ONLINE EXCLUSIVES
Microbial flora and antibiotic
resistance in odontogenic
abscesses in Upstate New York
Ann W. Plum, MD; Anthony J. Mortelliti, MD; Ronald E. Walsh, NP
Abstract
Abscesses in the head and neck frequently have odon-
togenic sources. As bacterial pathogens and antibiotic
resistance patterns may change over time and based on
location, we describe the current common bacteria found
in odontogenic abscesses, the prevalence of antibiotic re-
sistance, and differences in each between pediatric and
adult patients in Upstate New York. This is a retrospective
review of patients who underwent drainage of odontogen-
ic abscesses (n = 131) from 2002 to 2012 at an academic
institution. The medical records were reviewed for results
of abscess cultures, comorbidities, and drainage proce-
dures. Polymicrobial sources were identified in 60.3%
and monomicrobial in 33.6%. Overall, the most common
bacteria were alpha hemolytic Streptococci (33.6%),
Streptococcus milleri (32.1%), Prevotella (16.8%), and
coagulase-negative Staphylococcus (14.5%). Candida
and Morganella spp were more common in children than
in adults. Overall, antibiotic resistance was observed in
seven different pathogens. The most common antibiotic
resistances were to clindamycin and erythromycin, which
should be considered when deciding initial antibiotic ther-
apy, especially in adult patients, who trended in this study
toward having pathogens with higher rates of resistance.
Introduction
Head and neck abscesses commonly have an odonto-
genic source. The microbiology of these abscesses can
be influenced by a variety of factors, including location
in the oral cavity, the geographic location of the patient,
and host factors.1 It has been proposed that Strepto-
coccus salivarius and Veillonella spp often colonize the
From the Department of Otolaryngology and Communication
Sciences, State University of New York Upstate Medical University,
Syracuse, N.Y.
Corresponding author: Ann Plum, MD, Department of Otolaryngology-
Head and Neck Surgery, 170 Manning Dr., Chapel Hill, NC 27599.
Email: ann_plum@med.unc.edu
Previous presentation: Presented as a poster at the American Academy
of Otolaryngology–Head and Neck Surgery annual meeting;
September 21-24 2014; Orlando, Fla.
Volume 97, Number 1-2
tongue and buccal surfaces; Actinomycetes and Strepto-
coccus spp, such as sanguinis, mutans, and mitis, often
colonize tooth enamel; and Fusobacterium, Prevotella,
Porphyromonas, and Spirochetes colonize gingival sur-
faces.2 The breakdown in the mucosa over those regions
provides entry for these specific pathogens.2 Host fac-
tors that result in either mucosal breakdown (e.g., dental
procedures) or impaired immune systems (such as with
immunodeficiencies, chemotherapy, and poorly con-
trolled diabetes) play a crucial role in these infections.2
Often, these abscesses require antimicrobial therapy
and surgical intervention via dental extraction or inci-
sion and drainage.1 The initial antimicrobial of choice
is often penicillin or a penicillin derivative such as
amoxicillin-clauvulanic acid, except in the presence
of an allergy or in a patient recently on a penicillin for
this infection, in which case clindamycin is frequently
used.1,3 The incidences of antibiotic resistance and the
exact microbial flora within odontogenic abscesses
have varied in prior studies.1-8 Here, we characterize
the current microbial flora and associated antibiotic
resistances observed in Upstate New York.
Patients and methods
Adult and pediatric patients who presented to Upstate
Medical University with head and neck abscesses with
an underlying odontogenic source from 2002 through
2012 formed the study population. Patients 18 years or
older were considered adults (mean age: 37 years), and
those 17 years or younger were considered pediatric
patients (mean age: 11 years).
The diagnosis of an odontogenic abscess was made via
drainage that yielded purulent material. All patients were
deemed to have an odontogenic source by way of clinical
assessment by an otolaryngologist (resident or attending).
Drainage of the odontogenic abscesses was performed
via either a transoral or transcutaneous incision. A trans-
cutaneous approach was used in patients with deep neck
space involvement or in those with multiple loculations,
or when an external approach was determined, either
www.entjournal.com E27
 Plum, Mortelliti, Walsh

