Deep Neck Spaces Infections in Radiology

Abstract

Deep Neck Spaces radiology and review of deep neck infections at KingAbdulaziz University Hospital

Key words: infection, tomography and magnetic resonance

Khaled Al-Nourya, Alsaid Lotfyb

Otolaryngology Departmenta, King Abdulaziz University, Jeddah, Saudi ArabiaRadiology Departmentb, International Medical Centre, Jeddah, Saudi Arabia

The anatomy of the deep neck spaces is a complex one, and deep neck infections are a common clinical problem in allage groups. Otolaryngologists should be familiar with the detailed anatomy and divisions of the neck in order to properlydiagnose and treat patients. Thus, knowledge of the radiological anatomy is paramount for accurately localizing andevaluating the extent of the deep neck inflammatory processes. In this study, we present the anatomy of normal deep neckspaces in computerized tomography and magnetic resonance images. The computerized tomography images of 56 patientsthat presented with symptoms indicative of deep neck infections were also reviewed to illustrate the different patterns ofneck infections and their relationship with thoracic abnormalities. Because of the frequent thoracic abnormalities associatedwith neck infections, we recommend performing a CT scan of chest in the same sitting with the CT of neck and absenceof ring enhancement does not exclude the presence of abscess

ORIGINAL ARTICLE

Introduction as infections. Some neck infections result from suppurativeadenitis, with the primary infection often originating fromthe mucosa of the oral cavity, paranasal sinuses, andpharynx. Dental caries and periodontal disease of the teethare other sources frequently implicated in deep neckinfection12,13 .

Neck AnatomyA. Fascial Layers in the Neck

The fascial layers in the neck are either superficial ordeep. The superficial fascia lies just below the subcutaneoustissue and envelopes the platysma muscle. The deep fasciallayers comprise the investing, visceral, and prevertebralfasciae14 .The investing fascia envelopes thesternocleidomastoid and trapezius muscles and divides toform the capsules of the submandibular and parotid glands(Figure 1). Both the superficial fascia and the investinglayer surround the entire neck14. The visceral (pretracheal)fascia forms a circle in the center of the anterior neck(Figure 2). Within its bounds are the pharynx, trachea,esophagus, and thyroid and parathyroid glands. Posteriorly,the visceral fascia lies between the esophagus and thevertebral bodies. Anteriorly, the visceral fascia divides toform the capsule of the thyroid gland14.The prevertebralfascia encircles the spine and the paraspinal muscles(Figure 3). Anterior to the vertebral bodies, the prevertebralfascia divides into 2 layers: (a) the alar fascia (anterior)and (b) the true prevertebral fascia (posterior). Betweenthe alar and true prevertebral fasciae is the importantdanger space that is involved in the spread of infection14.

Infection of the neck is a common clinicalproblem in all age groups, especially among children andyoung adults1–3. Imaging studies that use computerizedtomography (CT) and magnetic resonance imaging (MRI)are frequently required to confirm a diagnosis and, moreimportantly, to localize the infectious process and todelineate an abscessed cavity. Hence, knowledge of theanatomy is paramount for localizing and evaluating theextent of the inflammatory process4–7.

Neck infections can be categorized as either deep(i.e.,intrinsic) or extrinsic8–10. Intrinsic infections includeLudwig’s angina, peritonsillar, parapharyngeal, andretropharyngeal inflammation, while extrinsic infectionscomprise cervical adenitis and submandibular, submental,and masticator space infections . The causative organismsof neck infections such as lymphadenitis are varied butcan be subdivided into bacterial, fungal, parasitic, or viral.Unfortunately, if the treatment of a neck infection isdelayed or inadequate, mediastinal extension may ensueand result in pain, dyspnea, asphyxia secondary toaspiration, swelling with mediastinal widening, anddysphagia f rom esophageal compression1 1 .

The source of neck infections is unknown in up to50% of cases. Known source are mainly extrinsic such

EJENTAS Egyptian Journal of ear, nose, throat and Allied Sciences Volume 11, December 2010

69


70

The carotid sheath is formed with contributions fromall 3 layers of the deep cervical fascia. The sheath surroundsthe common carotid artery, internal jugular vein, andvagus nerve. When the common carotid artery dividesinto the internal and external branches, the external carotidartery leaves the carotid sheath, and the internal carotidartery continues within the sheath to the base of the skull.The infrahyoid fascia, formed with contributions from theinvesting and visceral layers of the deep cervical fascia,surrounds the strap muscles14.

