Abstract
Objectives:
Hemifacial microsomia (HFM) is frequently associated with middle ear anomalies. This study aimed to investigate the association between middle ear anomalies and the severity of mandibular deformity in patients with HFM.
Methods:
We retrospectively analyzed patients with HFM treated at the Eye and ENT Hospital of Fudan University between 2017 and 2025. The severity of mandibular deformity was classified using the Pruzansky–Kaban classification system. External ear anomalies were graded according to the Marx classification, and middle ear morphology was evaluated using high-resolution computed tomography with the Jahrsdoerfer scoring system. Audiological outcomes were assessed using pure-tone audiometry. Correlations between mandibular classification and middle ear structures were analyzed using the Kruskal–Wallis H-test, and hearing outcomes were compared using independent samples t-tests.
Results:
A total of 51 patients were included in this study. External ear deformities and malleus–incus complex abnormalities were prevalent across all HFM subtypes. Incus–stapes articulation differed significantly between type I and type II HFM, and between type I and type III HFM. Significant differences were observed in stapes morphology between type I and type III HFM, as well as between type II and type III HFM. Middle ear cavity development and mastoid pneumatization were significantly reduced in type III HFM compared with type I HFM. Inner ear anomalies were infrequently observed and were limited to a small subset of patients with severe HFM. Patients with more severe mandibular deformities demonstrated larger air–bone gaps and poorer air-conduction thresholds.
Conclusion:
In patients with HFM, the severity of mandibular deformity is significantly associated with middle ear structural abnormalities, including the ossicular chain, middle ear space, and mastoid pneumatization. Assessment of middle ear morphology may provide valuable adjunctive information for disease stratification, hearing outcome prediction, and surgical planning in HFM.
Keywords
Introduction
Hemifacial microsomia (HFM) is the second most common congenital craniofacial anomaly and is generally regarded a developmental disorder of the first and second pharyngeal arches. The reported incidence ranges from 1 in 3600 to 1 in 5600 live births. 1 Clinically, HFM is characterized by unilateral mandibular hypoplasia, maxillary deficiency, microtia, and varying degrees of associated soft-tissue deformities.
Two classification systems are most commonly used to describe the severity of HFM: the OMENS classification and the Pruzansky–Kaban (PK) classification. The OMENS system provides a comprehensive assessment of craniofacial involvement, including the orbit (O), mandible (M), ear (E), facial nerve (N), and soft tissue (S). However, its evaluation of ear involvement is largely limited to the morphology of the auricle and external auditory canal, without detailed consideration of middle ear structures. 2 In contrast, the PK classification is more widely applied in surgical decision-making and focuses primarily on mandibular development: type I describes a small but normally shaped mandible; type IIa indicates a hypoplastic mandibular ramus with a normal glenoid fossa; type IIb describes abnormal morphology and displacement of the temporomandibular joint; and type III denotes the absence of the temporomandibular joint. 3 Notably, neither classification system adequately addresses the morphology of the middle ear cavity, ossicular chain, or mastoid pneumatization.
From an embryological perspective, the mandible, external ear, and middle ear structures share a close developmental relationship, originating largely from the first and second pharyngeal arches. This developmental relationship has important implications for mandibular reconstruction, auricular reconstruction, and hearing rehabilitation in patients with HFM. However, relatively few studies have focused on ear malformations, particularly middle ear abnormalities, in this patient population. 4 Previous studies have reported associations between mandibular hypoplasia and reduced middle ear cavity volume or reduced mastoid pneumatization, but detailed analyses of the ossicular chain morphology remain limited. 5
Therefore, this study aimed to characterize external and middle ear morphology, including detailed analysis of the ossicular chain, and to investigate the association between middle ear anomalies and mandibular deformity severity in patients with HFM. These findings may provide complementary information to existing HFM classification systems.
Patients and Methods
Patients Selection and Data Collection
Data were retrospectively collected between 2017 and 2025 at our institution. Patients diagnosed with HFM were included if adequate craniofacial and temporal bone computed tomography (CT) scans were available. Informed consent was obtained from all patients for participation in this retrospective study.
