Journ al of the Korean Radiological Society 1995 : 33(6) : 853-860

Odontogenic Versus Nonodontogenic Deep Neck Space Infections: CT Manifestations1

Hyung-Jin Kim, M.D. , Eui Dong Park, M.D., Jung Hee Kim, M.D. ,

Jae Hyoung Kim, M.D. , Eui Gee Hwang, M.D.2, Sung Hoon Chung, M.D.

Purpose : The purpose of this study was to evaluate computed tomographic (CT) findings of deep neck space infection(DNSI) with particular attentionto the differences in the spaces involved and in complications between odontogenic and nonodontogenic groups.

Materials and Methods: Forty-four patients(21 odontogenic and 23 nono­dontogenic) were included in this study. Among odontogenic DNSls, 15 had the dental infection in the second or third mandibular molar. We compared the CT features betw een odontogenic and nonodontogenic DNSls with special empha­sis on the differences in the spaces involved and in the rate and type of com­plications.

Results: In all patients, CT clearly differentiated abscess from cellulitis. The most common spaces involved in 21 patients with odontogenic DNSI were the parapharyngeal(n=18}, the submandibular( n=18}, the anterior viscera l(n=13},

꾀침 the masticator( n=9}, and the sublingua l(n=7) spaces. In contrast, in 23 patients with nonodontogenic DNSI, the anterior visceral space(n=14) was most fre­quently involved. The parapharyngeal, the submandib비ar， and the masticator spaces were statistically more frequently involved in odontogenic than in no­nodontogenic DNSl(p < .05) . Twenty-two patients had one or more com미 i ­

cationsshown by CT, ofwhich airwaycompromisewas morefrequentand severe in odontogenicthan in nonodontogenic DNSI.

Conclusion : We conclude that the parapharyngeal , the submandibular, and the masticator spaces are more significantly vulnerable in odontogenic DNSI than in nonodontogenic DNSI. The predilection for certain spaces of the neck in odonto­genic DNSI seems to originate from the intimate relationship of the mandibular molars to the adjacent deep neck spaces.

Index Words: Neck, infections Neck, CT Teeth

A thorough knowledge of the anatomy of the deep fascial layers and spaces of the neck is ess~ntial to understand the passageway of deep neck space infec­tion(DNSI) , because serious life - threatening compli­catons such as airway compromise , jugular vein th­rombosis , mediastinal involvement, pericarditis , pneu­monia , empyema, arterial erosion , or intracranial ex-

'Departmenl ofD iagnoslic Radiology, Gyeongsan Nalional Unversily Hospilal 2Departmenl olOlorhi nolaryngology, Gyeongsang Nalional Universily Hospilal Received Oclober 6, 1995; Accepled November 3, 1995 Address reprinl requesls 10 : Hyung-Jin Kim, M.D., Deparlmenl 01 Diagnoslic Radiology, Gyeongsang Nalional Unversily Hospilal ~ 92, Chi lam-dong, Chinju 660- 702, Korea

Te l. 82-591- 50-8202 Fax. 82-591- 758- 1568

tension may ensue if the diagnosis and the treatment are delayed. Recently , dental infection has become one of the most important sources of DNSI(1 , 2). It often has been described that the particular spaces of the neck tend to be involved in odontogen ic DNSI. This mainly results from the close relationship between the position of the dental roots and the origi n or insertion of the adjacent muscles of the jaw and the mouth floor(2 , 3). Although there have been many descriptions con­cerning the usefulness ofcomputed tomography(CT) in the evaluation of DNSI(4-14) , less attention has been paid on the different modes of spread between odonto­gen ic and nonodontogenic DNSls in the radiologic lit­erature(7 , 10, 14). However , because of the reported

853 -

Journal of the Korean Radiological Society 1995: 33(6) : 853-860

predilection of the certain spaces involvement in odon­togenic DNSI , its early detection by CT may lead one to search for an odontogenic source and help one direct to the appropriate management without delay. Accor­dingly , the purpose of this study was to evaluate the usefulness of CT in patients with DNSI with special em­phasis on the differences in the spaces involved and in complications between odontogenic and nonodonto­genic DNSls.

