Anatomy & Physiology
Comparison of Bony Nasolacrimal Canal Narrowing With or Without Primary Acquired Nasolacrimal Duct Obstruction in a Japanese Population Yasuhiro Takahashi, M.D., Ph.D.*, Kouhei Nakata, M.D., Ph.D.†, Hidetaka Miyazaki, D.D.S., Ph.D.‡, Akihiro Ichinose, M.D., Ph.D.§, and Hirohiko Kakizaki, M.D., Ph.D.* *Department of Ophthalmology, Aichi Medical University, Nagakute, Aichi, Japan; †Department of Radiology, Wakayama Medical University, Wakayama, Japan; ‡Department of Stomatology and Oral Surgery, Gunma University Graduate School of Medicine, Maebashi, Gunma, Japan; and §Department of Plastic Surgery, Kobe University School of Medicine, Kobe, Hyogo, Japan
Purpose: To compare the minimum diameter of the nasolacrimal canal and its location between patients with or without primary acquired nasolacrimal duct obstruction (NLDO) in a Japanese population. Methods: One hundred one patients with unilateral primary acquired NLDO (Group A, 101 affected sides; Group B, 101 unaffected sides) and 50 non-NLDO patients (Group C, 100 sides) were included. Anteroposterior and transverse diameters were measured at the canal entrance and the shortest point using contiguous 1-mm axial computed tomographic images. Canal shapes were classified into 2 types: the “funnel” type (a canal with both minimum diameters at the canal entrance) and the “hourglass” type (a canal with at least one minimum diameter in the canal). The distance from the entrance to the part with the shortest diameter was measured on sides with the hourglass type. Results: The funnel type was found more frequently in Groups A and B than in Group C (p < 0.050), although a difference was not found between Groups A and B (p = 0.778). The distance of the transverse diameter was significantly shorter in Groups A and B than Group C (p < 0.050), although no significant difference was found in this distance between Groups A and B (p = 1.000). There were no significant differences between the groups for each value except for the above mentioned (p > 0.050). Conclusions: Primary acquired NLDO patients exhibited the funnel type more frequently or there was a shorter distance from the entrance to the part with the shortest diameter than non-NLDO patients, which may enhance the risk of primary acquired NLDO. (Ophthal Plast Reconstr Surg 2014;30:434–438)
N
arrowing of the bony nasolacrimal canal (BNLC) is thought to be a key factor in the development of primary acquired nasolacrimal duct obstruction (NLDO).1–11 Such narrowing causes tear flow stagnation, debris accumulation, and mucosal adhesion in the nasolacrimal duct.2–5,8 Previous studies reported that narrow passages are found more often in women
Accepted for publication May 20, 2014. The authors have no financial or conflicts of interest to disclose. Address correspondence and reprint requests to Hirohiko Kakizaki, M.D., Ph.D., Department of Ophthalmology, Aichi Medical University, Nagakute, Aichi 480-1195, Japan. E-mail: [email protected] DOI: 10.1097/IOP.0000000000000238
434
and Caucasians, resulting in gender and racial differences in the incidence of primary acquired NLDO.2,5–11 However, the etiology of BNLC narrowing has mostly been investigated in normal populations.3–11 Although Janssen et al.2 demonstrated shorter minimum BNLC diameters in patients with primary acquired NLDO as compared to non-NLDO patients, Phillips and George12 reported a contradictory outcome in that there was no significant difference in the area of the BNLC entrance between NLDO and non-NLDO patients. The narrowest part of the BNLC can be located anywhere throughout the BNLC.5,8,13 However, dacryocystographic and dacryoendoscopic studies showed that the BNLC entrance was the point of obstruction in two-thirds of patients with primary acquired NLDO.14,15 It is assumed, therefore, that the BNLC with the narrowest section at the canal entrance is likely to develop primary acquired NLDO. However, the location of the narrowest section has not been compared between NLDO and non-NLDO patients. Therefore, the anatomical characteristics of the BNLC were examined, including the minimum diameter and its location, in patients with or without primary acquired NLDO in a Japanese population.
MATERIALS AND METHODS In this single institutional, case-controlled study, the records of Japanese patients that had undergone orbital CT at Aichi Medical University from April 2006 to May 2013 were collected. All patients with unilateral primary acquired NLDO were reviewed. The authors diagnosed primary acquired NLDO with a patient history, lacrimal irrigation, probing, dacryoendoscopy, and/or dacryocystography, and routinely performed CT in patients with NLDO to identify nasal pathology and to rule out secondary NLDO by a lacrimal sac tumor. Patients with incomplete primary acquired NLDO, sole canalicular obstruction, congenital NLDO, and secondary NLDO induced by a tumor, trauma, or sarcoidosis were excluded from this study. All patients with free lacrimal irrigation and no BNLC fracture despite blunt ocular trauma were also reviewed as age-matched non-NLDO patients. The patients were divided into 3 groups: Group A included the affected sides in patients with unilateral primary acquired NLDO; Group B comprised the unaffected sides of the same patients; and Group C included non-NLDO patients. The Institutional Review Board approval from Aichi Medical University (No. 13–016) was obtained and the tenets of the Declaration of Helsinki were followed. CT studies were performed using a high-speed scanner (Aquillion 64; Toshiba, Tokyo, Japan). Contiguous 1-mm axial Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
images were obtained through the bilateral optic nerves using a bone window algorithm (width: 2500; level: 500). Anatomical measurements were made by one of the authors (YT) using the caliper tool of a viewer (ShadeQuest/ViewR; Yokogawa Medical Solutions Corporation, Tokyo, Japan). The anteroposterior and transverse diameters of the BNLC entrance were measured first. The inferior orbital rim was used as the reference plane at the canal entrance.3,8,9 The anteroposterior and transverse diameters of the BNLC on every image to find the minimum diameter were then measured (Fig.). The authors classified the shape of the BNLC as 1 of the following 2 types.8 A “funnel” type was defined as a BNLC with both minimum anteroposterior and transverse diameters at the BNLC entrance. An “hourglass” type was defined as a BNLC with a minimum anteroposterior and/or transverse diameter at a certain point in the canal. The number of image slices was defined as the distance from the entrance to the section with the minimal diameter on sides with the hourglass type. The reduction in diameter was calculated on sides with the hourglass type by subtracting the minimum diameter from the diameter at the BNLC entrance. The percent decrease was calculated as the reduction in diameter/diameter at the BNLC entrance × 100. Patient age and measured values are expressed as mean value ± standard deviation. The gender distribution was compared using a χ2 test for independent variables between the groups and using a Student t test in each group. Age was compared between the groups and between genders in
A computed tomographic image, axial view. The anteroposterior (solid arrow) and transverse diameters (dotted arrow) are measured.
