Anda di halaman 1dari 4

YIJOM-3348; No of Pages 4

Int. J. Oral Maxillofac. Surg. 2016; xxx: xxx–xxx, available online at

Randomized Controlled Trial


Removal versus retention of Z. McNamara1, G. Findlay1,

P. O’Rourke2, M. Batstone1
Department of Oral and Maxillofacial

asymptomatic third molars in Surgery, Royal Brisbane and Women’s

Hospital, Herston, Queensland, Australia;
Statistics Unit, QIMR Berghofer Medical
Research Institute, Brisbane, Queensland,

mandibular angle fractures: a Australia

randomized controlled trial

Z. McNamara, G. Findlay, P. O’Rourke, M. Batstone: Removal versus retention of
asymptomatic third molars in mandibular angle fractures: a randomized controlled
trial. Int. J. Oral Maxillofac. Surg. 2016; xxx: xxx–xxx. # 2016 Published by Elsevier
Ltd on behalf of International Association of Oral and Maxillofacial Surgeons.

Abstract. The treatment dilemma provided by asymptomatic third molars in

mandibular angle fractures remains controversial. This prospective randomized
controlled trial was undertaken to determine whether there is an advantage to
extraction or retention of the third molar whilst repairing a mandibular angle
fracture. Sixty-four patients were allocated randomly to the two treatment groups.
All underwent open reduction and internal fixation (ORIF) with standard
postoperative care. The primary outcome measure was uncomplicated fracture
healing. Secondary measures were surgical duration, malocclusion, wound healing,
nerve injury, and return to theatre. All patients had uncomplicated fracture healing.
The incidence of nerve injury was 16% for the retention group compared with 39%
for the removal group (P = 0.038). The average operating time for ORIF and third
molar retention cases was 58.5 min and for ORIF and third molar removal cases was
66.3 min (P = 0.26). There was no statistically significant difference between
groups for wound healing, occlusion outcomes, or return to theatre. Given the
Key words: randomized controlled trial
additional risk of nerve injury and the additional operating time required for mandibular angle fractures; internal fixation;
removal of a third molar, in the absence of an absolute indicator for removal of the third molars.
third molar, it appears justifiable to advise retaining the tooth in the line of a
mandibular angle fracture. Accepted for publication 14 January 2016

The treatment of mandibular angle frac- controversial at present is the management 85% contained a third molar.2 Historical-
tures has evolved over the years from of an asymptomatic wisdom tooth in the ly, extraction of the tooth in the fracture
closed reduction with a period of inter- line of an angle fracture. line was advocated, as this was thought to
maxillary fixation, to open reduction This issue is important as this treatment decrease the risk of infection and the need
with internal fixation (ORIF).1 One aspect dilemma is common. A study by Ellis for removal of the wisdom tooth and
of the ORIF procedure that remains showed that out of 402 angle fractures, plating at a later date.3,4 However, over

0901-5027/000001+04 # 2016 Published by Elsevier Ltd on behalf of International Association of Oral and Maxillofacial Surgeons.

Please cite this article in press as: McNamara Z, et al. Removal versus retention of asymptomatic third molars in mandibular angle
fractures: a randomized controlled trial, Int J Oral Maxillofac Surg (2016),
YIJOM-3348; No of Pages 4

