Dis Child 49: , Erasmie U, Lundberg B, Ringertz H: Measurements of skull size and width of cranial sutures in children. Acta Radiologica 23: , Furuya Y, Edwards MSB, Alpers CE, Tress BM, Ousterhout DK, Norman D: Computerized tomography of cranial sutures. Part I: comparison of suture anatomy in children and adults. J Neurosurg 61:53 58, Goldstein SJ, Kidd RC: Value of computed tomography in the evaluation of craniosynostosis. Comput Radiol 6: , Haaga JR: Radiation dose managment: weighing risk versus benefit. AJR Am J Roentgenol 177: , Hall EJ: Lessons we have learned from our children: cancer risks from diagnostic radiology. Pediatr Radiol 32: , Ngo AV, Sze RW, Parisi MT, Sidhu M, Paladin AM, Weinberger E, et al: Cranial suture simulator for ultrasound diagnosis of craniosynostosis. Pediatr Radiol 34: , Oppermann LA: Cranial sutures as intramembranous bone growth sites. Dev Dyn 219: , Peitsch WK, Keefer CH, LaBrie RA, Mulliken JB: Incidence of cranial asymmetry in healthy newborn. Pediatrics 110:e72, Persing J, James H, Swanson J, Kattwinkel J, Committee on Practice and Ambulatory Medicine, Section on Plastic Surgery, et al: Prevention and management of positional skull deformities in infants. Pediatrics 112: , Slovis TL: Children, computed tomography radiation dose, and the As Low As Reasonably Achievable (ALARA) concept. Pediatrics 112: , Soboleski D, McCloskey D, Mussari B, Sauerbrei E, Clarke M, Fletcher A: Sonography of normal cranial sutures. AJR Am J Roentgenol 168: , Soboleski D, Mussari B, McCloskey D, Sauerbrei E, Espinosa F, Fletcher A: High-resolution sonography of the abnormal cranial suture. Pediatr Radiol 28:79 82, Sze RW, Parisi MT, Sidhu M, Paladin AM, Ngo AV, Seidel KD, et al: Ultrasound screening of the lambdoid suture in the child with posterior plagiocephaly. Pediatr Radiol 33: , Vannier MW: Radiologic evaluation of craniosynostosis, in Cohen MM Jr, MacLean RE (eds): Craniosynostosis: Diagnosis, Evaluation, and Management, ed 2. New York: Oxford University Press, 2000, pp Wall BF: Radiation protection dosimetry for diagnostic radiology patients. Radiat Prot Dosimetry 109: , 2004 Manuscript received March 18, Accepted in final form July 27, Address reprint requests to: Jan Regelsberger, M.D., Department of Neurosurgery, University Hospital of Hamburg, Martinistrasse 52, Hamburg, Germany. j.regelsberger@uke.uni-hamburg.de. 417