Table 1. Dental history of patients with
odontogenic abscess
Dental history
Dental fracture
Dental or oral procedure
Dental extraction
Unspecified procedure
Root canal
Dental implant
Arch bar removal
Prior dental infection
Prior dental abscess
reviewed. All patients included in this study had drainage
of their abscess and cultures taken from the abscess fluid.
Patients with an abscess at the site of a facial fracture,
Incidence rate (%) other than a fractured tooth, were excluded.
4.7 Medical records for each patient were reviewed to
20.2 assess for microbial flora and the rate of antibiotic re-
14.8 sistance, which were the primary outcomes of the study.
Secondary outcomes included complications related to
3.9
the abscess, complications related to drainage of the
1.6 abscess, and recurrence of the abscess.
0.8 Antibiotic resistance was determined by the minimum
0.8 inhibitory concentration. Pathogens with intermediate
12.8 minimum inhibitory concentrations were not consid-
ered to be resistant.
10.7

Results
clinically or radiographically, to be safer or to offer better The distribution of men (50.4%) and women (49.6%)
exposure. The drainage procedure was performed in the in the cohort was even. The average age was 35.1 years.
operating room in cases in which severe trismus pre- A compromised immune system was relatively rare,
vented exposure, intraoral drainage failed, or there was occurring in 3.1% of patients, and was caused by either
poor patient tolerance or deep space neck involvement. use of immunosuppressant medications (chemothera-
Transcutaneous drainage was performed by an peutic agents at the time of diagnosis of odontogenic
incision 2 cm below the mandible and elevation of a
subplatysmal flap, when applicable. The abscess pocket Table 2. Microbial flora found in odontogenic abscesses
was opened and all loculations were broken up. Irrigation
was used to flush the abscess pocket. A Penrose drain Pathogen Incidence (%)
was placed at the conclusion of these procedures and Polymicrobial 60.3
removed postoperatively. Monomicrobial 33.6
Intraoral drainage was performed by injecting 1%
No growth 6.1
lidocaine with 1:100,000 epinephrine in the mucosa
overlying the area of fluctuance in the oral cavity and Streptococcus 66.4
using either a #15 or a #11 blade to incise the mucosa Alpha hemolytic 33.6
over the fluctuance. Hemostat forceps were used to Milleri 32.1
open the abscess and break the loculations, and the Beta hemolytic 8.4
pocket was irrigated.
Non-hemolytic 3.8
Culture specimens were taken from the abscess pock-
et, either using the initial aspirate before incision and Mitis 0.8
drainage or from swabbing the pocket. All cultures were Dysgalactiae subsp equisimilis 0.8
placed in BBL CultureSwab Plus transport containers Prevotella 16.8
(Becton, Dickinson and Company; Franklin Lakes, Staphylococcus 15.3
N.J.), which contain Amies gel to preserve anaerobes
Corynebacterium 9.2
and aerobes. Cultures were grown in the institution’s
microbiology laboratory on nonselective media con- Neisseria 6.1
sisting of blood, chocolate, magnesium, and Columbia Bacteroides 5.3
agars for at least 3 days to determine growth. The mi- Haemophilus 4.6
crobiology laboratory using Clinical and Laboratory Candida 3.8
Standards Institute guidelines were used to determine
antimicrobial resistance. Eikenella corrodens 3.1
Internal Institutional Review Board approval was ob- Lactobacillus 2.3
tained before beginning this study. The medical records Stomatococcus 1.5
of all patients who were diagnosed with an odontogenic Gram-positive cocci not specified 1.5
abscess by physicians in the Department of Otolaryn-
Parvimonas micra (peptostreptococcus) 1.5
gology from 2002 through 2012 at the State University
of New York Upstate Medical University Hospital were Propionibacteria eubacterium 1.5

E28 www.entjournal.com ENT-Ear, Nose & Throat Journal January/February 2018

 Microbial flora and antibiotic resistance in odontogenic abscesses in Upstate New York