2 McNamara et al.

the years this view has been challenged Ethical approval for the study was Results
with the counter-argument that extraction granted by the necessary human research
Participant flow
of the third molar risks loss of bone, and ethics committee.
making reduction and plating more diffi- Sixty-four patients were enrolled in the
cult, increases the surgical time, and study. The randomization process allocat-
increases the risk to the inferior dental Participants ed 31 patients to the retention group and
nerve. Patients were assessed on presentation and 33 patients to the removal group. All
At present there is conflicting informa- standard imaging was obtained (panoram- allotted patients underwent treatment as
tion in the literature concerning the ques- ic radiographs and postero-anterior man- per their group.
tion of extracting or retaining an dible X-rays). Demographic data
asymptomatic third molar in the line of including age, sex, smoking status, diabe-
a mandibular angle fracture, with no pro- Patient demographics (Table 3)
tes, and indigenous ethnicity were docu-
spective randomized controlled trials con- mented. Examination findings related to There was no statistically significant
ducted.2,5–12 Thus clinicians have to use post-trauma alveolar nerve function were difference between the groups regarding
their best judgement rather than evidence- also recorded. Randomization then oc- sex (P = 0.19), mean age at injury
based medicine in weighing the benefits curred by the participant drawing one of (P = 0.78), or the average number of frac-
and risks of removing a third molar in the two possible envelopes, indicating which tures treated (P = 0.56). Furthermore,
line of an angle fracture against the ben- arm of the trial they would be included in. there was no statistically significant dif-
efits and risks of retaining it.2 The aim of Participants were not blinded to their al- ference between the groups regarding
this study was to compare the outcomes of lotted treatment. smoking status (P = 0.86), diabetic co-
fractures of the mandibular angle with morbidity (P = 1.00), or indigenous eth-
random allocation to removal or retention nicity (P = 0.28).
of the third molar tooth. Surgical details
Surgery was performed under general an- Primary outcome
Methods aesthesia predominantly by a maxillofa-
cial surgeon (GF). If not the primary The primary outcome measure was un-
Trial design complicated fracture healing. Each group,
operator, the surgeon (GF) was present
A prospective randomized controlled trial as an assistant. All angle fractures were removal and retention, had one return to
was performed. There were no changes to secured with a single 2.0-mm miniplate theatre for non-union. Both return cases
the methods after trial commencement. via a combined transoral incision and had an uneventful postoperative recovery
All patients presenting to the maxillo- transbuccal trocar. The fractured segments following the second surgery. Therefore
facial unit with a mandibular angle frac- were approximated visually with the den- all 64 cases had eventual complete frac-
ture requiring ORIF (Table 1) and who tition held in occlusion by the assistant ture healing.
were 18 years of age or older were con- whilst fixation was applied. Resorbable
sidered for inclusion in the study. Patients sutures were utilized for wound closure.
who could not give informed consent and Standard clinical and radiographic postop- Secondary outcome measures (Table 4)
patients with absolute indicators for the erative follow-up was undertaken. The average surgical time for retention
removal of third molars in angle fractures cases was 58.5 min and for removal cases
(Table 2) were excluded. was 66.3 min; however this difference
Randomization was accomplished by Outcomes
between the groups was not statistically
sealed envelopes containing allocation to The primary outcome measure was un- significant (P = 0.26).
one of the two study groups. Sixty-four complicated fracture healing. The second- All patients underwent X-ray postoper-
patients gave consent and were deemed ary measures were wound issues (wound atively. Furthermore, all patients had at
eligible for the trial. breakdown/infections/collections), opera- least one postoperative review. The fol-
tive duration, malocclusion, inferior den- low-up period for the removal and reten-
tal nerve (IDN) injury, and return to tion groups was similar: follow-up ranged
Table 1. Indicators for open reduction and theatre. from 1 to 164 days (mean 27 days, median
fixation of a mandibular angle fracture. Surgeons undertook each postoperative 12 days) in the removal group and from 1
1 Unfavourable fracture patient assessment, but they were not to 164 days (mean 26 days, median 27
2 Inability to obtain adequate occlusion by blinded to which group patients had been days) in the retention group (P = 0.49,
closed techniques allocated. Patients who did not attend Mann–Whitney U-test).
3 Infection (peri-apical/pericoronitis) clinic review appointments were deemed With regard to wound issues, the re-
4 Fracture of tooth/roots
to have had nil negative outcomes. moval group (9.1%) had a higher rate than
Categorical data were summarized as the retention group (0%), but this differ-
frequencies, and comparisons across allo- ence between the groups was not statisti-
Table 2. Absolute indicators for removal of cated groups were done by cross-tabula- cally significant (P = 0.09). The incidence
third molars.12 tions and x2 tests for significance. of malocclusion was similar in the reten-
1 Caries Continuous variables were summarized tion group (12.9%) and removal group
2 Mobile tooth as the mean and range, and comparisons (12.1%) (P = 0.92). Importantly the inci-
3 Infection (peri-apical/pericoronitis) between groups were done using the inde- dence of IDN injury showed a significant
4 Fracture of tooth/roots pendent samples t-test or the Mann–Whit- difference (P = 0.038), with the retention
5 Pathology associated with third molar ney U-test. All significance tests used a group (16.1%) showing a lower incidence
6 Preventing adequate fracture reduction two-sided P-value of 0.05. than the removal group (39.4%).