Since HFM predominantly affects 1 side of the face, with bilateral involvement observed only in severe cases, the following approach was used for case enumeration: in patients with bilateral HFM and bilateral microtia, each side was evaluated independently and counted as a separate case; in patients with bilateral HFM and unilateral microtia, only the affected side was counted as a single case.
Image Analysis
Mandibular deformity was classified according to the PK classification (types I-III; Figure 1). Middle ear morphology was evaluated using high-resolution CT, and middle ear malformations were graded using the Jahrsdoerfer scoring system. The Jahrsdoerfer score assesses the anatomical integrity of key middle ear structures, with a maximum score of 10 points: 2 points are assigned for the presence of the stapes, and 1 point is assigned for each of the following structures: patent oval window, well-defined middle ear space, facial nerve canal, malleus–incus complex, mastoid pneumatization, intact incus–stapes connection, round window, and external auditory canal. 6 Auricular deformities were classified according to the Marx classification (grades I–III). 7
Three-dimensional computed tomography reconstructions illustrating facial morphology in different types of HFM. (A, B) Type I HFM. (C, D) Type II HFM. (E, F) Type III HFM. Frontal views are shown in panels (A, C, E), whereas lateral views are shown in panels (B, right) and (D, F, left). HFM, hemifacial microsomia.
All CT scans were acquired with a slice thickness of 0.75 mm using bone window settings. Representative images of normal and hypoplastic middle ear cavities are presented in Figure 2. CT images were initially interpreted by experienced radiologists and subsequently reviewed by senior otolaryngologists to ensure comprehensive evaluation of both otologic and mandibular structures.
Representative computed tomography images of the middle ear cavity in patients with HFM. (A, B) A patient with Pruzansky–Kaban (PK) type I HFM showing a well-pneumatized mastoid and an almost normal tympanic cavity on the left side. (C, D) A patient with PK type III HFM demonstrating a non-pneumatized mastoid and a markedly hypoplastic middle ear cavity on the left side. Panels (A, C) show coronal views, whereas panels (B, D) show axial views. HFM, hemifacial microsomia.
Audiological Evaluation
Audiological assessments were performed using a Madsen Conera audiometer (GN OtometricsA/S, Copenhagen, Denmark). Pure-tone air-conduction and bone-conduction thresholds were measured at frequencies ranging from 0.25 to 8 kHz. The average air-conduction and bone-conduction thresholds were calculated using frequencies of 0.5, 1, 2, and 4 kHz. The air–bone gap (ABG) was calculated as the difference between the average air-conduction and bone-conduction thresholds.
Statistical Analysis
The Kruskal–Wallis H-test was performed to compare total Jahrsdoerfer scores among the 3 mandibular deformity groups (types I, II, and III). For audiological data that met assumptions of normality and homogeneity of variance, independent samples t-tests were used to compare hearing outcomes between groups. A 2-tailed P < .05 was considered statistically significant. Post hoc pairwise comparisons were performed with Bonferroni correction where appropriate to control for multiple testing. All statistical analyses were performed using SPSS version 20 (IBM Corp, Armonk, NY, USA).
Results
General Characteristics
A total of 51 patients were included, comprising 33 males and 18 females (male-to-female ratio, 1.83:1). The mean age was 10.76 years (range, 5-26 years). Right-sided involvement was observed in 27 patients, left-sided in 23 patients, and bilateral involvement in 1 patient. According to the PK classification, 30 patients were classified as type I, 10 as type II, and 11 as type III. In the type I HFM group, 1 patient had bilateral HFM with bilateral microtia and was therefore counted as 2 affected sides (2 cases) in the analysis.
In the type I group, males predominated (22/30), whereas gender distribution was relatively balanced in the type II and type III groups (Table 1). Right-sided involvement was more common in the type I group, while left-sided involvement predominated in the type II and type III groups. Mean age was comparable among groups. Two patients in the type III group had congenital heart disease: one with tetralogy of Fallot and atrial septal defect, and the other with an isolated atrial septal defect. Both had undergone cardiac surgery prior to enrollment in this study.