BNLC Narrowing and Primary Acquired NLDO
each group using a Student t test. The measurements were compared between the groups using a 1-way ANOVA and a Bonferroni correction. The measurement results were compared between genders in each group using a Student t test or a Mann-Whitney U test depending on the number of sides. For values 0.050). The analysis of gender differences showed that transverse diameters at the BNLC entrance were shorter in women than in men in Group A (p = 0.007) and Group B (p = 0.023). The transverse minimum diameter was also shorter in women than in men in Group A (p = 0.017). The percent decrease was larger in women than in men in Group A (p = 0.041). The funnel type tended to be more frequently observed in women than in men in Group A; however, the difference did not reach statistical significance (p = 0.060). In Group C, the reduction and percent decrease of the anteroposterior diameter were significantly larger in men than women (p < 0.050). All other measurements were similar in both groups (p > 0.050).
DISCUSSION The funnel type was more common among patients with NLDO (Groups A and B) than non-NLDO patients (Group C).
TABLE 1. Patient characteristics p: intergroup difference
Number Patients/sides M/F Age (range) (years) Total M F p Value: gender difference
Group A & B
Group C
Group A and B vs. C
101/101 26/75
50/100 20/30
0.073*
65.16 ± 12.83 (29–90) 61.31 ± 14.66 (29–87) 66.49 ± 11.96 (40–90) 0.076†
63.40 ± 13.43 (42–91) 60.80 ± 9.46 (50–81) 65.13 ± 15.44 (42–91) 0.268†
0.971†
No statistical significance using a *χ2 test for independence, or †Student t test. M, male; F, female.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
435
436 53 (52.48) 48 (47.52)
5.32 ± 1.13 (3.01–8.74) 4.96 ± 1.15 (2.57–8.74) 0.82 ± 0.53 (0.14–2.23) 15.19 ± 8.26 (2.65––34.98) 4.64 ± 3.05 (1.00–16.00)
5.43 ± 1.08 (3.39–8.62) 5.09 ± 1.16 (2.75–8.61) 0.79 ± 0.44 (0.16–1.80) 14.95 ± 8.26 (3.02–36.44) 4.28 ± 2.92 (1.00–16.00)
51 (50.50) 50 (49.50)
6.34 ± 1.37 (3.82–10.47) 6.21 ± 1.35 (3.75–9.62) 0.85 ± 0.58 (0.22–2.18) 12.87 ± 8.30 (4.27–32.74) 4.43 ± 2.71 (1.00–12.00)
6.43 ± 1.38 (3.88–9.46) 6.29 ± 1.40 (3.16–9.46) 0.79 ± 0.55 (0.21–2.08) 12.19 ± 7.29 (2.64–29.80) 4.11 ± 2.08 (1.00–10.00)
Group B
NLDO
Measurement value
Reduction, hourglass type: The reduction in diameter from the entrance to the point of minimum diameter. Percent decrease, hourglass type: The reduction in each diameter/diameter at the BNLC entrance × 100. Distance: The distance from the BNLC entrance to the point of minimum diameter. No statistical significance using a *one-way ANOVA, § Bonferroni correction, or a ║χ2 test for independence. Statistical significance using a †one-way ANOVA, ‡ Bonferroni correction, or a ¶χ2 test for independence. NLDO, nasolacrimal duct obstruction; BNLC, bony nasolacrimal canal.
Number of BNLC Funnel type (%) Hourglass type (%)
Anteroposterior diameter (mm) Entrance (range) Minimum (range) Reduction (range) Decreasing rate (range) (%) Distance (range) (mm) Transverse diameter (mm) Entrance (range) Minimum (range) Reduction (range) Decreasing rate (range) (%) Distance (range) (mm)
Group A
TABLE 2. A group comparison of measurements
34 (34.00) 66 (66.00)
5.43 ± 1.19 (3.52–10.94) 4.80 ± 0.86 (2.85–7.34) 1.00 ± 0.74 (0.11–4.53) 16.59 ± 8.99 (2.16–41.41) 6.48 ± 3.99 (1.00–16.00)
6.48 ± 1.30 (4.39–10.88) 6.35 ± 1.30 (4.02–10.88) 0.81 ± 0.48 (0.24–1.63) 11.88 ± 6.82 (4.17–24.29) 5.19 ± 3.94 (1.00–13.00)
Group C
0.018¶
0.004‡
Group A vs. C
0.008¶
0.711* 0.142* 0.172* 0.555* 0.002† 0.021‡
0.775* 0.766* 0.950* 0.929* 0.554*
Group B vs. C
p
0.778║
1.000§
Group A vs. B
Y. Takahashi et al. Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
M F M F
M F M F M F M F M F
M F M F M F M F M F
9 (34.62) 42 (56.00) 17 (65.38) 33 (44.00)
5.92 ± 1.46 (3.49–8.62) 5.26 ± 0.86 (3.39–7.05) 5.56 ± 1.51 (3.39–8.61) 4.93 ± 0.98 (2.75–7.02) 0.67 ± 0.45 (0.16–1.77) 0.85 ± 0.44 (0.28–1.80) 11.77 ± 8.05 (3.02–34.30) 16.48 ± 8.04 (5.47–36.44) 4.86 ± 3.68 (1.00–16.00) 4.00 ± 2.51 (1.00–10.00)
6.74 ± 1.29 (4.49–9.10) 6.33 ± 1.41 (3.88–9.46) 6.51 ± 1.31 (4.20–9.10) 6.21 ± 1.44 (3.16–9.46) 0.87 ± 0.44 (0.45–1.73) 0.74 ± 0.61 (0.21–2.08) 12.83 ± 4.47 (6.68–19.70) 11.82 ± 8.69 (2.64–29.80) 4.71 ± 2.56 (3.00–10.00) 3.75 ± 1.76 (1.00–6.00)
NLDO
10 (38.46) 43 (57.33) 16 (61.54) 32 (42.67)
5.75 ± 1.47 (3.14–8.74) 5.16 ± 0.95 (3.01–7.70) 5.27 ± 1.40 (2.64–8.74) 4.85 ± 1.03 (2.57–7.70) 0.89 ± 0.68 (0.19–2.23) 0.79 ± 0.45 (0.14–1.70) 14.66 ± 8.58 (3.85–30.31) 15.44 ± 8.24 (2.65–34.98) 5.57 ± 3.96 (1.00–16.00) 4.20 ± 2.48 (1.00–9.00)
6.62 ± 1.48 (4.33–9.42) 6.25 ± 1.33 (3.82–10.47) 6.43 ± 1.40 (4.33–9.25) 6.14 ± 1.33 (3.75–9.62) 0.81 ± 0.76 (0.22–2.18) 0.87 ± 0.49 (0.28–1.83) 11.02 ± 8.52 (4.27–23.14) 13.98 ± 8.42 (5.08–32.74) 5.33 ± 3.72 (1.00–12.00) 3.90 ± 1.91 (1.00–6.00)
Group B
Group C
15 (37.50) 19 (31.67) 25 (62.50) 41 (68.33)
5.42 ± 0.97 (4.21–8.62) 5.46 ± 1.32 (3.52–10.94) 4.83 ± 0.88 (2.85–7.34) 4.78 ± 0.85 (3.03–6.85) 0.97 ± 0.64 (0.11–2.45) 1.01 ± 0.80 (0.11–4.53) 17.03 ± 10.14 (2.16–33.03) 16.33 ± 8.34 (2.58–41.41) 7.08 ± 3.67 (2.00–15.00) 6.13 ± 4.17 (1.00–16.00)
6.63 ± 1.22 (4.78–10.88) 6.40 ± 1.34 (4.39–9.12) 6.43 ± 1.24 (4.52–10.88) 6.32 ± 1.34 (4.02–9.12) 1.14 ± 0.33 (0.67–1.45) 0.54 ± 0.43 (0.24–1.63) 16.74 ± 5.52 (9.45–24.29) 8.10 ± 5.26 (4.17–20.55) 6.71 ± 3.95 (3.00–13.00) 4.00 ± 3.71 (1.00–13.00)
