- Email: [email protected]
A range of malar and masseteric hypoplasia exists in Treacher Collins syndrome
Journal of Plastic, Reconstructive & Aesthetic Surgery (2013) 66, 43e46
A range of malar and masseteric hypoplasia exists in Treacher Collins syndrome Kenneth R. Wong, Miles J. Pfaff, Christopher C. Chang, Roberto Travieso, Derek M. Steinbacher* Yale University School of Medicine, Section of Plastic and Reconstructive Surgery, New Haven, CT, USA Received 27 January 2012; accepted 27 July 2012
KEYWORDS Treacher Collins syndrome; Zygoma; Masseter; Volume; Asymmetry; Three-dimensional imaging
Summary Background: Treacher Collins syndrome (TCS) is a facial dysostosis, the hallmark being bilateral malar hypoplasia. The purpose of this study is to morphologically classify the TCS malar deformity and to volumetrically characterise both the TCS zygoma and masseter muscle, including for lefteright symmetry, compared to controls. We hypothesise that the TCS zygoma will be smaller than controls and zygomatic deficiency will portend masseteric hypoplasia. Methods: Demographic and computed tomography (CT) data were recorded. The CT scans were converted into three-dimensional facial renderings, and the zygomatic morphology was grossly evaluated. A classification was reported based on malar structure and presence/ absence of normal zygomaticomaxillary complex articulations. The zygoma and masseter muscles were then digitally isolated using 3-D planning software (Materialise, Leuven, Belgium). Volumes and sidedness ratios were calculated and compared using two-sided t-tests. Results: 58 sides were identified (24 TCS: 34 controls), mean age of 60.0 months and normally distributed. The phenotypic dysmorphology was graded as mild, moderate or severe (I, II and III, respectively). TCS malar and masseteric volumes were significantly smaller than controls (p < 0.0001 in both cases). The TCS zygomatic sideeside symmetry ratio was 0.66 0.28, compared to 0.97 0.02 for controls (p Z 0.002). The TCS masseteric sideeside ratio was 0.74 0.20, compared to 0.92 0.09 for controls (p Z 0.001). Conclusions: A range of zygomatic hypoplasia exists in TCS (mildesevere). The decrease in malar volume occurs in concert with masseteric hypoplasia, and the left and right sides are not equally affected. ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Craniofacial Program, Yale Plastic and Reconstructive Surgery, 3rd Floor, Boardman Building, 330 Cedar Street, New Haven, CT 06520, USA. Tel.: þ1 203 785 4559; fax: þ1 203 785 7514. E-mail address: [email protected] (D.M. Steinbacher). 1748-6815/$ - see front matter ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bjps.2012.07.028
44
Introduction Treacher Collins syndrome (TCS, or mandibulofacial dysostosis) is an autosomal dominant syndrome caused by a TCOF1 gene mutation, affecting approximately 1 in 50,000 newborns.1,2 The anomalies of this syndrome are bilateral and include zygomatic hypoplasia, downwardslanting palpebral fissures, and ear abnormalities. Characteristic mandibular retrognathia can result in airway obstruction and imparts a convex facies. Zygomatic hypoplasia and associated lid deformities are pathognomonic of TCS and are requisite to the diagnosis.3 The loss of malar structure in TCS has been previously assessed, showing an incomplete zygomatic arch as the most frequent finding, with the malar body being more often spared.4,5 Interestingly, though not often reported, a degree of lefteright asymmetry in the TCS facial structures has also been described.4 The pterygoid muscles have been shown to be deficient, perhaps as a function of attachment to a hypoplastic mandible. However, this previous report did not evaluate the masseter, which is an important masticatory muscle originating from the zygoma.4 Given this anatomic relationship, a quantitative assessment of the TCS malaremasseteric complex, using high-resolution CT data, would be beneficial. The purpose of this study is, first, to classify the TCS zygoma based on phenotypic diminution and loss of usual zygomaticomaxillary complex (ZMC) anatomic articulations. Next, we will volumetrically characterise TCS zygoma and masseter muscle versus unaffected individuals. Finally, we will document the lefteright symmetry in these TCS facial structures compared to controls. We hypothesise that the TCS zygomatic volumes will be less than norms, and that masseteric hypoplasia will proportionally diminish with worsening malar dysmorphology. We also expect a range of deformity across the TCS population and anticipate that the left and right sides will be differentially affected.
