Abstract
The aim of the present study was to test whether there is a significant difference in the clinical outcomes between surgical and non-surgical treatment of mandibular condylar fractures. An electronic search was undertaken in February 2014. Eligibility criteria included clinical human studies, either randomized or not. The search strategy resulted in 36 publications. The estimates of an intervention were expressed as the risk ratio (RR) and mean difference (MD) in millimetres. A statistically significant effect was observed for the outcome of post-treatment malocclusion (RR 0.46, P < 0.00001), lateral deviation during maximum inter-incisal opening (RR 0.56, P = 0.0001, dichotomous; MD −0.75, P = 0.002, continuous), protrusion (MD 0.68, P = 0.01), and laterotrusion (MD 0.53, P = 0.03) favouring surgical treatment, and for infection (RR 3.43, P = 0.03) favouring non-surgical treatment. There was no statistically significant effect on temporomandibular joint pain (RR 0.81, P = 0.46) or noise (RR 1.44, P = 0.24), or maximum inter-incisal opening (MD 2.24, P = 0.14). The test for overall effect showed that the difference between the procedures significantly affected the incidence of post-treatment complications, favouring surgical treatment, when all dichotomous and continuous outcomes were analysed (RR 0.70, P = 0.006 and MD 1.17, P = 0.0006, respectively).
Approximately 11–16% of all facial fractures and 30–40% of all mandibular fractures (MFs) are fractures of the mandibular condyle. Most are not caused by direct trauma, but follow indirect forces transmitted to the condyle from a blow elsewhere. Consequently, mandibular condylar fractures (MCFs) are those most commonly missed. MCFs have a distinctive position in oral and maxillofacial surgery because, although in many cases good initial clinical results are achieved, serious late complications have been reported such as pain, restricted mandibular movement, muscle spasm and deviation of the mandible, malocclusion, pathological changes in the temporomandibular joint (TMJ), osteonecrosis, facial asymmetry, and ankylosis, irrespective of treatment performed or not.
There are two principal therapeutic modalities for these fractures: non-surgical (functional) and surgical. Historically, non-surgical treatment of MCFs by means of maxillomandibular fixation (MMF) followed by physiotherapy was the standard practice.
Arguments for non-surgical treatment include reduced overall morbidity, in most cases acceptable occlusal results, avoidance of typical surgical complications, a simpler procedure, and less risk of ankylosis and avascular necrosis. However, long-term complications such as pain, arthritis, open bite, deviation of the mandible on opening and closing, inadequate restoration of vertical height of the ramus leading to malocclusion, and ankylosis do occur with non-surgical treatment.
With the development of improved materials for fixation and the refinement of surgical techniques, open reduction and internal fixation (ORIF) has gained higher acceptance by surgeons for the management of MCFs, especially in severely displaced and dislocated fractures, in edentulous patients, in cases of loss of ramus height, and when a closed approach with manipulation cannot re-establish the pre-trauma occlusion or excursions, i.e., the tendency to treat operatively usually increases with increasing complexity of the fracture. The ORIF technique provides stable three-dimensional reconstruction, promotes primary bone healing, shortens the treatment time, and eliminates the need for early release of the MMF. A decreased dependence on MMF improves post-treatment respiratory care, nutritional intake, and oral hygiene measures. However, ORIF of MCFs is technically difficult due to the difficulty in manipulating the fragments in a small area, leaves a visible external scar, results in increased costs and hospitalization time, and carries the risk of facial nerve injury, damage to vessels such as the internal maxillary artery, and wound infection.
There has been considerable controversy regarding the treatment of MCFs, in particular whether they should be treated conservatively or surgically. Moreover, an increasing number of articles in the current literature report good results for surgically treated MCFs compared with non-invasive techniques. As the philosophies on the treatment of maxillofacial trauma alter over time, a periodic review of the different concepts is necessary to refine techniques and eliminate unnecessary procedures. This would form a basis for optimum treatment. Thus, in light of all the reported advantages of the surgical treatment of MCFs, the objective of this study was to conduct a systematic review and meta-analysis of studies published in the literature up to and including February 2014 in order to verify whether there is a significant difference in the clinical outcomes and post-treatment complications between the surgical and the non-surgical treatment of unilateral or bilateral MCFs, in patients of any age or gender.
Materials and methods
This study followed the guidelines of the PRISMA statement. A review protocol does not exist.
Objective
The purpose of the present review was to test the null hypothesis of no difference in the incidence of post-treatment complications for MCFs treated surgically or non-surgically, against the alternative hypothesis of a difference.
Search strategies
An electronic search without time or language restrictions was undertaken in February 2014 in the following databases: PubMed, Web of Science, and the Cochrane Oral Health Group Trials Register. The following terms were used in the search strategy: {Subject AND Adjective} { Subject : (condylar fracture [text words]) AND Adjective : (open closed OR surgical conservative OR surgical nonsurgical [text words])}.
The following terms were used in the search strategy on Web of Science, refined by the research area ‘dentistry oral surgery medicine’ and ‘otorhinolaryngology’: {Subject AND Adjective} { Subject : (condylar fracture [title]) AND Adjective : (open closed OR surgical conservative OR surgical nonsurgical [title])}.
The following terms were used in the search strategy on the Cochrane Oral Health Group Trials Register: (condylar fracture AND (open closed OR surgical conservative OR surgical nonsurgical)).
