3D analysis of condylar and mandibular remodeling one year after intra-oral ramus vertical lengthening osteotomy

Abstract

Objectives

Among the existing techniques for the correction of mandibular posterior vertical insufficiency (PVI), the intra-oral ramus vertical lengthening osteotomy (IORVLO) can be proposed as it allows simultaneous correction of mandibular height and retrusion. This study assessed the 3D morpho-anatomical changes of the ramus-condyle unit and occlusal stability after IORVLO.

Materials and methods

This retrospective analysis compared immediate and 1-year post-operative 3D CBCT reconstructions. The analysis focused on the condylar height (primary endpoint) and on the changes in condylar (condylar diameter, condylar axis angle) and mandibular (ramus height, Frankfort-mandibular plane angle, gonion position, intergonial distance, angular remodeling) parameters. Additionally, this analysis investigated the maxillary markers and occlusal stability.

Results

On the 38 condyles studied in 21 included patients (mean age 23.7 ± 3.9 years), a condylar height (CH) loss of 0.66 mm (p < 0,03) was observed, with no correlation with the degree of ramus lengthening (mean 13.3 ± 0.76 mm). Only one patient presented an occlusal relapse of Class II, but a 3.4 mm (28%) condylar diameter loss and a 33% condylar volume reduction with loss of 1 mm and 3.4 mm in CH and condyle diameter, respectively. A mean 3.56 mm (p < 0.001) decrease in ramus height was noted, mainly due to bone resorption in the mandibular angles.

Conclusion

This study confirms the overall stability obtained with IORVLO for the correction of PVI.

Clinical relevance

This study aims to precise indication of IORVLO, and to validate the clinical and anatomical stability of results.

Introduction

The mandibular posterior vertical dimension (PVD) holds a critical role in defining the esthetic and architectural parameters of the lower third of the face. PVD influences the position of the mandibular angle, the inferior jawline, as well as the occlusal plane. Salagnac et al. [1] defined the posterior mandibular height as the length of the ascending branch of the mandible, i.e., the F3 line in the Delaire cephalometric analysis [2]. Posterior vertical insufficiency (PVI) of the mandible may result from several etiologies including congenital disorders (craniofacial microsomia, Treacher-collins syndrome, condylomandibular dysplasia), acquired etiologies (condylar fractures, temporo-mandibular joint (TMJ) osteo-arthritis or ankylosis) or from constitutional condyle hypoplasia [3]. Unilateral PVI is characterized by shortening of the mandibular ramus causing a facial asymmetry with homolateral elevation of the angle, deviation of the chin, commissural line and occlusal plane tilting, as well as unilateral class II malocclusion. Bilateral PVI is responsible for the typical ‘bird face deformity’ as described by Obwegeser [4], with class II malocclusion and poorly defined jawline and mandibular angles, facial hyperdivergence, pathologic molar contact, anterior open bite, sleep apneas, stressed by small buccal volume and lingual dysfunction.

Several procedures have been described to correct PVI depending on whether the TMJ is affected. Functional orthopedic appliances to promote mandibular growth in children showed interesting results regarding the correction of mandibular height as evidenced in condylo-mandibular dysplasia [5]. The chondro-costal graft is usually reserved for the correction of PVI with absent or altered TMJ particularly regarding children, as it allows both reconstruction of the ramus and condyle, while maintaining the growth potential of the mandible [6]. TMJ prosthesis can be proposed for severe deformities in adults to correct both the mandibular asymmetry and the joint function [7]. In the case of functional TMJ, both distraction osteogenesis and ramus lengthening procedures are usually proposed. Cases of TMJ ankylosis following distraction have been described requiring caution when correcting mandibular vertical defects [8]. Conventional orthognathic procedures associating Le Fort I osteotomy and/or bilateral sagittal split osteotomy (BSSO) can be compromised especially for treating the anterior open bite in severe cases of PVI [9]. Handelman et al. described a 19% rate of condylar resorption in patients presenting pre-operative high mandibular angles and PVI [10]. We previously described and evaluated two surgical techniques for mandibular lengthening. The vertical ramus osteotomy (VRO) according to the Caldwell-Letterman technique with extra-oral approach revealed good architectural and aesthetic outcomes in patients with unilateral PVI [3]. The intra-oral ramus vertical lengthening osteotomy (IORVLO), consisting in a modification of the Obwegeser technique, takes its advantages in an intra-oral approach and less resorption of the mandibular angles [11]. To our knowledge, there is no study investigating the condylar changes after mandibular ramus lengthening procedures. We hypothesize that IORVLO induces significant changes due to the adaptation of the condyles and the mandible to the new occlusal and functional conditions. Furthermore, as post-orthognathic surgery adaptations may include bone remodeling and resorption of the mandibular ramus and condyles, we hypothesize that patients undergoing surgical correction of PVI are at risk of condylar resorption because of the increased vertical condylar strain and loading. Indeed, with the condyle in maximal push within the extremity of the glenoid cavity, and due to the compressive force exerted by the pterygo-masseteric sling, bone remodeling of mandibular condyle is likely to occur. Moreover, as most studies examined condylar modifications after osteotomy on two-dimensional cephalograms, it can be a source of bias related to superimposition of anatomical structures and it gives incomplete information about condylar complex anatomical changes. The aim of this study was to investigate the three-dimensional (3D) morpho-anatomical changes of the ramus-condyle unit and occlusal stability after IORVLO for the treatment of PVI.