MATERIALS and METHODS

We reviewed the CT scans of 60 patients with DNSI examined between March 1989 and May 1993 in our in­stitution. Of the 60 patients , the cause of DNSI could be identified by the medical history or physical examin­ation in 44 patients who formed the basis of this study. There were 28 men and 16 women with their ages ranging from 3 to 78 years(mean , 42 years). The diag­nosis was established by surgery in 31 patients and by the clinical signs and symptoms in 13 patients. The specific microorganisms were isolated from the cul­ture in 26 patients , among whom Streptococcus spe­cies was the most common aerobe demonstrated in 17 patients.

Among the causes of DNSI identified , dental infec­tion was the most frequent etiology(n=21) , followed by trauma including surgery and foreign body injury (n=10) , upper respiratory tract or tonsillar infection (n=8) , sep- ticemia(n=2) , skin furuncle(n=1) , erysip­elas(n =1) , and spread of infection by the infected sec­ond branchial cleft cyst(n=1). Among the 10 patients with trauma, there were three patients with fishbone in­jury , two in the cervical esophagus and one in the mouth floor , and one patient with acupuncture in the posterior neck, hematoma in the posterior neck follow­ing blunt injury, mandibular fracture , and hypoph­aryngeal perforation by blunt trauma each. In the re­maining three patients , DNSI developed after surgery , two in the anterior neck and one in the posterior neck.

We classified the cause into odontogenic only if all of the following three criteria were met : (1) the presence of the dental infection should be documented by a den­tist; (2) there should be an appropriate cause -and-ef

10

a b

fect chronological sequence between the dental prob­lem and the clinical manifestations ; and(3) other cau­ses could be excluded clinically. Of the 21 patients with the dental infection , the specific teeth infected were the second or third mandibular molar in 15 patients , the fi rst mandibul ar mol ar with or without the teeth anterior to it in four patients , and the second maxillary molar in one patien t. In the remaining one patient, there were widespread caries in the mandibular teeth.

CT was performed with a GE 9800 scanner( GE Medi­cal System , Milwaukee, Wis) after the IV administraton of the contrast material. Axial scans were obtained in all patients with a 5-10mm slice thickness and table incrementation through the region of interest. The ad­ditional coronal scans were obtained in some patients. If necessary , the gantry angle was modified to mini­mize artifacts from metallic dental reconstructions.

We reviewed CT scans, paying partic비ar attention to the location and extent of the infectious process , the presence of an abscess to be drained , and the compli­cation(s) associated with DNSI. Two radiologists inter­preted the CT scans and reached a consensus. I nvolve­ment of each space by the infectious process was do­cumented only on a present-or -absent basis. Abscess was considered to be present , ifthere was discrete low density area with peripheral rim enhancement at CT. If there was only soft tissue swelling or infiltration which 。bliterated the fascial plane(s) at CT, cellulitis was con­sidered. When abscess in one space was accompanied by the infiltration in the contiguous space(s) at CT, we considered the patient to have both abscess and cellu ­litis. We then compared the CT features between odon­togenic and nonodontogenic DNSls with special em­phasis on the differences in the spaces involved and in the rate and type of compl ications.

According to the various authors , different terms have been applied to describe the fasciae and spaces of the deep neck. Among those , we largely adopted the terminology described by Harnsberger(15) in this pa­per.

Fig . 1. A 74-year-old woman with the right third mandibular molar infection CT scans at the level of the ramus (a) and the body (b) of the mandible show multiple abscesses in the right masti­cato r( arrow in a) , submandibular(arr­。WS In 비 ， and sublingual (arrowheads in b) spaces. Note the higher attenuation (open arrow in a) in the right paraph­aryngeal space fat in comparison with the contralateral normal side

- 854 -

Hyung-Jin Kim, et al : Odontogenic Versus Nonodontogenic Deep Neck Space Infections

lection without significant complication were treated

with trial of needle aspiration followed by medication

without single case of failure. AII of seven patients

whose CT showed cellulitis alone improved with medi­

cation only. Of the 37 patients who had abscess , gas

bubbles within abscess were noted in 18 patients.

