You are on page 1of 11

The Use of Antibiotics in Odontogenic

Infections: What Is the Best Choice?

A Systematic Review
ao Roig Martins, DDS,* Otacı́lio Luiz Chagas Jr, DDS, MS, PhD,y
Bibiana Dalsasso Velasques, DDS,z Angelo Niemczewski Bobrowski, DDS,x
Marcos Britto Correa, DDS, PhD,k and Marcos Antonio Torriani, DDS, MS, PhD{
Purpose: Odontogenic infections are a common problem in dentistry, and their treatment often requires
the use of antibiotics besides the removal of the source of infection, which frequently makes it more diffi-
cult for clinicians to make a decision regarding the choice of antibiotic. This study aimed to answer the
following questions through the Patient, Intervention, Comparison, Outcome (PICO) format: When
should antibiotics be used in dental infections (DIs)? Which are the most effective drugs? How long should
antibiotics be administered?
Materials and Methods: This was a systematic review using the PubMed, Scopus, and Cochrane data-
bases without restriction as to the period researched. The variables analyzed in each article were the num-
ber of odontogenic infections in each study, type of study, surgical intervention performed, antibiotics
administered, statistical differences between groups studied, and patients’ evolution after treatment.
Results: The search included 1,109 articles. After the full reading of 46 articles, 16 were included in the
final review and 30 were excluded. A sample of 2,197 DI cases was obtained, in which 15 different anti-
biotics were used, with a 98.2% overall cure rate.
Conclusions: The studies showed that antibiotics were prescribed only in situations of regional and/or
systemic body manifestations. In the case of DIs, once drainage has been performed and/or the cause of
infection has been removed, all antibiotics tested are equally effective with respect to clinical cure, and the
choice of antibiotics is not as successful as the local intervention treatment procedure. When the real need
for antibiotic therapy is detected, antibiotics should be used for the shortest time possible until the pa-
tient’s clinical cure is achieved.
Ó 2017 American Association of Oral and Maxillofacial Surgeons
J Oral Maxillofac Surg 75:2606.e1-2606.e11, 2017

Dental infections (DIs) are a common problem, and fascial spaces, especially in patients in an immunocom-
their treatment is sometimes debatable in dentistry. promised or weak condition.1
They can range from low-level local infections, which Removal of the source of infection and surgical
are usually easily treated without more serious conse- drainage are the most important steps in DI treatment.
quences, to severe life-threatening infections in the Nevertheless, in many cases the use of antibiotics is

Received from Federal University of Pelotas, Pelotas, Brazil. Conflict of Interest Disclosures: None of the authors have any
*Author, University Hospital. relevant financial relationship(s) with a commercial interest.
yAssociate Professor, Oral and Maxillofacial Surgery Residency Address correspondence and reprint requests to Dr Torriani:
Program, University Hospital. School of Dentistry, Federal University of Pelotas–UFPel, Rua
zResident, Oral and Maxillofacial Surgery Residency Program, Gonçalves Chaves, 457, Third Floor, RS 96015-560, Brazil; e-mail:
University Hospital.
xFormer Resident, Oral and Maxillofacial Surgery Residency Received March 14 2017
Program, University Hospital. Accepted August 9 2017
kAdjunct Professor, Federal University of Pelotas. Ó 2017 American Association of Oral and Maxillofacial Surgeons
{Titular Professor, Oral and Maxillofacial Surgery Residency 0278-2391/17/31121-7
Program, University Hospital.