Reduction: The reduction of each diameter from the diameter at the BNLC entrance to the minimum diameter on sides with the hourglass shape. Percent decrease: The reduction of each diameter/the diameter at the BNLC entrance × 100 on sides with the hourglass shape. Distance: The distance from the BNLC entrance to the section of minimum diameter. No statistical significance as determined with a *Student t test, a †Mann-Whitney U test, or a ¶χ2 test for independence. Statistical significance according to a ‡Mann-Whitney U test or §Student t test. M, male; F, female; NLDO, nasolacrimal duct obstruction; BNLC, bony nasolacrimal canal.
Hourglass type (%)
No. BNLC Funnel type (%)
Distance (range) (mm)
Decreasing rate (range) (%)
Reduction (range)
Minimum (range)
Transverse diameter (mm) Entrance (range)
Distance (range) (mm)
Decreasing rate (range) (%)
Reduction (range)
Minimum (range)
Anteroposterior diameter (mm) Entrance (range)
Group A
Measurement value
TABLE 3. A summary of measurements and comparison between genders
0.060¶
0.444†
0.041‡
0.165†
0.017§
0.007§
0.592†
0.384†
0.299†
0.352*
0.190*
Group A
NLDO
0.097¶
0.357†
0.650†
0.980†
0.106*
0.023§
0.635†
0.368†
0.635†
0.345*
0.247*
Group B
p
0.546¶
0.356‡
0.765*
0.838*
0.722*
0.908†
0.091†
0.006‡
0.012‡
0.605*
0.337*
Group C
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
BNLC Narrowing and Primary Acquired NLDO
437
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
Y. Takahashi et al.
In addition, in comparison to the measurement values among the subjects with the hourglass type, the distance from the entrance to the section with the minimum diameter was shorter in patients with NLDO than in non-NLDO patients. This study suggests that the BNLC with the narrowest point at or near the entrance enhances the risk of primary acquired NLDO. A change in the lumen diameter of the lacrimal passage influences tear flow resistance.3 The BNLC entrance is, by nature, where the mucosal lacrimal duct lumen narrows.15 This is because the nasolacrimal duct is confined in the osseous passage, and the cavernous body of the nasolacrimal duct is more developed and thickened than that of the lacrimal sac wall.16 The funnel type leads to further reduction in the mucosal lacrimal duct lumen at the entrance. On the other hand, the finding that NLDO patients with the hourglass type showed a shorter distance indicates a steeper rate of BNLC narrowing from the entrance to the point of the minimum diameter than non-NLDO patients. Both findings may, therefore, contribute to elevation of tear flow resistance. This study demonstrated a lack of significant differences in the entrance and minimum diameters between the groups, which supports the results reported previously by Phillips and George.12 However, this contrasts with results reported by Janssen et al.2 One possible reason for this discrepancy may be racial characteristics of the populations used in each study. The anatomy of the bony lacrimal passage is known to differ in various races.4,6,10,11 Another point of discrepancy is the methods used for measurement in each study. Phillips and George12 used head x-rays, while Janssen et al.2 used CT with slices of 2-mm thickness. This study used CT with more narrow slices to increase the precision of the measurements. Although the lack of significant differences in the diameters between the groups appeared to be inconsistent with the etiology of primary acquired NLDO, it is necessary to consider that NLDO is influenced by multiple factors including thickening and/or fibrosis of the nasolacrimal duct mucosa.1,5,17,18 Gender differences in the anatomy of the bony lacrimal passage are thought to be the reason for a higher incidence of primary acquired NLDO in women.4–11,19 This study showed that the funnel type of canal tended to be more common in women and NLDO patients, indicating that a funnel-shaped BNLC may increase the incidence of NLDO in women. Female subjects also exhibited smaller transverse diameters at the BNLC entrance and at the narrowest point of the canal; the decrease in diameter was also steeper in female NLDO patients. However, as these values did not reveal statistically significant differences between NLDO and non-NLDO patients, this study was not able to verify that the gender differences in transverse diameter and/or magnitude of the reduction in diameter were associated with a higher incidence of primary acquired NLDO in women. Lagging obstruction on the contralateral side occasionally occurs in patients with unilateral primary acquired NLDO.5,8 The lack of any difference between Groups A and B may support this progression. Although probing and silicone intubation are occasionally performed in patients with NLDO,20 false passage formation is a troublesome complication. In patients with a firm NLDO at the BNLC entrance, a probe is likely to enter the orbit.21 In patients with an NLDO below the BNLC entrance, the risk of false passage formation is lower because the obstruction point is surrounded by the BNLC.21 This study demonstrated a higher incidence of the funnel type in patients with NLDO, suggesting that dacryocystorhinostomy is preferable to probing and silicone intubation. This study is limited by several factors. The first is the design, which was retrospective. Notably, the patients in Groups B and C may suffer from primary acquired NLDO in the future.