RESULTS

Abscess versus Cellulitis

Of the 44 patients, seven patients were judged to

have cellulitis alone and 37 patients to have both cellu­

litis and abscess based on C T. Incision and drainage

procedure was performed in 31 out of 37 patients with

abscess(es) and confirmed the CT findings. Six pati­

ents in whom CT showed a minimal amount offluid col-

Differences in the Spaces Involved between

Odontogenic and Nonodontogenic DNSls

In general , in the patients with a more obvious clini-

Table 1 . Differences in the Spaces Involved between Odontogenic and Nonodontogenic Deep Neck Space Inlections (DNSls)

No. 01 Patients

Anatomic Spaces Involved

Parapharyngeal space

Submandibular space

Anterior visceral space

Masticator space

Sublingual space

Carotid space

Retropharyngeal space

Posterior cervical space

Parotid space

Paraspi nal space

SpaceolBurns

Superlicial space

"Signilicant at p<.05 by X2 analysis

Odontogenic DNSI (n=21)

18"

18a

13

9a

7" 끼4 끼/L 끼/L

o 17

를 ! μ a b

a b

- 855 -

N。nodontogenic DNSI (n=23)

5

9

14

2

3

2

8

6

2

3

3

18

Fig. 2. A 79-year-old woman with the

right second maxillary molar inlection

CT scans at the level 01 the ramus (a)

and the angle (b) 01 the mandible show

multiple abscesses with air-Iluid level in

the right superlicial(arrows) and deep

(arrowheads in a) masticator , and pa­

rapharyngeal(open arrows) spaces

Fig. 3. Nonodontogenic deep neck sp­

ace inlection. a. CT scan 01 a 68-year-old woman with

the upper respi ratory tract inlection

shows an abscess in the right anterior

visceral space b. CT scan 01 a 60-year-old woman with

the upper respiratory tract inlection

shows an abscess in the retropharyn­

geal space.

Journal of the Korean Radiological Society 1995: 33(6) : 853-860

cal history such as trauma, tonsillitis , skin furuncle , or

the underlying neck cyst, it was easy to interpret CT findings concerning the primary site of infection. How­

ever, the massive involvement of the contiguous neck

spaces seen in many patients in our study prevented us

from telling the exact pathways of the spread of infec­tion or the initial site of infection by CT with confidence.

In many patients , infection seemed to violate the adjac­

ent neck spaces without respecting the fascial barrier Nevertheless, certain spaces of the neck were more

frequently involved in odontogenic DNSI than in nono­

dontogenic DNSI. The differences in the spaces invol­

ved between odontogenic and nonodontogenic DNSls are summarized in Table 1. Of the 21 patients with od­

ontogenic DNSI , the parapharyngeal and the subman­

dibular spaces were the two most common spaces

involved seen in 18 patients each , followed by the an­terior visceral , the masticator , and the sublingual sp­

aces(Fig. 1 and 2). In contrast , of the 23 patients with

nonodontogenic DNSI , the anterior visceral space(Fig.

Table 2. Differences in Complications between Odontogenic and Nonodontogenic Deep Neck Space Infections (DNSls)

랩

n”

뼈

빠 -5

4

얘

-

m

빼

N Total

N0.ofPatients(n=44)

17

6

No. of Patients

Compl ications

Airway compromise

Mediastinitis/

mediastinal abscess

Jugular vein thrombosis

Empyema

Pericarditis

Aortic pseudoaneurysm

aSignilicant at p< .05 by X2 analysis.

Odontogenic DNSI (n=21)

12a

2

nu

nu

nU

2

“1훌 -... ‘ ~훌 ‘ ....... if* f li요↓‘~"，...훌률톨빼b

a

c

Fig. 4. Complications of deep neck space infection. a. Mediastinal abscess: CT scan of a 39-year-old man with a trau­matic injury to the hypopharynx shows a gas-containing abscess involving the mediastinum b. Jugular vein thrombosis: CT scan of a 44-year-old man with the left second mandibular molar inlection shows an abscess containing gas bubbles within the left carotid sheath , The lat­erally displaced internal jugular vein is severely compressed and appears in crescentic shape(arrows) , Surgery conlirmed the partially eroded and thrombosed jugular vein ,

c. Pseudoaneurysm lormation : CT scan 01 a 52-year 이 d woman with the fishbone injury to the hypopharynx shows a large abnor­mal vascular structure(arrow) beneath the aortic arch , ac­companied by the mediastinal inlection. At surgery , inadvertent dissection of the mediastinum caused massive uncontrolled hemorrahge

856 -

Hyung -Jin Kim, et al: Odontogen ic Versus Nonodontogen ic Deep Neck Space Infections

3a) was most 1requently involved seen in 14 patients , followed by the submandibular , the retropharyngeal (Fig. 3b) , the posterior cervical , and the parapharyn­geal spaces. The parapharyngeal , the submandibular , and the masticator spaces were statistically more fre­quently involved in odontogenic than in nonodontoge­nic DNSI(p< .05). According to the specific teeth in1ec­ted , there was no significant difference in the spaces in­volvement among the patients with odontogenic DNSI.

