INTRODUCTION
An intimate knowledge of lateral nasal wall anatomy is essential in lacrimal surgery. A recent metaanalysis of level II and III studies shows that the success rate of the endonasal dacryocystorhinostomy (EnDCR) is comparable with the external DCR [1,2▪] . Failure of the endoscopic technique seems in large part due to a poor understanding of endonasal anatomy and location of the lacrimal sac on the lateral nasal wall, which leads to a poorly placed rhinostomy [3] .
The primary goal of EnDCR is to create a bony ostium large enough to allow complete exposure and marsupialization of the lacrimal sac [4,5] and avoidance of complications such as basal skull fracture, intranasal and orbital hemorrhage [6] , orbital fat prolapse and compromise of the ostium of the maxillary sinus [7,8] .
The maxillary line and the axilla of the middle turbinate are the major landmarks utilized by endonasal surgeons to localize the lacrimal sac. However, the frontal process of maxilla (FPM), the lacrimal bone, agger nasi air cell (ANC) and the uncinate process are also intimately related to the lacrimal sac and will be routinely encountered during endonasal surgery. This article reviews the studies that have helped to define the intranasal and external surface anatomy of the lateral nasal wall and established the relationship of the lacrimal sac to other important anatomical landmarks that have contributed to the increasing success of endonasal lacrimal surgery.
Box 1: no caption available
OSTEOLOGY OF THE LACRIMAL SAC FOSSA
The lacrimal fossa is formed by two bones, the FPM anteriorly and lacrimal bone posteriorly, articulating at the vertical lacrimomaxillary suture (LMS) within the lacrimal fossa (Fig. 1). The lacrimal fossa is bordered anteriorly by the anterior lacrimal crest (ALC) on the FPM and posteriorly by the posterior lacrimal crest (PLC) on the lacrimal bone. In Caucasians the average width of the lacrimal fossa, from the ALC to the PLC, is, approximately, 7.2–10.4 mm [9,10] . The lacrimal fossa measures, approximately, 16 mm in vertical height and is 2–4 mm deep. The lacrimal fossa is widest at its base, wherein it is confluent with the opening of the nasolacrimal canal, formed by the maxillary bone laterally and the lacrimal and inferior turbinate bones medially.
FIGURE 1: External view of the left lacrimal fossa and medial orbital wall in a human skull. The relative position of the lacrimal fossa to the anterior and posterior lacrimal crest, lacrimal, frontal, maxillary and ethmoid bone and their respective sutures are shown.
Variability has been demonstrated in different racial groups in the relative position of the LMS within the lacrimal fossa and hence the relative composition of the lacrimal fossa by the thicker FPM anteriorly and the thinner lacrimal bone posteriorly (Fig. 2). A cadaveric study of 47 orbits of Caucasian subjects found a maxillary bone dominant lacrimal fossa in 32%, wherein the LMS is located closer to the PLC [10] and creating the osteotomy using a bone punch alone can be more difficult. This anatomical configuration was seen in 12% in the Turkish [11] and 8.3% in the Indian cadaveric study; [12] with one case in which the entire lacrimal fossa was formed by the FPM (Table 1 ).
FIGURE 2: Transilluminated external view of the left lacrimal fossa and medial orbital wall in a human skull. The composition of the lacrimal fossa by the thin lacrimal bone posteriorly, which does not extend to the top of the lacrimal sac, and the thicker frontal process of the maxilla anteriorly is shown.
Table 1: Comparison of the relative bony composition of the lacrimal fossa in different racial groups
The lacrimal fossa shallows superiorly terminating at the frontomaxillary and frontolacrimal suture, but in some cases the lacrimal fossa may not have a well defined superior boundary [12] . The FPM appears to dominate at the upper level of the lacrimal fossa; the lacrimal bone not extending to this level (Figs 2 and 3). The Asian computerized tomography (CT) study by Woo demonstrated the contribution of the FPM to the lacrimal fossa to be 79% at the upper level of the fossa, 69% at the midlevel and 62.4% at the lower level [13] and the average thickness of the lacrimal bone was 3.3 mm, 2.6 mm and 2 mm at the these levels, respectively [13] . In contrast, in Caucasians the lacrimal bone within the lacrimal fossa has a mean thickness of 1.06 mm and easily penetrated during surgery. The height of the nasal bone was also shown to have a significant negative correlation with the thickness of the FPM within the lacrimal fossa [13] . This supports anecdotal experience that performing a DCR in Asians with a low nasal bridge can be more difficult.
