Imaging of the Paranasal Sinuses

pdf path

Image Gallery

Sinonasal inflammatory disease is extremely common, with nearly all individuals experiencing symptoms at some point due to infection, allergy or other causes. The vast majority of people do not require imaging, but those who do are often unresponsive to medical therapy or may have repeated episodes of sinusitis, or chronic sinusitis. Other reasons to image include an abnormal physical exam, such as the presence of a polyp or other mass, or an underlying condition that predisposes one to sinusitis, such as an immunocompromised state.1 Patients who have known or suspected sinusitis, and signs or symptoms that suggest a complication such as intracranial abscess or orbital cellulitis, may undergo imaging of the suspected site. In addition, patients planning for surgical intervention typically undergo preoperative imaging.

Imaging Techniques

While occasionally ordered by referring physicians, radiographs are limited in their depiction of sinonasal disease processes. Computed tomography (CT) is the workhorse modality in paranasal imaging. In the past, single-detector CT required patients to be imaged twice – once axially in the supine position and then coronally in the prone position with the neck extended. While this technique could determine the mobility of intrasinus material, it was uncomfortable for patients and was susceptible to significant streak artifact from dental hardware. Modern-day multidetector CT allows for patients to be scanned more rapidly in the supine position and then have the images reconstructed in the sagittal and coronal planes, which minimizes dental artifact. However, this makes assessment for air fluid levels and mobility of secretions more difficult.

CT is optimal for evaluating sinus anatomy and drainage pathways, as well as the surrounding bony structures. It can readily identify soft tissue and orbital extension. MRI is a secondary complementary modality and is particularly useful in cases in which a mass or intracranial extension is suspected. MRI can also be useful to assess extent of soft tissue or bony involvement, especially when CT is inconclusive. PET/CT (positron emission tomography/CT) is also a secondary complementary modality in sinonasal disease, typically used in cases of malignancy to assess for residual or recurrent tumor or metastatic spread. American College of Radiology (ACR) appropriateness criteria is useful in determining which patients should undergo sinonasal imaging and which modality is preferred.1


When evaluating the paranasal sinuses, it is useful to divide them into anterior and posterior. The anterior sinuses include the frontal, maxillary and anterior ethmoid air cells (Figure 1), and the posterior ethmoid and sphenoid sinuses make up the posterior sinuses (Figure 2). At birth, the maxillary sinuses are quite small, and they continue developing for about 15 years. Similarly, small ethmoid sinuses are present at birth and develop into late puberty. The sphenoid sinuses are the next to develop and generally aerate by age 3. The frontal sinuses are last to develop and are most variable. They are usually visible by age 6 and develop into adult years. Approximately 4% of the population has no frontal sinuses, and a much larger percentage has hypoplastic frontal sinuses.2

The anterior sinuses drain into the middle meatus. The frontal sinuses drain via the frontal sinus drainage pathway or nasofrontal duct (Figure 3), the maxillary sinuses drain via the ostiomeatal unit, and the anterior ethmoid air cells drain via the infundibulum. The ostiomeatal unit (OMU) is bordered by the bony uncinate process, which delineates the air/drainage passages of the maxillary sinus ostium, infundibulum, and hiatus semilunaris (Figure 4). Opacification of the OMU has been shown to correlate with an increased likelihood of active sinusitis. Therefore, the OMU is the major target of surgery, such as functional endoscopic sinus surgery (FESS), to reduce episodes of recurrent sinusitis.3

The posterior sinuses drain into the superior meatus through the sphenoethmoidal recesses (Figure 2). The naso- lacrimal duct is the only passage that drains into the inferior meatus.

Additional bony structures important to sinonasal anatomy include the fovea ethmoidalis, lateral lamella, cribriform plate, crista galli, perpendicular plate, nasal septum, lamina papyracea, and the nasal turbinates (Figure 5). The fovea ethmoidalis represents the bony roof of the ethmoid air cells. It connects to the cribriform plate via the lateral lamella. The crista galli protrudes superiorly in the midline from the cribriform plate. Along its inferior aspect, the cribriform plate connects with a thin vertically oriented bone, known as the perpendicular plate, which connects inferiorly to the nasal septum.

Within the nasal cavity are the turbinates — superior, middle and inferior, in ascending order of size. The lamina papyracea is quite thin and represents the lateral border of the ethmoid sinuses and medial wall of the orbit.

