Int. J. Oral Maxillofac. Surg.

2019; 48: 1227–1234

https://doi.org/10.1016/j.ijom.2019.02.003, available online at https://www.sciencedirect.com

Clinical Paper
Trauma

Surgical approaches for T. Imai1,2, Y. Fujita3, A. Motoki4,

H. Takaoka5, T. Kanesaki2,3,
Y. Ota1,6, S. Iwai1, H. Chisoku5,
M. Ohmae4, T. Sumi3, M. Nakazawa1,
condylar fractures related to N. Uzawa1
1
Department of Oral and Maxillofacial
Surgery II, Osaka University Graduate School

facial nerve injury: deep versus of Dentistry, Suita, Osaka, Japan;

2
Department of Oral and Maxillofacial
Surgery, Saiseikai Senri Hospital, Suita,

superficial dissection Osaka, Japan; 3Department of Oral and

Maxillofacial Surgery, Toyonaka Municipal
Hospital, Toyonaka, Osaka, Japan;
4
Department of Oral and Maxillofacial
Surgery, Rinku General Medical Centre,
Izumisano, Osaka, Japan; 5Department of
T. Imai, Y. Fujita, A. Motoki, H. Takaoka, T. Kanesaki, Y. Ota, S. Iwai, H. Chisoku, M. Oral and Maxillofacial Surgery, Higashiosaka
Ohmae, T. Sumi, M. Nakazawa, N. Uzawa: Surgical approaches for condylar City Medical Centre, Higashiosaka, Osaka,
fractures related to facial nerve injury: deep versus superficial dissection. Int. J. Oral Japan; 6Department of Oral and Maxillofacial
Maxillofac. Surg. 2019; 48: 1227–1234. ã 2019 International Association of Oral and Surgery, Itami City Hospital, Itami, Hyogo,
Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved. Japan

Abstract. The aim of this study was to investigate the probability of facial nerve injury
(FNI) in the treatment of condylar neck and subcondylar fractures (CN/SCFs) with
percutaneous approaches and to identify factors predicting FNI. The data of 80
patients with 87 CN/SCFs were evaluated retrospectively. The primary outcome
was FNI occurrence. The predictor variables were age, sex, aetiology, alcohol
consumption, fracture site and pattern (dislocation or not), concomitant fractures,
time interval to surgery, surgeon experience, plate type, and the dual classification
of percutaneous approaches. The approaches were classified based on whether
subcutaneous dissection traversed the marginal mandibular branch (MMB) deeply
(deep group: submandibular and retroparotid approaches) or superficially
(superficial group: transparotid, transmasseteric anteroparotid (TMAP), and high
cervical-TMAP approaches). Twenty-two patients (27.5%) suffered FNI, of whom
two in the deep group had permanent paralysis of the MMB. In the multivariate Key words: facial nerve paralysis; transmasse-
teric anteroparotid approach; condylar neck
logistic regression model, deeply traversing surgery approaches (odds ratio 12.4,
fractures; subcondylar fractures; new classifi-
P = 0.025) and the presence of a dislocated fracture (odds ratio 6.66, P = 0.012) cation.
were associated with an increased risk of FNI. These results suggest that
percutaneous approaches in the superficial group should be recommended for the Accepted for publication 14 February 2019
treatment of CN/SCFs to reduce the risk of FNI. Available online 2 March 2019

Successful open treatment of mandibular condylar neck and subcondylar fractures different routes of subcutaneous dissec-
condyle fractures depends on accurate (CN/SCFs) is still highly debated; percu- tion with similar incisions often confuse
diagnostic imaging, an appropriate surgi- taneous approaches are still performed in both surgeons and trainees4,5. Representa-
cal procedure, anatomical reduction of the most procedures2,3. tive examples include approaches via
fractured condyle, rigid osteosynthesis, With the advent of approaches purport- retromandibular and submandibular inci-
and careful postoperative management1. ed to minimize associated morbidities, sion. The former includes three main
The selection of surgical approaches for such as facial nerve injury (FNI), the methods differing by their relationship

0901-5027/0901227 + 08 ã 2019 International Association of Oral and Maxillofacial Surgeons. Published by Elsevier Ltd. All rights reserved.
1228 Imai et al.