FIGURE 3: Schematic diagram of the intranasal landmarks of the lateral nasal wall. Parasagittal view of the lateral nasal wall showing the relationship of the LF overlying the lacrimal bone and frontal process of maxilla, anterior and lateral to the axilla of the MT, superior to the IT, and underlying the UP and ANC. ANC, agger nasi cell; FS, frontal sinus; IT, inferior turbinate; LF, lacrimal fossa; MT, middle turbinate; UP, uncinate process .
Wormald et al. [3] used CT dacryocystogram (DCG) scans to establish the relationship of the lacrimal sac to the lateral nasal wall and found that the FPM progressively thickens toward the fundus of the sac reaching up to 15 mm. This explains why in the majority of patients a diamond burr rather than the punch is required to remove the bone above the axilla of the middle turbinate. Successful DCR surgery is dependent on the area of bony resection around the common canaliculus to be at least 3–5 mm in diameter [14] . Adequate bony clearance over the fundus of the lacrimal sac may require removal of 3–5 mm of frontal bone above the lacrimal fossa, that is, above the frontomaxillary suture (Fig. 4). In the presence of a large osteotomy with adequate superior bony clearance, complete marsupialization of the lacrimal sac, and in the absence of canalicular disease, stenting of the lacrimal system is also unnecessary [15,16] .
FIGURE 4: The fundus of the right lacrimal sac relative to the axilla of the middle turbinate. The fundus of the lacrimal sac extends 8 mm above the axilla of the middle turbinate and 5 mm above the common canalicular opening. LF, lacrimal fossa; MT, middle turbinate; UP, uncinate process .
THE RELATIONSHIP OF THE MIDDLE TURBINATE TO THE LACRIMAL SAC
There are three turbinates along the lateral nasal wall whose long axis is parallel to the floor of the nasal cavity. Beneath each turbinate lies its corresponding meatus. Within the inferior meatus is the nasolacrimal duct opening. The middle meatus contains the uncinate process , the bulla ethmoidalis, the frontal recess and the maxillary sinus ostium. The axilla of the middle turbinate is the most anterior point of its insertion to the lateral nasal wall and is formed by the fusion of the middle turbinate and the FPM [17] .
The axilla of the middle turbinate is a constant anatomical landmark, which can be usefully utilized in DCR surgery. In the majority of cases the lacrimal fossa usually lies superior, anterior and lateral to the axilla of the middle turbinate (Fig. 3).
Contrary to previous descriptions that the axilla of the middle turbinate was at the upper limit of the fundus of the lacrimal sac [18–20] , Wormald et al. [3] used CT DCG to show that the sac extends 8–10 mm above the axilla of the middle turbinate and that a significant part of the fundus of the sac (∼5 mm) lies above the common canalicular opening (Figs 4 and 5). Studies on Caucasian [21] , Asian [13] and Turkish [11] subjects have also reported that the fundus of the sac is an average of 6.10 mm ± 2.02 mm, 6.6 mm ± 1.3 mm and 4.73 mm ± 2.86 above the axilla, respectively. EnDCR may require concomitant nasal procedures, including anterior middle turbinectomy, to enable adequate access to the entire lacrimal fossa [22–24] .
FIGURE 5: Endoscopic view of the left nasal cavity of a human skull. The relationship of the lacrimal sac (dotted line) to the frontal process of the maxilla, lacrimal bone, UP and axilla of the MT are shown. The UP is anterosuperiorly attached to the frontal process of the maxilla, but not the lacrimal bone, which is anterior and lateral to the UP. Anteriorly, the lacrimal bone joins with the frontal process of the maxilla by the LMS. Posteriorly, it joins the lamina papyracea of the ethmoid bone. LMS, lacrimomaxillary suture; MT, middle turbinate; UP, uncinate process .