438
The second is the inclusion of only Japanese patients. The third is measurement errors inherent to CT, such as those caused by interslice gaps.2,7,8 In conclusion, the patients with primary acquired NLDO were more likely to have a funnel-type BNLC and exhibited a steeper rate of narrowing than non-NLDO patients. On the other hand, this study demonstrated a lack of significant differences in the entrance and minimum diameters between the groups. This study suggests that the BNLC with the narrowest point at or near the entrance enhances the risk of primary acquired NLDO.
REFERENCES 1. Linberg JV, McCormick SA. Primary acquired nasolacrimal duct obstruction. A clinicopathologic report and biopsy technique. Ophthalmology 1986;93:1055–63. 2. Janssen AG, Mansour K, Bos JJ, Castelijins JA. Diameter of the bony lacrimal canal: normal values and values related to nasolacrimal duct obstruction: assessment with CT. AJNR Am J Neuroradiol 2001;22:845–50. 3. Lee H, Ha S, Lee Y, et al. Anatomical and morphometric study of the bony nasolacrimal canal using computed tomography. Ophthalmologica 2012;227:153–9. 4. McCormick A, Sloan B. The diameter of the nasolacrimal canal measured by computed tomography: gender and racial differences. Clin Experiment Ophthalmol 2009;37:357–61. 5. Groessl SA, Sires BS, Lemke BN. An anatomical basis for primary acquired nasolacrimal duct obstruction. Arch Ophthalmol 1997;115:71–4. 6. Santos-Fernandez J. The measurement of the nasal canal according to the race. Am J Ophthalmol 1921;4:32–7. 7. Takahashi Y, Kakizaki H, Nakano T. Bony nasolacrimal duct entrance diameter: gender difference in cadaveric study. Ophthal Plast Reconstr Surg 2011;27:204–5. 8. Takahashi Y, Nakamura Y, Nakano T, et al. The narrowest part of the bony nasolacrimal canal: an anatomical study. Ophthal Plast Reconstr Surg 2013;29:318–22. 9. Shigeta K, Takegoshi H, Kikuchi S. Sex and age differences in the bony nasolacrimal canal. Arch Ophthalmol 2007;125:1677–81. 10. Fasina O, Ogbole GI. CT assessment of the nasolacrimal ca nal in a black African Population. Ophthal Plast Reconstr Surg 2013;29:231–3. 11. Post RH. Tear duct size differences of age, sex and race. Am J Phys Anthropol 1969;30:85–8. 12. Phillips CI, George M. Epiphora and the bony naso-lacrimal canal. Br J Ophthalmol 1956;40:673–80. 13. Groell R, Schaffler GJ, Uggowitzer M, et al. CT-anatomy of the nasolacrimal sac and duct. Surg Radiol Anat 1997;19:189–91. 14. Francisco FC, Carvalho AC, Francisco VF, et al. Evaluation of 1000 lacrimal ducts by dacryocystography. Br J Ophthalmol 2007;91:43–6. 15. Sasaki T, Nagata Y, Sugiyama K. Nasolacrimal duct obstruction classified by dacryoendoscopy and treated with inferior meatal dacryorhinotomy. Part I: Positional diagnosis of primary nasolacrimal duct obstruction with dacryoendoscope. Am J Ophthalmol 2005;140:1065–9. 16. Narioka J, Ohashi Y. Changes in lumen width of nasolacrimal drainage system after adrenergic and cholinergic stimulation. Am J Ophthalmol 2006;141:689–98. 17. Lee-Wing MW, Ashenhurst ME. Clinicopathologic analysis of 166 patients with primary acquired nasolacrimal duct obstruction. Ophthalmology 2001;108:2038–40. 18. Tucker N, Chow D, Stockl F, et al. Clinically suspected primary acquired nasolacrimal duct obstruction: clinicopathologic review of 150 patients. Ophthalmology 1997;104:1882–6. 19. Takahashi Y, Nakamura Y, Nakano T, et al. Horizontal orientation of the bony lacrimal passage: an anatomical study. Ophthal Plast Reconstr Surg 2013;29:128–30. 20. Connell PP, Fulcher TP, Chacko E, et al. Long term follow up of nasolacrimal intubation in adults. Br J Ophthalmol 2006;90:435–6. 21. Ikoma M. Indications of silicone intubation in our institution [in Japanese]. Ganka Rinsho Iho. 1998;92:1387–8.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
Comparison of Bony Nasolacrimal Canal Narrowing With or Without Primary Acquired Nasolacrimal Duct Obstruction in a Japanese Population Yasuhiro Takahashi, M.D., Ph.D.*, Kouhei Nakata, M.D., Ph.D.†, Hidetaka Miyazaki, D.D.S., Ph.D.‡, Akihiro Ichinose, M.D., Ph.D.§, and Hirohiko Kakizaki, M.D., Ph.D.* *Department of Ophthalmology, Aichi Medical University, Nagakute, Aichi, Japan; †Department of Radiology, Wakayama Medical University, Wakayama, Japan; ‡Department of Stomatology and Oral Surgery, Gunma University Graduate School of Medicine, Maebashi, Gunma, Japan; and §Department of Plastic Surgery, Kobe University School of Medicine, Kobe, Hyogo, Japan
Purpose: To compare the minimum diameter of the nasolacrimal canal and its location between patients with or without primary acquired nasolacrimal duct obstruction (NLDO) in a Japanese population. Methods: One hundred one patients with unilateral primary acquired NLDO (Group A, 101 affected sides; Group B, 101 unaffected sides) and 50 non-NLDO patients (Group C, 100 sides) were included. Anteroposterior and transverse diameters were measured at the canal entrance and the shortest point using contiguous 1-mm axial computed tomographic images. Canal shapes were classified into 2 types: the “funnel” type (a canal with both minimum diameters at the canal entrance) and the “hourglass” type (a canal with at least one minimum diameter in the canal). The distance from the entrance to the part with the shortest diameter was measured on sides with the hourglass type. Results: The funnel type was found more frequently in Groups A and B than in Group C (p < 0.050), although a difference was not found between Groups A and B (p = 0.778). The distance of the transverse diameter was significantly shorter in Groups A and B than Group C (p < 0.050), although no significant difference was found in this distance between Groups A and B (p = 1.000). There were no significant differences between the groups for each value except for the above mentioned (p > 0.050). Conclusions: Primary acquired NLDO patients exhibited the funnel type more frequently or there was a shorter distance from the entrance to the part with the shortest diameter than non-NLDO patients, which may enhance the risk of primary acquired NLDO. (Ophthal Plast Reconstr Surg 2014;30:434–438)
