How far we have come with the Management of Condylar Fractures? A Meta-Analysis of Closed Versus Open Versus Endoscopic Management

Abstract

Background

The treatment approaches for condylar fractures of the mandible include functional, closed reduction and open reduction–internal fixation. Recently endoscopic management of condylar fractures has been emphasized in the literature. We systematically review the studies comparing closed versus open versus endoscopic-assisted condyle fracture management with regard to the indications, effectiveness and complications of each modality.

Methods

A total of 11 articles were selected based on the inclusion and exclusion criteria from PubMed, Cochrane and clinical trials.gov. Differences in means and risk ratios were used as principal summary measures with p value < 0.05 as significant. For detection of any possible biases in sample sizes, the OR and its 95% CI for each study were plotted against the number of participants. Chi-square test, I² test and the Cochrane bias tool were used to assess the bias in and across studies.

Results

Except for deviation on opening there was no significant difference between open versus closed treatment of condylar fractures. Endoscopic approach and open surgical approaches differed only in terms of operating time and TMJ pain. There was no significant difference in facial nerve injury among the two groups.

Discussion

Closed reduction is particularly indicated for minimally displaced fractures; for moderate to severe displacement, open reduction is preferred. Open reduction can also be preferred over endoscopic approaches as there is no significant advantage of using latter. Limitations of the study included specific treatment according to the site of fracture not addressed, limited data regarding pediatric condylar fracture, lack of homogenous classification schemes, etc.

Introduction

Rationale: Mandible fractures continue to be one of the most common forms of facial trauma worldwide. The incidence of condyle fractures varies with different studies and regions. Overall it ranges from 16 to 42% [1–3]. The complications of condylar fracture include pain, restricted mandibular movement, muscle spasm and deviation of the mandible, malocclusion, pathological changes in the TMJ (temporomandibular joint), osteonecrosis, facial asymmetry and TMJ ankylosis. The goals of management are to restore normal occlusion along with adequate mouth opening, to minimize the risk of TMJ ankylosis and other possible pathologies, to decrease morbidity and increase quality of life of the patient. The two principal approaches recommended for the management are functional and surgical. Earlier, these fractures were commonly managed by closed reduction (CR); however, in recent times, open reduction is more preferred due to its improved results such as quality of life, functional and radiological outcomes [4–6]. There is another recently evolving modality of treatment, which is known as endoscopic-assisted open reduction and internal fixation (EAORIF). This approach minimizes the pitfalls of external surgical approach such as facial nerve injury, salivary fistula and external scarring while providing the benefits of anatomic reduction [7]. There are numerous comparative studies in the literature that has been published which proves the efficacy of one method over the other; however, the debate regarding the treatment of choice for a condylar fracture still remains. Review of literature reveals that various authors such as Al-Moraissi EA [8], Li et al. [9], Chrcanovic BR [10], Yao et al. [11], Nussbaum et al. [12], Duan et al. [13], Kyzas et al. [14], Han et al. [15], Liu et al. [16] and Berner et al. [17] have carried out meta-analyses and systematic reviews to identify the differences in the clinical outcomes of closed and open reduction of condylar fractures [8–17]. However, none of the studies have compared endoscopic management with open or closed treatment. Therefore, we conducted a systematic review and meta-analysis of the literature to analyze the clinical outcomes of condylar fracture treatment comparing the closed, open and endoscopic techniques.

Objectives

• To systematically compare the effectiveness of closed versus open versus endoscopic-assisted management of mandibular condylar fractures according to the PRISMA guidelines and PICOS model.

• To draw out a conclusion regarding the effectiveness of each treatment according to the type of condyle fracture.

• To evaluate and discuss the various complications of each treatment modality.

• To evaluate the outcome terms of occlusion, maximal interincisal opening, protrusion, laterotrusion, TMJ pain, nerve injuries and operating time.

Methods

Equator guidelines (equator-network.org) and PRISMA guidelines (prisma-statement.org) were followed in preparing the manuscript [18]. The study has been registered in Prospero (ID:–CRD42020223184).

Eligibility Criteria

The PICOS (Population, Intervention, Comparison, Outcome, Study Design) [19, 20] model has been used as a search strategy tool to assess the eligibility of the studies.

Focus question: What is the best treatment modality for the management of condylar fractures of the mandible?

Inclusion Criteria

• Randomized controlled trials evaluating the effectiveness of closed, open and/or endoscopic management of condyle fracture.

• Studies with adequate data on follow-up and functional outcome.

Exclusion Criteria

• Case reports, case series and observational studies, cadaver studies, animal studies.

