J Oral Maxillofac Surg 64:1093-1103, 2006

Severe Odontogenic Infections, Part 1: Prospective Report

Thomas R. Flynn, DMD,* Rabie M. Shanti, Michael H. Levi, ScD, Arthur K. Adamo, DDS, Richard A. Kraut, DDS, and Norman Trieger, DMD, MD
Purpose: The purpose of this study was to prospectively evaluate a series of patients with severe

odontogenic infections (OI). Patients and Methods: In this study, 37 consecutive hospitalized patients with odontogenic infection were treated with intravenous penicillin (PCN) (unless allergic), and prompt incision and drainage. Standardized data collection included demographic, preadmission, time-related, preoperative, anatomic, treatment, microbiologic, and complications information. Appropriate descriptive statistics were computed. Results: The sample consisted of 37 subjects (38% female) with a mean age of 34.9 years. Three subjects (8%) had immunocompromising diseases. Caries was the most frequent dental disease (65%) and the lower third molar was the most frequently involved tooth (68%). Trismus and dysphagia were present on admission in over 70% of cases. The masticator, perimandibular (submandibular, submental, and/or sublingual), and peripharyngeal (lateral pharyngeal, retropharyngeal, and/or pretracheal) spaces were infected in 78%, 60%, and 43% of cases, respectively. Abscess was found in 76% of cases. PCN-resistant organisms were identied in 19% of all strains isolated and in 54% of patients with sensitivity data. PCN therapeutic failure occurred in 21% of cases and reoperation was required in 8%. Length of hospital stay was 5.1 3.0 days. No deaths occurred. Conclusions: This study indicated that PCN resistance, resulting in PCN therapeutic failure, was unacceptably high in this sample. Alternative antibiotics, such as clindamycin, should be considered in hospitalized patients with OI. Masticator space infection occurred much more frequently than previously reported. Trismus and dysphagia should be appreciated as signicant indicators of severe OI. 2006 American Association of Oral and Maxillofacial Surgeons J Oral Maxillofac Surg 64:1093-1103, 2006 Most published case series of severe odontogenic infections are retrospective.1-9 As such, they are subject to errors due to missing data, misclassication or misinterpretation of clinical records, and inconsistent treatment methods. Therefore, we designed a prospective descriptive study of 37 consecutive cases of severe odontogenic infections (OI), dened as those warranting hospital admission. Standardized data
*Assistant Professor, Oral and Maxillofacial Surgery, Harvard School of Dental Medicine; and Associate Visiting Surgeon, Massachusetts General Hospital, Boston, MA. Howard Hughes Medical Institute-National Institutes of Health Research Scholar, National Institutes of Health, Bethesda, MD; and Predoctoral Candidate, Harvard School of Dental Medicine, Boston, MA. Co-Director, Microbiology, Monteore Medical Center; and Associate Professor of Pathology (Clinical), Albert Einstein College of Medicine, Bronx, NY. Director, Oral and Maxillofacial Surgery, North Bronx Healthcare Network; and Associate Clinical Professor of Oral and Maxillofacial Surgery, Albert Einstein College of Medicine, Bronx, NY.

were collected from each case, and uniform treatment methods were used. The specic aims of this study were: 1) to accumulate prospective descriptive data to characterize severe OI, and 2) to determine the success rate of intravenous penicillin (PCN) for treatment of severe OI. Our hypothesis was that intravenous PCN, combined with prompt surgical incision and drainage
Professor and Chairman, Department of Oral and Maxillofacial Surgery, Monteore Medical Center/Albert Einstein College of Medicine, Bronx, NY. Chairman Emeritus and Professor, Oral and Maxillofacial Surgery, Department of Oral and Maxillofacial Surgery, Monteore Medical Center/Albert Einstein College of Medicine, Bronx, NY. Address correspondence and reprint requests to Dr Flynn: Harvard School of Dental Medicine, 188 Longwood Avenue, Boston, MA 02115; e-mail: thomas_ynn@hsdm.harvard.edu
2006 American Association of Oral and Maxillofacial Surgeons

0278-2391/06/6407-0015$32.00/0 doi:10.1016/j.joms.2006.03.015

1093

1094 (I&D) of all affected anatomic spaces, would result in improvement in swelling, fever, and white blood cell count (WBC) by 48 hours after surgery.

SEVERE ODONTOGENIC INFECTIONS

Patients and Methods

STUDY DESIGN/SAMPLE

In this study we used a prospective case series design, in which all consecutive patients with OI severe enough to justify hospitalization were treated with intravenous PCN (unless allergic) and incision and drainage of all affected anatomic deep fascial spaces as soon as possible during the hospital stay. The subjects enrolled in this study presented for care between March 1996 and June 1999 at 1 of 4 large urban hospitals served by the Monteore Medical Center Department of Dentistry, including Monteore Medical Center, Jack Weiler Hospital at the Albert Einstein College of Medicine, North Central Bronx Hospital, and Jacobi Medical Center. A total of 37 subjects were enrolled in this study based on the following criteria: severe OI (as determined by an attending oral and maxillofacial surgeon) and hospital admission. Informed consent was obtained using forms and procedures developed for this institutional review boardapproved study. The criteria for hospital admission were: OI causing swelling in one or more of the deep fascial spaces of the head and neck, impending threat to the airway or vital structures, fever greater than 101F, need for general anesthesia, or the need for inpatient control of a concomitant systemic disease. Potential subjects were excluded from this study according to the following criteria: pregnancy, nonodontogenic cause (eg, trauma-related or upper respiratory infection), and refusal of consent. Previously published nomenclature and descriptions of the deep fascial spaces were used for the purposes of this study.10,11
TREATMENT METHODS

