EDUCATION EXHIBIT Failed Endoscopic Sinus Surgery: Spec ...€¦ · Introduction Sinusitis is a...

177EDUCATION EXHIBIT

Benjamin Y. Huang, MD • Kristen M. Lloyd, MD • John M. DelGaudio, MD • Eric Jablonowski, AS • Patricia A. Hudgins, MD

Since its introduction over 2 decades ago, functional endoscopic si-nus surgery (FESS) has revolutionized the surgical management of chronic sinusitis. Performed over 200,000 times annually in the United States to treat medically refractory sinusitis, FESS has success rates as high as 98%. When surgical failure occurs, it is typically due to postoperative scarring or unaddressed outflow tract obstruction in the region of the frontal recess. The most common causes of surgical fail-ure in the frontal recess include remnant frontal recess cells, a retained uncinate process, middle turbinate lateralization, osteoneogenesis, scarring or inflammatory mucosal thickening, and recurrent polyposis. Computed tomography (CT) of the paranasal sinuses has become in-dispensable in evaluation of patients with FESS failure, particularly in the frontal recess, a location that can be difficult to visualize at endos-copy. Familiarity with the complex anatomy of the frontal recess and knowledge of the most common causes of surgical failure are essential for proper interpretation of sinus CT images obtained in patients be-ing considered for revision FESS of the frontal sinus. ©RSNA, 2009 • radiographics.rsnajnls.org

Failed Endoscopic Sinus Surgery: Spec-trum of CT Findings in the Frontal Recess1

ONlINE-ONly CME

See www.rsna .org/education /rg_cme.html

lEARNING OBJECTIVESAfter reading this article and taking the test, the reader

will be able to:

Discuss the anat- ■

omy of the frontal recess and the types of accessory frontal recess cells that may contribute to frontal recess stenosis.

Describe com- ■

monly performed endoscopic proce-dures used to treat frontal sinusitis and their postoperative CT appearances.

List common CT ■

findings associated with failed endo-scopic sinus surgery in the frontal recess.

Abbreviation: FESS = functional endoscopic sinus surgery

RadioGraphics 2009; 29:177–195 • Published online 10.1148/rg.291085118 • Content Codes: 1From the Department of Radiology, University of North Carolina School of Medicine, 101 Manning Dr, CB 7510, Chapel Hill, NC 27599-7510 (B.Y.H.); and the Departments of Radiology (K.M.L., E.J., P.A.H.) and Otolaryngology (J.M.D., P.A.H.), Emory University School of Medicine, At-lanta, Ga. Presented as an education exhibit at the 2007 RSNA Annual Meeting. Received May 2, 2008; revision requested June 2 and received July 30; accepted August 1. P.A.H. is a shareholder in Amirsys; all other authors have no financial relationships to disclose. Address correspondence to B.Y.H. (e-mail: [email protected]).

©RSNA, 2009

See last page

TEACHING POINTS

Note: This copy is for your personal non-commercial use only. To order presentation-ready copies for distribution to your colleagues or clients, contact us at www.rsna.org/rsnarights.

178 January-February 2009 radiographics.rsnajnls.org

IntroductionSinusitis is a common medical problem in the United States, affecting 14%–16% of adults and accounting for approximately 11.6 million office-based outpatient visits annually (1,2). The current mainstay of treatment for sinusitis con-tinues to be medical therapy, and the majority of patients diagnosed with sinusitis respond ade-quately to a combination of antibiotics, deconges-tants, mucolytics, and steroids (3). However, in a significant proportion of patients with sinusitis, medical management alone is insufficient to re-lieve symptoms, necessitating referrals to rhinolo-gists for consideration of surgical management.

First described over 2 decades ago (4), func-tional endoscopic sinus surgery (FESS) has be-come the treatment of choice for patients with medically refractory rhinosinusitis. FESS pro-cedures are now performed more than 200,000 times per year in the United States (5), with published success rates of 76%–98% for primary FESS and 65%–78% for revision cases (6). Al-though the majority of patients who undergo FESS for chronic sinusitis experience significant symptomatic relief, up to 23% of patients ulti-mately require revision surgery for continued or recurrent sinus symptoms after initial surgery (7,8).

In patients presenting with sinusitis after FESS, the frontal sinus outflow tract is the re-gion where disease recurrence is most likely to occur (9). In addition, most medically refractory disease of the frontal sinuses can be attributed to obstruction at the level of the frontal recess (10). The frontal recess is a notoriously difficult area to treat with endoscopy owing to its anterior location and its tight confines between the orbit and anterior skull base. Furthermore, the frontal recess has a significant predilection for stenosis after FESS (11). The prevalence of frontal si-nusitis after FESS has not been established, but published series have reported persistent frontal sinusitis symptoms at short-term follow-up in 2%–11% of post-FESS patients (12) and long-term failure rates of 15%–20% (11), with up to 11% of these patients ultimately requiring revi-sion surgery (12,13).

Frontal sinusitis after FESS can pose a num-ber of challenges to the sinus surgeon, not the least of which is elucidating the cause of surgical

failure. Evaluation of patients in whom FESS has failed typically includes computed tomog-raphy (CT) of the paranasal sinuses to identify potential causes of sinus outflow tract stenosis. Radiologists tasked with interpreting these scans need to be familiar with the complex anatomy of the frontal recess and the processes that may contribute to surgical failure in order to generate accurate and meaningful reports for the referring rhinologists.

The goal of this article is to review several of the most salient issues related to CT evaluation of patients with recurrent frontal sinus disease after FESS. Specific topics discussed include scanning technique, with a focus on the utility of multiplanar reformation; anatomy of the frontal recess and its many variants, including frontal re-cess cells; common endoscopic procedures used to treat disease in the frontal sinus and frontal re-cess; and common entities contributing to failure of FESS.

CT TechniqueTiming is critical when imaging patients with chronic sinusitis. The presence of mucosal thick-ening is a nonspecific finding and can be seen in 90% of individuals in the presence of simple viral upper respiratory tract infections. Furthermore, changes due to simple cases of acute maxillary sinusitis may be evident for 4–6 weeks. Therefore, CT of the paranasal sinuses should not be per-formed until 4–6 weeks after initiation of medical therapy, and scanning should be delayed in pa-tients with acute upper respiratory infections (9).

