Rationale

The orthopaedically stable mandibular condyles ground the normal temporomandibular joint (TMJ) function. Orthognathic surgery is intended to change the spatial position of anatomical structures, and therefore certainly can disrupt this balance. The resulting compression of the condyle initiates its adaptive remodeling , which has been proven in animal experiments and in clinical studies [1, 2]. The remodeling of the articular surface is a natural compensatory biological response to the mechanical load that has arisen, which provides homeostasis of the TMJ form and function and the occlusion. If the resulting TMJ compression exceeds the articular adaptivity, the remodeling becomes pathological and can outcome in condylar resorption and skeletal relapse of the dentofacial deformity [3]. According to the literature, the correlation has been proven between significant condyle dislocation and the resulting remodeling [4, 5]. The dislocated condyle tends to return to its optimal position, causing postoperative instability and possibly complicating the orthodontic follow-up. Therefore, it is necessary to minimize the surgical dislocation of the mandibular condyle fragment. This determines the urgency of the problem, and also determines the need for correct control of the result of the surgical treatment.

Objective

To evaluate linear and rotational changes in the position of the mandibular condyle fragments after orthognathic surgery.

Materials and methods

The retrospective study included 64 condylar processes of 32 patients (9 men and 23 women, aged 18–41 years; 24.1 on average). All patients underwent double-jaw orthognathic surgery for the skeletal form of a Class II or III dentofacial anomaly. All patients underwent preoperative orthodontic decompensation of the dentition. The following were the exclusion criteria: craniofacial syndrome, active temporomandibular joint diseases, pronounced post-traumatic deformation of the mandibular condyles. Depending on the type of skeletal deformity, the patients were divided into group 1, Class II skeletal deformity (n=16) and group 2, Class III skeletal deformity (n=16).

All patients underwent double-jaw orthognathic surgery: the upper jaw Le For I osteotomy and bilateral sagittal split osteotomy (BSSO) of the lower jaw with the short-split [6]. To fix the distal and proximal fragments of the mandible passively, we performed selective grinding of excessive contact or posterior flexion osteotomy of the tooth-bearing fragment. The bivector seating positioned the mandibular condyle fragments. Then the mandibular fragments were fixed with titanium plates, mono- and bicortical screws. The necessity and number of bicortical screws were determined intraoperatively to increase the inter-fragmental stability.

Multispiral computed tomography (MSCT) of the skull (Siemens SOMATOM, Germany) was performed before (T0) and 1–3 days after the surgery (T1). The three-dimensional visualization software superimposed the three-dimensional images before and after the surgical treatment semi-automatically (based on preselected points of the base of the skull and voxel-based). Then the mandibular rami were segmented manually, and the resulting volumes were combined into “before/after” pairs separately for the left and right rami for each patient. Thus, three-dimensional images were obtained of 64 pairs of mandibular rami. For orientating the volumes, the Frankfurter plane (drawn through the points of the left and right porion [Po] and right orbital [Or]) was used as a reference of the axial plane. The frontal plane was drawn through the skull base point (Ba), perpendicular to the Frankfurter plane. The sagittal plane was carried out through the nasion (Na) and skull base (Ba) points.

To quantify the change in the position of the mandibular condyle fragment, a three-dimensional coordinate system was introduced, with the X axis reflecting the mediolateral, the Y axis, the upper-lower, and the Z axis, anterior-posterior displacement. The coordinates of the following points were measured: the lateral pole (LP) of the condyle, the medial pole (MP) of the condyle, the condylion (Co), gonion (Go), coronoid process (Cor) and the most concave point of the mandibular notch (Notch) points. These points were chosen to describe both the linear and the angular changes. To assess the linear displacement of the condylar processes after the surgery, we calculated the difference in the coordinates for each point under study. To assess the rotational changes of the mandibular condyle fragments, we calculated the angles between the vectors drawn through the LP and MP points in the axial view (angle α in the {x;z} coordinate plane), the LP and MP points in the frontal view (angle β in the {x;y} coordinate plane) and the Co and Go points in the sagittal view (angle γ in the {y;z} coordinate plane). The coordinate vectors were calculated separately before and after surgery (the corresponding points in the postoperative period are indicated by the same letters with apostrophe, for example, MP’, LP’). All measurements were carried out in pairs of the obtained mandibular condyle fragments' volumes separately for the right and left sides.