Materials and methods

Data collection

This retrospective observational study included patients who underwent uni- or bilateral IORVLO procedure at the Nantes University Hospital (Nantes, France) between January 2014 and December 2021. Patients were identified using the hospital’s electronic medical records system. The inclusion criteria encompassed patients who had undergone IORVLO surgery for PVI correction, regardless of the indication, and who had both immediate and 1-year post-operative clinical examinations and cone-beam computerized tomography (CBCT) scans available. The patient medical charts were reviewed, and data documenting the age, gender, surgical indication, affected side, possible postoperative complications, follow-up duration and occlusal relapse were compiled.

Surgical technique

IORVLO was performed as previously described by Grimaud et al. in this clinical indication [11]. Osteotomy line started from the lingula and followed the mandibular external oblique ridge up to the middle or distal part of the second mandibular molar. The anterior vertical line stopped about 5–6 mm above the basilar edge. Then, the lower cutting line reached the posterior ramus edge, parallel to the basilar edge, leaving the mandibular angle and basilar edge on the medial valve and allowing simultaneous posterior vertical lengthening and sagittal mandibular lengthening. Fixation was obtained with a horizontal four-hole anterior plate combined with a posterior and vertical two or four-hole plate according to surgeon’s preference. A posterior open bite was created on the affected side, and a preformed interocclusal splint was positioned and progressively reduced in length to promote secondary maxillary teeth egression. The IORVLO could be associated with other conventional orthognathic procedures including Le Fort 1 osteotomy (LF1), contralateral sagittal split osteotomy (SSO) and/or a genioplasty. An orthodontic preparation was usually needed, and the postoperative intermaxillary elastic fixation lasted 6 weeks.

Workflow of three-dimensional (3D) imaging analysis

Image acquisition was performed in preoperative time (T0), immediately after surgery (T1) and at least 1 year after the procedure (T2) using a CBCT device (NewTom VGi, QR, Verona, Italy). Following parameters were used for image acquisition: wide field 15 × 15 cm, 110 kV, tube current and exposure time adjusted automatically by the machine according to the scout views. The DICOM files were anonymized and processed using the open-source software 3DSlicer version 5.2.2 (www.slicer.org) and SlicerCMF (cmf.slicer.org) with dedicated tools of 3D imaging analysis. Automated orientation of T1 was performed according to the Frankfurt and the midsagittal planes. Imaging analysis was performed by the same examiner (S.B.) as illustrated in Fig. 1.

Fig. 1.

Workflow of the three-dimensional analysis. T1 immediate postoperative imaging, T2 last postoperative imaging, 3D three-dimensional

The post-surgical CBCT images were automatically aligned with the oriented T1 scans using voxel-based superimposition [12]. The reference area for measurement was chosen as the cranial base which remained unaffected by the surgical procedure. The creation of virtual Toolkit (vtk) models was performed automatically using an artificial intelligence-based tool [13]. Automated segmentation of cranial base, maxilla and mandible was achieved. When required, the segmentation underwent additional refinement using the ITK-SNAP software (version 3.8.0; http://www.itksnap.org). Automated identification of landmarks was conducted to predict the precise localization of the cephalometric landmarks essential for the analysis [12]. All the landmarks were then manually controlled and adjusted. Chosen landmarks and related anatomical parameters are listed in Fig. 2 and Table 1.