to the parotid: transparotid6, retroparo- were mentally capable of undergoing a cases with concomitant fractures of the
tid7,8, and transmasseteric anteroparotid neuromotor examination, and who had mandibular body that were managed sur-
(TMAP)9,10 approaches, while the latter postoperative clinical follow-up data cov- gically. Surgeries were performed by spe-
includes approaches via a low subman- ering at least 3 months10,15,16. cialists certified in oral and maxillofacial
dibular incision with sub-platysmal surgery or under the instruction of a spe-
dissection (traditional submandibular cialist if not certified. All surgeons served
Study variables
approach, i.e., Risdon approach) and a their apprenticeship at Osaka University.
high submandibular incision with super- Documented variables were age, sex, The percutaneous approaches applied
platysmal dissection (high cervical- cause of trauma, consumption of alcoholic included the traditional submandibular
TMAP (HC-TMAP) approach11 or high beverages before the accident, fracture (Risdon) approach and others described
submandibular transmasseteric approach12, location (condylar neck or subcondylar)14 below.
i.e., modified Risdon approach). and pattern (non-displaced/deviated, In the transparotid approach, an incision
Few studies have compared the rate displaced, or dislocated)17, concomitant approximately parallel to the posterior
of FNI among multiple percutaneous mandibular and midfacial fractures, inter- border of the mandibular ramus was
approaches using statistical methods13. val from trauma to the operation (days), used6,19,22,23. The superficial musculoapo-
Recently published systematic reviews experience of the operator (a certified neurotic system (SMAS) and parotid cap-
have suggested that the routes are more specialist or not certified), surgical ap- sule were incised anteriorly. The parotid
closely associated with the probability of proach, and type of miniplate for fixation. was bluntly dissected towards the ramus
such injury than the incision site2,3. How- The primary outcome was the occur- with no attempt to visualize the facial
ever, the surgical pathway has not been rence of FNI, and the patients were divid- nerve branches.
evaluated as a risk factor for FNI with ed into groups according to FNI presence In the retroparotid approach, an incision
potential variables, such as patient back- or absence at 1 week after surgery. This was initiated just below the earlobe
ground characteristics or the site and pat- outcome was also checked at the postop- and continued obliquely and inferiorly
tern of the fractures. erative 1-, 3-, and 6-month follow-ups. towards the mandibular angle7,8. The sub-
This study investigated the risk of FNI The motor response of the major branches cutaneous tissue was dissected posteriorly
in patients undergoing extraoral surgeries was visually checked18, including the abil- around the surface of the parotid capsule,
for CN/SCFs and explored factors predict- ity to wrinkle the forehead (temporal followed by superoanterior retraction of
ing FNI. Focusing on the marginal man- branch), to completely close the eyes the parotid and surrounding soft tissues for
dibular branch (MMB) as a critical branch (zygomatic branches), to puff the cheeks the posterior border of the ramus.
in FNI, a simplified dual classification of (buccal branch), and to symmetrically The TMAP approach9 can be performed
percutaneous approaches according to the show a smile (marginal mandibular via only a retromandibular incision (mod-
anatomical relationship between the sub- branch). ified form)15,24 or with pretragal extension
cutaneous dissection route and the MMB Other clinical evaluations included the (original form)9,10. The original procedure
was employed. inter-incisal distance, malocclusion, and was employed at the study hospitals.
postoperative surgical complications. An Dissection proceeded anteriorly on the
inter-incisal distance of less than 35 mm SMAS, and the anterior edge of the parotid
Patients and methods was defined as restricted mouth opening19. was carefully separated with protection of
Malocclusion was categorized into three the buccal branches. Retracting the parotid
Study design and samples
grades20: recovery to pre-injury occlusion posteriorly and the posterior edge of the
This retrospective cohort study was ap- with intercuspation of teeth (grade I), mild masseter muscle anteriorly, the perioste-
proved by the internal review boards of the malocclusion requiring prosthetic recon- um was dissected to the ramus.
authors’ institutions. Data for cases of struction or orthodontic therapy (grade II), In the HC-TMAP approach, the skin
surgical CN/SCFs repair were acquired and severe malocclusion requiring re- was incised for a 5-cm length from 0.5
from the electronic charts of patients trea- operation (grade III). Postoperative surgi- to 1 cm below the lower border of the
ted between January 2010 and August cal complications included surgical site mandible11,25. Subcutaneous dissection
2018 at Osaka University Dental Hospital, infection, parotid-associated complica- was performed on the platysma, which
Saiseikai Senri Hospital, Toyonaka tions (sialocele, salivary fistula, and Frey was then transected and retracted to ex-
Municipal Hospital, Rinku General syndrome), and hypertrophic scar21. Ab- pose the masseter muscle. The masseter
Medical Centre, Higashiosaka City normal findings on postoperative radiog- muscle was cut superior to the MMB and
Medical Centre, and Itami City Hospital. raphy, such as plate breakage, were also anterior to the parotid, following which
Data that included paper-based records evaluated using panoramic radiographs or sub-periosteal access to the ramus was
were also obtained as far back as Septem- CT images. achieved.
ber 2006 at Osaka University Dental Hos- With these approaches, the condyle was
pital and Saiseikai Senri Hospital. The five reduced under digital compression on the
Surgical management
general hospitals listed are affiliated with occlusal aspect of the ipsilateral lower
Osaka University and have adopted poli- At the study institutions, the following dentition or by inferior traction using a
cies of mutual exchange of surgeons. CN/SCFs have generally been referred twisted wire around the screw that was
The selection criteria were patients over for open treatment: (1) unilateral dis- temporarily inserted into the ramus19, if
the age of 15 years with CN/SCFs14, who placed or dislocated fractures with maloc- necessary. In principle, two miniplates
had undergone surgical treatment with clusion; (2) bilateral displaced fractures were set in a triangular fashion with one
rigid internal fixation by percutaneous associated with a symphyseal fracture of below the sigmoid notch and the other
approach, who had pre- and postoperative the mandible or midface fractures. The along the posterior edge of the ramus.
panoramic radiographs or computed to- authors occasionally referred non- The plates were predominantly locking
mography (CT) images available, who displaced forms for open treatment in miniplates (AO LOCK Mandible 2.0 or
 Surgical approaches for condylar fractures 1229

