This study aimed to compare the joint function and morphology achieved following condylar reconstruction using sternoclavicular grafts (SCG) versus transport distraction osteogenesis (TDO) in temporomandibular joint (TMJ) ankylosis patients. Twenty-two patients with TMJ ankylosis underwent TMJ reconstruction with SCG or TDO ( n = 11 each). Radiographic and clinical evaluations were performed at 1 week and at 1, 3, and 6 months post-surgery. Clinical criteria examined included the duration of surgery, mean postoperative mouth opening, excursive jaw movements, and pain scores. The radiographic evaluation 6 months postoperatively (computed tomography) included subjective assessment of joint morphology and measurements of the mean condylar height, width achieved, and amount of condylar resorption. The χ 2 test and Student t -test were used to compare qualitative and quantitative variables, respectively. Similar mean mouth opening (SCG = 31.8 mm, TDO = 32.1 mm at 6 months), excursive movements, and pain scores were observed in the two groups throughout follow-up. Mean condylar resorption was significantly greater in the TDO group (TDO = 7.0 mm, SCG = 2.7 mm; P = 0.005). The duration of reconstruction surgery was greater in the SCG group ( P = 0.035). A greater incidence of complications was observed with TDO. In conclusion, based on the protocols used in this study, SCGs are superior to TDO in terms of condylar morphology, stability, and surgical safety.
“It is true that it may be necessary to remove the condyle in cases of ankylosis, but some provision should always be made to reconstruct the joint”. The value of reconstruction of the ramus–condyle unit (RCU) following osteoarthrectomy has been stressed since Smith and Robinson made this modest statement in 1953 . The quest for the ideal reconstructive modality using autogenous tissues has been a matter of contention for surgeons for decades, as they have endeavoured to fashion a temporomandibular joint (TMJ) from innumerable materials, tissues, and virtually every bone in the body. In spite of exhibiting variable success rates, autogenous bone grafts have stood the test of time, yielding adequate joint morphology and function, and they are still considered by some surgeons as a preferred method for TMJ reconstruction .
Among the various donor sites explored, the sternoclavicular graft (SCG), originally introduced in 1971 , remains a favourable option, owing to its unique histomorphological resemblance to the TMJ . Although it has hitherto failed to gain widespread popularity like the costochondral graft (CCG), the role of the SCG in RCU reconstruction may be revisited in an attempt to circumvent the unreliable growth characteristics of the CCG.
However, trends change. The 21 st century has seen a revolution in the art of osseous reconstruction with the introduction of transport distraction osteogenesis (TDO), which represents a promising, less invasive, alternative technique for RCU reconstruction. TDO provides autogenous bone while obviating donor site morbidity and a period of immobilization, which are the undesirable corollaries of autogenous bone grafting. In addition, the early restoration of jaw function and the common histological observation of a pseudo-disc at the joint interface suggest favourable joint function and morphology following TDO .
Some surgeons believe that due to these benefits, TDO supersedes all preceding modalities for RCU reconstruction to the extent that it might replace the autogenous graft as the gold standard in the future. However, comparative clinical studies that definitively support this claim are limited. Animal studies comparing the histology of the condyle reconstructed with the CCG and TDO have been reported, which have established that both techniques are effective modalities for reconstruction . However, human clinical studies comparing the morphological and functional outcomes between the two methods are lacking. Moreover, the overwhelming advantages of TDO have not yet been weighed against the SCG for adult RCU reconstruction.
In light of the above, the current study was conducted with the aim of comparing the morphological reconstruction and the functional outcomes of the RCU achieved with autogenous SCG versus TDO following osteoarthrectomy in adults with unilateral TMJ ankylosis.
