Objectives To determine objective data to improve the methods of identification of the anterior ethmoidal artery during endoscopic dissection.
Study Design Cadaveric dissection of adult human heads.
Methods A 0°, 4-mm rigid endoscope was used to guide uncinectomy and frontoethmoidectomy. The location of the anterior ethmoidal artery was first determined visually and then confirmed by passing a needle through the anterior ethmoidal foramen from the orbit into the nose in all cases. The distances were endoscopically measured using a simple ruler between two nasal landmarks and the anterior ethmoidal artery.
Results Fifty-six nasal fossae in 28 cadavers were dissected endoscopically. The median distance between the artery and the “axilla” formed by the anterior attachment of the middle turbinate to the lateral nasal wall was 20 mm (range, 17–25 mm), irrespective of the side. The measurement differed by less than 2 mm between the sides in the same individual. The median distance between the artery and the “axilla” formed by the medial and lateral crura of the lower lateral cartilage (superomedial edge of the nostril) was 62 mm (range, 55–75 mm) for both sides. The artery was found to be in direct alignment with the two “axillae” formed by the middle turbinate and the nostril edge.
Conclusions The distance between the ethmoidal artery and the axilla of the middle turbinate showed the least intraindividual and interindividual variations. The tip of the endoscope (or the ruler) points directly at the anterior ethmoidal artery in the fovea ethmoidalis when its edge is aligned with the two nasal landmarks. These simple guidelines can aid the identification of the artery in endoscopic frontoethmoidectomy.
Since the first functional endoscopic sinus surgery course was held in the United States in 1985, the technique has spread worldwide. Initial surgical complications have been serious and sometimes catastrophic, but increasing surgical experience has been shown to reduce the complication rate. 1 This led to the recognition that safety and accuracy of endoscopic sinus surgery can be improved by a thorough knowledge of the endoscopic anatomy of the nose and paranasal sinuses, which should be gained through adequate cadaveric dissection in the first instance. More recently, intraoperative computed tomography (CT)–guided navigation of the sinonasal tract represents a further attempt to increase the safety of endoscopic surgery.
Irrespective of the surgical techniques and technical assistance available, the safety of the procedure is enhanced by identification of the anatomical margin of the paranasal sinuses with particular attention to the internal carotid artery and the optic nerve posteriorly, the fovea ethmoidalis and cribriform plate superiorly, and the lamina papyracea laterally. The carotid artery and the optic nerve can be identified through an enlarged sphenoidal sinus opening, and palpation of the orbit can help to gauge the proximity of the lamina papyracea. In experienced hands, the fovea ethmoidalis can be recognized by its white, smooth, ridged appearance or by finding the frontoethmoidal recess, the frontal sinus ostium, and the anterior ethmoidal artery. The anterior ethmoidal artery is a particularly important landmark, because it is attached to the skull base. If it is identified with certainty during endoscopic removal of any frontoethmoidal disease, inadvertent injury to the artery resulting in profuse epistaxis and even intraorbital and intracranial bleeding can be avoided. In addition, damage to the anterior skull base and cerebrospinal fluid leak from this area can be avoided, because the intranasal artery is at the level of the cribriform plate. The purpose of this study was to identify new guidelines that may improve the identification of the anterior ethmoidal artery during endoscopic dissection.
MATERIALS AND METHODS
Fifty-six nasal fossae in 28 cadavers were dissected endoscopically using a 4-mm, 0° Hopkins nasal endoscope. Uncinectomy and anterior ethmoidectomy were performed to reach the anterior skull base so as to locate the anterior ethmoidal artery. The anterior ethmoidal artery was identified visually in its endoscopic location (Fig. 1) whenever possible; then its intraorbital location at the anterior ethmoidal foramen was determined. The endoscopic location, whether identified with certainty or not, was confirmed by passing a short, straight needle through the anterior ethmoidal foramen from the orbit into the nose (Fig. 2) with the artery in situ in all cases. This ensured that the true location of the anterior ethmoidal artery was always marked by a needle for distance measurement.