Fig. 1. Dual classification of percutaneous approaches for condylar neck and subcondylar fractures. The cervicofacial division of the facial nerve
contains nerve fibres for the marginal mandibular branch (MMB). In this study, the MMB system was defined as the segment shown with a bold
line (left panel). The skin incisions are generally located inferoposterior to the MMB system. The point of focus is whether the route of
subcutaneous dissection traverses the MMB system deeply or superficially. The coloured shapes indicate the main dissection area under the
superficial musculoaponeurotic system/platysma. The middle panel shows the deeply traversing group (DepG). The surgical path begins with
dissection in the deep direction, remaining inferoposterior to the MMB, followed by traversing under the MMB. The right panel shows the
superficially traversing group (SupG). The dissection route superficially traverses the MMB system and then deeply proceeds to the ramus
superoanteriorly to the MMB through the interspace between the facial nerve branches, with penetration through (i.e., TP) or travelling anterior to
the parotid (i.e., TMAP and high cervical-TMAP). BB, buccal branch; MMB, marginal mandibular branch; RP, retroparotid; TMAP,
transmasseteric anteroparotid; TP, transparotid.

MatrixMANDIBLE 2.0, Synthes, Paoli, system’ and five percutaneous approaches parotid approaches were categorized into
PA, USA; or Lorenz Plating System were classified into two groups depending the deep group, whereas transparotid,
Mandible, Biomet Inc., Jacksonville, on whether the subcutaneous dissection TMAP, and HC-TMAP were categorized
FL, USA). Alternatively, a single heavier traversed the MMB system deeply (deep into the superficial group. In other words,
plate was placed along the ramus. A group) or superficially (superficial group) approaches via a retromandibular incision
specialized osteosynthesis system of (Fig. 1). were divided into two groups (retroparotid
three-dimensional subcondylar miniplates Based on this classification, the tradi- approach into the deep group, and trans-
(MatrixMANDIBLE Subcondylar Plate, tional submandibular (Risdon) and retro- parotid and TMAP into the superficial
Synthes, Paoli, PA, USA) has been used
more recently. For some subcondylar
fractures, bioabsorbable plate systems
were also used (GrandFix; Johnson and
Johnson, Tokyo, Japan). A suction drain
was placed in contact with the lateral
aspect of the ramus and was removed
within a few days postoperative.
A liquid diet was started in principle
the day after the operation, followed by a
soft diet for approximately 1 month. Jaw
exercises with or without functional
mouth-opener activator was provided at
the outpatient clinic, and self-rehabilita-
tion was encouraged.

Dual classification of percutaneous

approaches
Fig. 2. Sectional models of percutaneous approaches for condylar neck and subcondylar
The inter-branch space of the facial nerve
fractures. The arrows indicate the dissection routes with the skin incision-to-ramus approach
through which the surgeon performs the on the right side of the coronal (left panel) and horizontal sections (right panel). The marginal
condyle dissection is important. In order to mandibular branch (MMB) system (stars) is located within the retracted flap in the deep group
treat the various percutaneous approaches approaches (Risdon and RP approaches), but not in the superficial group approaches (TP,
as dual variables, the MMB and its central TMAP, and high cervical-TMAP). C, condyle; MM, masseter muscle; P, parotid; R, ramus; RP,
side (the cervicofacial division of the retroparotid; SMAS, superficial musculoaponeurotic system; TMAP, transmasseteric antero-
facial nerve) were defined as ‘the MMB parotid; TP, transparotid.
1230 Imai et al.