The following formulas were used for the calculations:

For linear displacement:

∆x = x(T1) – x(T0),

∆y = y(T1) – y(T0),

∆z = z(T1) – z(T0).

For the rotational change, the coordinates were calculated of the vectors of the condylar axis in the axial view in the pre- and postoperative periods, respectively:

={MP_x – LP_x; MP_z – LP_z},

={MP'_x – LP'_x; MP'_z – LP'_z}

The angle between the vectors of the condylar axis in the axial view after the surgical intervention was calculated using the following formula:

.

Similarly, the coordinates of the vectors and angles between them were calculated for the axis of the condyle in the frontal view and the axis of the condylar process in the sagittal view.

Results and Discussion

A number of factors influence the position of the condyle because of the surgery: the magnitude and direction of the maxillo-mandibular complex movement, the fixation method and passivity, the precision of intraoperative positioning of mandibular condyle fragments, and compression of perimandibular tissues [7]. The postoperative position of the condyle is a surgical outcome factor. Excessive compression in the condylar displacement in its contact with the surface of the articular fossa can remodel pathologically the articular surfaces and subchondral bone. Clinically, the pathology can be manifested by TMJ disorder symptoms or late skeletal relapse of dento-facial deformity [8, 9].

Computed tomography is the method of choice for monitoring postoperative outcome. The literature describes methods for the TMJ study that include an assessment of the position of the mandibular condyle by the analysis of two-dimensional slices of tomograms in fixed views [10]. Mostly, the width of the articular gap at certain points is measured — a decrease in the posterior and upper articular spaces shows the posterior-upper or distal condylar displacement. However, this method cannot assess rotational changes in mandibular condyle fragments. For example, the torque of the condylar process can maintain the preoperative position of the apex of the mandibular condyle, with its displaced lateral poles outside the slice under analysis. Combined with mediolateral displacement, this position can be regarded radiologically as an increase in the upper articular space, leading to overdiagnosis of the lower and/or anterior position of the condyle (the so-called “condylar sag”). Such diagnostic errors significantly complicate the further treatment of the patient by both a surgeon and an orthodontist. A method is also proposed for measuring two-dimensional angles formed by lines drawn through fixed points of the condyle and the generally accepted planes [11]. The development and implementation of three-dimensional technologies enabled an accurate mandibular condyle fragments’ position measuring, including direct and visual comparison with superimposition of the volumes of pre- and postoperative images [12].

We aimed to investigate changes in the spatial position of the mandibular condyle fragments after orthognathic surgery. At the moment, there are no studies in the Russian literature with the analysis of data from three-dimensional reconstruction of the volumes of computed tomograms.

The results of linear and angular changes are shown in Tables 1–3. The average advancement of the tooth-bearing fragment of the mandible at point B in class II patients was 12.2±0.8 mm, and in class III patients it was 1.8±0.9 mm.

Table 1. The linear changes of condyle position in group 1 patients after surgery

Note. If the value of the number is positive it means the direction to the right, up and forward, respectively, if it’s negative — to the left, down and to the back, respectively, R — right side, L — left side, * — p<0.05, statistically significant.

Table 2. The linear changes of condyle position in group 2 patients after surgery

Note. If the value of the number is positive it means the direction to the right, up and forward, respectively, if it’s negative — to the left, down and to the back, respectively, R — right side, L — left side, * — p< 0.05, statistically significant.

Table 3. Rotational changes of condyle-bearing fragments in group 1 and 2 patients

Note. ∠α show rotation of condyle-bearing fragments in axial plane (axis {x;z}), ∠β — in coronal plane (axis {x;y}), ∠γ — in sagittal plane (axis {y;z}), If the value of the number is positive it means the counterclockwise rotation of the condyle, if it’s negative — clockwise rotation, R — right side, L — left side, * — p<0.05, statistically significant.