A manual search of journals on the subject was also performed, including the British Journal of Oral and Maxillofacial Surgery, International Journal of Oral and Maxillofacial Surgery, Journal of Craniofacial Surgery, Journal of Cranio-Maxillofacial Surgery, Journal of Maxillofacial and Oral Surgery, Journal of Oral and Maxillofacial Surgery, and Oral Surgery Oral Medicine Oral Pathology Oral Radiology and Endodontology.
The reference lists of the identified studies and relevant reviews on the subject were also scanned for possible additional studies. Moreover, online databases providing information on clinical trials in progress were checked ( http://clinicaltrials.gov; http://www.centerwatch.com/clinical-trials; http://www.clinicalconnection.com ).
Inclusion and exclusion criteria
Eligibility criteria included clinical human studies—randomized controlled trials (RCTs), controlled clinical trials (CCTs), or retrospective—comparing the clinical outcomes between surgical and non-surgical treatment of MCFs, and reporting the incidence of post-treatment complications. The following were excluded: case reports, technical reports, animal studies, in vitro studies, and reviews papers.
Study selection
The titles and abstracts of all reports identified through the electronic searches were assessed. The full text was obtained for studies appearing to meet the inclusion criteria and for studies for which there were insufficient data in the title and abstract to make a clear decision.
Quality assessment
The quality assessment was performed using the recommended approach for assessing the risk of bias in studies included in Cochrane reviews. The classification of the risk of bias potential for each study was based on the following four criteria: sequence generation (random selection in the population), allocation concealment (steps must be taken to secure strict implementation of the schedule of random assignment by preventing foreknowledge of the forthcoming allocations), incomplete outcome data (clear explanation of withdrawals and exclusions), and blinding (measures to blind study participants and personnel from knowledge of which intervention a participant received). Incomplete outcome data was also considered addressed when there were no withdrawals and/or exclusions. A study that met all the criteria mentioned above was classified as having a low risk of bias. A study that did not meet one of these criteria was classified as having a moderate risk of bias. When two or more criteria were not met, the study was considered to have a high risk of bias.
Data extraction and meta-analysis
The following data were extracted (when available) from the studies included in the final analysis: year of publication, study design, number of patients, patient age range and/or mean age, follow-up period, number of MCFs, associated MFs, fixation methods, surgical approach, length of operation, post-treatment MMF, use of antibiotics and/or chlorhexidine, inclusion criteria for patients, post-treatment radiological assessment, and post-treatment complications. Authors were contacted via e-mail for possible missing data.
The post-treatment complications evaluated were infection, post-treatment disturbance in occlusion, malunion, non-union, TMJ noise/click/sound, and TMJ pain, all dichotomous outcomes. The continuous outcomes evaluated were the duration of the operation, maximum inter-incisal opening (MIO), laterotrusion (lateral excursion), protrusion (protrusive movement), and the lateral deviation during MIO. When provided in different units (number of observations, or mean ± standard deviation), the outcome was evaluated as both a dichotomous and a continuous outcome. The results of different follow-ups were included in the meta-analysis when the information was provided by the study. The incidences of wound dehiscence, facial nerve injury, hardware failure (fracture or loosening), and hardware removal for the surgical group were recorded in a table. The statistical unit for the dichotomous outcomes was the number of MCFs treated by surgery or not. Weighted mean differences were used to evaluate the continuous outcomes.
Whenever outcomes of interest were not clearly stated, the data were not used for analysis. The I 2 statistic was used to express the percentage of the total variation across studies due to heterogeneity, with 25% corresponding to low heterogeneity, 50% to moderate heterogeneity, and 75% to high heterogeneity. The inverse variance method was used for random-effects or fixed-effects model. Where statistically significant ( P < 0.10) heterogeneity was detected, a random-effects model was used to assess the significance of treatment effects. Where no statistically significant heterogeneity was found, the analysis was performed using a fixed-effects model. The estimates of an intervention for dichotomous outcomes were expressed as the risk ratio (RR) and for continuous outcomes as the mean difference (MD) in millimetres, both with a 95% confidence interval (CI). Only if there were studies with similar comparisons reporting the same outcome measures was a meta-analysis to be attempted.
A funnel plot (plot of effect size versus standard error) was drawn. Asymmetry of the funnel plot may indicate publication bias and other biases related to sample size, although asymmetry may also represent a true relationship between trial size and effect size.
The data were analysed using the statistical software Review Manager (version 5.2.8, The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark, 2014).
Results
Literature search
The study selection process is summarized in Fig. 1 . The search strategy resulted in 400 entries. The initial screening of titles and abstracts resulted in 94 full-text papers, of which 13 were cited in more than one research of terms. Assessment of the full-text reports of the remaining 81 articles led to the exclusion of 53 because they did not meet the inclusion criteria: 21 were reviews articles; nine studies did not evaluate post-treatment complications (three evaluating condylar motion, one evaluating masticatory motion, one evaluating bite force, one evaluating occlusion only, one evaluating facial symmetry, one evaluating TMJ anatomical–radiological aspects, and one evaluating intraoperative methods to determine the treatment approach); 10 articles did not correctly report the incidence of post-treatment complications; five studies were performed in animals (in vivo); four articles were not published in English (two in Japanese, one in Korean, and one in Chinese), for which contact with the authors was attempted in order to obtain detailed information, but without success; three papers were the same article published in different journals; and one article did not report the number of fractures and patients in one of the groups. Additional hand-searching of the reference lists of selected studies yielded eight additional papers. Thus, a total of 36 publications were included in the review.