Materials and methods
Twenty-three adult patients with unilateral TMJ ankylosis were included in the study and divided randomly (alternately) into two surgical groups based on the reconstruction modality that they would receive ( Table 1 ). One patient was later excluded to maintain uniformity. The prior approval of the institutional review board, permission from the local ethics committee, and written informed consent from the patient/patient’s parents were obtained.
|Group I||Autogenous sternoclavicular graft (SCG)|
|Group II||Transport distraction osteogenesis (TDO)|
Surgery for TMJ ankylosis release
All patients underwent osteoarthrectomy under general anaesthesia via the Al Kayat and Bramley approach ( Fig. 1 A). An ipsilateral coronoidectomy/coronoidotomy, with/without contralateral coronoidectomy/coronoidotomy (via an intraoral approach) was also performed if and when required to achieve the strict goal of an intraoperative inter-incisal opening of ≥35 mm. In both patient groups, the temporal end of the resection was contoured to simulate a glenoid fossa and lined with a partial thickness temporalis myofascial flap ( Fig. 1 B). This was followed by reconstruction of the defect with either the SCG or TDO, with all 22 reconstructions performed by a single surgeon.
Ramus–condyle unit reconstruction
Group I patients received SCGs. An autogenous, non-vascularized, split-thickness, corticocancellous SCG, approximately 4–5 cm in length, was harvested from the medial half of the right clavicle along with the fibrocartilaginous disc attached to its sternal end via a 4–5-cm long horizontal incision made 1–2 cm above the medial end of the clavicle ( Fig. 2 A, B). The right clavicle was selected for all cases due to the operator’s comfort and to maintain uniformity in the study. The graft was then contoured to facilitate its adaptation to the recipient site and fixed, via a submandibular incision, over the lateral aspect of the ramus using positional screws ( Fig. 2 C).
Group II patients underwent TDO. The mandibular ramus was exposed via a submandibular incision, and an L-shaped osteotomy was marked over the lateral aspect of the ramus, representing the transport disc for condylar reconstruction ( Fig. 3 A). The vertical limb of the ‘L’ was designed parallel to a vector that would position the transport disc into the glenoid fossa and would move in close contact with the main segment along its long arm. Within the limits of the short remaining mandibular ramus, the length of the transport disc was adjusted so as to be equal to or exceed the distance to be covered to reach the glenoid fossa.
The cephalic end of the transport disc was rounded off to conform to the condylar anatomy. A corticotomy was performed over the marking and a semi-buried unidirectional stainless steel distraction device (Synthes, Zuchwil, Switzerland) was fixed across it ( Fig. 3 B), with its port projecting externally in the submandibular region. The osteotomy was completed using chisels, and experimental activation of the distractor device was done. The manoeuvre was then reversed to close the gap and maintain bone contact.
The maximum inter-incisal opening achieved intraoperatively was recorded in millimetres: the total inter-incisal distance achieved after actively stretching the jaw using a mouth gag was measured. In addition, the duration (in minutes) of the reconstructive part of the surgery, as well as the total duration of the surgery, was recorded for both groups ( Table 2 ).
|Maximum inter-incisal opening (mm)||After osteoarthrectomy ± coronoidectomy/coronoidotomy|
|Total duration of the procedure (min)||Measured from the beginning of pre-auricular incision to placement of the last suture|
|Duration of reconstruction (min)||Measured from the beginning of supraclavicular (group I)/ submandibular (group II) incision to placement of the last suture|
For group I patients, a clavicle support bandage was provided postoperatively for 6 months and the patient was advised against loading of the right arm with heavy objects or the patient’s own body weight and avoiding trauma to the right shoulder region.
For group II patients, distraction was initiated on day 7 postoperative and was continued at a rate of 0.5 mm every 12 h until resistance was met on further activation or until the predetermined amount of distraction was achieved (as estimated by the size of the intraoperative defect created), after which a radiographic assessment was done via panoramic radiographs to assess the position of the transport disc. A consolidation period of a number of days equalling three times the total amount of distraction performed was adopted , after which the distractor was removed, unless it was indicated for earlier removal due to infection or hardware failure.
Functional jaw movements were started 1 week postoperatively in both groups. Active jaw stretching with a mouth gag was initiated 4 weeks after the surgery, with 8 to 10 daily sessions, each 10 minutes long. This was continued for at least six subsequent months.