Occasionally, a small, thin, concave web of membrane extending downward and slightly posteriorly from the undersurface of the middle turbinate near its anterior attachment was present. When present (in two cases), the membranous web was removed using cutting biopsy forceps until the hard bone of the middle turbinate and the lateral nasal wall were reached. This served to facilitate endoscopic frontoethmoidal dissection and standardize the narrow angle between the anterior end of middle turbinate and the maxilla (anterior turbinate axilla). Thus the narrow angle formed by the solid bony middle turbinate and the maxilla with no redundant mucosal covering was used as an intranasal landmark. Intranasal measurement was performed using a thin, 80-mm-long and 4-mm-wide plastic ruler that was sufficiently slim for insertion into the middle meatus. The ruler was trimmed from a longer plastic ruler that is widely used in ordinary operating theaters.
The distances between the artery and the anterior turbinate axilla (Figs. 3 and 4) between the artery and the “axilla” formed by the medial and lateral crura of the lower lateral cartilage (superomedial edge of the nostril) (Fig. 5) were measured with the ruler. The artery-to-anterior turbinate axilla distance was read under direct, close-up endoscopic vision. The ruler was placed in such a way that its tip touched the mid point of the ethmoidal artery and its edge was against the anterior turbinate axilla and then the uppermost medial corner of the nostril. The alignment of the artery with the two “axillas” was noted.
In this study of 28 cadavers and 56 nasal wall dissections, the anterior ethmoidal artery was always present. The mean age of the cadavers was 65 years (range, 25–90 y). There were 14 female and 14 male cadavers. In all but two dissections, the anterior ethmoidal artery was larger than the posterior ethmoidal artery. No skull base perforation occurred during any of the dissections.
The mean distance between the anterior ethmoidal artery and the anterior turbinate axilla in the entire group was 21 mm on both sides and was almost exactly the same irrespective of side and sex. The distance was 21 mm on the left side for both sexes (median, 20 mm; range, 18–25 mm; SD = 1.8 mm), 21 mm on the right side in females, and 22 mm on the right side in males (median for both sexes, 20 mm; range, 17–25 mm; SD = 2.1 mm). The mean distance between the artery and the superomedial nostril edge was 63 mm on the right side (median, 62 mm; range, 55–72 mm; SD = 3.6 mm) and 64 mm on the left (median, 62 mm; range, 60–75 mm; SD = 3.8 mm). In males, the mean distance between the artery and the nostril edge was 64 mm on the right side and 65 mm on the left. In females, the distance values were 61 mm and 63 mm, respectively. The interindividual variation in the distance measured between the artery and the nostril edge was up to 20 mm (range, 55–75 mm); the distance between the artery and the anterior turbinate axilla showed less variation, up to 8 mm (range, 17–25 mm). The intraindividual variation for the latter distance between the two sides was less than 2 mm (range, 0–2 mm) with two exceptions (cadavers 13 and 19), each of which showed a 3-mm difference. The individual results of the 56 dissections are presented in Table I.
Table Table 1.. Endoscopic Measurements for 28 Cadavers in Chronological Order.
During the cadaveric dissections, it was found that the anterior ethmoidal artery, the anterior turbinate axilla, and the superomedial edge of the nostril were in a straight line (Fig. 6).
The anterior ethmoidal artery is a key landmark for the fovea ethmoidalis and the base of the anterior cranial fossa. It separates the frontal recess anteriorly from the anterior ethmoidal cell posteriorly at the anterior roof of the nasal fossa. The superior boundary of the frontal recess slopes down posteriorly at a 15° angle and becomes more horizontal at the junction with the fovea ethmoidalis. Just behind the junction, the anterior ethmoidal artery transverses the medial orbital wall to the junction of the lateral lamella of the lamina cribrosa and the cribriform plate.
