Radiographic assessment of the influence of metopism in frontal sinus morphology – a systematic review

Objective: This systematic literature reviews aimed to revisit radiographic observational studies the find out the potential influence of metopism in frontal sinus (FS) development. Methodology: Medline, Scopus, SciELO, LILACS and Web of Science databases were searched. Only cross-sectional and case-control original studies were eligible. Two reviewers independently participated in each phase of this review. General, methodological and outcome-related information of each study were extracted. Risk of bias was assessed with the Joanna Briggs Institute (JBI) Critical Appraisal tool for observational studies. Based on the limitations of each study, recommendations were made following EQUATOR/STROBE guidelines. Results: Depicted in an adapted PRISMA flowchart, three-hundred and sixty-five studies were identified and seven (published between 1959-2019) remained for qualitative analysis. Without considering studies with overlapping samples, five-hundred and nine skulls with metopism were detected. FS aplasia reached a prevalence of nearly 13%. Four studies (three with overlapping samples) observed a tendency of FS underdevelopment in skulls with metopism, while three studies suggested the opposite. Higher risk of bias was found in the latter. Conclusion: The available and most recent scientific literature points toward an eventual influence of metopism in FS underdevelopment. Due to the study of overlapping samples and the lack of standardized reporting protocols of methods and outcomes, however, the evidence extracted in the present systematic literature review does not confirm the association of FS underdevelopment and metopism.        


Introduction
The articulating system of the human skeleton may be classified in groups according to function as synarthroses, amphiarthroses and diarthroses (Moore et al., 2017). These groups indicate that articulating bones have slight (or non-existing), moderate and broad movement between each other, respectively (Schünke et al., 2014). Particularly in the synarthroses, most of the joints are fibrous, such as the cranial sutures (Totora & Nielsen, 2017). Depending on the edges of the articulating bones and the arrangement of their interface, sutures may be planebetween bones with smooth linear surfaces (e.g. intermaxillary suture), serratedbetween bones with interdigitation (e.g. sagittal suture), overlappingwhen the edge of one bone is covered by the edge of the other (e.g. squamosal suture), and schindylesisa type of suture that fits bones ridge-to-groove (e.g. vomero-sphenoid suture) (Craigie, 1838;Rawlani & Rawlani, 2013).
The metopic suture (MS) represents one of the six primary sutures of the infant skull (Miura et al., 2009). From Nasion (intersection between internasal and frontonasal sutures) to the anterior fontanelle (latter replaced by the Bregmaintersection between coronal and sagittal sutures), the MS describes an indentate midsagittal pathway dividing the frontal bone in half (Skrzat et al., 2004). Early in life, this suture plays an important part in enabling the development of the encephalus together with other sutures and fontanelles (Lipsett & Steanson, 2019). Additionally, because it is anatomically displayed in the anterior neurocranium, the MS may have a secondary role in the symmetrical development of the upper third of the face (forehead). Whereas most of the cranial sutures reach synostosis during adolescence and young adulthood, the MS has estimated closure interval between three and nine months after birth (Vu et al., 2001). Metopism occurs when persisting MS are found in adults skulls (Bademci et al., 2007;Zdilla, et al., 2018;Vinchon et al., 2019).
Studies have pointed prevalence rates of metopism up to 7% depending on population (Aksu et al., 2014). However, statistics might vary not only based on ancestry, but also based on individual's systemic condition. Syndromic patients or individuals with preexisting morphological alterations (anatomical variation, anomaly or monstrosity), such as severe dentomaxillofacial discrepancies (Nelke et al., 2014) might present metopism more often.
Obscurity remains on the influence of metopism in morphological alterations of the frontal sinuses (FS) . These are cavities bilaterally present in the frontal bone split by an intersinus septum, beneath the anterior cortical plate .
Didactically, FS belong to the four types of paranasal sinuses, together with maxillary, Development, v. 9, n. 10, e5719108993, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i10.8993 5 sphenoidal and etmoidal air cells (Silva et al., 2014). In practice, the FS have clinical and forensic applications. In the former, they reduce the weight of the skull and participate in immune systems by producing mucus to be distributed into the upper airways (Fahrioglu et al., 2018). In the latter, the distinctive morphology of the FS may be used for human identification and anthropological investigations (Silva et al., 2008;Silva et al., 2009;Silva et al., 2013;Silva et al., 2017;Furtado et al., 2018). What clinical and forensic applications have in common is the study of the FS by means of radiographic qualitative or quantitative assessment as a non-invasive approach to exam sinus integrity, outline, volume and content (Beaini et al., 2014;Sahlstrand-Johnson et al., 2011;Ribeiro, 2000).
Compiling studies to better understanding the potential influence of metopism in morphological alterations of the FS is a current need in science. This study revisited original articles that radiographically assessed FS in individuals with metopism in the search for potential associations. The hypothesis addressed in the present systematic review was that FS morphology is indeed affected by metopism. In order to test the hypothesis, methods and outcomes of these eligible studies were analyzed to find out the answer for the following question: "What is the available evidence to affirm morphological alteration of the FS influenced by metopism in radiographs of the human skull?"

