Characterizing the skull base in craniofacial microsomia using principal component analysis


The aim of this study was to compare the anatomical differences in the skull base between the affected and non-affected side in patients with craniofacial microsomia (CFM), and to compare the affected and non-affected sides with measurements from a normal population. Three-dimensional computed tomography scans of 13 patients with unilateral CFM and 19 normal patients (age range 7–12 years) were marked manually with reliable homologous landmarks. Principal component analysis (PCA), as part of a point distribution model (PDM), was used to analyse the variability within the normal and preoperative CFM patient groups. Through analysis of the differences in the principal components calculated for the two groups, a model was created to describe the differences between CFM patients and normal age-matched controls. The PDMs were also used to describe the shape changes in the skull base between the cohorts and validated this model. Using thin-plate splines as a means of interpolation, videos were created to visualize the transformation from CFM skull to normal skull, and to display the variability in shape changes within the groups themselves. In CFM cases, the skull base showed significant asymmetry. Anatomical areas around the glenoid fossa and mastoid process showed the most asymmetry and restriction of growth, suggesting a pathology involving the first and second pharyngeal arches.

Craniofacial microsomia (CFM) is the second most common congenital craniofacial anomaly after cleft lip and palate . The prevalence is between 1 in 3000 and 1 in 5000 live births . CFM has a heterogeneous presentation, mainly characterized by hypoplasia in the auricular, mandibular, and maxillary anatomical regions . Many of the clinical features involve structures that arise from the first and second pharyngeal arches, thus involvement of the adjacent anatomical structures might also occur within this congenital craniofacial condition .

The aetiology or underlying cause of CFM remains a subject of discussion in the literature. The different theories include a sporadic event, disturbed migration of the cranial neural crest cells, and a hereditary role in genetics . Another hypothesis is stapedial artery disruption causing ischaemic necrosis to the anatomical features in the first and second pharyngeal arches .

The varied phenotypic presentations of CFM may be due to the wide variety of structures that arise from the first and second pharyngeal arches . The Pruzansky–Kaban classification is the most commonly used classification system to describe mandibular deformity in CFM and was used in this study . The skull base has a close relationship with the facial skeleton, and the morphology of the skull base has an influence on facial asymmetry . CFM is mainly characterized by the facial asymmetry, and thus far only one study has evaluated the skull base . This previous study found that the skull base axis was not deviated compared to those of age-matched controls and that there was little difference in morphological measurements with increasing severity of CFM .

The data contained within conventional three-dimensional computed tomography (3D-CT) scans can be utilized in mathematical techniques such as geometric morphometrics to analyse complex shapes. In this study, principal component analysis (PCA) was performed using manually landmarked 3D-CT scans to identify the global complex shape of different skulls. Using this analysis, the differences between the affected and unaffected populations could be visualized and described. This technique has been used successfully in the analysis of Apert syndrome and Crouzon–Pfeiffer syndrome .

The aim of this study was to determine the anatomical differences in the skull base between the affected and non-affected sides in patients with CFM, and to determine the differences between the CFM patients and the normal population.

Materials and methods

Data collection

Patients diagnosed with unilateral CFM with suitable preoperative 3D-CT scans available, aged between 7 and 12 years, and without any history of craniofacial skeletal surgery were included. Patients were classified with the Pruzansky–Kaban system, and those with types 1–2B were included ( Tables 1 and 2 ). Patients classified as Pruzansky–Kaban type 3 were excluded since essential anatomical features are missing, making them inappropriate for the study analysis. Patients with bilateral CFM were also excluded, since the affected sides would nullify each other during the analysis. After applying the eligibility criteria, a total of 13 patients with unilateral CFM (eight right-sided and five left-sided CFM) who had preoperative 3D-CT scans were available for analysis.

Table 1
Distribution of the CFM population by age, sex, disorder, and affected side.
Age in years Female Male CFM Right side Left side
7 1 1 2 1 1
8 2 1 3 1 2
9 0 0 0 0 0
10 2 3 5 3 2
11 2 1 3 3 0
12 0 0 0 0 0
Total 7 6 13 8 5

CFM, craniofacial microsomia.

Table 2
Distribution of the CFM population by age and Pruzansky–Kaban classification.
Age in years 1 2A 2B Total
7 0 2 0 2
8 0 3 0 3
9 0 0 0 0
10 1 1 3 5
11 0 1 2 3
12 0 0 0 0
Total 1 7 5 13

CFM, craniofacial microsomia.

The control group comprised patients with an unaffected craniofacial skeleton, aged between 7 and 12 years ( Table 3 ). Nineteen normal patients were included in the study as the control group.

Table 3
Distribution of the normal population by age and sex.
Age in years Female Male Total
7 2 0 2
8 1 1 2
9 6 2 8
10 3 0 3
11 1 1 2
12 2 0 2
Total 15 4 19

The 3D-CT scans were taken at Great Ormond Street Hospital using a 16-slice Siemens Somatom Sensation spiral CT scanner set to 0.75 collimation (Siemens Medical Solutions, Malvern, PA, USA) and at Erasmus MC using a 6-slice Siemens spiral CT scanner (Emotion 6; Siemens, Munich, Germany), with a fixed slice thickness of 0.8 mm. All scans were saved as DICOM files (Digital Imaging and Communications in Medicine) and were converted into a University College London (UCL) proprietary format. The files were loaded into 3D voxel-imaging software (Robin’s 3D; Robin Richards 2013). For all 3D-CT scans, a Hounsfield unit range of between 223 HU and 431 HU was chosen as the threshold for data imaging of the bony tissue surface. The mandible and the top of the cranium were separated and segmented off from the rest of the craniofacial skeleton to facilitate the accurate placement of the landmarks on the skull base surface.


An accurate and reliable set of homologous landmarks had to be identified for the comparison of the normal and CFM patient scans. To increase the reliability and repeatability of the landmarks, they were located at anatomical points of the skull base. An iterative process was used to test different landmark sets and to determine which distribution of landmarks best described the morphology of the skull base in normal and CFM patients. The landmarks were mainly placed around the anterior and middle skull base, as these areas are of surgical interest and are expected to be affected in CFM cases ( Figs 1 and 2 ). Thus, a smaller number of landmarks were located on the posterior skull base. It was important that the set of landmarks used captured all key shape features of the skull base.

Dec 14, 2017 | Posted by in Oral and Maxillofacial Surgery | Comments Off on Characterizing the skull base in craniofacial microsomia using principal component analysis
Premium Wordpress Themes by UFO Themes