group), as were those via a submandibular Table 1. Bivariate analysis of variables grouped by patients with and without facial nerve injury.
incision (Risdon into the deep group and Facial nerve injury, n (%)
HC-TMAP into the superficial group) Study variables P-value
(Fig. 2). Present Absent
Sample size
Patients, n = 80 22 (27.5) 58 (72.5)
Statistical analyses Fractures, n = 87 22 (25.3) 65 (74.7)
Categorical and continuous variables were Patient demographics
presented as the frequency (percentage) Age (years), mean  SD 40.3  18.7 48.5  20.7 0.082
and as the mean and standard deviation (or Sex 0.441
median (interquartile range)), respective- Male 12 (54.5) 38 (65.5)
ly. To assess the association with FNI, Female 10 (45.5) 20 (34.5)
Fisher’s exact test was applied for cate- Aetiology 0.100
Falling 7 (31.8) 27 (46.6)
gorical variables and the Wilcoxon rank-
Motor vehicle accident 9 (40.9) 10 (17.2)
sum test for continuous variables. Univar- Assault 1 (4.5) 12 (20.7)
iate and multivariate logistic regression Bicycle accident 2 (9.1) 4 (6.9)
models were utilized to obtain the odds Sports 2 (9.1) 2 (3.4)
ratios (OR) and 95% confidence intervals Falling from a height 1 (4.5) 3 (5.2)
(CI) of FNI occurring and to assess the Alcohol consumption 3 (13.6) 19 (32.8) 0.101
effects of study variables. Additional lo- Condylar fracture
gistic regression analyses allowing dupli- Fracture sitea 0.279
cation of cases with bilateral surgeries Condylar neck 8 (36.4) 14 (24.1)
were performed as sensitivity analyses19. Subcondylar 14 (63.6) 44 (75.9)
Statistical significance was considered as a Fracture patterna 0.015
P-value of less than 0.05. Statistical anal- Dislocated 7 (31.8) 5 (8.6)
yses were conducted using R 3.4.1 soft- Non-dislocated 15 (68.2) 53 (91.4)
ware program (R Foundation for Concomitant facial fractures
Statistical Computing, Vienna, Austria). Mandible 12 (60.0) 36 (67.9) 0.585
Midface 2 (9.0) 6 (10.3) 1.000
Operation
Results Interval to surgery (days), median (IQR) 5.0 (3.25–8.75) 6.5 (4.25–10.0) 0.236
A total of 80 patients with 87 fracture sides Experience of surgeons 1.000
were eligible for the analysis; 73 patients Certified in OMS 17 (77.3) 46 (79.3)
Not certified in OMS 5 (22.7) 12 (20.7)
(91.2%) underwent unilateral surgery and Percutaneous approacha 0.017
seven patients (8.8%) underwent bilateral Deeply traversing 21 (95.5) 40 (69.0)
surgery. The mean age of the patients was Superficially traversing 1 (4.5) 18 (31.0)
46.3 years. The cause of the trauma Plates for fixationa
was predominantly falling (34 patients, Miniplate 16 (72.7) 46 (79.3) 0.690
42.5%), followed by motor vehicle acci- 3D subcondylar plate 3 (13.6) 8 (13.8)
dents (19 patients, 23.8%). The median Combination of those above 1 (4.5) 1 (1.7)
interval between the trauma and the sur- Bioabsorbable plate 2 (9.1) 3 (5.2)
gery was 6.0 days. Regarding the fracture 3D, three-dimensional; IQR, interquartile range; OMS, oral and maxillofacial surgery. SD,
site, 25 fractures (28.7%) were located in standard deviation
a
the condylar neck and 62 (71.3%) in the Target fracture per case to avoid analyzing duplicate cases with bilateral surgeries.
subcondylar areas. The fracture pattern
included dislocation in 13 sides (15.0%), in five patients (6.2%) with six fractures (OR 9.45, 95% CI 1.18–75.80;
displacement in 37 (42.5%), and devia- (6.9%). P = 0.035 (patient level) and OR 9.33,
tion/non-displacement in 37 (42.5%). The primary outcome of FNI was ob- 95% CI 1.17–74.30; P = 0.035 (fracture
Concomitantly with the CN/SCFs, 48/80 served in 22 (27.5%) of 80 patients who level)) and fractures with dislocation (OR
patients (60.0%) had extra-condylar frac- underwent surgery for 87 CN/SCFs. This 4.95, 95% CI 1.37–17.80; P = 0.015
tures in the mandible and eight (10.0%) complication was unilaterally observed (patient level) and OR 4.59, 95% CI
had midfacial fractures. even in patients with bilateral surgeries. 1.34–15.70; P = 0.015 (fracture level))
Surgeries in 63 patients (78.8%) were A statistical comparison between patients were associated with the probability of
performed by specialists certified in oral with and without the outcome indicated FNI. In the multivariate model, back-
and maxillofacial surgery. In the dual that the dual classification (P = 0.017) and ward-selection stepwise regression identi-
classification of the approaches, 61 fractures with dislocation (P = 0.015) fied the following as significant factors:
patients (76.2%) with 66 fractures were associated with an increased risk deep group surgery (OR 12.40, 95% CI
(75.9%) underwent procedures of the deep of FNI (Table 1). No statistically signifi- 1.38–112.00; P = 0.025 (patient level) and
group and 19 patients (23.8%) with 21 cant differences were detected among the OR 14.10, 95% CI 1.53–130.00; P = 0.020
fractures (24.1%) underwent procedures other variables examined. (fracture level)) and fractures with dislo-
of the superficial group. Osteosynthesis Logistic regression analyses were per- cation (OR 6.66, 95% CI 1.52–29.10;
was mainly achieved using miniplates: formed at the patient level and at the P = 0.012 (patient level) and OR 7.17,
62 patients (77.5%) with 68 fractures fracture level (Table 2). The univariate 95% CI 1.66–30.90; P = 0.008 (fracture
(78.2%). Bioabsorbable plates were used model showed that deep group surgery level)).
 Surgical approaches for condylar fractures 1231