Description of the studies
Detailed data for the 36 studies included are listed in Tables 1 and 2 . When the of one or more authors of an article was available, the author was contacted to obtain any missing data. An was not available anywhere for the authors of three studies. An e-mail was sent to the authors of 33 articles. All e-mails were sent within a period of 24 h. The authors of six (18%) studies replied. Five RCTs, 13 CCTs, and 18 retrospective studies were included in the meta-analysis.
| Authors and year published | Study design | Total patients ( n , per group) | Patient age range (average), years | Follow-up period | Aetiology (%) | MCFs (per group) (level) | Dislocated/non-dislocated MCFs | Bilateral MCFs (number of patients) | Associated MFs ( n ) | MF fixation methods (number of fractures) (G1) | Surgical approach ( n ) (G1) | Post-treatment MMF (number of patients) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Takenoshita et al., 1990 | RA (single centre) | 36 (G1, 16; G2, 20) | NM | 2 years | NM | 36 (16, G1; 20, G2) (NM) | NM | 6 (G1) 3 (G2) |
13 (G1) 13 (G2) |
2.0-mm miniplate, screws, stainless wires, or a Kirschner pin | Pre-auricular + submandibular (16) | 3 weeks (16, G1) 3 weeks (20, G2) |
| Hidding et al., 1992 | RA (single centre) | 34 (G1, 20; G2, 14) | 17–60 (31, G1) 18–50 (29, G2) |
Range 1–5 years | NM | 34 (20, G1; 14, G2) (all neck) | 20/0 (G1) 14/0 (G2) |
0 (G1) 0 (G2) |
NM | Wire or plating osteosynthesis | Submandibular (20) | 2 weeks (20, G1) 2 weeks (14, G2) |
| Moritz et al., 1994 | RA (single centre) | 76 (G1, 25; G2, 51) | NM (23, G1) NM (32, G2) |
NM | NM | 109 (29, G1; 80, G2) (29, head, 8 neck, 72, subcondylar) | 0/29 (G1) 40/40 (G2) |
4 (G1) 29 (G2) |
NM | Mini-dynamic compression plates (29) |
Intraoral (29) | 2 weeks (51, G2) |
| Worsaae and Thorn, 1994 | RA/RCT a (single centre) | 52 (G1, 24; G2, 28) | 21–70 (36, G1) 18–71 (38, G2) |
Mean 21 months (G1) Mean 30 months (G2) |
NM | 52 (24, G1; 28, G2) (all subcondylar) | 24/0 (G1) 28/0 (G2) |
0 (G1) 0 (G2) |
11 (G1) 11 (G2) |
Wire osteosynthesis | Submandibular (24) | Mean 42 days (24, G1) Mean 28 days (28, G2) |
| Widmark et al., 1996 | RA (single centre) | 32 (G1, 19; G2, 13) | 17–73 (39.5, G1) 17–76 (30, G2) |
1 year | NM | 35 (21, G1; 14, G2) (all subcondylar) | 21/0 (G1) 14/0 (G2) | 2 (G1) 1 (G2) |
Body (5, G1), coronoid (2, G1) | 2.0-mm miniplate (21) | Submandibular (21) | 2 weeks (7, G1) 3 weeks (13, G2) |
| Joos and Kleinheinz, 1998 | CCT (single centre) | 51 (G1, 25; G2, 26) | 16–69 (23.7) | 10 days, 6 weeks, 3, 6, and 12 months | Traffic accident (49.9%), fall, (24.5%), assault (12.8%), occupational (3.2%), sport (4.3%), recreational (5.3%) | 51 (25, G1; 26, G2) (all neck) | 9/16 (G1) 4/22 (G2) |
0 (G1) 0 (G2) |
52% (G1) 62% (G2) |
2.0-mm titanium miniplate (25) | Retro-mandibular (25) | 10 days (26, G2) |
| Newman, 1998 | RA (single centre) | 61 (G1, 9; G2, 52) | 12–80 (NM) | Mean 64 months (6 months to 13 years) | Fall (52%), traffic accident (33%), assault (8%), sports (5%), work (2%) | 122 (10, G1; 112, G2) (8 diacapitular, 59 neck, 55 subcondylar) | 4/6 (G1) 2/110 (G2) | 9 (G1) 52 (G2) |
30 (predominantly at parasymphysis) | 2.0-mm titanium miniplate (8), Brown–Obeid technique (2) | Submandibular and pre-auricular (8), retro-mandibular (2) | Mean 37 days (39, G2) |
| Oezmen et al., 1998 | RA (single centre) | 30 (G1, 20; G2, 10) | 16–59 (31.5) | Range 6–24 months | NM | 37 (24, G1; 13, G2) (6 head, 28 subcondylar, 3 NM) | NM | 4 (G1) 3 (G2) |
NM | Transfixation screw (24) | NM | 2 weeks (30, G1 + G2) |
| Santler et al., 1999 | CCT (single centre) | 150 (G1, 37; G2, 113) | 6–73 (24, G1) 8–71 (24, G2) |
