Available online at www.sciencedirect.com

American Journal of Otolaryngology–Head and N

eck Medicine and Surgery 33 (2012) 56–63www.elsevier.com/locate/amjoto

Deep neck space infections: a retrospective review of 173 cases☆

Salih Bakir, MDa,⁎, M. Halis Tanriverdi, MDb, Ramazan Gün, MDa,A. Ediz Yorgancilar, MDa, Müzeyyen Yildirim, MDa, Güven Tekbaş, MDc,

Yılmaz Palanci, MDd, Kaan Meriç, MDe, İsmail Topçu, MDa

aDepartment of Ear Nose & Throat, Dicle University School of Medicine, Diyarbakir, TurkeybDepartment of Family Medicine, Dicle University School of Medicine, Diyarbakir, Turkey

cDepartment of Radiology, Dicle University School of Medicine, Diyarbakir, TurkeydDepartment of Public Health, Dicle University School of Medicine, Diyarbakir, Turkey

eDepartment of Radiology, Diyarbakir State Hospital, Diyarbakir, Turkey

Received 25 October 2010

Abstract Purpose: The purpose of this study is to review our recent experience with deep neck infections and

☆ Declaration of iauthors alone are resp

⁎ Correspondingof Medicine, 21280, D

E-mail address: d

0196-0709/$ – see frodoi:10.1016/j.amjoto.2

emphasize the importance of radiologic evaluation and appropriate treatment selection in those patients.Materials and Methods: The records of 173 patients treated for deep neck infection at theDepartment of Otolaryngology and Head and Neck Surgery of Dicle University Hospital during theperiod from 2003 to 2010 were retrospectively reviewed. Their demography, symptoms, etiology,seasonal distribution, bacteriology, radiology, site of deep neck infection, durations of the hospitaladmission and hospital stay, treatment, complications, and outcomes were evaluated. The findingswere compared to those in the available literature.Results: Dental infection was the most common cause of deep neck infection (48.6%). Peritonsillarinfections (19.7%) and tuberculosis (6.9%) were the other most common cause. Pain, odynophagia,dysphagia, and fever were the most common presenting symptoms. Radiologic evaluation wasperformed on almost all of the patients (98.3%) to identify the location, extent, and character (cellulitisor abscesses) of the infections. Computed tomography was performed in 85.3% of patients. The mostcommon involved site was the submandibular space (26.1%). In 29.5% of cases, the infection involvedmore than one space. All the patients were taken to intravenous antibiotic therapy. Surgical inter-vention was required in 95 patients (59.5%), whereas 78 patients (40.5%) were treated with intra-venous antibiotic therapy alone. Life-threatening complications were developed in 13.8% of cases;170 patients (98.3%) were discharged in stable condition.Conclusion: Despite the wide use of antibiotics, deep neck space infections are commonly seen.Today, complications of deep neck infections are often life threatening. Although surgical drainageremains the main method of treating deep neck abscesses, conservative medical treatment areeffective in selective cases.

© 2012 Elsevier Inc. All rights reserved.

1. Introduction

Deep neck space infection (DNI) means infection in thepotential spaces and fascial planes of the neck, either with

nterest: The authors report no conflicts of interest. Theonsible for the content and writing of the manuscript.author. Department of ENT, Dicle University Schooliyarbakir, Turkey. Tel.: +90 412 2488001/4492.rsalihbakir@hotmail.com (S. Bakir).

nt matter © 2012 Elsevier Inc. All rights reserved.011.01.003

abscess formation or cellulitis [1]. Despite the prevalenceand the complications incidence of DNI has been diminishedwith improved diagnostic techniques and widespread avail-ability of antimicrobial therapy, these infections are stillserious and potentially life threatening today as in the past.The DNIs may arise from several focuses in the head andneck, including teeth, adenotonsillar tissue, and salivaryglands [1-3]. The origin of DNI is different in many publi-cations. In the preantibiotic era, most of DNIs arose fromtonsillitis or pharyngitis [2,3]. Today, dental infections are

Fig. 1. Distribution of age (n = 173).