Materials and methods This was a retrospective analysis performed in concordance with the Yale University Institutional Review Board (protocol # 1101007932). Patients with confounding trauma or pathology were excluded. Demographic information and age at computed tomography (CT) scan were recorded for both TCS and control subjects. Three-dimensional (3-D) scans
K.R. Wong et al. were obtained and each individual zygoma was analysed grossly for morphologic continuity and classified into one of three groups, based on the amount and location of agenesis. The 3-D scans were then further isolated using Surgicase CMF (version 5.0.0.32, Materialise, Leuven, Belgium). The malar bones were digitally segmented from the craniofacial skeleton along exposed articulations (the zygomaticofrontal, zygomaticosphenoid, zygomaticomaxillary and zygomaticotemporal sutures). The masseter muscles, including both deep and superficial portions, were segmented using the muscle-density setting of Surgicase CMF. Volume in cubic millimetres (mm3) was obtained for each structure (zygoma and masseter). Volume was then compared between the TCS cohort and controls, with mean group differences calculated by two-tailed t-tests (p < 0.05 was considered significant). The volumetric relationship between each respective zygoma and masseter was evaluated using the Pearson correlation coefficient. The lefteright symmetry for each patient was determined by comparing the morphologic grades attributed to each side. Volumetric symmetry was calculated using an intrapatient sidedness ratio, in which the volume of the smaller side was divided by that of the larger side (for both zygoma and masseter). Mean group differences between TCS and controls were calculated here as well, by twotailed t-tests.
Results Three-dimensional CT images were obtained for 12 TCS patients and 17 control subjects (24 and 34 sides, respectively). Mean age was 60.0 months for both groups. The malar volumetric data were normally distributed, as seen by the respective skewness and kurtosis values (controls: 0.36, 0.42; TCS: 1.23, 0.54). Gender stratification showed 7 M:5 F and 10 M:7 F for TCS and controls, respectively. A range of hypoplasia and aplasia was noted among the TCS zygomata, from a small zygoma with all articulations preserved (except along the arch), which we classified as type I (mild), to near-total malar absence, which we designated as type III (severe). A type II (moderate) zygoma fell between these extremes (Figure 1). Based on this classification, we stratified our TCS cohort as follows: 17% type I, 21% type II and 63% type III. The TCS malar volumes (cohort of types IeIII combined) were significantly smaller compared to controls:
Figure 1 Examples of the Treacher Collins zygomatic deformity, shown through lateral 3-D computed tomography images. For type I (left), shape is normal but hypoplastic, typically with a cleft of the arch. The type II zygoma (centre) is misshapen, more hypoplastic and there is no extension from the temporal bone to meet the deformed zygomatic arch. The type III zygoma (right), is severely diminutive in volume and lacks malar prominence. This severe zygoma has a slight arch artefact directed anteriorly, but other type III zygomata with more aplasia only have discontinuous structural buds.
Malar and masseteric in TCS
45
861 1004 mm3 versus 2820 868 mm3, respectively (p < 0.0001). Analysing for correlation with classification type showed volumetric decrease with increasing severity score (Table 1). TCS masseters were also significantly smaller than controls: 3158 3254 mm3 versus 9395 4167 mm3, respectively (p < 0.0001), and masseteric volumetric decrement typically correlated proportionally with that of the zygoma (Table 2). The Pearson correlation coefficient (r) between malar and masseteric volume in TCS was 0.44 (p < 0.20), with a coefficient of determination (r2) being 0.20. This contrasted with r Z 0.74 (p < 0.001) and r2 Z 0.54 between malar and masseteric volume, in normal controls (Table 2). The majority of the TCS patients (83%) had matching left and right classification types. However, volumetric sidedness ratios showed that the mean TCS zygoma lefteright ratio was significantly smaller (p Z 0.002) than that of normal controls (0.66 0.28, versus 0.97 0.02, respectively) (Table 3). Similarly, the mean TCS masseter ratio was significantly smaller (p Z 0.001) than that of normal controls (0.74 0.20, compared to 0.92 0.09, respectively) (Table 3).