vested by either aspiration or by swab sampling of open surgical wounds. All spaces that were opened were copiously irrigated and maintained using latex Penrose or Jackson-Pratt type drains. Postoperative CT scanning was performed when indicated based on the patients progress and response to treatment. All patients received PCN intravenously at a dose of 2 million units every 4 hours, unless they gave a history of PCN allergy or presented with signs and symptoms of necrotizing fasciitis. Clindamycin 900 mg every 8 hours was administered intravenously to PCN-allergic subjects. PCN therapeutic failure (PTF) was dened as: 1) development of an allergic or toxic reaction to the antibiotic; 2) development of necrotizing fasciitis, in which case broad-spectrum antibiotic therapy was indicated; or 3) no improvement of temperature, WBC, and swelling after 48 hours of continuous intravenous therapy with the same antibiotic, plus a postoperative CT scan demonstrating adequate surgical drainage of all anatomic deep fascial spaces affected by cellulitis or abscess. If inadequate surgical drainage was shown on postoperative CT scan, then the operation was repeated, with appropriate drainage of all spaces affected by cellulitis or abscess (Fig 1).
DATA COLLECTION

All patients were subjected to the same treatment protocol. The patient was prepared for surgery as soon as possible after hospital admission. Appropriate preoperative medical workup was performed, including history and physical examination, complete blood cell count, urinalysis, appropriate imaging studies, and medical consultation when necessary. Preoperative imaging methods included periapical and panoramic dental x-rays, as well as preoperative computed tomography (CT) scanning in selected cases. After establishment of a secure airway, the skin and mucosa were prepared with antiseptic solution. I&D was performed for all anatomic fascial spaces that were involved by either cellulitis or abscess. Specimens for culture and sensitivity testing were har-

The demographic variables recorded were age, gender, and race. Preadmission variables were: smoking, drug allergies, preadmission antibiotic therapy, and the presence of immunocompromising diseases (such as diabetes, human immunodeciency virus [HIV] seropositivity, use of immunosuppressive medications, severe kidney disease, and cancer chemotherapy within the previous year). The time-related variables included the number of preoperative days of pain, preoperative days of swelling, length of stay (LOS), operating room time, time between admission and surgery, and season of the year. Preoperative clinical variables included causative teeth, number of teeth involved, dental diagnosis (such as caries, periodontal disease, or pericoronitis), dyspnea, dysphagia, trismus (maximum interincisal opening 20 mm), WBC, and admission core temperature. For purposes of statistical analysis, certain variables were grouped together. For example, upper teeth were grouped into categories of upper anterior, upper non-third-molar posterior, and upper third molars. Anamnestic data were obtained from the subjects in a standardized fashion, limited to the current episode of infection, and veried by the attending surgeon. The anatomic variables included deep fascial spaces involved by cellulitis or abscess, number of spaces affected, and severity score (SS). A severity score (low 1, moderate 2, or severe 3) was developed for this study by categorizing the deep

FLYNN ET AL

1095

fascial spaces according to their proximity to the airway, likelihood of preventing access to the airway, or proximity to vital structures, such as the contents of the mediastinum or the cranial cavity (Table 1). The admission SS was the sum of the severity ratings for each of the anatomic spaces that were affected by

Table 1. SEVERITY SCORES FOR SEVERE ODONTOGENIC INFECTIONS

Severity Score Severity score 1 (Low risk to airway or vital structures) Severity score 2 (Moderate risk to airway or vital structures)

Anatomic Space Vestibular Subperiosteal Space of the body of the mandible Infraorbital Buccal Submandibular Submental Sublingual Pterygomandibular Submasseteric Supercial temporal Deep temporal (or infratemporal) Lateral pharyngeal Retropharyngeal Pretracheal Danger space (space 4) Mediastinum Intracranial infection

Severity score 3 (High risk to airway or vital structures)

NOTE. The severity score for a given subject is the sum of the severity scores for all of the spaces involved by cellulitis or abscess, based on clinical and radiographic examination. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

cellulitis or abscess, as determined by clinical and radiographic examination. For example, a subject whose infection involved the right buccal, right submandibular, and right lateral pharyngeal spaces was given an SS of 6, which is the total of 1 for the buccal space, 2 for the pterygomandibular space, and 3 for the lateral pharyngeal space. For purposes of statistical analysis, masticator space infection was dened as infection involving any or all of the following spaces: pterygomandibular, submasseteric, supercial temporal, or deep temporal (including the infratemporal portion of the deep temporal space).