N
arrowing of the bony nasolacrimal canal (BNLC) is thought to be a key factor in the development of primary acquired nasolacrimal duct obstruction (NLDO).1–11 Such narrowing causes tear flow stagnation, debris accumulation, and mucosal adhesion in the nasolacrimal duct.2–5,8 Previous studies reported that narrow passages are found more often in women
Accepted for publication May 20, 2014. The authors have no financial or conflicts of interest to disclose. Address correspondence and reprint requests to Hirohiko Kakizaki, M.D., Ph.D., Department of Ophthalmology, Aichi Medical University, Nagakute, Aichi 480-1195, Japan. E-mail: [email protected] DOI: 10.1097/IOP.0000000000000238
434
and Caucasians, resulting in gender and racial differences in the incidence of primary acquired NLDO.2,5–11 However, the etiology of BNLC narrowing has mostly been investigated in normal populations.3–11 Although Janssen et al.2 demonstrated shorter minimum BNLC diameters in patients with primary acquired NLDO as compared to non-NLDO patients, Phillips and George12 reported a contradictory outcome in that there was no significant difference in the area of the BNLC entrance between NLDO and non-NLDO patients. The narrowest part of the BNLC can be located anywhere throughout the BNLC.5,8,13 However, dacryocystographic and dacryoendoscopic studies showed that the BNLC entrance was the point of obstruction in two-thirds of patients with primary acquired NLDO.14,15 It is assumed, therefore, that the BNLC with the narrowest section at the canal entrance is likely to develop primary acquired NLDO. However, the location of the narrowest section has not been compared between NLDO and non-NLDO patients. Therefore, the anatomical characteristics of the BNLC were examined, including the minimum diameter and its location, in patients with or without primary acquired NLDO in a Japanese population.
MATERIALS AND METHODS In this single institutional, case-controlled study, the records of Japanese patients that had undergone orbital CT at Aichi Medical University from April 2006 to May 2013 were collected. All patients with unilateral primary acquired NLDO were reviewed. The authors diagnosed primary acquired NLDO with a patient history, lacrimal irrigation, probing, dacryoendoscopy, and/or dacryocystography, and routinely performed CT in patients with NLDO to identify nasal pathology and to rule out secondary NLDO by a lacrimal sac tumor. Patients with incomplete primary acquired NLDO, sole canalicular obstruction, congenital NLDO, and secondary NLDO induced by a tumor, trauma, or sarcoidosis were excluded from this study. All patients with free lacrimal irrigation and no BNLC fracture despite blunt ocular trauma were also reviewed as age-matched non-NLDO patients. The patients were divided into 3 groups: Group A included the affected sides in patients with unilateral primary acquired NLDO; Group B comprised the unaffected sides of the same patients; and Group C included non-NLDO patients. The Institutional Review Board approval from Aichi Medical University (No. 13–016) was obtained and the tenets of the Declaration of Helsinki were followed. CT studies were performed using a high-speed scanner (Aquillion 64; Toshiba, Tokyo, Japan). Contiguous 1-mm axial Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
images were obtained through the bilateral optic nerves using a bone window algorithm (width: 2500; level: 500). Anatomical measurements were made by one of the authors (YT) using the caliper tool of a viewer (ShadeQuest/ViewR; Yokogawa Medical Solutions Corporation, Tokyo, Japan). The anteroposterior and transverse diameters of the BNLC entrance were measured first. The inferior orbital rim was used as the reference plane at the canal entrance.3,8,9 The anteroposterior and transverse diameters of the BNLC on every image to find the minimum diameter were then measured (Fig.). The authors classified the shape of the BNLC as 1 of the following 2 types.8 A “funnel” type was defined as a BNLC with both minimum anteroposterior and transverse diameters at the BNLC entrance. An “hourglass” type was defined as a BNLC with a minimum anteroposterior and/or transverse diameter at a certain point in the canal. The number of image slices was defined as the distance from the entrance to the section with the minimal diameter on sides with the hourglass type. The reduction in diameter was calculated on sides with the hourglass type by subtracting the minimum diameter from the diameter at the BNLC entrance. The percent decrease was calculated as the reduction in diameter/diameter at the BNLC entrance × 100. Patient age and measured values are expressed as mean value ± standard deviation. The gender distribution was compared using a χ2 test for independent variables between the groups and using a Student t test in each group. Age was compared between the groups and between genders in
A computed tomographic image, axial view. The anteroposterior (solid arrow) and transverse diameters (dotted arrow) are measured.