• Studies with inadequate data on functional outcome.

• Studies comparing different methods or techniques of closed, open and/or endoscopic management.

Population: # (Adults) or (elderly) or (children) or (pediatric) or (geriatric) or (mandible fracture) or (mandibular fracture) or (condyle fracture) or (condylar fracture) or (diacapitular fracture) or (condyle head) or (condyle neck) or (subcondyle) or (condyle base).

Intervention: # (open reduction) or (closed reduction) or (endoscopic).

Comparison: # (endoscopic) or (closed) or (open) or (internal fixation).

Outcome: # (occlusion) or (occlusal disturbances) or (TMJ pain) or (joint pain) or (protrusion) or (protrusive) or (laterotrusive) or (maximal mouth opening) or (maximal interincisal opening) or (nerve injury) or (time).

Study Design: # (clinical trials) or (randomized trials) or (nonrandomized trials) or (controlled clinical trials) or (quasicontrolled trials).

Filters:

Language: English only.

Species: Human.

Ages: middle aged, young, aged, older.

Journal categories: dental, oral surgery, head and neck surgery, otolaryngology, maxillofacial surgery, plastic surgery.

Search dates: 1946–February 2020.

Study Selection and Data Extraction

The study selection was done in accordance with the PRISMA guidelines by two independent reviewers. Disagreements were resolved through discussion. Corresponding authors were contacted when data were incomplete or unclear. With respect to the listed question of our systematic review, data were sought for predictor variables; condylar fractures, closed reduction, open reduction, endoscopic management. Both reviewers evaluated the primary outcome, which was the treatment technique (closed, open and/or endoscopic). The secondary outcomes evaluated were functional upshots and operating time. Finally, funding sources of the selected studies were assessed. Electronic and manual data resources were consulted using databases: PubMed/MEDLINE, the Cochrane Library and clinicaltrial.gov.in for studies published until February 2020. The results were limited to studies written in English. The terms which were imported in the search strategy on various databases were condylar fractures, closed reduction, open reduction, endoscopic management.

Literature search on PubMed/MEDLINE was based on terms: condylar[All Fields] AND (“fractures, closed”[MeSH Terms] OR (“fractures”[All Fields] AND “closed”[All Fields]) OR “closed fractures”[All Fields] OR (“fractures”[All Fields] AND “closed”[All Fields]) OR “fractures, closed”[All Fields]) AND reduction[All Fields] AND (“open fracture reduction”[MeSH Terms] OR (“open”[All Fields] AND “fracture”[All Fields] AND “reduction”[All Fields]) OR “open fracture reduction”[All Fields] OR (“open”[All Fields] AND “reduction”[All Fields]) OR “open reduction”[All Fields]) AND (“endoscopy”[MeSH Terms] OR “endoscopy”[All Fields] OR “endoscopic”[All Fields]) AND (“organization and administration”[MeSH Terms] OR (“organization”[All Fields] AND “administration”[All Fields]) OR “organization and administration”[All Fields] OR “management”[All Fields] OR “disease management”[MeSH Terms] OR (“disease”[All Fields] AND “management”[All Fields]) OR “disease management”[All Fields].

The following terms were used in the search strategy on the Cochrane Library, the database for systematic review: condylar fractures, closed reduction, open reduction, endoscopic management. We found the following data: Cochrane Reviews— 3, Cochrane Protocols—0, Trials— 50, Editorials—0, Special collections—0, Clinical Answers—0 and Other Reviews—0.

Quality of the Studies

Quality assessment of the selected studies was executed by Newcastle–Ottawa scale. Scale was applied for clinical trials to judge each included study on selection of studies, comparability of trials and the ascertainment of either the exposure or outcome of interest. Stars were awarded such that the highest quality studies were awarded up to nine stars. The oxford level of evidence 2011 [21, 22] was used to assess the strength of each study. The level of evidence of our selected studies was of III and IV categories.

The Oxford 2011 Levels of Evidence

Level Category of evidence.

I SR (with homogeneity) of RCT.

Individual RCT.

II SR (with homogeneity) of cohort studies.

Individual cohort study (including low-quality RCT. For example.

 < 80% follow-up).

‘Outcome’ research; ecological studies.

III SR (with homogeneity) of case–control studies.

Individual case–control study.

IV Case series and poor-quality cohort and case–control studies.

V Expert opinion without explicit critical appraisal or based on.

physiology, bench research or first principles.

SR = systematic review, RCT = randomized controlled trials.