Table 2. Logistic regression model to determine the association between facial nerve injury and mandibular approach (i.e., deep group) also
study variables. had paralysis of the MMB at their 6-month
Univariate model Multivariate model visit. Regarding the distribution of branches
with impaired function, 20 patients showed
OR (95% CI) P-value OR (95% CI) P-value
FNI only in the MMB and the remaining
Variable (n = 80 patients) two patients showed FNI in multiple
Age 0.98 (0.95–1.00) 0.110 branches, one with hemifacial dysfunction
Sex 0.367 (Fig. 3) and the other with an injury of the
Male Ref.
MMB and buccal branch.
Female 1.58 (0.58–4.30)
Fracture site 0.277 No surgical site infection was evident
Subcondylar Ref. postoperatively. No patients suffered any
Condylar neck 1.80 (0.62–5.17) salivary complications that required addi-
Fracture pattern 0.015 0.012 tional intervention or had limited mouth
Non-dislocated Ref. Ref. opening at the 3-month follow-up.
Dislocated 4.95 (1.37–17.80) 6.66 (1.52–29.10) Three patients (3.8%) with concomitant
Experience of surgeons 0.842 mandibular body fractures underwent
Not certified in OMS Ref. prosthodontic occlusal reconstruction
Certified in OMS 0.88 (0.27–2.89) (grade II malocclusion). Plate breakage
Approach applied 0.035 0.025
Superficially traversing Ref. Ref.
occurred in one case with a displaced
Deeply traversing 9.45 (1.18–75.80) 12.40 (1.38–112.00) condylar neck fracture. No patients had
Plates for fixation 0.816 hypertrophic skin scarring.
Miniplates Ref.
3D subcondylar plate 1.08 (0.25–4.57) 0.919
Combination of those above 2.87 (0.17–48.70) 0.464 Discussion
Bioabsorbable plate 1.92 (0.29–12.50) 0.497
Variable (n = 87 fractures)a
Among the diverse postoperative compli-
Fracture site 0.363 cations associated with percutaneous
Subcondylar Ref. approaches for CN/SCFs21, FNI is pro-
Condylar neck 1.61 (0.58–4.52) bably the largest concern for both the
Fracture pattern 0.015 0.008 patient and the surgeon2,3. The few studies
Non-dislocated Ref. Ref. comparing the clinical outcomes among
Dislocated 4.59 (1.34–15.70) 7.17 (1.66–30.90) multiple percutaneous approaches have
Approach applied 0.035 0.020 not conducted multivariate analyses for
Superficially traversing Ref. Ref. factors related to FNI13. Furthermore,
Deeply traversing 9.33 (1.17–74.30) 14.10 (1.53–130.00)
studies on surgery for CN/SCFs using
Plates for fixation 0.820
Miniplates Ref. these analyses have only focused on a
3D subcondylar plate 1.22 (0.29–5.15) 0.788 single approach18,19. In the present study,
Combination of those above 3.25 (0.19–55.00) 0.414 univariate and multivariate logistic anal-
Bioabsorbable plate 1.62 (0.27–9.71) 0.594 yses revealed that the dual classification
3D, three-dimensional; CI, confidence interval; OMS, oral and maxillofacial surgery; OR, odds of deep group surgery and the presence of
ratio; Ref., reference. a dislocated fracture were significantly
a
Allowing duplicate cases. associated with the risk of FNI.
Of the deep group surgeries, the tradi-
tional submandibular approach is the pre-
The relationships between dislocated the 22 patients (27.5%) with FNI, seven ferred extraoral choice for the repair of
fracture and FNI with stratification by per- (31.8%, or 8.8% of the total study patients) mandibular body or angle fractures. How-
cutaneous approach applied are summa- had prolonged paralysis for more than 3 ever, for the treatment of CN/SCFs, the
rized in Table 3. No case of FNI was months. Of the prolonged paralysis cases, superior traction of the flap often causes an
encountered in patients who underwent two (28.6%, or 2.5% of the total study injury to the MMB because this branch,
the TMAP or HC-TMAP approach. Of patients) treated using the traditional sub- which is located at the inferior end of the

Table 3. Relationship between the approaches applied and fracture components.

Deeply traversing group Superficially traversing group
Approach
Traditional High cervical-
submandibular Retroparotid Transparotid TMAP TMAP
Fracture site CN SC CN SC CN SC CN SC CN SC
4/8 13/46 4/7 0/5 0/3 1/5 0/5 0/3 0/2 0/3
Fracture pattern
Dislocated 1/1 3/4 3/4 0/2 0/2
Non-dislocated 3/7 10/42 1/3 0/5 0/3 1/3 0/5 0/3 0/3
Facial nerve injury 17/54 (31.5) 4/12 (33.3) 1/8 (12.5) 0/8 (0) 0/5 (0)
sides/total sides, n (%)
CN, condylar neck; SC, subcondylar; TMAP, transmasseteric anteroparotid.
1232 Imai et al.

Fig. 3. Severe postoperative facial nerve weakness. A patient with a dislocated condylar neck fracture (first (left) panel) underwent open treatment
with reduction and internal fixation using the retroparotid approach (second panel). The patient displayed a unilateral severely imbalanced facial
appearance (third panel) with almost normalized features at the 3-month follow-up (last panel) and complete recovery of the nerve function at the
6-month follow-up.