Anatomic Variants

Several variants of the paranasal sinuses are of primary importance in surgical planning. Also important, but less so, is that they can increase the likelihood of developing sinus disease. Ethmoid air cell variants include a prominent agger nasi cell, type I-IV frontal cells, prominent ethmoid bullae, Haller cell, and Onodi cell. A common variant of the middle turbinate is called a concha bullosa, which describes pneumatization and enlargement of the middle turbinate.

Much confusion exists regarding the agger nasi cell. The agger nasi cell describes the anterior-most ethmoid air cell when it lies anterolateral and inferior to the fronto-ethmoidal recess, and anterior and superior to the attachment of the middle turbinate. It is essentially a normal finding seen in most everyone. However, when prominent, it can predispose one to sinus disease.3

The frontal air cell variants are related to the agger nasi cell and have 4 types. Type I is a single cell located superior to the agger nasi cell (Figure 6). Type II is a tier of cells above an agger nasi cell that may “invade” into the frontal sinus. Type III is a single massive cell that invades into the frontal sinus and attaches to the anterior table. Type IV is a single isolated cell in the frontal sinus with no obvious connection to the frontal recess.4 A Haller cell is an ethmoid air cell that projects along the inferomedial aspect of the orbit and can predispose to ostiomeatal unit obstruction (Figure 7). Lastly, the Onodi, or sphenoethmoid, cell is the posterior-most ethmoid air cell when lateral and superior to the sphenoid sinus. The presence of an Onodi cell is of clinical importance, as its accidental surgical violation presents a significant risk for optic canal violation and optic nerve damage.3

Pneumatization of various bony structures can be seen, including the crista galli, pterygoid processes and clivus. Deviation of the nasal septum is extremely common and may be considered an anatomic variation as well. It can predispose one to nasal and OMU obstruction, and sinusitis and can be exacerbated by an osseous spur.

A pseudo-anatomic variant of the paranasal sinuses is an encephalocele, or herniation of intracranial contents through a congenital or acquired defect in the cribriform plate (Figure 8). The herniated substance could be neural tissue (encephalocele), cerebrospinal fluid (meningocele), or both (meningoencephalocele). The greater the depth of the lateral lamella, typically more than 8 mm, the greater the risk of injury or defect to the cribriform plate. If present, MRI can be useful in determining the presence of brain tissue herniation.5

Paranasal Sinus Pathologies

Common benign paranasal sinus pathologies include inflammatory mucosal disease, mucocele, osteoma, noninvasive fungal sinusitis (allergic or mycetoma), mucous retention cysts, and polyps. Inflammatory mucosal disease, which simply describes sinus mucosal thickening, is seen in both asymptomatic patients and those with acute sinusitis (Figure 9). In the absence of adequate clinical history, differentiating between the two can be difficult. Certain coexisting findings, such as the presence of air-fluid levels, tend to favor an acute process. However, this finding is nonspecific and may also be seen in other circumstances, such as nasogastric tube insertion and prolonged supine positioning.3 Mucosal thickening can also affect the nasal turbinates, often asymmetrically. Usually this occurs in a normal cyclical pattern, although nasal obstruction can result depending on the degree of thickening.

In many cases, the cause of sinonasal inflammatory disease may not be readily apparent and is likely multifactorial due to an allergy and/or infectious agents with a component of obstruction. There are instances, however, in which the cause of sinonasal inflammation is more readily identifiable on imaging, such as obstructive anatomic variant or mass, surgical hardware, or odontogenic sinusitis. Odontogenic sinusitis occurs secondary to the intimate relationship between the maxillary teeth and maxillary sinuses, allowing for direct spread of infection. This can be from a discrete lesion, such as a periapical abscess (Figure 10), or could be due to bony dehiscence and direct communication between the oral cavity and maxillary sinus.6

Complications of sinonasal inflammatory disease include bony wall thickening or osteitis from chronic inflammation (Figure 11), mucocele formation, or orbital or intracranial involvement (Figure 12). A mucocele is an obstructed and dilated mucus-filled sinus (Figure 13). While it can occur in any sinus, the frontal sinus is most common.3 Surgical intervention is usually required.

Noninvasive fungal sinusitis is characterized as sinus opacification with high-density secretions on CT. This, however, is not specific and can also be seen with chronic inspissated, or desiccated, secretions (Figures 14 and 15) with subsequent calcification; therefore, a clinical workup is required for differentiation (i.e., IgE and allergen testing).7

Osteomas of the paranasal sinuses are relatively common. They are bone density and are most commonly located within the frontal and ethmoid sinuses (Figure 16). While there is an increased risk of sinus obstruction and resultant sinusitis with osteomas, the vast majority are asymptomatic.8

Mucous retention cysts, or dilated secretory glands, appear as round, dome-shaped, soft-tissue densities in the maxillary sinus that are typically found incidentally, similar to polyps. They are less common in the other sinuses and typically do not require additional workup or follow-up.9 Polyps, on the other hand, may become more extensive and problematic, leading to obstruction and osseous remodeling. The presence of diffuse sinus and nasal cavity opacification leads one to consider sinonasal polyposis. Infundibular enlargement is almost always present.10 The opacification typically represents a combination of the polyps and the resultant obstructive inflammation (Figures 14 and 15).