required.1 It is incorrect, however, to think that every articles, the PubMed (
infection requires antibiotics. There are situations in pubmed), Scopus (
which they are not helpful or are even contraindi- form.url?zone=TopNavBar&origin=searchadvanced),
cated. The dental surgeon’s lack of knowledge on and Cochrane (
various principles of drug therapy with antibiotics index.php) databases were used.
can lead him or her to unnecessarily opt for antibiotics,
thus characterizing a patient overtreatment situation. SEARCH STRATEGY
Antibiotic overuse and improper indication by dentists After a brief reading on the topic, a search was per-
have been well documented. The US Centers for Dis- formed using the following key words: ‘‘tooth and bac-
ease Control and Prevention estimates that approxi- terial infections,’’ ‘‘periapical abcesses,’’ ‘‘periodontal
mately one third of all prescribed antibiotics are abcess,’’ ‘‘infection control, dental,’’ ‘‘pericoronitis,’’
unnecessary.2 ‘‘odontogenic infections,’’ and ‘‘anti-bacterial agents.’’
This study aimed to perform a systematic literature Search lines adapted to each database are shown
review regarding the choice of antibiotics in odonto- in Table 2.
genic infection treatment. We hypothesized that if
the source of infection were removed, no considerable STUDY SELECTION
differences between antibiotics would be found. This
For all articles found, the titles and abstracts were
article aimed to answer the following questions: 1)
read by 2 previously trained reviewers (J.R.M. and
In which DI treatment situations should antibiotics ^ .N.B.). For initially selected articles, the full texts
be used? 2) Which drugs are the most effective? 3)
were read, and they were submitted to the application
How long should antibiotics be administered?
of our exclusion (Table 3) and inclusion (Table 1)
criteria for final inclusion. After the analyses were
Materials and Methods compared, disagreements between reviewers were
For the purpose of obtaining answers to the afore- settled through further discussion with senior evalua-
mentioned questions, we designed and implemented tors (M.A.T. and O.L.C.).
a systematic review based on The Cochrane Collabora-
tion’s recommendations for systematic reviews. The DATA COLLECTION
study population included all English-language publi- Once the complete reading of the articles was per-
cations that addressed systemic antibiotic use in DI sit- formed, the following information of interest was
uations, without restriction as to the period
To be included in the study sample, publications had Table 2. LINE OF SEARCH USED FOR EACH DATABASE
to follow choice criteria as listed in Table 1. Descriptive
Database Line of Search
literature reviews, clinical reports, series of clinical re-
ports, and expert opinions were excluded. Cases of al- PubMed ((((((((tooth[MeSH Terms]) AND bacterial
veolitis, periodontitis, and infected odontogenic cysts infections[MeSH Terms])) OR periapical
were not thought to be odontogenic infection cases. abscesses[MeSH Terms]) OR periodontal
The variables analyzed in each article included in this abscess[MeSH Terms]) OR infection
systematic review were the number of odontogenic in- control, dental[MeSH Terms]) OR
fections in each study, type of study, surgical interven- pericoronitis[MeSH Terms]) OR
tion performed, antibiotics administered, statistical odontogenic infections[Title/Abstract])
differences between the groups studied, and AND anti-bacterial agents[MeSH Terms]
patients’ evolution after treatment. To search for Scopus (TITLE-ABS-KEY(tooth and ‘‘bacterial
infection’’) OR TITLE-ABS-KEY(periapical
abscess) OR TITLE-ABS-KEY(pericoronitis)
OR TITLE-ABS-KEY(periodontal abscess)
OR TITLE-ABS-KEY(odontogenic
Specification of antibiotics used infection)) AND TITLE-ABS-KEY(anti-
Indication of No. of patients treated bacterial agents)
Mention of whether incision and drainage and/or Cochrane (tooth AND bacterial infections) OR
removal of cause were performed periapical abscesses OR periodontal
Description of patients’ evolution according to abscess OR infection control, dental OR
intervention performed pericoronitis OR odontogenic infections
Report of clinical follow-up of patients treated AND anti-bacterial agents

Martins et al. Antibiotics and Odontogenic Infections. J Oral Max- Martins et al. Antibiotics and Odontogenic Infections. J Oral Max-
illofac Surg 2017. illofac Surg 2017.


a high risk. The articles and their respective assess-
ments are shown in Table 6.
Descriptive reviews in literature, reports of clinical cases, After common data grouping, a sample of 2,197 DIs
series of clinical cases, and expert opinions from 16 studies, in which the number of patients
Studies involving cases of alveolitis, periodontitis, and ranged from 21 to 759 per article, was obtained. All
infected odontogenic cysts
cases involved local intervention, whether drainage,
Martins et al. Antibiotics and Odontogenic Infections. J Oral Max- removal of cause, or both. Of the patients, 55 were
illofac Surg 2017.
treated with local interventions whereas 2,142 were
treated with 1 or more of the 15 antibiotics mentioned
obtained: number of DI cases, type of study, local inter- in the studies. The overall cure rate with the antibi-
vention performed, antibiotics used, and patient out- otics of choice was obtained from the included
comes. All data were collected and tabulated using a studies’ reporting on how many patients were cured
Microsoft Excel spreadsheet (Microsoft Office Profes- and how many showed treatment failure, with 98.2%
sional Plus 2010; Microsoft, Redmond, WA). of patients cured. Treatment failure occurred in 39
cases, 66.7% of which had been treated with
drugs in the macrolide group (roxithromycin in 5,
QUALITY EVALUATION erythromycin in 8, and azithromycin in 13), 7.9%
Included articles were submitted to methodologic with b-lactam antibiotics (ampicillin in 1 and
quality assessment, which included PRISMA amoxicillin-clavulanate in 6), and 15.3% with novobi-
(Preferred Reporting Items for Systematic Reviews ocin (in 6), and the therapeutic approach had to be
and Meta-Analyses),3 CONSORT (Consolidated Stan- revised. However, the latter situation was reported in
dards of Reporting Trials),4 QUOROM (Quality of Re- only 2 studies. Treatment with ampicillin failed in 1 pa-
porting of Meta-Analyses),5 MOOSE (Meta-analysis of tient, who was afterward prescribed clindamycin, and
Observational Studies in Epidemiology),6 and STROBE 6 patients who did not respond to novobiocin were
(Strengthening the Reporting of Observational Studies administered penicillin. All of the patients showed suc-
in Epidemiology)7 statement criteria, to check the cessful results with the new therapy.9,10
strength of the scientific evidence available in the cur- Drugs in the b-lactam antibiotic group were the
rent literature for clinical decision making. The classi- most used in the assessed studies, especially the peni-
fication of potential risk of bias of each study followed cillin subgroup. Tables 7 and 8 show the number of ar-
the criteria used by Bobrowski et al,8 namely, popula- ticles in which each antibiotic was used, the number
tion (sample) random selection, inclusion and exclu- of patients treated with each antibiotic, and the num-
sion criteria definitions, follow-up loss reports, ber of patients with treatment failure.
validated measurements, and statistical analysis. Six studies reported adverse antibiotic effects in their
Studies that included all the aforementioned criteria findings, including nausea, vomiting, dizziness, itchy
were classified as having a low risk of bias, those that skin, and gastrointestinal disorders, the latter being
did not include 1 criterion were classified as having a the most frequent. The criteria used to evaluate patient
moderate risk of bias, and those that did not include outcomes were based mainly on clinical features (pain,
2 or more criteria were classified as having a high swelling, fever, and lymphadenopathy), although some
risk of bias. studies used additional tests for assessment, such as
blood counts, leukocyte counts, and urine tests.
The study, updated up to December 6, 2015, yielded
1,109 results. After article titles and abstracts were The study aimed to perform a systematic literature
read and duplicates were excluded, 46 potentially rele- review on the use of antibiotics in DI treatment. We
vant articles were obtained, which were fully ac- believed that if the source of infection were removed,
cessed, analyzed, and submitted to the application of there would be no significant differences in antibiotic
our inclusion and exclusion criteria. Thirty articles choice. The aim of the study was to determine when to
were excluded after full reading. These are listed in use antibiotics in DIs, which drug is the most effective,
Table 4 along with the reasons for exclusion. Sixteen and how long the drug should be administered.
studies were included in the final systematic review Despite controversies, DIs have known treatment
sample (Table 5). The selection and assessment pro- pathways, and the results of this systematic review
cess flowchart is shown in Figure 1. allow setting some principles of treatment for this
As for the quality assessment determining the risk of type of clinical picture.
bias in each included study, 6 articles showed a low To discuss the issues involving antibiotic therapy in
risk of bias, 4 showed a moderate risk, and 6 showed DIs, it was assumed that for any local interventions