Fig. 2.

Anatomical parameters and cephalometric landmarks studied

Table 1.

List of cephalometric landmarks used for the 3D analysis

Reference landmarks
Skull landmarks
Nasion	Na	Middle of the fronto-nasal suture
Metanasion	M	Intersection frontal, nasal and maxillary sutures
Sella	S	Middle of the turcic sella
Basion	Ba	Most anterior point of the foramen magnum
Porion	Po	Highest point of the external auditory duct
Orbitale	Or	Most inferior point of the inferior orbital rim
Maxillar landmarks
Anterior nasal spine	ANS	Extremity of anterior nasal spine
Posterior nasal spine	PNS	Most posterior point of the palatal bone
Naso-palatal	Np	Anterior ridge of the naso-palatal canal
Mandibular landmarks
Pogonion	Pg	Most anterior point of the mandibular symphysis
Mentalis	Me	Lowest point on mandibular symphysis
Gonion	Go	Intersection of the tangent at the posterior edge of the ramus and the basilar edge of the horizontal branch of the mandible
Condylar landmarks
Condylion	Co	Highest point of the condyle
Lateral condylian	LCo	Most lateral point of the condyle
Medial condylian	MCo	Most medial point of the condyle
Dental landmarks
Upper molars	U6M	Extremity of the disto-vestibular cuspid of the maxillary first molar
Lower molars	L6M	Extremity of the disto-vestibular cuspid of the mandibular first molar

Both qualitative and quantitative analyses were performed. The qualitative analysis involved overlaying skeletal models at different time-points and subtracting corresponding points of T1 and T2 for shape analysis. The corresponding skeletal changes were visualized using software-generated color maps after selecting values from the color bar. The aim of the quantitative analysis was to compare positional changes of each selected landmark between T1 and T2. We obtained displacement vectors and their components in antero-posterior (x-axis), left–right (y-axis), and supero-inferior (z-axis) directions using the automatic quantification of 3D components (AQ3DC) tool. Results were recorded as displacement vectors within a 3D coordinate system represented by (x, y, z). Positive x, y, and z values indicated anterior, lateral, and cranial displacement due to condyle symmetry. Additionally, the root mean square 3D displacement was calculated. The midpoint coordinates were automatically computed in all three dimensions for each paired landmark. 3D angular measurements (in degrees, °) were decomposed into yaw, pitch, and roll components.

Endpoints

The primary endpoint was the difference of condylar height (CH), defined as the height between the highest point of the condyle and a plane passing through the lowest point of the sigmoid notch and parallel to the Frankfurt plane, between T1 and T2 postoperative CBCTs. Correlation coefficients were calculated to isolate predictive factors of CH reduction. The secondary endpoints included: the changes in condylar parameters (condylar diameter (CD) defined as the LCo-MCo distance, condylar axis angle defined as the angle between the LCo-MCo line and the mid-sagittal plane, condylar volume), mandibular parameters (ramic height (RH), Frankfort-mandibular plane angle (FMA), gonion position, intergonial distance, angular remodeling), maxillary stability, and dental extrusion (Fig. 2).

Statistical analysis

Statistical analysis was performed using SPSS Statistics for Windows (Version 26.0 IBM Corp. 2019, Armonk, NY: IBM Corp). Intra-rater reliability was determined using the intra-class correlation coefficient (ICC) for a single operator. Data were analyzed using descriptive statistics: absolute and relative frequencies for categorical data, mean, and standard deviation for continuous quantitative variables. Statistical significance of T1 and T2 distances and angles differences was tested using two-tailed paired t tests (alpha = 0.05). Correlations coefficients were calculated by Spearman correlation tests.

Results

Study population

Twenty-seven patients received uni- or bilateral IORVLO during the study period; six patients were secondarily excluded due to incomplete medical records. The study ultimately included 21 patients, consisting of 17 females and four males with a mean age of 23.7 ± 3.9 years (ranging from 15 to 41 years old). Among the etiologies reported, most patients presented with bilateral condyle hypoplasia resulting in anterior open bite and hyperdivergent class II malocclusion (n = 10), and idiopathic class II malocclusion without anterior open bite (n = 7) (Table 2). The mean pre-operative FMA angle was 33.9 ± 1.3°.

