The bony cranio-cervical junction (CCJ) is formed by the articulation between the condyles of the occipital bone and the lateral masses of the atlas, as well as by the three joints between the atlas and the axis: two between the lateral masses and the articular facets of the axis, and one between the anterior arch of the atlas and the dens.

The embryogenesis of this region is a delicate and complex process, during which numerous anomalies can occur, including: 

• Dysgenesis of the odontoid process (os odontoideum, os avis, bifid dens, or hypoplasia/aplasia of the odontoid process); 
• Dysgenesis of the basiocciput (bifid clivus or basioccipital dysplasia with basilar impression, platybasia, etc.); 
• Anomalies of the proatlas; 
• Anomalies of the first cervical sclerotome (with hypoplasia/aplasia or different possible defects of various components of C1). 

This poster will focus on anomalies of the proatlas, which lead to the presence of some interesting anatomical variants. During embryogenesis, the CCJ is formed from the occipital somites and the first three cervical somites. The caudal part of the fourth occipital somite and the cranial part of the first cervical somite form a structure called the proatlas, from which the basion and the apical segment of the odontoid process originate medially, and the occipital condyles and part of the opisthion laterally. Ventrally to the proatlas a mesenchymal structure called the hypochordal bow (or hypochordal arch) is formed, which normally becomes the anterior clival tubercle of the basiocciput (Pang and Thompson 2011).

When incomplete regression of the hypochordal bow occurs, manifestations of the “occipital vertebra” may appear: if it persists entirely, it is referred to as a pre-basioccipital arch, otherwise partial persistences may be found, with small bumps or actual bony processes, which depending on their position, are named third condyle, basal process and paracondylar processes (Pang and Thompson 2011; Prescher, Brors, and Adam 1996).

Anomalies in ossification processes can also lead to the presence of an additional variant, represented by the presence of an epitrasverse process of the atlas. The basal, paracondylar, and epitrasverse processes may all be unilateral or bilateral (Shapiro and Robinson 1976).

Unlike the aforementioned processes, which represent incomplete assimilation of the first cervical sclerotome, alterations in segmentation processes may lead to varying degrees of atlanto-occipital fusion. 

We decided to report two additional relevant anatomical variants, the ponticulus posticus and ponticulus lateralis: though not part of the “occipital vertebra” spectrum, they are a “must-know” for the radiologist as they may have significant clinical implications.