PURPOSE: Amongst the craniosynostosis syndromes, patients with FGFR2 gene mutations are at increased risk of hydrocephalus and Chiari I malformation requiring early intervention and distraction of the posterior cranial vault. This is generally not seen in patients suffering from FGFR3 or TWIST mutations. We hypothesize that patients with FGFR2 mutations are more likely to suffer from growth restriction of the posterior vault when compared with patients with FGFR3 and TWIST mutations.
METHODS: A retrospective chart review of patients treated at the Children’s Hospital of Philadelphia (CHOP) from 1980 to 2009 was performed. Patients with a clinical diagnosis of syndromic craniosynostosis and known mutations of the FGFR2, FGFR3 or TWIST genes were included in the study. Patients with insufficient radiographic records were excluded. Earliest available CT scans were imported into SurgiCase 5.0 software (Materialise; Leuven, Belgium) and 3D endocranial views were obtained. Three measurements of posterior fossa dimensions were obtained: (1) The distance between each internal auditory meatus (IAM-IAM) (2) the angle formed by right IAM, midpoint of the foramen magnum, and left IAM (IAM-FM-IAM), and (3) the angle formed by the right IAM the dorsum sellae and the left IAM (IAM-DS-IAM). Two additional measurements were made based on the above calculations: (1) the length between DS and a point halfway between IAM to IAM (ANT-Length) and (2) the length between FM and a point halfway between IAM to IAM (POST-Length). Comparisons between the following subgroups of FGFR2 and non-FGFR2 patients were made: 1) all patients, 2) patients under 12 months of age, and 3) patients over 12 months of age. Individual patient subgroups were compared among themselves and with published norms for cranial base growth. Appropriate statistical tests were performed.
RESULTS:
The study included a total of 68 patients--41 patients with FGFR2 and 27 with non-FGFR2 (10 TWIST, 17 FGFR3) gene mutations. In patients under twelve months of age, no statistically significant differences existed between the two groups of patients.
In the subgroup of patients older than 12 months of age, several parameters were significantly different. The IAM-IAM distance was smaller in FGFR2 patients than non-FGFR2 patients (mean 40.2 mm vs. 49.10 mm, p =0.002). The IAM-FM-IAM angle was more acute in patients with FGFR2 mutations than non-FGFR2 patients (mean 99.13 vs. 107.42 degrees, p<0.001); and the IAM-DS-IAM angle was more acute for FGFR2 than non-FGFR2 patients (mean 83.89 vs. 90.47 degrees, p = 0.012). The measurements of cranial base length were not significantly different although the FGFR2 patients trended towards shorter ANT-Length (22.55 vs. 24.59mm, p = 0.10) and POST-Length 17.13 vs. 18.03mm) p = 0.18.
CONCLUSION: This data suggests that, when compared with patients with FGFR3 and TWIST mutations (and established norms), patients with FGFR2 mutations are more likely to suffer from horizontal growth restriction of the posterior fossa. Furthermore, these discrepancies in posterior fossa growth develop after one year of life.