Table 2.

Characteristics of study population

Patients’ characteristics
Age, mean ± SD (years)	23.7 (± 3.9)
Indications for IORVLO, n (%)	10 (47.6)
Bilateral condyle hypoplasia	7 (33.3)
Idiopathic Class II malocclusion	2 (9.5)
Juvenile polyarthritis	1 (4.8)
Oral-facial-digital syndrome (OFDS) type 1	1 (4.8)
Unilateral condyle hypoplasia
Preoperative symptoms, n (%)	4 (19.0)
Lingual dysfunction	6 (28.6)
TMJ disorders
Surgical procedures, n (%)	4 (19.0)
Unilateral IORVLO	17 (81.0)
Bilateral IORVLO	17 (81.0)
Associated LF1	19 (90.5)
Genioplasty
Mean surgical vertical lengthening	13.3 ± 0.76 mm
Pre-operative FMA angle	33.9 ± 1.3°

Seventeen patients benefited from bilateral IORVLO, and four from unilateral IORVLO (three on the left side and one right side); an associated contralateral SSO was performed in the unilateral cases. An associated LF1 osteotomy was performed in 17 patients (monobloc maxillary osteotomy (n = 15) or fragmented osteotomy (n = 2)); 19 patients received an associated genioplasty (Table 2). Mean ramus lengthening was 13.3 ± 0.76 mm. All the patients had immediate postoperative class I occlusion; one patient exposed clinical occlusal relapse of an hyperdivergent Class II with anterior open bite condition after 1 year.

Measurement methods validation

Control points positioned on not-affected cranio-facial areas (Na, M, S, Ba) showed a mean displacement < 0.2 mm between T1 and T2. The intra-class correlation test realized to control reproducibility of results turned out a coefficient of 0.99, indicating a good reliability of the results.

Primary endpoint

Out of the 38 condyles studied, there was a significant mean reduction of 0.66 ± 0.57 mm in CH between T1 and T2 (p = 0.03) (Table 3). No significant correlation was observed between the changes in CH and the primary FMA angle, the thickness of the mandibular basilar bone, the lengthening amplitude, or the patient’s age. Out of all the patients, the one who had a relapse of occlusal and anterior open bite was also the only one who experienced a CH loss of over 2.0 mm. This patient had a CH loss of 1.8 mm for the left condyle and 5.2 mm for the right condyle (Fig. 3).

Table 3.

Quantitative analysis

Quantitative analysis	T1	T2	T2-T1	p
Primary endpoint: condylar height (mm)	11.94	11.28	−0.66	0.03
Secondary endpoints:
Condylar parameters
Diameter (mm)	15.39	14.44	−0.95	0.001
Yaw (°)	62.91	65.18	2.27	0.03
Roll (°)	83.37	81.28	−2.09	0.02
Volume (mm3)	775	724	−7.70	0.01
Mandibular parameters
Ramus height (mm)	57.90	54.34	−3.56	< 0.001
FMA (°)	23.5	29.2	5.70	< 0.001
Intergonial distance (mm)	82.6	82.18	4.42	< 0.001
Displacement of anatomical landmarks	Vertical	Sagittal	Frontal	3D
Control
Na	0.02	0.06	0.05	0.05
M	0.12	0.10	0.05	0.15
Ba	0.01	0.05	0.04	0.08
Mandible
Me	1.23	4.30	0.30	4.66
Co	1.03	0.22	−0.31	1.66
LCo	1.29	−0.94	−0.171	2.11
MCo	1.57	0.25	−0.4	1.82
Go	5.07	0.54	−2.28	7.5
Maxilla
ANS	0.30	0.31	0.10	0.49
PNS	0.08	0.10	0.09	0.02
Np	0.20	0.14	0.05	0.13
Dental
Upper molars	−1.06
Lower molars	1.72

Fig. 3.