Before the widespread availability of spiral and, more recently, multidetector CT, direct coronal scanning was preferred because the coronal plane best demonstrates the anterior os-tiomeatal unit and skull base and has traditionally been thought to replicate most closely the actual view of the endoscopist. Recently, direct coronal scanning has fallen out of favor because modern multidetector CT scanners are capable of acquir-ing sections only a few tenths of a millimeter in thickness, allowing data to be acquired axially and then reconstructed with exquisite detail in any plane. Advantages of axial scanning include improved patient comfort during scanning and avoidance of artifacts created by dental restora-tions (which frequently plague scans acquired coronally). Furthermore, raw data from thin-sec-tion axial scanning are now commonly used for intraoperative stereotactic guidance systems.

TeachingPoint

RG ■ Volume 29 • Number 1 Huang et al 179

The use of multiplanar reformation (sagittal and coronal) has been shown to improve pre-operative understanding of the frontal recess. Specifically, review of sagittal images significantly improves one’s ability to identify and measure the frontal recess (14) and is critical in assessment of obstructing anterior ethmoid cells (14,15). Kew et al (14) found that preoperative review of both sagittal and coronal reformatted images together significantly altered surgical planning in over one-half of cases when compared with review of coronal scans alone.

We advocate thin-collimation contiguous heli-cal scanning with a maximum section thickness of no greater than 1 mm. Scans should be recon-structed in the axial, coronal, and sagittal planes by using both soft-tissue and high-resolution-algorithm bone windows with a reconstruction thickness of 3 mm or less. If available, transfer of raw data to a workstation is also extremely beneficial because it allows end users to further adjust parameters such as image brightness, con-trast, section thickness, and section plane, which can be particularly useful for teasing out complex bony anatomy.

Anatomy of the Frontal Recess and Common Variants

Familiarity with the complex and highly vari-able anatomy of the frontal recess is critical to its evaluation at CT. Along with the regions of the ethmoid infundibulum–middle meatus and the

sphenoethmoidal recess, the frontal recess rep-resents one of the three anatomic “tight spots” that are implicated as sites of obstruction leading to sinusitis. At its most basic level, the frontal recess can be conceptualized as an inverted fun-nel within the anterior ethmoid complex through which the frontal sinus drains. The tip or apex of the funnel lies at the frontal sinus ostium, which can be easily identified on sagittal CT images as a “waist” located at the level of the nasofrontal process, which demarcates the level of the floor of the frontal sinus (Fig 1). The frontal recess typi-cally flares out inferiorly and posteriorly to form the wider opening of the funnel. Taken together, the funnel-shaped inferior portion of the frontal sinus (commonly referred to as the frontal in-fundibulum), the frontal ostium, and the frontal recess make up the frontal sinus outflow tract (10).

The boundaries of the frontal recess are typi-cally formed by the agger nasi cell anteriorly; the lamina papyracea laterally; the most anterior and superior portion of the middle turbinate medially; and the ethmoid bulla, its associated bulla la-mella, and the suprabullar cell (if present) poste-riorly (Fig 1). From the frontal recess, secretions drain to the middle meatus of the nasal cavity via one of two routes: In approximately 40% of cases, secretions from the frontal recess drain into the ethmoid infundibulum and subsequently into

Figure 1. Normal frontal recess anatomy. Coronal (a) and sagittal (b) CT images show the right frontal re-cess (dotted red line), which is bounded anteriorly and laterally by an agger nasi cell (white arrow) and a type 1 frontal cell (black arrow), medially by the middle turbinate, and posteriorly by the ethmoid bulla and bulla lamella. The nasofrontal process (arrowhead in b) forms the floor of the frontal sinus and demarcates the level of the frontal sinus ostium.

TeachingPoint

TeachingPoint


The frontal recess may be pneumatized by various anterior ethmoid cells, which are collec-tively known as frontal recess cells. These cells are normal anatomic variations that are present in some combination in most individuals. The clini-cal relevance of frontal recess cells lies in their potential for causing frontal sinusitis by obstruct-ing frontal sinus outflow at the level of the frontal recess (18). Any endoscopic surgical procedure aimed at clearing frontal recess obstruction must

the middle meatus through the hiatus semilunaris (16,17). In the remaining cases, the frontal recess drains into the middle meatus. On rare occasions, when the bulla lamella does not extend to the skull base, the frontal recess may drain directly into the suprabullar recess (10).

Figure 2. Agger nasi cell. Coronal (a) and parasagittal (b) drawings show the location of the agger nasi cell (blue area), which forms part of the anterior and lateral boundaries of the frontal recess.

Types of Frontal Recess Cells

Cell Type DescriptionBest Planes for Viewing

Frontal cell Type 1 Single cell above the agger nasi cell; it does not extend into the frontal sinus,

and its posterior wall is a free partition in the frontal recessCoronal,

sagittal Type 2 A tier of two or more cells above the agger nasi cell; its posterior wall is a free

partition in the frontal recessCoronal,

sagittal Type 3 Single large cell above the agger nasi cell; it extends into the true frontal sinus,

and its posterior wall is a free partition in the frontal sinus or recessCoronal,

sagittal Type 4 Isolated cell in the frontal sinus; its anterior or inferior margin is the anterior

table or floor of the frontal sinus, and its posterior wall is a free partition in the frontal sinus

Coronal, sagittal

Supraorbital eth-moid cell

A cell extending over the orbit from the frontal recess; its posterior wall is the skull base, and it may mimic a septated frontal sinus

Axial, coronal

Frontal bullar cell A cell above the ethmoid bulla pneumatizing into the frontal sinus; its superior wall is the skull base, and its anterior wall extends into the frontal sinus

Sagittal

Suprabullar cell A cell above the ethmoid bulla; its superior wall is the skull base, and its ante-rior wall does not extend into the frontal sinus

Sagittal

Inter–frontal sinus septal cell

A pneumatized frontal sinus septum; it is associated with a pneumatized crista galli

Axial, coronal


bullar cells, suprabullar cells, and inter–frontal sinus septal cells (Table) (18). These cells can be further categorized into one of three major groups based on their locations relative to the frontal sinus (14).

Frontal cells, along with the agger nasi cell, constitute the anterior group of frontal recess cells. Frontal cells are present at CT in 20%–33% of pa-tients, making them extremely common anatomic variants (19,22). The anterior boundaries of these cells are made up of the anterior wall of the fron-tal recess or the frontal sinus, and these cells do not extend posteriorly to abut the skull base (14). There are four types of frontal cells described un-der the system known as the Kuhn classification (18).

Type 1–3 frontal cells all sit atop the agger nasi cell. These cells are best demonstrated on coronal and sagittal CT images (14). Type 1 frontal cells, seen in up to 37% of frontal recess sides (14,19), are defined as single anterior ethmoid cells within the frontal recess sitting above the agger nasi cell. These cells do not extend into the frontal sinus (Figs 1, 3).