Demographic and Clinical Characteristics of Patients With HFM.
External and Inner Ear Anomalies
One patient with bilateral HFM and microtia was analyzed with each side counted as an independent case. According to the Marx classification, most patients presented with grade III microtia (n = 46), with only 1 patient having a normal auricle (Table 2). Among patients with microtia, 30 had type I mandibular deformity. The prevalence of microtia was similar between the type II and type III groups.
External and Inner Ear Features in Patients With HFM According to PK Classification.
Abbreviation: PK: Pruzansky–Kaban.
With regard to the external auditory canal, atresia was observed in 46 patients, stenosis in 4, and a normal canal in 1. All patients in the type III group demonstrated complete canal atresia. No apparent association was observed between the severity of mandibular deformity and external ear deformity. Of note, 2 patients with type III HFM had unilateral mandibular involvement but bilateral type III microtia with bilateral canal atresia.
CT evaluation of the inner ear revealed normal inner ear and facial nerve anatomy in most patients. Three patients in the type III group demonstrated fusion between the lateral semicircular canal and vestibule. The facial nerve canal was not clearly visible on CT in 4 patients, all of whom had normal facial function. In contrast, 3 patients presented with facial paralysis despite normal facial nerve canal imaging on temporal bone CT.
Middle Ear Malformations and Hearing Outcomes
Middle ear morphology was assessed with a specific focus on the ossicular chain, middle ear space, and mastoid pneumatization (Table 3). According to the Jahrsdoerfer scoring system, malleus–incus complex malformation was present in 47 of 52 ears, with no significant differences among the 3 groups (P = .466). A normal incus–stapes connection was identified in 24 of 31 type I cases, whereas 19 of 21 type II or type III cases showed abnormalities of this structure, representing a statistically significant difference among groups. Post hoc pairwise comparisons demonstrated significant differences between type I and type II (P = .001) and between type I and type III (P < .001). A normal stapes was identified in 33 of 41 type I or type II cases, whereas all 11 type III cases demonstrated stapes abnormalities (type I vs type III and type II vs type III: P < .001). Most type I HFM patients had well-developed middle ear space and mastoid pneumatization, whereas most type II and type III patients showed underdevelopment of these structures. These differences were statistically significant (type I vs type III: P < .001 for mastoid pneumatization; P = .002 for middle ear space). Total Jahrsdoerfer scores did not differ between type II and type III HFM (P = .846), but were lower in type I HFM than in the other 2 groups (type I vs type II: P = .016; type I vs type III: P < .001). Because no difference in total Jahrsdoerfer scores was found between type II and type III HFM, these types were combined for subsequent hearing outcome analyses. Mean bone‑conduction thresholds were similar between type I and type II to type III HFM (Table 4; P = .173), whereas type II to type III patients had higher air‑conduction thresholds (P = .011) and larger ABGs (P = .022) than type I patients.
Middle Ear Structures Assessed by the Jahrsdoerfer Scoring System in Patients With HFM According to the PK Classification.
Abbreviation: PK: Pruzansky–Kaban.
Data are presented as number of cases with normal versus abnormal structures. Statistical comparisons were performed using the Kruskal–Wallis H-test. A P < .05 was considered statistically significant.
Comparison of Audiological Parameters Between Patients With Mild and Severe Mandibular Deformity.
Abbreviations: HFM, hemifacial microsomia; PK, Pruzansky–Kaban; dB HL, decibels hearing level; SD, standard deviation.
Discussion
Previous studies have reported variable findings regarding sex distribution in patients with HFM. In the present cohort, males predominated, with a male-to-female ratio of 1.83:1 (33:18). This differs from the equal sex distribution reported by Horgan et al 8 and shows a greater male predominance than that reported by Cohen et al. 2 These differences may be attributed to variations in sample size, patient selection, and study methodology. Consistent with prior reports, 9 right-sided involvement was more common in our cohort. Although the affected side of the mandible and microtia generally corresponded, a small number of patients demonstrated discordant involvement, underscoring the phenotypic heterogeneity of HFM.