57S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

the most common causes of DNI [4-6]. The DNIs aregenerally polymicrobial. Streptococci, Peptostreptococcusspp, Staphylococcus aureus and anaerobes are the organismsmost commonly cultured from deep neck abscesses [3,4].The main complications include respiratory obstruction,mediastinitis, pleural empyema, pericarditis, jugular veinthrombosis, and septic shock [4,5]. Complications can evenresult in death [4]. The advent of modern imaging techniqueshas made it possible to diagnose these complications earlierand to localize them exactly [5]. Management of deep neckinfections has usually been based on prompt surgicaldrainage of purulent abscesses through an external approachor nonsurgical treatment with on the basis of appropriateantibiotics [7]. The purpose of this study is to review ourrecent experience with DNI and emphasize the importance ofradiologic evaluation and appropriate treatment selection inthose patients.

2. Materials and methods

In this study, the records of 173 patients treated for DNIat the Department of Otolaryngology and Head and NeckSurgery of Dicle University Hospital between January 2003and August 2010 were retrospectively reviewed. Theirdemography; symptoms; etiology; seasonal distribution; bac-teriology; radiology; site of deep neck infection; durations ofthe hospital admission; and hospital stay, treatment, complica-tions, and outcomes were evaluated. The findings were com-pared to those in the available literature.

3. Results

There were 80 (46.2%) male and 93 (53.8%) femalepatients, with a female-to-male ratio of 1.16/1. The mean agewas 25.1 years (±15.5) (range, 3–69 years) (Fig. 1). Theduration of admission ranged from 2 to 33 days with anaverage of 6.6 ± 4.7 days.

The seasonal distribution of patients presenting with deepneck infections: autumn (43.4%), summer (24.3%), spring(16.8%), and winter (15.6%) (Fig. 2, Table 1).

Pain was present in almost all cases. After pain, the othercommon complaint was neck swelling (66%), odynophagia(48%), dysphagia (44%), fever (35%), dysphonia (28%),trismus (27%), otalgia (13%), dyspnea (12%), and drainingfistulas in the neck (2%).

Physical examination revealed that 77 patients (45%) hadfever (N37.5°C). The white blood cell (WBC) count washigher than 10 000 cells/mm3 (cells per cubic millimeter) in98 cases (56%). In addition, 57 patients (33%) had a WBCcount of more than 15 000 cells/mm3, and 23 patients (15%)had a WBC count of more than 20 000 cells/mm3.

Considering clinical and radiological evidence, the causesof deep neck infections were identified in 144 patients(83.2%). The most common cause of deep neck infection

was odontogenic (84 cases, 48.6%). Odontogenic causeswere diagnosed through dental consultations. Orthopanto-grams of the mandible were obtained in 26 cases. The secondmost common cause of deep neck infection was peritonsillarabscess (34 cases, 19.7%). In 12 patients, the abscess wascaused by tuberculosis (6.9%), and in 10 patients, an infectedsalivary glands were found (5.8%). In 3 patients, abscesswas caused by branchial cleft cyst (1.7%), and in 1 patient,caused by thyroiditis (0.6%). In the remaining 29 patients(16.8%), the origin of the DNI remained unclear. The etiol-ogy of deep neck infections is recorded in Fig. 3.

There were 4 patients (2.3%) with diabetes mellitus (DM)in our study. There was no case of known liver, lung, kidneydisease or malignancies, trauma, intravenous drug abuse,or immunodeficiency.

The results of bacterial cultures were available for 34of the 96 cases who underwent surgical treatment orneedle aspiration (35.4%). Anaerobic and aerobic cultureswere obtained. Anaerobes account for 5 (14.7%) of thepositive cultures. The cultures of 20 patients (58.8%)were polymicrobial. The most common bacteries wereanaerobic Peptostreptococcus (21.3%), and Staphylococcusepidermidis (19.7%).

Radiologic evaluation was performed almost all of thepatients to identify the location, extention, and character(cellulitis or abscesses) of the infections (170 patients, 98.3%).In 28 patients (16.5%), ultrasonography was the only imagingprocedure. For 145 DNI patients, computed tomography (CT)was performed (85.3%), and in 14 of those, an additionalmagnetic resonance imaging also was performed. Neck ultra-sonography and magnetic resonance imaging of the neckwere performed less relatively to the CT. In 26 patients,orthopantograms of the mandible were indicated.