Statistical Analysis

Statistical software RevMan (Review Manager [Computer program], version 5.3, Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration, 2014) was used for statistical analysis. Differences in means and risk ratios were used as principal summary measures. The overall estimated effect was categorized as significant where p < 0.05. For detection of any possible biases in sample sizes, the OR and its 95% CI for each study were plotted against the number of participants. We used a Chi-square-based Q test to assess heterogeneity. The significance of the pooled OR was calculated by the Z-test. Additionally, I-square value is another test of heterogeneity. Cochrane risk assessment tool was utilized to assess the bias across studies.

Results

Study Selection

A total of 11 studies met the inclusion and exclusion criteria and were selected for qualitative and quantitative analysis (Fig. 1).

Fig. 1.

Study selection according to PRISMA guidelines

Study Characteristics

In total, 580 patients with 271 (46.72%) patients treated with open reduction, 250 (43.10%) with closed reduction, whereas 56(9.66%) patients with endoscopic-assisted surgery were encompassed (Table 1). The included studies were Worsaae et al. [23], Eckelt et al. [24], Landes et al. [25], Schneider et al. [26], Schmelzeisen et al. [27], Danda et al. [28], Singh et al. [29], Kotrashetti et al. [30], Rastogi et al. [31], Shiju et al. [32], Anehosur et al. [33]. Patients' age ranged from 20 to 40 years old, and male patients were in the majority. Three studies (Danda et al. 28; Singh et al. 29; Kotrashetti et al. 30) mentioned the etiology, involving fall, road traffic accident and assault. Three studies (Worsaae et al. 23; Landes et al., 25; Kotrashetti et al. 30) reported associated mandibular fractures like ramus and parasymphysis fracture. Classification schemes varied in each study; however, on a general note the fractures included were diacapitular fractures, condylar head, neck, base and/or subcondylar fractures. Closed reduction was performed either with rigid fixation, elastic guidance or a combination of both. Open reduction and internal fixations were performed through a variety of surgical approaches: retromandibular, periangular, submandibular, preauricular, transoral, etc. We developed a database into which we mentioned the individual study characteristics (Tables 2, 3 and 4).

Table 1.

Demographic data

Authorref no	Country	No. of cases	Surgical approach used	Type of fracture
Worsaae et al. [23]	Denmark	101	Open versus closed treatment	Unilateral dislocated subcondylar fractures
Eckelt et al. [24]	Germany	88	Open versus closed treatment	Condylar fractures
Landes et al. [25]	Germany	22	Open versus closed treatment	Diacapitular fractures
Schneider et al. [26]	Germany	66	Open versus closed treatment	Condylar fractures
Schmelzeisen et al. [27]	Germany	74	ORIF versus EAORIF	Uni-/bilateral condylar neck fractures
Danda et al. [28]	India	32	Open versus closed treatment	Unilateral subcondylar and condylar neck fractures
Singh et al. [29]	India	40	Open versus closed treatment	Mandibular subcondylar fractures
Kotrashetti et al. [30]	India	22	Open versus closed treatment	Subcondylar fractures
Rastogi et al. [31]	India	50	Open versus closed treatment	Condylar fracture
Shiju et al. [32]	India	50	Open versus closed treatment	Condylar fractures
Anehosur et al. [33]	India	32	Endoscopic versus retromandibular approach	Condylar fractures

Table 2.