A more severe form of nasal polyposis is intracranial or massive polyposis (Figures 15). This can be seen in a condition called Samter’s triad, which is seen in patients in their third to fourth decades of life. It is associated with asthma and aspirin sensitivity, and patients may have massive polyps extending into the orbits or intracranially.11

A polyp seen most often in the third through fifth decades of life that arises in the maxillary sinus and extends through the ostium into the nasal cavity, and commonly through the choana into the nasopharynx, is called an antrochoanal polyp (Figure 17). Antrochoanal polyps are considered benign and may cause smooth osseous remodeling without destruction.3

Conversely, an inverted (or inverting) papilloma can be locally aggressive. It typically arises from the lateral nasal wall and can protrude into the sinus and nasal cavity, often causing obstruction. Inverted papillomas require surgical resection due to locally aggressive behavior and imaging features that overlap with squamous cell carcinoma. In addition, approximately 5% of inverted papillomas harbor squamous cell carcinoma (Figure 18).12

Other aggressive paranasal sinus pathologies include invasive fungal sinusitis, Wegener’s granulomatosis and neoplasms. These are seen as soft-tissue densities within the paranasal sinuses, causing bony destruction. Invasive fungal sinusitis (Figures 19 and 20) is an aggressive infection in immunocompromised patients that will typically involve the adjacent paranasal soft tissues, orbits and even extend intracranially. At times, invasive fungal sinusitis may develop too quickly to destroy bone. Wegener’s granulomatosis is an idiopathic necrotizing granulomatous vasculitis. Early on it appears as nonspecific sinonasal inflammation. As the disease progresses, there is often bony destruction, first involving the nasal septum (Figure 21).3 There is often simultaneous involvement of the lower respiratory tract, kidneys, skin and joints, so history is critical.

Neoplasms commonly seen in the sinonasal cavities include squamous cell carcinoma (Figure 22), lymphoma, sarcoma (Figure 23), salivary gland tumors and metastases. Squamous cell carcinoma is the most common sinonasal malignancy. It is more common in men over 60, and the majority of cases occur in patients with a history of tobacco use.13 On MRI, squamous cell carcinoma enhances following contrast administration and typically demonstrates bony destruction. No reliable imaging features differentiate it from inverted papilloma or the other sinonasal neoplasms.


Two types of surgery have traditionally been used to treat sinonasal inflammatory disease. The older method is the Caldwell Luc procedure, which requires an approach through the anterior maxillary sinus wall. The mucosa of the maxillary sinus is then stripped in its entirety. A drainage pathway is then created for maxillary sinus contents through the lateral wall of the nasal cavity into the inferior meatus. On CT, common findings after a Caldwell Luc include anterior and medial wall bony defect (Figure 21), osseous wall thickening, and collapse of affected maxillary sinus.14 Due to complication rates and the radical nature of the procedure, FESS was developed. First performed in the United States in the mid-1980s, FESS has supplanted Caldwell Luc as the preferred surgical method to treat sinonasal inflammatory disease. It involves the placement of an endoscope through the nasal passages and removal of the sinus walls and septations to open/create drainage pathways for inflammatory/infectious debris (Figure 24). Surgical planning CT studies are often used for guidance, making it important to delineate the anatomic variants described above.


Fractures of the sinus walls most commonly involve the maxillary and ethmoid sinuses. Orbital blowout fractures are the result of blunt trauma to the orbit. Due to weakness caused by the presence of the infraorbital canal, these fractures can go through the orbital floor into the maxillary sinus. The medial orbital wall or lamina papyracea is also a weak point, allowing fracture into the ethmoid air cells (Figure 25). In both cases, there is a risk of extraocular muscle herniation through the fracture defect with or without muscle entrapment, resulting in muscle dysfunction and diplopia.3 Fractures through the lateral wall of the maxillary sinus often occur as part of a zygomaticomaxillary complex or tripod fracture (Figure 26). The lateral orbital wall and zygomatic arch are also typically involved in this fracture complex.