Type 2 cells, seen in up to 19% of frontal recesses, are defined as a tier of two or more an-terior ethmoid cells sitting above the agger nasi cell (14,19) (Fig 4). Type 3 cells are defined as a single massive cell sitting above the agger nasi cell and pneumatizing into the frontal sinus (Figs 5, 6); type 3 cells are seen in approximately 6%–8% of frontal sinus sides (14,19).

address these variant cells; failure to do so may result in surgical failure. For this reason, the ra-diologist’s report must accurately characterize any frontal recess cells present by using standard accepted nomenclature. In the remainder of this section, we briefly review the classification of frontal recess cells.

Agger Nasi CellAgger nasi is a Latin term literally meaning “nasal mound” (10). At rhinoscopy, the agger nasi ap-pears as an eminence located on the lateral nasal wall at the leading edge of the middle turbinate; it represents the intranasal portion of the frontal process of the maxilla (Fig 2). As noted earlier, the agger nasi serves as the anterior limit of the frontal recess. Pneumatization of the agger nasi (resulting in the so-called agger nasi cell) occurs in 78%–98.5% of individuals (14,17,19–21). When present, agger nasi cells are considered the most anterior of all ethmoid cells (21) and can pneumatize posteriorly to narrow the frontal recess (10). Coronal and sagittal reformatted CT images are most helpful in identifying the agger nasi cell. On coronal images, the agger nasi ap-pears as a laterally placed sinus below the frontal sinus and anterior to the middle turbinate. Sagit-tal images demonstrate the anterior location of the air cell (Fig 1).

Other Frontal Recess CellsIn addition to agger nasi cells, there are several other named frontal recess cells, which include frontal cells, supraorbital ethmoid cells, frontal

Figure 3. Type 1 frontal cell. Coronal (a) and parasagittal (b) drawings show a type 1 frontal cell (blue area) sitting atop an agger nasi cell anteriorly in the frontal recess.

TeachingPoint


Figure 4. Type 2 frontal cells. (a, b) Coronal (a) and parasagittal (b) drawings show a tier of type 2 frontal cells (blue areas) sitting atop an agger nasi cell. (c, d) Coronal (c) and sagittal (d) CT images show a tier of two type 2 frontal cells (arrows) sitting directly atop an agger nasi cell (*).

Figure 5. Type 3 frontal cell. Coronal (a) and parasagittal (b) drawings show a type 3 frontal cell (blue area) sitting atop an agger nasi cell. The type 3 cell extends superiorly from the frontal recess through the frontal ostium and into the frontal sinus.


dered anteriorly by the anterior frontal sinus table, with their posterior walls representing free partitions in the frontal sinus (Fig 7). Recogni-tion of type 4 cells often requires both coronal

The rare type 4 frontal cell, seen in only 2.4% of frontal recesses (19), is unique among the frontal cells in that it does not abut the agger nasi cell. Rather, type 4 cells are defined as isolated air cells located within the frontal sinus, bor-

Figure 6. Type 3 frontal cell and frontal bullar cell. Sagittal CT image obtained through the frontal recess shows a type 3 frontal cell (white ar-row) sitting above an agger nasi cell (*) and extending superiorly into the fron-tal sinus. Also note the frontal bullar cell (black arrow), which pneumatizes along the skull base from the posterior frontal recess into the frontal sinus.

Figure 7. Type 4 frontal cell. (a, b) Coronal (a) and parasagittal (b) drawings show a type 4 frontal cell (blue area) situated entirely within the right frontal sinus and bordered by the anterior frontal sinus wall. The type 4 cell does not abut the agger nasi cell. (c, d) Coronal (c) and sagittal (d) CT images show an opacified type 4 frontal cell (arrow) in the frontal sinus.


and the frontal sinus. They typically drain into the lateral aspect of the frontal recess. Up to 15% of adults may have one or more supraorbital eth-moid cells (23), with approximately 5% of frontal sinuses having multiple supraorbital cells (14). Supraorbital ethmoid cells can obstruct frontal sinus drainage; preoperative identification is es-sential because these cells can be readily mistaken for the frontal ostium during endoscopic dissec-tion (23).

Identification of these cells at CT requires review of both axial and coronal images (Fig 8) (14). Supraorbital ethmoid cells may mimic the appearance of a septated frontal sinus or may give the appearance of multiple frontal sinuses. On axial images, the supraorbital ethmoid cell is situ-ated behind the frontal sinus. On coronal images, what appears to be the lateral compartment of a septated frontal sinus is actually a separate su-praorbital ethmoid cell located posterior and lat-eral to the frontal sinus. The supraorbital ethmoid cell ostium can be found anterior to the canal for the anterior ethmoid artery on sequential coronal images. Sagittal reformation can occasionally be helpful in characterizing supraorbital cells (23).

Frontal bullar cells represent pneumati-zation of the anterior skull base in the poste-rior frontal recess with extension into the true

and sagittal CT reformation. On sagittal images, these cells have been described as having the ap-pearance of a “balloon on a string,” with the cell itself representing the balloon and the narrow outflow tract of the cell representing the string (14). However, most often type 4 cells have no identifiable connection to the frontal recess at imaging. In patients with frontal sinus disease, an isolated aerated cell abutting the anterior sinus wall in an otherwise opacified frontal sinus (the so-called cell within a cell) usually represents a type 4 frontal cell (10). It has been observed that frontal mucosal thickening is more prevalent in patients with type 3 and type 4 frontal cells than in those without frontal cells (22).

Supraorbital ethmoid cells, frontal bullar cells, and suprabullar cells make up the posterior group of frontal recess cells. All of the cells in this group are located along the posterior wall of the frontal recess and are bordered posteriorly or superiorly by the anterior skull base (14).

Supraorbital ethmoid cells are anterior eth-moid air cells that extend superiorly and laterally over the orbit from the frontal recess. These cells represent pneumatization of the orbital plate of the frontal bone posterior to the frontal recess

Figure 8. Supraorbital ethmoid cell. (a) Coronal CT image obtained slightly posterior to the level of the frontal recess shows a supraorbital ethmoid cell (arrow) extending over the left orbit. The opening of this eth-moid cell is closely related anatomically to the canal for the anterior ethmoid artery (arrowheads). In this case, it would be difficult to distinguish this cell from posterior pneumatization of the frontal sinus on the basis of coronal images alone. (b) Axial CT image shows the supraorbital ethmoid cell (arrow), which is clearly differ-entiated from the frontal sinus (*) by a discrete bony septum.