THE RELATIONSHIP OF THE MAXILLARY LINE TO THE LACRIMAL SAC
The maxillary line is considered the most medial projection of the FPM (Fig. 6). The maxillary line is the indentation of the FPM on the lateral nasal wall. It is a curvilinear eminence which runs from the axilla of the middle turbinate along the lateral nasal wall to end at the root of the inferior turbinate [17,20] . Contrary to earlier studies suggesting that the maxillary line corresponds to the position LMS [9,11,20] , more recent studies [25] and our own clinical experience support that the maxillary line, at the level of the superior one-third of the lacrimal fossa, corresponds to the anterior extent of the upper portion of the uncinate process at its junction with the FPM or middle turbinate. At the level of the middle and inferior one-third of the lacrimal fossa, however, there is frequently 3–6 mm of thick FPM extending posterior and lateral to the maxillary line before the junction with the uncinate process is encountered, with 2–4 mm of intervening lacrimal bone, which can only be identified once the overlying mucosa has been removed.
FIGURE 6: Intraoperative photograph of the left lateral nasal wall. The location of the maxillary line (dotted line) relative to the axilla of the middle turbinate and frontal process of the maxilla is demonstrated. The maxillary line begins superiorly at the middle turbinate attachment and then curves down along the lateral nasal wall to end at the root of the inferior turbinate.
THE RELATIONSHIP OF THE UNCINATE PROCESS TO THE LACRIMAL SAC FOSSA
The uncinate process is a thin sickle-shaped bone, slightly thicker than the lacrimal bone, oriented in the parasagittal plane, whose anterior aspect curves slightly laterally to attach to the lateral nasal wall at the FPM [26] (Fig. 5). Inferiorly, the uncinate process attaches to the ethmoidal process of the inferior turbinate. The uncinate process is lined by mucosa on both its medial and lateral aspects, with its posterior edge free within the middle meatus. The small two-dimensional cleft between the posterior edge of the uncinate process and the ethmoid bulla is the hiatus semilunaris. Lateral to the uncinate process , the infundibulum is the three-dimensional space in which the maxillary sinus and anterior ethmoid air cells drain.
The uncinate process is divided into three parts (Fig. 5): the inferior horizontal portion attaches to the ethmoidal process of the inferior turbinate; its middle third attaches to the lacrimal bone [13] ; and its superior third extends to a varying degree into the frontal recess forming the posterior lateral wall of the ANC and continues to insert into the lamina papyracea, skull base, middle turbinate or combination of these [17,26,27] . Standard ethmoidal pneumatization usually places the middle third of the uncinate process on the posterior lacrimal bone at its junction with the lamina papyracea.
When starting the bony rhinostomy in EnDCR the FPM is approached, and the articulation of the lacrimal bone is found in the lower one-third of the lacrimal fossa (Fig. 3). The thin lacrimal bone is elevated off the posterior half of the lower lacrimal sac up to the insertion of the uncinate process . In this area, the uncinate process is in a retrolacrimal position, approximately, 3–4 mm of thin lacrimal bone overly the posteroinferior lacrimal sac before the uncinate process is encountered and does not need resection [28] . Once this area of the lower one-third of the lacrimal fossa is exposed, the sac is exposed in an upward direction.
To expose the fundus of the sac in the superior one-third of the lacrimal fossa requires the opening of the ANC, which is in a superoposterior relation to the lacrimal fossa (Fig. 3), by removing the anterior wall of the ANC. The medial wall of the ANC is the upper continuation of the uncinate process , is usually preserved but may need to be removed in some cases in which there is poor pneumatization of the ANC. This anterior approach to the lacrimal fossa via the FPM obviates the need to begin with an uncinectomy to protect the orbit from inadvertent entry [29] .
The uncinate process can easily be overlooked because it is a very thin structure. Elevation of the mucosal flap to expose the posterior edge of the FPM and lacrimal bone will often detach the root of the uncinate process , but this is distinct from a formal uncinectomy. In contrast, some authors that have advocated deliberate cleavage of the uncinate process to gain access to the lacrimal fossa [13,30–32] , supported by reports that the anterior part of the uncinate process overlaps the lacrimal fossa, at least partially, in 94.8% [21,30] .