Deep Neck Spaces radiology and review of deep neck infections, Al-Noury & Lotfy

The submandibular space corresponds to the submentaland submandibular triangles. The roof of the submandibularspace is the mylohyoid muscle (Figure 4). The structureswithin the visceral fascia are said to be in the visceralspace (Figure 5). The larynx, trachea, esophagus, thyroidgland, and parathyroid glands are included in the visceralspace. Structures within the prevertebral fascia, includingthe spine, the paraspinal muscles, the vertebral artery, andthe phrenic nerve, are in the prevertebal space (Figure 6).The tissues between the pharynx and the vertebral columnare often grouped together as the retropharyngeal space.There are, in actuality, 4 distinct spaces between thepharynx (or esophagus) and the vertebral bodies. Fromanterior to posterior, they are (1) the visceral space, (2)the true retropharyngeal space, (3) the danger space, andthe (4) prevertebral space. The true retropharyngeal spacelies between the visceral fascia and the alar fascia. Theretropharyngeal lymph nodes are located within the trueretropharyngeal space (Figure 7) . The danger space liesbetween the alar fascia and the true prevertebral fascia,extending from the skull base to the diaphragm. Thedanger space is so termed because infections can spreadvertically along the entire length of the neck and torso17.

Figure 1: Axial MRI T1WI (A) and CT (B) images of theinvesting fascia, as outlined in red.

Figure 2: Axial MRI T1 WI (A) and CT (B) images of thevisceral fascia, as outlined in red.

Figure 3: Axial MRT1 WI (A) and CT images (B) of theprevertebral fascia, as outlined in red.

B. Neck Spaces

The cervical fascia divides the neck into several distinctcompartments (spaces). One should be aware of thesespaces and their defining landmarks, not only becausethey affect the spread of infection, but also because theyinfluence the differential diagnosis of pathologicfindings15,16.

Figure 4: Axial CT image of the submandibular space, asoutlined in red.

The parapharyngeal space lies between the layers ofthe deep cervical fascia, with the pharynx medial, thecarotid sheath posterior, and the parotid gland lateral. Thestyloid process of the temporal bone, along with itsassociated muscles and ligaments, divides theparapharyngeal space into the prestyloid and neurovascularcompartments.

Figure 5: Axial CT image of the visceral space, as outlined in red.

Figure 7: Axial CT scan of a patient with nasopharyngeal cancer(indicated by the short, blue arrow) and a retropharyngeal lymphnode within the true retropharyngeal space (long, red arrow)


71

The prestyloid parapharyngeal space contains fat andextends only to the level of the hyoid bone. Theneurovascular space corresponds to the carotid sheath andextends the length of the neck (Figure 8).


Figure 6: Axial CT image of the prevertebral space, as outlined in red.

Figure 8: Axial CT image of the parapharyngeal space, as outined in red.

Below the level of the hyoid bone, there are fat-filled spaces anterior and posterior to the carotid sheath.These are the anterior and posterior cervical spaces. Theposterior cervical space corresponds to the posteriortriangle and contains the spinal accessory and dorsalscapular nerves, along with the spinal accessory lymphnode chain (Figure 9).

Figure 9: Spinal accessory lymph nodes in the posterior cervicalspace (indicated by the arrow)

Patients and MethodsWe reviewed the CT images of patients that presented

to our department from January 2001 through March 2008because of suspected neck infections. For all of thesepatients, a study of the neck and chest had been performedusing both multislice CT 4 slices (General ElectricLightspeed) and 16 slices (Siemens Sensation). For eachpatient, 100 cm3 of 300 mg/ml non-ionic contrast materialwas intravenously administrated at a rate of 3.5 cm3/s.The medical records of the patients were reviewedfollowing the approval of the hospital ethics committee.The CT images were reviewed for:

1.The neck spaces involved2.Abnormalities in the neck lymph nodes3.Presence of ring enhancing lesions4.Presence of air pockets5.Mediastinal extensions6.Parenchymal lesions in the lungs7.Pleural effusions8.Mediastinal lymph nodes9.Chest wall abnormalities

ResultsFifty-six patients (38 males and 18 females) were

included in our study. The mean age was 43.3 years. Theyoungest individual was 2 years of age and had HIV, andthe oldest individual was 73 years old. Twenty patients(35.7 %) had diabetes mellitus, 4 had HIV (7.1 %), 2 (3.6%) were infected with a branchial cleft cyst, and 6 (10.7%) had underlying dental problems.