According to clinical, surgical and imaging findings, 122(70.5%) had one involved space. The most common oneinvolved site was the submandibular space (26.1%), followedby the peritonsillar space (14.5%), the parapharyngeal space(11.6%), the submental space (10.4%), the retropharyngealspace (3.5%), the parotid space (2.9%), the carotid space(0.5%), the masseter space (0.5%), and the anterior visceralspace (0.5%). In 51 patients (29.5%), the infection involved

Fig. 2. The seasonal distribution of DNI patients (n = 173).

58 S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

more than one space. If 2 or more spaces were concurrentlyinvolved in a significant way, they were classified as extendedspaces. The diagnostic criteria of Ludwig's angina are definedas the simultaneous involvement of the sublingual, submylo-hyoid, and submental spaces, either as cellulitis or abscesses.Twenty-seven patients (15.6%) were evaluated as Ludwig'sangina, and 24 patients (13.9%) were evaluated as extendedspaces. According to clinical and imaging findings, thedistribution of involved spaces and sites is recorded in Fig. 4.

In our department, all patients received antimicrobialtherapy after admission. The antibiotic regimen has to covermostly gram-positive aerobes and anaerobes implicated indeep neck infections, also considering the rising incidence ofpolymicrobic infections. Empirical intravenous antibiotics(β-lactamase-resistant β-lactam antibiotics, the third-genera-tion cephalosporin antibiotics, metronidazole, clindamycin)were administered before the culture results were available,then the antibiotics regimen was modified based on theculture and sensitivity results. We mostly chose ampicillin/sulbactam 1.5 g 4 times per day plus metronidazole 500 mg

Table 1The seasonal distribution of patients according to the origin of DNI

Origin Seasons Total

Spring Summer Autumn Winter

No. ofcases

% No. ofcases

% No. ofcases

% No. ofcases

%

Dental 15 17.9 24 28.6 31 36.9 14 16.7 84Tonsil 5 14.7 6 17.6 15 44.1 8 23.5 34Tuberculosis 0 .0 2 16.7 10 83.3 0 .0 12Unknown 7 24.1 4 13.8 14 48.3 4 13.8 29Salivary

gland1 10.0 5 50.0 4 40.0 0 .0 10

Thyroid 1 100.0 0 .0 0 .0 0 .0 1Neck cyst 0 .0 1 33.3 1 33.3 1 33.3 3Total 29 16.8 42 24.3 75 43.4 27 15.6 173

3 times per day. The second most preferred option wasceftriaxone 1 g 2 times per day plus metronidazole 500 mg3 times per day. Supportive medical treatments (analgesicsand antipyretics, intravenous fluids, mouthwashes, intrave-nous steroids) were performed when required. Consider-ing the clinical condition and imaging, 78 patients (40.5%)were treated with intravenous antibiotic therapy alone,whereas surgical intervention was required in 95 patients(59.5%). Distribution of our treatment approach is shown inFig. 5 and Table 2. The surgical procedures ranged from asimple drainage by a topical anesthesia to a wide incisionand drainage with a general anesthesia. In 11 DNI patientswith mediastinitis, we performed mediastinotomy. Patientswith abscesses caused by dental infection were referredto the Department of Oral and Maxillofacial Surgery forfurther treatment.

There were 24 patients (13.8%) who developed life-threatening complications. Descending mediastinitis (11patients, 6.3%) was the most frequently occurring compli-cation in our series. Eight patients (4.6%) had upper airwaydistress. Of them, 6 patients (3.4%) required temporarytracheotomy. The other rare but serious complications in ourseries were sepsis (3 patients, 1.7%) and thrombosis of theinternal jugular vein (2 patients, 1.1%). All these complica-tions were developed in patients with extended space ab-scesses and Ludwig's angina abscesses.

The mean duration of hospital stay was 8.9 ± 5.8days (range, 2–36 days) (Fig. 6). Crude mortality was1.7% (3 cases); 170 patients (98.3%) were discharged instable condition.