Authors Year Type of Study Reason for Exclusion*

Cope et al11 2014 Systematic review 1

Chi et al12 2014 Retrospective 2
Farmahan et al13 2014 Retrospective 2, 3
Loyola-Rodrigues et al14 2014 Laboratorial 4
Kara et al15 2014 Retrospective 2, 3, 4
Fedorowicz et al16 2013 Systematic review 5
Rasteniene et al17 2015 Retrospective 3
Gr€onholm et al18 2013 Retrospective 2
Lee and Kanagalingam19 2011 Retrospective 2
Flynn20 2011 Systematic review 1
Saito et al21 2011 Retrospective 2, 3
Akinbami et al22 2010 Prospective 2, 3
Rao et al23 2010 Prospective 3
Carter and Layton24 2009 Retrospective 2
Marioni et al25 2008 Retrospective 3
Warnke et al26 2008 Retrospective 3
Youssef et al27 2007 Literature review 2
Al-Nawas and Maeurer28 2008 Prospective 3
Ndukwe et al29 2007 Prospective 3, 7
Rega et al30 2006 Retrospective 2
Flynn et al31 2006 Prospective 6
Flynn et al32 2006 Prospective 7
Marioni et al33 2006 Retrospective 2, 3
Uluibau et al34 2005 Retrospective 2, 3
Wang et al35 2005 Retrospective 2, 3
Bross-Soriano et al36 2004 Retrospective 2, 3
Matthews et al37 2003 Systematic review 1
Bridgeman et al38 1995 Retrospective 2, 3, 4, 7
Har-El et al39 1994 Retrospective 2, 3
Lo Bue et al40 1993 Randomized clinical trial 7
* The reasons for exclusion were as follows: 1) failed to collect data from cases without checking reference articles, 2) did not
mention antibiotics used in treatment or their dosage, 3) did not describe separate evolution of patients for each antibiotic
group, 4) did not include clinical follow-up of patients who underwent antibiotic therapy, 5) did not deal with dental infections,
6) mentioned only results from other studies, and 7) included clinical situations not considered by method.
Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.

performed in patients—basically drainage (incision or removal of the cause, such as exodontics or

channel route) and removal of the cause of infection endodontics. In such cases, drainage must be
(mainly extraction and endodontics)—as well as for performed immediately, postponing the removal of
all studies in this review, including prospective the infection focus until clinical conditions are more
studies, medically compromised patients were not favorable.10,43,44
included. In an ideal treatment situation, drainage The first question this systematic review sought to
and removal of the cause of infection are performed answer regarding the use of antibiotics in DIs was
during the first contact with the patient; however, when to use them. All studies included in the review
there are situations that do not allow this to happen. involved DIs at stages at which there were either
Local intervention was used as part of treatment in regional or systemic manifestations of the organism
all study analyses, but some studies involved cases in to the infection. Brennan et al45 compared a group
which drainage was not possible or necessary and receiving penicillin with a group receiving no antibi-
removal of the cause as a single local intervention otics in patients with a toothache who had sought
was performed.41,42 However, in a DI at an advanced emergency assistance; they found that the use of anti-
stage, the opposite often occurs: An abscess that microbials did not prevent DI development, thus
could be drained is present, but the patient offers showing that antibiotics commonly prescribed for a
unfavorable clinical conditions for the procedure for toothache do not prevent the development of