Range 6–46 months (24, G1) Range 6–103 months (39, G2) |
Traffic accident (55%), assault (7%), fall (21%), sports (9%), work accident (5%), horse kick (3%) | 189 (43, G1; 146, G2) (27, head, 162 neck) | 132/57 (G1 + G2) | 6 (G1) 33 (G2) |
13 (G1) 59 (G2) |
2.0-mm miniplates (13), anchor screws (10), open reduction without fixation (9), microplates (4), wire osteosynthesis (4), screws (2), and removal of the small fragment (1) | NM | NM (107 patients, 132 fractures, G2) |
| Throckmorton and Ellis, 2000 | CCT (single centre) | 136 (G1, 62; G2, 74) | NM | 6 weeks, 6 months, 1, 2, and 3 years | Most assault b | 136 (62, G1; 74, G2) (101 subcondylar, 35 neck) | NM | 0 (G1) 0 (G2) |
39 (G1) 44 (G2) |
Mini-dynamic compression plates (most fractures), 2 lag screws |
Retro-mandibular (62) | NP (G1 + G2) (in G2, training elastics without MMF was applied) |
| De Riu et al., 2001 | RA (single centre) | 39 (G1, 20; G2, 19) | 13–69 (NM, G1) 13–25 (NM, G2) |
Range 5–6 years (G1) Range 8–12 years (G2) |
Most traffic accident and assault b | 49 (27, G1; 22, G2) (19 neck, 25 subcondylar, 5 NM) | 10/12 (G1) 14/8 (G2) |
7 (G1) 3 (G2) |
NM | 2.0-mm titanium miniplate (27) | Submandibular or pre-auricular (27) | 3–5 days (20, G1) 5–7 days (19, G2) |
| Haug and Assael, 2001 | RA (single centre) | 20 (G1, 10; G2, 10) | NM (37, G1; 36, G2) | 2.3 years (G1) 4.4 years (G2) |
Motor vehicle (45%), motorcycle (10%), assault (20%), fall (20%) | 20 (10, G1; 10, G2) (all subcondylar) | NM | 0 (G1) 1 (G2) |
NM | 2.0-mm miniplate b (10) | Submandibular (10) b | 2–6 weeks (10, G2) |
| Hyde et al., 2002 | CCT (multi-centre) | 54 (G1, 33; G2, 21) | >16 | Mean 14.5 months (range 1–36) | NM | 54 (33, G1; 21, G2) (NM) | NM | 0 (G1) 0 (G2) |
Parasymphysis (30) | 2.0-mm miniplate (33) | Retro-mandibular (33) | 7–10 days (21, G2) |
| Yang et al., 2002 | RA (single centre) | 66 (G1, 36; G2, 30) | 19–70 (26, G1) 17–68 (25, G2) |
1 and 2 weeks, 1, 2, 3, 4, 6, and 12 months | NM | 66 (36, G1; 30, G2) (30 neck, 36 subcondylar) | 14/0 (G1) 4/12 (G2) (not all fractures were evaluated) |
0 (G1) 0 (G2) |
27 (G1) 23 (G2) |
2.0-mm miniplate (36) | Endoscope-assisted intraoral (26), or pre-auricular (10) | 1 week (36, G1) 3 weeks (30, G2) |
| Villarreal et al., 2004 | RA (single centre) | 84 (G1, 10; G2, 74) | 5–81 (27) | Mean 14.6 months (range 3–31, G1) Mean 7.6 months (range 1–33, G2) |
Traffic accident (63.1%), casual accident (16.7%), assault (9.5%) sports (9.5%), fall (1.2%) | 104 (12, G1; 92, G2) (37 head, 28 neck, 39 subcondylar) | 82/22 (G1 + G2) | 2 (G1) 18 (G2) |
Symphysis (35), body (7), angle (9), ramus (2), dentoalveolar (1) | 2.0-mm miniplate (5), wire osteosynthesis (1) c | Pre-auricular (11), coronal (1) | Mean 23.7 days (10, G1) Mean 25.3 days (G2) (71, G1 + G2) |
| Hlawitschka et al., 2005 | RA (single centre) | 43 (G1, 14; G2, 29) | 15–74 (30, G1) 14–77 (28, G2) |
Mean 11 (G1) and 20 (G2) months | NM | 49 (15, G1; 34, G2) (all diacapitular) | 0/15 (G1) 0/34 (G2) |
1 (G1) 5 (G2) |
NM | Titanium compression screw (10), titanium micromesh (4), absorbable screw (1) | Auricular (15) | 1 day (14, G1) 10 days (29, G2) |
| Landes and Lipphardt, 2005 | CCT (single centre) | 42 (G1, 32; G2, 10) | 9–79 (36, G1) 15–60 (30, G2) |
1, 6, and 12 months | NM | 47 (36, G1; 11, G2) (all subcondylar) | 36/0 (G1) 0/11 (G2) |
4 (G1) 1 (G2) |
14 (NM) | One (90%) or two (10%) 2.0-mm miniplates | Retro-mandibular (36) | 2 weeks (10, G2) |
| Stiesch-Scholz et al., 2005 | CCT (single centre) | 27 (NM) | 19–70 (32) | NM | NM | 37 (24, G1; 13, G2) | 2/35 (G1 + G2) | 10 (G1 + G2) | NM | NM | NM | NM |
| Landes and Lipphardt, 2006 | CCT (single centre) | 35 (G1, 24; G2, 11) | 6–79 (33, G1) 6–34 (15, G2) |