flap, thus becomes directly loaded26,27. In ‘anatomical nerve-free window’ located (33.3%; 10.5%7–40%8), transparotid
the retroparotid approach, also in the superior to the MMB10,30. Although flap (12.5%; 3.2%23–30%22), TMAP (0%;
deep group, oblique manipulation of the retraction at the superior level of the MMB 0%10–7.7%24), and HC-TMAP (0%;
CN/SCFs with superoanterior traction of can stretch the buccal branch, possibly 0%25–0.6%30) approaches. However, the
the bulky flap that includes the whole leading to an asymmetric upper lip, the relatively elevated incidence in the retro-
parotid may be unavoidable for accurate abundant interconnections may allow for parotid approach (33.3%) may be due to
reduction and stable fixation. For the pa- the preservation of the nerve function in its high rate of application for condylar
tient with whole FNI on the affected side cases treated with TMAP and HC-TMAP, neck fractures with dislocation. The inci-
in the present study (Fig. 3), this may have as no FNI was noted with these approaches dence of FNI in the traditional subman-
been due to the traction force applied to in this study. A systematic review showed dibular approach (31.5%) is considered to
the facial nerve trunk, as described above. that TMAP and HC-TMAP were more reflect the intrinsic disadvantage of the
Compared with the deep group surger- suitable than the transparotid approach high chance of FNI due to the necessity
ies, the superficial group surgeries provide for condylar neck fractures2; however, of heavy superior traction of the flap. In
more perpendicular access to the condyle, no statistical comparison of the rate of contrast, the superficial group approaches
which facilitates efficient handling and FNI among the superficial group appro- showed reasonable outcomes with a
helps reduce the fracture fragment and aches was performed in the present study low incidence of FNI (1/21 sides, 4.8%),
stable fixation with plates and thereby because of the small number of cases although the sample size was small.
decreases the incidence of FNI. A cadaver treated with each approach. The investigation of additional cases
study demonstrated that the traction force Another independent factor related to in the future will help to determine the
on the soft tissues in the HC-TMAP ap- FNI was fracture with dislocation, al- reliability of this result, showing a low
proach (i.e., superficial group) was signif- though with a lower OR than for the dual incidence of FNI in the superficial group
icantly lower than that in the Risdon classification. Previous studies on the approaches.
approach (i.e., deep group)28. transparotid approach have also reported Salivary gland-associated complica-
Interconnections occur between the zy- dislocated fractures as a predictor of tions are more likely in approaches
gomatic and buccal branches in more than FNI in multivariate models18,19. This through the parotid than in those circum-
70% of subjects29. Such features are pre- may reflect the difficulty in reduction of venting the parotid. To avoid these
dominantly anterior to the masseter mus- the dislocated condyle, which requires a problems with transparotid approaches,
cle. However, the MMB is vulnerable to strong traction force over a long manipu- gentle blunt dissection and tight closure
surgical injuries due to the low rate of lation time. of the parotid fascia is crucial19. With
interconnections, which generally occur in In the present study, the overall inci- the TMAP and HC-TMAP approaches,
no more than 17% of subjects, leading to dence of FNI was high (27.5%, patient- dissection into the masseter muscle but
the occurrence of visible postoperative level; 25.3%, fracture-level). This is as- not into the parotid can reduce the risk
complications29. A comparative study on cribed to the high rate of adopting the deep of salivary complications, as shown in the
approaches for CN/SCFs showed that per- group approaches (traditional submandib- present study. Of note, complications have
manent paralysis was more frequent with ular and retroparotid approaches) among also been reported for the modified TMAP
the traditional submandibular approach the total approaches; the deep group approach16. Dissection through a narrow
(11%) than with the transparotid approach approaches were performed for 66 of 87 manipulating space via a small 2.5- to
(3.6%)13. In the present study, the tradi- fracture sides, with a high incidence of 3-cm retromandibular incision, unlike
tional submandibular approach was the FNI (21/66 sides, 31.8%), as shown in the regular incision employed by the pres-
only procedure that provoked permanent Table 3. Each of the five approaches in ent authors, might cause an unexpected
damage (two of 54 fracture sites treated this study showed an incidence of FNI injury to the parotid.
with this approach, 3.7%). within the range of those reported previ- A limitation of this study is the retro-
FNI was unlikely to occur using the ously in the literature (Table 4): traditional spective collection of data for patients
dissection routes in the superficial group, submandibular (incidence of FNI in who underwent surgery performed by dif-
particularly TMAP and HC-TMAP, be- this study, 31.5%; incidence in previous ferent surgeons at any of six institutions.
cause these routes pass through the studies, 5.3%27–48.1%26), retroparotid Although a multicentre approach in obser-
 Surgical approaches for condylar fractures 1233

vational clinical research, unlike a small

High cervical-TMAP
study at a single centre, allows for an

Stepwise dissection

High cervical, high

increased number of patients and en-

Possible/possible
hanced generalizability of the results,

submandibular

0%25–0.6%30
leading to improved efficiency31, the clin-

Developing

BB, ZMB
ical outcomes associated with surgical

Anterior
interventions can be biased by a larger

None
number of surgeons performing various
surgical approaches using diverse techni-
ques. However, the five collaborative

Retromandibular  pretragal
hospitals involved in this study are all

Anteriorly circumnavigating
affiliated with Osaka University and have
Easier accessibility from the perpendicular direction

adopted policies of mutual exchange of

Simple and short distance Stepwise dissection

surgeons who have received postgraduate

Possible/possible

training in oral and maxillofacial surgery

0%10–7.7%24
at Osaka University. Therefore, it is be-
Developing

lieved that the outcomes of this study were

extension
Superoanterior to the MMB system

based on treatment by surgeons with a

TMAP

BB, buccal branch; FN, facial nerve; MMB, marginal mandibular branch; TMAP, transmasseteric anteroparotid; ZMB, zygomatic branch.
Rare

common background and that the out-

BB
Superficially traversing group

comes were relatively reliable compared

with those of samples collected from
Outside the retracted flap

completely independent institutions. An-

other limitation is that individual varia-
Table 4. Profiles of the dual classification and major percutaneous approaches for condylar neck and subcondylar fractures.