FIGURE 1. A, Contrast-enhanced CT of Case #20, 5 days after intraoral and extraoral I&D with drains placed in the right pterygomandibular (solid arrow), anterior compartment of the right lateral pharyngeal (open arrow), and posterior compartment of the right lateral pharyngeal (arrowhead) spaces. Note that the swollen oropharyngeal tissues are in circumferential contact with the endotracheal tube. This CT shows adequate surgical drainage of all infected spaces. In this case, the criteria for PTF were met and penicillin was replaced with clindamycin. Rapid improvement ensued. B, Contrast-enhanced CT of Case #18, 5 days after intraoral I&D of the left pterygomandibular space. Note the penrose drain in that space (open arrow). The infection has extended into the left and right lateral pharyngeal spaces (white arrows) and the retropharyngeal space (arrowhead). This patient needed reoperation, with tracheotomy under local anesthesia and repeat I&D of all affected spaces. Reprinted from Flynn TR: Surgical management of orofacial infections. Atlas Oral Maxillofac Surgery Clin North Am 8:99, 2000, with permission from Elsevier. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

1096

SEVERE ODONTOGENIC INFECTIONS

Table 2. DEMOGRAPHIC VARIABLES

Mean SD Age (years) Gender Male Female Ethnicity African-American/black White Hispanic Asian 34.9 15.8

Range 1476

n (Cases)

% of Cases

23 14 20 8 8 1

62 38 54 22 22 3

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

The recorded treatment variables included the anatomic spaces drained and the presence or absence of pus at the time of drainage. If pus was present, the stage of infection was recorded as abscess; if not, the stage was recorded as cellulitis. The product at I&D was categorized as pus if it could be described as a creamy yellow to grey uid, or ecks or curds of pus suspended in a bloody uid drained from any of the spaces that were explored at surgery. Cellulitis was recorded when the product at I&D consisted only of serosanguineous uid. Treatment variables also included the number of spaces drained, type and number of drains used, number of teeth extracted, and antibiotic(s) used. The anesthesia treatment variables were: type of anesthesia (general or local), difculty of intubation, type of anesthesia for intubation (general, intravenous sedation, or topical anesthesia), and type of intubation (endotracheal or tracheotomy). The microbiologic variables were: genus and species identication, oxygen requirements (anaerobic or aerobic, including facultative), PCN and clindamycin sensitivity, and number of species isolated per case. For purposes of statistical analysis, certain groups of species were evaluated together, such as Prevotella and Porphyromonas, Streptococcus milleri group species, and PCN-resistant strains. Although multiple cultures were taken in some cases, only the results of the initial culture taken at the onset of treatment were used in the statistical analyses. The complications recorded were: PTF, facial nerve decit, need for reoperation, emergency airway management, spread of infection into the chest or the brain, trigeminal nerve decit, and death.
DATA MANAGEMENT AND ANALYSES

Results
A total of 37 subjects (23 male, 14 female) from 14 to 76 years of age (mean 34.9 15.8) were enrolled in this study. There were 20 (54%) African-American/ black, 8 (22%) Hispanic, 8 (22%) Caucasian, and 1 (3%) Asian patient (Table 2). Three subjects (8%) were PCN-allergic, 3 (8%) had immunocompromising diseases (2 insulin-dependent diabetics and 1 HIV-seropositive individual with a CD4 count of 400 cells/L). There were 15 (41%) smokers, although this variable was not recorded for 4 subjects (11%). At the time of entry into this study, 20 (54%) subjects were taking various oral antibiotics, predominantly penicillins or other beta-lactam antibiotics, as detailed in Table 3. The most frequent dental disease leading to severe odontogenic infection was caries (65%), followed by pericoronitis (22%) and periodontal disease (22%) (Table 4). These percentages total more than 100% because multiple dental diseases were present in

Table 3. PREADMISSION VARIABLES

n (Cases) Penicillin allergy Immune system compromise Diabetes HIV seropositivity Smoking Not recorded Yes No Preadmission antibiotics No antibiotic Penicillin Clindamycin Penicillin cephalexin Erythromycin Penicillin clindamycin Penicillin metronidazole 3 2 1 4 15 18 17 10 5 2 1 1 1

% of Cases 8 5 3 11 41 49 46 27 14 5 3 3 3

Data were recorded prospectively on standardized collection forms. A database was constructed using Microsoft Excel (Microsoft, Redmond, WA) and imported into SPSS 12.0 (SPSS, Inc, Chicago, IL) for statistical analysis. Descriptive statistics were computed for all of the study variables.