BNLC Narrowing and Primary Acquired NLDO
each group using a Student t test. The measurements were compared between the groups using a 1-way ANOVA and a Bonferroni correction. The measurement results were compared between genders in each group using a Student t test or a Mann-Whitney U test depending on the number of sides. For values 0.050). The analysis of gender differences showed that transverse diameters at the BNLC entrance were shorter in women than in men in Group A (p = 0.007) and Group B (p = 0.023). The transverse minimum diameter was also shorter in women than in men in Group A (p = 0.017). The percent decrease was larger in women than in men in Group A (p = 0.041). The funnel type tended to be more frequently observed in women than in men in Group A; however, the difference did not reach statistical significance (p = 0.060). In Group C, the reduction and percent decrease of the anteroposterior diameter were significantly larger in men than women (p < 0.050). All other measurements were similar in both groups (p > 0.050).
DISCUSSION The funnel type was more common among patients with NLDO (Groups A and B) than non-NLDO patients (Group C).
TABLE 1. Patient characteristics p: intergroup difference
Number Patients/sides M/F Age (range) (years) Total M F p Value: gender difference
Group A & B
Group C
Group A and B vs. C
101/101 26/75
50/100 20/30
0.073*
65.16 ± 12.83 (29–90) 61.31 ± 14.66 (29–87) 66.49 ± 11.96 (40–90) 0.076†
63.40 ± 13.43 (42–91) 60.80 ± 9.46 (50–81) 65.13 ± 15.44 (42–91) 0.268†
0.971†
No statistical significance using a *χ2 test for independence, or †Student t test. M, male; F, female.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
435
436 53 (52.48) 48 (47.52)
5.32 ± 1.13 (3.01–8.74) 4.96 ± 1.15 (2.57–8.74) 0.82 ± 0.53 (0.14–2.23) 15.19 ± 8.26 (2.65––34.98) 4.64 ± 3.05 (1.00–16.00)
5.43 ± 1.08 (3.39–8.62) 5.09 ± 1.16 (2.75–8.61) 0.79 ± 0.44 (0.16–1.80) 14.95 ± 8.26 (3.02–36.44) 4.28 ± 2.92 (1.00–16.00)
51 (50.50) 50 (49.50)
6.34 ± 1.37 (3.82–10.47) 6.21 ± 1.35 (3.75–9.62) 0.85 ± 0.58 (0.22–2.18) 12.87 ± 8.30 (4.27–32.74) 4.43 ± 2.71 (1.00–12.00)
6.43 ± 1.38 (3.88–9.46) 6.29 ± 1.40 (3.16–9.46) 0.79 ± 0.55 (0.21–2.08) 12.19 ± 7.29 (2.64–29.80) 4.11 ± 2.08 (1.00–10.00)
Group B
NLDO
Measurement value
Reduction, hourglass type: The reduction in diameter from the entrance to the point of minimum diameter. Percent decrease, hourglass type: The reduction in each diameter/diameter at the BNLC entrance × 100. Distance: The distance from the BNLC entrance to the point of minimum diameter. No statistical significance using a *one-way ANOVA, § Bonferroni correction, or a ║χ2 test for independence. Statistical significance using a †one-way ANOVA, ‡ Bonferroni correction, or a ¶χ2 test for independence. NLDO, nasolacrimal duct obstruction; BNLC, bony nasolacrimal canal.
Number of BNLC Funnel type (%) Hourglass type (%)
Anteroposterior diameter (mm) Entrance (range) Minimum (range) Reduction (range) Decreasing rate (range) (%) Distance (range) (mm) Transverse diameter (mm) Entrance (range) Minimum (range) Reduction (range) Decreasing rate (range) (%) Distance (range) (mm)
Group A
TABLE 2. A group comparison of measurements
34 (34.00) 66 (66.00)
5.43 ± 1.19 (3.52–10.94) 4.80 ± 0.86 (2.85–7.34) 1.00 ± 0.74 (0.11–4.53) 16.59 ± 8.99 (2.16–41.41) 6.48 ± 3.99 (1.00–16.00)
6.48 ± 1.30 (4.39–10.88) 6.35 ± 1.30 (4.02–10.88) 0.81 ± 0.48 (0.24–1.63) 11.88 ± 6.82 (4.17–24.29) 5.19 ± 3.94 (1.00–13.00)
Group C
0.018¶
0.004‡
Group A vs. C
0.008¶
0.711* 0.142* 0.172* 0.555* 0.002† 0.021‡
0.775* 0.766* 0.950* 0.929* 0.554*
Group B vs. C
p
0.778║
1.000§
Group A vs. B
Y. Takahashi et al. Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
M F M F
M F M F M F M F M F
M F M F M F M F M F
9 (34.62) 42 (56.00) 17 (65.38) 33 (44.00)
5.92 ± 1.46 (3.49–8.62) 5.26 ± 0.86 (3.39–7.05) 5.56 ± 1.51 (3.39–8.61) 4.93 ± 0.98 (2.75–7.02) 0.67 ± 0.45 (0.16–1.77) 0.85 ± 0.44 (0.28–1.80) 11.77 ± 8.05 (3.02–34.30) 16.48 ± 8.04 (5.47–36.44) 4.86 ± 3.68 (1.00–16.00) 4.00 ± 2.51 (1.00–10.00)
6.74 ± 1.29 (4.49–9.10) 6.33 ± 1.41 (3.88–9.46) 6.51 ± 1.31 (4.20–9.10) 6.21 ± 1.44 (3.16–9.46) 0.87 ± 0.44 (0.45–1.73) 0.74 ± 0.61 (0.21–2.08) 12.83 ± 4.47 (6.68–19.70) 11.82 ± 8.69 (2.64–29.80) 4.71 ± 2.56 (3.00–10.00) 3.75 ± 1.76 (1.00–6.00)
NLDO
10 (38.46) 43 (57.33) 16 (61.54) 32 (42.67)
5.75 ± 1.47 (3.14–8.74) 5.16 ± 0.95 (3.01–7.70) 5.27 ± 1.40 (2.64–8.74) 4.85 ± 1.03 (2.57–7.70) 0.89 ± 0.68 (0.19–2.23) 0.79 ± 0.45 (0.14–1.70) 14.66 ± 8.58 (3.85–30.31) 15.44 ± 8.24 (2.65–34.98) 5.57 ± 3.96 (1.00–16.00) 4.20 ± 2.48 (1.00–9.00)
6.62 ± 1.48 (4.33–9.42) 6.25 ± 1.33 (3.82–10.47) 6.43 ± 1.40 (4.33–9.25) 6.14 ± 1.33 (3.75–9.62) 0.81 ± 0.76 (0.22–2.18) 0.87 ± 0.49 (0.28–1.83) 11.02 ± 8.52 (4.27–23.14) 13.98 ± 8.42 (5.08–32.74) 5.33 ± 3.72 (1.00–12.00) 3.90 ± 1.91 (1.00–6.00)
Group B
Group C
15 (37.50) 19 (31.67) 25 (62.50) 41 (68.33)
5.42 ± 0.97 (4.21–8.62) 5.46 ± 1.32 (3.52–10.94) 4.83 ± 0.88 (2.85–7.34) 4.78 ± 0.85 (3.03–6.85) 0.97 ± 0.64 (0.11–2.45) 1.01 ± 0.80 (0.11–4.53) 17.03 ± 10.14 (2.16–33.03) 16.33 ± 8.34 (2.58–41.41) 7.08 ± 3.67 (2.00–15.00) 6.13 ± 4.17 (1.00–16.00)
6.63 ± 1.22 (4.78–10.88) 6.40 ± 1.34 (4.39–9.12) 6.43 ± 1.24 (4.52–10.88) 6.32 ± 1.34 (4.02–9.12) 1.14 ± 0.33 (0.67–1.45) 0.54 ± 0.43 (0.24–1.63) 16.74 ± 5.52 (9.45–24.29) 8.10 ± 5.26 (4.17–20.55) 6.71 ± 3.95 (3.00–13.00) 4.00 ± 3.71 (1.00–13.00)
