- Email: [email protected]
Histopathologic Study of Human Lacrimal Gland
Histopathologic Study of Human Lacrimal Gland Statistical Analysis with Special Reference to Aging Hiroto Obata, MD, 1 Seiichiro Yamamoto, MS, 2 Hajime Horiuchi, MD, 1 Rikuo Machinami, MD1 Purpose: Histopathologic changes in human lacrimal gland were investigated, and the relation between histopathologic parameters and patient age and sex, as well as the histopathologic differences between palpebral and orbital lobes of the lacrimal gland were analyzed. Methods: Samples of the main human lacrimal gland that included the palpebral lobes and orbital lobes were taken in 80 autopsies. A statistical analysis was made based on light microscope observations with the following histopathologic changes as parameters: (1) fibrosis (focal, lobular, and diffuse); (2) acinar atrophy (focal, lobular, and diffuse); (3) periductal fibrosis; (4) interlobular ductal dilatation; (5) interlobular ductal proliferation; (6) lymphocytic foci; (7) periductal lymphocytic infiltration; and (8) fatty infiltration. Results: The incidences of these parameters in the palpebral and orbital lobes ranged from 3.8% to 35.0%. Lobular fibrosis, lobular atrophy, diffuse fibrosis, diffuse atrophy, periductal fibrosis, lymphocytic infiltration, and fatty infiltration were more frequent in the orbital lobes with statistical significance, whereas interlobular ductal dilatation was more frequent in the palpebral lobes. There were statistically significant correlations between age and diffuse fibrosis, diffuse atrophy, and periductal fibrosis in the orbital lobes of women, and periductal fibrosis in the palpebral lobes of men. Diffuse fibrosis and diffuse atrophy in the orbital lobes were observed more frequently in elderly women than in elderly men. Conclusions: Various histopathologic changes were observed in the human lacrimal gland. Diffuse fibrosis, diffuse atrophy, and periductal fibrosis predominantly found in elderly women suggested a relation with keratoconjunctivitis sicca in postmenopausal women. The authors speculate the periductal fibrosis is related to the decrease in tear fluid outflow with age and that interlobular ductal dilatation in the palpebral lobes may be caused by stenosis of the excretory duct in the fornix of conjunctiva. Ductal pathologic changes may be important in lacrimal gland dysfunction. Ophthalmology 1995;102:678-686
Originally received: May 17, 1994. Revision accepted: January 3, 1995. 1 Department of Pathology, Faculty of Medicine, University of Tokyo, Tokyo. 2 Department of Epidemiology and Biostatistics, Faculty of Medicine, University of Tokyo, Tokyo. Presented as a poster at the ARVO Annual Meeting, Sarasota, May 1993. The authors received no specific financial support for this study, and have no financial interest in any company or product discussed in it.
678
Ocular dryness is a common condition found in elderly patients. As reported previously, the quality and quantity of tear fluid change during aging. 1- 5 Histologic studies of the human lacrimal gland and of the cause of these changes have been few, and neither the effect of aging nor Reprint requests to Hiroto Obata, MD, Department of Pathology, Faculty of Medicine, University ofTokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113, Japan.
Obata et al
Human Lacrimal Gland
Table 1. Age and Sex Distribution Age (yrs)
0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 Total
No. of Cases
2 2 0 0 12 18 28 16 2 80
Males
1 1 0 0 7 12
20 13
1
55
Females
1 1 0 0
5
6 8 3 1 25
sex on the histopathology of the human lacrimal gland has been evaluated fully. 6- 8 The main human lacrimal gland is comprised of the palpebral and orbital lobes which are continuous with each other at the lateral edge of the aponeurosis of the levator palpebrae superioris. 9 The orbital lobe lies in the lacrimal fossa on the anterior lateral part of the orbit. The palpebral lobe lies below the aponeurosis of the levator palpebrae superioris and is in contact with the superior and lateral fornix of the conjunctiva. Because of its anatomic characteristics, the palpebral lobes may be exposed to foreign antigens. In addition to its basic role of the secretion of tear fluid, the lacrimal gland has an important function in the defense system of the ocular surface and in providing the local and systemic immunologic mechanisms of the ocular surface. In addition, it forms part of the conjunctival-associated lymphoid tissue. 10 Previous histopathologic studies of the human lacrimal gland6- 8 did not state which portions of the palpebral or orbital lobes had been removed, nor the size samples of the lacrimal gland. Therefore, differences in the palpebral and orbital lobes were not investigated, and the extents of histologic changes were not evaluated fully. For our study, we completely removed both lobes at autopsy and investigated in detail the extent of the various histopathologic changes in each lobe of the main lacrimal gland. We also made a statistical analysis of the relation between histopathologic change and the patient's age and sex to elucidate the pathogenesis of ocular dryness.