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

FLYNN ET AL

1097

Table 4. PREOPERATIVE CLINICAL VARIABLES

n (Cases) Dental etiology Caries Pericoronitis Periodontitis Needle track infection (after dental procedures) Postoperative infection (third molar exodontia) Teeth involved Lower third molars Other lower posteriors Upper third molars Other upper posteriors Upper anteriors Lower anteriors Dyspnea Dysphagia Trismus (MIO 20 mm) White blood cell count on admission ( 103) Admission core temperature (F) Number of teeth involved 24 8 8 2 1 25 18 3 3 0 0 5 29 27

% of Cases 65 22 22 5 3 68 49 8 8 0 0 14 78 73

Mean SD

Range

14.9 4.2 101.3 1.3 1.5 0.9

5.926.0 98.0104.4 15

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

some patients. The most frequently involved tooth was the lower third molar (68%), followed by other lower posterior teeth (bicuspids and rst and second molars). Anterior teeth gave rise to no infection severe enough to warrant hospitalization in this study. Table 4 lists the distribution of causative teeth, dental etiologies, and other preoperative clinical variables. On admission, 5 subjects (14%) reported dyspnea, 29 (78%) complained of dysphagia, and 27 (73%) had trismus (MIO 20 mm). Maximum interincisal opening (MIO) was not recorded in 3 subjects. The initial core temperature ranged from 98.0F to 104.4F, with a mean SD of 101.3 1.3. The initial mean SD WBC was 14.9 4.2, with a range of 5.9 to 26.0 103/uL (Table 4). Subjects presented with a history of 8.2 14.6 days of preoperative swelling (range, 1 to 71). After admission to the hospital, which was dened as leaving the emergency room or clinic for the hospital oor or operating room, surgery was performed 5.1 7.5

hours later, with a range of 0.2 to 23.3 hours. The duration of surgery, dened as the time between entry into and leaving the operating suite, was 2.1 0.7 hours, with a range of 0.9 to 3.8 (Table 5). The largest number of cases, 15, occurred during the summer (Table 5). A variable number and combination of fascial spaces were infected (Table 6). The mean SD number of infected spaces per case was 3.3 1.5 (range, 1 to 8). The admission SS was 5.9 3.1 (range, 1 to 16). The most commonly infected space was the pterygomandibular (22 cases, 60%), followed closely by the submandibular (20 cases, 54%) and lateral pharyngeal (16 cases, 43%). At surgery, 3.1 1.8 spaces were drained (range, 1 to 8). Twenty-eight cases (76%) yielded pus at I&D, and pus was not encountered in 9 cases (24%). LOS was 5.1 3.0 days, with a range of 1 to 14 days. Airway management was by endotracheal intubation in all but 3 cases, 18 (49%) by beroptic laryn-

Table 5. TIME-RELATED VARIABLES

Mean SD Days of preoperative swelling Time between admission and OR (h) Duration of surgery (h) Season of occurrence Summer Spring Autumn Winter 8.2 14.6 5.1 7.5 2.1 0.7

Range 171 0.223.3 0.93.8

n (Cases)

% of Cases

15 8 7 7

41 22 19 19

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

1098

SEVERE ODONTOGENIC INFECTIONS

Table 6. ANATOMIC VARIABLES

Mean SD Number of spaces affected Severity score Spaces affected Pterygomandibular Submandibular Lateral pharyngeal Buccal Space of body of mandible Submasseteric Deep temporal (including infratemporal) Sublingual Submental Supercial temporal Infraorbital Retropharyngeal Infratemporal Maxillary sinus Parotid Groups of spaces affected Masticator Perimandibular (submandibular, sublingual, and submental) Space 3 (lateral pharyngeal, retropharyngeal, and pretracheal) 3.3 1.5 6.0 3.1

Range 18 116

n (Cases)

% of Cases

22 20 16 15 13 9 6 6 4 3 2 2 1 1 1 29 22 16

59 54 43 41 35 24 16 16 11 8 5 5 3 3 3 78 60 43

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

goscopy, and 16 (43%) by direct laryngoscopy. One case required urgent tracheotomy under local anesthesia at reoperation for extension of infection with dyspnea (Table 7). Topical anesthetics and mild sedation were most commonly used for analgesia during beroptic intubation, and general anesthesia was most commonly used for direct laryngoscopy. Intravenous PCN was used initially in 33 cases. Because of PCN allergy, intravenous clindamycin (900

mg every 8 hours) was used initially in 3 cases. One patient was admitted with necrotizing fasciitis, and intravenous gentamicin, clindamycin, and metronidazole were used for broad-spectrum antibiotic coverage, as shown in Table 7. Bacterial species identication and PCN and clindamycin sensitivity were performed in the last 24 consecutive cases enrolled in this study. Cultures yielded no growth in 2 of those cases (8%). The

Table 7. TREATMENT VARIABLES

n (Cases) Stage of infection Cellulitis (no pus at I&D) Abscess (pus at I&D) Antibiotics used Penicillin Clindamycin Gentamicin Metronidazole Clinda Airway management techniques Fiberoptic intubation Direct laryngoscopic intubation No intubation Tracheotomy (at reoperation) Number of spaces drained Length of hospital stay (days) 9 28 33 3 1 18 16 3 1

% of Cases 24 76 89 8 3 49 43 8 3

Mean SD

Range

3.1 1.8 5.1 3.0

18 114

Abbreviations: I&D, incision and drainage; Clinda, clindamycin. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