4. Discussion

The current study found that the most common cause ofDNI was dental infections (48.6%). Most reports indicated asignificant prevalence of DNI that were caused by dental

Fig. 3. Distribution of etiology (n = 173).

59S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

infections [2,4-6,8]. Parhiscar and Har-El [2] found 43% in2001, Bottin et al [5] found (42%) in 2003, Huang et al[4] found (42%) in 2004, Marioni et al [8] found 38.8% in2008, and Eftekharian et al [6] found 49% in 2009 that themost common cause of deep neck infection was odontogenicin DNI.

We found that the second most common cause of deepneck infection was tonsillar infections (19.7%) (Fig. 3).Before the widespread use of antibiotics, several studiesshowed that most DNI cases (70–80%) resulted from com-

Fig. 4. Distribution of involved s

plicated tonsillopharyngeal infections [7], whereas nowa-days, a decreased incidence (8–16%) in pharyngotonsillaronset was described [2,6]. In children, the most encounteredcauses still remain acute tonsillitis and pharyngitis [3].According to some studies, upper airway infections are stillthe most common cause of deep neck infections [8-12].

Causes of deep neck infections may differ in variousstudies. According to some recent reports, poor dentalhygiene and intravenous drug abuse have become the mostcommon causes of DNI in adults, followed by foreign body

paces and sites (n = 173).

Fig. 5. Distribution of our treatment approach (n = 173).

60 S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

ingestion and infections of unknown origin [13,14]. Unlikeother reports, the present study showed that, followed bydental and upper faryngeal infections, tuberculosis infectionwas the third common cause (6.9%) (Fig. 3). Today, tuber-culosis disease is a still common in our country and the otherdeveloping and underdeveloped countries especially east ofthe world. Every organ can be affected by tuberculosis.Tuberculous involvement of the neck is also possible.Because of deep neck abscess formation seen on CT in all,we performed surgical drainage and mass biopsy. Consid-ering the biopsy results, tuberculosis was suspected in 12patients and then confirmed by other investigations.Tuberculosis disease should be considered in the differentialdiagnosis in patients with DNI.

The age distribution showed that most of our patientswere young and middle aged. As shown in Fig. 1, 80.9% of

Table 2Distribution of our treatment approach according to the origin of DNI

Origin Treatment Total

Antimicrobialtherapy

Surgery andantimicrobial therapy

No. of cases % No. of cases %

Dental 57 67.9 27 32.1 84Tonsil 10 29.4 24 70.6 34Tuberculosis 5 41.7 7 58.3 12Unknown 19 65.5 10 34.5 29Salivary gland 3 30.0 7 70.0 10Thyroid 0 .0 1 100.0 1Neck cyst 1 33.3 2 66.7 3TOTAL 95 59.5 78 40.5 173

our patients were under the age of 40 and 95.4% of patientswere younger than 50 years. The mean age was 25.1 years.The prevalence of DNI is comparatively high in young andmiddle-aged adults in our series. Because dental and tonsilinfections constituted a large part of the etiology in DNI inour series. These infections are more frequent in youngerages and middle-aged adults [12].

In our study, there was a slight predominance of women,whereas other studies showed a male dominancy or anequal distribution [4-6]. This result is consistent with ourcountry conditions. According to the Saydam et al. [15]report, which was supported by the World Health Organi-zation, prevalence of caries was higher in women than inmen in our country.

Our report has not shown association with trauma, intra-venous drugs abuse, chemotherapeutic treatments, chronic

Fig. 6. Distribution of mean hospital stay (n = 173).

61S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

hepatitis, chronic pulmonary diseases, renal insufficiency,autoimmune disease, or HIV infection, which were describedin some other reports [4,5,16]. There were 4 patients withDM in our study. Huang et al [4] reported that DM is themost common risk factor among the systemic disease thathas been associated with the development of deep neckinfections and noted a major incidence (30.3%). However,our study showed a lower incidence when compared (2.3%).This situation can be explained by age differences. Huang etal. [4] reported that the mean age was 49.5 (±20.5) years, and52.4% of the patients were older than 50 years, whereas inour study, the percentage of patients younger than 50 yearswas 95.4, and the mean age was 25.1 years (±15.5).