Patients were assessed regularly for a minimum of 6 months and the following parameters were employed to evaluate the functional and morphological outcomes of the reconstructed RCU: (1) maximum inter-incisal opening. (2) Maximum protrusive and laterotrusive excursions. (3) Pain on function, assessed using a numerical rating scale with scores varying from 0 to 10, as described by the patient, with 0 representing no pain at all and 10 being the worst possible pain that could be experienced by the patient. All three of these parameters were measured at 1 week and then at 1, 3, and 6 months postoperative at follow-up visits. (4) Condylar morphology, evaluated radiographically at 6 months postoperative on panoramic radiographs and computed tomography (CT) scans using the parameters listed in Table 3 . (5) Any postoperative complications, which were duly recorded and promptly managed.
|Subjective evaluation||Anatomical similarity of the reconstructed RCU to a normal condyle as assessed on panoramic radiographs and 3D reconstructed CT images|
|Maximum width of the formed condyle (mm)||Greatest mediolateral dimension of the reconstructed RCU on coronal CT sections|
|Height of the condyle (mm)||Greatest cephalocaudal dimension of the reconstructed RCU on sagittal and coronal CT sections, measured from the sigmoid notch to the most cephalic point on the reconstructed RCU|
|Amount of resorption (mm)||Distance between the glenoid fossa and the most cephalic point on the reconstructed RCU|
|Degree of resorption||Amount of resorption as measured on the CT sections was graded as:
(a) None, ≤3 mm
(b) Mild, 4–6 mm
(c) Severe, ≥7 mm
The data obtained for the 22 patients were analysed statistically using IBM SPSS Statistics version 20.0 software (IBM Corp., Armonk, NY, USA). The clinical profile of the patients was analyzed using Pearson’s χ 2 test for qualitative variables and the Student t -test for quantitative variables. The 5% probability level was considered as statistically significant, i.e. P < 0.05.
The study was conducted on 22 adult patients suffering from unilateral TMJ ankylosis, presenting at a mean age of 24.5 years (range 18–48 years). The male to female ratio was 1.4:1. The duration of illness ranged from 2 to 23 years (mean 11.1 years), and a mean preoperative mouth opening of 4.8 mm was recorded (range 0–28 mm). There was an equal distribution of right and left TMJ involvement. Six of the 22 patients (27.3%) had undergone a previous surgery for the release of TMJ ankylosis. Gross facial asymmetry was seen in six cases (27.3%); however, all patients were informed that this would be addressed at a later stage.
All patients underwent an aggressive osteoarthrectomy, in which a bony mass ranging from 1.5 cm to 2 cm in size was excised. Following excision, the fossa was lined with a temporalis myofascial flap. An ipsilateral coronoidectomy/coronoidotomy was required in 20 patients (90.9%) and a contralateral coronoidotomy in 14 patients (63.6%) to achieve a mean intraoperative mouth opening of 38.5 mm (range 35–45 mm).
Reconstruction of the RCU was done using an autogenous SCG in 11 patients (group I); a corticocancellous SCG of mean 4.8 cm in length and 1.5 cm in width was used to reconstruct the condylar unit. TDO was used as a reconstructive modality in the remaining 11 patients (group II). The average duration of reconstruction was significantly longer for the SCG group: mean 134.2 min compared to 92.7 min with TDO ( P < 0.05) ( Table 4 ).
|All patients||Group I (SCG)||Group II (TDO)||P -value|
|N = 22||( n = 11)||( n = 11)|
|Mean total duration ofsurgery||192.5||192.9||192.0||0.32|
|Mean duration of reconstruction||113.5||134.2||92.7||0.035|
The total distraction performed in group II patients ranged from 20 mm to 25 mm (mean 24.7 mm) and the average consolidation period was 15 weeks (range 5–27 weeks). Despite the previously mentioned guidelines for the optimum consolidation period, one patient required device removal before the calculated time (removed after 5 weeks of placement) owing to a persistent infection at the distractor site. All patients were followed up for a period of 6 to 33 months. A radiographic assessment was performed at 6 months post-surgery and regular clinical assessments of joint function were done. Adverse events, if any, were noted and appropriately managed. No evidence of re-ankylosis was seen in any of the 22 patients during this period, as evidenced by regular clinical evaluation.