Suprabullar cells are nearly identical to frontal bullar cells, with the only distinguishing feature being that suprabullar cells lie entirely below the level of the frontal sinus ostium and do not ex-tend into the frontal sinus (Fig 10). Like frontal bullar cells, suprabullar cells sit above the eth-moid bulla and form a portion of the posterior

frontal sinus. These cells lie above the ethmoid bulla (typically the largest anterior ethmoid cell, bordered posteriorly by the basal lamella of the middle turbinate) and, when present, define a portion of the posterior boundary of the frontal recess and frontal sinus. Frontal bullar cells are best demonstrated at sagittal CT, where they ap-pear as ethmoid cells sitting atop the ethmoid bulla and extending into the frontal sinus (14) (Figs 6, 9).

Figure 9. Frontal bullar cell. Parasagittal drawing shows a frontal bullar cell (blue area), which is situated along the posterior boundary of the frontal recess and pneumatizes into the frontal sinus.

Figure 10. Suprabullar cell. (a) Parasagittal drawing shows a suprabullar cell (blue area) situ-ated along the posterior boundary of the frontal recess. (b) Sagittal CT image obtained through the frontal recess shows a large suprabullar cell (arrow) sitting above the ethmoid bulla (*) and situated along the posterior aspect of the frontal recess. The cell is bordered superiorly by the skull base; however, unlike a frontal bullar cell, the suprabullar cell does not extend above the level of the frontal ostium into the frontal sinus.


plasty may be performed to provide better endo-scopic access and to improve antrostomy patency.

Most endoscopic procedures directed at the frontal sinus are revision surgeries performed

wall of the frontal recess (14). These cells are the CT correlate to the suprabullar recess (also known as the sinus lateralis), which appears as a cleft above the ethmoid bulla when viewed endo-scopically (14). Suprabullar cells are best demon-strated on sagittal CT images. During endoscopic frontal sinusotomy, both suprabullar and frontal bullar cells can be mistaken for the skull base; thus, failure to recognize their presence preopera-tively can result in incomplete surgical dissection (14).

The final group of frontal recess cells is the medial type, which is made up of the inter–fron-tal sinus septal cell. This cell represents pneuma-tization of the inter–frontal sinus septum; when extensive, such pneumatization can extend into the crista galli. These cells drain into the medial frontal recess and can impinge on the frontal sinus ostium (10,14). Axial and coronal are the best planes for demonstrating inter–frontal sinus septal cells (14) (Fig 11).

Common Endo- scopic Procedures

Used in the Frontal RecessAs mentioned earlier, the frontal sinus is the most difficult of the four paranasal sinuses to treat endoscopically owing to its location and com-plex anatomy. Serious complications can occur because of the proximity of the frontal recess to the anterior ethmoid artery, orbit, and anterior cranial fossa (11). In addition, the region of the frontal recess is extremely susceptible to postop-erative scarring, resulting in frontal recess steno-sis. For these reasons, surgical dissection in the region of the frontal recess is often avoided ini-tially, and primary surgery for anterior ethmoid sinus disease (which may or may not involve the frontal sinus) is generally directed at the anterior ostiomeatal complex. Typically, these surgeries consist of a combination of uncinectomy, anterior ethmoidectomy, and middle meatal antrostomy, which is generally sufficient to clear disease in the frontal sinus and frontal recess (6,24). Septo-

Figure 11. Inter–frontal sinus septal cell. (a) Coro-nal CT image obtained slightly anterior to the frontal recess shows an inter–frontal sinus septal cell (arrow), which arises from the frontal sinus septum. (b) Axial CT image shows the inter–frontal sinus septal cell (white arrow). Note how the frontal sinus ostium (black arrow) is narrowed. Also note the pneumatiza-tion of the crista galli (arrowhead).


severe complicated frontal sinus disease or in patients in whom a more conservative frontal sinusotomy has failed. Similar to the endoscopic frontal recess approach, the extended frontal sinusotomy removes any obstructing disease in the frontal recess region not addressed by prior surgeries. In addition to removing those struc-tures typically removed in a Draf type I surgery, the Draf type II procedure also involves enlarging the frontal sinus ostium by removing any fron-tal recess cells protruding into the frontal sinus or by resecting the frontal sinus floor between the lamina papyracea and nasal septum, includ-ing the anterior portion of the middle turbinate (24,26,27). At postoperative imaging, it can be difficult to distinguish this procedure from a Draf type I surgery. Sagittal reformatted images can be helpful in this regard, as they will demonstrate a more extensive ethmoid resection and removal of the frontal sinus floor (the nasofrontal beak) after Draf type II surgery (Fig 13).

in patients who have already undergone failed ostiomeatal complex surgery (24). A number of endoscopic frontal sinus drainage procedures have been developed, most of which are varia-tions on the classification system of Draf (24–27). In recent years, use of the Draf nomenclature has fallen out of favor, although it is still used by some rhinologists.

The endoscopic frontal recess approach, previ-ously referred to as the Draf type I frontal sinus-otomy, is the least invasive of these procedures and consists of removal of obstructing disease in-ferior to the level of the frontal ostium. This pro-cedure is indicated in patients with chronic fron-tal sinus disease and clinical evidence of outflow tract obstruction at the level of the frontal recess. The anterosuperior ethmoid cells (including the agger nasi cell), the uncinate process surrounding the frontal ostium, and any frontal recess cells are resected (Fig 12). Care is taken to ensure that the mucosa of the frontal recess and frontal ostium is preserved (26,27).

Draf type II procedures, or extended frontal sinusotomies, are performed in patients with

Figure 12. Endoscopic frontal recess approach (Draf type I procedure). Coronal (a) and parasagittal (b) drawings show the area of resection in green. This surgery consists of removal of obstructing structures, including anterosuperior ethmoid cells (agger nasi cell and any obstruct-ing frontal recess cells) and the uncinate process. The dissection does not extend above the frontal ostium, hence the nasofrontal beak (best seen on the sagittal image) remains.


septum, and anterior middle turbinates (Fig 14). These procedures are performed for the most severe cases of recalcitrant frontal sinus disease, as an alternative to frontal sinus obliteration (26–28).

During evaluation of paranasal sinus CT scans from patients being considered for frontal sinus FESS, the size of the frontal recess should be noted, as a small frontal recess limits the poten-tial for enlarging the frontal sinus outflow tract without performance of a more extensive frontal sinusotomy (25).