No. of No. of DIs: ATBs SSD Between

Authors DIs Type of Study Surgical Intervention Administered Groups Patient Evolution and Comments

Igoumenakis et al46 179 Prospective Extraction or no intervention 91: AMP-SULB, MET, and Yes There was a statistically
(2015) exodontia significant association
86: AMP-SULB and MET between extraction and
infection resolution time.
Cachovan et al47 (2011) 31 Double-blind randomized I&D and/or removal of cause 15: MXF No MXF showed faster pain
clinical trial (extraction, RCT) during 16: CLI reduction and clinical
study improvement, but both were
Matijevic et al48 (2009) 90 Randomized clinical trial I&D and/or removal of cause 30: AMX Yes The ATB groups showed a
(extraction) during study 30: CEF reduced treatment time
30: surgical treatment only compared with patients who
were treated with only surgery.
Al-Nawas et al49 (2009) 21 Randomized clinical trial I&D during study 10: MXF No AMX-CLAV showed remission of
11: AMX-CLAV clinical signs and symptoms in
a shorter period than MXF.
Chardin et al41 (2009) 81 Double-blind randomized Patients without need for 42: AMX for 3 days No There were no differences in
clinical trial drainage; removal of 39: AMX for 7 days AMX treatment for 3 or 7 days.
cause during study
Rush et al50 (2007) 60 Randomized clinical trial I&D and/or removal of 31: CLI No The groups showed similar
cause (extraction, RCT) 29: AMP-SULB results.
during study


Kuriyama et al51 (2005) 112 Prospective study I&D during study; removal 65: PNC V or AMX No Patients who underwent I&D had
of cause after study 24: PNC V and MET a faster evolution than those
2: ERI who underwent drainage
9: MET through the pulp chamber,
6: ERI and MET, 400 mg regardless of ATB used. All
6: AMX-CLAV ATBs were clinically effective.
Al-Belasy and Hairam44 60 Randomized clinical trial I&D and removal of cause as 20: AZI Yes The ATB groups had a reduced
(2003) soon as conditions were 20: ERI treatment time compared with
favorable during study 20: without ATBs patients who underwent only
surgery. AZI reduced swelling
in a shorter period than ERI.
Adriaenssen52 (1998) 292 Randomized clinical trial Patients without need for 144: AZI No The groups showed similar
immediate drainage; 148: AMX-CLAV results.
removal of cause during
Martin et al53 (1997) 759 Prospective I&D and/or removal of cause 546: AMX Not applicable In 748 patients, resolution of
(extraction) during study 141: CLI signs and systemic symptoms
72: ERI presented from 2 to 3 days.
Eleven patients did not show
improvement because of
drainage failure at the first
consultation and underwent a
new drainage procedure.
Fazakerley et al54 100 Double-blind randomized Drainage through incision 33: AMX Yes Within 2 days, CFD showed a
(1993) clinical trial or canal during study 33: CFD statistically significantly
34: PNC V greater remission in pain,
temperature, and edema. After
5 days, all ATBs produced the
same results.
Deffez et al55 (1992) 176 Double-blind randomized When needed, I&D during 85: ROX Not applicable ATB efficacy was the same with
clinical trial study 91: ERI or without associated surgery.
Mangundjaja and 106 Double-blind randomized I&D during study; removal 54: AMP No The groups showed similar
Hardjawinata9 (1990) clinical trial of cause after study 52: CLI results.
Gilmore et al43 (1988) 55 Double-blind randomized I&D during study; removal 27: PNC V Not applicable The groups showed similar
clinical trial of cause after study 28: CLI results.
Hood56 (1978) 24 Randomized clinical trial I&D and removal of cause 18: MET Not applicable The groups showed similar
during study 19: PNC G and PNC V results.
Brown et al10 (1958) 51 Noncontrolled clinical Surgery when conditions 51: NVB Not applicable Of the patients, 45 responded
study were favorable during positively to ATBs and showed
study resolution of the case. Six
patients did not respond well,
and ATB therapy was changed
to PNC.
Abbreviations: AMP, ampicillin; AMX, amoxicillin; ATB, antibiotic; AZI, azithromycin; CEF, cephalexin; CFD, cephradine; CLAV, potassium clavulanate; CLI, clindamycin; DI, dental
infection; ERI, erythromycin; I&D, incision and drainage; MET, metronidazole; MXF, moxifloxacin; NVB, novobiocin; PNC, penicillin; RCT, root canal therapy; ROX, roxithromy-
cin; SSD, statistically significant difference; SULB, sulbactam.
Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.


FIGURE 1. Flow diagram for study selection.

Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.

infection. These findings appear to confirm that anti- condition resolution. However, the same patients,
biotic therapy recommendation in situations in which when compared with groups of patients who were
there is no regional or systemic involvement is unnec- prescribed antibiotics plus local intervention, showed
essary; rather, antibiotics are recommended only in cir- clinical sign and symptom remission within a signifi-
cumstances in which the patient’s immune defenses cantly longer period, which is proof that, when prop-
are unable to control infection, which can be deter- erly prescribed, antibiotics are excellent adjuvants and
mined through signals and clinical symptoms indi- should be used in these situations. Otherwise, the pre-
cating its spread, such as pronounced edema scriber will give no chance for the organism to control
(cellulite), limited mouth opening, tachycardia, the infection, contribute to resistant bacterium selec-
dysphagia, general malaise, and fever.57 Other studies tion, and subject the patient to the many adverse ef-
have confirmed the role of local intervention. Igoume- fects that these drugs can cause, increasing
nakis et al46 conducted a prospective study and found treatment costs.
a statistically significant association between the in- The second question was which drug is the most
fected element’s extraction and the infection’s resolu- effective. The findings of this review show that such
tion time, suggesting that in life-threatening cases, questioning only makes sense from the moment the
extraction of the involved tooth should be considered, antibiotic therapy is performed in association with
even when it is restorable. Al-Belasy and Hairam44 and some local intervention. The articles included in the
Matijevic et al48 reported that a total of 55 patients study showed that when drainage and/or removal of
with clinical signs and symptoms that could have justi- the cause of infection was properly carried out, all
fied antibiotic therapy were treated with drainage and/ the tested antibiotics were equally effective with
or removal of the cause and even so obtained clinical respect to clinical cure. Only 2 studies found


Random Definition of Reporting of Estimated

Selection in Inclusion and Loss to Validated Statistical Potential
Authors Year Population Exclusion Criteria Follow-Up Measurements Analysis Risk of Bias

Igoumenakis et al46 2015 Yes Yes Yes Yes Yes Low

Cachovan et al47 2011 Yes Yes No Yes Yes Moderate
Matijevic et al48 2009 Yes Yes No Yes Yes Moderate
Al-Nawas et al49 2009 Yes Yes Yes Yes Yes Low
Chardin et al41 2009 Yes Yes Yes Yes Yes Low
Rush et al50 2007 Yes No No No Yes High
Kuriyamaet al51 2005 Yes Yes No Yes Yes Moderate
Al-Belasy and Hairam44 2003 Yes Yes Yes Yes Yes Low
Adriaenssen52 1998 Yes Yes No Yes Yes Moderate
Martin et al53 1997 Yes Yes No Yes No High
Fazakerley et al54 1993 Yes Yes Yes Yes Yes Low
Deffez et al55 1992 Yes Yes No Yes No High
Mangundjaja and 1990 Yes Yes Yes Yes Yes Low
Gilmore et al43 1988 Yes Yes No Yes No High
Hood et al56 1978 Yes Yes No No No High
Brown et al10 1958 Yes No No Yes No High
Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.

statistically significant differences between antibi- and the product cost, besides the prescriber’s clinical
otics, but these referred to the clinical sign and symp- experience, should be taken into account.
tom remission period rather than final treatment The articles in our study showed that penicillin is
efficacy.44,54 Therefore, it is reasonable to infer that the most used and, therefore, first-choice drug, which
antibiotics do not differ as to their degree of efficacy can be associated with b-lactamase inhibitors if
but rather with reference to the total treatment needed. Flynn et al58 reported that 19% of the micro-
period, even when the continued increase in organisms collected from their patients were peni-
bacterial resistance episodes is considered. cillin resistant, and approximately 21% of their pa-
This review shows that the correct diagnosis and tients had treatment failure with penicillin.
local intervention should be given the greatest atten- Conversely, Kuriyama et al51 reported that clinical out-
tion by the surgeon, with the choice of antibiotic play- comes had not changed, showing a good clinical
ing a secondary role, provided that the antibiotic used response even when 37.5% of the micro-organisms
fits in with the action spectrum that has been proved collected from their patients were penicillin resistant.
effective in DI treatment. The safety of antibiotic use These data show that the presence of antibiotic-



No. of articles in which antibiotic was used 5 5 2 2 3 1 1 1

No. of patients treated with antibiotic 690 157 54 208 104 19 30 33
No. of patients with treatment failure 0 6 1 0 0 0 0 0

Note: Kuriyama et al51 stated that a group of 65 patients were given either AMX or PNC V but did not specify how many patients
received which antibiotic. These patients are not included in this table. In addition, there were drug combinations in some
studies, so a patient may have been prescribed more than 1 antibiotic mentioned in this table.
Abbreviations: AMP, ampicillin; AMX, amoxicillin; CEF, cephalexin; CFD, cephradine; CLAV, potassium clavulanate; PNC, peni-
cillin; SULB, sulbactam.
Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.



No. of articles in which antibiotic was used 2 5 2 5 2 1 1

No. of patients treated with antibiotic 25 268 218 191 164 85 51
No. of patients with treatment failure 0 0 0 8 13 5 6

Note: There were drug combinations in some studies, so a patient may have been prescribed more than 1 antibiotic mentioned in
this table.
Abbreviations: AZI, azithromycin; CLI, clindamycin; ERI, erythromycin; MET, metronidazole; MXF, moxifloxacin; NVB, novo-
biocin; ROX, roxithromycin.
Martins et al. Antibiotics and Odontogenic Infections. J Oral Maxillofac Surg 2017.