Possible/possible

Retromandibular

tions, such as the anatomical distribution

3.2%23–30%22

of nerve branches or the extensibility

Transparotid

Penetrating

MMB, BB

and thickness of the retracted flap, are

Smaller

likely to affect the probability of FNI

Known

Low

but have yet to be evaluated. In addition,

a statistical analysis of prolonged FNI was
not performed due to the small number of
MMB (with inferiorly extended

samples.
Posteriorly circumnavigating

Knepil et al.4 and the present authors’

Popular and simple Relatively long distance of
Slightly difficult/possible

group5 have proposed a model to facilitate

Tendency from the inferoposterior direction

the recognition of the anatomical relation-

ships of the nerve branches and parotid
Retromandibular

through the dissection course that should

Inferoposterior to the MMB system

10.5%7–40%8
Retroparotid

be applied before selecting a percutaneous

dissection

approach for CN/SCFs. The dual classifi-

incision)
Known

cation employed in this study may be a

Rare
Within the retracted flap
Deeply traversing group

framework reflecting the risk of FNI and

indicating the importance of the relation-
ship between the subcutaneous route and
Difficult/possible

5.3%27–48.1%26
Out of the field
Submandibular

Submandibular

the MMB systems.

Well-known

In conclusion, when selecting a per-

cutaneous approach for CN/SCFs, pro-
Larger

MMB

cedures with superficial dissection

None

traversing the MMB system are recom-

mended because of the significantly lower
Possible main exposed FN branches in the surgical

probability of FNI than with approaches

Surgical manipulation of the fractured condyle

via deeply traversing routes. Surgeons

Dissection area in the deep direction to the

treating maxillofacial fractures should be

Application to displaced condylar neck/

aware of the different subcutaneous dis-

Awareness in maxillofacial surgeons
Volume of the retracted soft tissues

section routes to the condyle and choose

the most appropriate approach based on
the profile.
Relationship to the parotid

Prevalence of FN injuries
Salivary complications
subcondylar fractures
Procedure technique

Acknowledgements. The authors thank Dr

Location of MMB

Eiji Nakatani of the Division of Medical

Skin incision

Statistics, Translational Research Centre

for Medical Innovation, Foundation for
mandible

Biomedical Research and Innovation at

field

Kobe in Japan, for statistical consultation.