The anatomical spacePeritonsillarSubmandibularParapharyngealRetropharyngealSublingualAnterior cervicalPosterior cervicalParotid space

Number of patients24141484822

Table 1: The neck spaces involved in infection as determined by CT.


72


Figure 10: Axial CT images of a patient with an infected secondbranchial cyst extending to the mediastinum (indicated by thered arrows)

Figure 11: Axial CT of patient with Ludwig’s angina.

Figure 12: Axial CT of patient with necrotizing fasciitis andthickening of skin subcutaneous tissues and muscles. Smallsubmental lymph nodes (arrows) and air pockets in several neckspaces are also observed (circle).

The spaces of neck involved, as determined by CT, wereas follows (Table 1): -− Lymph nodal abnormalities were identified in 36 patients(32 %). These abnormalities were in the form of:

1.Enlargement (using a 1.5 short axis diameter as thelimit in all levels)2.Abnormal enhancement (significant or ringenhancement)3.Central hypodensity

− Ring enhanced fluid collections were found in 30 patients(53.6 %).− Air pockets were detected in 20 patients (35.7 %).− 4 patients with air pockets (7.1 %) and without ringenhancement were shown to have pus during needleaspiration.− 12 patients (21.4 %) had mediastinal extensions in theform of mediastinal ring enhancing lesions or air pockets.− 4 patients (7.1 %) had lung parenchymal changes in theform of lung abscess and infiltrations.

− 2 patients (3.7 %) had mediastinal (right hilar) lymphnodal enlargement.− 4 patients (7.6 %) had multiple chest wall abscesses.− 8 patients (14.3 %) had a pleural effusion.− The most common organisms isolated in the pyogenicinfections were Staphylococcus aureus (14 patients, 25%) and Klebsiella pneumoniae (8 patients, 14.3 %).− Mycobacterium TB was the source of the infection in24 patients (42.9 %).− None of our patients (0 %) had jugular vein thrombosis.− 2 patients had a large, infected branchial cleft cyst thatextended inferiorly to the mediastinum (Figure 10).− 4 patients (7.1 %) had Ludwig’s angina (Figure 11).− 6 patients (10.7 %) had necrotizing fasciitis (Figure 12).

DiscussionThe chief modality for the evaluation of a neck infection

is contrast helical CT carried out in the axial plane. 32out of the 56 patients (57%) in our study were diabetic (n= 20), had HIV (n = 4), suffered from dental problems (n= 6), or had a preexisting pathology (n = 2) that consistedof a large branchial cyst extending into the mediastinum.We believe that these factors played a predisposing rolein the neck infections for this group of patients. In supportof dental problems playing a role in causing infection,Chowdhury et al and Kim et al have found that dentalcaries and periodontal disease of the teeth are frequentlyimplicated in deep neck infections12, 13.

In our study, the most frequently infected neck spacewas the peritonsillar space (n = 24) (42.9%), followed bythe parapharyngeal (n = 14) (25%) and the submandibular(n = 14) (25%) spaces. Consistent with these findings, ina large study that included 184 patients with deep neckinfections, Wang et al, found that the neck spaces mostfrequently involved in infection were the parapharyngeal(n = 77) (41.9%), peritonsillar (n = 59) (32.1%) andsubmandibular (n = 55) (29.9%) spaces18. In anotherretrospective study that included 117 children, Ungkanontet al.found the most frequent neck space involved wasthe peritonsillar space (49%), followed by theretropharyngeal space (22%), the submandibular space(14%), the buccal space (11%), and the parapharyngealspace, indicating that children may differ from adults intheir susceptibility to different types of infections1. In ourstudy, the only child (below 16 years of age) was a 2 yearold, HIV-positive male with a diffuse infective processand early abscess formation in the submandibular region.Based on its presence, Mycobacterium tuberculosis waspresumed to be the causative agent.