ease. Samples were taken from 55 men and 25 women. Mean age(± standard deviation) was 59± 16 years (range, 1-87 years) (males: 60 ± 14 years; range, 2-82 years; and females: 56± 18 years; range, 1-87 years). There was no significant difference in age between the male and female patients (Table I). The mean postmortem time was 8.9 ± 5.5 hours. In general, the main lacrimal gland consisting of the palpebral and orbital lobes was removed by means of an intracranial approach made through the superior orbital bones (Figs IA and IB), but some glands were removed by means of a transconjunctival approach. Lacrimal gland tissues removed from the right orbit were cut into two portions, the palpebral and orbital lobes, then immediately fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, and kept overnight. The fixed palpebral or orbital lobe tissue was embedded in paraffin then cut into 4-J,Lm-thick sections at the widest cut surface obtainable, after which the sections were deparaffinized then stained with hematoxylin-eosin. The respective palpebral and orbital lobe sections were examined by two observers
Subjects and Methods Tissues Eighty samples of human lacrimal gland that included the palpebral and orbital lobes were obtained from autopsies done at the University of Tokyo Hospital. Autopsies were excluded from the study if the patients had autoimmune diseases (including Sjogren syndrome), rheumatoid arthritis, endocrine diseases, head irradiation, amyotrophic lateral sclerosis, 11 and graft-versus-host dis-
Figure 1. A, human lacrimal gland in the right orbit. The superior orbital bone has been removed. The orbital lobe of the lacrimal gland (arrowheads) lies in the lacrimal fossa on the lateral part of the orbit. SR = superior rectus muscle; MR = medial rectus muscle; SO = superior oblique muscle. B, entire main lacrimal glands from the lefr and right orbits. The palpebral lobes are above and the orbital lobes below the dotted line. Each orbital lobe is approximately one and a half times larger than the palpebral lobe. R = right; L = left.
679
Ophthalmology
Volume 102, Number 4, April 1995
'··
' .· -.;..;
'·· ·-;.~-'
t
••:
Figure 2. Focal atrophy without fibrosis. A palpebral lobe section from a 43-year-old woman. Acinar atrophy is present in part of one lobule. Atrophic acini look like intralobular ducts. Eosinophilic amorphous material, interpreted as the stasis of tear fluid, is present in some lumens (hematoxylin- eosin; original magnification X20).
(HO and HH) under a light microscope in a masked manner without prior knowledge of age and sex.
Histopathologic Findings and Definition We examined histopathologic changes in each lobe of the lacrimal gland samples and established the following changes as the study parameters: l. Atrophy (Figs 2-4) (focal, lobular, and diffuse);
2. Fibrosis (Fig 4) (focal, lobular, and diffuse); 3. Periductal fibrosis (Fig 5); 4. Interlobular ductal changes (interlobular ductal dilatation [Fig 6] and interlobular ductal proliferation [Fig 7]);
Figure 4. Lobular atrophy with lobular fibrosis. A palpebral lobe section from a 74-year-old man. Both acinar atrophy and periacinar fibrosis are present throughout one lobule. Normal acini are present on the left side of the section (hematoxylin- eosin; original magnification, X33).
5. Lymphocytic changes (periductal lymphocytic infiltration [Fig 8]; lymphocytic foci [Fig 9]); and 6. Fatty infiltration (Fig 10). These phenomena often were associated with each other in the same specimen (e.g., lobular atrophy, lobular fibrosis, and peri ductal fibrosis are all present in Fig 11).
Grading The following criteria were used to grade the fibrosis and atrophy: "focal" was defined as change present only within one lobular architecture, "lobular" as change present in less than 50% of the lobules per section, and "diffuse" as more than 50% change. Periductal fibrosis was classified as grade 0 if not present, grade 1 if present in less than 50% of the intralobular and interlobular ducts, and grade 2 if present in more than 50% of them.