The results were processed statistically using the licensed STATISTICA software v. 10 (Statsoft, USA). To assess the compliance of quantitative data with the normal distribution, the Shapiro–Wilk test was used. With a distribution corresponding to the normal, quantitative data were presented as an average ± standard deviation (M±SD). Quantitative variables with a distribution different from normal were described using median (Me) and lower and upper quartiles [Q25; Q75]. The Mann–Whitney U test was used to determine the differences in quantitative variables between two independent groups. The differences in the quantitative variables for two periods were compared using the Wilcoxon W-test. The p=0.05 values were considered significant.

The orthognathic surgery resulted not only in linear displacements of the condyle, but also in angular parameters changes as an upper-lateral torque of the entire mandibular condyle fragment.

For example, in Group I, a significant linear displacement of the right LP point occurred only along the Y axis, the median of displacement was 1.61 [1.26; 1.92], p=0.001; that of the right MP point occurred along the Z axis (median −1.47 [−1.61; −0.85]), p=0.001; that of the right Co point occurred along the Y and Z axes (median 0.48 [−0.06; −0.87], p=0.017, and −0.37 [−1.01; −0.12], p=0.028, respectively). When estimating the Go coordinates on the right, the significant linear displacements were revealed along the X, Y, and Z-axes (median offsets −4.96 [−6.17; −3.70], p=0.001; 1.18 [0.66; 1.71], p=0.002; and 4.55 [3.52; 5.96], p=0.001). It is worth noting that the median of the upper-lower displacement (Y-axis) of the LP point in the Group I was higher on both sides compared to Group II (right p=0.011, left p=0.001). The median of the Go coordinates displacement along the Z-axis on the right (p=0.002) and the X-axis on the left (p=0.014) in the Group I significantly exceeded that in Group II.

The rotational changes in mandibular condyle fragments showed significant intergroup differences in the angle γ on the right (5.21 [4.14; 6.93] ° in Group I vs. 3.02 [−0.61; 4.12] ° in Group II, p=0.003) and in the angles β (8.32±5.38° in Group I vs. 3.93±3.56° in Group II, p=0.018) and γ (6.42±3.04° in Group I vs. 0.46±2.11 in Group II, p<0.001) on the left.

Pronounced rotation in the frontal and sagittal planes in the Group I may probably be associated with a more significant advancement of the maxillo-mandibular complex. In the frontal plane, the right mandibular condyle fragment is rotated clockwise, and the left one is rotated counterclockwise (Fig. 1). A significant displacement in the frontal plane is probably caused by the features of the anatomical shape of the mandible, which is a posteriorly expanding parabola. During the advancement and rotation of the tooth-bearing fragment, interfragmental bone interferences occur causing the lateral rotation of mandibular condyle fragments (Fig. 2, 3). The counterclockwise rotation of the tooth-bearing fragment in the sagittal plane is also associated with anterior displacement of the proximal fragment, and inevitably occurs in skeletal Class II treating (Fig. 4).

Fig. 1. The counterclockwise rotation of the left condyle-bearing fragment (coronal plane).

The angle between line MP-LP and line MP’-LP’ was determined as ∠β. MP, MP’ — medial pole of condyle before and after surgery. LP, LP’ — lateral pole of condyle before and after surgery. Blue image — the position of condyle-bearing fragment before surgery, yellow image — after surgery.

Fig. 2. Schematic drawings of condyle-bearing fragments displacement during mandibular advancement.

a — bony interference occurs after large mandibular advancement; b — rotation and torque of condyle-bearing fragments, caused by distal part of the tooth-bearing fragment; c — condyle-bearing fragments displacement amplitude.

Fig. 3. Positional changes of condyle-bearing fragment after mandibular advancement caused by parabolic shape of mandible.