1, 6, and 12 months | NM | 50 (33, G1; 17, G2) (all head) | 13/20 (G1) 0/17 (G2) |
9 (G1) 6 (G2) |
6 (NM) | 1.2-mm microplate (33) | Pre-auricular (33) | 2 weeks (11, G2) |
| Ishihama et al., 2007 | RA (multi-centre) | 55 (G1, 18; G2, 37) | 16–74 (34) | Range 6 months to 2.5 years (NM) | Fall (40.3%), traffic accident (34.3%), bicycle (19.4%), assault (3%), sports (1.5%), unknown (1.5%) | 110 (36, G1; 74, G2) (40 head, 67 neck, 27 subcondylar) d | 27/9 (G1) 58/16 (G2) |
18 (G1) 37 (G2) |
Parasymphysis (12, G1; 25, G2), body (7, G1; 6, G2) | 2.0-mm miniplate (6), lag screw (4), wire osteosynthesis (1), open reduction without fixation (2), extracorporeal osteosynthesis by vertical ramus osteotomy (2), condylectomy and artificial caput placement (2) and condylectomy (10) | Submandibular (36) | 1–3 weeks (37, G2) |
| Carneiro et al., 2008 | RA (single centre) | 30 (G1, 11; G2, 19) | NM (32) | NM | NM | 30 (11, G1; 19, G2) (NM) | NM | 0 (G1) 0 (G2) |
NM | NM | NM | NM |
| Landes et al., 2008a | CCT (single centre) | 24 (G1, 11; G2, 14) e | 7–14 (10.4, G1) 5–14 (9.3, G2) |
1, 2, and 5 years | NM | 25 (11, G1; 14, G2) (17 subcondylar, 5 head, 3 diacapitular) | 11/0 (G1) 0/14 (G2) | 1 bilateral fracture, each one treated in a different group | NM | One (55%) or two (45%) 2.0-mm titanium or biodegradable miniplate for neck fractures, 1.2-mm microplate for high fractures | Retro-mandibular (neck fractures), pre-auricular (high fractures) | 2 weeks (14, G2) Patients younger than 12 years had guided occlusion by a removable orthodontic appliance for an average of 3 months to spare the tooth buds a trauma by set screw insertion |
| Landes et al., 2008b | CCT (single centre) | 129 (G1, 87; G2, 42) | 17–80 (39, G1) 15–73 (36, G2) |
12 months | NM | 158 (106, G1; 52, G2) (15, diacapitular, 23 neck, 120 subcondylar) | 106/0 (G1) 0/52 (G2) |
19 (G1) 10 (G2) |
42 (NM) | 2.0-mm miniplate, 1.2-mm microplate | Retroangular, pre-auricular | 2 weeks (42, G2) |
| Landes et al., 2008c | CCT (single centre) | 22 (G1, 9; G2, 13) | 16–54 (27, G1) 12–56 (31, G2) |
1, 2, 4, and 12 weeks, 6 and 12 months | NM | 26 (11, G1; 15, G2) (all diacapitular) | 0/11 (G1) 0/15 (G2) |
2 (G1) 2 (G2) |
3 (G1) 2 (G2) |
1.2-mm microplate (11) | Pre-auricular (11) | 2 weeks (13, G2) |
| Schneider et al., 2008 | RCT (multi-centre) | 66 (G1, 36; G2, 30) | >18 (32) | 6 months | NM | 79 (42, G1; 37, G2) (23 head, 14 neck, 42 condylar base) | 42/0 (G1) 37/0 (G2) (displaced) |
6 (G1) 7 (G2) |
NM | Miniplates (33), miniscrews (4), or lag screws (5) | Submandibular, peri-angular (12), retro-mandibular (5), pre-auricular (12), or transoral (13) | 10 days (30, G2) |
| Danda et al., 2010 | RCT (single centre) | 32 (G1, 16; G2, 16) | NM | Mean 22.3 months (range 9–39, G1) Mean 21.5 months (range 4–42, G2) |
Traffic accident (75%), assault (18.7%), fall (6.3%) | 32 (16, G1; 16, G2) (all subcondylar or neck) | NM | 0 (G1) 0 (G2) |
NM | One or two 2.0-mm miniplates (16) | Pre-auricular, submandibular, transmasseteric anterior parotid, or retro-mandibular | 2 weeks (16, G1) 4 weeks (16, G2) |
| Rasheed et al., 2010 | RCT (single centre) | 60 (G1, 30; G2, 30) | NM | 1, 3, 6, and 12 months | NM | 60 (30, G1; 30, G2) (32 subcondylar, 28 neck) | NM | 0 (G1) 0 (G2) |
NM | 2.0-mm titanium miniplate (30) | Pre-auricular (30) | 4–6 weeks (30, G2) |
| Singh et al., 2010 | RCT (single centre) | 40 (G1, 18; G2, 22) | NM (30.6) | 6 weeks, 3 and 6 months | Traffic accident (60%), assault (30%), others (10%) | 40 (18, G1; 22, G2) (all subcondylar) | NM | 0 (G1) 0 (G2) |
NM | 2.0-mm titanium miniplate (18) | Retro-mandibular (18) | Mean 20 days (22, G2) |