Main characteristics of data from included studies

Authorref no	Aetiology	Patient age (mean/average) years	Number and type of condylar fracture	Concomitant mandibular fracture	Open reduction	closed reduction	Endoscopic	Comments	Follow-up visit
Worsaae et al., 23	NM	36	101 (52 follow-up), unilateral Dislocated mandibular subcondylar fractures	Mandibular ramus	40 (24 follow-up), submandibular approach, double 0.3-mm-stainless steel wires; rigid MMF for 6 weeks, then elastic MMF for 1 week	61 (28 follow-up), rigid MMF for 6 weeks, then elastic MMF for 1 week	–	malocclusion, mandibular asymmetry, impaired masticatory function and pain located to the affected joint or masticatory muscles—CR	2 years
Eckelt et al., 24	NM	32	79, condylar base, neck or in the condyle itself	NM	42; submandibular, periangular, retromandibular, preauricular, transoral approach; using miniscrews, miniplates and lag screws	37; short-term elastic maxillomandibular immobilization applied to arch bars for 10 days	–	No severe clinical complications in either treatment group, no facial nerve damage in surgical group	6 months
Landes et al., 25	NM	27–31 yrs	26, unilateral (22) or bilateral (4) nondisplaced nondislocated condylar head fractures or partial condylar head fractures	Other mandibular fractures (5)	9 patients; preauricular approach, osteofixation performed with microplusplates of 1.2 mm strength in H- and T-shape, 6–8 mm length screws of 1 mm diameter	13 patients, two weeks of intermaxillary-guided occlusion via 2 mandibulomaxillary rubber bands suspended over 14 *2.4 mm set screws	–	Disocclusion, insufficient translations on opening	1 year
Schneider et al., 26	NM	NM	79, condylar head, neck, base	NM	36 patients, submandibular, periangular, retromandibular, preauricular, transoral approach, using miniscrews, miniplates and lagscrews	30 patients, short-term elastic MMF for 10 days and adjuvant conservative measures	–	CR of bilateral condylar fractures, a higher level of pain and a higher level of functional impairment	6 months
Schmelzeisen et al., 27	NM	26–27 yrs	74, uni-/bilateral condylar neck fractures	NM	34, nonendoscopic open reduction and internal fixation using an extraoral (submandibular, preauricular, retromandibular) approach	–	40, transoral endoscopically assisted approach	For the ORIF group, pain, implant breakage, superficial infection, hematoma and hypoesthesia for the endoscopic patients: technical problem that led to inadequate reduction, mild swelling followed by nonunion; facial nerve damage in both Endoscopy is reliable and may offer advantages for selected cases, particularly concerning lower occurrence of facial nerve damage and better cosmesis	1 year
Danda et al., 28	Road traffic accident, assault or fall	> 18 yrs	32; base, neck	NM	16; preauricular, submandibular, transmasseteric anterior parotid or retromandibular approach, miniplates; elastic MMF for 2 weeks	16; rigid MMF for 2 weeks, then elastic MMF for 2 weeks	–	Transient facial paralysis	4–42 months
Singh et al., 22	Motor vehicle, assault or others	30.6 yrs	40	NM	18; retromandibular approach, 2.0 mm titanium miniplates; elastic MMF for 3to5 days	22; elastic MMF for 1to5 week (mean = 20 days)	–	Insignificant	6 months
Kotrashetti et al., 30	Road traffic accident	20–40 yrs	22; subcondylar fractures	Parasymphysis	10; retromandibular approach, titanium miniplates	12; elastic MMF for 2 to3 days, then rigid MMF for 3to4 weeks	–	Facial paralysis	6mths
Rastogi et al., 31	NM	> 18 yrs	50; sub-condylar or in the condylar neck region	NM	25; single titanium mini-plate 2.0 mm four-hole with gap through a retromandibular approach	25; MMF using Erich’s arch bar for 2 weeks followed by guiding elastics for 1 or 2 weeks	–	NO No significant clinical difference exists between patients undergoing closed treatment and open method	6 months
Shiju et al., 32	NM	> 18 yrs	50; unilateral displaced subcondylar and condylar neck	NM	25; retromandibular or submandibular approach followed by elastic maxillomandibular fixation (MMF) for 1 or 2 weeks	25; rigid MMF using Erich’s arch bar for 2 weeks followed by guiding elastics for 1 or 2 weeks	–	No significant clinical difference exists between patients undergoing closed treatment and open method	6 months
Anehosur et al., 33	NM	26–31 yrs	32; low condylar neck and subcondyle	NM	16; Retromandibular (RM) approach	–	16; EAORIF	transient facial nerve weakness was higher in the RM group	6 months

Table 3.