Please cite this article in press as: McNamara Z, et al. Removal versus retention of asymptomatic third molars in mandibular angle
fractures: a randomized controlled trial, Int J Oral Maxillofac Surg (2016),
YIJOM-3348; No of Pages 4

Third molar removal/retention in angle fractures 3

Table 3. Demographics of patients included in the study and average number of fractures patients, patient compliance, quality of
treated. post-surgical care, and definitions of com-
Study group plications may vary.
Several weaknesses of this study should
Retention (n = 31) Removal (n = 33) P-value
be acknowledged. Firstly, neither the pa-
Sex 0.19 tient nor the operating surgeon nor the
Male 30 29 assessor was blinded to the treatment al-
Female 1 4 location. For several reasons this is impos-
Age, years 0.78
sible, but most notably the postoperative
Mean 23 23
Range 18–46 18–44 X-ray reveals a tooth socket or the pres-
Average number of fractures treated 1.71 1.79 0.56 ence of a tooth depending on allocation.
Secondly, like most studies on facial trau-
ma, patient return for follow-up was poor,
Table 4. Secondary outcome measures. which limits the validity of results, partic-
Study group ularly long-term factors such as nerve
recovery. Thirdly, some of the values
Retention (n = 31) Removal (n = 33) P-value measured may have achieved statistical
Surgical time, min significance had the study recruited more
Mean 58.5 66.3 0.26 patients. For example, wound issues
Range 14–109 19–122 showed a trend towards increased pro-
Wound issues, % 0% 9.1% 0.09 blems in the removal group. Finally the
Malocclusion, % 12.9% 12.1% 0.92
principal surgeon (GF) is very experi-
IDN injury, % 16.1% 39.4% 0.038
enced and thus it may not be possible to
IDN, inferior dental nerve. extrapolate the results achieved to other
Return to theatre occurred twice during statistical significance (P = 0.09), a trend Importantly this prospective trial
this study. Each group, removal and reten- towards greater wound complications with looked at the risk to the IDN when con-
tion, had one occurrence. In the retention third molar removal was seen in the pres- sidering retaining or removing the third
case, the patient had continuing postoper- ent study. molar in an angle fracture. A statistically
ative pain and a subsequent follow-up X- Importantly, a recent systematic review significant difference was found between
ray showed a non-union of the angle frac- and meta-analysis also showed that there the groups (P = 0.038). Perhaps under-
ture. For the removal case, the patient had was no significant statistical difference standably, patients in the removal group
an ongoing wound infection at the surgical between removing or retaining the tooth reported a higher incidence of decreased
site and a persistent malocclusion. A fol- in the line of fracture with regard to the sensation in the distribution of the IDN
low-up X-ray also showed non-union of occurrence of postoperative infection.13 (39.4% vs. 16.1% in the retained group).
the angle fracture. Both return cases had However it is important to note that this During removal of the third molar, the
an uneventful postoperative recovery fol- analysis included only three prospective displacement of the fracture fragments
lowing the second surgery. trials, none of which were randomized. during the procedure and extra time taken
A previous non-randomized prospec- to complete the procedure, with subse-
tive trial conducted by Rai and Pradhan quent swelling, increase the patient risk
in India commented on the discrepancy in when compared to the risk in the retained
Randomization provided two groups with occlusion between the groups.14 They group patients.
a similar composition in terms of sex, reported that the removal group (16.7%) Given the additional risk of nerve injury
ethnicity, age, and smoking status. It had a higher occlusion discrepancy com- and the additional operating time required
was expected that this randomization pared to the retention group (13.3%).14 for removal of a third molar, in the ab-
would also result in similar levels of pre- Although the result was not statistically sence of an absolute indicator for removal
operative paresthesia. The same surgeon significant, they hypothesized that the re- of the third molar the authors would advise
(GF) was present during all operative moval of the third molar causes a loss of the retaining of an asymptomatic wisdom
procedures thus removing one cause of contact and continuity in the fracture, and tooth in the line of a mandibular angle
bias. Following the surgery, both groups thus more chances of displacement or fracture.
had a similar follow-up regimen and ob- collapse during fixation.14 The results of
servation time period. It is believed that the present study regarding malocclusion
this process enabled a meaningful study of in the two groups, although not significant,
comparison between the two groups. differed from their study. The present None.
Ellis previously reported a 19.5% infec- study showed a minimal difference be-
tion rate when the tooth was present in the tween the groups: retention group 12.9%
Competing interests
angle fracture as compared to 19.0% when and removal group 12.1%. So perhaps the
the tooth was extracted.2 The results of the loss of contact and continuity in the frac- None.
present study differ from those of that ture is less of a problem than hypothe-
previous study. In the retention group, sized.
Ethical approval
none of the 31 patients (0%) had postop- It is acknowledged that direct compari-
erative wound issues, while in the removal son of studies from different units and Ethical approval for the study was granted
group, three of the 33 patients (9.1%) had countries is difficult, as the aetiology of by the Human Research and Ethics Com-
such issues. Although this did not reach fractures, socio-economic status of the mittee of the Prince Charles Hospital,