Discussion Zygomatic hypoplasia is the phenotypic cornerstone of TCS, both observed and qualitatively described, since E. Treacher Collins’s report of symmetrically absent malar development in 1900.6,7 Such dysmorphology typically occurs in concert with lower lid abnormalities, in line with the embryologic defect. Prior to the development of 3-D CT scans, knowledge of TCS skeletal anatomy was limited to two-dimensional (2-D) roentgenographic studies, syndromic skulls in museums and autopsies.4 These experiences and reports were limited by the non-lethality and rarity of the condition. Thus, Marsh et al. used early 3-D CT imaging techniques for their patients and found that skeletal aplasia most consistently affected the zygomatic process of the temporal bone and least affected the zygomatic body.4 This report also first mentioned the lefteright asymmetry in TCS facial bone and pterygoid muscle anatomy. Subsequently, Kay and Kay studied the zygomatic arches of 25 TCS subjects, noting that 36% of arches were unidentifiable and 24% were misshapen.5 Teber et al. defined the two minimum TCS diagnostic criteria as hypoplastic zygomatic arches and downward-sloping palpebral fissures.3 One component of our study was to examine the overall TCS malar morphology and to devise a classification scheme based on increasing severity of the deformity. Such a classification aids in standardised description and Table 1
Volume by Treacher Collins malar grade. Controls (n Z 34)
Type I (n Z 4)
Type II (n Z 5)
Table 2 Volumetric correlation of zygomata and masseter muscles. Zygomata (mm3) TCS (n Z 24) Controls (n Z 34)
2820 868
Pearson coefficient
3158 3254
r Z 0.44 (p < 0.20) r Z 0.74 (p < 0.001)
9395 4167
TCS, Treacher Collins syndrome.
communication, and facilitates surgical planning related to the TCS zygomatic deformity. Similar to previous studies, we confirmed that the minimal malar deficiency in TCS is lack of continuity along the zygomatic arch. We found an incomplete zygomaticetemporal articulation in 23/24 sides (the remaining side was spanned by a narrow osseous bridge). The next most-frequent open articulation was the zygomaticosphenoid suture, followed by incomplete connections at the zygomaticofrontal and zygomaticomaxillary regions. These recurring patterns served as the basis for our proposed classification system. The type I deformity is represented by a small zygomatic body, with presence of all articulations except for the arch. Type II is designated by diminished malar structure and absence along the lateral orbital wall (zygomaticosphenoid suture). While the inferolateral orbital rim is preserved in the moderate type, the arch is incomplete and shows no evidence of a zygomatic process from the temporal bone. Lastly, the type III deformity is distinguished by minimal volume, often entailing only remnants of malar structure at the zygomaticofrontal or maxillary region. The sequential lack of fusion in the inferolateral orbit and arch is reminiscent of arrested embryological development expected in a first and second pharyngeal arch deformity. Not surprisingly, the type III zygoma was seen most frequently in our TCS cohort, and represents the pathognomonic orbitozygomatic deformity. The absence of all normal ZMC articulations, and the near aplasia of the zygoma, accounts for the malar flattening and lid and lateral canthal malposition, classic in TCS. This would also be expected to result in a subnormal malar volume. The lack of a zygomatic arch and a hypoplastic malar body would also be anticipated to presage a diminished masseter muscle. As expected, the TCS malar volume was much smaller compared to age-matched controls. Our analysis volumetrically quantified the hypoplasia and gross dysmorphology posited in the above-mentioned classification. Considering Table 3 Symmetry comparison of Treacher Collins syndrome and controls.