The anterior insertion of the uncinate process may be onto the retrolacrimal bone, lacrimal bone, FPM or axilla of the middle turbinate. Studies by Fayet et al. [21] , Woo et al. [13] and Zhang et al. [17] compared the anterior insertion of the uncinate process at the upper, middle and inferior third of the lacrimal fossa to various structures; the lacrimal bone, FPM or to the axilla of the middle turbinate (Table 2 ). Unlike the study by Fayet et al. [21] on Caucasian subjects, in which the anterior insertion of the uncinate process was retrolacrimal in 32 and 5.2% at the inferior and middle one-third of the lacrimal fossa, respectively, Woo et al. [13] found no cases in which the uncinate process was attached posterior to the lacrimal fossa and always overlapped with the lacrimal fossa, suggesting an uncinectomy was required more frequently in Asians. Zhang et al. [17] showed similar results in cadavers and a CT scan study on 42 Asian lacrimal systems; the uncinate process attached laterally to the FPM in all individuals, as well as to the lacrimal bone in 79%. In the remaining 21% the anterosuperior portion of the uncinate process only attached to the FPM but not to the lacrimal bone. Soyka et al. [28] analyzed the CT scans of 60 patients undergoing DCR and found the uncinate process covering at least 50% of the lacrimal fossa in 63% of individuals.
Table 2: The position of the anterior insertion of the uncinate process along the superior, middle and inferior third of the lacrimal fossa
THE RELATIONSHIP OF THE AGGER NASI CELL TO THE LACRIMAL SAC FOSSA
The ANC appears as a mound or eminence in the lateral nasal wall, anterior to the origin of the middle turbinate and is present in 80–98.5% of cases [33,34▪▪] . It can usually be seen on the CT scan anterior to the middle turbinate [27] . Depending on the degree of pneumatization, ANC can be bounded anteriorly by the FPM, anterolaterally by the nasal and lacrimal bones, posterolaterally by the lamina papyracea, superiorly by the frontal recess, inferolaterally and posteriorly by the uncinate process . An extensively pneumatized ANC can displace the insertion of the middle turbinate medially, anteriorly and superiorly and may even reach the floor of the frontal sinus. Most of ANC is anterior to the uncinate process , but the posterior half of the ANC has an intimate relationship with the upward extension of the uncinate process [35] .
The ANC is closely related to the posterior–superior aspect of the lacrimal fossa [36,37] , with frequent extension of the air cell over the superior half of the lacrimal fossa (Fig. 3). This was appreciated by Whitnall et al. [36] who found the ANC extended anterior to the LMS in 86 of 100 skulls. More recent studies have demonstrated the ANC to extend anterior to the PLC, and, thus facing the upper and posterior part of the lacrimal fossa in 41–90% of cases [13,17,28,34▪▪,37] . It is, therefore, unsurprising that in 78% of failed DCRs compared with 20% of successful DCRs ethmoid air cells were found, which extended to the medial side of the lacrimal sac [38] . As the bone above the axilla of the middle turbinate is removed, the mucosal lining of the ANC is visualized, which can be marsupalized and opposed to the posterior lacrimal sac flap.
THE RELATIONSHIP OF THE AGGER NASI CELL TO THE UNCINATE PROCESS
The interaction between the upper portion of the uncinate process and the ANC is important in endonasal surgery. Embryonically, both the uncinate process and the ANC originated from the first ethmoturbinal. Most of the ANC is anterior to the uncinate process , but the posterior half of the ANC has an intimate relationship with the superior attachment of the uncinate process [35] (Fig. 3). Zhang et al. [17] showed that the ANC was medially, superiorly and inferiorly bounded by the uncinate process , and that the posterior pneumatization of the ANC pushed the posterosuperior attachment of the uncinate process backward to the lamina papyracea to form the terminal recess. Pneumatization of the ANC with or without frontal ethmoidal cells effects the superior attachment of uncinate process [39] . Full marsupialization of the ANC involves the removal of anterosuperior and posterosuperior portions of the uncinate process as well as the posterosuperior portion of the FPM and finally the connection between the uncinate process and the middle turbinate [39] . In patients whom there is no ANC (1.5%) in the bone above the axilla of the middle turbinate is thicker because of the absence of this cell [40] .
CONCLUSION
The studies summarized in this article have helped to improve our appreciation of the anatomical landmarks of the lateral nasal wall that are relevant to EnDCR surgery. This knowledge enables accurate localization of the lacrimal fossa and full exposure the lacrimal sac, thereby improving the success and safety of this procedure.