Study Design
Secondary data was assessed in this study. Ethical clearance was obtained. The nature of the study was qualitative, and consisted of a systematic literature review (Pereira et al., 2018). The review was designed following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (Moher et al., 2015) and based on Cochrane (Higgins & Green, 2011) guidelines.

Eligibility Criteria
Studies exclusively dedicated to the investigation of metopism and its influence on the development of the FS by means of imaging were included. Studies with retrospective or prospective sampling from image databases were included. Literature reviews (systematic or not), case reports, case series, letter to the editor, editorials, abstracts in proceedings of Research, Society and Development, v. 9, n. 10, e5719108993, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i10.8993 6 conferences and unpublished academic material were excluded. Original observational studies that did not state clearly their intention to investigate the influence of metopism on FS morphology were excluded as well. Finally, studies that clearly investigated samples of individuals with developmental disorders, systemic diseases/conditions and history of trauma also fit to the exclusion criteria (to avoid the bias of condition over FS anatomy).

Research and Sources
Population (P), variables (V) and outcomes (O) retrospectively represented by human skull, metopism and radiographic alterations of FS morphology were used to establish the research question previously presented.
Medline, LILACS, SciELO, Scopus and Web of Science were searched as primary data sources. A search string of keywords (predominantly built of Medical Subject Headings [MeSH] and Descriptors in Health Sciences [DeCS]) and Boolean operator ("AND", "OR" and "NOT") was built. The terms were clustered in two main groups with variations based on their plural forms. The first, related to the MS, namely "metopic suture", "metopism", "interfrontal suture" and "median frontal suture"; the second, related to the FS, namely "frontal sinus" and "paranasal sinus". The search was performed on November 4th 2019.
Restriction of the year of publication was not applied, while language of publication was restricted to English, Spanish and Portuguese.
The obtained studies were imported in Mendeley™ (Mendeley Ltd., London, UK) software for the removal of duplicates. Eventually remaining duplicates were manually removed.

Study Selection
Two main trained reviewers were enrolled during study selection. In every selection phase, a third reviewer was included to enable consensus in case of disagreement. The first phase of study selection consisted of the title reading. The reviewers read and excluded titles that were not related to the research topic. They were not blind for authorships and journals.
Next, the abstracts were read and excluded based on eligibility criteria. Studies with missing abstracts proceeded to the next phase without exclusion. The third phase established full-text reading and exclusion. Studies excluded in this phase had the reason for their exclusion registered. Research, Society and Development, v. 9, n. 10, e5719108993, 2020 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v9i10.8993 7

Data Extraction
Were extracted from each study publication data, such as authorship, year and country of publication, name of the article and name of the journal. Methodological information consisted of sample size, sample distribution for individuals with metopic suture, distribution of males and females, age interval (minimum and maximum age), type of image exam, imaging device, number of examiners and type of analysis (quantitative or qualitative).
Extracted data related to outcomes were atrophy, hypertrophy or agenesis of the FS (right, left or bilateral).

Risk of Bias in Eligible Studies
The Joanna Briggs Institute (JBI) Critical Appraisal tool for assessing the risk of bias in observational cross-sectional studies was used (Munn et al., 2015). The eligible studies were rated based on their percentage of positive answers to the questions inherent to the JBI tool. High risk of bias was found when less than 50% of positive answers occurred. Moderate risk of bias ranged between 50 and 69% of positive answers, while low risk of bias was found in studies with more than 70% of positive answers. Given the possible limitations of the studies, recommendation will be provided for improvement of future investigations following the Enhancing the Quality and Transparency of Health Research (EQUATOR) and Strengthening the Reporting of Observational Studies in Epidemiology (STROBE).

Risk of Bias of Eligible Studies
Three studies reached low risk of bias with 100% of positive answers (Guerram et al., 2014;Sandre e ta l., 2017;, two studies reached low risk of bias with 75% of positive answers (Nikolova et al., 2016;Nikolova & Toneva, 2019) and two studies had high risk of bias with 37.5% of positive answers (Marciniak & Nizankowski, 1959;Bilgin et al., 2013) (Table 5). Based on the limitations found within each manuscript, recommendations for the improvement and standardization of reporting methods and outcomes were established ( Figure   2).