Table 3. Differences in prevalence of microbial flora in odontogenic The odontogenic abscess resulted in
pockets between pediatric and adult patients complications in 3.6% of patients. These
complications included osteomyelitis
Pathogen Pediatric (%) Adult (%) p Value
of the mandible (1.6%), pathologic
Polymicrobial 70 59.5 0.516 fracture of the mandible (0.8%), myo-
Monomicrobial 20 34.7 0.342 cardial infarction (0.8%), mediastinitis
Streptococcus 50 67.8 0.254 (0.8%), and death (0.8%). One adult
Alpha hemolytic 20 34.7 0.342
patient had a complication from the
drainage procedure (excessive bleeding
Milleri 30 32.2 0.889
from injury to a facial vein during an
Beta hemolytic 0 9.1 0.317 external approach).
Non-hemolytic 0 4.1 0.509 The odontogenic abscesses were usu-
Mitis 0 0.8 0.772 ally polymicrobial (table 2). No growth
Dysgalactiae subsp Equisimilis 0 0.8 0.772
after 5 days was uncommon. The most
common pathogens were Streptococcus
Prevotella 10 17.4 0.549
spp (alpha hemolytic and milleri),
Staphylococcus 10 15.7 0.631 Prevotella spp, and Staphylococcus spp.
Oropharyngeal oral flora 20 10.7 0.379 The three most common types of strep-
Corynebacterium 10 9.1 0.92 tococcal strains were alpha hemolytic,
Neisseria 10 5.8 0.596
milleri, and beta hemolytic.
The prevalence of most of the odon-
Bacteroides 0 5.8 0.435
togenic abscess pathogens was similar
Haemophilus 0 5.0 0.472 between pediatric patients (n = 10) and
Candida 20 2.5 0.00544 adults (n = 121), with the exception of
Albicans 10 0 0.00048 Candida spp and Morganella morganii
Tropicalis 10 2.5 0.186
being significantly higher in the pedi-
atric patients (table 3).
Eikenella corrodens 0 3.3 0.562
Overall, antibiotic resistance was
Morganella morganii 10 0 0.00048 observed in seven pathogens (table
4). Clindamycin and erythromycin
abscess for the treatment of lymphoma) or an under- were the most common antibiotics to which resistance
lying immunologic disorder (history of human immu- occurred. Stomatococcus, Staphylococcus, Corynebacte-
nodeficiency virus and on highly active antiretroviral rium, alpha hemolytic Streptococcus, Lactobacillus, and
therapy, or B-cell deficiency). Streptococcus milleri all demonstrated the presence of
The most common medical comorbidities in the clindamycin resistance. Penicillin-resistant strains of
cohort were hypertension (13.2%), diabetes (12.4%), Stomatococcus and Lactobacillus spp were observed.
and asthma (9.3%). A history of dental procedures was Staphylococcus, Corynebacterium, alpha hemolytic
also common (20.2%), with some patients undergoing Streptococcus, and S milleri all had strains resistant to
multiple different dental procedures (table 1). Likewise, erythromycin. S milleri also demonstrated tetracycline
a history of odontogenic infections and abscesses was resistance, Staphylococcus demonstrated methicillin re-
present in 12.9 and 10.7%, respectively. sistance, and M morganii demonstrated both ampicillin
Overall, 67.2% of patients received at least one dose and cefazolin resistance. Antibiotic resistance in patho-
of antibiotic before incision and drainage and before gens was more common in adult patients, except for M
cultures were obtained. More pediatric patients than morganii and its resistance to ampicillin and cefazolin.
adult patients received antibiotics before culture (80%
vs. 66.1%). Some 58.9% of the cohort underwent intra- Discussion
oral incision and drainage of their abscess. The details Studies suggest that head and neck abscesses second-
regarding the drainage procedure were unavailable for ary to odontogenic sources are polymicrobial in 59 to
1.6%. Multiple drainage procedures were necessary in 89% of cases.2,4,5 This rate is similar to the rate in our
19.4%. Incision and drainage in the operating room study (60.3%). Streptococcus, Prevotella, and Staphylo-
was the initial drainage procedure in 38.9% of adults coccus spp were the most common pathogens identi-
and 40% of pediatric patients. fied in our study. Most studies are consistent in that
The parapharyngeal space was involved in 11.6%; no Prevotella and Streptococcus spp are among the com-
patients had involvement of the retropharyngeal space. mon sources.1,2,4-9