FLYNN ET AL

1099 has risen steadily in 4 recent studies, from 33% of cases in 1991 to 55% in 1995, and 54% of cases in the current study.12-14 In the current study, there were 10 cases receiving PCN with culture and sensitivity testing in which PCN-resistant organisms were later identied. PCN failed in 6 (60%) and was successful in 4 (40%) of those cases. If we can predict an increasing rate of PCN resistance among the pathogens of OI, then the rate of PTF in severe OI can also be expected to increase. It should be noted that in 2000, Kuriyama et al15 found an increased rate of resistance to beta-lactam antibiotics in subjects with OI who had received such antibiotics prior to sampling. They recommended beta-lactamase stable antibiotics in patients with unresolved infections that have previously received betalactam antibiotics. These data were not available during the patient treatment period of this study, which ended in 1999. This study provides clinical evidence to support the laboratory ndings of increased resistance among bacteria cultured from OI.12-15 Although clindamycin has recently been recommended for widespread use in dentistry,16 multiple clinical studies that have compared PCN and clindamycin found success rates with PCN17,18 or ampicillin19 at 97% of cases or higher. Only the study of Kannangara et al in 198020 found a clinical PCN success rate of 80% versus clindamycin at 100%. All of the PCN failures in Kannangaras study were in mandibular fractures that harbored Bacteroides fragilis, not in OI.20 In this study of patients with OI severe enough to warrant hospitalization, penicillins clinical success rate was 79%. A comparison with clindamycin was not part of the study design. It is impossible to compare the complication rate in this study with that of other published reports because of differences in study design, patient population, cause of infection, and the lack of a common method of calibrating severity. Such calibration may be possible in the future by using C-reactive protein, WBC or SS, or a combination thereof.1,3,21 As in other reports,7 the most frequently identied causative tooth was the lower third molar in 25 (68%) cases, followed by other lower posterior teeth in 18 (49%) cases. Upper posterior teeth were involved much less frequently, and no anterior teeth were identied as causing severe OI in this study. Pericoronitis was not diagnosed as frequently in this study as in others.7-9 In this study, the most frequently infected deep fascial spaces were, in descending order, pterygomandibular (n 22 patients), submandibular (n 20), lateral pharyngeal (n 16), and buccal (n 15). If infections involving any portion of the masticator

various species isolated, the number of strains and percent of cases in which they were identied, and the PCN resistance rate for each species are listed in Table 8. A total of 90 strains were isolated, with a mean SD of 3.8 2.0 species per case (range, 0 to 8). Two cases (8%) yielded aerobic bacteria only, 4 cases (17%) yielded anaerobes only, and 16 cases (67%) yielded a mixed ora (Table 9). Thus, anaerobes were present in 20 of 24 (83%) of cases with culture data. Seventeen (19%) of the 90 isolated strains were PCN-resistant; 1 or more PCN-resistant organisms were found in 13 (54%) of the 24 cases with antibiotic sensitivity data. Four clindamycin-resistant strains were identied, one each of Streptococcus milleri, Eikenella corrodens, and Streptococcus mitis, and one strain of Klebsiella pneumoniae that was also resistant to PCN. Clindamycin-resistant strains were identied in 4 (17%) cases with sensitivity data. Complications, including penicillin therapeutic failure (PTF), facial nerve decit, reoperation, and death are listed in Table 10. The criteria for PTF were met in 7 (21%) of the 33 cases that received PCN. Only 6 complications other than PTF occurred in this study: 3 facial nerve decits that were improving at last follow-up and 3 reoperations. Extension of infection into the mediastinum or cranial cavity did not occur. No deaths occurred. A second operation was required in 3 cases (8%): minor intraoral I&D (n 2) and tracheotomy and drainage of right and left lateral pharyngeal and retropharyngeal spaces (n 1). In the latter case, the subject developed dysphagia, dyspnea, and increasing swelling 5 days after initial drainage of the right pterygomandibular space. Repeat CT scan showed extension of the infection into both lateral pharyngeal spaces and the retropharyngeal space (Fig 1). After tracheotomy under local anesthesia, all of the infected spaces were drained and the antibiotic was changed from PCN to clindamycin. Rapid improvement ensued. PCN-resistant organisms were later identied in this case. Table 11 lists each of the 37 cases and clinically important data for each case.

Discussion
In this study we prospectively evaluated 37 consecutive patients with severe OI and managed them with a standardized protocol of high-dose intravenous PCN (unless the patient was PCN allergic) and early incision and drainage. We encountered a PTF rate of 21% in severe OI requiring hospitalization. Such a failure rate is clinically unacceptable. The isolation of 1 or more PCN-resistant strains in OI

1100

SEVERE ODONTOGENIC INFECTIONS

Table 8. DETAILED MICROBIOLOGIC DATA (N 24)