In endoscopic frontoethmoidectomy, localization of the anterior ethmoidal artery is an important step before further dissection can be continued accurately, safely, and confidently. The alternative is to identify the frontal recess or frontal sinus ostium, which may be difficult, even with the aid of a sinus probe. Endoscopic localization of the anterior ethmoidal artery, which serves as a landmark for the skull base and ethmoidal roof, has been advocated by Kennedy. 2
The technique for localization of the anterior ethmoidal artery initially described by Stammberger 3 makes use of the superior limit of the anterior wall of the bulla ethmoidalis, which acts as a superior pointer to the junction of the posterior wall of the frontal recess with the roof of the anterior ethmoidal cell. The artery is 1 to 2 mm posterior to this skull base junction. However, the position of the anterior ethmoidal artery in the ethmoidal canal is several millimeters higher than the level of the cribriform plate. As the artery passes from the ethmoidal canal into the olfactory fossa intracranially, it passes through a junction formed by the lateral lamina and the cribriform plate. This potentially fragile bony junction is susceptible to trauma, which can result in cerebrospinal fluid leak. Damage to the ethmoidal artery causes profuse hemorrhage and may also result in a surgical eye emergency when an orbital hematoma forms rapidly as a result of retraction of the lacerated artery into the orbit. For these reasons, surgical exploration of the anterior ethmoidal artery is performed with great caution and, in practice, limited dissection may have to be performed to save the anterior ethmoidal artery from damage and to avoid major complications.
Anatomy atlases and cadaveric study such as that by Becker 4 using cadaveric head sections are useful reference materials that visually demonstrate the relationship of the anterior ethmoidal artery to the frontal recess and the anterior ethmoidal air cells. All these visual representations have the shortcoming that they cannot be applied directly in endoscopic dissection, especially since the morphology of the intervening normal or diseased anterior ethmoidal air cells can be variable.
Ohnishi and Yanagisawa 5 showed the endoscopic anatomy of the anterior ethmoidal artery based on the endoscopic appearance of the anterior skull base after frontoethmoidectomy, but how many cases were collated and how the artery was safely approached and exposed were unclear. A classic study of the anatomy of the ethmoidal arteries by Kirchner et al. 6 was based on cadaveric skull base dissection via a cranial approach to the orbit; their findings were mainly useful for transorbital ethmoidal artery ligation and were not applicable in endoscopic sinus surgery. In contrast, our study describes a simple technique that involves a suitable ruler and alignment of the endoscopic shaft to simple landmarks to locate the anterior ethmoidal artery endoscopically. This technique is potentially useful in the clinical setting.
This study has shown that the anterior ethmoidal artery is at a mean distance of 21 mm from the anterior turbinate axilla on both sides and that the difference between the sexes is negligible. The shortest distance measured between these two sites was 17 mm on the right side and 18 mm on the left, although the median distance was 20 mm on either side. In addition, the position of the artery can be confirmed by its distance from the superomedial edge of the nostril (between the medial and lateral crura), which was, on average, 62 to 64 mm. This mean distance was 61 mm on the right side and 63 mm on the left in females, and the mean values were 64 mm and 65 mm, respectively, in males. Thus the intersex variation for this mean distance was up to 4 mm. The interindividual variation in this distance was up to 18 mm. This distance was not less than 60 mm, notwithstanding the only exception, in a 90-year-old female cadaver. In general, the ratio of the distance measured from the anterior ethmoidal artery to the middle turbinate axilla to the distance measured from the artery to the edge of the nostril was approximately 1 to 3.
The greatest possible interindividual variation occurred in the distance measurement between the artery and the nostril edge (up to 18 mm); the distance measurement between the artery and the anterior turbinate axilla showed less variation (up to 8 mm). The intraindividual variation for the latter distance between the two sides was less than 2 mm with two exceptions (cadavers 13 and 19), which showed a 3-mm difference. Therefore the distance measurement between the artery and the anterior turbinate axilla should be more useful in the clinical setting. These measurement variations, no matter how small, remain important, and great caution must still be exercised in dissecting the frontoethmoidal region. Visual recognition of the endoscopic structures should continue to be the standard in determining how to proceed during endoscopic sinus surgery.