| Sforza et al., 2011 | CCT (single centre) | 21 (G1, 9; G2, 12) | 18–50 (27) | Mean 26 months (range 6–66) | NM | 21 (9, G1; 12, G2) (9 neck, 9 diacapitular, 3 NM) | 5/4 (G1) 0/12 (G2) |
0 (G1) 0 (G2) |
0 | NM | NM | NM |
| Gupta et al., 2012 | CCT (single centre) | 28 (G1, 10; G2, 18) | 2–65 (28.4) | 1, 4, 8, and 12 weeks | Traffic accident (53.6%), assault (10.7%), fall (28.6%), horse kick (3.5%), hit by a block of wood (3.5%) | 34 (11, G1; 23, G2) (NM) | NM | 1 (G1) 5 (G2) |
Symphysis (2, G1; 3, G2), parasymphysis (5, G1; 12, G2) | 2.0-mm stainless steel miniplate | Pre-auricular and/or retro-mandibular | 4 weeks (14, G2) |
| Handschel et al., 2012 | RA (single centre) | 105 (NM) | NM | >1 year | NM | 111 (83, G1; 28, G2) (12 diacapitular, 66 neck, 33 NM) | 66/45 (G1 + G2) | 6 (G1 + G2) | NM | One 2.0-mm titanium miniplate (41), two 2.0-mm titanium miniplates (10), locking 2.0-mm miniplate (23), Würzburger tension screw (2), others (7) | Intraoral (14), pre-auricular (3), submandibular (38), or retro-mandibular (28) | 10 days or 3 weeks (NM, G2) |
| Kokemueller et al., 2012 | CCT (multi-centre) | 75 (G1, 31; G2, 44) | NM (38, G1) NM (32, G2) |
8–12 weeks, 1 year | NM | 75 (31, G1; 44, G2) (all neck) | NM | 0 (G1) 0 (G2) |
NM | NM | Endoscope-assisted transoral (31) | Mean 11 days (range 7–14; 31, G1) Mean 33 days (range 14–56; 44, G2) |
| Singh et al., 2012 | RA (single centre) | 44 (G1, 24; G2, 20) | 19–55 (28.2) | 1, 2, 3, and 4 weeks, 2, 3, and 6 months | Traffic accident (65%), assault (20%), fall (12%), sports (3%) | 88 (48, G1; 40, G2) (all subcondylar) | NM | 24 (G1) 20 (G2) |
Symphysis/parasymphysis (25), body (10) angle (9) | 2.0-mm titanium miniplate (48) | Retro-mandibular (48) | 3–5 days (24, G1) 21–35 days (20, G2) |
| Kotrashetti et al., 2013 | RCT (single centre) | 22 (G1, 10; G2, 12) | NM | 3 and 6 months | Traffic accident (100%) | 22 (10, G1; 12, G2) (all subcondylar) | 10/0 (G1) 12/0 (G2) |
0 (G1) 0 (G2) |
16 (7, G1; 9, G2) Parasymphysis (40%) | 2.0-mm titanium miniplate (10) | Retro-mandibular (10) | 2–3 days, elastics; 3–4 weeks, wires (12, G2) |
| Leiser et al., 2013 | RA (single centre) | 37 (G1, 10; G2, 27) | 19–42 (30, G1) 12–66 (27, G2) |
Mean 49.2 (G1) and 28.2 (G2) months | Traffic accident (30%), assault (24%), fall (46%) | 38 (11, G1; 27, G2) (all subcondylar) | 11/0 (G1) 27/0 (G2) |
1 (G1) 0 (G2) |
0 | One or two 2.0-mm titanium miniplate (11) | Anteroparotid transmasseteric (11) | 14 days (27, G2) |
| Reddy et al., 2013 | RA (single centre) | 124 (NM) | NM | NM | Traffic accident (64%), assault (17%), fall (8.6%), sports (9.6%) | 175 (110, G1; 65, G2) (88 subcondylar, 32 head, 55 neck) | 39/136 (G1 + G2) | 51 (G1 + G2) | Symphysis (45), parasymphysis (33), body (3) angle (3) | NM (110) | Retro-mandibular (65), pre-auricular (30), submandibular (15) | NM (NM, G2) |
a The study had two different design approaches. From 1980 to 1983, all patients were treated non-surgically, while from 1983 to 1989 patients were treated in a randomized manner depending on the day of admission: surgically on even days and non-surgically on uneven days.
b Unpublished information was obtained by personal communication with one of the authors.
c Of the 12 MCFs treated surgically, six were fixed, the fragment was replaced as a free graft in five, and in one case fixation was not possible.
d There were 137 MCFs initially, but only 110 were followed up.
e Here the sum of the number of patients for both groups ( n = 25) is greater than the number of patients in the study ( n = 24) because one patient had a bilateral fracture: one fracture was treated surgically and the other was treated non-surgically.