The endoscopic modified Lothrop procedure (formerly known as a Draf type III procedure and occasionally referred to as a median drain-age procedure) and the endoscopic transseptal frontal sinusotomy consist of contiguous bilateral enlargement of frontal sinus drainage. This is achieved by removal of the frontal sinus floor on both sides and removal of adjacent parts of the inferior inter–frontal sinus septum, superior nasal

Figure 13. Extended frontal sinusotomy (Draf type II procedure). (a, b) Coronal (a) and parasagittal (b) drawings show the area of resection in green. This procedure can be difficult to distinguish from the less inva-sive endoscopic frontal recess approach (Draf type I surgery) at postoperative CT. Unlike the Draf type I pro-cedure, Draf type II surgery includes resection of the frontal sinus floor and may extend into the frontal sinus, resulting in a less pronounced or absent nasofrontal beak on sagittal images (cf Fig 12). (c, d) Coronal (c) and sagittal (d) CT images show the typical postoperative appearance after an extended frontal sinusotomy. The posterior ethmoid cells have also been resected.


moval of the agger nasi cell or frontal recess cells (7,11). Not only do residual air cells obstruct the frontal recess, but they also serve as platforms for scar tissue to form (29). Published series have re-ported the presence of retained agger nasi cells in 49%–93% of patients undergoing revision FESS (6,19,31). Although they did not break down their results based on specific types of frontal re-cess cells, Chiu and Vaughan (29) reported find-ing residual agger nasi or ethmoid bulla remnants in 79% of cases and unopened supraorbital eth-moid or other frontal recess cells in 11.9% of pa-tients undergoing revision FESS, whereas Musy and Kountakis (6) reported incomplete anterior ethmoidectomies in 64% of patients undergoing revision FESS. DelGaudio et al (19) reported finding frontal cells (types 1–4) at CT in 21.9% of frontal recess sides in patients undergoing revi-sion FESS and intersinus septal cells in an addi-tional 11.4% of frontal recess sides.

CT is invaluable in the identification and char-acterization of frontal recess cells prior to revision FESS. Proper identification of frontal recess cells requires reviewing images in multiple imaging planes (Fig 15). The optimal imaging planes for identifying each type of frontal recess cell are summarized in the Table.

CT Findings Associated with FESS

Failure in the Frontal RecessMost cases of recurrent frontal sinusitis after FESS can be attributed to stenosis of the fron-tal recess, and a number of relatively common causes of postoperative frontal recess stenosis have been described in the rhinology literature. These include frontal recess obstruction caused by (a) inadequate removal of the agger nasi and frontal recess cells, (b) a retained superior por-tion of the uncinate process, (c) lateralization of the middle turbinate, (d) osteoneogenesis sec-ondary to chronic inflammation or mucosal strip-ping, (e) scarring or inflammatory mucosal thick-ening, and (f) recurrent polyposis (13,27,29,30). These entities are discussed individually in this section.

One key point to keep in mind is that patients with few or no symptoms after FESS may still have endoscopic or radiologic evidence of persis-tent or recurrent disease, but the treatment may still be considered a success (12). Therefore, ob-servation of any of these entities on a postopera-tive CT scan does not in itself indicate surgical failure. The CT results must be interpreted in the context of the overall clinical picture.

Residual Frontal Recess CellsPerhaps the most common reason for stenosis of the frontal recess after FESS is inadequate re-

Figure 14. Modified Lothrop procedure (Draf type III procedure). (a) Coronal drawing shows the area of resection in green. This procedure consists of contiguous bilateral enlargement of the frontal outflow tract, a result achieved by removal of the frontal sinus floors and adjacent parts of the inferior inter–frontal sinus septum, superior nasal septum, and anterior middle turbinates. (b) Coronal CT im-age obtained at the level of the frontal recess shows these findings in addition to a defect of the superior nasal septum. This defect should not be mistaken for an unintended surgical septal perforation.

TeachingPoint


nalis (12). On the other hand, when the uncinate process attaches to the skull base or the middle turbinate, it serves as the medial wall of the fron-tal recess, directing secretions into the ethmoid infundibulum prior to passage into the middle meatus (Fig 16) (17).

In patients who have not undergone surgery, one of the most common causes of obstruction at the level of the frontal recess is a medially displaced uncinate process. This occurs when disease in the recessus terminalis displaces the uncinate process medially so that it lies close to or even against the middle turbinate (12). Oc-

Retained Uncinate ProcessThe uncinate process has a variable relationship to the frontal recess, and its superior insertion dictates the direction of frontal recess drain-age into the middle meatus. When the uncinate process attaches to the lamina papyracea or the agger nasi, its anterior portion forms the lateral wall of the frontal recess, funneling frontal recess drainage directly into the middle meatus. In these cases, the ethmoid infundibulum terminates in a blind-ending recess known as the recessus termi-

Figure 15. Retained frontal recess cells. Postoperative coronal (a) and sagittal (b) CT images obtained through the left frontal recess show complete opacification of the frontal sinus and frontal recess. There are remnant opacified frontal recess cells, including an agger nasi cell (*) and a tier of type 2 frontal cells (arrows), which narrow the frontal ostium and frontal recess.

Figure 16. Effect of the superior attachment of the uncinate process on frontal recess drainage. Large arrow = superior aspect of the frontal recess. (a) Coronal CT image shows the uncinate process (small ar-rows) attached to the lamina papyracea. As a result, the ethmoid infundibulum terminates in a blind recess known as the recessus terminalis (*). In this case, frontal recess drainage (dashed red line) passes directly into the middle meatus. (b) Coronal CT image from another patient shows the uncinate process (small ar-rows) attached to the skull base at the junction of the cribriform plate and lateral lamella. Therefore, frontal recess drainage (dashed red line) is directed into the ethmoid infundibulum.


lateralize and obstruct the frontal recess (Fig 18) (28). It has been suggested that recurrent fron-tal sinusitis occurs significantly more often in patients who have undergone partial middle tur-binate resection than in surgically treated patients with intact middle turbinates (32), but recent studies have disputed this claim (33,34).

Nonetheless, rather than resecting a middle turbinate that has become “floppy” as a result of surgical manipulation, some surgeons choose to medialize the middle turbinate by creating small abrasions on the medial aspect of the middle turbinate and on the adjacent nasal septum (a process known as “Bolgerization”), effectively causing an adhesion between the two structures (9). Therefore, it is not uncommon to observe a medially deviated middle turbinate that appears

casionally, the uncinate process and middle tur-binate may become fused. It is not uncommon for surgeons to ignore the superior attachment of the uncinate process (17), and retained uncinate processes have been reported in 37% of patients undergoing revision FESS (6). Not surprisingly, if a medialized uncinate process is left behind during FESS, then the frontal recess will have a tendency to restenose postoperatively (12).