resistant bacteria does not necessarily mean there will parameters are the major indicators of the odonto-
be treatment failure. However, in the case of failure or genic infection picture evolution; however, additional
allergy, clindamycin and the macrolide class of drugs, tests should be performed when necessary.
particularly azithromycin, are viable alternatives, Sources of bias in this systematic review include
together with moxifloxacin, which has shown prom- publication bias (unpublished studies were not read)
ising results in clinical trials. Furthermore, Farmahan and language (only articles in English were analyzed).
et al57 reported that 95% of patients were discharged Nonrandomized and nonblinded clinical trials are
before antibiogram results were obtained. This finding more likely to show favorable results in the treatment
may raise questions regarding the therapeutic value of group than in the control group.59 According to Schulz
culture and sensitivity tests in any odontogenic infec- et al,60 inadequate randomization methods can over-
tion situation, which perhaps could be more state the estimated effect of treatment by up to 41%,
adequately used in more serious situations in which and when these methods are not well described, the
there is evidence of the need to use more spe- effect may be approximately 30%. Although all clinical
cific drugs. trials included in our study were randomized—most of
The third question that this systematic review which were double blinded—some did not clearly
sought to answer was how long antibiotics should be describe their randomization methods. Two prospec-
given. On the basis of the literature reviewed, one tive studies were included that, although having
cannot make a decision based on scientific evidence met all criteria, did not show methods as selective as
as to the optimal duration of antibiotic therapy in their clinical trials.
DIs. Two studies analyzed this theme. Chardin et al41 This study faced some restrictions. The ideal study
compared amoxicillin treatment for 3 and 7 days and to answer the guiding questions would have been a
found no statistically significant differences between double-blinded randomized clinical trial to test all ma-
groups, suggesting that patients’ exposure to antibi- jor antibiotics under the same conditions at different
otics should be reduced. Similarly, Martin et al53 re- treatment times. Because there are no such studies, an-
ported that of 759 patients who underwent drainage swers to these questions based on conclusions from
and/or removal of the cause and in whom antibiotic different studies that tested different antibiotics
therapy had been initiated on the first day, 98.6% were sought.
showed infection resolution within 2 or 3 days and In conclusion, in DI cases, once drainage and/or
had drug treatment interrupted without the need for removal of the cause of infection has been performed,
any other new intervention; they claimed that in all antibiotics tested were equally effective with
most DI cases, the duration of antibiotic therapy can respect to clinical cure. Therefore, most of the sur-
be limited to 2 or 3 days safely, once some local inter- geon’s attention should be directed toward proper
vention has been performed. By knowing that the pro- drainage and/or removal of the infection focus inas-
longed use of antibiotics only serves the purpose of much as the choice of antibiotics is not as successful
selecting resistant bacterial species,57 these studies in the treatment as local action. Antibiotics are only
may point out that, once drainage and/or removal of recommended in regional and/or systemic body mani-
the cause of infection has been performed, antibiotic festation situations and should be used as an aid to
therapy does not require a long cycle but rather only fight infection once the major treatment is surgical.
the patient’s follow-up to evaluate his or her evolution, When the real need for antibiotic therapy is deter-
preferably on a daily basis; its duration should be mined, this should be administered for the shortest
defined in accordance with clinical sign and possible period until the clinical cure of the patient
symptom remission. These findings show that clinical is obtained. We suggest that further randomized
MARTINS ET AL 2606.e10