Type III (n Z 15)
Avg. age 60.0 93.8 106.0 35.5 (mos) Volume 2820 868 2356 1066 1678 345 190 133 (mm3)
861 1004
Masseters (mm3)
Zygomatic ratio Masseteric ratio
Treacher Collins
Controls
Significance
0.66 0.28
0.97 0.02
p Z 0.002
0.74 0.20
0.92 0.09
p Z 0.001
46 TCS malar volumes in light of this classification showed a trend inversely correlated with increasing severity grade. Our mean age for both TCS and controls was 5 years, though the range was from infancy to older childhood. However, this age variability was conserved in both groups, which should balance any growth impact on malar volume. Additionally, compared to previous studies,4 these TCS subjects exhibited a much tighter age grouping. We also hypothesised that diminished and absent malar structure in TCS would correspond to masseteric hypoplasia (Figure 2). Indeed, in parallel with zygomatic volume, the mean TCS masseter was smaller than the mean control masseter: 3158 3254 mm3 versus 9395 4167 mm3, respectively (p < 0.0001). This is not surprising as the masseter muscle originates from the zygomatic arch and a portion of the malar body. Statistically, there was a positive correlation (r Z 0.44) between malar and masseteric volume in Treacher Collins patients (though weak). The coefficient of determination (r2 Z 0.20) estimated that only a fifth of the variation of one variable (e.g., masseteric volume) was explained by its relationship to the other variable (e.g., malar volume). This contrasted with the strong, positive correlation (r Z 0.74) and coefficient of determination (r2 Z 0.54) between malar and masseteric volume in normal controls. We expected that with an embryologic defect affecting the zygoma, the surrounding and attached structures, including the masseter, would be proportionally hypoplastic.8 Additionally, we posited that a diminished site of muscle origin (e.g., the zygoma) remains underdeveloped due in part to lack of biomechanical stimulation (i.e., the functional matrix). Our data do corroborate that both the TCS masseter and zygoma are hypoplastic, and significantly so, compared to controls. However, statistically, with these small numbers, we cannot intercede with certainty that one is because of the other. Interestingly, in a few TCS cases the smaller masseter was actually associated with the larger zygoma and vice versa. The interplay between muscle and bone appears more orderly and predictable in normal controls. Treacher Collins is frequently described as a ‘symmetric’ syndrome,9 but perhaps the more accurate term is ‘bilateral’. Phenotypically, our analyses show that there is
Figure 2 3-D CT scan highlighting the hypoplastic masseter of a TCS patient with type III zygomata.
K.R. Wong et al. reproducibility from left to right sides in the characterisation of malar deformity (83% demonstrated the same sideeside malar classification). However, volumetrically, there was significant variation. The average TCS patient exhibited 34% lefteright zygomatic volume asymmetry compared to only a 3% difference in normal controls. This sidedness ratio was similar for the masseters in TCS versus controls, thus quantifying the variable penetrance from side-to-side. This has important treatment implications because even though each zygoma may look anatomically similar in a single TCS patient, there may be significant volumetric differences that would portend differential reconstructive augmentation.
Conclusions A range of zygomatic hypoplasia exists in TCS (mildemoderateesevere), with near total aplasia being most common. The decrease in malar volume typically occurs in concert with masseteric hypoplasia, and the left and right sides are not symmetrically affected.
Conflict of interest None.
Acknowledgements We thank the Charles W. Ohse Grant for Surgical Research and the Yale University School of Medicine Office of Student Research for supporting this study. We appreciate the assistance of Kevin Callender and Elizabeth Roberto at the Yale University Statistical Laboratory.
References 1. Edwards SJ, Fowlie A, Cust MP, Liu DT, Young ID, Dixon MJ. Prenatal diagnosis in Treacher Collins syndrome using combined linkage analysis and ultrasound imaging. J Med Genet 1996;33: 603e6. 2. Jabs EW, Li X, Coss CA, Taylor EW, Myeres DA, Weber JL. Mapping the Treacher Collins syndrome locus to 5q31.3-q33.3. Genomics 1991;11:193e8. 3. Teber OA, Gillessen-Kaesbach G, Fischer S, et al. Genotyping in 46 patients with tentative diagnosis of Treacher Collins syndrome revealed unexpected phenotypic variation. Eur J Hum Genet 2004;12:879e90. 4. Marsh JL, Celin SE, Vannier MW, Gado M. The skeletal anatomy of mandibulofacial dysostosis (Treacher Collins syndrome). Plast Reconstr Surg 1986;78:460e70. 5. Kay ED, Kay CN. Dysmorphogenesis of the mandible, zygoma, and middle ear ossicles in hemifacial microsomia and mandibulofacial dysostosis. Am J Med Genet 1989;32:27e31. 6. Collins ET. Cases with symmetrical congenital notches in outer part of each lower lid and defective development of malar bones. Trans Ophthal Soc U K 1900;20:190e2. 7. Franceschetti A, Klein D. The mandibulofacial dysostosis: a new hereditary syndrome. Acta Ophthalmol Scand 1949;27: 143e224. 8. Poswillo D. The pathogenesis of the Treacher Collins syndrome (mandibulofacial dysostosis). Br J Oral Surg 1975;13:1e26. 9. Poswillo D. Otomandibular deformity: pathogenesis as a guide to reconstruction. J Maxillofac Surg 1974;2:64e72.
