6

In recent years endoscopic sinus surgery (ESS) has become accepted as the treatment of choice for chronic sinusitis that is resistant to medical management.1 As ESS has become more widely adopted so the understanding of the complex and varied anatomy of the sinuses has improved.^2,3 However, the frontal recess and frontal sinus remain a challenge for surgeons. The anatomy is complex, varied, and can be confusing.^4,5 To better understand the anatomy of the paranasal sinuses, it is important to be aware of the embryology of the turbinates and sinuses. There are six embryological lamellae, or “ridges,” that form from the lateral nasal wall and give rise to important structures in the nose. Early in the fetal development these lamellae fuse to form four lamellae. Persistence of the fifth lamella will result in the presence of a supreme turbinate. This is rare and only present in 15% of the population. The first lamella forms the uncinate process, the second the bulla ethmoidalis, the third the middle turbinate, the fourth the superior turbinate, and the fifth (if present) the supreme turbinate (Fig. 6.1). The frontal, anterior ethmoid, and maxillary sinuses pneumatize from the furrow between the uncinate and bulla ethmoidalis. The posterior ethmoids pneumatize from the furrow between the middle and superior turbinates and the sphenoid sinus from the furrow above the superior turbinate.

The key to safe surgery in the frontal recess is a clear understanding of the anatomy. This chapter explains how two- dimensional (2D) computed tomography (CT) scans in coronal, parasagittal, and axial planes can be used to create a three- dimensional (3D) picture of the anatomy of the frontal recess. Such a 3D picture allows the surgeon to plan a surgical approach to the frontal recess so that each cell in the frontal recess can be entered in a predetermined sequential manner and then removed. This allows the surgeon to be able to turn to the CT scan at any point during the dissection and identify the cell that is currently being dissected. This mental picture gives the surgeon greater confidence that the complex anatomy of the frontal recess and frontal sinus is fully understood and removal of obstructing cells can be safely achieved. Insecurity during dissection in the frontal recess may result in either inadequate surgery with ESS failure or may increase the risk of injury to the skull base, orbit, and the anterior ethmoid artery.^4,6

Basic Anatomy of the Frontal Recess and Frontal Sinus

A common reason for ESS failure is inadequate removal of cells obstructing the outflow of the frontal sinus.^4,6 The location of the frontal recess creates anxiety for the surgeon as operating in this region places the lateral wall of the olfactory fossa (the thinnest part of the skull base), the anterior skull base (fovea ethmoidalis), the anterior ethmoidal artery, and the orbit at risk. The anterior wall of the frontal recess is formed by the thick bone of the frontal process of the maxilla, the so-called “beak” of the frontal process (Fig. 6.2). The size of this beak will vary according to the degree of pneumatization of the agger nasi cell. If there is a large agger nasi cell then the beak will be small. If, however, the agger nasi cell is absent or under-pneumatized, then the beak will extend significantly into the frontal recess and create a narrow frontal ostium as the beak approaches the forward projecting anterior skull base. Thus the antero-posterior distance from the skull base to the frontal beak is largely determined by the pneumatization of the agger nasi cell (Fig. 6.3).

The medial wall of the frontal recess is formed by the lateral wall of the olfactory fossa. The height of this wall is determined by the level of the cribriform plate. Keros⁷ classified the depth of the olfactory fossa as a Keros type 1 (< 3 mm), type 2 (3–7 mm), and a type 3 (> 7 mm). Depending on the Keros type, a variable amount of the lateral wall of the olfactory fossa will be exposed during dissection in this region. The bone of the lateral wall of the olfactory fossa varies in thickness between 0.05 and 0.2 mm and provides little resistance to penetration.⁸

The lateral wall of the frontal recess is formed by the lamina papyracea and the posterior wall by the upward continuation of the anterior face of the bulla ethmoidalis. On occasion this anterior wall of the bulla ethmoidalis may not reach the skull base and a suprabullar recess is formed (Fig. 6.3). The frontal recess is then continuous with this recess.

The roof of the frontal recess is formed by the fovea ethmoidalis. This bone is relatively thick and normally provides significant resistance to penetration. In a study conducted in our department we found that the right fovea ethmoidalis was higher than the left in 59% of patients.⁹ It should also be noted that the roof (fovea ethmoidalis) may slope, thereby placing the medial aspect of the roof at a lower level than the lateral aspect. The anterior ethmoidal artery and nerve runs across the fovea ethmoidalis at a 45-degree angle from lateral to medial (Fig. 6.4). In most instances it can be found behind the upward continuation of the bulla ethmoidalis. However, when this is absent and a suprabullar recess is present, the anterior ethmoidal artery will be in the frontal recess. The anterior ethmoidal artery may lie in a mesentery suspended from the skull base in 14 to 43% of patients (in our study the incidence was 34%).⁹ It is important that the CT scan is reviewed carefully before surgery to establish if the anterior ethmoidal artery is against the skull base or in a mesentery and if a suprabullar recess is present or not (Fig. 6.5).