| Authors and year published | Inclusion criteria | Post-treatment radiological assessment | Hardware failure (fracture, loosening) (G1) | Wound dehiscence (G1) | Facial nerve injury (G1) | Hardware removal (G1) |
|---|---|---|---|---|---|---|
| Takenoshita et al., 1990 | NM | NM | NM | NM | NM | NM |
| Hidding et al., 1992 | Dislocated neck fractures | In G1, 19 out of 20 joints (95%) were anatomically reconstructed, with only one displacement. In G2 there was only one joint in correct alignment whereas 13 joints were in malposition (93%) | NM | NM | NM | NM |
| Moritz et al., 1994 | NM | The G1 patients had improvement of the condyle position after surgery, which did not occur in G2 | NM | NM | 0 | NM |
| Worsaae and Thorn, 1994 | Unilateral dislocated low subcondylar fractures | No correlation between the degree of radiographically recorded dislocation or angulation of the condylar fragment and the number of complications was found | NM | NM | 0 | NM |
| Widmark et al., 1996 | Subcondylar fractures | Good anatomic positioning in G1 in all but 2 patients, who had a medially dislocated condyle with increased condyle–fossa distance | NM | 0 | 1 | NM |
| Joos and Kleinheinz, 1998 | Low condylar neck fractures with displacement or dislocation | The G2 had on average 4 mm more correction of the ramus height than the surgically treated G1 patients | NM | NM | NM | NM |
| Newman, 1998 | Bilateral condylar fractures | NM | NM | NM | 0 | NM |
| Oezmen et al., 1998 | NM | In 8 patients (80%) from G2, the condylar process healed in a dislocated position with severe disfigurement of the condylar head. None of the patients in G1 showed any axis misalignment or deformation of the condylar head | NM | NM | 0 | NM |
| Santler et al., 1999 | NM | NM | NM | NM | NM | NM |
| Throckmorton and Ellis, 2000 | Unilateral condylar fractures, absence of any history of TMJ dysfunction, sufficient dentition to allow MMF and reproduce occlusal relationships, absence of severe pre-traumatic dysgnathia | At 6 weeks, surgery had uprighted the condylar processes in G1 so that there was a significant change in the coronal displacement and ramus length (but not for the sagittal position) between pretreatment and 6 weeks post-treatmently. G2 showed no significant change in any displacement variables | NM | NM | NM | NM |
| De Riu et al., 2001 | NM | G1 patients showed TMJ morphology similar to that on the contralateral unaffected side. Neither differences in the height of the rami nor alterations of the glenoid fossa were recorded, compared to the frequent ramus asymmetry observed in G2 patients | NM | NM | 0 a | NM |
| Haug and Assael, 2001 | Subcondylar fractures, non-comminuted, absence of head fractures | The G1 patients had improvement of the condyle position after surgery, which did not occur in G2 a | 0 a | 0 a | 0 | 0 a |
| Hyde et al., 2002 | Unilateral condylar fractures with deranged occlusion | NM | NM | NM | 3 | NM |
| Yang et al., 2002 | Unilateral condylar fractures, absence of midface fractures | NM | NM | NM | 3 | NM |
| Villarreal et al., 2004 | NM | The post-treatment coronal displacement of the condyle was between −60° (medial) and 20° (lateral) with a mean of −6.07° ± 15.06°, with no statistically significant relationship with the method of treatment. There were statistically significant differences between the preoperative and post-treatment coronal and sagittal displacement | NM | NM | NM | 0 a |
| Hlawitschka et al., 2005 | Diacapitular fractures | In G1 significant improvement in the position of the major fragments was achieved. In G2 considerable misalignment, distinctive changes in condylar form and resorption of the fractured condyle were frequently seen | 0 | 0 | 1 | 0 |
| Landes and Lipphardt, 2005 | Unilateral or bilateral condyle fractures located at the sigmoid notch or subcondylar, absence of any history of TMJ dysfunction, sufficient dentition to reproduce occlusal relationships, absence of severe pre-traumatic dysgnathia | Angular repositioning of the condyle in relation to the ascending ramus was better than vertical reduction of the condylar height | 1 | NM | 2 | 1 |
| Stiesch-Scholz et al., 2005 | NM | NM | NM | NM | 0 | NM |
| Landes and Lipphardt, 2006 | Unilateral or bilateral condylar head and high condylar fracture, absence of any history of TMJ dysfunction, sufficient dentition to reproduce occlusal relationships, absence of severe pre-traumatic dysgnathia | NM | 0 | NM | 3 | 0 |
| Ishihama et al., 2007 | Bilateral condylar fractures, absence of maxillary fractures | The G1 patients had improvement of the condyle position after surgery, which did not occur in G2 a | NM | NM | NM | 36 a,b |
| Carneiro et al., 2008 | Unilateral condylar fractures | On the fractured side in only 26.7% was the morphology of the mandibular condyle not preserved, while on the non-fractured side most patients (73.3%) displayed a preserved morphology | NM | NM | NM | NM |
| Landes et al., 2008a | Patients under 14 years of age, absence of any history of TMJ dysfunction, sufficient dentition to reproduce occlusal relationships, absence of severe pre-traumatic dysgnathia | NM | 2 (fracture) | NM | 2 | 2 |
| Landes et al., 2008b | Neck and head fractures, absence of any history of TMJ disorder, sufficient dentition or prosthetic rehabilitation to check the occlusion, absence of prevalent severe dysgnathia | Shortening of the ascending ramus and fracture angulation had not substantially improved when compared with the preoperative values for G2. For G1, both parameters improved substantially | 5 (fracture) | NM | 7 | 5 |
| Landes et al., 2008c | High non-displaced, non-dislocated diacapitular fractures, absence of previous history of TMJ dysfunction, sufficient dentition to reproduce an occlusion, absence of severe pre-traumatic dysgnathia | Angular rectification in G1 was successful; in G2 a slight enlargement of the angulation was encountered at follow-up | 1 (broken) | NM | 0 | 1 |
| Schneider et al., 2008 | No history of TMJ pathologies, no pre-existing skeletal discrepancies with malocclusion | At 6 months follow-up in G2 shortening of the ascending ramus and fracture angulation had not improved substantially when compared with the preoperative values. In G1, both parameters improved substantially | NM | NM | 0 | NM |
| Danda et al., 2010 | Unilateral displaced subcondylar or neck condylar fractures, absence of condylar head fractures, sufficient dentition to reproduce normal occlusion, absence of any associated midface fractures, absence of any history of TMJ dysfunction, absence of a history of occlusal disturbances or skeletal malocclusion | Four patients (25%) in G2 and 14 patients (87.5%) in G1 had an anatomic reduction of the condyle radiographically; the difference was statistically significant | NM | NM | 2 | NM |