Acknowledgements
We wish to thank Mr Randy E. Veridick (FOPS) and Dr Rowan Valentine for their expertise and assistance in the preparation of the images for this study.
Financial support and sponsorship
None .
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
▪ of special interest
▪▪ of outstanding interest
REFERENCES
1. Huang J, Malek J, Chin D, et al. Systematic review and meta-analysis on outcomes for endoscopic versus external dacryocystorhinostomy. Orbit 2014; 33:81–90.
2▪. Marcet MM, Kuk AK, Phelps PO. Evidence-based review of surgical practices in endoscopic endonasal dacryocystorhinostomy for primary acquired nasolacrimal duct obstruction and other new indications. Curr Opin Ophthalmol 2014; 25:443–448.
This recent article reviews how the evolution of the endonasal dacryocystorhinostomy technique has resulted in improved outcomes and has helped establish this approach as a viable alternative to external DCR for primary acquired nasolacrimal duct obstruction.
3. Wormald PJ, Kew J, Van Hasselt A. Intranasal anatomy of the nasolacrimal sac in endoscopic dacryocystorhinostomy. Otolaryngol Head Neck Surg 2000; 123:307–310.
4. Chan W, Selva D. Ostium shrinkage after endoscopic dacryocystorhinostomy. Ophthalmology 2013; 120:1693–1696.
5. Cukurova I, Caner Mercan G, Cetinkaya E, et al. Endoscopic dacryocystorhinostomy: outcomes using mucosal flap preserving technique. Eur Arch Otorhinolaryngol 2013; 270:1661–1666.
6. Andrew N, Selva D. Postoperative haemorrhage in powered endoscopic dacryocystorhinostomy. Clin Experiment Ophthalmol 2014; 42:262–265.
7. Shams PN, Selva D. Acute postoperative rhinosinusitis following endonasal dacryocystorhinostomy. Eye (London, England) 2013; 27:1130–1136.
8. Fayet B, Racy E, Assouline M. Complications of standardized endonasal dacryocystorhinostomy with unciformectomy. Ophthalmology 2004; 111:837–845.
9. Chastain JB, Cooper MH, Sindwani R. The maxillary line: anatomic characterization and clinical utility of an important surgical landmark. Laryngoscope 2005; 115:990–992.
10. Shams PN, Abed SF, Shen S, et al. A cadaveric study of the morphometric relationships and bony composition of the Caucasian nasolacrimal fossa. Orbit 2012; 31:159–161.
11. Orhan M, Saylam CY, Midilli R. Intranasal localization of the lacrimal sac. Arch Otolaryngol Head Neck Surg 2009; 135:764–770.
12. Bisaria KK, Saxena RC, Bisaria SD, et al. The lacrimal fossa in Indians. J Anat 1989; 166:265–268.
13. Woo KI, Maeng HS, Kim YD. Characteristics of intranasal structures for endonasal dacryocystorhinostomy in asians. Am J Ophthalmol 2011; 152:491–498.e491.
14. Welham RA, Wulc AE. Management of unsuccessful lacrimal surgery. Br J Ophthalmol 1987; 71:152–157.
15. Chong KK, Lai FH, Ho M, et al. Randomized trial on silicone intubation in endoscopic mechanical dacryocystorhinostomy (SEND) for primary nasolacrimal duct obstruction. Ophthalmology 2013; 120:2139–2145.
16. Cannon PS, Chan W, Selva D. Incidence of canalicular closure with endonasal dacryocystorhinostomy without intubation in primary nasolacrimal duct obstruction. Ophthalmology 2013; 120:1688–1692.
17. Zhang L, Han D, Ge W, et al. Anatomical and computed tomographic analysis of the interaction between the
uncinate process and the agger nasi cell. Acta Otolaryngol 2006; 126:845–852.
18. Steadman MG. Transnasal dacryocystorhinostomy. Otolaryngol Clin North Am 1985; 18:107–111.
19. McDonogh M, Meiring JH. Endoscopic transnasal dacryocystorhinostomy. J Laryngol Otol 1989; 103:585–587.
20. Metson R, Woog JJ, Puliafito CA. Endoscopic laser dacryocystorhinostomy. Laryngoscope 1994; 104:269–274.
21. Fayet B, Racy E, Assouline M, Zerbib M. Surgical anatomy of the lacrimal fossa a prospective computed tomodensitometry scan analysis. Ophthalmology 2005; 112:1119–1128.
