Wormian bones and nonsyndromic craniosynostosis: A comparative study using computed tomography

Introduction : Wormian bones (WB) are accessory bones positioned within the sutures or fontanelles. Although several studies have associated them with genetic disorders, ethnicity, and skull deformations, their relationship with nonsyndromic craniosynostosis (CS) has not been established, especially among Brazilians. Therefore, this study was conducted to analyze the incidence and association of WB with nonsyndromic CS in Brazilian children. Methods: An observational and retrospective study was conducted using computed tomography with 3D reconstruction in Brazilian children aged <3 years to compare the incidence and number of WB between regular children (group 1) and those with nonsyndromic CS (group 2). Results: A total of 140 children, comprising 62.9% boys (p < 0.001) with a mean age of 8.78 months, were included in this study. The most common types of CS were trigonocephaly (34; 48.6%), scaphocephaly (25; 35.7%), anterior plagiocephaly (5; 7.1%), posterior plagiocephaly (3; 4.2%), and brachycephaly (3; 4.2%). WB were more common in regular children (41; 58.6%) than in children with CS (28; 40.0%) (p = 0.028). Regular children had an average of 2.12 WB versus 1.32 WB in children with nonsyndromic CS (p = 0.024). Conclusion: The significantly less incidence of WB in Brazilian children with nonsyndromic CS was quite different from that reported in other studies and could indicate the tendency of this disease to have disturbance in ossification of the cranial vault, the impact of ethnicity, and probably the lack of additional compensatory skull growth.


INTRODUCTION
Wormian bones (WB) or intrasutural bones are isolated ossicles with variable size, thickness, and shape, which can be found in the sutures or fontanelles of the skull. They received this eponym by Thomas Bartholin to honor the Dutch anatomist Olaus Wormius [1], despite the fact that the first description of WB was attributed to Paracelsus (1460-1541), who named a bone located in the posterior fontanelle termed as "ossiculum antiepilepticum" [2,3]. Such bones also received an interesting designation of the "Inca bone" when located in the posterior fontanelle based on observations in Peruvian mummies [4] (Figure 1).
The prevalence of WB in the general population is variable and apparently dependent on ethnicity and age, with the highest proportion ever described belonging to the Chinese population, reaching 80% of cases. Furthermore, the pediatric population presents an average rate of WB of 53% [3,[5][6][7]. Beyond demographic factors, genetic and skull deformities have also been considered as factors associated with the incidence of WB, and the relationship with imperfect osteogenesis has already been consolidated [8][9][10].
Provided that WB would represent an abnormal site of ossification and could impact skull growth, it was surmised that their presence could be altered in patients with Figure 1-Different types and distribution of WB are represented on the CT scan with 3D reconstruction. WB are observed on the bregma (*) in a newborn with trigonocephaly (A) and simultaneously on the lambda termed as "Inca bone" (thin arrow) and the lambdoid suture (arrowhead) in a healthy child (B). Numerous small WB were observed in a normal child in the lambdoid suture bilaterally (C). An Inca bone is observed in a child with scaphocephaly (D). craniosynostosis (CS). For this purpose, some studies have correlated a higher incidence of WB with nonsyndromic CS, which is a disease that progresses with the premature fusion of one or more skull sutures and affects approximately one each of two thousands of live births [11][12][13]. Sanchez-Lara et al [14] found that the incidence of WB is approximately 3.5 times higher in patients with CS. However, given the wide variation in the incidence of WB among the global population, it is not clear whether this correlation is reproducible. Due to this lack of knowledge, it is important to clarify the incidence of WB in Brazilian children with and without nonsyndromic CS. Therefore, this study was conducted to evaluate the correlation between the prevalence of WB in normal Brazilian children and those with nonsyndromic CS and to describe the patterns of the position and quantity of WB.

Study design and eligibility criteria
Following the ethical board approval under the protocol (30620120.7.0000.5134), a retrospective study was designed to examine data from the medical records of patients aged <3 years and attended by the pediatric or pediatric neurosurgery at a single private hospital between January 2016 and December 2019.
For the study design, the following two primordial groups were formed: a group consisting of previously healthy children who were attended at the emergency department due to minor head trauma with indication to undergo computed tomography (CT) scan of the skull (group 1) and another group consisting of children with nonsyndromic CS who underwent surgery at the same single institution by the pediatric neurosurgery team (group 2).
The following patients were excluded from this study: those who were attended due to head trauma who presented any intracranial hemorrhages or skull fractures; patients with abnormalities not related to the trauma, such as cysts, ventriculomegaly, or brain dysmorphology; and all patients with osteogenesis imperfecta and syndromic CS such as Crouzon and Apert syndrome.

Data collection
The following variables were analyzed in this study: (1) demographic characteristics, (2) the type of nonsyndromic CS, and (3) the presence, position, and number of WB.
Patients' CT scans were obtained in the DICOM format and transferred to the OsiriX imaging program (OsiriX, Geneva, Switzerland) [15], which was validated using ANVISA to assure the use of medical findings. The images were evaluated by the senior author of this study who assured the normality of the brain and analyzed the surface of the skull using the 3D reconstruction images.
The significance level was set at 5% with 95% confidence intervals. Qualitative variables were analyzed using the twoproportion equality test. The mean of the incidence rates between the case and control groups was compared by the analysis of variance.