..."" - . ,·;~,~ ~
' ;
.·
I
}
~- }
~
, / ".
--~.\ )$
Figure 3. Lobular atrophy without fibrosis. A palpebral lobe section from a 44-year-old man. Acinar atrophy covers the whole of one lobule, but no periacinar fibrosis is present. Eosinophilic amorphous material is present in the lumens of the dilated intralobular ducts (asterisk). Normal acini are present on the left side of the section (hematoxylin- eosin; original magnification, X25).
680
A.
,'~
'"
Figure 5. Periductal fibrosis. An orbital lobe section from a 62-year-old man. H yalinized thick connective tissue is present around the interlobular duct (arrows). The ductal epithelium is atrophic (hematoxylin-eosin; original magnification, X50).
Obata et al · Human Lacrimal Gland
Figure 6. Interlobular ductal dilatation. A palpebral lobe section from a 64-year-old man. Cystic dilatation of the interlobular ducts with atrophic ductal epithelium (arrows) is present. Stasis of the tear fluid is seen as eosinophilic amorphous material in the cystic ducts (asterisk) (hematox· ylin-eosin; original magnification, Xl6).
Periductal lymphocytic infiltration was defined as lymphocytic infiltration of 50 or more cells found only around the inter- or intralobular ducts (Fig 8). Lymphocytic foci was defined as lymphocytic infiltration regardless of the periductal region (Fig 9). The lymphocytic foci parameter was assessed statistically and included the periductal lymphocytic infiltration findings because most lymphocytic foci were at the edges of the interlobular or intralobular ducts. The criteria used to grade lymphocytic foci and periductallymphocytic infiltration were grade 0 if not present per section, grade 1 if present in one focus per section, and grade 2 if in two or more foci per section. Fatty infiltration was defined as replacement of 30% or more of the parenchyma by adipose tissue. The presence or absence of interlobular ductal dilatation and interlobular ductal proliferation was investigated.
Statistical Analysis All statistical analyses were performed using the SAS version 6 statistical software package (SAS Institute, Inc,
Figure 8. Periductal lymphocytic infiltration. A palpebral lobe section from a 64-year-old woman. Lymphocytic infiltration can be seen around the intralobular duct (hematoxylin-eosin; original magnification XSO).
Cary, NC). McNemar's test was used to compare the histopathologic differences between the palpebral and orbital lobes. Because it is clear that the palpebral and orbital lobes are not independent of each other, we statistically analyzed the parameters in each lobe in the current study. To estimate the relation between the parameters and age and sex, ( 1) Spearman's rank correlation test was used to analyze the relation between the parameters and age in each sex group, and (2) odds ratios (OR) were calculated with 95% confidence interval (95% CI) to examine the relation between the parameters and sex in each age group. On the parameters having grade 1 and 2, the presence or absence was statistically analyzed. An OR of more than 1 means that a parameter is observed more frequently in men than in women. To analyze the relation between the parameters and sex, the samples were divided into two age groups: group 1, 40 to 59 years of age (n = 30); group 2, 60 to 87 years of age (n = 46). Four patients younger than 39 years of age were omitted from the current study
Figure 9. Lymphocytic foci. An orbital lobe section from a 72-year-old man. Lymphocytic infiltration is present around the acinar lobule but not in the periductal region (hematoxylin-eosin; original magnification, X40).
681
Ophthalmology
Volume 102, Number 4, April1995
Figure 10. Fatty infiltration. An orbital lobe section from a 48-year-old woman. Fat cells have infiltrated the acinar lobules (hematoxylin-eosin; original magnification, X16).
because of a possible bias. In addition, for the parameters that have no interaction with age and sex, by using Breslow-Day test for interaction, Cochran-Mantel-Haenszel test was performed to combine the effects over strata.