Bottin et al [5] reported that there was a slight prepon-derance of cases presenting in the summer season (35%), butin our series, there was an apparent preponderance of casespresenting in the autumn season (43.4%) (Fig. 2, Table 1).

The most frequent symptoms of our patients werepain, neck swelling, odynophagia, and dysphagia, whichwere similar to other series [5-7]. Although the frequencyof dyspnea is not common relatively than the othersymptoms, the presence of dyspnea may be the sign ofserious complications.

Fever was present in 45%, and 56% had high white bloodcell count of over 10 000 cells/mm3. Neither fever norleukocytosis are constant findings in deep neck infections[6]. Widespread diffusion of empirical broadspectrum anti-biotic and anti-inflammatory treatments may cause maskedpresentations of deep neck infections without pain, fever, orleukocytosis [5].

The results of pus cultures from either surgery or needleaspiration were available in 35.4% of patients; 58.8% of thepositive cultures had polymicrobial growth. Many of ourpatients had received antibiotic therapy; before admission,our clinic might have yielded the significant rate of negativecultures. The limited number of positive cultures did notallow any conclusions. The positive cultures in our studywere much similar to the latest reports [5,6,17].

Developing imaging techniques have made the manage-ment of deep neck infections better. Contrast-enhanced CT(CCT) scan is highly sensitive (91%) and very useful toidentify the extent of the deep neck infections and distin-guish cellulitis from abscesses [3]. Contrast-enhanced CThelps to decide whether surgical intervention is indicated[18]. Those cases with radiologic findings of cellulitis aresupposed to have better prognosis and respond earlier tomedical treatment, whereas those with abscesses behavemore aggressively, have more complications, and may re-quire surgical treatment [7]. However, ultrasonographycannot always identify small or deep abscess and cannotprovide the specific anatomical information necessary forsurgical intervention [18]. Magnetic resonance imaging hassimilar prognostic value to CCT scanning, but it is moreexpensive and requires longer scanning time when comparedwith CCT, so that not commonly preferred in imaging deepneck infection [18]. Based on this evidence, we considered

CCT to be part of the routine investigation in DNI patients(85.3%; 145 patients). Based on the clinic and radiologicalfindings, in 70.5% of patients, the infection involved onespace, and in 29.5% of patients, the infection involved morethan one space (Fig. 4). The most commonly involved sitewas the submandibular space (26.1%), which has beenreported as the most commonly involved site in most paststudies [3,6,19,20].

According to our group and most investigators world-wide, management of deep neck abscess has usually beenbased on prompt surgical drainage of purulent abscessesthrough an external approach [1-6]. On the contrary, PlazaMayor et al. [7] suggested broad-spectrum intravenous anti-biotics and high-dosed oral or intravenous corticosteroids foralmost all patients with any DNI. In our series, 78 patients(40.5%) were treated successfully with only intravenousantibiotic therapy. In the remaining 95 patients (59.5%),surgical procedures were required (Fig. 5 and Table 2). It hasbeen stated that, in patients with cellulitis, the infection canbe controlled successfully with IV antibiotics alone in mostDNI cases [6,13,17]. If there is a small amount of abscessand no impending complications are noted, medical therapymay be sufficient [6,7,17]. In those cases, medical treatmentdid not seem to increase complication rates or mortality[6,7]. Medical treatment could also be considered in selectedcases [17].

Whenever a DNI patient is admitted, empirical antibiotictherapy should be administered before the culture results areavailable. Empirical antibiotic treatments must cover gram-positive and gram-negative aerobic and anaerobic pathogens[17]. We used usually penicillin or third-generation cepha-losporin plus metronidazole or clindamycin combinationdepending on the case severity, the most probable focus andthe existence of a previous treatment. This regimen coversmost gram-positive, anaerobic, as well as β-lactamase-producing bacteria [3]. In our clinic, the initial antibiotictherapy was penicillin and metronidazole usually. Penicillinshould be the drug of choice for aerobic bacteria. Wepreferred clindamycin for severe DNI, which providesadequate therapy against anaerobes that were resistant topenicilin [1]. The third-generation cephalosporins were usedinstead if poor clinical response was noted or when com-plications had developed [1]. If required, the antibioticswere modified depending on the result of culture andsensitivity reports.