Main characteristics of included studies

Author	Occlusal disturbances	Max interincisal opening	Deviation on opening	Protrusion	Laterotrusion	TMJ pain	Operating time (hrs)
Worsaae et al [23] CR versus ORIF	4 versus 2 patients	50 mm versus 46 mm	–	Protrusion: 7 mm versus 7 mm	Laterotrusion fractured side: 9 mm versus 10 mm Laterotrusion contralateral side: 7mm versus 9 mm Mean = 8 ± 1.41 mm versus 9.5 ± 0.707	6 versus 1 patients	–
Eckelt et al [24] CR versus ORIF	6 versus 0 PATIENTS 20% versus 0	40.93 ± 6.8 versus 46.47 ± 5.3 mm	Deviation left in mm: 0.77 ± 1.68 versus 0.03 ± 0.167 Deviation right: 1.30 ± 1.75 versus 0.47 ± 1.34 Mean deviation = 1.035 ± 0.374 versus 0.25 ± 0.311	4.7 ± 2.5 versus 7.33 ± 2.0 mm	Left: 6.33 ± 2.7 versus 8.44 ± 3.02 mm Right: 7 ± 2.7 versus 9.06 ± 2.7 mm Sum of both: 13.33 ± 4.97 versus 17.50 ± 5.1 mm Mean = 6.665 ± 0.47 versus 8.75 ± 0.43 mm	13.53 ± 17.14 versus 2.89 ± 6.52 (VAS SCORE) NO. OF PATIENTS WITH PAIN: 17 versus 8 PATIENTS	–
Landes et al [25] CR versus ORIF	13% (1 patient) versus 0	41 ± 7.3 versus 39 ± 8.9 mm	Mean Deviation fractured side: 10.4 ± 2.6 versus 6.7 ± 2.9 mm Mean Deviation contralateral side: 11.2 ± 2.2 versus 10.8 ± 1.8 mm	7.5 ± 1.8 versus 6.1 ± 3.1	8 ± 2.3 versus 8.1 ± 3.1	0 versus 22% patients	–
Schneider et al [26] CR versus ORIF	–	42 versus 45 mm (unilateral mean values) Bilateral = 36 versus 48.33 mm(mean value) Mean maximum opening (unilateral & bilateral) = 39 ± 4.24 mm versus 46.665 ± 2.35 mm	–	5.7 versus 7.4 mm	15 mm versus 18 mm	25 versus 1(VAS score) (bilateral) Unilateral =  11 mm versus 3 mm Mean vas score18 ± 9.8 versus 2 ± 1.41 mm	–
Schmelzeisen et al [27] ORIF versus EAORIF	17 versus 14 patents	31 versus 13 mm	–	–		65 versus 94% 20 versus 34 patients	53.5 versus 86.5 min
Danda et al [28] CR versus ORIF	4 versus 1 patients	40.062 versus 42.125 mm	–	6.93 versus 7.37 mm(protrusion)	Fractured side: 6.5 mm versus 8 mm Non-fractured side: 7.56 versus 8.06 mm Mean = 7 ± 0.707 versus 8.03 ± 0.042	6 versus 2 Patients	–
Singh et al [29] CR versus ORIF	9% versus 5% patients 2 versus 1 patient	33.5 ± 1.89 versus 39.6 ± 2.22	1.18 ± 1.29 versus 0.38 ± 0.84 mm	4.1 ± 0.77 versus 5.9 ± 1.10 mm	9.86 ± 1.64 versus 12.55 ± 1.33 mm	12 patient versus 2 patient Vas score =  5.27 ± 5.43 versus 1.11 ± 3.30	–
Kotrashetti et al [30] CR versus ORIF	8.3% versus 0 1 versus 0 patient	58.3% versus 90% adequate mouth opening at 6 months  > 40 mm	–	–		33% versus 0 4 versus 0 patients	–
Rastogi et al [31] CR versus ORIF	9/9 versus 7/7	–	70% versus 0% patients had deviation on mouth opening	–		–	–
Shiju et al [32] CR versus ORIF	7/7 versus 9/9	–	70% versus 0%	–		–	–
Anehosur et al [33] ORIF versus EAORIF	(16 versus 0) patients	43.63 ± 7.5 versus 42.53 ± 7.4	0	–	0 patients	0 versus 0 patients over 6 months	RM- 107 ± 19.7 min EAORIF- 155 ± 18.2 min

Table 4.

Comparison of facial nerve function between ORIF versus EAORIF

Author no	ORIF (%)	EAORIF (%)
Anehosur et al 33	56.25	6.25
Schmelzeisen et al 27	6	8

The outcomes of the meta-analysis with a fixed-effect model are shown in Table 5, which shows level of heterogeneity among the studies.

Main Results of Meta-Analysis

Occlusal disturbances: Except Schneider et al [19] all studies reported occlusal disturbances after closed (CR), open (ORIF) and endoscopic treatments (EAORIF). For the closed versus open categories, the difference in occlusal disturbances was insignificant with p value > 0.05, heterogeneity > 50% and RR (relative risk) of 4.46 (Table 5).

Table 5.

Meta-analysis for CR versus ORIF

Results	Heterogeneity test				Odd ratio OR	Relative risk RR	95% CI	Overall test
	X2	df	p value	I2				Z-test	p value
Occlusal disturbances	21.01	7	0.521	78%	5.31	4.46	7.90–11.774	0.128	0.371
Max interincisal opening	1.72	6	3.810	42%	11.62	9.62	33.04–45.32	4.001	0.110
Deviation on opening	0.521	3	0.370	41%	5.10	0.24	0.28–9.69	2.812	0.031*
Protrusion	0.98	6	0.919	67%	11.86	8.35	4.12–7.45	0.920	2.110
Laterotrusion	0.49	6	1.391	67%	11.12	6.30	6.45–12.76	0.971	1.08
TMJ pain	17.09	6	0.760	45%	11.52	8.02	0.24–24.89	2.99	0.512

*p value < 0.05 is significant I² > 50%—heterogeneity

Among the EAORIF and ORIF groups both the included studies, Schmelzeisen et al. [22] and Anehosur et al. [28], reported malocclusion which was higher in the ORIF group as compared to the EAORIF group. However, the difference was statistically insignificant with a p value > 0.05, heterogeneity < 50% and RR of 9.66 (Table 6).

