Asymmetric dentofacial deformity frequently results in aesthetic and functional impairments, typically characterized by pronounced jaw deviation, imbalance, and occlusal disorders (1–4). The causes of asymmetric dentofacial deformities are primarily associated with genetic factors, developmental abnormalities, and trauma (1, 5). Additionally, studies have identified pathological changes in the temporomandibular joint (TMJ), such as excessive load or positional deviations of the condyle, as significant contributors to jaw asymmetry (1). Asymmetric joint movement may result in either overgrowth or undergrowth of the condyle, subsequently contributing to facial deformity (6). Currently, the primary treatment for patients with severe skeletal asymmetry involves orthodontic treatment combined with orthognathic surgery (7). The occlusal plane will be adjusted through maxillary Lefort I osteotomy. After bilateral sagittal split mandibular ramus osteotomy (BSSRO), asymmetric movement of the distal mandibular segment facilitate the adjustment of the mandible's length, height, and width on both sides. The jaws will be repositioned to restore facial aesthetics and improve occlusion (1, 8). However, due to the variability in condylar position during surgery, changes in condyle after orthognathic surgery are highly intricate, affecting the long-term outcomes of the surgery (1, 9, 10).

Advancements in imaging technology have led to significant changes in the methods used to analyze TMJ alterations (11, 12). The Cone-Beam Computed Tomography (CBCT) enhanced the accuracy of condylar position assessment, overcoming the limitations of two-dimensional imaging, and become widely utilized in orthognathic surgery research (13–15). Nowadays, machine learning-based assessment emerged to analyze the image data automatically and effectively. Our previous research has developed a fully automated quantitative method using deep learning, capable of identifying the condyle in CBCT images and measuring changes in its position and volume before and after surgery. This method enables a more objective and accurate evaluation of postoperative condylar changes in patients with skeletal malocclusion, providing a scientific foundation for assessing surgical outcomes and clinical treatments (16). In the previous study, we found that condylar resorption in both sides after bimaxillary orthognathic surgery, correlated with counterclockwise rotation in the sagittal plane in patients with skeletal class II malocclusion. In the current study, we used a new sample with asymmetric dentofacial deformity.

Currently, studies on condylar changes in patients with Angle class II and class III jaw deformities have revealed several similar trends (17). In Angle class II patients, the condyle often exhibits significant posterior displacement and remodeling due to the anterior displacement of the distal mandibular segment (14, 18, 19). In contrast, patients with Angle Class III, who present with protruding mandibles, typically require resection of part of the proximal segment and posterior displacement of the distal segment. Although these patients undergo some remodeling and movement of the condyle, such changes have a limited impact on bone recurrence and clinical outcomes (13, 14). However, research on condylar changes in patients with asymmetric dentofacial deformities remains limited (1, 11, 12). Systematic studies evaluating condylar changes before and after surgery, as well as the long-term effects of these changes on joint function, are lacking in this patient population.

This study aimed to measure condylar volume, position, direction in patients with asymmetric dentofacial deformities before and after surgery using an established automated measurement program. The hypothesis was that condylar kinematics and spatial orientation would be significantly different between the deviated and non-deviated sides from pre-surgery to 1-week and 6-month post-surgery.

Approximately one-third of patients with dentofacial deformities present with asymmetric deformity (20). We found 122 patients with dentofacial deformity and 30% (36 case) of them had asymmetry. Orthodontic-orthognathic treatment is currently the primary approach to correct asymmetric maxillofacial deformities. However, due to changes in occlusion and jaw movement after surgery, studies suggest that the mandibular condyle may showing remodeling and displacement (21, 22). The extent of condylar remodeling plays a critical role in maintaining postoperative occlusal stability, preventing bony recurrence, and preserving TMJ health (23).

In our earlier study, we developed a reliable and efficient method for assessing condylar changes using deep learning (16), which provides reliable and stable CBCT image analysis. In that study, we focused on the patients with skeletal class II malocclusion and found condylar resorption in both sides after bimaxillary orthognathic surgery, correlated with counterclockwise rotation in the sagittal plane. In the current study, we used a new sample who were diagnosed with asymmetric dentofacial deformity.

We found the average volume of the condyle on the deviated side is smaller than that on the non-deviated side, which is consistent with Chou's findings (24). However, some studies have reported that the deviated condyle is larger than the non-deviated condyle (3, 25, 26). This discrepancy may because of the different marking point of the condylar extent in other studies.