In the course of these cadaveric dissections, it was found that the anterior ethmoidal artery, the anterior turbinate axilla, and the superomedial edge of the nostril were in a straight line. This discovery was extremely helpful in the endoscopic search for the anterior ethmoidal artery, because it improved the orientation of the frontoethmoidectomy. The linear relationship between these three structures aids in pointing to the position of the anterior ethmoidal artery.
If the shaft of the endoscope were rested against the upper nostril edge but well below the superior edge of the anterior turbinate axilla, it would be highly likely that the tip of the endoscope would be pointed at other structures such as the posterior ethmoidal artery behind the anterior ethmoidal artery. The linear relationship is particularly useful when the ethmoidal artery traverses the ethmoidal canal within the ethmoidal roof rather imperceptibly, showing no dehiscence, or within a mesentery, or, indeed, without an identifiable raised bony ridge. However, it must also be emphasized that the dissections in this study were performed exclusively in the Chinese population and extrapolation of these data to other ethnic groups should not be presumed.
The experience of this and other studies has shown that the anterior ethmoidal artery can vary in size or be absent altogether (although the latter instance was not featured in this series). Furthermore, it may be impossible to identify the anterior ethmoidal artery as it crosses the skull base imperceptibly within bone. Under these circumstances, the dissections should be performed with great care to avoid dangerous and futile attempts to identify the artery. Instead, other regional landmarks such as the frontal sinus ostium should be employed to guide the endoscopic dissection. Identification of the frontal sinus ostium is facilitated by the use of a 45° rigid nasoendoscope, which provides a better view of the frontal recess than the 30° or 70° versions. Under direct endoscopic vision, angled instruments such as curved curettes or Giraffe forceps are introduced to remove remnants of uncinate, ethmoidal bulla, or agger nasi cells, which can bend upward, follow the concavity of the frontal recess, and obscure the ostium. This principle, which was nicknamed “uncapping the egg,” was applied successfully for some years in the Graz University Medical School (Austria). The key to the principle is to locate the edges of these obstructing bony plates (or “eggshell”) in the frontal recess (or the “egg-holder”). In practice, intraoperative bleeding and difficult anatomical orientation can make this dissection challenging, even with appropriately angled endoscopes and forceps. However, if the anterior ethmoidal artery can be identified first, as suggested by the methods proposed in this study, the identification and removal of bony plates obscuring the frontal sinus ostium can be improved, since these remnants lie anterior to the artery.
The anterior ethmoidal artery, a crucial landmark in frontoethmoidal dissection, can be identified in experienced hands by direct observation of the artery at the junction of the frontal recess and the ethmoidal roof. This can be aided by using the front wall of the ethmoidal bulla as a superior pointer. This study has shown that the minimal distance between the artery and the middle turbinate axilla was 17 mm (median distance, 20 mm). The minimal distance between the artery and upper edge of the nostril was, almost always, not less than 60 mm (median distance, 62 mm). The former measurement is more useful, because it showed less intersex and interindividual variation. Furthermore, by placing the side of a slim ruler (or the endoscope) against the well-defined upper edge of the middle turbinate axilla and the superomedial edge of the nostril, the position of the artery is directly in front of the tip of the instrument. This consistent relationship may be a more valuable and convenient guide in identification of the artery than the somewhat variable distance taken from an intranasal landmark, although knowing how far one can dissect safely is also important.
The potential advantages of these techniques and measurements are that they are simple to follow, they are not intrusive or time-consuming to the surgeon, and they are likely to aid the endoscopic identification of the anterior ethmoidal artery. Therefore, serious complications during endoscopic dissection of the frontoethmoidal region may be prevented by careful adoption of the technique described in this study.