Results Predominant Histopathologic Parameters The proportions with the histopathologic parameters in the palpebral and orbital lobes are shown in Figure 12. They ranged from 3.8% to 35.0%. The most common parameter was periductal fibrosis (grade 2) in the orbital lobes (28 patients, 35.0%), and the least common was interlobular ductal dilatation in the palpebral lobes (3 patients, 3.8% ). No periductal fibrosis (grade 2) was found in the palpebral lobes. Histopathologic differences between the palpebral and orbital lobes are shown in Figure 12. Lobular fibrosis, lobular atrophy, diffuse fibrosis, diffuse atrophy, periductal fibrosis, lymphocytic foci, and fatty infiltration were more frequent in the orbital lobes with statistical significance (P < 0.05, McNemar's test), whereas interlobular ductal dilatation was more frequent in the palpebral lobes with statistical significance (P < 0.05, McNemar's test).
test). The correlation coefficients of these parameters were 0.66, 0.58, and 0.49, respectively, indicating that these were age-related changes. The relation between the parameters in the palpebral lobes and sex in each age group is shown in Table 2. The parameters whose 95% confidence interval for odds ratio does not include one indicate that there is an elevated risk of sex difference. In the palpebral lobes, no sex difference was found. The relation between the parameters in the orbital lobes and sex by age group is shown in Table 2. In the orbital lobes, the parameters in which the 95% CI does not include one were as follows: lobular fibrosis, diffuse fibrosis, and focal atrophy in age group 2, and fatty infiltration in age group 1. Of these parameters, lobular fibrosis and focal atrophy were observed more frequently in men than in women (indicated by OR > 1), whereas diffuse fibrosis and fatty infiltration were observed more frequently in women than in men (OR < 1). In addition, it was suggested that diffuse atrophy in the orbital lobes in age group 2 also was observed more frequently in women than in men, as shown by the OR and its 95% CI. The parameters that have not shown interaction effect were (I) in the palpebral lobes, focal, lobular, and periductal fibrosis; lymphocytic foci; periductal lymphocytic infiltration; and interlobular ductal proliferation, and (2) in the orbital lobes, focal, lobular, and periductal fibrosis; lymphocytic foci; and periductal lymphocytic infiltration (P > 0.20; because the power of Breslow-Day test is not large, we use alpha = 0.20 as the rejection criteria). As a result ofCochran-MantelHaenszel test for these parameters, there were statistically significant associations between age and interlobular ductal proliferation in the palpebral lobes and periductal fibrosis in the orbital lobes, whereas there was statistically significant association between sex and focal fibrosis in the orbital lobes, which was more frequently observed in men.
Relation between Parameters and Age and Sex The relation between the parameters in the palpebral lobes and age in each sex group is shown in Table 2. Statistically significant correlations existed for age and periductal fibrosis in men (P < 0.05, Spearman's rank correlation test). The positive correlation coefficient (0.29) means agerelated change. Similarly, the relation between the parameters in the orbital lobes and age by sex is shown in Table 2. Statistically significant correlations existed for age and diffuse fibrosis, diffuse atrophy, and periductal fibrosis in women (P < 0.05, Spearman's rank correlation
682
Figure 11. Lobular atrophy, lobular fibrosis, and periductalfibrosis. An orbital lobe section from a 68-year-old man. Acinar atrophy, periacinar fibrosis, and periductal fibrosis are present (hematoxylin-eosin; original magnification, X10).
Obata et al · Human Lacrimal Gland Focal fi brosis (p=0.34) Lobular fibrosis (p=0.004)* Diffuse fibrosis (p=0.003)* Focal atro phy (p=0.51) Lobular atrophy (p=0.007)* Diffuse atrophy (p=0.003)* Periductal fibrosis (p
G1 Figure 12. Prevalence of the h istopathologic parameters in the palpebral and orbital lobes. Numerals on the bars give the number of subjects (percentage of total patients). For the statistical analysis, the McNemar's test was used (*P < 0.05). Gl = grade 1; G2 = grade 2.