Volume 97, Number 1-2 www.entjournal.com E29

 Plum, Mortelliti, Walsh

The preva lence of Table 4. A description of the pathogens determined to have antibiotic resistance
Staphylococci in odonto- and the incidences of the antibiotic resistance
genic oral cavity abscess-
es is uncertain, although Total % % Resistant in % Resistant
several studies report Bacteria Antibiotic resistant pediatric patients in adult patients
their presence. 1,5,6,9
In Stomatococcus Clindamycin 50.0 N /A 50.0
our study, alpha hemo- Penicillin 50.0 N/A 50.0
lytic Streptococcus was Staphylococcus Clindamycin 10.0 0 10.5
seen in 33.6%, S milleri
Erythromycin 10.0 0 10.5
in 32.1%, and beta he-
molytic Streptococcus in Oxacillin 10.0 0 10.5
8.4% of abscesses. Alpha Penicillin 10.0 0 10.5
hemolytic streptococcal Corynebacterium Clindamycin 8.3 0 9.1
strains include S viridans, Erythromycin 8.3 0 9.1
mutans, and mitis, which
Streptococcus Clindamycin 2.3 0 2.4
prior studies suggest
as the more common (alpha hemolytic) Erythromycin 2.3 0 2.4
streptococcal strains, a Lactobacillus Clindamycin 33.3 N/A 33.3
finding consistent with Penicillin 33.3 N/A 33.3
our study.1,4-7 Streptococcus milleri Erythromycin 35.7 33.3 38.5
S milleri has also been
Clindamycin 33.3 0 3.5
reported to be extremely
common in odontogenic Tetracycline 2.4 0 2.6
abscesses in the United Morganella morganii Ampicillin 100.0 100.0 N/A
Kingdom, as in our study Cefazolin 100.0 100.0 N/A
in Upstate New York.5
Incidences of each bac-
terial strain do vary, and the variation observed in the and erythromycin was higher than the incidence of
literature may reflect the geographic region and local penicillin resistance, which is consistent with several
bacterial flora.1-9 studies.4-7,11 This finding suggests that continuing with
In comparing pathogenic species between pediatric a penicillin derivative (i.e., amoxicillin-clavulanic
and adult patients, a difference in Morganella and Can- acid) as first line seems to be a reasonable approach,
dida spp was apparent between the two study cohorts, reserving clindamycin for penicillin-allergic patients.
but the smaller size of the pediatric cohort in this study Overall, for most pathogens with antibiotic resistance,
compared with the adult cohort might have magnified resistance was more common in adult patients, which
the difference. may be a result of medical comorbidities, poor denti-
Penicillin resistance was relatively uncommon in tion, and exposure.
our study population. Only three abscesses grew or- The current literature is composed of smaller studies
ganisms resistant to penicillin, due to Lactobacillus, that do not consider the possible difference in resis-
Stomatococcus, and Morganella spp. The literature tance rates between pediatric and adult patients or
varies in terms of penicillin resistance, ranging from 0 that antibiotic resistance and prevalence of particular
to 100% depending on the organism, date, and location pathogens might differ based on age. This study pro-
of the study.1,2,4-6,9-11 vides the initial look at the effect of age on odontogenic
No streptococcal strains showed penicillin resistance abscesses and shows that antibiotic resistance and the
in our study, which is similar to the finding in several type of pathogens may differ. However, due to the small
studies that show low resistance rates of streptococcal number of pediatric patients in this study, a larger-scale
strains to penicillin, given that streptococcal strains study focusing on pediatric patients with odontogenic
do not produce beta lactamase.2,4-6,9 Although not seen abscesses is needed.
in this cohort, Prevotella, Veillonella, Haemophilus, There are several limitations to this study. First, this
and Eikenella spp may exhibit resistance to penicillin is a single institution study. Nevertheless, 123 abscesses
through the production of beta lactamase.11,12 with cultures grew organisms, so the results would not
Clindamycin and erythromycin resistance was pres- be expected to change drastically with inclusion of more
ent in multiple bacterial species and abscesses in our centers. Several patients were on antibiotics before their
series, and the incidence of resistance to clindamycin abscesses were drained, which could alter the flora.