Retropharyngeal space infections were historicallybelieved to occur almost exclusively in children less than6 years of age19. However, with the increase in theimmunocompromised patient population, these infectionshave become more frequent in adults, occurring morecommonly in males than in females (2:1) . In our study,all the patients that had retropharyngeal space infectionwere adults (n = 8). In conformity with the generalpopulation ratio, 6 male patients and 2 female patient hadthis type of infection. The neck space infections wereeither peritonsillar (n = 6) or parapharyngeal (n = 2).Patients with retropharyngeal space infections often presentwith fever, chills, odynophagia, sore throat, dysphagia,


73


nausea, vomiting, respiratory distress, neck pain, andstiffness. Interestingly, however, in perhaps the first knowncase in our department, one of these patients presentedbecause of odynophagia, dysphagia, nausea and vomiting.In the first study, we had him swallow barium and thenperformed a CT of his neck and chest hence, in the absenceof signs of infections, barium swallow can help to delineatethe infectious process in the retropharyngeal neck space.

The danger point for the spreading of infection is thespace situated dorsal to the retropharyngeal space betweenthe alar and prevertebral fascia . This space extends fromthe skull base to the level of the posterior diaphragm.Infections that involve the retropharyngeal space may enterthe danger space and thus extend into the mediastinum.

In our study, 12 patients (21.4%) had mediastinalextensions of the infectious process: 6 from retropharyngealspace infection and 6 from anterior cervical space infection.In addition, suppurative uncontrolled infection of thetonsils may result in a peritonsillar abscess (quinsy) orrarely in a tonsillar abscess. A peritonsillar abscess is anaccumulation of pus around the palatine tonsils. If thisabscess extends outside the tonsillar fossa, it may involvethe lateral retropharyngeal or parapharyngeal spaces . Inour study, out of 24 patients with a peritonsillar infection,6 had retropharyngeal extension and 10 had parapharyngealextension.

Tuberculosis of the head and neck can involve thecervical lymph nodes, larynx, temporal bone, sinonasalcavity, eye, pharynx, thyroid gland, and skull base20. TBis a rare cause of cervical lymphadenopathy in childrenin the United States and is largely seen in children withAIDS21.The only child included in our study was an AIDSpatient and had a diffuse neck infection and an enhancedlymph node with central hypodensity. He was shown tohave TB. The other patient with AIDS in our study alsohad a TB neck abscess. Other patients with TB in ourstudy (n = 20) had chest wall abscesses (n = 4), mediastinallymph nodes (n = 2), lung abscesses (n = 2) and pleuraleffusion (n = 4).

The most common organism found in pyogenicinfections in our study was Staphylococcus aureus (n =14) (25%), followed by Klebsiella pneumonia (n = 8)(14.3%). In the study of Wang et al, Klebsiella pneumoniawas the organism most commonly isolated frominfections18.

If an abscess is clearly present, a CT scan usuallyshows a central, low attenuation region surrounded by anenhanced rim . However, the low attenuation center doesnot necessarily imply the presence of actual pus, as asimilar appearance can result from lymph nodes undergoingearly liquefaction (pre suppurative phase) and in thosethat have undergone complete liquefaction necrosis(suppurative phase). Again, the absence of a rimenhancement does not preclude the presence of pus.

Vural et al. studied the accuracy of CT scans of deepinfections in the pediatric population22. They concludedthat CT scans have important limitations in differentiatingabscesses and cellulites and that and clinical findingsbesides a CT-based diagnosis should guide the decisionfor surgery22.

The presence of thoracic manifestations with a neckinfection in our study was a clear finding. Out of the 56patients, 20 (35.7%) had one or more of the following:Mediastinal lesions as ring enhancing lesions or air pockets(n = 4)Lung abscess or infiltrations (n = 4)Mediastinal lymphadenopathy (n = 2)Chest wall abscesses (n = 4)Bilateral or unilateral pleural effusion (n = 8)

We assumed that mediastinal lesions were basicallydue to extension of the infection through the danger spaceor anterior cervical space. The lung infiltrations were seenmainly in patients with TB while lung abscesses werefound in a patients with necrotizing fasciitis and multipleneck abscesses. The enlarged mediastinal lymph noderight hilar was found in a patient with a TB neck and achest wall abscess. The presence of a pleural effusion wasseen in 4 patients with TB and 4 patients with a pyogenicinfection.