G2
Lymphocytic G 1 ~~=~~::=~::!__-, foci (p=0.008)* G2 G1 Periductal lymphocytic infiltratio n G2 (p=0.35) Interlobular ductal dilatation (p=0.008)* Interlobular ductal proliferation (p=0.046)*
~~::=:::=:::=::::~~
1----,
Fatty infiltration (p
0
10
20
30
c=J :Palpebral lobe, IM8:Orbital lobe Discussion Our study showed various histopathologic changes in human lacrimal glands obtained postmortem, some of which were age related, including (I) diffuse fibrosis, diffuse atrophy, and periductal fibrosis in the orbital lobes of female patients, and (2) periductal fibrosis in the palpebral lobes of male patients. Examples of age-related changes are the histologic findings of a 17-year-old boy and an 87year-old woman shown in Figure 13. Histologic studies of the salivary glands of the elderly show that, in general, acinar elements are decreased and replaced with adipose and fibrotic tissue. 12•13 Prager 6 pointed out that a decrease in lacrimal gland weight, fibrosis, and acinar atrophy are changes that occur with
%
senility, but they did not do a statistical study. Damato et al 7 reported that periductal and periacinar fibrosis, acinar atrophy, and dilated and tortuous ducts are age-related changes based on results of histologic examinations of 99 human lacrimal glands. Moreover, Roen et al8 reported that chronic inflammation, periductal fibrosis, and enlarged ducts are aging changes based on results of histologic examinations of 33 lacrimal glands. Neither report, however, compared the incidences of individual histologic changes in the two lobes. We removed the lacrimal gland completely, including both lobes, and examined the histopathologic findings statistically in detail. Our results showed that diffuse fibrosis and diffuse atrophy of women in the orbital lobes are age-related changes. The atrophic acinar cells showed scanty cy-
683
Volume 102, Number 4, April1995
Ophthalmology
Table 2. Relationship between the Parameters and Age and Sex Age
Sex
Female
Male
cc
Histologic Parameters
cc
p
Group 2t (n = 46)
Group 1* (n = 30)
(n = 25)
(n =55)
p
OR
95% CI
OR
95% CI
Palpebral Lobes Focal fibrosis Lobular fibrosis Diffuse fibrosis Focal atrophy Lobular atrophy Diffuse atrophy Periductal fibrosis Lymphocytic foci Periductallymphocytic infiltration Interlobular ductal dilatation Interlobular ductal proliferation Fatty infiltration Focal fibrosis Lobular fibrosis Diffuse fibrosis Focal atrophy Lobular atrophy Diffuse atrophy Periductal fibrosis Lymphocytic foci Periductallymphocytic infiltration Interlobular ductal dilatation Interlobular ductal proliferation Fatty infiltration CC = correlation coefficient; OR • Group 1 = 40-59 yrs of age. t Group 2 = 60-87 yrs of age.
=
0.085 0.063 0.12 0.14 0.057 0.099 0.29 0.086
0.54 0.16 0.37 0.32 0.68 0.47 0.03'1' 0.53
0.31 -0.085 -0.028 0.27 -0.085 -0.028 0.34 -0.026
0.13 0.36 0.89 0.20 0.69 0.89 0.098 0.90
6.7 1.2 0.56 2.7 1.2 0.56 0.53 0.43
0.33-137 0.69-14.7 0.03-9.87 0.26-27.5 0.09-14.7 0.03-9.87 0.06-4.41 0.09-2.05
1.9 5.7 1.9 0.92 4.7 2.8 0.52 0.83
0.20-18.1 0.30-109 0.09-42.9 0.20-4.26 0.24-90.8 0.13-57.8 0.12-2.23 0.20-3.41
-0.004 -0.018 0.25 -0.01
0.98 0.90 0.062 0.94
-0.003 0.24 0.17 -0.14
0.99 0.26 0.41 0.50
0.43 0.84 1.8 0.56
0.09-2.05 0.12-6.03 0.07-49.8 0.03-0.72
0.52 0.52 0.40 1.9
0.12-2.23 0.10-2.60 0.08-2.13 0.09-42.9
0.004 0.19 0.05 0.13 0.22 0.05 0.17 0.14
0.98 0.16 0.67 0.34 0.11 0.71 0.20 0.29
-0.20 -0.19 0.66 -0.20 -0.05 0.58 0.49 0.36
0.34 0.36 0.0003'1' 0.34 0.82 0.0026'1' 0.013'1' 0.073
5.8 0.71 1.9 3.6 0.50 1.9 1.9 1.9
0.61-55.7 0.13-3.99 0.17-20.6 0.36-35.5 0.08-3.06 0.17-20.6 0.39-9.70 0.42-8.92
5.7 8.7 0.12 6.8 3.1 0.26 0.64
0.30-109 1.01-75.0 0.03-0.55 2.77-129 0.58-16.4 0.06-1.06 0.27-4.39 0.17-2.40
0.07 0.09 0.22 0.24
0.63 0.53 0.11 0.08
0.21 -0.20 0.35 -0.35
0.32 0.34 0.08 0.08
3.8 0.72 1.9 0.13
0.79-18.1 0.03-19.2 0.17-20.6 0.03-0.72
1.3 0.65 0.62 1.4
0.34-4.74 0.09-42.9 0.15-2.59 0.32-6.37
odds ratio; CI
=
Orbital Lobes
1.1
confidence interval.