| Rasheed et al., 2010 | Unilateral condylar fractures, dentition complete enough to apply stable Erich arch bars, absence of other facial fractures (except in the mandible) | NM | NM | NM | NM | NM |
| Singh et al., 2010 | Unilateral subcondylar fractures, sufficient dentition to reproduce the occlusal relationships. Degree of displacement of the condylar fragment in the coronal plane: 10° to 35° and/or shortening of the height of the ascending ramus of the mandible >2 mm, absence of condylar head or neck fractures. No previous history of TMJ dysfunction | For G2, at 6 months follow-up, shortening of the ascending ramus and fracture angulation had not improved substantially when compared with the preoperative values. For G1, both parameters improved substantially | NM | NM | 0 | NM |
| Sforza et al., 2011 | Unilateral condylar fractures | NM | NM | NM | NM | NM |
| Gupta et al., 2012 | NM | NM | NM | NM | 2 | NM |
| Handschel et al., 2012 | NM | NM | 8 (loosening) | NM | 0 | 37 |
| Kokemueller et al., 2012 | Unilateral condylar neck fractures, displacement of the condyle with an inclination >30° and/or severe functional impairment such as malocclusion or an open bite, with or without dislocation of the condylar fragment, non-high or non-intracapsular condylar neck fractures | NM | 1 (fracture) | NM | 1 | NM |
| Singh et al., 2012 | Bilateral condylar fractures, all patients partially or totally dentulous | NM | 0 | 0 | 2 | 0 |
| Kotrashetti et al., 2013 | Displaced subcondylar fractures, absence of pan-facial trauma, patients 20–40 years old | NM | 0 | NM | 1 | 0 |
| Leiser et al., 2013 | Low dislocated a subcondylar fractures with no need for further reduction of associated facial bone fractures | The G1 patients had improvement of the condyle position after surgery, which did not occur in G2 a | 0 a | 0 | 0 | 0 a |
| Reddy et al., 2013 | NM | NM | 5 (loosening) | 0 | 19 | 5 |
a Unpublished information was obtained by personal communication with one of the authors.
b Removal of plates was a standard procedure in the department where the study was conducted.
Many studies had the inclusion criterion of unilateral MCF only. Three studies evaluated only patients with bilateral MCFs. Two studies included diacapitular (intracapsular) MCFs only, three studies included only subcondylar MCFs, and one study included only condylar head and high condylar fractures. The absence of head fractures was also mentioned, as well as the absence of midface fractures, maxillary fractures, pan-facial fractures, or any other facial fracture, except for associated MFs. One study assessed only patients younger than 14 years of age, and another evaluated only patients in the age range of 20–40 years. The absence of any history of TMJ disorder/dysfunction was cited as an inclusion criterion in nine studies, sufficient dentition to allow MMF and reproduce occlusal relationships was cited in 10 studies, and the absence of severe pre-traumatic dysgnathia was mentioned as an inclusion criterion in eight studies. Ten studies did not report the inclusion criteria. The classification of Spiessl and Schroll for MCFs was used in 11 articles.
In total 1982 patients were enrolled in the 36 studies, with 1094 MCFs in the surgical treatment group and 1307 MCFs in the non-surgical treatment group. The presence of associated MFs was reported in 19 studies, but only seven studies reported the precise location of the fractures. The most prevalent associated MF was fracture of the symphysis/parasymphysis region, with 78.3% (227/290) of the reported associated MFs of known location. Fourteen studies reported the aetiology of the MCFs; traffic accident (including motor vehicle and bicycle accidents) was the most prevalent aetiology in 11 studies, fall was the most prevalent aetiology in two studies, and assault/altercation was the most prevalent in one study.
The maximum follow-up period varied between 3 months and 13 years. Some studies had short maximum follow-up periods, like 3 months or 6 months. Four studies did not report the follow-up period.
Concerning the surgical treatment group, 22 studies performed the fixation using one 2.0-mm miniplate in one or more patients, and five studies used two 2.0-mm miniplates, whereas one study used a locking 2.0-mm miniplate, five studies used 1.2-mm microplates, six studies used wire osteosynthesis, two studies used mini-dynamic compression Plates, one study used titanium micromesh, and eight studies used screws (transfixation screws, lag screws, miniscrews, anchor screws, compression screws, absorbable screws, Würzburger tension screws). Some studies performed open reduction without fixation or condylectomy. Six studies did not report the fixation method used in the patients in the surgical treatment group. Twelve studies also performed MMF in some or all patients in the surgical treatment group. Concerning the non-surgical treatment group, 14 studies performed the MMF for up to 2 weeks, whereas in 16 studies the MMF was performed for more than 2 weeks. One study did not perform MMF in any group, three studies did not report the MMF period, and two studies did not report whether MMF was performed or not.
Six studies did not report the type of surgical approach used. Four studies exclusively used the submandibular approach, seven studies the retromandibular approach, three studies the pre-auricular approach, two studies an endoscope-assisted intraoral approach, one study the auricular approach, one study the intraoral approach, and one study the anteroparotid transmasseteric approach. Twelve studies used more than one approach. Twenty-four studies mentioned that post-treatment physiotherapy was performed by the patients.
Only one study provided information on the mean operation time. Three studies made use of prophylactic antibiotics for a period of 5–8 days, whereas one study did not provide details. No study reported on the use of chlorhexidine rinse. Information on post-treatment radiological assessment was provided in 19 articles; all stated that the shortening of the ascending ramus and the fracture angulation had improved substantially when compared with the preoperative values in the surgical treatment group. In one study, no correlation between the degree of radiographically recorded dislocation or angulation of the condylar fragment and the number of complications was found.
Quality assessment
Each trial was assessed for risk of bias; the scores are summarized in Table 3 . All 36 studies were judged to be at high risk of bias.