Study population
A total of 140 patients were evaluated (70 normal children and 70 children with CS). Regarding demographic analysis, the mean age of the patients was 8.78 months (1 day to 36 months) (SD = 7.76), with 62.9% of them being male (p < 0.001). A statistical difference was noted in terms of age, with the normal group being older with a mean age of 11.06 months versus 6.53 months for the group with CS (p = 0.047).
Regarding the quantity of WB, group 1 showed a higher incidence, with a mean value of 1.32 compared with 2.12 in group 1 (p = 0.024). The vast majority of WB was found in the lambda "Inca bone" in group 2 or the unilateral lambdoid suture in group 1. The bregma was the most unusual site to have WB and more common in patients with trigonocephaly ( Figure 1) (Table 1).
No statistical difference was observed between the types of CS in terms of WB quantity (p = 0.137), despite a trend toward more WB in patients with trigonocephaly. Overall, WB were infrequent and less numerous than those in healthy children.

DISCUSSION
WB are accessory cranial bones that develop from independent ossification centers within sutures and have been described in an average half of normal children [16,17]. Similarly, in the present study, the overall incidence of WB in Brazilian healthy children was 58.6%, and the lambdoid suture was found to be the primary site of these bones under normal condition (43.9%); no WB were observed on the coronal suture. These findings were different from those observed by Marti et al [2] who reported the presence of WB along the lambdoid suture in 64% of cases but similar to those observed by Jeanty et al [16] who conducted prenatal detection of WB and observed 50% of WB along the lambdoid suture or 25% along the coronal suture [16]. However, a higher incidence of the "Inca bone" was observed in children with CS, indicating that they are more susceptible to show WB in the lambdoid suture, and especially children with trigonocephaly displayed WB on the bregma. Some reports have mentioned the bregma as well as orbits as unusual sites to find WB [18,19]. Therefore, the presence of WB on the bregma and lambdoid suture may indicate abnormal ossification on these sites, affecting patients with impaired growth of metopic and posterior sagittal sutures, respectively.
Despite the qualitative differences concerning the position of WB in children with CS observed in the present study, another important finding was the markedly less incidence of WB in such patients than in regular children, showing an overall rate of 41%. To our knowledge, such an unexpected result could have two implications as follows: (1) the expression of ethnicity on the minor incidence of WB in Brazilian children with CS and (2) as a disease with impaired bone growth, CS has less propensity to develop an alternative ossification path to compensate for the restriction, assuming the role of WB as accessory bones. However, some authors consider WB as an anatomical variant or having a specific function in bone growth. According to Pryles et al. [7], the prevalence of congenital malformations and central nervous system abnormalities such as microcephaly and macrocephaly, hydrocephalus, CS, cerebral palsy, epilepsy, and even learning difficulties in a population of children with WB ranged from 93% to 100%, based on which they mentioned that WB were a true marker of developmental abnormalities of the central nervous system. However, no patients with central nervous system abnormality were found in the present study. Agrawal et al [19] reported that WB have been shown to develop in relation to the specific location where synostosis begins. In cases of unilateral synostosis, such as those occurring along the coronal or lambdoid sutures, there was an increased frequency of WB on the contralateral sides. In cases of midline synostosis, which begins along the metopic or sagittal sutures, WB were formed along the midline of the skull.
Regarding sex distribution, discrepancies have been reported in the literature, and some studies have reported increased prevalence among men [16] or no differences between sexes [14]. In the present study, a higher male prevalence was observed, with a statistical significance. However, this difference was in fact due to the relative frequencies of head trauma and CS, which are more prevalent in men. Considering this fact, a bias was observed in this result.
This study has some limitations such as the retrospective collection of data, which could have some imprecisions, and the impossibility to examine the regular population more appropriately due to some bias regarding sex distribution. However, this study has the merit of describing for the first time the difference in the distribution of WB among Brazilian children using high-quality images in a single institution, which has an adequate control of variables. Although the exact role of WB in skull growth remains a speculation and more studies would be required to clarify this topic, it is clear that the major importance of recognizing these bones in regular children is to avoid the misdiagnosis of skull fractures and exclude the pseudoclosure of the anterior fontanelle. The pediatrician must be aware of this condition and request a 3D reconstruction of CT scan to avoid unnecessary hospital admission or parents' complaint.

CONCLUSIONS
This study demonstrated for the first time the high incidence of WB in a healthy Brazilian pediatric population, which accounted for 58.6% of cases. Moreover, these bones were less common in children with nonsyndromic CS who were younger than 3 years of age compared with similar studies conducted worldwide. This result could indicate the impact of ethnicity and the loss of compensatory skull growth observed in this type of disease. Therefore, to our knowledge, WB are not a sign of skull deformation and reveal the loss of ossification ability.

Ethical approval
This study was performed in line with the principles of the Declaration of Helsinki. Approval was granted by the local Ethics Committee

Consent to participate
The patient gave consent to use his information and images for research proposes.

Consent for publication
The patient gave consent to use his information and images for publication.