# of Strains No growth (2 cases) Aerobes (including facultative species) All S. milleri group species (as detailed below) S. constellatus S. intermedius S. anginosus S. milleri S. milleri I S. milleri II All other S. viridans species (excluding S. milleri) S. mitis* S. sanguis Other streptococci (as detailed below) -hemolytic streptococcus Group F -hemolytic streptococcus Group C -hemolytic streptococcus not Group ABCDFG Other aerobic/facultative species (as detailed below) Gemella morbillorum S. acidominimus Corynebacterium spp. Lactobacillus acidophilus Staphylococcus coagulase negative Staphylococcus epidermidis Klebsiella pneumoniae* Enterococcus spp. Anaerobes All Prevotella and Porphyromonas species (as detailed below) P. oralis P. oris P. buccae P. intermedia P. melaninogenica P. denticola P. gingivalis P. loeschii F. nucleatum All Peptostreptococcus species (as detailed below) P. micros P. prevotii Other anaerobic species (as detailed below) Veillonella spp. Wolinella spp. Capnocytophaga spp. Actinomyces odontolyticus Actinomyces israelii Bacillus gracilis Bacillus spp. Bacteroides fragilis Bacteroides ureolyticus Propionibacterium acnes Hemophilus inuenzae Bidobacterium spp. Clostridium spp. Eikenella corrodens* None 12 3 2 0 4 1 2 2 1 1 3 1 1 1 12 1 1 2 1 4 1 1 1 23 2 1 10 6 1 1 1 1 5 12 11 1 21 1 1 2 1 1 4 2 1 1 3 1 1 1 1

% of Cases 8 50 13 8 0 17 4 8 8 4 4 13 4 4 4 50 4 4 8 4 17 4 4 4 63 8 4 42 25 4 4 4 4 21 50 46 4 88 4 4 8 4 4 17 9 4 4 13 4 13 4 4

% of Strains PCN-Resistant

0 0 0 0 0 0 0 0 0 33 0 100 0 58 0 100 100 100 100 35 0 100 30 67 0 0 0 0 25 0 0 0 0 0 0 0 0 0 0 0 0 0 0

*Clindamycin-resistant strain. Not all strains were tested for antibiotic sensitivity. Blank cells none of the strains were tested for antibiotic sensitivity. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

FLYNN ET AL

1101 ability of and facility in beroptic intubation techniques among anesthesiologists, and they are consistent with recent results found in the otolaryngology literature.23 The most frequent pathogens isolated in the 24 cases for which complete species identication was performed were: Prevotellae, Viridans streptococci (including the S. milleri group), and Peptostreptococci. These results are consistent with the studies of Heimdahl et al,24 Lewis et al,25 Sakamoto et al,26 and others.12,15,27,28 The design of this study did not include a prospective evaluation of the utility of computed tomography (CT) in the pre- and postoperative evaluation of severe OI. This may have led to misclassication of the anatomic location of infection in some cases, which is a limitation of this study. The accuracy of abscess detection in head and neck infections is improved by the combination of clinical examination and contrastenhanced CT.29 During the course of the study, however, we observed several advantages of CT. A preoperative CT scan can be useful in identifying airway displacement or effacement and in visually demonstrating the airway problem to the anesthesiologist before the airway management plan is formulated. This has led to a routine policy of passing the beroptic cable and endotracheal tube through the naris on the side opposite the infection, toward which the airway has been deviated. This maneuver has led to easier and more rapid beroptic intubation. In addition, postoperative CT was very useful in identifying correct drain placement as well as undrained loculations of pus or extension of infection during treatment. In descending necrotizing mediastinitis, Freeman et al30 reported a reduced mortality of 0 in 10 cases, using a protocol of open thoracotomy for direct mediastinal drainage and postoperative CT taken every 48 to 72 hours in patients with lack of clear improvement after surgery. They reported using ranges of 3 to 15 CT scans per case and 4 to 8 operations per case. Laparotomy for extension of infection into the abdominal cavity was necessary in 30% of cases, and CT was most useful in identifying the need for laparotomy.30 Recently, CT has been used to trace the anatomic pathways of the spread of infection arising from maxillary and mandibular teeth.31-33 The results of this study indicate that PCN resistance, resulting in PCN therapeutic failure, was unacceptably high in this group of patients. Alternative antibiotics, such as clindamycin, should be considered in hospitalized patients with OI. Anatomic involvement of the masticator space, resulting in trismus, was diagnosed much more frequently in this study than has been previously reported. Trismus and dysphagia on presenta-

Table 9. SUMMARY MICROBIOLOGIC DATA (N 24)

Culture and Sensitivity Results No growth Oxygen requirements Aerobes only Anaerobes only Mixed aerobes and anaerobes Antibiotic resistance Penicillin Clindamycin

# of Cases 2 2 4 16 13 4

% of Cases 8 8 17 67 54 17

Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

space (submasseteric, pterygomandibular, supercial or deep temporal) are considered, this space represented 78% of cases (n 29). Masticator space involvement was identied much more frequently in this study than in other anatomic studies of severe OI, where the submandibular space was the most frequently reported location.4-6,8,9 Dysphagia and trismus were presenting symptoms in a high proportion of subjects, at 29 (78%) and 27 (73%) cases, respectively. It appears from this study, therefore, that the most common manifestation of severe OI may be infection involving the masticator and/or submandibular spaces due to pericoronitis or caries of a mandibular posterior tooth, most commonly the lower third molar. Trismus and dysphagia were the most frequent presenting sign and symptom, and these should therefore be considered highly suggestive of severe OI. The incidence of trismus and dysphagia were not reported in previous studies of OI.1-9,22 The most frequently used airway management technique in this study was beroptic intubation under light sedation in 18 (49%) cases, followed by direct laryngoscopic intubation under general anesthesia in 16 (43%) cases. Airway management technique was selected by the anesthesiologist in consultation with the oral and maxillofacial surgeon. These data reect the perceived increasing avail-