Please cite this article in press as: McNamara Z, et al. Removal versus retention of asymptomatic third molars in mandibular angle
fractures: a randomized controlled trial, Int J Oral Maxillofac Surg (2016),
YIJOM-3348; No of Pages 4

4 McNamara et al.

Brisbane, Queensland, Australia (HREC/ analysis of 57 cases. Oral Surg Oral Med complications. Rev Stomatol Chir Maxillo-
10/QPCH/174). Oral Pathol 1994;78:426–31. fac 2005;106:146–8.
6. Spiessl B. Closed fractures. In: Spiessl , 12. Shetty V, Freymiller E. Teeth in the line of
editor. Internal fixation of the mandible. fracture: a review. J Oral Maxillofac Surg
Patient consent Berlin: Springer; 1989. p. 199. [chapter 5]. 1989;47:1303–6.
Not required. 7. Thaller SR, Mabourakh S. Teeth located in 13. Bobrowski A, Sonego C, Chagas O. Postop-
the line of mandibular fracture. J Craniofac erative infection associated with mandibular
Surg 1994;5:16–9. angle fracture treatment in the presence of
References 8. Gerbino G, Tarello F, Fasolis M, De Gioanni teeth on the fracture line: a systematic review
PP. Rigid fixation with teeth in the line of and meta-analysis. Int J Oral Maxillofac
1. Anderson T, Alpert B. Experience with rigid
mandibular fractures. Int J Oral Maxillofac Surg 2013;42:1041–8.
fixation of mandibular fractures and imme-
Surg 1997;26:182–6. 14. Rai S, Pradhan R. Tooth in the line of
diate function. J Oral Maxillofac Surg
9. Rubin MM, Koll TJ, Sadoff RS. Morbidity fracture: its prognosis and its effects on
1992;50:555–60. discussion 560–1.
associated with incompletely erupted third healing. Indian J Dent Res 2011;22:495–6.
2. Ellis E. Outcomes of patients with teeth in
molars in the line of mandibular fractures. J
the line of mandibular angle fractures treated
Oral Maxillofac Surg 1990;48:1045–7. dis- Address:
with stable internal fixation. J Oral Maxil- Zeb McNamara
cussion 1048.
lofac Surg 2002;60:863–5. Royal Brisbane and Women’s Hospital
10. Anastassov DT, Vuvakis VM. Mandibular
3. Bradley RL. Treatment of fractured mandi- Herston
fracture complications associated with the
ble. Am Surg 1965;31:289–90. Queensland
third molar lying in the fracture line. Folia
4. Rowe NL, Killey HC. Fractures of the facial Australia
Med (Plovdiv) 2000;42:41–6.
: 181–2. 2nd ed. London: Livingstone; 1968
skeleton. Tel: +61 0736468111
5. Marker P, Eckerdal A, Smith-Sivertsen C. 11. Soriano E, Kankou V, Morand B, Sadek H,
Incompletely erupted third molars in the line Raphaël B, Bettega G. Fractures of the man-
of mandibular fractures. A retrospective dibular angle: factors predictive of infectious

Please cite this article in press as: McNamara Z, et al. Removal versus retention of asymptomatic third molars in mandibular angle
fractures: a randomized controlled trial, Int J Oral Maxillofac Surg (2016),