The uncinate process, in the past, was thought to be the key to the frontal recess.8 This text adopts an alternate approach and suggests that the agger nasi cell is the key that unlocks the frontal recess.^10,11 The agger nasi cell is present in more than 90% of patients.¹² As the agger nasi cell is the key, it is important to understand the interaction between the uncinate and the agger nasi cell. The interaction between the upward continuation of the uncinate and the agger nasi cell is often poorly understood. The attachment of the root of the uncinate into either the lamina papyracea, skull base, or middle turbinate have been well described (Fig. 6.6),^5,8,13 but how this upward continuation of the uncinate interacts with the agger nasi cell and anterior ethmoidal cells in the frontal recess is sometimes poorly understood.

Attachment of the Uncinate to the Lamina Papyracea

In most cases the uncinate/medial wall of agger nasi cell implants on the lamina papyracea. In a large proportion of these patients, this upward extension will give off a leaflet of bone to the bulla lamella forming a plate of bone that divides the frontal recess vertically from posterior to anterior14,15 as it extends from the bulla ethmoidalis to the medial wall of the agger nasi cell and onto the frontal beak. In this situation the frontal sinus will drain medial to this plate (Fig. 6.7).

The relationship of a single large agger nasi cell to the frontal sinus ostium is better understood by viewing the coronal and parasagittal scans. This example shows the simplest anatomical configuration of the frontal recess. The next important step is to decide where the frontal sinus drains in relation to these cells.16–18 This concept is illustrated in Fig. 6.8.

Attachment of the Uncinate to the Middle Turbinate

The second anatomical variation to consider is that of a larger agger nasi cell. A large cell may push the upward continuation of the uncinate medially so that it attaches to the middle turbinate (Fig. 6.9).

The following series of CT scans and operative dissection pictures illustrates the upward continuation of the uncinate which forms the medial wall of the agger nasi cell and has been pushed by this cell to insert on the middle turbinate before progressing superiorly, forming the roof of the agger nasi and then implanting on the lamina papyracea (Figs. 6.10 and 6.11).

Attachment of the Uncinate to the Skull Base

The third scenario involves the further upward continuation of the uncinate onto the skull base. In a small percentage of patients the uncinate may have no relationship with the agger nasi cell. Usually in this configuration the uncinate will progress superiorly to implant on the skull base. The following series of CT scans and anatomical dissections illustrates this variation (Fig. 6.12).

The uncinate can be seen passing medial to the agger nasi cell and implanting at the junction of the middle turbinate and skull base. The broken line indicates the position of the parasagittal scan.

Alternatively the uncinate process may form the medial wall of a frontoethmoidal cell that is sitting above the agger nasi cell. This frontoethmoidal cell may push the upward continuation of the uncinate superiorly to attach onto the skull base (Fig. 6.13). The variations associated with the frontoethmoidal cells are considered later, in the discussion on the classification of the frontal ethmoidal cells. The following cadaver dissection and CT scan illustrate a cell on the right side pushing the insertion of the uncinate onto the skull base (Fig. 6.13).

The Agger Nasi Cell¹⁰ (Video 10)

There are a large number of possible variations in the anatomy of the frontal recess. To gain a functional understanding of the anatomy of the frontal recess the simplest configurations should be understood first before more complex variations are tackled. The simplest anatomical configuration is the single agger nasi cell without frontal ethmoidal cells. The agger nasi cell is the most anterior ethmoidal cell and is present in 93% of people.¹² The agger nasi cell forms a bulge on the lateral nasal wall anterior to the middle turbinate (Fig. 6.14). If sequential coronal CT scans are evaluated in an anterior to posterior direction, the agger nasi cell can be seen before the middle turbinate comes into view (Fig. 6.15).^5,12

Note that the uncinate only has a relationship with the posterior half of the agger nasi cell and not the anterior half which is why the uncinate cannot be seen on the coronal CT scans taken through the anterior half of the agger nasi cell, as seen in Fig. 6.15. This relationship can be viewed in Figs. 6.16 and 6.17.

The Transition from Frontal Sinus to Frontal Recess on the Coronal CT Scans

The surgeon reviewing a patient’s CT scans before surgery needs to understand how the anterior coronal CT scan through the agger nasi cell relates to the frontal beak and frontal sinus and how to be able to tell on sequential coronal CT scans when the transition from frontal sinus to frontal recess occurs. Figure 6.16 is a diagrammatic illustration of how a coronal CT scan through the anterior half of the agger nasi cell and through the frontal sinus can be identified on a coronal CT scan.