T1 (red) and T2 (beige) superimpositions of the right condyle of the only patient exposing occlusal relapse. We observe a circumferential resorption affecting not only the condylar height but also diameter of head and neck

Secondary endpoints

Condyles exposed a loss of diameter of 0.95 ± 0.5 mm between T1 and T2 (p = 0.001). Condylar axis angle showed a mean yaw lateral rotation of + 2.27° (p = 0.03) and a mean roll horizontalization of − 2.09° (p = 0.02). A statistically significant condylar volume decrease of 7.7 ± 0.3% (p = 0.01) was observed. Upon analyzing all the data, we found that the condyle had reductions in height, diameter, and volume. Additionally, there was lateral rotation in the axial plane and flattening, with more resorption being observed in the anterior and superolateral parts. Co point exposed a mean displacement of 0.22 ± 0.8 mm, −0.31 ± 0.28 mm, and 1.03 ± 0.34 mm in x-, y-, and z-axis, respectively. The only patient who had occlusal relapse showed a condylar height loss of 5 mm, a condylar diameter loss of 3.4 mm (28%), and a 33% condylar volume reduction (Fig. 3). We observed a significant difference in the reduction of condylar volume depending on whether the patient had a pre-operative anterior open bite (p = 0.049), but we did not find any significant differences in condylar diameter (p = 0.06) or CH (p = 0.08). The condyle was positioned higher, more medially, and anteriorly in the glenoid fossa 1 year after the ramus lengthening procedure when observing Co, LCo, and MCo displacements (Table 3).

The qualitative analysis showed that the superior, antero-lateral condylar surfaces, and posterior condylar neck were the most affected by bone resorption (Fig. 4). No major remodeling was observed except for the one patient with occlusal relapse, in whom a circumferential resorption of both condyles was observed.

Fig. 4.

Qualitative analysis of left condylar morphology. Superior line: Superimpositions of immediate (T1, red) and 1-year post-operative (T2, beige) condylar reconstructions. Inferior line: point-to point 3-D models comparison. Note the superior and lateral condylar extremities being more affected by resorption

Regarding the displacement of maxillary landmarks (ANS, PNS, Np) in both operated and non-operated maxillary bones, non-operated patients did not show any significant displacement, while patients who underwent a Le Fort I osteotomy had a mean 3D distance between T1 and T2 that was less than 0.5 mm in all maxillary landmarks.

The study of the mandible showed that the distance between gonion and condylion decreased by 3.56 ± 1.16 mm (p < 0.001) between T1 and T2 (Fig. 5). Gonion (Go) showed a mean advancement, medialization, and ascension of 0.52 ± 1.33 mm, 2.28 ± 0.72 mm, and 5.07 ± 0.98 mm, respectively. There was a significant transversal mandibular narrowing with a decrease in intergonial distance of 4.4 ± 1.8 mm (p = 0.004). The mean post-operative FMA angle was 23.6 ± 1.86°, and there was a mean FMA opening of 4.9 ± 1.18° (p < 0.001) at the 1-year follow-up. There was no significant correlation between the thickness of the inner valve posterior segment and gonion ascension (ρ = − 0.133; p = 0.471) (Table 4).

Fig. 5.

Qualitative analysis of mandibular morphology. Left: Mandibular modifications between immediate post-operative (beige) and 1-year post-operative (red). Right: point to point 3-D models comparison. Red and yellow zones represent the bone resorption zones, and blue zone represent the bone creation zones. Green zones are identical between T1 and T2 models. Main mandibular remodeling was observed around the gonial area in all the patients. Bone resorption occurred around the posterior end of the osteotomy line, affecting the lowered mandibular angle, posterior part of basilar edge, and posterior edge of the ramus. All the patients presented posterior ramus resorption, the bone loss increasing from superior to inferior ramus. Osteogenesis was observed between internal and external valve in all the patients. No reshaping was observed in the sigmoid notch and coronoid process areas

Table 4.