Complications of DNSI Twenty - two patients had one or more complications

shown by CT(Table 2). These were airway compro­mise, mediastinal involvement(Fig. 4a) , jugular vein thrombosis(Fig. 4b) , empyema, pericarditis , and pseu­doaneurysm 01 the aortic arch(Fig. 4c). Among the vari­ous complications , airway compromise was statisti­cally more frequent in odontogenic than in nonodon­togenic DNSI((p < .05).

DISCUSSION

Deep neck space infection involves the spaces 01 the head and neck which are surrounded and compart­mentalized by the three layers of the deep cervical fascia. It can spread via Iymphatic vessels , blood ve­ssels , or direct extension , leading to multiple contigu­ous spaces involvement, and thus be categorized into transpatial disease(16). Since the advent of antibiot­ics, the incidence of DNSI has continuously declined , but it still occasionally causes serious life - threaten­ing complications. For the early diagnosis and the ap­propriate treatment to reduce a morbidity and mor-

Buccalspace

tality , a comprehensive understanding of the perti­nent cervical anatomy is crucial , described else­where(1 , 13, 17 -19).

Etiology of DNSI In the preantibiotic era, pharyngeal infection was the

most common cause of DNSI. However, with the intro­duction of antibiotics, various other portals of entry have been recognized more importantly than before(1 , 11 -13, 20) ; these are intravenous drug abuse, odonto­genic in1ection , skin infection , ear infection , cervical adenitis , trauma, and so forth. It has also been reported that the cause 01 DNSI could not be found up to 50% of the cases(20). Concerning odontogenic infection , the prevalence has been reported in 9-34%(1 , 12, 20). If we include the 16 patients with unknown etiology in our study, odontogenic infection comprises 35%(21/6이 01 the etiology of DNSI.

Abscess versus Cellulitis There are different opinions about the optimal timing

of surgical therapy for DNSI. While recent investigators advocate early incision and drainage during the cellu­litis stage(20) , the more or less prevailing traditional thought is that no drainage should be attempted during the stage of cellulitis(5 , 9). It is often difficult to deter­mine clinically whether a patient with a pain1ul tender and swollen neck has an abscess or has cell비 itis

alone, because the characteristic fluctuation of an ab­scess frequently cannot be felt in the region 01 the deep neck(12, 13). In the study 0151 patients with neck ab­scess , Tom and Rice(12) reported that 11uctuation was noted only in 27%. One of the most important values 01 CT in the evaluation of DNSI is its high reliability on the

Mylohyoid

Fig . 5. Diagram demonstrating the different pathways 01 the den- Fig. 6. Diagram demonstrating the different pathways 01 the den-tal inlection according to the relationship between the buccinator tal inlection according to the relationship between the mylohy이 d

muscle attachment to the alveolar process and the position ofthe muscle attachment to the mandible and the position of the root root apices ofthe molar teeth. If a maxillary molar with a long root apices of the mandibular teeth. If a mandibular teeth with a long extending beyond the buccinator attachment is infected , inlection root extending beyond the mylohyoid ridge is infected as shown can reach the buccal space with ease, from which it can spread on the right side of the diagram, infection can directly spread to into the masticator or submandibular space. On the contrary , if a the submandibular space. In contrast, if a mandibular teeth with a mandibular molar with a short root above the buccinator attach- short root above the mylohyoid ridge is inlected as shown on the ment is infected , inlection is preponderantly directed to the oral left side of the diagram, infection tends to involve the sublingal vestibule. spaceinitially

- 857 -

Journal of the Korean Radiological Society 1995 ; 33(6 ) : 853-860

differentiation between abscess and cellulitis , thus it is useful for treatment planning and management of the patients. The distinguishing feature of abscess vs ce-11 비 itis on CT scan is the enhancing rim that represents the wall of abscess. 1 n our study , surgery confirmed the CT findings in all of 31 patients whose CT showed a con­siderable amount of abscess fluid. In six patients in whom CT showed a minimal amount of abscess fluid , we simply aspirated the fluid followed by medication in­stead of incision and drainage. With this method , we could achieve the favorable clinical outcome in most of the cases