Table 10. COMPLICATIONS (N 37)

# of Cases Penicillin therapeutic failure (n 33)* Facial nerve decit Need for reoperation Death 7 3 3 0

% of Cases 21 8 8 0

*33 of 37 subjects received penicillin; 3 subjects received clindamycin because of penicillin allergy and 1 received clindamycin, gentamicin, and metronidazole because of necrotizing fasciitis. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

1102

SEVERE ODONTOGENIC INFECTIONS

Table 11. SELECTED DATA FOR ALL CASES

No. of No. of Pus PCN Case Age WBC Severity Infected Infected OR Time at Resistant LOS no. Race Gender (years) (103) Score Spaces Teeth (minutes) I&D PTF Strains (days) Reoperation 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 B W B A W H B B W B B H B B W B B B W B W H H B H B B W B B B H B B H W H F M M F M F M M M M F M F F F M M F F M M M F F M M M M F M F M M F M M M 22 22 23 45 26 43 21 33 76 54 27 46 21 45 60 14 41 23 22 42 31 45 29 19 25 24 39 28 31 23 75 36 29 21 28 71 30 12.6 11.6 13.0 13.7 13.8 15.6 15.4 14.2 15.3 13.0 17.0 11.7 26.0 6.7 14.7 12.1 15.1 21.6 13.6 16.1 12.4 16.9 10.4 16.2 18.2 22.1 24.9 13.8 16.8 12.7 5.9 11.3 14.8 15.3 19.8 12.6 13.0 7 4 3 4 3 3 7 5 5 3 13 2 8 5 6 2 8 4 6 9 7 16 6 5 12 5 9 5 6 8 1 5 5 6 9 3 5 4 2 2 2 2 3 4 3 4 3 6 2 4 2 3 1 4 5 3 4 2 8 6 2 6 4 5 2 4 4 1 3 2 3 3 2 2 2 1 1 1 2 3 1 2 1 2 1 3 2 1 3 1 2 1 3 1 1 5 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 99 145 82 94 116 102 70 75 55 60 200 113 154 110 90 115 175 100 120 225 115 182 180 135 190 110 157 185 130 110 80 160 90 120 175 106 112 Y N Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y N N N Y Y N Y Y Y Y Y Y N Y N N N Y Y Y N N N N N N N N C N Y N N Y Y C N Y Y Y C NF N N N N N N N N N N N N N Y N 3 4 3 3 3 6 3 4 4 4 9 4 7 5 6 3 6 14 7 9 2 14 7 4 6 3 10 4 3 3 3 3 4 4 4 5 1 N N N N N Y N N N N N N N N N N N Y N N N N N N N N Y N N N N N N N N N N

Y Y Y N Y Y Y N Y N N N N N Y Y N N N N Y N Y Y

Abbreviations: A, Asian; B, African-American/black; H, Hispanic; W, white; F, female; M, male; WBC, white blood cell count on admission; OR, operating room; I&D, incision and drainage; PTF, penicillin therapeutic failure; PCN, penicillin; LOS, length of hospital stay; Y, yes; N, no; C, subject received clindamycin; NF, subject received multiple antibiotics for necrotizing fasciitis. Flynn et al. Severe Odontogenic Infections. J Oral Maxillofac Surg 2006.

tion should be appreciated as signicant indicators of severe OI. Directions for future research include: biochemical and clinical predictors of outcomes in severe OI, and molecular methods for bacterial identication and antibiotic sensitivity testing.34 Acknowledgments
The authors wish to acknowledge Mauricio Wiltz, DDS, and all of the Oral and Maxillofacial Surgery residents at the Monteore Medical Center for their assistance in the care of patients and gathering of data. This study was supported in part by the Monteore Medical Center Department of Dentistry and the Massachusetts General Hospital Department of Oral and Maxillofacial Surgery Education and Research Fund.

References
1. Dodson TB, Barton JA, Kaban LB: Predictors of outcome in children hospitalized with maxillofacial infections: A linear logistic model. J Oral Maxillofac Surg 49:838, 1991 2. Biederman GR, Dodson TB: Epidemiologic review of facial infections in hospitalized pediatric patients. J Oral Maxillofac Surg 52:1042, 1994 3. Peters ES, Fong B, Wormuth DW, et al: Risk factors affecting hospital length of stay in patients with odontogenic infections. J Oral Maxillofac Surg 54:1386, 1996 4. Huang TT, Liu TC, Chen PR, et al: Deep neck infection: Analysis of 185 cases. Head Neck 26:854, 2004 5. Chen MK, Wen YS, Chang CC, et al: Predisposing factors of life-threatening deep neck infection: Logistic regression analysis of 214 cases. J Otolaryngol 27:141, 1998