Correlations table between pre-operative, surgical parameters, and anatomic parameters represented by the Spearman correlation coefficient ρ

	Δ in condylar height	Δ in condylar diameter			Go ascension		Volumic loss
	ρ	p	ρ	p	ρ	p	ρ	p
Age	0.2	0.23	0.12	0.47	−0.31	0.07	0.14	0.44
Ramus lengthening	−0.15	0.39	0.05	0.77	0.06	0.72	0.16	0.39
Supra-basilar thickness	0.23	0.2	0.14	0.43	−0.13	0.48	0.02	0.91
Pre-op FMA	−0.02	0.9	0.21	0.22	−0.11	0.93	−0.11	0.95
T1–T0 FMA	0.13	0.43	0.12	0.53	−0.29	0.09	0.12	0.44

Discussion

Condylar resorption and orthognathic surgery

Condylar shape is determined by its mechanical regime depending on its position and movements, and orthognathic surgery can alter the strains applying on mandibular condyles. New compression forces induce condylar resorption and make the stable condyle position finding critical in orthognathic surgery. Indeed one of the factors described in literature explaining BSSO-induced resorption is the new positional change of the condyle after mandibular rotation [14]. The development of CR is believed to be caused by the production of oxidative free-radicals [15] and hypoperfusion/reperfusion due to mechanical shearing forces, resulting from condyle malposition. Inflammation caused by either systemic auto-immunity or condylar malposition can lead to an imbalance between bone apposition and resorption, particularly in stressed condylar location on the superior or antero-superior condylar sides in BSSO [16] and leading to a decrease in the PVD. This PVD loss can result in facial asymmetry or anterior open bite, which are the conditions that the surgery was intended to treat. While some studies explored the management of CR management through pharmacology [17] or secondary TMJ surgery [18, 19], the first approach should be to detect at-risk situations, manage pre-operative risk factors, and perform the most appropriate surgical treatment. In cases where hyperdivergent class II patients are already at risk for CR risk factors preoperatively, our study shows that IORVLO can provide satisfactory results in terms of occlusal and bone remodeling after 1 year.

Condyle resorption may affect 0.3% of patients after an orthognathic procedure [20, 21], reaching 19–21% of the patients with high mandibular angles (i.e., with hyperdivergence) [10]. Risk factors for bone resorption include patient-related factors such as age, gender, high mandibular angles, hormonal impregnation, and procedure-related factors such as two-jaw surgery, clockwise rotation of maxilla, and long sagittal lengthening [22–25]. In our study, the only patient who experienced condylar resorption and relapse had all these risk factors. Condyle resorption is critical to comprehend and prevent as bone loss is a major cause of occlusal relapse and surgical failure [26]. While many authors have searched the effect of orthognathic procedures on condyle remodeling, only a few studies have reported on 3D changes, and those few were only for conventional mandibular advancement [27] or setback [28]. Bidimensional assessments can be biased due to the superimposition of the complex and variable anatomic structures of condyles. This study is, to our knowledge, the first to investigate the condylar changes after IORVLO and to assess the modifications of the condylar unit in response to the mechanical strain regime induced by IORVLO. With IORVLO, leaving the mandibular angle on the inner valve limits the frontal displacement of rami and condyles [29] and applies limited stress to the condyle, thus limiting remodeling. However, an increase in vertical stress is possible due to the compression of the condyle in the glenoid cavity under the action of the pterygo-masseteric sling. An excess of vertical stress may lead to TMJ ankylosis as observed in rami vertical distraction inflicting extreme vertical strain on the joint and condyle [8], illustrating the extreme caution needing to be taken when treating PVI.

If condylar resorption was at work in our series of patients, its amplitude was limited, and no ankylosis was observed after 1 year. The definition of condylar resorption varies in the literature. Hoppenreijs et al. considered pathologic condylar resorption a mean ramus height loss over to 6% [30]. In a series of 56 patients benefiting from an advancement BSSO for mandibular hypoplasia, Xi et al. considered a resorption of more than 17% to be responsible for a possible surgical relapse [31]. Regarding the other condylar parameters, we found a slight height loss of 0.66 mm and volumetric loss of 7.7%. These results are less than those observed by Wang et al. [32] who showed a height loss of 1 mm and volumetric loss of 12% in a condylar shape analysis at 1-year following orthognathic surgery. Yin et al. reported a mean condylar height loss of 1 mm, with 25% of patients presenting more than 2 mm of height loss [33]. Moreover, we found no correlation between the amplitude of surgical movement and condylar remodeling. These results are consistent with those of Park et al. who showed no correlation between the condylar volume and the skeletal movement in class III patients [34]. In contrast, Xi et al. found a significant correlation between pre-operative high mandibular angle and volumetric resorption after BSSO [31]. There are various explanations for the low rate of condylar remodeling after IORVLO: [1] the surgical procedure itself in which the mandibular angle is separated from the ramus condyle unit with a limited effect on condylar position into the glenoid cavity, as evidenced by the low displacement of the condylar points; [2] the use of a thick occlusal splint consistently for at least 6 weeks after the procedure that could reduce stress on the condyle. Our results highlighted the spontaneous closure of the open bite generated by progressive dental extrusion (mean molar extrusion of 1.02 mm for upper molars).

