The study was designed as a prospective study including patients who underwent a unilateral mandibular reconstruction with preservation of the condylar head at Lanzhou University’s Second Hospital’s maxillofacial surgery department between October 1, 2019, and March 1,2022, However, the last patients were included in September 1, 2021, with a follow-up period of at least six months. The study was approved by the ethical committee of the Stomatology School of Lanzhou University with the approval No (LZUKQ-2019-047). The study followed the guidelines of the Declaration of Helsinki.

The inclusion criteria were as follows: consecutive patients above the age of 18 with a unilateral mandibular tumor, mandibular reconstruction with condylar head preservation, patients who have CBCT in all follow-up steps with a follow-up after at least six months, and no history of previous TMJ surgery. Patients with a bilateral reconstruction, early recurrence of malignancy, immediate flap failure due to venous thrombosis, and patients without CBCT were excluded.

The sample size was calculated using G. power analysis software (University of Dusseldorf, Dusseldorf, Germany) at a significant level equal to 0.05, power equal to 99%, and an effect size of differences in the measure of Superior joint space at three-time intervals as reported in a previous study [11]. The power analysis revealed a need for 30 subjects to be enrolled in the present study.

CBCT one week before surgery T1, two weeks after surgery T2, then after at least six months T3. All CBCTs and related data forms were collected, and slices were adjusted to obtain the same value as the CBCT image. The Cone Beam Computed Tomography analysis was achieved from Cone Beam CBCT images by New Tom VGI Imaging System (QR R Italy); all Subjects were scanned using a standard protocol which included a 16.0 cm x 16.0 cm field of view, standardized head position, maximum teeth intercuspation, the horizontal plane (HP) parallel to the floor, exposure parameter settings (tube voltage =110 kV, filament current = 29.43 mAs, total scan time = 1.8 s), and image acquisition at 0.3 mm voxel size. The TMJ images were analyzed in 3D using the Invivo Anatomage version 6 software (Anatomage, San Jose, CA, USA). To assess the mandibular condylar position relative to the cranial base, we evaluated according to the relationship of the condyle with the horizontal plane, vertical plane, and mid-sagittal plane. TMJ space measures were defined by selecting TMJ points and analyzed as a 3D mold based on three planes: coronal, axial, and sagittal (X, Y, and Z). Tables 1, 2, 3 and 4 show 3D craniofacial skeletal landmarks, reference planes and lines temporomandibular landmarks, 3D measurements of the condylar position, and three-dimensional measurements of the mandibular condyle and TMJ spaces. Figure 1 shows the cranial-facial skeleton reference planes and lines. Figure 2 demonstrates measurements of the condylar position on the tumor and non-tumor sides.

Three planes. Three planes. MSP: midsagittal plane, VP: vertical plane, HP: horizontal plane

A Condyle angulation and position in the vertical plane (VP), CdA: anterior condylar point, CdP: posterior condylar point, V-CL: vertical condylar inclination; B Condyle position and angulation in the midsagittal plane (MSP). AP-CL Anteroposterior condylar inclination, CdM: medial condylar point, CdL: lateral condylar point, CdS: superior condylar point, C-VP: vertical condylar position, C Position, and angulation to the horizontal plane (HP), CdM: medial condylar point, CdL: lateral condylar point, CdS: superior condylar point

The orientation of the coordination system was set based on the points of facial skeletal midline: Nasion, incisive foramen, and basion, confirmed by Garcia et al. as correct sites and validated by the previous studies with different objectives [12,13,14]. The lateral landmarks were determined by the right orbital (Or) point and two porion (Pr) points; second, the following landmarks were adjusted to the exact position of the selected points and digitized separately using three planes slice locator (sagittal, coronal, and axial) Fig. 3.

Three planes slice locator

Volumetric joint space: A TM line was drawn from the inferior point of the auditory meatus (IM) to the inferior point of the articular tubercle (AT) to determine the inferior border of the whole joint space Fig. 4, To measure the whole joint space we used the cubic three-dimensional analysis of volumetric joint space by sectioning the whole joint space into six sections, where every section had a thickness of 1.5 mm and a width of 80 mm at an interval of 0.75 mm, and then calculated the space with the equation of sigma volume:

$$v\cong _ A\left(_}\right)\Delta x$$

Three-dimensional Volumetric measurement of the TMJ spacel

To further evaluate the reproducibility of the results, a random sample of 30% of the total analyzed sample was measured twice at three-week intervals by the same assessor (S.N).

Gender, age, primary location, pathology, and defect type were all recorded for each patient. For the clinical outcome, the study used the Helkimo index scoring system, which accurately assesses the TMJ function [15, 16]. Helkimo Ai represents subjective symptoms of TMJ dysfunction, whereas the Di index evaluates the TMJ dysfunction by clinical assessment of TMJ pain, impaired TMJ function, muscular pains, and reduced mouth movement.

In this study, the Ai was obtained and divided into three categories: Ai 0, Ai I, and Ai II. Ai 0 indicates no symptoms are present. Ai I denotes moderate symptoms with at least one of the following: stiffness during exercise, muscle fatigue, and stiffness in the morning or during exercise. Ai II indicates severe symptoms, including at minimum one of the following: joint noise, locking of the joint, mouth opening restriction, dislocation, mandibular pain, and masticatory muscle pain or TMJ caused by motion. For objective clinical assessment (Di), joint dysfunction, mouth discomfort, limited mandible movement, and muscular soreness were used. Every feature was given a score ranging from 0 to 5 on a scale of 1 to 5. The total score for each feature was as follows: 0 = Di 0 (normal function), 1–4 = Di I, 5–9 = Di II, and 10–30 = Di III (worst function) [17].

Two highly qualified surgeons with over twenty years of experience treated all reconstruction cases. A two-team procedure was used, the head and neck team performed the mandibular ostectomy, in view of surgical margin and subsequent fixation, and preserved the condylar head in the glenoid fossa, and reproduced the original occlusion and condylar position after flap positioning. The other team used the approach to harvesting the fibula flaps. All flaps were osteotomized according to the defect size and according to the design of virtual surgical planning using a cutting guide template.

For contouring. Intermaxillary fixation was used for occlusion (IMF). The condylar head and fibula were fixed in place using reconstruction plates. After that, microsurgical vascular anastomosis was performed. After resection and reconstruction of the defect, we performed IMF for two weeks, then removed and made the second CBCT T2.

The Statistical Package for Social Sciences (IBM SPSS Statistics) software, version 25 (IBM Corp.) for Windows, was used to analyze the data. This study includes 30 patients, which means we have 60 joints (30 joints on the Tumor sides and 30 joints non-Tumor side). The significance level was set at P < 0.05. Shapiro-Wilk test was used to check for normal distribution. The repeated-measures analysis of variance (ANOVA) was conducted to check the presence of different time-dependent changes (T1–T2, T1–T3, and T2–T3) of the condylar positional change within the tumor side and non-tumor side groups, as well as in between them.

To evaluate reproducibility and reliability, researchers used intra-class correlation coefficients (ICCs) and absolute and relative technical error of measurement tests (the absolute technical error of measurement [TEM], relative TEM [rTEM], and a coefficient of reliability [R]). The student’s t-test was used to statistically analyze data between tumor and non-tumor sides. It was carried out using GraphPad Prism 9, which efficiently performs basic statistical tests commonly used by laboratory and clinical researchers.