This protocol has been compiled using the PRISMA-P (Preferred Reporting Items for Systematic reviews and Meta-Analyses for Protocols) guidelines [20] and is registered with PROSPERO (#CRD42019148060). The systematic review conformed to the PRISMA guidelines [21].

Ethics approval and consent for participation were not required for this study. This study was approved by the University of Western Cape Biomedical Science Research Ethics Committee (BM205/3).

The literature search included primary studies evaluating the morphology of the sella turcica based on lateral radiographs or via digital imaging from a lateral view, such as cone-beam computed tomography (CBCT). Only studies on postnatal human participants with genetic syndromes primarily involving the craniofacial complex were included in the review. Reviews and letters to the editor were excluded. No restriction was placed on grey literature, language, and time frame. The inclusion criteria were deliberately left broad to attempt to find as many relevant papers that relate to the primary and secondary objectives of this study as possible, as the morphology of the sella turcica may be noted as an incidental finding in some published studies.

The sella turcica is a saddle-like bony structure on the upper surface of the sphenoid bone. The most posterior part of the sella turcica is known as the hypophyseal or pituitary fossa that houses the pituitary gland [22]. It has an anterior and a posterior border, known as the tuberculum sellae and the dorsum sellae, respectively [14]. The dorsum sellae is continuous with the clivus and the posterior clinoid process. The two anterior and posterior clinoid processes project over the pituitary fossa [10,14].

As noted in the introduction, the morphology of the sella turcica is often described subjectively and qualitatively as shapes that broadly describe the shape of the sella turcica itself and its relation to the tuberculum and dorsum sellae [17]. Axelsson et al [18] created a detailed classification consisting of six morphological types of the sella turcica: (a) “normal,” (b) oblique anterior wall, (c) double contour of the floor, (d) sella turcica bridge, (e) irregularity (notching) in the posterior wall, and (f) pyramidal shape of the dorsum sellae. Andredaki et al [17] established the need for more quantitative methods to measure the size and shape of the sella turcica by using specialized software. In the review, we considered any method used to describe the morphology of the sella turcica.

A syndrome is defined as a set of symptoms occurring together or a pattern of multiple malformations thought to be pathogenetically related and not known to represent a single sequence or polytopic field defect [23]. According to Kjaer [24], a craniofacial syndrome is characterized by morphological and developmental deviations in the cranial tissue components, including the teeth. Humans with genetic syndromes primarily involving the craniofacial bony complex were included in this study. However, primary endocrine syndromes or syndromes caused by neoplasia as well as empty sella syndrome were excluded.

Imaging of the sella turcica for orthodontic and craniofacial growth assessment purposes is conventionally performed using either digital or nondigital lateral cephalometric radiographs. For this review, we also included studies involving lateral skull radiography and CBCT providing a lateral view of the sella turcica region [25]. Moreover, we also included studies providing an adequate description of the lateral view of the sella turcica with novel forms of imaging, given that measurements are correlated to a standard cephalometric radiograph.

The primary objective was to determine if genetic syndromes involving the craniofacial complex are associated with an abnormal morphology of the sella turcica based on radiographic analysis. Thus, we needed to first establish what is considered a “normal” morphology of the sella turcica. Axelsson et al [18] described the U-shaped morphology of the sella turcica as normal based on a subjective, visual assessment. Andredaki et al [17] used a software program to determine normative reference standards of the sella turcica. Therefore, we allowed flexibility in the definition of “normal” morphology as there is no accepted standard at present. If an adequate description of the “normal” reference morphology were provided in the included studies, it would be reported. Deviation from the “normal” was defined as “abnormal.”

The secondary outcome measures relate to the findings of morphological aberrations of the sella turcica in individuals with the same syndrome as well as between them and individuals with no genetic syndromes if a comparator was used in the given study. This was done in the same manner as determining the primary outcome measure and then observing whether trends exist within and between studies.

An electronic search of several databases and journals was conducted using a planned search strategy (Table 1). SCOPUS, PubMed/MEDLINE, Web of Science, ProQuest, and WorldCat, as well as all databases within EBSCOhost, were searched. Each electronic database was searched using a tailored keyword or MeSH term. The results of the search were documented, reported, and compared between databases. Where papers were found to be relevant, the reference lists were manually searched for additional relevant papers. Additional searches were conducted using the search term “sella turcica genetic syndrome” with Google Scholar, ISI Citation Indices, and Web of Science ISI proceedings (conference proceedings).

Unpublished and ongoing studies were also sought from online registries, including the NIH Health Services Research Projects in Progress, the ISRCTN registry, the UK Clinical Trials Gateway, and the WHO International Clinical Trials Platform. In cases wherein studies were found to be potentially relevant and full-text papers could not be found, the authors were contacted to gain more information and determine whether the study should be included.

Papers were screened in two stages—initially by title and abstract screening and then by full-text screening. Two reviewers conducted the screening and data extraction independently, and any failure of consensus was resolved by a senior third party. Reasons for exclusion of studies will be provided in the report. Interrater agreement was calculated and reported.

Each reviewer independently performed data extraction. An electronic data extraction form was custom-made, piloted, and amended as required for this review. The data collection form included the fields enlisted in Textbox 1.

All data generated or analyzed during this study will be included in the published study.

Each reviewer independently conducted an assessment of the study quality and risk of bias by using the risk of bias tools recommended by the Joanna Briggs Institute, for each included case study, case series, case-control study, and cohort study. Interrater agreement was then calculated. These assessments will be summarized and included in the report.

We will provide a narrative synthesis of the findings from the included studies. The narrative synthesis will be structured according to themes in the following manner, to provide a coherent summary of findings: characteristics of included studies; baseline characteristics of the population (eg, age and sex); nature of the syndrome (a brief overview of the syndromes included with a summary of craniofacial manifestations); any genetic information considered pertinent; imaging techniques used; techniques used to describe sella turcica morphology; methods of classification of sella turcica morphology; prevalence of sella turcica anomalies; patterns of sella turcica abnormalities within and between syndromes; and any additional findings. Limitations and gaps in the reporting and methodologies will be noted and discussed.

This study aimed to determine if genetic syndromes involving the craniofacial complex are associated with abnormal sella turcica morphology and to identify patterns of malformations if they exist. Current evidence suggests that there is an association between morphological abnormalities of the sella turcica and abnormalities in craniofacial development and that these abnormalities appear to be patterned among patients with particular syndromes [1]. However, no systematic review has been conducted to confirm this.

The sella turcica is considered a borderline area between developmental fields [26]. The anterior part of the sella turcica is derived from the neural crest cells, whereas most of its posterior wall and floor is derived from the notochord mesoderm [27]. Thus, abnormalities in the anterior part of the sella turcica may be associated with malformations in the facial bones, and abnormalities in the posterior wall and floor may be associated with abnormalities in the spine [27]. The sella turcica bridge between the anterior and posterior walls is also more commonly observed in individuals with severe malocclusions and other craniofacial malformations [5,28]. Owing to this complex development and the close association between the central nervous system, peripheral nervous system, and endocrine system, malformation of the sella turcica may be affected by many genetic and nongenetic factors. However, the patterning of malformations within certain genotypes indicates that genes play a central role in these malformations. Thus, attempting to understand these malformation patterns may provide us with new avenues of genetic and craniofacial pathology research.

The results of this systematic review are expected to provide a comprehensive list of morphological variations of the sella turcica, which will aid in the identification of syndromes associated with the craniofacial complex. It may also provide insights into patterns of craniofacial malformations that may be genetically linked, thus adding to the body of evidence relating to genetics and craniofacial malformations.