Growth Outcome and Jaw Functions Are Better After Gap Arthroplasty Plus Costochondral Graft Reconstruction Than Gap Arthroplasty Alone in Pediatric Temporomandibular Joint Ankylosis Patients: A Cluster Randomized Controlled Trial

It is not known if the muscle matrix that becomes functional after gap arthroplasty (GA) in temporomandibular joint ankylosis (TMJA), induces growth of the mandible or the reconstructive arthroplasty with costochondral graft (CCG) is responsible for growth. The study aimed to evaluate the mandibular growth and functional outcome with the use of CCG/GA in the management of pediatric TMJA. The investigators designed a cluster randomized controlled trial on pediatric (3 to 16 years) TMJA patients. Treatment applied (CCG and GA), was the primary predictor variable. Patients were divided into CCG and GA groups. The primary outcome variable was growth. Secondary outcome variables included etiology and duration of ankylosis, maximal incisal opening (MIO), reankylosis, occlusion, laterotrusion, chin deviation, facial asymmetry, occlusal tilt, and complications. The distance condylion (Co) to gnathion (Gn) was used to measure mandibular length. Ramal height was measured from Co- gonion (Go). Lower facial height was measured from the anterior nasal spine to Gn. Generalized estimating equations were used to calculate the regression coefficient adjusted for the cluster. The patient was considered as a cluster and the unit of analysis was joint. Fifty-six {n = 28 in each group, (n = 33 joint in the CCG group and n = 31 joints in GA group)} patients were analyzed. The median follow-up was 33-months (31.93 ± 15.24) in the CCG group and 32-months (32.85 ± 17.84) in the GA group. Intergroup comparison between the CCG and GA group showed a statistically significant difference in mandibular length (CCG = 77.51 ± 9.31 and GA = 66.66 ± 8.32 mm, P < .001), ramal height (CCG = 44.21 ± 7.3 and GA = 31.87 ± 8.4 mm, P < .001), and statistically insignificant difference in lower facial height (CCG = 52.53 ± 6.1 and GA = 50.19 ± 6.3 mm, P = 0.14) at follow-up. Statistically, significant improvement was seen in MIO in both groups (