The uncinate process is most easily identified on coronal CT images, which clearly demonstrate its relationship to the frontal recess and sur-rounding structures (Fig 17).

Lateralized Middle Turbinate RemnantIn patients who have undergone middle turbinate manipulations, including partial middle turbinate resection, the amputated anterior stump may

Figure 17. Retained uncinate process. Postopera-tive coronal CT image obtained through the frontal recess shows a remnant uncinate process (arrow) on the left. The uncinate process is attached to the lamina papyracea and forms a recessus terminalis (*), which is opacified. The frontal recess is opaci-fied at the level of the recessus terminalis. Also note the medialized left middle turbinate adherent to the nasal septum; this appearance is a normal and often expected postsurgical finding.

Figure 18. Lateralized middle turbinate remnant. Postoperative coronal (a) and axial (b) CT images show a lateralized middle turbinate remnant (arrow). The middle turbinate remnant is adherent to the lamina papyra-cea, narrowing and obstructing the frontal recess.


Failure to preserve the normal mucosa will tend to result in scarring and osteoneogenesis (6,10). Osteoneogenesis, also referred to as osteitis or hyperostosis, refers to bone remodeling and new bone formation caused by chronic inflamma-tion. Sinus CT demonstrates osteoneogenesis in 36%–64% of patients with chronic rhinosinusitis (35,36), and the presence of hyperostosis at pre-operative CT has been shown to be a predictor of poorer surgical outcome after FESS (36).

There is a much higher prevalence of osteo-neogenesis among patients who have undergone FESS than among those without a history of si-nus surgery (41% vs 5%, respectively) (35). The relatively high prevalence of osteoneogenesis in postoperative patients is believed to be related to a combination of surgical mucosal trauma, persistent inflammation, and chronic refractory infection. As for the reason why the presence of osteoneogenesis may contribute to surgical fail-ure, it has been suggested that osteitic bone rem-nants serve as an inflammatory nidus, inducing overlying mucosal edema and hypertrophy, which in turn may contribute to frontal recess stenosis (35).

In normal individuals, the bony septa of the ethmoid sinuses have an average thickness of ap-proximately 0.5 mm, with the upper limit of nor-mal being about 1 mm. In addition, the middle turbinate measured at its midpoint on axial im-ages averages 1.5 mm in thickness, with a normal upper limit of 2.5 mm (36). At CT, osteoneogen-esis appears as thickening of the ethmoid septa or sinus walls and is often accompanied by scarring or mucosal edema (Fig 19).

adherent to the adjacent nasal septum on postop-erative CT scans (Fig 17). This is a normal and often expected postoperative finding.

Regardless of whether middle turbinate ma-nipulation does in fact have a detrimental effect on the frontal sinus after surgery, middle tur-binate lateralization is a well recognized cause of frontal recess stenosis in surgically treated individuals. A lateralized middle turbinate rem-nant is among the most common findings seen at preoperative endoscopy or CT. In one series, a lateralized middle turbinate was present in 78% of patients undergoing revision FESS (6), although other series have reported much lower rates of postoperative middle turbinate lateraliza-tion, ranging from 22% to 36% (8,29). Coronal and axial reformatted CT images are extremely helpful in demonstrating the presence of middle turbinate remnant lateralization.

In patients who have frontal recess stenosis secondary to a lateralized middle turbinate, a frontal sinus rescue procedure can be performed (28). This procedure involves removal of the an-terior aspect of the middle turbinate stump along with the mucous membrane over its medial as-pect. The membrane on the lateral aspect of the stump is preserved and draped medially over the denuded area where the stump has been resected, resulting in a large opening into the frontal sinus with preserved mucosa (11).

OsteoneogenesisAmong the primary objectives of FESS is pres-ervation of normal mucociliary function (6).

Figure 19. Osteoneogenesis. Postoperative coronal (a) and axial (b) CT images show thickening of remnant ethmoid septa (arrows), an appearance indicative of osteoneogenesis. There is associated mucosal thickening, which may represent inflamed edematous mucosa, scar tissue, or secretions, causing sinus opacification.


to quantify the amount of disease in chronic sinusitis, perhaps the most widely accepted of which is the Lund-Mackay system (39,40). How-ever, these grading systems are primarily research tools and are not widely used in the clinical set-ting. In evaluation of the frontal recess, it is gen-erally sufficient to report the presence or absence of mucosal thickening and, when mucosal thick-ening is present, a representative measurement of the degree of mucosal thickness.

Recurrent PolyposisSinonasal polyps are among the most frequent complications of sinusitis and among the most common findings in patients undergoing sinus surgery. They are formed by expansion of fluid in the deep lamina propria of the sinonasal Schneiderian mucosa and are the most common expansile lesions in the nasal cavity. Polyps have been associated with multiple causes including infectious rhinosinusitis, cystic fibrosis, aspirin intolerance, and allergic fungal sinusitis (41). Sinonasal polyposis is considered a predictor of poorer surgical outcome, and surgical fail-ure rates as high as 75% have been reported in patients with extensive sinonasal polyposis before surgery (42). Recurrent polyposis is seen in 29.9%–40% of patients undergoing revision FESS (6,8,29).

On CT images, polyps are usually homoge-neous soft-tissue masses with smooth convex borders and may be single or multiple (Fig 20). As they enlarge, they may fill a sinus and go on to remodel or even destroy adjacent bony structures (41).

Scarring and Inflam- matory Mucosal ThickeningScarring and inflammatory mucosal thickening in the frontal recess are extremely common after FESS, even in asymptomatic patients not requir-ing revision surgery. The presence of mucosal thickening has not been found to correlate with the presence of symptoms after FESS. In one series, mucosal disease was present in the frontal recess at postoperative endoscopy in 39 of 40 patients, but only three of 40 patients were symp-tomatic (37). Frontal recess scarring at endos-copy has been reported in approximately 15% of frontal sinuses not requiring revision (13).

On the other hand, it has been reported that mucosal inflammatory disease is a significant factor in two-thirds of cases of frontal sinusitis in patients undergoing revision frontal sinus sur-gery (19). Similarly, frontal recess scarring has been reported in 8.7%–50% of patients requiring revision FESS (6,13). Thus, although mucosal disease probably plays a role in most cases of re-current frontal sinusitis after FESS, its presence at CT or endoscopy does not necessarily corre-late with the presence of symptoms after surgery. Another important point to keep in mind when one is evaluating the sinus mucosa is that chronic sinus disease may be underdiagnosed with CT, as mucosal disease may be endoscopically visible even when it is not seen at CT (9).