Developmental Relationship Between Mandibular and Ear Anomalies
Microtia does not result from a single genetic mutation or environmental factor and is often considered part of the HFM phenotypic spectrum. Additionally, HFM has been proposed to represent a more severe craniofacial manifestation within the oculo-auriculo-vertebral spectrum, which includes Goldenhar syndrome. 10 In the present study, auricular deformity severity did not correlate with the degree of mandibular hypoplasia, a finding consistent with previous observations. 11 This lack of correlation may be explained by differences in embryologic origin. Although the mandible and middle ear ossicles are primarily derived from the first pharyngeal arch, the auricle develops from 6 hillocks of His with contributions from both the first and second pharyngeal arches. Consequently, auricular morphology is highly variable and may be influenced by complex genetic and environmental factors. 12
The high prevalence of severe microtia and canal atresia in our cohort likely reflects referral bias, as most patients were referred specifically for auricular reconstruction. This may explain why auricular deformities in our cohort appeared more severe than those reported in previous studies. 13 Unlike the findings of Chai et al, 13 who reported a correlation between mandibular deformity severity and external ear deformity, our study did not demonstrate such a correlation, which may reflect population-based differences in HFM phenotypes.
In contrast to the external ear, middle ear structures demonstrated a stronger association with mandibular development. The middle ear cavity contains the ossicular chain, which is essential for sound transmission and lies in close anatomical proximity to the mandible. Embryologically, Meckel’s cartilage is a transient embryonic structure that gives rise to the mandible anteriorly and contributes to the malleus and incus posteriorly. 14
In the present study, significant differences were observed among HFM groups in the incus–stapes connection, stapes morphology, middle ear cavity size, and mastoid pneumatization. Conversely, no significant differences were observed in the malleus–incus complex, oval window, round window, or facial nerve canal. These findings may be interpreted in the context of proposed pathogenic mechanisms of HFM, including vascular ischemia, disruption of Meckel’s cartilage development, and abnormal migration or differentiation of cranial neural crest cells (CNCCs).
Vascular ischemia is the most widely accepted pathogenic mechanism and may account for the predominantly unilateral presentation of HFM. The consistent absence of a normal stapes in patients with type III HFM, together with its relative preservation in type I and type II cases, supports the hypothesis of stapedial artery disruption. However, vascular disruption alone cannot fully explain the high prevalence of malleus–incus complex abnormalities observed across all HFM subtypes.
Disruption of Meckel’s cartilage development may provide an alternative explanation, as its derivatives contribute to both the mandible and the ossicular chain. 15 This shared embryological origin may account for the close association between mandibular hypoplasia and malleus–incus abnormalities. Additionally, disrupted CNCC migration and differentiation may contribute to the variable involvement of adjacent skeletal and soft tissue structures. 16 These mechanisms likely interact to varying degrees in individual patients, with genetic susceptibility and environmental factors—such as maternal diabetes or teratogenic exposures—further modifying phenotypic expression. 17
An interesting laterality pattern was observed in this study, with right-sided predominance in type I cases and left-sided predominance in type II and type III cases. Although the underlying mechanism remains unclear, this finding may be related to asymmetric vascular development or differential susceptibility to disruption of the first and second pharyngeal arches during embryogenesis. Further studies are needed to clarify this observation.
Inner ear and vestibular structures were largely preserved in our cohort, even in patients with severe mandibular deformity. This preservation likely reflects their distinct embryological origin from the otic placode rather than the pharyngeal arches. Rare inner ear anomalies, such as lateral semicircular canal–vestibule fusion, were observed exclusively in type III HFM, consistent with previous reports. 18 Facial nerve paralysis and congenital heart disease, identified in a minority of severe HFM cases, suggest more widespread embryological disruption.