Differences in the Spaces Involved between Odontogenic and Nonodontogenic DNSls

Although the massive involvement of the contiguous neck spaces often made it difficult to make sure about the exact pathways of the spread of infection in this study, our results showed that there is a different tend­ency of the space involvement between odontogenic and nonodontogenic DNSls. This knowledge of the pre­dilection sites involved in odontogenic DNSI is import­ant because it may lead a radiologist to warn the cl in­ician to search for an underlying or hidden odontogenic source and help guide to remove it without delay. In our study , the parapharyngeal , the submandibular, and the masticator spaces were statistically more frequently involved in odontogenic than in nonodontogenic DNSI. Although the sublingual space was also more frequ­ently involved in odontogenic(7/21) than in nonodon­togenic(3/23) DNSI , it was not statistically significan t.

More than one anatomical relationships between the mandibular or maxillary molars and the adjacent deep neck spaces may result in the different spaces involve­ment between the two groups. Several important ana­tomic details are described below. First , in the region of the mandibular and maxillary molars, it is the attach­ment of the buccinator muscle to the base of the al­veolar process that directs the path of the dental infec­tion(3). If there is an infection of the mandibular or maxillary molars which have long roots extending be­yond the buccinator attachment, as in the second or third mandibular and maxillary molars, it can reach the buccal space with ease , from which it can spread into the masticator or submandibular space(3 , 19) (Fig. 5). Second, infection from a mandibular molar that directly perforates the buccal or lingual plate of the mandible can invade the superficial or deep masticator space, respectively(3). Third , the relation of the apices of the mandibular teeth to the mandibular attachment of the mylohy이 d muscle is such that, in general , the roots of the second and third mandibular molars extend beyond the mylohyoid ridge , while the roots of the anterior ones never reach so far. Consequently , dental infec­tions anterior to the second mandibular molar tend first to involve the sublingal space , while those of the sec­ond or third mandibular molar can directly involve the

submandibular space(2 , 3, 19, 22) (Fig. 6). Fourth , once the submandibular space is involved , there is more chances of the spread of infection to the parapharyn­geal space because the fascial barrier between the parapharyngeal space and the submylohoid space is the thinnest providing the path of least resistance(3) . Last, although the sublingual and submandibular spa­ces are separated by the mylohy이 d muscle, they are freely communicating posteriorly to each other , mak­ing the infection of one space involve the other space with ease(13).

Our original thought was that there should be a dif­ferent pattern of the spaces i nvolvement regardi ng the specific tooth i nfected. Unfortunately , however , we could not prove it in our study. The great part of this nonspecificity may be attributed to the subjects in­cluded in our study. In the patients who have infection in the anterior teeth , the clinical detection and the treat­ment usually can be made more promptly without the aid ofthe sophisticated imaging study such as CT. Most of the serious odontogenic infection requiring CT arise from the mandibular molars. Li kewise , if we exclude the one patient who had widespread caries in the man­dibular teeth , all patients except one who had infection in the second maxillary molar had an infection focus in the mandibular molars , 15 patients in the second or third molar , and four patients in the first molar. Wide­spread infection seen in many patients in our study also masked the primary site of infection. Interestingly, although it was not included in the results , among the four patients who had infection in the first mandibular molar, the sublingual space was involved in only one patien t. The other three patients had the submandi­bular space involvement sparing the sublingual space. This finding suggests that the relation of the apex of the first mandibular molar to the mandibular attachment of the mylohy이 d muscle may be subject to the individual variation . In cases where it extends below the mylo­hyoid ridge , the submandibular space can be directly involved ; on the other hand , in cases where it lies ab­ove the mylohy이d ridge , the sublingual space is in­volved first(Fig. 6).