It is impossible to differentiate scarring from inflamed edematous mucosa at CT because both appear as mucosal thickening, and they should be reported as such. Within the paranasal sinuses, mucosal thickness up to 3 mm may be normal (38). Various grading systems have been proposed

Figure 20. Recurrent polyposis. Postoperative coronal (a) and sagittal (b) CT images show near-complete opacification of the frontal sinuses and polypoid soft tissue completely opacifying the frontal recess, findings consistent with recurrent polyposis. Note the convex soft-tissue borders formed by the polyps (arrows in b).


tion of CT scans obtained in patients being con-sidered for frontal recess revision surgery requires a clear understanding of frontal recess anatomy, which in turn requires reviewing images in multi-ple imaging planes, including—at the very least—the axial, coronal, and sagittal planes. In addition, a working knowledge of the most common causes of surgical failure in the frontal recess, which include remnant frontal recess cells, a retained uncinate process, middle turbinate lateraliza-tion, osteoneogenesis, scarring or inflammatory mucosal thickening, and recurrent polyposis, will ensure that these entities are not overlooked dur-ing revision surgery. Recognition and effective communication of the presence of these findings should lead to better patient care and may reduce the likelihood of additional surgical failures.

References 1. Anand VK. Epidemiology and economic impact

of rhinosinusitis. Ann Otol Rhinol Laryngol Suppl 2004;193:3–5.

2. Shashy RG, Moore EJ, Weaver A. Prevalence of chronic sinusitis diagnosis in Olmstead County, Minnesota. Arch Otolaryngol Head Neck Surg 2004;130(3):320–323.

3. Maccabee M, Hwang PH. Medical therapy of acute and chronic frontal sinusitis. Otolaryngol Clin North Am 2001;34(1):41–47.

4. Kennedy DW, Zinreich SJ, Rosenbaum AE, Johns ME. Functional endoscopic sinus surgery: theory and diagnostic evaluation. Arch Otolaryngol 1985; 111(9):576–582.

5. Gross CW, Schlosser RJ. Prevalence and economic impact of rhinosinusitis. Curr Opin Otolaryngol Head Neck Surg 2001;9(1):8–10.

6. Musy PY, Kountakis SE. Anatomic findings in patients undergoing revision endoscopic sinus sur-gery. Am J Otolaryngol 2004;25(6):418–422.

7. Ramadan HH. Surgical causes of failure in endo-scopic sinus surgery. Laryngoscope 1999;109(1): 27–29.

8. Schaitkin B, May M, Shapiro A, Fucci M, Mester SJ. Endoscopic sinus surgery: 4-year follow-up on the first 100 patients. Laryngoscope 1993;103(10): 1117–1120.

9. Kennedy DW, Senior BA. Endoscopic sinus sur-gery: a review. Prim Care 1998;25(3):703–720.

10. McLaughlin RB Jr, Rehl RM, Lanza DC. Clini-cally relevant frontal sinus anatomy and physiology. Otolaryngol Clin North Am 2001;34(1):1–22.

11. Kuhn FA, Javer AR. Primary endoscopic manage-ment of the frontal sinus. Otolaryngol Clin North Am 2001;34(1):59–75.

Other Causes of Frontal Recess ObstructionIn addition to the entities described earlier, there are several entities to keep in mind when evalu-ating patients with symptoms of sinusitis after FESS. These include new or previously unde-tected mucoceles, neoplasms, or fibro-osseous lesions that may obstruct the frontal recess. It can be difficult to distinguish between polyps, muco-celes, mucous retention cysts, and true tumors at CT because they all appear as soft-tissue lesions within a sinus. However, certain imaging features can be useful in differentiating these entities.

Mucoceles are the most common expansile le-sions found in the paranasal sinuses. They repre-sent respiratory epithelium–lined cystic lesions of a sinus caused by chronic obstruction at the level of the sinus ostium. At CT, a mucocele appears as an expanded, airless sinus that is filled with homogeneous mucoid material. The presence of air within an affected sinus effectively rules out the possibility of a mucocele (41).

In contrast, mucous retention cysts represent obstructed submucosal mucinous glands. Mu-cous retention cysts are identical to and indistin-guishable from polyps at CT. Unless they become very large, both mucous retention cysts and pol-yps are almost always partially surrounded by air, a feature that helps distinguish them from muco-celes (41).

The presence of frank bone destruction should raise suspicion for a tumor, but large polyps can also cause thinning and deossification of bone, which may simulate destructive bone changes. If CT findings suggest the possibility of a tumor, contrast-enhanced magnetic resonance imaging should be performed.

ConclusionsThe emergence of FESS in the treatment of chronic sinusitis has significantly expanded the role of CT evaluation of the paranasal sinuses. CT is an invaluable adjunct to diagnostic nasal endoscopy in identifying various causes of frontal recess obstruction after FESS. Proper interpreta-


27. Smith TL, Rhee JS, Loehrl TA. Surgical manage-ment of frontal sinusitis. Curr Opin Otolaryngol Head Neck Surg 2001;9(1):42–47.

28. Sonnenburg RE, Senior BA. Revision endoscopic frontal sinus surgery. Curr Opin Otolaryngol Head Neck Surg 2004;12(1):49–52.

29. Chiu AG, Vaughan WC. Revision endoscopic fron-tal sinus surgery with surgical navigation. Otolar-yngol Head Neck Surg 2004;130(3):312–318.

30. Friedman M, Landsberg R, Schults RA, Tanyeri H, Caldarelli DD. Frontal sinus surgery: endoscopic technique and preliminary results. Am J Rhinol 2000;14(6):393–403.

31. Bradley DT, Kountakis SE. The role of agger nasi cells in patients requiring revision endoscopic frontal sinus surgery. Otolaryngol Head Neck Surg 2004;131(4):525–527.

32. Swanson PB, Lanza DC, Vining EM, Kennedy DW. The effect of middle turbinate resection upon the frontal sinus. Am J Rhinol 1995;9(4):191–195.

33. Fortune DS, Duncavage JA. Incidence of frontal si-nusitis following partial middle turbinectomy. Ann Otol Rhinol Laryngol 1998;107(6):447–453.

34. Giacchi RJ, Lebowitz RA, Jacobs JB. Middle tur-binate resection: issues and controversies. Am J Rhinol 2000;14(3):193–197.