A range of malar and masseteric hypoplasia exists in Treacher Collins syndrome Kenneth R. Wong, Miles J. Pfaff, Christopher C. Chang, Roberto Travieso, Derek M. Steinbacher* Yale University School of Medicine, Section of Plastic and Reconstructive Surgery, New Haven, CT, USA Received 27 January 2012; accepted 27 July 2012
KEYWORDS Treacher Collins syndrome; Zygoma; Masseter; Volume; Asymmetry; Three-dimensional imaging
Summary Background: Treacher Collins syndrome (TCS) is a facial dysostosis, the hallmark being bilateral malar hypoplasia. The purpose of this study is to morphologically classify the TCS malar deformity and to volumetrically characterise both the TCS zygoma and masseter muscle, including for lefteright symmetry, compared to controls. We hypothesise that the TCS zygoma will be smaller than controls and zygomatic deficiency will portend masseteric hypoplasia. Methods: Demographic and computed tomography (CT) data were recorded. The CT scans were converted into three-dimensional facial renderings, and the zygomatic morphology was grossly evaluated. A classification was reported based on malar structure and presence/ absence of normal zygomaticomaxillary complex articulations. The zygoma and masseter muscles were then digitally isolated using 3-D planning software (Materialise, Leuven, Belgium). Volumes and sidedness ratios were calculated and compared using two-sided t-tests. Results: 58 sides were identified (24 TCS: 34 controls), mean age of 60.0 months and normally distributed. The phenotypic dysmorphology was graded as mild, moderate or severe (I, II and III, respectively). TCS malar and masseteric volumes were significantly smaller than controls (p < 0.0001 in both cases). The TCS zygomatic sideeside symmetry ratio was 0.66 0.28, compared to 0.97 0.02 for controls (p Z 0.002). The TCS masseteric sideeside ratio was 0.74 0.20, compared to 0.92 0.09 for controls (p Z 0.001). Conclusions: A range of zygomatic hypoplasia exists in TCS (mildesevere). The decrease in malar volume occurs in concert with masseteric hypoplasia, and the left and right sides are not equally affected. ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Craniofacial Program, Yale Plastic and Reconstructive Surgery, 3rd Floor, Boardman Building, 330 Cedar Street, New Haven, CT 06520, USA. Tel.: þ1 203 785 4559; fax: þ1 203 785 7514. E-mail address: [email protected] (D.M. Steinbacher). 1748-6815/$ - see front matter ª 2012 British Association of Plastic, Reconstructive and Aesthetic Surgeons. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.bjps.2012.07.028
44
Introduction Treacher Collins syndrome (TCS, or mandibulofacial dysostosis) is an autosomal dominant syndrome caused by a TCOF1 gene mutation, affecting approximately 1 in 50,000 newborns.1,2 The anomalies of this syndrome are bilateral and include zygomatic hypoplasia, downwardslanting palpebral fissures, and ear abnormalities. Characteristic mandibular retrognathia can result in airway obstruction and imparts a convex facies. Zygomatic hypoplasia and associated lid deformities are pathognomonic of TCS and are requisite to the diagnosis.3 The loss of malar structure in TCS has been previously assessed, showing an incomplete zygomatic arch as the most frequent finding, with the malar body being more often spared.4,5 Interestingly, though not often reported, a degree of lefteright asymmetry in the TCS facial structures has also been described.4 The pterygoid muscles have been shown to be deficient, perhaps as a function of attachment to a hypoplastic mandible. However, this previous report did not evaluate the masseter, which is an important masticatory muscle originating from the zygoma.4 Given this anatomic relationship, a quantitative assessment of the TCS malaremasseteric complex, using high-resolution CT data, would be beneficial. The purpose of this study is, first, to classify the TCS zygoma based on phenotypic diminution and loss of usual zygomaticomaxillary complex (ZMC) anatomic articulations. Next, we will volumetrically characterise TCS zygoma and masseter muscle versus unaffected individuals. Finally, we will document the lefteright symmetry in these TCS facial structures compared to controls. We hypothesise that the TCS zygomatic volumes will be less than norms, and that masseteric hypoplasia will proportionally diminish with worsening malar dysmorphology. We also expect a range of deformity across the TCS population and anticipate that the left and right sides will be differentially affected.