clinical trials using rigorous methods should be per- 22. Akinbami BO, Akadiri O, Gbujie DC: Spread of odontogenic in-
fections in Port Harcourt, Nigeria. J Oral Maxillofac Surg 68:
formed to expand knowledge on antibiotic use in
2472, 2010
DI treatment. 23. Rao DD, Desai A, Kulkarni RD, et al: Comparison of maxillofacial
space infection in diabetic and nondiabetic patients. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 110:e7, 2010
References 24. Carter LM, Layton S: Cervicofacial infection of dental origin pre-
senting to maxillofacial surgery units in the United Kingdom: a
1. Gill Y, Scully C: Orofacial odontogenic infections: Review of national audit. Br Dent J 206:73, 2009
microbiology and current treatment. Oral Surg Oral Med Oral Pa- 25. Marioni G, Rinaldi R, Staffieri C, et al: Deep neck infection with
tol 70:155, 1990 dental origin: analysis of 85 consecutive cases (2000 -2006). Acta
2. Swift JQ, Gulden WS: Antibiotic therapy—Managing odonto- Otolaryngol 128:201, 2008
genic infections. Dent Clin N Am 46:623, 2002 26. Warnke PH, Becker ST, Springer IN, et al: Penicillin compared
3. Moher D, Liberati A, Tetzlaff J, et al: Preferred Reporting Items with other advanced broad spectrum antibiotics regarding anti-
for Systematic Reviews and Meta-Analyses: The PRISMA state- bacterial activity against oral pathogens isolated from odonto-
ment. J Clin Epidemiol 62:1006, 2009 genic abscesses. J Craniomaxillofac Surg 36:462, 2008
4. Moher D, Jones A, Lepage L, CONSORT Group (Consolidated 27. Youssef W, D’innocenzo R, Mehra P: Antibiotic therapy in the
Standards for Reporting of Trials): Use of the CONSORT state- management of severe odontogenic infections: a comparison
ment and quality of reports of randomized trials: A comparative of two treatment regimens. Oral Surg Oral Med Oral Pathol
before-and-after evaluation. JAMA 285:1992, 2001 Oral Radiol Endod 106:505, 2008
5. Moher D, Cook DJ, Eastwood S, et al: Improving the quality of 28. Al-Nawas B, Maeurer M: Severe versus local odontogenic bacte-
reports of meta-analyses of randomized controlled trials: The rial infections: comparison of microbial isolates. Eur Surg Res 40:
QUOROM statement. Quality of Reporting of Meta-Analyses. 220, 2008
Lancet 354:1896, 1999 29. Ndukwe KC, Ugboko VI, Akinwande JA, et al: Bacterial infec-
6. Stroup DF, Berlin JA, Morton SC, et al: Meta-analyses of observa- tions of the head and neck fascial spaces in Nigerians. West
tional studies in epidemiology: A proposal for reporting. Meta- Afr J Med 26:126, 2007
analysis of Observational Studies in Epidemiology (MOOSE) 30. Rega AJ, Aziz SR, Ziccardi VB: Microbiology and antibiotic sensi-
group. JAMA 283:2008, 2000 tivities of head and neck space infections of odontogenic origin.
7. von Elm E, Altman DG, Egger M, et al: The Strengthening the Re- J Oral Maxillofac Surg 64:1377, 2006
porting of Observational Studies in Epidemiology (STROBE) 31. Flynn TR, Shanti RM, Levi MH, et al: Severe odontogenic infec-
statement: Guidelines for reporting observational studies. Lan- tions, part 1: Prospective report. J Oral Maxillofac Surg 64:
cet 370:1453, 2007 1093, 2006
8. Bobrowski AN, Sonego CL, Chagas Junior OL: Postoperative 32. Flynn TR, Shanti RM, Hayes C: Severe odontogenic infections,
infection associated with mandibular angle fracture treatment part 2: prospective outcomes study. J Oral Maxillofac Surg 64:
in the presence of teeth on the fracture line: A systematic review 1104, 2006
and meta-analysis. Int J Oral Maxillofac Surg 42:1041, 2013 33. Marioni G, Castegnaro E, Staffieri C, et al: Deep neck infection in
9. Mangundjaja S, Hardjawinata K: Clindamycin versus ampicillin in elderly patients. A single institution experience (2000–2004).
the treatment of odontogenic infections. Clin Ther 12:242, 1990 Aging Clin Exp Res 18:127, 2006
10. Brown R, Thomassen PR, Singler JM: The use of novobiocin for 34. Uluibau IC, Jaunay T, Goss NA: Severe odontogenic infections.
treatment of infections of odontogenic origin. Oral Surg Oral Aust Dent J 50:S74, 2005
Med Oral Patol 11:598, 1958 35. Wang J, Ahani A, Pogrel MA: A five-year retrospective study of
11. Cope A, Francis N, Wood F, Mann MK, Chestnutt IG: Systemic an- odontogenic maxillofacial infections in a large urban public hos-
tibiotics for symptomatic apical periodon-titis and acute apical pital. Int J Oral Maxillofac Surg 34:646, 2005
abscess in adults. Status and date: New, published 6, 2014. 36. Bross-Soriano D, Arrieta-G omez JR, Prado-Calleros H, et al: Man-
12. Chi TH, Tsao YH, Yuan CH: Influences of patient age on deep agement of Ludwig’s angina with small neck incisions: 18 years
neck infection: clinical etiology and treatment outcome. Otol experience. Otol Head Neck Surg 130:712, 2014
Head Neck Surg 151:586, 2014 37. Matthews DC, Sutherland S, Basrani B: Emergency management
13. Farmahan S, Tuopar D, Ameerally PJ: The clinical relevance of of acute apical abscesses in the permanent dentition: a system-
microbiology specimens in head and neck space infections of atic review of the literature. J Can Dent Assoc 69:660, 2003
odontogenic origin. Br J Oral Maxillofac Surg 52:629, 2014 38. Bridgeman A, Wiesenfeld D, Hellyar A, Sheldon W: Major maxil-
14. Loyola-Rodriguez JP, Garcia-Cortes JO, Martinez-Martinez RE, lofacial infections. An evaluation of 107 cases. Aust Dent J 40:
et al: Molecular identification and antibiotic resistant bacteria 281, 1995
isolated from primary dentition infections. Aust Dent J 59:497, 39. Har-El G, Aroesty JH, Shaha A, Lucente FE: Changing trends in
2014 deep neck abscess: a retrospective study of 110 patients. Oral
15. Kara A, Ozsurekci Y, Tekcicek M, et al: Length of hospital stay Surg Oral Med Oral Pathol 77:446, 1994
and management of facial cellulitis of odontogenic origin in chil- 40. Bue AL, Sammartino R, Chisari G, et al: Efficacy of azithromycin
dren. Pediatr Dent 36:18E, 2014 compared with spiramycin in the treatment of odontogenic in-
16. Fedorowicz Z, Van Zuuren EJ, Farman AG, et al: Antibiotic use fections. J Antimicrob Chemother 31:119, 1993
for irreversible pulpitis. Cochrane Database Syst Rev 12, 2013 41. Chardin H, Yasukawa K, Nouacer N, et al: Reduced susceptibility
17. Rasteniene_ R, P uriene_ A, Aleksej
uniene_ J, et al: Odontogenic to amoxicillin of oral streptococci following amoxicillin expo-
maxillofacial infections: A ten-year retrospective analysis. Surg sure. J Med Microbiol 58:1092, 2009
Infect 16:305, 2015 42. Andriaenssen CF: Comparison of the efficacy, safety and tolera-
18. Gr€onholm L, Lemberg KK, Tj€aderhane L, et al: The role of unfin- bility of azithromycin and co-amoxiclav in the treatment of acute
ished root canal treatment in odontogenic maxillofacial infec- periapical abscesses. J Int Med Res 26:257, 1998
tions requiring hospital care. Clin Oral Invest 17:113, 2013 43. Gilmore WC, Jacobus NV, Gorbach SL, et al: A prospective
19. Lee YQ, Kanagalingam J: Deep neck abscesses: the Singapore double-blind evaluation of penicillin versus clindamycin in the
experience. Eur Arch Oto Rhino Laryngol 268:609, 2011 treatment of odontogenic infections. J Oral Maxillofac Surg 46:
20. Flynn TR: What are the antibiotics of choice for odontogenic in- 1065, 1988
fections, and how long should the treatment course last? Oral 44. Al-Belasy F, Hairam AR: The efficacy of azithromycin in the treat-
Maxillofac Surg Clin N Am 23:519, 2011 ment of acute infraorbital space infection. J Oral Maxillofac Surg
21. Saito CTMH, Gulinelli JL, Mar~ao HF, et al: Occurrence of odonto- 61:310, 2003
genic infections in patients treated in a postgraduation program 45. Brennan MT, Runyon MS, Batts JJ, et al: Odontogenic signs and
on maxillofacial surgery and traumatology. J Craniofac Surg 22: symptoms as predictors of odontogenic infection: A clinical
1689, 2011 trial. J Am Dent Assoc 137:626, 2006