Implications for Hearing Outcomes and Surgical Planning
These findings have important clinical implications for hearing assessment and surgical decision-making in patients with HFM and ear malformations. Although malleus abnormalities were common across all subtypes, increasing severity of mandibular deformity was associated with a higher likelihood of stapes abnormalities, reduced middle ear cavity volume, and decreased mastoid pneumatization. Patients with more severe mandibular hypoplasia demonstrated poorer air-conduction thresholds and larger ABGs, reflecting the cumulative effect of middle ear structural abnormalities. Consistent with previous studies, well-developed middle ear space and adequate mastoid pneumatization were associated with better hearing outcomes.
From a surgical standpoint, these anatomic patterns may assist in preoperative risk stratification. In patients with advanced mandibular hypoplasia and limited middle ear space, ossicular reconstruction or canalplasty may be technically challenging and less likely to yield satisfactory hearing improvement. In such cases, early consideration of bone-conduction hearing devices may provide more predictable auditory rehabilitation. Conversely, patients with preserved middle ear volume and mastoid pneumatization may be more suitable candidates for reconstructive procedures.
Children with HFM and microtia often present early in life with hearing loss and speech delays and are frequently referred for craniofacial or auricular reconstruction. 19 Comprehensive preoperative evaluation, including audiological assessment and high-resolution temporal bone CT, is therefore essential. 20 Systematic evaluation of middle ear morphology may help determine whether patients are candidates for canalplasty or whether bone-conduction devices are more appropriate, thereby facilitating individualized treatment planning.
Integrating these radiologic and audiologic findings into routine surgical planning can also inform the timing and type of intervention, helping surgeons balance reconstructive goals with realistic expectations for hearing improvement. Such an approach may contribute to more individualized rehabilitation strategies and improved functional outcomes.
Considerations for Classification Systems
Existing HFM classification systems, including the PK and OMENS classifications, primarily focus on mandibular morphology and external craniofacial features, with limited assessment of middle ear anatomy. 8 Although these systems remain essential for craniofacial evaluation, 21 they may not fully capture the spectrum of otologic involvement or the extent of hearing impairment. In our cohort, middle ear features—particularly the incus–stapes connection, stapes morphology, middle ear cavity volume, and mastoid pneumatization—demonstrated stronger associations with mandibular deformity severity than did auricular morphology alone.
Rather than proposing a novel classification system, our findings suggest that detailed evaluation of middle ear morphology may provide valuable complementary information to existing classification frameworks. Incorporating systematic assessment of ossicular chain integrity and middle ear volume into routine preoperative evaluation may improve prediction of hearing outcomes and facilitate more precise surgical planning.
Conclusion
In patients with HFM, the severity of mandibular deformity is significantly associated with middle ear structural abnormalities, including ossicular chain integrity, middle ear cavity volume, and mastoid pneumatization. These middle ear features demonstrate strong correlations with hearing outcomes and provide valuable complementary information to existing mandibular classification systems. Systematic preoperative evaluation of middle ear morphology may enhance disease stratification, improve prediction of auditory function, and facilitate individualized surgical planning in this patient population.
Footnotes
Acknowledgements
The authors would like to thank the staff of the Eye and ENT Hospital of Fudan University for their assistance in data collection and clinical evaluation.
Ethical Considerations
This study was approved by the Ethics Committee of the Eye and ENT Hospital of Fudan University (approval no: 2023179-1) and was conducted in accordance with the Declaration of Helsinki.
Consent to Participate
Written informed consent was obtained from all patients or their legal guardians for participation in this study.
Author Contributions
Yang Zhang and Yaoyao Fu designed the work. Jinchao Yu and Xinhai Ye acquired and analyzed data. Yang Zhang, Yaoyao Fu, and Tianyu Zhang drafted, revised, and approved the manuscript. Yaoyao Fu and Tianyu Zhang agree to be accountable for all aspects of the work.
Funding
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Yaoyao Fu was supported by the Clinical Distinguished Physician Training Program 2026–2028 (Grant No. pp25039). Tianyu Zhang was supported by the Natural Science Foundation of Shanghai (Grant No. 23Y21900200).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.