FLYNN ET AL
6. Krishnan V, Johnson JV, Helfrick JF: Management of maxillofacial infections: A review of 50 cases. J Oral Maxillofac Surg 51: 868, 1993 7. Indresano AT, Haug RH, Hoffman MJ: The third molar as a cause of deep space infections. J Oral Maxillofac Surg 50:33, 1992 8. Haug RH, Hoffman MJ, Indresano AT: An epidemiologic and anatomic survey of odontogenic infections. J Oral Maxillofac Surg 49:976, 1991 9. Storoe W, Haug RH, Lillich TT: The changing face of odontogenic infections. J Oral Maxillofac Surg 59:739, 2001 10. Flynn TR: Anatomy and surgery of deep fascial space infections, in Kelly JJ: Oral and Maxillofacial Surgery Knowledge Update 1994. Rosemont, IL, American Association of Oral and Maxillofacial Surgeons, 1994, pp 79-107 11. Flynn TR: Anatomy of oral and maxillofacial infections, in Topazian RG, Goldberg MH, Hupp JR: Oral and Maxillofacial Infections (ed 4). Philadelphia, PA, WB Saunders, 2002, pp 188-213 12. Brook I, Frazier EH, Gher ME: Aerobic and anaerobic microbiology of periapical abscess. Oral Microbiol Immunol 6:23, 1991 13. Lewis MAO, Parkhurst CL, Douglas CWI, et al: Prevalence of penicillin resistant bacteria in acute suppurative oral infection. J Antimicrob Chemother 35:785, 1995 14. von Konow L, Kondell PA, Nord CE, et al: Clindamycin versus phenoxymethylpenicillin in the treatment of acute oral infections. Eur J Clin Microbiol Infect Dis 11:1129, 1992 15. Kuriyama T, Nakagawa K, Karasawa T, et al: Past administration of beta-lactam antibiotics and increase in the emergence of beta-lactamase-producing bacteria in patients with orofacial odontogenic infections. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 89:186, 2000 16. Brook I, Lewis MAO, Sandor GKB, et al: Clindamycin in dentistry: More than just effective prophylaxis for endocarditis? Oral Surg Oral Med Oral Pathol Oral Radiol Endod 100:550, 2005 17. von Konow L, Kondell PA, Nord CE: Clindamycin versus phenoxymethylpenicillin in the treatment of acute orofacial infections. Eur J Clin Microbiol Infect Dis 11:1129, 1992 18. Gilmore WC, Jacobus NV, Gorbach SL, et al: A prospective double-blind evaluation of penicillin versus clindamycin in the treatment of odontogenic infections. J Oral Maxillofac Surg 46:1065, 1988 19. Mangundjaja S, Hardjawinata K: Clindamycin vs. ampicillin in the treatment of odontogenic infections. Clin Ther 12:242, 1990

1103
20. Kannangara DW, Thadepalli H, McQuirter JL: Bacteriology and treatment of dental infections. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 50:103, 1980 21. Ylijoki S, Suuronen R, Jousimies-Somer H, et al: Differences between patients with or without the need for intensive care due to severe odontogenic infections. J Oral Maxillofac Surg 59:867, 2001 22. Chow AW, Roser SM, Brady FA: Orofacial odontogenic infections. Ann Intern Med 88:392, 1978 23. Shockley WW: Ludwig angina: A review of current airway management. Arch Otolaryngol Head Neck Surg 125:600, 1999 24. Heimdahl A, Von Konow L, Satoh T, et al: Clinical appearance of orofacial infections of odontogenic origin in relation to microbiological ndings. J Clin Microbiol 22:299, 1985 25. Lewis MAO, MacFarlane TW, McGowan DA: A microbiological and clinical review of the acute dentoalveolar abscess. Br J Oral Maxillofac Surg 28:359, 1990 26. Sakamoto H, Kato H, Sato T, et al: Semiquantitative bacteriology of closed odontogenic abscesses. Bull Tokyo Dent Coll 39:103, 1998 27. 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 90:600, 2000 28. Stefanopoulos PK, Kolokotronis AE: The clinical signicance of anaerobic bacteria in acute orofacial odontogenic infections. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 98:398, 2004 29. Miller WD, Furst IM, Sandor GKB, et al: A prospective blinded comparison of clinical examination and computed tomography in deep neck infections. Laryngoscope 109:1873, 1999 30. Freeman RK, Valliere E, Verrier ED, et al: Descending necrotizing mediastinitis: An analysis of the effects of serial surgical debridement on patient mortality. J Thorac Cardiovasc Surg 119:260, 2000 31. Obayashi N, Ariji Y, Goto M, et al: Spread of odontogenic infection originating in the maxillary teeth: Computerized tomographic assessment. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 98:223, 2004 32. Oshima A, Ariji Y, Goto M, et al: Anatomical considerations for the spread of odontogenic infection originating from the pericoronitis of impacted mandibular third molar: Computed tomographic analyses. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 98:589, 2004 33. Yonetsu K, Izumi M, Nakamura T: Deep facial infections of odontogenic origin: CT assessment of pathways of space involvement. AJNR Am J Neuroradiol 23:468, 2002 34. Flynn TR, Stokes LN, Lee AM, et al: Molecular microbiology of orofacial infections. J Oral Maxillofac Surg 60:72, 2002 (suppl 1)