22. Figueira E, Al Abbadi Z, Malhotra R, et al. Frequency of simultaneous nasal procedures in endoscopic dacryocystorhinostomy. Ophthal Plast Reconstr Surg 2014; 30:40–43.
23. Fayet B, Katowitz WR, Racy E, et al. Endoscopic dacryocystorhinostomy: the keys to surgical success. Ophthal Plast Reconstr Surg 2014; 30:69–71.
24. Hull S, Lalchan SA, Olver JM. Success rates in powered endonasal revision surgery for failed dacryocystorhinostomy in a tertiary referral center. Ophthal Plast Reconstr Surg 2013; 29:267–271.
25. Raikos A, Waidyasekara P, Morrison AK. Surgical and topographic anatomy of the maxillary line: an important landmark for endoscopic nasal surgery. Ann Anat 2015; 197:24–28.
26. Ercan I, Cakir BO, Sayin I, et al. Relationship between the superior attachment type of
uncinate process and presence of agger nasi cell: a computer-assisted anatomic study. Otolaryngol Head Neck Surg 2006; 134:1010–1014.
27. Stammberger HR, Kennedy DW. Paranasal sinuses: anatomic terminology and nomenclature. Ann Otol Rhinol Laryngol Suppl 1995; 167:7–16.
28. Soyka MB, Treumann T, Schlegel CT. The agger nasi cell and
uncinate process , the keys to proper access to the nasolacrimal drainage system. Rhinology 2010; 48:364–367.
29. Tsirbas A. Lacrimal fossa anatomy. Ophthalmology 2006; 113:1475–1476.author reply 1476-1477.
30. Fayet B, Racy E, Assouline M. Systematic unciformectomy for a standardized endonasal dacryocystorhinostomy. Ophthalmology 2002; 109:530–536.
31. McDonogh M. Endoscopic transnasal dacryocystorhinostomy. Results in 21 patients. S Afr J Surg 1992; 30:107–110.
32. Sham CL, van Hasselt CA. Endoscopic terminal dacryocystorhinostomy. Laryngoscope 2000; 110:1045–1049.
33. Lee WT, Kuhn FA, Citardi MJ. 3D computed tomographic analysis of frontal recess anatomy in patients without frontal sinusitis. Otolaryngol Head Neck Surg 2004; 131:164–173.
34▪▪. Liang J, Hur K, Merbs SL, Lane AP. Surgical and anatomic considerations in endoscopic revision of failed external dacryocystorhinostomy. Otolaryngol Head Neck Surg 2014; 150:901–905.
In this study, 25 repeat endoscopic dacryocystorhinostomies were performed on 22 patients after failed external dacryocystorhinostomy and a CT scan demonstrated anterior ethmoid pneumatization with agger nasi cells in 88% (22/25) of patients, and 95% (21/22) of these partially overlapped the medial aspect of the lacrimal fossa, suggesting that variability of the anterior ethmoid anatomy may contribute to surgical failure following external dacryocystorhinostomy.
35. Kim KS, Kim HU, Chung IH, et al. Surgical anatomy of the nasofrontal duct: anatomical and computed tomographic analysis. Laryngoscope 2001; 111:603–608.
36. Whitnall SE. The relations of the lacrimal fossa to the ethmoidal cells. Ophthal Rev 1911; 30:321–325.
37. Blaylock WK, Moore CA, Linberg JV. Anterior ethmoid anatomy facilitates dacryocystorhinostomy. Arch Ophthalmol 1990; 108:1774–1777.
38. Gokcek A, Argin MA, Altintas AK. Comparison of failed and successful dacryocystorhinostomy by using computed tomographic dacryocystography findings. Eur J Ophthalmol 2005; 15:523–529.
39. Wormald PJ. The agger nasi cell: the key to understanding the anatomy of the frontal recess. Otolaryngol Head Neck Surg 2003; 129:497–507.
40. Tsirbas A, Wormald PJ. Mechanical endonasal dacryocystorhinostomy with mucosal flaps. Otolaryngol Clin North Am 2006; 39:1019–1036.viii.