Complications of DNSI The most important role of CT in the evaluation of

DNSI is probably the identification of serious compli­cations. A variety of complications associated with DN­SI have been reported ; airway compromise , jugular vein thrombosis , mediastinal involvement , pericardi­tis , pneumonia, empyema, arterial erosion , intracra­nial extension , or necrotizing fasciitis(5 , 8, 9, 23 -25). Recognition of these potentially life -threatening com­plications as soon as possible should lessen morbidity and mortality by the prompt surgical procedures. In ou r study , airway compromise and mediastinal involve­ment were the two most frequent complications of DNSI. Although the reason is not clear , the ai rway com-

858 -

Hyung-Jin Kim , et al: Odontogenic Versus Nonodontogenic Deep Neck Space Infections

promise was more frequent and severe in odontogenic

than in nonodontogenic DNSI , similar to the results of

Tom and Rice(12). Mediastinal involvement was less

prevalent in odontogenic DNSI in our study. The spread

of DNSI into the mediastinum can occur via the anterior

visceral space or the retropharyngeal space , either of

which has components that cross the thoracic inlet(13)

In addition , mediastinitis can also result from suppu­

ration in the parapharyngeal space spreading down

through the carotid sheath(11) . A single mortality case

in our series occurred in a patient with fishbone i 미 ury

to the cervical esophagus. In this patient, the medias­

tinal extension of DNSI caused the erosion of the aortic

arch resulting in the formation of a pseudoaneurysm

(Fig.4c)

We conclude that somewhat different modes of sp­

read of infection exist between odontogenic and nono­

dontogenic DNSls and this difference originates from

the intimate relationship of the mandibular molars to

the adjacent deep neck spaces. CT should be used as

the principal diagnostic tool in the evaluation and man­

agement of DNSI.

REFERENCES

1. Everts EC , Echevarria J. Diseases ofthe pharynx and deep neck

infections.ln :Paparella MM , Shunrick DA, eds. Otolaryngology,

Head and Neck‘ Vol. 3. Philadelphia : W‘ B. SaundersCo., 1980 :

2302-2322

2. Chow AW, Roser SM , Brady FA. Orofacial odontogenic infec

tions. Ann Int Med 1978 ; 88 : 392-402

3. Spilka CJ. Pathways of dental infections. J Oral Surg 1966 ; 24

111 -124

4. Merhar GL, Col’ey DP , Clark RA , Herwig SR. Computed tomo­

graphic demonstration of cervical abscesses and jugular vein

thrombosis. Arch OtolaryngoJ 1981 ; 1 07 : 313-315

5. Endicott JN , Nelson RJ , Saraceno CA Diagnosis and manage­

ment decisions in infections of the deep fascial spaces of the

head and neck utilizing computerized tomography. Laryngo­

scope 1982 ; 92 : 630-633

6. Holt GR , McManus G, Newman RK , Potter JL, Tinsley PP

Computed tomography in the diagnosis of deep-neck infections

Arch Otolaryngo/1982 ; 1 08: 693-696

7. Braun IF , Hoffman JC, Reede D, Grist W. Computed tomography

。f the buccomasseteric region : 2. Pathology. AJNR 1984 ; 5

611 -616

8. Wenig BL , Shikowitz J, Abramson AL. Necrotizing fasci itis as a

lethal complication of peritonsillar abscess. Laryngoscope 1984

; 94 : 1576-1579

9. Nyberg DA, Jeffrey B, Brant-Zawadzki M, et al. Computed tom­

ography of cervical infections. J Comput Assist Tomogr 1985 ; 9

288-296

10. Hardin CW‘ Harnsberger HR , Osborn AG, et al. Infection and

tumor of the masticator space : CT evaluation. Radiology 1985 ;