Less common are fractures through the frontal sinus and sphenoid sinus. Fractures through the frontal bone can involve the walls of the frontal sinus or less commonly the fovea ethmoidalis and cribriform plate. In fractures through both the anterior and posterior walls of the frontal sinus, there is a risk of intracranial infection. Fractures through the fovea ethmoidalis and cribriform plate run the risk of encephaloceles and cerebrospinal fluid leaks. Fractures involving the walls of the sphenoid sinus typically occur in severe skull base trauma. In those instances, the carotid canal must be examined for injury due to its intimate relationship with the lateral walls of the sphenoid sinus. CT angiography may be considered to evaluate for internal carotid artery injury.15


Evaluation of the paranasal sinuses requires detailed knowledge of the anatomy and common anatomic variants. The anatomic variants are not only important in that they predispose one to sinus inflammatory disease, they are important for surgical mapping. While it may be difficult to differentiate asymptomatic sinus opacification from acute sinusitis without a clinical history, imaging can greatly aid in directing the course of treatment for sinusitis by unmasking alternative causes such as obstructing lesions or odontogenic sources. Imaging is also important in delineating complications of sinusitis, such as orbital or intracranial involvement. Finally, benign and aggressive sinonasal masses, postsurgical changes, and post-traumatic injuries can be readily recognized based on their characteristic patterns.


  1. Cornelius RS, Martin J, Wippold FJ, et al. ACR appropriateness criteria sinonasal disease. J Am Coll Radiol 2013;10(4):241-246.
  2. Harnsberger R. Handbook of head and neck imaging. 2nd edition. St Louis, MO: Mosby; 1995:340-395.
  3. Som P, Shugar, J, Brandwein, M. Sinonasal cavities. In: Som P, Curtin H. Head and Neck Imaging, 4th edition. St Louis, MO: Mosby; 2003:1-438.
  4. Tuncyurek O, Songu M, Adibelli ZH, et al. Frontal infundibular cells: pathway to the frontal sinus. Ear Nose Throat J 2012;91(3):E29-32.
  5. Hedlund G. Congenital frontonasal masses: developmental anatomy, malformations, and MR imaging. Pediatr Radiol 2006;36(7):647-662.
  6. Mehra P, Murad, H. Maxillary sinus disease of odontogenic origin. Otolaryngol Clin North Am 2004;37(2): 347-364.
  7. Fatterpekar GM, Delman BN, Som PM. Imaging the paranasal sinuses. Anat Rec 2008;291(11):1564-1572.
  8. Strek P, Zagólski O, Składzień J, et al. Osteomas of the paranasal sinuses. Med Sci Monit 2007;13(5): CR244-250.
  9. Wang, JH, Jang YJ, Lee BJ. Natural course of retention cysts of the maxillary sinus: long-term follow-up results. Laryngoscope 2007;117(2):341-344.
  10. Drutman J, Babbel RW, Harnsberger HR, et al. Sinonasal polyposis. Semin Ultrasound CT MR 1991;12(6):561-574.
  11. Majithia A, Tatla T, Sandhu G, et al. Intracranial polyps in patients with Samter’s triad. Am J Rhinol 2007;21(1):59-63.
  12. Mendenhall WM, Hinerman RW, Malyapa RS, et al. Inverted papilloma of the nasal cavity and paranasal sinuses. Am J Clin Oncol 2007;30(5):560-563.
  13. Michel J, Fakhry N, Mancini J, et al. Sinonasal squamous cell carcinomas: clinical outcomes and predictive factors. Int J Oral Maxillofac Surg 2014; 43(1):1-6.
  14. Nemec SF, Peloschek P, Koelblinger C, et al. Sinonasal imaging after Caldwell-Luc surgery: MDCT findings of an abandoned procedure in times of functional endoscopic sinus surgery. Eur J Radiol 2009;70(1):31-34.
  15. York G, Barboriak D, Petrella J, et al. Association of internal carotid artery injury with carotid canal fractures in patients with head trauma. AJR Am J Roentgenol 2005;184(5):1672-1678.
Back To Top

Kazmi KS, Shames JP.  Imaging of the Paranasal Sinuses.  J Am Osteopath Coll Radiol.  2015;20(7):27.

About the Author

Khuram S. Kazmi, M.D., Jason P. Shames, M.D.

Khuram S. Kazmi, M.D., Jason P. Shames, M.D.

Department of Radiologic Sciences, Hahnemann University Hospital / Drexel University College of Medicine, Philadelphia, PA


Copyright © The American College of Osteopathic Radiology 2017