Postoperative mouth opening and excursive movements
Comparable mean mouth opening and excursion values were obtained for the two groups of patients at all stages of assessment ( P > 0.05). At 6 months, mean mouth opening was greater than 3 cm in both groups (mean 31.8 mm for group I and mean 32.1 mm for group II). A considerable progressive increase in mean protrusion and lateral excursion values was seen in both groups; protrusion increased from a mean 1.2 mm at 1 week to 2.4 mm at 6 months. Similarly, between the 1-week and 6-month postoperative visits, a mean increase of 2.3 mm was seen in ipsilateral movements and of 0.9 mm was seen in contralateral movements. Although the observed mean excursion values were higher for the SCG group, the difference between the two groups was not statistically significant ( P > 0.05). Additionally, it was observed that at all stages, ipsilateral movements showed markedly greater values than contralateral movements ( Table 5 ).
|All patients||Group I (SCG)||Group II (TDO)||P -value|
|N = 22||( n = 11)||( n = 11)|
|Mean postoperative mouth opening (mm)|
|Mean protrusion (mm)|
|Mean ipsilateral movement (mm)|
|Mean contralateral movement (mm)|
Pain scores at all follow-up visits were similar in the two groups. On day 1, 90.9% (10/11) of group I patients had a pain score ≤5, while for group II this was 72.7% (8/11). At 3 months, nine patients in group I and all patients in group II had a pain score of 0 or 1. By 6 months, 72.7% of group I patients and 90.9% of group II patients were pain-free.
Sixteen of the 22 patients (72.7%) presented an anterior open bite post-surgically. This was corrected using guiding elastics and/or manual training. A post-distraction occlusal imbalance was seen in three group II patients (27.3%), where the ipsilateral occlusal prematurity caused a contralateral open bite and conversion of a class I molar relationship to a class III relationship.
Subjective evaluation of the 6-month postoperative panoramic radiographs revealed excellent graft uptake and remodelling in all 11 patients in group I, with the superior end assuming a morphology similar to that of a normal mandibular condyle, along with a favourable relationship to the fossa. Furthermore, the inferior end showed positive osseous union with the mandibular ramus ( Fig. 4 A, B). CT scans corroborated these findings and revealed well-healed, minimally resorbed grafts simulating the condylar anatomy ( Fig. 4 C–F).
The reconstructed RCU in group II patients, however, exhibited a more varied morphology. Three cases showed encouraging results with minimal resorption at the leading end of the transport disc and positive contact with the ramus at the ramal end on both the panoramic radiographs and CT images. Regrettably, the shape of the remodelled transport disc did not bear as striking a resemblance to the mandibular condyle as envisioned, although its adaptation to the glenoid fossa was found to be satisfactory in these three cases ( Fig. 5 A–E). However, a large percentage of group II patients (72.7%, 8/11) showed considerable resorption of the transport disc, which was appreciable on both the panoramic radiographs and CT images. Overall, the resulting anatomy was variable; in general, large gaps were seen between the fossa and the leading edge of the disc, with no similarity in shape or form to the mandibular condyle. Some cases displayed irregularly shaped osseous masses in the vicinity of the fossa, with incomplete mineralization of the regenerate at the trailing edge, while others showed transport discs that were conical, bifid, or irregular at the superior end ( Fig. 6 A–D). Notably, complete osseous fusion at the inferior end was not observed in any of the 11 cases.
The mean condylar height and width achieved was greater in the SCG group than in the TDO group. However, the differences between the groups were not statistically significant ( P > 0.05). Overall, the mean condylar height and mean condylar width achieved for the whole patient population were 10.7 mm and 10.5 mm, respectively. Significantly greater resorption was seen in the reconstructed RCUs of group II patients ( P = 0.005), with 45.5% (5/11) of patients showing a severe degree of resorption (>7 mm). None of the group I patients exhibited severe resorption, while 72.7% (8/11) of patients in group I showed no resorption ( Table 6 ).