157: 413-417

11 . Stiernberg CM. Deep-neck space infections : diagnosis and man­

agemenl. Arch Otolaryngol Head Neck Surg 1986 ; 112: 1274

1279 12. Tom MB , Rice DH. Presentation and management of neck ab­

scess: a retrospective analysis. Laryngoscope 1988 ; 98 : 877-

880

13. Smoker WRK , Harnsberger HR , Reede DL, Holliday RA, Som PM ,

Bergeron RT. The neck.ln : Som PM , Bergeron RT, eds. Head and

Neck Imaging. SI. Louis : Mosby-Year Book , 1991 : 497-591

14. Holliday RA, Perdergast NC. Imaging inflammatory processes of

the oral cavity and suprahyoid neck. Oral Maxillofac Clin North

Am 1992 ; 4 : 215-240

15. Harnsberger HR. Introduction to the suprahyoid neck. In : Ha­

rnsberge r HR, ed. Handbooks in Radiology : Head and Neck

Imaging. Chicago : Year Book Medical Publishers , 1990 ‘ 3-28

16. Vogelzang P, Harnsberger HR , Smoker WRK. Multispatial and

transpatial diseases of the extracranial head and neck. Semin

UltrasoundCT MR1991 ; 12: 274-287

17. Grodinsky M, Hol yoke EA. Fasciae and fascial spaces of head ,

neck and adjacent regions. Am J Anat1938 ; 63: 367-408

18. Levitt GW. Cervical fascia and deep neck infections. Laryngo­

scope 1970 ; 80 : 409-435

19. Paonessa DF, Goldstein JC. Anatomy and physiology of head

and neck infections(with emphasis on the fascia of the face and

neck). Otolaryngol Clin North Am 1976 ; 9 : 561-580

20. WrightNL. Cervical infections. Am J Surg 1967 ; 113: 379-386

21 . Fl yn n TR. Odontogenic infections. Oral Maxillofac Clin North Am

1991 ;3: 311-329

22. Tschiassny K. Ludwig ’ s angina : Surgical approach based on

anatomical and pathological criteria. Ann Otol Rhin Laryngol

1947 ; 56 : 937-945

23. Wills PI , Vernon RP Jr. Complications of space infections in the

head and neck. Laryngoscope 1981 ; 91 : 1129-1136

24. Beck HJ , Salassa JR , McCaffrey TV , Hermans PE. Li fe-threaten­

ing soft tissue infections of the neck. Laryngoscope 1984 ; 94 ’

354-362

25. Moss RM , Kunpittaya S, Sorasuchart A Cervical necrotizing

fasciitis: an uncommon sequela to dental infection. Ann 0101

Rhinol Laryngo/1990 ; 99 : 643-646

폈

Journal of the Korean Radiological Society 1995 : 33(6) : 853-860

대 한 방사 선 의 학 회 지 1995 : 33( 6) : 853-860

CT를 이용한 치성과 비치성 심부 경부 감염의 비교: 파급된 경부 공간을 중심으로1

1 경상대학교의과대학진단방사선과학교실

2경상대학교 의과대학 이비인후과학교실

김형진 · 박의동 · 김정희 · 김재형 · 황으171 2 . 정성훈

목 적 :본 논문의 목적은 CT를 이용하여 치성과 비치성 심부 경부 감염에 있어서 침범된 경부 곰간과 합병증의 차이를 알

아보기 위힘이다.

대상 및 방법 :심부 경부 갑염으로 확진되고 임상적으로 그 원인을 파악할 수 있었던 44명의 환자를 대상으로 하였으며 이

중 치성 감염자가 21 명， 비치성 감염자가 23명이었다.21명의 치성 감염자중 15명에서 선행이 된 치아 병변은 제 2 또는 제 3

하악 대구치에 있었다. 저자들은 치성과 비치성 심부 경부 감염의 두 군에서 침범된 심부 경부 공간과 합병증의 앙상에 상이

한 점이 있는지에 주안점을 두고 CT 소견을 분석하였다.

결 과 : 전예에서 CT로 농앙과 봉와직염의 구분이 가능하였다. 21 명의 치성 감염자군에서 가장 흔히 침범되는 심부 경부

공간은 각각 18명에서 관찰된 부인두강과 하악하강이었고 내장강이 13명， 저작강이 9명， 설하강이 7멸에서 침범되었다. 반면

23명의 비치성 감염자군에서는 내장강이 14명으로 가장 흔히 침범되었다. 부인두강， 하악하강과 저작근강은 치성 감염자군

에서 비치성 감염자군보다 통계학적으로 유의하게 더 흔히 침범되었다. 전체 환자중 22명에서 하나 또는 그 이상의 함병증

이 CT에서 관찰되었는데 이중 기도 압박의 소견은 치성 감염자군에서 통계학적으로 유의하게 더 흔하였다.

결 론.치성 감염시 심부 경부 공간들중 부인두강， 하악하강과 저작근강은 비치성 감엽때보다 쉽게 침범되며 치성 감염에

의한 이러한 특정 공간의 침범은 하악 대구치와 그에 인접한 경부 곰간틀간의 긴밀한 해부학적 관계에 기인한 것으로 사료

된다.

- 860 -