46. Igoumenakis D, Giannakopoulos N, Parara E, et al: Effect of caus- 53. Martin MV, Longman LP, Hill JB, Hardy P: Acute dentoalveolar in-
ative tooth extraction on clinical and biological parameters of fections: An investigation of the duration of antibiotic therapy.
odontogenic infection: A prospective clinical trial. J Oral Maxil- Br Dent J 183:135, 1997
lofac Surg 73:1254, 2015 54. Fazakerley MW, McGowan P, Hardy P, Martin MV: A comparative
47. Cachovan G, B€ oger RH, Giersdorf I, et al: Comparative efficacy study of cephradine, amoxycillin and phenoxymethylpenicillin
and safety of moxifloxacin and clindamycin in the treatment in the treatment of acute dentoalveolar infection. Br Dent J
of odontogenic abscesses and inflammatory infiltrates: a phase 174:359, 1993
II, double-blind, randomized trial. Antimicrob Agents Chemo- 55. Deffez J, Scheimberg A, Rezvani Y: Multicenter Double-Blind
ther 55:1142, 2011 Study of the Efficacy and Tolerance of Roxithromycin versus
48. Matijevic S, Lazi Z, Kuljı́c-Kapulica N, Nonkovic Z: Empirical Erythromycin Ethylsuccinate in Acute Orodental Infection in
antimicrobial therapy of acute dentoalveolar abscess. Vojnosanit Adults. Diagn Microbiol Infect Dis 15:1335, 1992
Pregl 66:544, 2009 56. Hood FJC: The place of metronidazole in the treatment of acute
49. Al-Nawas B, Walter C, Morbach T, et al: Clinical and microbiolog- oro facial infection. J Antimicrob Chemother 4:71, 1978
ical efficacy of moxifloxacin versus amoxicillin/clavulanic acid 57. Pallasch TJ: Pharmacokinetic principles of antimicrobial ther-
in severe odontogenic abscesses: a pilot study. Eur J Clin Micro- apy. Periodontol 2000 10:511, 1996
biol Infect Dis 28:75, 2009 58. Flynn TR, Shanti RM, Levi MH, et al: Severe odontogenic infec-
50. Rush DE, Abdel-Haq N, Zhu J, et al: Clindamycin Versus Unasyn tions, part 1: Prospective report. J Oral Maxillofac Surg 64:
in the Treatment of Facial Cellulitis of Odontogenic Origin in 1093, 2006
Children. Clin Pediatr (Phila) 46:154, 2007 59. Jadad AR, Moore RA, Carroll D, et al: Assessing the quality of re-
51. Kuriyama T, Absi EG, Williams DW, Lewis MAO: An outcome ports of randomized clinical trials: Is blinding necessary? Con-
audit of the treatment of acute dentoalveolar infection: Impact trol Clin Trials 17:112, 1996
of penicillin resistance. Br Dent J 198:759, 2005 60. Schulz KF, Chalmers I, Hayes RJ, Altman DG: Empirical evidence
52. Andriaenssen CF: Comparison of the Efficacy, Safety and Tolera- of bias: Dimensions of methodological quality associated with
bility of Azithromycin and Co-amoxiclav in the Treatment of estimates of treatment effects in controlled trials. JAMA 273:
Acute Periapical Abscesses. J Intern Med Res 26:257, 1998 408, 1995