Recent studies found evidence for an association between CM-1 and genetic disorders but failed to identify specific gene mutations related to CM-1 (102,103). In our cohort, 29 patients (11.3%) had a documented genetic disease, reinforcing the idea of a genetic susceptibility to CM-1. Some studies have reported up to 12% of familial forms among CM-1 patients series (8). In our opinion, rather than a direct relation between a specific gene mutation and the presence of CM-1, hindbrain herniation should be understood as the consequence of multiple pathophysiological conditions that either alter musculo-ligamentous balance (i.e. RASopathies, collagen diseases, etc.) or modify brain-to-bone proportion (hypophosphatemic rickets, McCune Albright syndrome, etc.). Therefore, we think that genetic susceptibility to CM-1 is multifactorial and that only few patients carry a Mendelian genetic disorder.
Craniosynostosis
CM-1 is rare (1.1-9%) among all non-syndromic isolated craniosynostosis and generally not correlated to clinical presentation (104–106), but it is more frequent (55-60%) when only lambdoid craniosynostosis is considered (107,108). It is found in more than 50% of multisuture
31 and syndromic craniosynostosis, supposedly induced by the disproportion between hindbrain growth and posterior fossa volume (109) and may be aggravated by venous hypertension induced by jugular foramen stenosis that can be associated (110). In these cases, a posterior vault expansion is often recommended as the first intention surgical treatment (110), which may resolve tonsillar herniation without the need for suboccipital decompression (108,111). As the surgical strategy for these patients is discussed in a cranial vault remodeling perspective, we decided not to include CM-1 patients with a previously diagnosed craniosynostosis in our cohort. Still, we found 17 patients (6.3%) with a non-diagnosed craniosynostosis, among which 12 (4.7%) were sagittal synostosis, three (1.2%) were bilateral lambdoid and sagittal synostosis and two (0.8%) were oxycephalies. A simple suboccipital decompression was performed in 11 (64.7%) and dura was opened in six (35.3%). Four patients (23.5%) required a second intervention due to persistence or aggravation of symptoms. Most of the patients with non-diagnosed craniosynostosis in our cohort clustered in group 2. Patients’ characteristics in this group were similar with or without including the craniosynostosis patients in the analysis. The presence of these non-diagnosed craniosynostosis in group 2 did not modify our analyze, but we emphasize the need to early identify these patients within all the CM-1 patients. Like Valentini et al. (112), we think that the realization of 3D-CT scan or black bone MRI (113) may be of interest in every CM-1 patient considered for surgery in order to correctly identify a craniosynostosis that may not be clinically obvious.
Syringomyelia
Syringomyelia is present in 23-75% patients with CM-1 (20,48). Pathophysiology is not precisely elucidated, but the underlying mechanism of formation of a cystic cavity within the spinal cord is thought to be a blockage of CSF flow at the level of the foramen magnum due to tonsillar herniation, responsible for a pressure gradient (114).
32 A higher rate of syringomyelia compared to the whole cohort was found in groups 3 and 4.
Syrinx had a trend to be wider (9,8[1-20] vs 7,7[0-16], p=0,1) and was more extended (2,28[0-3] vs 1,36[0-4], p<0,0001) in group 4 compared to group 3. We think that different mechanisms underlie the CSF blockage at the level of the craniovertebral junction responsible for the development of a syrinx within these two groups: either a clapet-like arachnoid veil obstructing the obex and the Magendie in group 3, or a crowdedness of the foramen magnum by the tonsils associated to upper cervical spine rotational instability.
Syringobulbia was rarely found in our cohort, only in 2%, but it was most common in group 2 (5.4%) which is interestingly the group with the lowest total rate of syringomyelia (29.7%). We postulate that different mechanisms underlie the development of a syringobulbia and a syringomyelia, and that syringobulbia should therefore be considered separately from syringomyelia in the analyze of the origin of CM-1. Patients in group 2 presented with a significantly higher rate of bulbar compression symptoms, especially central apneas, compared to the whole cohort (73.7% vs 26.5%, p<0.0001). Moreover, patients from group 2 had a significantly higher obex than the whole cohort. The association between a high obex position and a higher incidence of clinical and radiological bulbar dysfunction lead us to hypothesize that there is a pathophysiological correlation between these phenomenon: a lower obex position might be associated with an impaired CSF circulation at the level of the obex, whereas a higher obex might preserve CSF circulation between the fourth ventricle and the ependymal central canal but lead to a dysfunction of the bulbar respiratory nuclei (Figure 6). Whether the obstruction of the obex is a cause or a consequence of its descent through the foramen magnum is debatable.
33 Figure 6 - Speculations about the relation between obex position, CSF circulation, syringobulbia and syringomyelia – 6A: when obex is high, tonsillar herniation exerts a direct bulbar compression, possibly creating a syringobulbia, but CSF keeps circulating through the obex;
6B: when obex is low, CSF cannot circulate through the obex, which leads to the development of a syringomyelia; direct bulbar compression is less important, therefore syringobulbia is unlikely to appear. Green circles materialize the bulbar ventral and dorsal respiratory groups of the brainstem.
(CSF: Cerebro-Spinal Fluid)
Scoliosis
In our cohort, the presence of scoliosis and severe syringomyelia were highly correlated (correlation coefficient of 0.86), which had already been described (54). The risk of scoliosis has been shown to be increased with a syrinx diameter greater than 6 mm (96), which pleads for the neurogenic nature of scoliosis in these patients. This direct causal relation between spinal deformities and CM-1 had also been postulated, and some authors have shown that posterior fossa decompression in scoliosis-associated CM-1 gave the best chance for syrinx reduction and scoliosis improvement, especially before the age of 10 and below a Cobb angle of 20-30°
(115,116).
34