Materials and methods This was a retrospective analysis performed in concordance with the Yale University Institutional Review Board (protocol # 1101007932). Patients with confounding trauma or pathology were excluded. Demographic information and age at computed tomography (CT) scan were recorded for both TCS and control subjects. Three-dimensional (3-D) scans
K.R. Wong et al. were obtained and each individual zygoma was analysed grossly for morphologic continuity and classified into one of three groups, based on the amount and location of agenesis. The 3-D scans were then further isolated using Surgicase CMF (version 5.0.0.32, Materialise, Leuven, Belgium). The malar bones were digitally segmented from the craniofacial skeleton along exposed articulations (the zygomaticofrontal, zygomaticosphenoid, zygomaticomaxillary and zygomaticotemporal sutures). The masseter muscles, including both deep and superficial portions, were segmented using the muscle-density setting of Surgicase CMF. Volume in cubic millimetres (mm3) was obtained for each structure (zygoma and masseter). Volume was then compared between the TCS cohort and controls, with mean group differences calculated by two-tailed t-tests (p < 0.05 was considered significant). The volumetric relationship between each respective zygoma and masseter was evaluated using the Pearson correlation coefficient. The lefteright symmetry for each patient was determined by comparing the morphologic grades attributed to each side. Volumetric symmetry was calculated using an intrapatient sidedness ratio, in which the volume of the smaller side was divided by that of the larger side (for both zygoma and masseter). Mean group differences between TCS and controls were calculated here as well, by twotailed t-tests.
Results Three-dimensional CT images were obtained for 12 TCS patients and 17 control subjects (24 and 34 sides, respectively). Mean age was 60.0 months for both groups. The malar volumetric data were normally distributed, as seen by the respective skewness and kurtosis values (controls: 0.36, 0.42; TCS: 1.23, 0.54). Gender stratification showed 7 M:5 F and 10 M:7 F for TCS and controls, respectively. A range of hypoplasia and aplasia was noted among the TCS zygomata, from a small zygoma with all articulations preserved (except along the arch), which we classified as type I (mild), to near-total malar absence, which we designated as type III (severe). A type II (moderate) zygoma fell between these extremes (Figure 1). Based on this classification, we stratified our TCS cohort as follows: 17% type I, 21% type II and 63% type III. The TCS malar volumes (cohort of types IeIII combined) were significantly smaller compared to controls:
Figure 1 Examples of the Treacher Collins zygomatic deformity, shown through lateral 3-D computed tomography images. For type I (left), shape is normal but hypoplastic, typically with a cleft of the arch. The type II zygoma (centre) is misshapen, more hypoplastic and there is no extension from the temporal bone to meet the deformed zygomatic arch. The type III zygoma (right), is severely diminutive in volume and lacks malar prominence. This severe zygoma has a slight arch artefact directed anteriorly, but other type III zygomata with more aplasia only have discontinuous structural buds.
Malar and masseteric in TCS
45
861 1004 mm3 versus 2820 868 mm3, respectively (p < 0.0001). Analysing for correlation with classification type showed volumetric decrease with increasing severity score (Table 1). TCS masseters were also significantly smaller than controls: 3158 3254 mm3 versus 9395 4167 mm3, respectively (p < 0.0001), and masseteric volumetric decrement typically correlated proportionally with that of the zygoma (Table 2). The Pearson correlation coefficient (r) between malar and masseteric volume in TCS was 0.44 (p < 0.20), with a coefficient of determination (r2) being 0.20. This contrasted with r Z 0.74 (p < 0.001) and r2 Z 0.54 between malar and masseteric volume, in normal controls (Table 2). The majority of the TCS patients (83%) had matching left and right classification types. However, volumetric sidedness ratios showed that the mean TCS zygoma lefteright ratio was significantly smaller (p Z 0.002) than that of normal controls (0.66 0.28, versus 0.97 0.02, respectively) (Table 3). Similarly, the mean TCS masseter ratio was significantly smaller (p Z 0.001) than that of normal controls (0.74 0.20, compared to 0.92 0.09, respectively) (Table 3).