Red line — the geometric shape of mandible. Green dotted lines — directions of condyle-bearing fragments displacement.

Fig. 4. Rotational changes of the condyle-bearing fragment caused by large mandibular advancement in Class II patients (sagittal plane).

Co — condylion, Go, Go’ — gonion before and after surgery. The angle between these lines was determined as ∠γ.

Thus, the separate analysis of the displacement of the selected points confirms the upper-lateral torque of the proximal fragments of the mandible (Fig. 5). Similar results are described by Ruo-han Ma et al. (2019) [13].

Fig. 5. The condyle-bearing fragment superior and lateral torque.

Go — gonion, MP — medial pole of condyle, LP — lateral pole of condyle, Cor — coronoid process. Arrows show the directions of points displacement that reveal the positional changes of condyle-bearing fragment.

The discovered upper-lateral torque pattern of mandibular condyle fragment displacement can explain the overdiagnosis of the posterior position of the condyles: when analyzing individual two-dimensional slices of computed tomograms, an axial rotation of mandibular condyle fragment can be regarded as a decrease in the posterior articular spaces (Fig. 6, 7). The true corpus-directed distal shift of the condyles was observed only in 2 cases (n=4), both in Group I, with a significant discrepancy in the volumes of articular elements against the background of a condyle decrease (presumably because of the earlier condylar resorption ended long before the initial treatment). Nevertheless, such disproportion is often observed in the skeletal class II patients and cannot be considered the only reason for postoperative malposition. Therefore, the role is logical for the direction and magnitude of maxillo-mandibular advancement, the geometry of the tooth-bearing fragment of the mandible, and the compression of perimandibular tissues in response to the movements. Further study is needed to correlate these variables.

Fig. 6. The «posterior» displacement of condyle.

Measurement of the condyle in sagittal view doesn’t take into account rotational changes of condyle-bearing fragment. a — the condylar position before surgery (sagittal plane); b — the condylar position after surgery (sagittal plane). The «posterior» displacement of condyle; c — differences between 3D models of condyles. MP, MP’ — medial pole of condyle before and after surgery. LP, LP’ — lateral pole of condyle before and after surgery. Cor, Cor’ — coronoid process before and after surgery. Go, Go’ — gonion before and after surgery. Yellow image — position of condyle-bearing fragment before surgery, blue image — after surgery. Red arrows — directions of points displacement.

Fig. 7. The rotation of condyle-bearing fragment (axial plane).

The angle between line MP-LP and line MP’-LP’ was determined as ∠α. MP, MP’ — medial pole of condyle before and after surgery. LP, LP’ — lateral pole of condyle before and after surgery. Cor, Cor’ — coronoid process before and after surgery. The rotation is often considered as posterior displacement of condyle in analysis of sagittal sections of CT scans.

Conclusion

The linear and rotational changes were analyzed in the coordinates of the selected points of volumetric reconstruction of the skull MSCT, and the change were assessed in the position of the mandibular condyle fragments resulted from the orthognathic surgery. Our data showed a trend to the upper-lateral torque of the entire mandibular condyle fragment. The revealed torque of proximal fragments can be mistaken for distalization of the condyle when studying two-dimensional MSCT slices.

Orthognathic surgery is the cause of the inevitable change in the mandibular condyle fragment position. In a significant condyle dislocation, the excessive mechanical load in the TMJ initiates the pathological remodeling, which can lead to the development of TMJ disorder symptoms and skeletal relapse because of condylar resorption. Therefore, it is necessary to minimize the surgical dislocation of the mandibular condyle fragment.

Further studies are required to understand the causes of this phenomenon and develop methods for its prediction and prevention at the stage of preoperative modeling. These studies should correlate the direction and degree of mandibular condyle fragments dislocation with several factors influencing the mutual disposition of articular structures because of orthognathic surgery. A study with a long postoperative observation would allow us to assess the dynamics of the condyle structure and volume depending on its position in the early postoperative period, as well as to develop recommendations to accelerate the functional rehabilitation.

The authors declare no conflict of interests.