As a result of jaw repositioning and occlusal changes, the condyle may undergo varying degrees of remodeling after orthognathic surgery (21–23, 27). In each side, there were no significant difference in condylar volume between T0T1, suggesting that condylar volume remained stable during the immediate postoperative phase. Between T1T2, we found the deviated side condylar volume decreased, aligning with many studies that condylar remodeling occurs after orthognathic surgery (22). However, we found the non-deviated condylar reduction after surgery was minimal. Many studies indicate that patients with skeletal Class II dentofacial deformities exhibit progressive bilateral condylar resorption and remodeling postoperatively (19, 28). In contrast, the condyle of patients with Class III deformities exhibited greater postoperative stability (29). This phenomenon may result from the anterior movement of the distal mandibular segment (30). Postoperatively, musculofascial tension induces a posterior force, pushing the condyle into the fossa (31). This generates compressive loading on the condylar head, initiating adaptive remodeling. For the deviated side, the movement of the distal mandibular were similar to skeletal Class II patients, thereby resulting in similar outcomes.

Most of the changes observed were very small in magnitude or even practically insignificant, and some of the smaller changes may fall within the error margin. However, even relatively small but true changes in TMJ position could lead to major occlusal problems affecting dental relationships.

Several studies suggest that condyles undergo three-dimensional displacement immediately after orthognathic surgery (32). In the anterior-posterior direction, we had not found significantly difference. However, Baek et al. (33) reported the condyle displaces posteriorly but Kim et al. (34) observed the contrary phenomena. We found condyles shifted outward laterally and downward vertically, in both sides immediately after surgery. Choi et al. (5), Zhang et al. (13) and Svetlana Tyan et al. (35) reported the similar phenomena. And we noticed that the position of condyle gradually returns to its original orientation over the long-term post-surgery. However, we find that the condyle on the deviated side has moved upward beyond its original position, as shown in some studies (36–38, 39). It may be related to the remodeling of the condyle. Numerous studies of patients with Class II deformities have identified a correlation between condylar resorption and upward condylar displacement. Thus, both operation and postoperative orthodontic treatment, should avoid imposing excessive loads on the deviated condyle.

In the rotation change of condyle, either on the deviated side or the non-deviated side, we investigated that the condyles rotated clockwise in the sagittal plane, aligning with findings by Ma et al. (11). In contrast, Hsu et al. (18) reported counterclockwise rotation in the sagittal plane in their cohort whereas other studies (17, 19, 40) documented negligible sagittal-plane rotational changes postoperatively. Notably, excessive counterclockwise mandibular rotation in the sagittal plane during orthognathic procedures has been associated with elevated risks of progressive condylar resorption (30). Many studies demonstrated that the condyle rotates anteriorly in the axial plane postoperatively (11, 15, 18, 19, 41), consistent with our findings. Some scholars have suggested that this phenomenon may be related to the fact that most surgeons are right-handed (15). We found that the mostly the condylar rotation returned to the original direction in the long-term, the deviated side condyle had the trend to back to its original direction in the axial plane.

Notably, we found that the condyle on the deviated side rotated laterally in the coronal plane, while the other side rotated medially. These aligns with some prior research findings (19, 22, 42–44), but some others believe that the condyle has no rotational changes in the coronal direction (42, 45).

Some scholars attribute this to the changes in mandibular body width, which is associated with the mandible's “V”-shaped structure and the movement of the distal bone segment (18). For example, on the non-deviated side, posterior movement of the distal segment causes the outward displacement of the proximal segment, causing medial rotation of the condyle. Podčernina et al. (19) concluded that the bilateral condyles in skeletal Class III patients rotated medially after surgery, resembling the medial rotation on the non-deviated side condyle after the posterior shift of the distal bone segment. However, some studies have reported findings that contradict our results (46, 47).

Orthognathic surgery not only alters mandibular positioning and occlusal relationships, but also directly impacts the morphology of condyle (Figure 6). Condylar displacement and adaptive remodeling after surgery are critically associated with clinical outcomes (12, 15). Emerging evidence suggests that despite successful restoration of occlusal relationships postoperatively, persistent TMJ instability may predispose patients to TMJ discomforts and relapse (17, 19). Comprehensive evaluation of postoperative temporomandibular joint (TMJ) alterations and predicting the reshaping of condyle after surgery, enables surgeons to develop holistic surgical strategies.

The current study still has some limitations, including the application of artificial intelligence to analyze 3D biomedical images. Due to lack of robust mandibular orientation, it remains insufficiently standardized to support reliable measurement of small changes in the condyles. Another limitation is the lack of reliability testing and more samples from multicenter are needed for further analyze. Moreover, we cannot reveal the relation between condylar changes and TMJ function. Future investigations will incorporate patients with TMJ discomfort to delineate associations between postoperative condylar kinematics and clinical symptomatology, thereby advancing our understanding of biomechanical contributors to TMJ dysfunction in orthognathic outcomes. And it also needs to quantify relationships between mandibular movement magnitude, condylar displacement/rotation, and volumetric remodeling.

For patients with asymmetric dentofacial deformities, orthognathic surgery causes positional and rotational changes. Condyle in the both sides tend to return to its original position in the long run, but there are still measurable rotations in the coronal plane. Condyle on the deviated side undergo greater remodeling than the non-deviated side.