Discussion Zygomatic hypoplasia is the phenotypic cornerstone of TCS, both observed and qualitatively described, since E. Treacher Collins’s report of symmetrically absent malar development in 1900.6,7 Such dysmorphology typically occurs in concert with lower lid abnormalities, in line with the embryologic defect. Prior to the development of 3-D CT scans, knowledge of TCS skeletal anatomy was limited to two-dimensional (2-D) roentgenographic studies, syndromic skulls in museums and autopsies.4 These experiences and reports were limited by the non-lethality and rarity of the condition. Thus, Marsh et al. used early 3-D CT imaging techniques for their patients and found that skeletal aplasia most consistently affected the zygomatic process of the temporal bone and least affected the zygomatic body.4 This report also first mentioned the lefteright asymmetry in TCS facial bone and pterygoid muscle anatomy. Subsequently, Kay and Kay studied the zygomatic arches of 25 TCS subjects, noting that 36% of arches were unidentifiable and 24% were misshapen.5 Teber et al. defined the two minimum TCS diagnostic criteria as hypoplastic zygomatic arches and downward-sloping palpebral fissures.3 One component of our study was to examine the overall TCS malar morphology and to devise a classification scheme based on increasing severity of the deformity. Such a classification aids in standardised description and Table 1
Volume by Treacher Collins malar grade. Controls (n Z 34)
Type I (n Z 4)
Type II (n Z 5)
Table 2 Volumetric correlation of zygomata and masseter muscles. Zygomata (mm3) TCS (n Z 24) Controls (n Z 34)
2820 868
Pearson coefficient
3158 3254
r Z 0.44 (p < 0.20) r Z 0.74 (p < 0.001)
9395 4167
TCS, Treacher Collins syndrome.
communication, and facilitates surgical planning related to the TCS zygomatic deformity. Similar to previous studies, we confirmed that the minimal malar deficiency in TCS is lack of continuity along the zygomatic arch. We found an incomplete zygomaticetemporal articulation in 23/24 sides (the remaining side was spanned by a narrow osseous bridge). The next most-frequent open articulation was the zygomaticosphenoid suture, followed by incomplete connections at the zygomaticofrontal and zygomaticomaxillary regions. These recurring patterns served as the basis for our proposed classification system. The type I deformity is represented by a small zygomatic body, with presence of all articulations except for the arch. Type II is designated by diminished malar structure and absence along the lateral orbital wall (zygomaticosphenoid suture). While the inferolateral orbital rim is preserved in the moderate type, the arch is incomplete and shows no evidence of a zygomatic process from the temporal bone. Lastly, the type III deformity is distinguished by minimal volume, often entailing only remnants of malar structure at the zygomaticofrontal or maxillary region. The sequential lack of fusion in the inferolateral orbit and arch is reminiscent of arrested embryological development expected in a first and second pharyngeal arch deformity. Not surprisingly, the type III zygoma was seen most frequently in our TCS cohort, and represents the pathognomonic orbitozygomatic deformity. The absence of all normal ZMC articulations, and the near aplasia of the zygoma, accounts for the malar flattening and lid and lateral canthal malposition, classic in TCS. This would also be expected to result in a subnormal malar volume. The lack of a zygomatic arch and a hypoplastic malar body would also be anticipated to presage a diminished masseter muscle. As expected, the TCS malar volume was much smaller compared to age-matched controls. Our analysis volumetrically quantified the hypoplasia and gross dysmorphology posited in the above-mentioned classification. Considering Table 3 Symmetry comparison of Treacher Collins syndrome and controls.