Mandibular reshaping

Regarding the changes in ramus parameters, a postoperative loss was observed in ramus height, mainly due to the bone remodeling in mandibular angles rather than the condylar resorption. Patients should be informed about this bone loss despite important gain of vertical height with a mean lengthening of 13 mm, and no functional consequences in terms of occlusion or TMJ function are to be expected. Other authors described the same modifications of the jaw line and the mandibular angles which they attributed to detachment of the medial pterygoid muscle during the procedure [35, 36]. Postoperative resorption of mandibular angles may explain the cephalometric accentuation of hyperdivergence evidenced by the FMA angle widening after 1 year without any occlusal relapse. In the frontal plane, post-operative stability of gonial angles was described [29], but no long-term stability was assessed before this study. Our results show that there is no long-term widening of gonial angles, but a global narrowing due to the gonial resorption. Finally, mandibular angle remodeling can cause the lower screw to loosen, then we recommend to remove the osteosynthesis material 6–12 months after the procedure.

Management of PVI

Unlike the correction of mandibular sagittal defects which is well codified, the management of PVI has been little studied. The BSSO, as described by Epker and Dal Pont, does not allow the correction of the PVI because of the required alignment of mandibular basilar edge. Obwegeser split may modify the PVD, but the technique is difficult to apply to patients with facial hyperdivergence and thin mandibular angles. In severe facial hyperdivergence with anterior open bite, the association of a BSSO to a LF I procedure requires important clockwise rotation of the maxilla with high risk of relapse. Ferri et al. [37] modified the Wolford BSSO technique to allow vertical augmentation by sectioning pterygomasseteric sling, specifically to correct hyperdivergent class II malocclusion. Inverted L-osteotomy can also be proposed in association to bone grafting because of the small bony surface contact, allowing a good correction of the PVI but with a certain donor site morbidity [38]. The Caldwell-Letterman technique, initially described for the treatment of class III malocclusion, consists in a retrospigian vertical ramus osteotomy and can be proposed to correct vertical and sagittal mandibular insufficiencies [39]. This technique offers good clinical results; however, it is most often performed using an extra-oral approach and does not modify mandibular angle position and has therefore a limited esthetic effect on the jawline. Kater and Paulus described a short oblique osteotomy which can have consequent vertical component if performed with an oblique enough split [40]. We currently propose the technique described by Grimaud et al. [11] allowing for a good mandibular lengthening with an intra-oral approach and a supra-basilar osteotomy line [41]. His study is the first to assess the clinical stability and anatomic remodeling at 1 year obtained after this procedure. These data are encouraging in proposing IORVLO as the procedure of choice for mandibular lengthening in PVI patients with functional TMJ.

Study limitations

Despite this being the first study to investigate condylar bone remodeling in 3D after mandibular lengthening, this study suffers from some limitations. The first lies in the inclusion of a small number of patients. The results are nonetheless robust, given the use of relevant software tools, and the time and experience required to carry out this type of assessment. Moving forward, further orthognathic follow-up and consequences studies should incorporate 3D analysis in our sense, given the promising tools being developed by different teams to ease 3D cephalometrics [42]. Additionally, since our analysis provides precise data on changes in condylar anatomy, prospective studies are necessary to determine the long-term clinical implications of these changes, especially concerning the TMJ function and symptoms, and modifications to the jawline due to the gonial angular remodeling.

Conclusion

This study confirms the overall stability obtained with IORVLO for the correction of PVI correction. Despite significant mandibular lengthening, usually exceeding 1 cm, no major condylar resorption was observed. However, a loss of mandibular ramus bone height was observed, mainly due to a bone remodeling in the mandibular gonial angle area that could interfere with the esthetic results of the jawline in patients.