E30 www.entjournal.com ENT-Ear, Nose & Throat Journal January/February 2018

 MICROBIAL FLORA AND ANTIBIOTIC RESISTANCE
IN ODONTOGENIC ABSCESSES IN UPSTATE NEW YORK

The number of patients with deep space neck abscesses

or sepsis was too small to analyze separately, and antibi-
otic resistance may change based on the severity of the
infection.9 Likewise, we could not analyze the impact
of immunodeficiency on microbial flora or antibiotic
resistance given the small number of patients with
immunodeficiency in the study.
In addition, at our facility we do not test for resistance
to amoxicillin-clavulanic acid. There does not appear to
be a significant difference between adult and pediatric
odontogenic abscess sources, other than antibiotic re-
sistance tending to be more common in adult patients.
The most common antibiotics to which resistance was
observed were clindamycin and erythromycin, which
should be considered when choosing initial antibiotic
therapy, especially in adult patients.

Acknowledgments
A special thank you to Alice Daniels, who assisted in
gathering patient charts, and Lisa Craner, who assisted
in creating the list of patients for this study.

References
1. Walia IS, Borle RM, Mehendiratta D, Yadav AO. Microbiology
and antibiotic sensitivity of head and neck space infections of
odontogenic origin. J Maxillofac Oral Surg 2014;13(1):16-21.
2. Levi ME, Eusterman VD. Oral infections and antibiotic therapy.
Otolaryngol Clin North Am 2011;44(1):57-78.
3. López-Píriz R, Aguilar L, Giménez MJ. Management of
odontogenic infection of pulpal and periodontal origin. Med
Oral Patol Oral Cir Bucal 2007;12(2):E154-9.
4. Sobottka I, Wegscheider K, Balzer L, et al. Microbial analysis of a
prospective, randomized, double-blind trial comparing moxifloxacin
and clindamycin in the treatment of odontogenic infiltrates and
abscesses. Antimicrob Agents Chemother 2012;56(5):2565-9.
5. Robertson D, Smith AJ. The microbiology of the acute dental
abscess. J Med Microbiol 2009;58(Pt2):155-62.
6. Rega AJ, Aziz SR, Ziccardi VB. Microbiology and antibiotic
sensitivities of head and neck space infections of odontogenic
origin. J Oral Maxillofac Surg 2006;64(9):1377-80.
7. Kuriyama T, Karasawa T, Nakagawa K, et al. Bacteriologic
features and antimicrobial susceptibility in isolates from orofacial
odontogenic infections. Oral Surg Oral Med Oral Pathol Oral
Radiol Endod 2000;90(5):600-8.
8. Eick S, Pfister W, Straube E. Antimicrobial susceptibility of
anaerobic and capnophilic bacteria isolated from odontogenic
abscesses and rapidly progressive periodontitis. Int J Antimicrob
Agents 1999;12(1):41-6.
9. Al-Nawas B, Maeurer M. Severe versus local odontogenic
bacterial infections: Comparison of microbial isolates. Eur Surg
Res 2008;40(2):220-4.
10. Baumgartner JC, Xia T. Antibiotic susceptibility of bacteria
associated with endodontic abscesses. J Endond 2003;29(1):44-7.
11. Kuriyama T, Williams DW, Yanagisawa M, et al. Antimicrobial
susceptibility of 800 anaerobic isolates from patients with
dentoalveolar infection to 13 oral antibiotics. Oral Microbiol
Immunol 2007;22(4):285-8.
12. Kinder SA, Holt SC, Korman KS. Penicillin resistance in the
subgingival microbiota associated with adult periodontitis. J Clin
Microbiol 1986;23(6):1127-33.

Volume 97, Number 1-2 www.entjournal.com E31