Reduction: The reduction of each diameter from the diameter at the BNLC entrance to the minimum diameter on sides with the hourglass shape. Percent decrease: The reduction of each diameter/the diameter at the BNLC entrance × 100 on sides with the hourglass shape. Distance: The distance from the BNLC entrance to the section of minimum diameter. No statistical significance as determined with a *Student t test, a †Mann-Whitney U test, or a ¶χ2 test for independence. Statistical significance according to a ‡Mann-Whitney U test or §Student t test. M, male; F, female; NLDO, nasolacrimal duct obstruction; BNLC, bony nasolacrimal canal.
Hourglass type (%)
No. BNLC Funnel type (%)
Distance (range) (mm)
Decreasing rate (range) (%)
Reduction (range)
Minimum (range)
Transverse diameter (mm) Entrance (range)
Distance (range) (mm)
Decreasing rate (range) (%)
Reduction (range)
Minimum (range)
Anteroposterior diameter (mm) Entrance (range)
Group A
Measurement value
TABLE 3. A summary of measurements and comparison between genders
0.060¶
0.444†
0.041‡
0.165†
0.017§
0.007§
0.592†
0.384†
0.299†
0.352*
0.190*
Group A
NLDO
0.097¶
0.357†
0.650†
0.980†
0.106*
0.023§
0.635†
0.368†
0.635†
0.345*
0.247*
Group B
p
0.546¶
0.356‡
0.765*
0.838*
0.722*
0.908†
0.091†
0.006‡
0.012‡
0.605*
0.337*
Group C
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
BNLC Narrowing and Primary Acquired NLDO
437
Ophthal Plast Reconstr Surg, Vol. 30, No. 5, 2014
Y. Takahashi et al.
In addition, in comparison to the measurement values among the subjects with the hourglass type, the distance from the entrance to the section with the minimum diameter was shorter in patients with NLDO than in non-NLDO patients. This study suggests that the BNLC with the narrowest point at or near the entrance enhances the risk of primary acquired NLDO. A change in the lumen diameter of the lacrimal passage influences tear flow resistance.3 The BNLC entrance is, by nature, where the mucosal lacrimal duct lumen narrows.15 This is because the nasolacrimal duct is confined in the osseous passage, and the cavernous body of the nasolacrimal duct is more developed and thickened than that of the lacrimal sac wall.16 The funnel type leads to further reduction in the mucosal lacrimal duct lumen at the entrance. On the other hand, the finding that NLDO patients with the hourglass type showed a shorter distance indicates a steeper rate of BNLC narrowing from the entrance to the point of the minimum diameter than non-NLDO patients. Both findings may, therefore, contribute to elevation of tear flow resistance. This study demonstrated a lack of significant differences in the entrance and minimum diameters between the groups, which supports the results reported previously by Phillips and George.12 However, this contrasts with results reported by Janssen et al.2 One possible reason for this discrepancy may be racial characteristics of the populations used in each study. The anatomy of the bony lacrimal passage is known to differ in various races.4,6,10,11 Another point of discrepancy is the methods used for measurement in each study. Phillips and George12 used head x-rays, while Janssen et al.2 used CT with slices of 2-mm thickness. This study used CT with more narrow slices to increase the precision of the measurements. Although the lack of significant differences in the diameters between the groups appeared to be inconsistent with the etiology of primary acquired NLDO, it is necessary to consider that NLDO is influenced by multiple factors including thickening and/or fibrosis of the nasolacrimal duct mucosa.1,5,17,18 Gender differences in the anatomy of the bony lacrimal passage are thought to be the reason for a higher incidence of primary acquired NLDO in women.4–11,19 This study showed that the funnel type of canal tended to be more common in women and NLDO patients, indicating that a funnel-shaped BNLC may increase the incidence of NLDO in women. Female subjects also exhibited smaller transverse diameters at the BNLC entrance and at the narrowest point of the canal; the decrease in diameter was also steeper in female NLDO patients. However, as these values did not reveal statistically significant differences between NLDO and non-NLDO patients, this study was not able to verify that the gender differences in transverse diameter and/or magnitude of the reduction in diameter were associated with a higher incidence of primary acquired NLDO in women. Lagging obstruction on the contralateral side occasionally occurs in patients with unilateral primary acquired NLDO.5,8 The lack of any difference between Groups A and B may support this progression. Although probing and silicone intubation are occasionally performed in patients with NLDO,20 false passage formation is a troublesome complication. In patients with a firm NLDO at the BNLC entrance, a probe is likely to enter the orbit.21 In patients with an NLDO below the BNLC entrance, the risk of false passage formation is lower because the obstruction point is surrounded by the BNLC.21 This study demonstrated a higher incidence of the funnel type in patients with NLDO, suggesting that dacryocystorhinostomy is preferable to probing and silicone intubation. This study is limited by several factors. The first is the design, which was retrospective. Notably, the patients in Groups B and C may suffer from primary acquired NLDO in the future.