In our patients, besides antibiotic therapy, supportivemedical treatment was required because of complaints.Odinophagia, dysphagia, and fever was common, soanalgesics and antipyretics, intravenous fluids for rehydra-tion, and mouthwashes (in dental and peritonsillar infection)were provided. Intravenous steroids (methylprednisolone,60 mg) were used (where possible) because of strong anti-inflammatory effects for a few days in patients showingimportant local edema and dyspnea.

Medical versus medical and surgical treatment was deter-mined by imaging and clinical progress. In patients with

62 S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

significant abscess formation seen on CT, early open surgicaldrainage is the most appropriate method of treating a deepneck infection [4]. In our department, the treatment of deepneck infection consists of using only antimicrobial therapyprimarily in the absence of abscess and presence of thecellulitis and surgical drainage required primarily in thepresence of the abscesses. Surgical exploration may alsorequired when there is airway compromise, clinical signs ofsepsis occur or if there is poor response to antimicrobialtherapy within the first 48 hours [6,13,17].

The main life-threatening complications include des-cending mediastinitis, respiratory obstruction, pleural effu-sion, pneumonia, pericarditis, jugular vein thrombosis,venous septic emboli, carotid artery rupture, hepatic failure,adult respiratory distress syndrome, septic shock, anddisseminated intravascular coagulopathy [4,5]. The mortal-ity rate may reach 40%, while these serious complicationsoccur [4]. Descending mediastinitis is one of the most life-threatening complications of DNI [14]. In our series, themost frequently occurring complication was mediastinitis(11 cases; 6.3%). In three patients with mediastinitis haddeveloped sepsis. Three of them died related to septicshock. The remaining 8 patients with mediastinitis recov-ered with effective intravenous antibiotics after surgicaldrainage. In reviewing those 11 mediastinitis cases, wefound the origin arose from dental infection. Surgicaltreatment is essential in patients with mediastinitis [21].Access by cervical incision has a lower risk thanthoracotomy and avoids pleural contamination, whereasthoracotomy is more invasive and associated with the riskof respiratory complications that may worsen the prognosis[21]. In patients with descending mediastinitis, Kinzer et al[21] performed collar mediastinotomy, whereas Wheatleyet al. [22] recommended transthoracic mediastinal drainageroutinely. In 11 DNI patients with mediastinitis, weperformed mediastinotomy.

Other complications in our series were airway distress andthrombosis of the internal jugular vein. It is worth em-phasizing that airway support is the priority in patients withdeep neck infections [17]. Temporary tracheotomy wasrequired for 6 patients. The tracheostomy rate was consid-erably low (3.4%).

Our series had a mean length of hospitalization of 8.9 ±5.8 days (range, 2–36 days), which was similar to otherseries [5,15,20]. According to our results; the hospitalizationtime of patients with dental origin was longer than the otherpatients (Fig. 6).

Odontogenic infections comprise one of the mostdangerous causes of DNI [21], because in our series, wehad 3 deaths and 3 of them were related to dental infection.The mortality rate in our study was 1.7%, which was close tothat of some previous reports [1,4,5,16]. In a recent study,Marioni et al [17] reported that none of their patients died ofdeep neck infection or its complications. Despite the decreas-ing in death rates, even now, complications of DNIs shouldnot be underestimated.

5. Conclusion

Despite the wide use of antibiotics, deep neck spaceinfections are commonly seen. Our results demonstrate thattuberculosis must be considered as possible causes of DNIs.Clinical evidence and early radiologic diagnosis withcontrast-enhanced CT provide valuable information indefining the origin, location and extension of neck infec-tions. Our treatment approach to deep neck infections(medical or medical plus surgical treatment) was determinedby clinical (presentation, complications, response to anti-biotics in the first 48 hours) and radiological evidence.Although surgical drainage remains the main method oftreating deep neck abscesses, conservative medical treatmentare effective in selective cases. The combination of appro-priate intravenous antibiotic therapy, drainage, and securingof airway are recognized cornerstones of treatment.