If line 1 is drawn in the coronal plane, the frontal beak can be seen as continuous ridge of bone with the frontal sinus above it (diagonally shaded area in Figs. 6.16 and 6.17). This line (line 1) is anterior to the uncinate and one can still see the continuous ridge of bone of the beak (floor of the frontal sinus). This makes it simple to differentiate the frontal sinus (above the beak) from the frontal recess. Line 2 in Fig. 6.17 shows a coronal cut through the uncinate process behind the beak. This illustrates the transition from the frontal sinus to the frontal recess with loss of the continuity of bone (illustrated as the “frontal beak” in Fig. 6.16) and by the presence of the uncinate. This coronal cut illustrates the posterior part of the agger nasi cell’s relationship to the superior extension of the uncinate process. This part of the uncinate forms the medial and posterior medial wall of the agger nasi cell and represents the relationship between the anterior agger nasi cell (shaded with dots) and the frontal beak and the floor of the frontal sinus (diagonally shaded area).

The diagonally shaded area in Figs. 6.16 and 6.17 is the frontal sinus above the frontal beak. The frontal beak forms the floor of the frontal sinus (Fig. 6.2). Frontal ethmoidal cells pushing into this shaded area are classified as Kuhn type 3 cells (Table 6.1).¹⁴ From these diagrams it can be seen that most of the agger nasi cell is anterior to the uncinate but the posterior half of the agger nasi cell has an intimate relationship with the upward extension of the uncinate process (Fig. 6.18).¹⁰

Transition from Frontal Sinus to Frontal Recess on the Axial Scans

The surgeon should also be able to differentiate on the axial scans when transition occurs from the frontal sinus to the frontal recess. The frontal sinus scans should be viewed from the top down (cranial to caudal). The frontal sinus is relatively easy to identify because, as it narrows toward the frontal ostia, it forms a square (Fig. 6.19). At this level the posterior wall of the two frontal sinuses forms a straight line (Fig. 6.19B). As the skull base turns posteriorly these squares elongate posteriorly but still maintain a roughly rectangular shape. This is the transition stage from frontal sinus to frontal recess (Fig. 6.19D,E). As the posterior ends of these boxes become pointed the scans reach the frontal recess (Fig. 6.19E). Figure 6.19A,B shows the square formation of the frontal sinuses. The transition region is between Fig. 6.19D and Fig. 6.19E. Note how the bone of the anterior wall changes at each of these levels. At Fig. 6.19A,B, the bone of the anterior wall of the frontal sinus is even and relatively flat and not very thick. The anterior wall bone becomes much thicker as the upper region of the frontal beak is reached (Fig. 6.19C). At Fig. 6.19D,E the anterior wall is curved indicating that the nasion has been reached. In Fig. 6.19D the nasion is fully developed and the frontal beak bone is thick. Also note how the posterior wall has become pointed (Fig. 6.19D,E) as the axial cut goes into the frontal recess toward the anterior ethmoidal arteries. This artery is seen in its canal in Fig. 6.19D on the right side.

Table 6.1 Kuhn Classification of Frontal Recess and Frontal Sinus Cells

A single agger nasi cell in the frontal recess is only one of the many anatomical variations. In 1995 Fred Kuhn classified the cells seen in the frontal recess and frontal sinus as presented in Table 6.1.¹⁴

This classification still forms the basis of our current classification although we have made significant modifications (Table 6.2). One of the important modifications is defining the types of cells that occur in the frontal recess and frontal sinus more precisely. The first cells to be considered are the frontal ethmoidal cells. For a cell to be called a frontal ethmoidal cell it needs to satisfy two criteria: first it must be an anterior ethmoidal cell and second it should be in close proximity to the frontal process of the maxilla. The frontal process of the maxilla is the bone forming the anterior wall of the frontal recess (Figs. 6.2 and 6.20).

This bone goes on to form the frontal beak. The frontal ethmoidal cells are further divided depending on how many there are and how far these cells extend into the frontal sinus through the frontal ostium.¹⁶ Kuhn classified frontal ethmoidal cells into types 1 to 4.¹⁴ However, we have modified this classification by clearly defining a frontal ethmoidal cell and by redefining the type 3 and 4 cells (Table 6.2).

The Building Block Concept for the Reconstruction of the Anatomy of the Frontal Recess

To reconstruct the cells in the frontal recess in three dimensions, building blocks are arranged, one block for each cell.^10,18 This is done as an aid to the dissection of the frontal recess. When operating in this area the surgeon needs to know exactly which cell is being dissected and in what sequence each cell will be opened so that the frontal recess can be safely and competently cleared. To build a mental picture of the cells in the frontal recess, the coronal CT scans are viewed first. The first cell seen on the coronal CT that is anterior to and directly above the insertion of the middle turbinate is, in most cases, the agger nasi cell^10,18 (Fig. 6.14). This cell should now be identified on the parasagittal CT scan (Fig. 6.21). Make sure that you use all the information available to you to confirm that the cell seen on the coronal is indeed the cell that you have identified on the parasagittal. Ask yourself if the cell has air or whether it is partially or completely opacified. Check on both the coronal and para-sagittal scans that the cell that you have identified has the expected amount of opacification. A building block is now placed for this cell (Fig. 6.21).

Stay updated, free articles. Join our Telegram channel

Join