Table 6.

Meta-analysis of ORIF versus EAORIF

Results	Heterogeneity test			Odd ratio OR	Relative risk RR	95% CI	Overall test
	X2	df	p value				Z-test	p value
Occlusal disturbances	1.31	2	0.721	12.01	9.66	16.15–24.124	6.845	0.171
Max interincisal opening	3.871	2	0.221	2.71	0.51	13.00–43.12	1.511	0.110
Deviation on opening	0.951	1	0.117	0.796	0.661	0.44–6.22	4.21	0.800
TMJ pain	1.203	2	1.311	8.32	5.10	0.15–33.10	6.70	0.006*
Operating time (hrs)	8.272	2	0.127	5.795	2.81	53.03–155.24	7.226	0.042*
Facial nerve	4.45	2	0.720	6.201	5.001	6.01–56.02	8.310	1.200

I² < 50% in all parameters—lack of heterogeneity. p value < 0.05—significant

Maximum interincisal opening: Except Rastogi et al [26] and Shiju et al [27] all studies reported maximal interincisal opening after either of the treatment methods. Among the CR and ORIF groups the difference was statistically insignificant with p value > 50%, heterogeneity < 50% and RR of 9.62, 95% confidence interval (CI) 33.04–45.32 (Table 5).

The difference was also statistically insignificant between the EAORIF and ORIF groups; p value > 50%, heterogeneity < 50% and RR 0.51, 95% CI: 13.00–43.12 (Table 6).

Deviation on opening: Between the CR and ORIF groups, patients with closed reduction showed more deviation as compared to the ORIF groups and the results were statistically significant; p value < 0.05, heterogeneity < 50% and RR 0.24, 95% CI: 0.28–9.69 (Table 5).

Between the ORIF and EAORIF groups the difference was insignificant with p value < 50%, heterogeneity < 50%, RR 0.661; 95% CI: 0.44–6.22 (Table 6).

Protrusion: Only six studies (Worsaae et al. 23, Eckelt et al. 24; Landes et al. 25; Schneider et al. 26; Danda et al. 28; Singh et al. 29) reported laterotrusion and protrusion after ORIF and CR treatment. It was more with OR group. The heterogeneity test showed that p value = 0.919, I² = 67%. The difference was statistically insignificant (RR = 8.35, 95% CI [4.12–7.45], p value > 0.05; Table 5).

Laterotrusion: Only six studies (Worsaae et al. 23, Eckelt et al. 24; Landes et al. 25; Schneider et al. 26; Danda et al. 28; Singh et al. 29) reported laterotrusion after ORIF and CR treatment. It was more with ORIF group. The heterogeneity test showed that p value = 1.391, I² = 67%. The difference was statistically insignificant (RR = 6.30, 95% CI [6.45–12.76], p value > 0.05; Table 5).

TMJ pain: Both the studies by Schmelzeisen et al [20], and Anehosur et al [26] evaluated pain and revealed that it was more in case of EAORIF than ORIF group. The heterogeneity test was conducted using Chi-square test and showed that p value = 1.311, I² =  < 50%. The difference was statistically significant (RR = 5.10, 95% CI [0.15–33.10], p value < 0.05; (Table 6)). Seven studies (Worsaae et al [16]; Eckelt et al [17]; Landes et al [18]; Schneider et al [19]; Danda et al [21]; Singh et al [22] and Kotrashetti et al [23]) reported TMJ pain after ORIF and CR treatments. It was more in CR group. The heterogeneity test was conducted using Chi-square test and showed that p value = 0.760, I² = 45%. The difference was statistically insignificant (RR = 8.02, 95% CI [0.24–24.89], p value > 0.05; (Table 5).

Facial nerve weakness: Both the studies by Schmelzeisen et al., 27 and Anehosur et al., 33 evaluated facial nerve weakness and revealed that it was more in case of ORIF than EAORIF group. The heterogeneity test was conducted using Chi-square test and showed that p value = 0.720, I² = < 50%. The difference was statistically insignificant (RR = 5.001, 95% CI [6.01–56.02], p value > 0.05; Table 6).