35. Lee JT, Kennedy DW, Palmer JN, Feldman M, Chiu AG. The incidence of concurrent osteitis in patients with chronic rhinosinusitis: a clinicopatho-logical study. Am J Rhinol 2006;20(3):278–282.

36. Kim HY, Dhong HJ, Hyun JL, et al. Hyperostosis may affect prognosis after primary endoscopic si-nus surgery for chronic rhinosinusitis. Otolaryngol Head Neck Surg 2006;135(1):94–99.

37. Jacobs JB, Lebowitz RA, Lagmay VM, Damiano A. Conservative approach to inflammatory nasofron-tal duct disease. Ann Otol Rhinol Laryngol 1998; 107(8):658–661.

38. Rak KM, Newell JD 2nd, Yakes WF, Damiano MA, Luethke JM. Paranasal sinus on MR images of the brain: significance of mucosal thickening. AJR Am J Roentgenol 1991;156(2):381–384.

39. Lund VJ, Mackay IS. Staging in rhinosinusitis. Rhi-nology 1993;31(4):183–184.

40. Zinreich SJ. Imaging for staging of rhinosinusitis. Ann Otol Rhinol Laryngol Suppl 2004;193:19–23.

41. Som PM, Brandwein MS. Inflammatory diseases. In: Som PM, Curtin HD, eds. Head and neck imaging. 4th ed. St Louis, Mo: Mosby, 2003; 193–259.

42. Kennedy DW. Prognostic factors, outcomes, and staging in ethmoid sinus surgery. Laryngoscope 1992;102(12 pt 2 suppl 57):1–18.

12. Orlandi RR, Kennedy DW. Revision endoscopic frontal sinus surgery. Otolaryngol Clin North Am 2001;34(1):77–90.

13. Friedman M, Bliznikas D, Vidyasagar R, Joseph NJ, Landsberg R. Long-term results after endoscopic sinus surgery involving frontal recess dissection. Laryngoscope 2006;116(4):573–579.

14. Kew J, Rees GL, Close D, Sdralis T, Sebben RA, Wormald PJ. Multiplanar reconstructed computed tomography images improve depiction and under-standing of the anatomy of the frontal sinus and recess. Am J Rhinol 2002;16(2):119–123.

15. Lee WT, Kuhn FA, Citardi MJ. 3D computed tomographic analysis of frontal recess anatomy in patients without frontal sinusitis. Otolaryngol Head Neck Surg 2004;131(3):164–173.

16. Som PM, Shugar JMA, Brandwein MS. Anatomy and physiology. In: Som PM, Curtin HD, eds. Head and neck imaging. 4th ed. St Louis, Mo: Mosby, 2003; 87–147.

17. Landsberg R, Friedman M. A computer-assisted anatomical study of the nasofrontal region. Laryn-goscope 2001;111(12):2125–2130.

18. Bent JP, Cuilty-Siller C, Kuhn FA. The frontal cell as a cause of frontal sinus obstruction. Am J Rhinol 1994;8(4):185–191.

19. DelGaudio JM, Hudgins PA, Venkatraman G, Ben-ingfield A. Multiplanar computed tomographic analysis of frontal recess cells: effect on frontal isthmus size and frontal sinusitis. Arch Otolaryngol Head Neck Surg 2005;131(3):230–235.

20. Bolger WE, Butzin CA, Parsons DS. Paranasal sinus bony anatomic variations and mucosal ab-normalities: CT analysis for endoscopic surgery. Laryngoscope 1991;101(1):56–64.

21. Van Alyea OE. Ethmoid labyrinth: anatomic study with consideration of the clinical significance of its structural characteristics. Arch Otolaryngol Head Neck Surg 1939;29:881–902.

22. Meyer TK, Kocak M, Smith MM, Smith TL. Coronal computed tomography analysis of frontal cells. Am J Rhinol 2003;17(3):163–168.

23. Owen RG, Kuhn FA. Supraorbital ethmoid cell. Otolaryngol Head Neck Surg 1997;116(2):254–261.

24. Smith M, Smith T. Frontal sinus drainage pro-cedures: postoperative imaging appearance. Neu- rographics 2001;1(1):article 3. http://www. neurographics.org/Smith/1.shtml. Accessed March 21, 2008.

25. Draf W. Endonasal micro-endoscopic frontal sinus surgery: the Fulda concept. Op Tech Otolaryngol Head Neck Surg 1991;2:234–240.

26. Weber R, Draf W, Kratzsch B, Hosemann W, Schaefer SD. Modern concepts of frontal sinus sur-gery. Laryngoscope 2001;111(1):137–146.

This article meets the criteria for 1.0 AMA PRA Category 1 CreditTM. To obtain credit, see www.rsna.org/education /rg_cme.html.

RG Volume 29 • Number 1 • January-February 2009 Huang et al

Failed Endoscopic Sinus Surgery: Spectrum of CT Findings in the Frontal Recess Benjamin Y. Huang, MD, et al

Page 178 Therefore, CT of the paranasal sinuses should not be performed until 4–6 weeks after initiation of medical therapy, and scanning should be delayed in patients with acute upper respiratory infections (9). Page 179 The use of multiplanar reformation (sagittal and coronal) has been shown to improve preoperative understanding of the frontal recess. Specifically, review of sagittal images significantly improves one’s ability to identify and measure the frontal recess (14) and is critical in assessment of obstructing anterior ethmoid cells (14,15). Page 179 The boundaries of the frontal recess are typically formed by the agger nasi cell anteriorly; the lamina papyracea laterally; the most anterior and superior portion of the middle turbinate medially; and the ethmoid bulla, its associated bulla lamella, and the suprabullar cell (if present) posteriorly (Fig 1). Page 181 In addition to agger nasi cells, there are several other named frontal recess cells, which include frontal cells, supraorbital ethmoid cells, frontal bullar cells, suprabullar cells, and inter–frontal sinus septal cells (Table) (18). Page 189 Most cases of recurrent frontal sinusitis after FESS can be attributed to stenosis of the frontal recess, and a number of relatively common causes of postoperative frontal recess stenosis have been described in the rhinology literature.

RadioGraphics 2009; 29:177–195 • Published online 10.1148/rg.291085118 • Content Codes:

EDUCATION EXHIBIT Failed Endoscopic Sinus Surgery: Spec ...€¦ · Introduction Sinusitis is a...

Documents

Transcript of EDUCATION EXHIBIT Failed Endoscopic Sinus Surgery: Spec ...€¦ · Introduction Sinusitis is a...