1234 Imai et al.

Funding. No funding source. 9. Wilson AW, Ethunandan M, Brennan PA. treatment of mandibular condylar process
Transmasseteric antero-parotid approach fractures. J Oral Maxillofac Surg 2000;58:
for open reduction and internal fixation of 950–8.
condylar fractures. Br J Oral Maxillofac 22. Manisali M, Amin M, Aghabeigi B, Newman
Competing interests. The authors declare
Surg 2005;43:57–60. L. Retromandibular approach to the mandib-
that they have no competing interests.
10. Narayanan V, Ramadorai A, Ravi P, Nirvi- ular condyle: a clinical and cadaveric study. Int
kalpa N. Transmasseteric anterior parotid J Oral Maxillofac Surg 2003;32: 253–6.
approach for condylar fractures: experience 23. Dalla Torre D, Burtscher D, Widmann G,
Ethical approval. Ethical approval was of 129 cases. Br J Oral Maxillofac Surg Pichler A, Rasse M, Puelacher W. Surgical
given by the IRB of Osaka University 2012;50:420–4. treatment of mandibular condyle fractures
Graduate School of Dentistry (H29-E27, 11. Trost O, Abu El-Naaj I, Trouilloud P, using the retromandibular anterior transpar-
H29-E27-1, H29-E27-2). Danino A, Malka G. High cervical trans- otid approach and a triangular-positioned
masseteric anteroparotid approach for open double miniplate osteosynthesis technique:
reduction and internal fixation of condylar a clinical and radiological evaluation of
Patient consent. Written patient consent fracture. J Oral Maxillofac Surg 124 fractures. J Craniomaxillofac Surg
was obtained to publish the clinical photo- 2008;66:201–4. 2015;43:944–9.
12. Meyer C, Zink S, Chatelain B, Wilk A. 24. Salgarelli AC, Anesi A, Bellini P, Pollastri G,
graphs.
Clinical experience with osteosynthesis of Tanza D, Barberini S, Chiarini L. How to
subcondylar fractures of the mandible using improve retromandibular transmasseteric
TCP plates. J Craniomaxillofac Surg anteroparotid approach for mandibular con-
References 2008;36:260–8. dylar fractures: our clinical experience. Int J
1. Rozeboom A, Dubois L, Bos R, Spijker R, de 13. Handschel J, Ruggeberg T, Depprich R, Oral Maxillofac Surg 2013;42:464–9.
Lange J. Open treatment of unilateral man- Schwarz F, Meyer U, Kubler NR, Naujoks 25. Trost O, Trouilloud P, Malka G. Open reduc-
dibular condyle fractures in adults: a system- C. Comparison of various approaches for the tion and internal fixation of low subcondylar
atic review. Int J Oral Maxillofac Surg treatment of fractures of the mandibular fractures of mandible through high cervical
2017;46:1257–66. condylar process. J Craniomaxillofac Surg transmasseteric anteroparotid approach. J
2. Al-Moraissi EA, Louvrier A, Colletti G, 2012;40:e397–401. Oral Maxillofac Surg 2009;67:2446–51.
Wolford LM, Biglioli F, Ragaey M, Meyer 14. Gellrich NC, Schoen R. Condyle, ascending 26. Tasanen A, Lamberg MA. Transosseous wir-
C, Ellis E. Does the surgical approach for ramus, and coronoid process fractures. In: ing in the treatment of condylar fractures of the
treating mandibular condylar fractures affect Ehrenfeld M, Manson PN, Prein J, editors. mandible. J Maxillofac Surg 1976;4:200–6.
the rate of seventh cranial nerve injuries? A Principles of internal fixation of the cranio- 27. Widmark G, Bagenholm T, Kahnberg KE,
systematic review and meta-analysis based maxillofacial skeleton. New York: Thieme; Lindahl L. Open reduction of subcondylar
on a new classification for surgical 2012. p. 158–67. fractures: a study of functional rehabilitation.
approaches. J Craniomaxillofac Surg 15. Tang W, Gao C, Long J, Lin Y, Wang H, Liu Int J Oral Maxillofac Surg 1996;25:107–11.
2018;46:398–412. L, Tian W. Application of modified retro- 28. Adnot J, Feuss A, Duparc F, Trost O. Re-
3. Rozeboom AVJ, Dubois L, Bos RRM, Spij- mandibular approach indirectly from the traction force necessary to expose the man-
ker R, de Lange J. Open treatment of con- anterior edge of the parotid gland in the dibular neck in Risdon and high cervical
dylar fractures via extraoral approaches: a surgical treatment of condylar fracture. J anteroparotid transmasseteric approaches:
review of complications. J Craniomaxillofac Oral Maxillofac Surg 2009;67:552–8. an anatomic comparative study. Surg Radiol
Surg 2018;46:1232–40. 16. Hou J, Chen L, Wang T, Jing W, Tang W, Anat 2017;39:1079–84.
4. Knepil GJ, Kanatas AN, Loukota RJ. Clas- Long J, Tian W, Liu L. A new surgical 29. Gosain AK. Surgical anatomy of the facial
sification of surgical approaches to the man- approach to treat medial or low condylar nerve. Clin Plast Surg 1995;22:241–51.
dibular condyle. Br J Oral Maxillofac Surg fractures: the minor parotid anterior ap- 30. Zrounba H, Lutz JC, Zink S, Wilk A. Epi-
2011;49:664–5. proach. Oral Surg Oral Med Oral Pathol demiology and treatment outcome of surgi-
5. Imai T, Nakazawa M, Uzawa N. Four-step Oral Radiol 2014;117:283–8. cally treated mandibular condyle fractures: a
chart of percutaneous approaches to the 17. MacLennan W. Consideration of 180 cases five years retrospective study. J Craniomax-
mandibular condyle: a proposal of a visual- of typical fractures of the mandibular con- illofac Surg 2014;42:879–84.
ized system for intuitive comprehension. J dylar process. Br J Plast Surg 1952;5:122–8. 31. Sprague S, Matta JM, Bhandari M, Dodgin
Oral Maxillofac Surg 2019;77:238–9. 18. Shi D, Patil PM, Gupta R. Facial nerve D, Clark CR, Kregor P, Bradley G, Little L.
6. Ellis E, Dean J. Rigid fixation of mandibular injuries associated with the retromandibular Multicenter collaboration in observational
condyle fractures. Oral Surg Oral Med Oral transparotid approach for reduction and fix- research: improving generalizability and ef-
Pathol 1993;76:6–15. ation of mandibular condyle fractures. J ficiency. J Bone Joint Surg Am 2009;91
7. Chossegros C, Cheynet F, Blanc JL, Bour- Craniomaxillofac Surg 2015;43:402–7. (Suppl 3):80–6.
ezak Z. Short retromandibular approach of 19. Kanno T, Sukegawa S, Tatsumi H, Karino M,
subcondylar fractures: clinical and radiolog- Nariai Y, Nakatani E, Furuki Y, Sekine J. Address:
Does a retromandibular transparotid ap- Tomoaki Imai
ic long-term evaluation. Oral Surg Oral Med
proach for the open treatment of condylar Department of Oral and Maxillofacial
Oral Pathol Oral Radiol Endod
fractures result in facial nerve injury. J Oral Surgery II
1996;82:248–52. Osaka University Graduate School of
8. Spinzia A, Patrone R, Belli E, Dell’Aversana Maxillofac Surg 2016;74:2019–32.
20. Liao HT, Wang PF, Chen CT. Experience Dentistry
Orabona G, Ungari C, Filiaci F, Agrillo A, 1–8Yamadaoka
De Riu G, Meloni SM, Liberatore G, Piom- with the transparotid approach via a mini-
Suita
bino P. Open reduction and internal fixation preauricular incision for surgical manage-
Osaka 565-0871
of extracapsular mandibular condyle frac- ment of condylar neck fractures. J Cranio-
Japan
tures: a long-term clinical and radiological maxillofac Surg 2015;43:1595–601.
Tel.: +81 6 6879 2941; Fax: +81 6 6879 2170
follow-up of 25 patients. BMC Surg 21. Ellis E, McFadden D, Simon P, Throckmor- E-mail: tomoimai@dent.osaka-u.ac.jp
2012;14:68. ton G. Surgical complications with open