In conclusion, CT scan is usually a sufficient techniquein the detection of the extension of various neck infectionsand in guiding percutaneous procedures. Because of thefrequent thoracic abnormalities associated with neckinfections, we recommend performing a CT scan of chestin the same sitting with the CT of neck. This may provideinformation about the extent of the disease or give cluesabout the diagnosis. Absence of ring enhancement doesnot exclude the presence of a abscess. A clinical diagnosisbesides a CT-based one must play a complementary rolebefore any interventional procedures.


74


References

1. Ungkanont K, Yellon RF, Weissman JL, et al. Head and neckspace infections in infants and children. Otolaryngol HeadNeck Surg 1995;112(3):375–82.

2.Chow AW. Life threatening infections of the head and neck.Clin Infec Dis 1992;14:991–1004.

3.Nagy M, Pizzuto M, Backstrom J, et al. Deep neck infectionsin children: a new approach to diagnosis and treatment.Laryngoscope 1997;107 (12):1627–34.

4.Harnsberger HR, Osborn AG. Differential diagnosis of headand neck lesions based on their space of origin. 1. Thesuprahyoid part of the neck. Am J Roentgenol1991;157:147–54.

5.Levitt GW. Cervical fascia and deep neck infections.Otolaryngol Cl in North Am 1976;9:703–16.

6.Mukherji SK, Castillo M. A simplified approach to the spacesof the extracranial head & neck. Radiol Clin North Am1998;36(5):761–80.

7.Paonessa DF, Goldstein JC. Anatomy and physiology of headand neck infections (with emphasis on the fascia of the faceand neck). Otolaryngol Clin North Am 1976;9(3):561–80.

8.Johnson JT. Abscesses and deep space infections of the headand neck. Infect Dis Clin North Am 1992;6(3):705–17.

9. Johnson JT. Changing concepts in the management of deepneck infection. AAO-HNS Bulletin 1992;11:17–21.

10. Marra S, Hotaling AJ. Deep neck infections. Am J Otolaryngol1996;17:287–98.

11.Blomquist IK, Bayer AS. Life-threatening deep fascial spaceinfections of the head and neck. Infect Dis Clin North Am1988;2(1):237–64.

12. Chowdhury K, Bloom J, Black M, et al. Spontaneous andnon-spontaneous internal jugular vein thrombosis. HeadNeck 1990;12:168–73.

13. Kim HJ, Park ED, Kim JH, et al. Odontogenic versusnondontogenic deep neck space infections. J Comput AssistTomogr 1997;21(2):202–8.

14. Weber AL, Siciliano A. CT and MR imaging evaluation ofneck infections with clinical correlations. Rad Clin of NorthAm 2000;38(5):941–68.

15. Bielamowicz SA, Stroper IS, Jabour BA, et al. Spaces andtriangles of the head and neck. Head Neck 1994:16:383-7.

16. Shah RR, Lwein JS. Imaging of the infrahyoid neck.Neuroimaging Clin North A

17. Weissman JL, Carrau RL. Anterior facial vein andsubmandibular gland together: Predicting the histology ofsubmandibular masses with CT or MR imaging. Radiology1998;208:441–6.

18. Wang LF, Kuo WR, Lin CS, et al. Space infection of thehead and neck. Kaohsiung J Med Sci. 2002;18(8):386–92.

[19] Gianoli GJ, Espinola TE, Guarisco JL, et al. Retropharyngealspace infection: changing trends. Otol Head Neck Surg.1991;105:92–100.

20. Moon WK, Han MH, Chang KH, et al. CT and MR imagingof head and neck tuberculosis. Radiographics1997;17(2):391–402.

21. Lee KC, Tami TA, Lalwani AK, et al. Contemporarymanagement of cervical tuberculosis. Laryngoscope1992;102(1):60–4.

22. Vural C, Gungor A, Comerci S. Accuracy of computerizedtomography in deep neck infection of pediatric population.Am J Otolaryngol 2003;24(3):143–8.

Documents

Deep Neck Spaces Infections in Radiology