Operating time: Endoscopic-assisted management is associated with an increased operating time as compared to the ORIF group. The difference is statistically significant with p value < 0.05, I² < 50% and RR 2.81, 95% CI; 53.03–155.24 (Table 6).

Assessment of bias: Cochrane risk assessment tool was utilized to assess the bias across studies. Anehosur et al. [28] and Worsaae et al. [23] are associated with lowest risk of bias, whereas Danda et al. have the highest number of bias (Fig. 10).

Fig. 10.

Bias of included studies

Forest plots were made to evaluate the relationship of various outcomes between CR and ORIF groups, Fig. 2: occlusal disturbances; Fig. 3: maximal interincisal opening; Fig. 4: deviation on opening; Fig. 5 and 6: protrusion and laterotrusion, respectively; Fig. 7: TMJ pain.

Fig. 2.

Forest plot graph representing occlusal differences between CR and ORIF

Fig. 3.

Forest plot graph for maximum interincisal opening between CR versus ORIF

Fig. 4.

Forest plot for deviation on opening between CR versus ORIF

Fig. 5.

Forest plot for protrusion between CR versus ORIF

Fig. 6.

Forest plot for laterotrusion between CR versus ORIF

Fig. 7.

Forest plot for TMJ pain between CR versus ORIF

Functional outcomes were also assessed between ORIF and EAORIF groups, Fig. 8: TMJ pain; Fig. 9: facial nerve weakness.

Fig. 8.

Forest plot for TMJ pain between ORIF versus EAORIF

Fig. 9.

Forest plot for facial nerve weakness between ORIF and EAORIF

Discussion

Zide and Kent [34] have classically described criteria for ORIF of condylar fractures. However, newer criteria advocates for closed reduction in displacement less than 10⁰ and ramus height shortening < 2 mm. ORIF is currently indicated with displacement > 45⁰ and ramus height shortening > 15 mm. The intermediate group (displacement 10–45⁰ and ramus shortening of 2–15 mm) can be managed either with open or closed treatment. Nevertheless, ORIF has been used with greater frequency over the past decade [35, 36]. The closed treatment of condylar fractures suffers from various complications. Majority of the patients suffers from a long-term functional and/or aesthetic problem [37, 38].

Worsaae et al. [23] study showed that almost one-third of the patients treated with conservative methods suffer from some or the other kind of complications. The study showed no difference in maximal mouth opening, protrusive or laterotrusive movements; however, the translation toward the nonfractured side was significantly lower in the CR group as compared to the ORIF group. The most common occlusal disturbance was lack of posterior contact in the CR group, and it was significantly more as compared to the ORIF group. In a nutshell, they favored ORIF for unilateral dislocated subcondylar fractures.

Eckelet et al [24] considered the borderline condyle fracture cases, displacement of 10⁰–45⁰ and ramus shortening of ≥ 2 mm. This study showed significant differences in the favor of ORIF in terms of maximal interincisal opening, protrusion, laterotrusion, deviation of mandible and TMJ pain. The study also favored ORIF in terms of fracture reduction and stability and recommended although both CR and ORIF can yield almost similar results in the moderately displaced fracture group; ORIF tends to produce better functional results.

Landes et al [25] compared CR versus ORIF in nondisplaced and nondislocated condylar head fractures. The overall success rates were not much different (75% CR vs. 78% ORIF). The only advantage ORIF had in these high condylar fractures was repositioning of the fractured fragment, but with remodeling. The CR group failed to reposition the fractured fragment but also encountered less remodeling.

Schneider et al [26] showed irrespective of the site of fracture (head, neck or base) ORIF had better results in both unilateral and bilateral fractures particularly when the displacement is more than 10⁰ and ramus shortening ≥ 2 mm.

Singh et al [29] also showed significant differences in functional outcome between CR and ORIF of unilateral subcondylar fractures of displacement > 10⁰ and ramus shortening of ≥ 2 mm.

Anehosur et al [33] compared EAORIF with ORIF in condylar fractures. The mean facial nerve weakness over a period of 6 months was more in the ORIF group (56.25%) compared with 6.25% in the EAORIF group. The results were statistically significant. The operating time required for EAORIF was more as compared to the ORIF group, although the EAORIF was performed through intraoral approach. This could be because of the difficulties in visibility and technical challenges. In this study occlusal derangements were also more in the ORIF group.