438
The second is the inclusion of only Japanese patients. The third is measurement errors inherent to CT, such as those caused by interslice gaps.2,7,8 In conclusion, the patients with primary acquired NLDO were more likely to have a funnel-type BNLC and exhibited a steeper rate of narrowing than non-NLDO patients. On the other hand, this study demonstrated a lack of significant differences in the entrance and minimum diameters between the groups. This study suggests that the BNLC with the narrowest point at or near the entrance enhances the risk of primary acquired NLDO.
REFERENCES 1. Linberg JV, McCormick SA. Primary acquired nasolacrimal duct obstruction. A clinicopathologic report and biopsy technique. Ophthalmology 1986;93:1055–63. 2. Janssen AG, Mansour K, Bos JJ, Castelijins JA. Diameter of the bony lacrimal canal: normal values and values related to nasolacrimal duct obstruction: assessment with CT. AJNR Am J Neuroradiol 2001;22:845–50. 3. Lee H, Ha S, Lee Y, et al. Anatomical and morphometric study of the bony nasolacrimal canal using computed tomography. Ophthalmologica 2012;227:153–9. 4. McCormick A, Sloan B. The diameter of the nasolacrimal canal measured by computed tomography: gender and racial differences. Clin Experiment Ophthalmol 2009;37:357–61. 5. Groessl SA, Sires BS, Lemke BN. An anatomical basis for primary acquired nasolacrimal duct obstruction. Arch Ophthalmol 1997;115:71–4. 6. Santos-Fernandez J. The measurement of the nasal canal according to the race. Am J Ophthalmol 1921;4:32–7. 7. Takahashi Y, Kakizaki H, Nakano T. Bony nasolacrimal duct entrance diameter: gender difference in cadaveric study. Ophthal Plast Reconstr Surg 2011;27:204–5. 8. Takahashi Y, Nakamura Y, Nakano T, et al. The narrowest part of the bony nasolacrimal canal: an anatomical study. Ophthal Plast Reconstr Surg 2013;29:318–22. 9. Shigeta K, Takegoshi H, Kikuchi S. Sex and age differences in the bony nasolacrimal canal. Arch Ophthalmol 2007;125:1677–81. 10. Fasina O, Ogbole GI. CT assessment of the nasolacrimal ca nal in a black African Population. Ophthal Plast Reconstr Surg 2013;29:231–3. 11. Post RH. Tear duct size differences of age, sex and race. Am J Phys Anthropol 1969;30:85–8. 12. Phillips CI, George M. Epiphora and the bony naso-lacrimal canal. Br J Ophthalmol 1956;40:673–80. 13. Groell R, Schaffler GJ, Uggowitzer M, et al. CT-anatomy of the nasolacrimal sac and duct. Surg Radiol Anat 1997;19:189–91. 14. Francisco FC, Carvalho AC, Francisco VF, et al. Evaluation of 1000 lacrimal ducts by dacryocystography. Br J Ophthalmol 2007;91:43–6. 15. Sasaki T, Nagata Y, Sugiyama K. Nasolacrimal duct obstruction classified by dacryoendoscopy and treated with inferior meatal dacryorhinotomy. Part I: Positional diagnosis of primary nasolacrimal duct obstruction with dacryoendoscope. Am J Ophthalmol 2005;140:1065–9. 16. Narioka J, Ohashi Y. Changes in lumen width of nasolacrimal drainage system after adrenergic and cholinergic stimulation. Am J Ophthalmol 2006;141:689–98. 17. Lee-Wing MW, Ashenhurst ME. Clinicopathologic analysis of 166 patients with primary acquired nasolacrimal duct obstruction. Ophthalmology 2001;108:2038–40. 18. Tucker N, Chow D, Stockl F, et al. Clinically suspected primary acquired nasolacrimal duct obstruction: clinicopathologic review of 150 patients. Ophthalmology 1997;104:1882–6. 19. Takahashi Y, Nakamura Y, Nakano T, et al. Horizontal orientation of the bony lacrimal passage: an anatomical study. Ophthal Plast Reconstr Surg 2013;29:128–30. 20. Connell PP, Fulcher TP, Chacko E, et al. Long term follow up of nasolacrimal intubation in adults. Br J Ophthalmol 2006;90:435–6. 21. Ikoma M. Indications of silicone intubation in our institution [in Japanese]. Ganka Rinsho Iho. 1998;92:1387–8.
© 2014 The American Society of Ophthalmic Plastic and Reconstructive Surgery, Inc.