References

[1] Wang LF, Kuo WR, Tsai SM, et al. Characterizations of life-threatening deep cervical space infections: a review of one hundredninety-six cases. Am J Otolaryngol 2003;24:111-7.

[2] Parhiscar A, Har-El G. Deep neck abscess: a retrospective review of210 cases. Ann Otol Rhinol Laryngol 2001;110:1051-4.

[3] Ungkanont K, Yellon RF, Weissman JL, et al. Head and neck spaceinfections in infants and children. Otolaryngol Head Neck Surg 1995;112:375-82.

[4] Huang TT, Liu TC, Chen PR, et al. Deep neck infection: analysis of185 cases. Head Neck 2004;26:854-60.

[5] Bottin R, Marioni G, Rinaldi R, et al. Deep neck infection: a present-day complication. A retrospective review of 83 cases (1998-2001). EurArch Otorhinolaryngol 2003;260:576-9.

[6] Eftekharian A, Roozbahany NA, Vaezeafshar R, et al. Deep neckinfections: a retrospective review of 112 cases. Eur Arch Otorhinolar-yngol 2009;266:273-7.

[7] Plaza Mayor G, Martinez-San Millan J, Martinez-Vidal A. Isconservative treatment of deep neck space infections appropriate?Head Neck 2001;23:126-33.

[8] Marioni G, Rinaldi R, Staffieri C, et al. Deep neck infection with dentalorigin: analysis of 85 consecutive cases (2000-2006). Acta Otolaryngol2008;128:201-6.

[9] Çağlı S, Yüce İ, Güney E. Deep neck infections: results of 50 cases.Erciyes Med J 2006;28:211-5.

[10] Miman MC, Öncel S, Kalcıoğlu T. Derin boyun enfeksiyonlarınaklinik yaklaşım. KBB İhtisas Dergisi 2001;8:206-13.

[11] İynen İ, Şan İ, Bozkuş F. Deep neck infection: an analysis of 82patients. J Harran Univ Med Fac 2009;6:25-8.

[12] Boscolo-Rizzo P, Marchiori C, Montolli F, et al. Deep neck infections:a constant challenge. ORL J Otorhinolaryngol Relat Spec 2006;68:259-65.

[13] Ridder GJ, Technau-Ihling K, Sander A, et al. Spectrum andmanagement of deep neck space infections: an 8-year experience of234 cases. Otolaryngol Head Neck Surg 2005;133:709-14.

[14] Lee JK, Kim HD, Lim SC. Predisposing factors of complicated deepneck infection: an analysis of 158 cases. Yonsei Med J 2007;48:55-62.

[15] Saydam G, Oktay I, Moller I. Oral health status analysis. Ankara:Ankara: WHO European Region-Ministry of Health; 1990.

[16] Har-El G, Aroesty JH, Shaha A, et al. Changing trends in deep neckabscesses. Oral Surg Oral Med Oral Pathol 1994;77:446-50.

[17] Marioni G, Staffieri A, Parisi S, et al. Rational diagnostic andtherapeutic management of deep neck infections: analysis of 233consecutive cases. Ann Otol Rhinol Laryngol 2010;119:181-7.

63S. Bakir et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 33 (2012) 56–63

[18] Smith II JL, Hsu JM, Chang J. Predicting deep neck space abscessusing computed tomography. Am J Otolaryngol 2006;27:244-7.

[19] Agarwal AK, Sethi A, Sethi D, et al. Role of socioeconomic factors indeep neck abscess: A prospective study of 120 patients. Br J OralMaxillofac Surg 2007;45:553-5.

[20] Tosun F, Akkaya A, Sağlam M. A retrospective study in patients withdeep neck infections. T Klin J E N T 2001;1:134-40.

[21] Kinzer S, Pfeiffer J, Becker S, et al. Severe deep neck space infectionsand mediastinitis of odontogenic origin: clinical relevance and impli-cations for diagnosis and treatment. Acta Otolaryngol 2009;129:62-70.

[22] Wheatley MJ, Stirling MC, Kirsh MM, et al. Descending necrotizingmediastinitis: transcervical drainage is not enough. Ann Thorac Surg1990;49:780-4.