Discussion
The uncertainty surrounding the influence of metopism (persistent metopic suture in adults) in the abnormal development of the frontal sinuses dates at least from 61 years ago (Marciniak & Nizankowski, 1959). Without clear answers in the scientific literature, this systematic review compiled the available information from articles filtered based on eligibility in order to contribute to the improvement of current knowledge.
Seven studies emerged as eligible after inclusion and exclusion criteria and deep reading of titles, abstract and full-texts. Other studies that previously appeared in reference lists of the eligible studies were not included for reasons registered in Appendix I. Some of the articles not included were not exclusively dedicated to investigate frontal sinus development in metopic individuals/skulls (Torgersen, 1950;Baaten et al., 2003) in other words, they did not establish the influence of MS in FS as primary topic and outcome of interest. The inclusion of these articles would lead to extracted and tabulated data with too many gaps and missing information. Consequently, their risk of bias (for not properly fitting to the eligibility criteria) would be high and known beforehandjustifying their exclusion. It is important to note that some studies related to frontal sinus (Welcker, 1862;Rochlin & Rubaschewa, 1934;Monteiro et al., 1957) were detected from the reference lists of eligible articles (Nikolova et al., 2016)however, they were excluded firstly based on language of publication. Another study was filtered out because it was not found full-text (Monteiro & Ramos, 1953) (the unsuccessful dedicated search for the full-text version was documented).
Discussing these methodological decisions is essential to clarify this study and point out the way taken for selecting the ideal studies.
The sample, as depicted in this review, was divided into skulls from living and skulls from osteological collections. The strategies to acquire images from both types of samples were retrospective or prospective. The first is ideally designed for the living in order to avoid the exposure to ionizing radiation exclusively for research. Hence, retrospectively revisiting image databases with radiographs, and CT and MRI scans previously taken for therapeutic (diagnostic, intraoperative and follow-up) purposes is a feasible strategy. The second is commonly addressed in forensic osteological collections, in which radiographs are prospectively taken from the skulls. All the eligible studies designed sampling under ethical protocols (not exposing the living to radiation for morphological analysis of the FS). It is important to note, that the used collections were contemporary. So, the anatomy of the human skulls were all from modern humans. Additionally, most of the studies adults. This is an important methodological aspect in order to reduce te bias of including individuals in which MS is not yet persistent, such as in young childhood. Currently, it is known that MS might close before the first year of life (Vu et al., 2001). However, some authors say that it may extend up to the first eight years after birth (Vikram et al., 2014). Picking adults guarantees a proper margin to safely sampling skulls with persistent MS.
Radiographic assessment of the FS was a pre-requisite to inclusion in this review.
Despite recently published ( More accurate measurements, however, are achieved with computed tomography or any image modality that allows three-dimensional navigation resizing proportions and distances 1:1. The two studies that used computed tomography as image modality to assess the FS diverged in relation to their final opinionone pointed to a trend of metopism manifesting in skull with underdeveloped FS (Sandre et al., 2017), while the other did not found evidence to support this the influence of metopism in underdeveloped sinuses (Bilgin et al., 2013).
Finding out whether or not metopism affects FS development was the core purpose of this systematic literature review. The four studies that presented outcomes relating metopism with underdeveloped FS (not considering overlapping samples) had low risk of bias (two of them with 100% score). In the other hand, two out of the three studies that found opposite outcomes had high risk of bias (with score 37.5%) and the only study with low risk of bias had the smallest sample among the eligible studies (Sandre et al. 2017). In general, the studies that related metopism and FS alterations indicated first hypoplasia as the most common alteration, followed by aplasia. A drawback, however, emerges because sex information might naturally influence on the size of FS, which may be smaller in females. Specific analysis of FS and metopism based on sex were hampered because detailed information were not clearly described among studies. In order to cluster individuals within groups based on size, the studies must establish techniques for the proper classification. Among the qualitative analyses, Guerram's approach was the most used, while Ribeiro's approach enabled most of the quantitative analyses. These techniques were originally designed for 2D analyses in radiographs. Validation in 3D scenarios could interesting to translate their approach to more contemporary image modalities. In practice, the 3D analysis of FS would contribute to a multiplanar view of their pneumatizationwhich may extend along the vertical and horizontal plates of the frontal bone . Although expressing clinical manifestations very discretely, metopism may play a role in determining the morphology (outline) of the skull. Understanding if this phenomenon has influence over FS development is part of the anatomical study of the human skull. In order to improve future studies, the present review suggestion a few recommendations for reporting methods on the investigation of the influence of metopism in FS development.
The methodological variances among the studies led to the recommendation of reporting guidelines based on well-known scientific standards, namely EQUATOR and STROBE. These guidelines were designed to enable homogeneity in future studies. Most of the studies in this systematic literature review did not report study design (e.g. experimental or observational, subdividing). Ethical approval was mentioned in the text of a single study (Sandre et al. 2017). Eligibility criteria for sampling was clearly defined in most of the studies but not necessarily in detail. Date of sample and interval of data collection were poorly reported, as well as the rationale behind sample size (e.g. calculation vs. convenience).
Examiner reproducibility training, calibration and testing was mentioned in three studies only ad shortly describedit is essential not only in metric studies but also for the allocation of FS based on size groups. Evidently, authors have the autonomy to design their own articles, but it is the duty of the present study to point out best practices to replace the existing limitations.
Based on the exposed, future studies are encouraged to be carried out in a systematic and standardized way so better comparisons (eventually by means of meta-analysis) can be accomplished.

Final Considerations
Outcomes of the scientific literature dedicated to verify the possible influence of metopism in FS development via radiographic analysis varied among eligible studies. Data reported among studies were heterogeneous and hampered comparative and inferential analyses. Lower risk of bias weighted in favor of studies that support a tendency of metopism in skulls with underdeveloped FS. Primary original studies with standardized protocol for reporting methods and results are encouraged to enable optimal and clear data extraction in the future.