Schmelzeisen et al [27] on the other hand reported transient facial nerve weakness which was more in the endoscopic group. This may be attributed to excessive stretching of the nerve during osteosynthesis. Nevertheless, during follow-up the recovery was better in the endoscopic group as compared to the ORIF group. Also, with regard to overall patient satisfaction, there was no significant differences between the two groups. Recovery from extra oral scars and transient nerve weakness does not seem to alter the overall satisfaction of the patients.

Garcı´a et al [39] used endoscopes intraorally as well as extra orally to treat mandibular condyle fractures. The intraoral approach is particularly suitable for laterally displaced fractures. Medially displaced fractures may require an additional extra oral exposure. The authors prefer to use a single nonbridged miniplate as the bridged plate may interfere with plating. Scho¨n et al. [40] used endoscopes with submandibular incisions for medially displaced fractures, comminuted or condylar neck fractures. The transoral approach was used in laterally displaced fractures.

The current meta-analysis showed that there was no such significant difference between ORIF and EAORIF with regard to mouth opening, deviation, etc. TMJ pain was reported by one study, and it was more in the endoscopic group. The operating time was univocally greater in the endoscopic group in both the studies. Concerning closed versus open reduction our review showed only significant differences in terms of mandibular deviation on opening, and it was more in the closed group.

The current meta-analysis has certain limitations,

• The study selection was not based on a specific site of condylar fractures (head, neck base). Our search was also not selective regarding type of condylar fractures, whether they were unilateral or bilateral.

• Every study has classified fractures differently. This inherent lack of uniformity among classification may add to the selection bias of the study.

• Few studies did not quantify the results regarding malocclusion, deviation, protrusion, TMJ pain, etc. This made the results of meta-analysis more subjective.

• Most results of heterogeneity test were poor and influenced the validity of overall effects to some extent.

• Risk assessment done using Cochrane risk assessment tool represents presence of various high-risk bias in many studies (Fig. 10).

• None of the studies included pediatric fractures. Management of pediatric mandibular condyle fractures is even more controversial.

• Most of the studies did not mention aetiology of the fractures. The understanding of the aetiology and particular fracture pattern may aid in deciding the specific treatment plan for a patient.

• The studies are heterogeneous in terms of associated mandibular fractures. Most of the studies failed to discuss the effect of concomitant mandible fracture in the management of condyle fracture and in the selection of a particular approach (closed, open or endoscopic).

• Most of the studies for endoscopic fracture have used a single miniplate defying the biomechanical principles of condylar fracture management which require either two plates (one along the posterior border and one along the sigmoid notch—the line of tension) or a 3D plate.

The generalization that can be drawn from the study is that:

• Condylar head fractures (diacapitular) can still be managed with closed reduction. Closed reduction does not reduce the fracture or bring a medially displaced fragment into position. Nevertheless, the bone remodeling is lesser as compared to surgical reduction. Regarding functional outcomes condylar head fracture seems to be managed equally better with closed or open reduction.

• Unilateral or bilateral minimally displaced fractures (< 10⁰) can be managed comparably with either closed or open reduction.

• Bilateral fractures or displacement > 10⁰, ramus height shortening ≥ 2 mm are best managed with ORIF regardless of the site of fracture (head, neck, base).

• The difference in facial nerve damage between ORIF and EAORIF from this meta-analysis is insignificant.

• Meticulously handled cases resulting in minimal scar formation or almost negligible at the end of year do not bother the patient much.

• Endoscopy can be used to treat condylar neck or base/subcondylar fractures. Intraoral along with transbuccal approaches may be necessary for laterally displaced or minimally displaced fractures. Medially displaced or moderate to severe displacement and/or comminution may demand for extraoral approaches.

• Facial nerve injury may or may not occur with endoscopic approach. During osteosynthesis excessive stretching may lead to transient paresis which eventually recovers.

• Considerable time and a steep learning curve are required for endoscopic-assisted treatment of condylar fractures.

• Continuous miniplates are better than bridged plates owing to its difficulty in plating in endoscopic-assisted fixation techniques.

Conclusion

Closed or open reduction of condylar fracture is equally effective in diacapitular fractures and minimally displaced neck and base fractures. Moderate to severe displacement with considerable ramus height shortening warrants the use of open reduction. Endoscopic approaches are associated with considerable technical challenges, and efficacy over open reduction is still not validated, dictating its use only in selected cases. The study confirms that further clinical trials are necessary particularly with endoscopic management of condylar fractures: both intraorally and extra orally. Studies in the form of systematic reviews, clinical trials and observational studies are also required for the management of pediatric and geriatric condylar fractures.

Funding

None.

Declarations

Conflict of interest

The authors declared that they have no conflict of interest.