Type III (n Z 15)
Avg. age 60.0 93.8 106.0 35.5 (mos) Volume 2820 868 2356 1066 1678 345 190 133 (mm3)
861 1004
Masseters (mm3)
Zygomatic ratio Masseteric ratio
Treacher Collins
Controls
Significance
0.66 0.28
0.97 0.02
p Z 0.002
0.74 0.20
0.92 0.09
p Z 0.001
46 TCS malar volumes in light of this classification showed a trend inversely correlated with increasing severity grade. Our mean age for both TCS and controls was 5 years, though the range was from infancy to older childhood. However, this age variability was conserved in both groups, which should balance any growth impact on malar volume. Additionally, compared to previous studies,4 these TCS subjects exhibited a much tighter age grouping. We also hypothesised that diminished and absent malar structure in TCS would correspond to masseteric hypoplasia (Figure 2). Indeed, in parallel with zygomatic volume, the mean TCS masseter was smaller than the mean control masseter: 3158 3254 mm3 versus 9395 4167 mm3, respectively (p < 0.0001). This is not surprising as the masseter muscle originates from the zygomatic arch and a portion of the malar body. Statistically, there was a positive correlation (r Z 0.44) between malar and masseteric volume in Treacher Collins patients (though weak). The coefficient of determination (r2 Z 0.20) estimated that only a fifth of the variation of one variable (e.g., masseteric volume) was explained by its relationship to the other variable (e.g., malar volume). This contrasted with the strong, positive correlation (r Z 0.74) and coefficient of determination (r2 Z 0.54) between malar and masseteric volume in normal controls. We expected that with an embryologic defect affecting the zygoma, the surrounding and attached structures, including the masseter, would be proportionally hypoplastic.8 Additionally, we posited that a diminished site of muscle origin (e.g., the zygoma) remains underdeveloped due in part to lack of biomechanical stimulation (i.e., the functional matrix). Our data do corroborate that both the TCS masseter and zygoma are hypoplastic, and significantly so, compared to controls. However, statistically, with these small numbers, we cannot intercede with certainty that one is because of the other. Interestingly, in a few TCS cases the smaller masseter was actually associated with the larger zygoma and vice versa. The interplay between muscle and bone appears more orderly and predictable in normal controls. Treacher Collins is frequently described as a ‘symmetric’ syndrome,9 but perhaps the more accurate term is ‘bilateral’. Phenotypically, our analyses show that there is
Figure 2 3-D CT scan highlighting the hypoplastic masseter of a TCS patient with type III zygomata.
K.R. Wong et al. reproducibility from left to right sides in the characterisation of malar deformity (83% demonstrated the same sideeside malar classification). However, volumetrically, there was significant variation. The average TCS patient exhibited 34% lefteright zygomatic volume asymmetry compared to only a 3% difference in normal controls. This sidedness ratio was similar for the masseters in TCS versus controls, thus quantifying the variable penetrance from side-to-side. This has important treatment implications because even though each zygoma may look anatomically similar in a single TCS patient, there may be significant volumetric differences that would portend differential reconstructive augmentation.
Conclusions A range of zygomatic hypoplasia exists in TCS (mildemoderateesevere), with near total aplasia being most common. The decrease in malar volume typically occurs in concert with masseteric hypoplasia, and the left and right sides are not symmetrically affected.
Conflict of interest None.
Acknowledgements We thank the Charles W. Ohse Grant for Surgical Research and the Yale University School of Medicine Office of Student Research for supporting this study. We appreciate the assistance of Kevin Callender and Elizabeth Roberto at the Yale University Statistical Laboratory.
References 1. Edwards SJ, Fowlie A, Cust MP, Liu DT, Young ID, Dixon MJ. Prenatal diagnosis in Treacher Collins syndrome using combined linkage analysis and ultrasound imaging. J Med Genet 1996;33: 603e6. 2. Jabs EW, Li X, Coss CA, Taylor EW, Myeres DA, Weber JL. Mapping the Treacher Collins syndrome locus to 5q31.3-q33.3. Genomics 1991;11:193e8. 3. Teber OA, Gillessen-Kaesbach G, Fischer S, et al. Genotyping in 46 patients with tentative diagnosis of Treacher Collins syndrome revealed unexpected phenotypic variation. Eur J Hum Genet 2004;12:879e90. 4. Marsh JL, Celin SE, Vannier MW, Gado M. The skeletal anatomy of mandibulofacial dysostosis (Treacher Collins syndrome). Plast Reconstr Surg 1986;78:460e70. 5. Kay ED, Kay CN. Dysmorphogenesis of the mandible, zygoma, and middle ear ossicles in hemifacial microsomia and mandibulofacial dysostosis. Am J Med Genet 1989;32:27e31. 6. Collins ET. Cases with symmetrical congenital notches in outer part of each lower lid and defective development of malar bones. Trans Ophthal Soc U K 1900;20:190e2. 7. Franceschetti A, Klein D. The mandibulofacial dysostosis: a new hereditary syndrome. Acta Ophthalmol Scand 1949;27: 143e224. 8. Poswillo D. The pathogenesis of the Treacher Collins syndrome (mandibulofacial dysostosis). Br J Oral Surg 1975;13:1e26. 9. Poswillo D. Otomandibular deformity: pathogenesis as a guide to reconstruction. J Maxillofac Surg 1974;2:64e72.