i' P < 0.05: Spearman's rank correlation test. toplasm. There is strong evidence that these atrophic changes are related to the decrease of tear proteins with age4 •5 (i.e., the decrease of quality of tear fluid, but the mechanisms of fibrosis and acinar atrophy are not clear). The most interesting results found were ductal changes. Moreover, periductal fibrosis in the palpebral lobes of men and in the orbital lobes of women were related to advanced age. We consider periductal fibrosis to be an important factor related to the decrease in the outflow of tear fluids with age. Because both ductal epithelial cells and acinar cells are responsible for electrolyte and water secretion, 14•15 periductal fibrosis and the atrophy of ductal epithelial cells may interfere with the normal physiologic function of the ducts.
684
Of the various ductal pathologic changes found, interlobular ductal dilatation in the palpebral lobes is of great importance. It suggests that stenosis or obstruction of the orifices of the excretory duct in the fornix of the conjunctiva causes cystic dilatation of the interlobular ducts in the palpebral lobes, an indication of the obstruction of tear fluid outflow. Moreover, stenotic or obstructive changes in the ducts may be caused by various types of subclinical conjunctivitis because the palpebral lobe is in anatomic contact with the fornix of the conjunctiva. Damato et aC reported that patients with cystic dilatation of the ducts were significantly older than patients without it. In contrast, our results showed that interlobular ductal dilatation was not related to age.
Obata et al · Human Lacrimal Gland
Figure 13. Histologic findings of age-related changes. A, normal histologic finding of a 17-year-old boy. B, histologic finding of an 87-year-old woman shows the acinar atrophy and periductal fibrosis (arrows) at the same magnification (hematoxylin-eosin; original magnification, XSO).
Interlobular ductal proliferation in the palpebral lobes was shown to be an age-related ductal change in our study. It probably is the same as the tortuous abnormal ducts described in Damato's7 report. The meaning of these findings, however, is not clear. In histologic examinations of tissues from 239 autopsies, Waterhouse 16 found focal lymphocytic infiltration in the lacrimal and salivary glands of elderly patients who had no other evidence of glandular disease. Nasu et al 17 reported focal lymphocytic infiltration of the lacrimal glands in 63.5% of 115 patients who underwent autopsy who had no autoimmune diseases. In our study, lymphocytic foci (grade 2) was the predominant parameter of lymphocytic infiltration, the incidence being 31.3%. Waterhouse 16 defined focal adenitis of the submandibular, parotid, and lacrimal glands as occurring when two or more lymphocytic foci are present. In Sjogren syndrome, the earliest histologic finding in the salivary glands is described as periductallymphocytic infiltration.18 Therefore, we consider focal lymphocytic infiltration of the lacrimal gland an important finding because it suggests subclinical dacryoadenitis. The fatty infiltration that replaces the parenchyma in the salivary glands generally is believed to accompany advanced age. 13 Our results do not agree with this belief. Fatty infiltration was more frequent in the orbital lobes
than in the palpebral lobes. Fatty infiltration in the orbital lobes may be related to the lobe's anatomic characteristics. The lacrimal gland has no definite capsule, and the orbital lobe is surrounded by orbital adipose tissue. Moreover, we found that fatty infiltration in middle-aged group, which was more frequently observed in women than in men, was a sex-associated finding. The lacrimal gland is considered to be responsible for various sex-associated differences, such as mucosal immunity and morphology, based on a study that showed that the lacrimal glands of male rats produce significantly more IgA and secretory component than do the glands of female rats. 19 Moreover, a morphometric study has shown that the acinar area in the lacrimal glands of males is larger than that of females in several species, including humans. 20 We found that diffuse fibrosis and diffuse atrophy were observed more frequently in elderly women than in elderly men. This result is of great importance because it suggests a relation with keratoconjunctivitis sicca encountered more frequently in postmenopausal women. 21 In contrast, we found that lobular fibrosis and focal atrophy were observed more frequently in elderly men than in elderly women. This sex difference was considered to be a result from the relation with diffuse fibrosis and diffuse atrophy more frequently observed in elderly women. In the only report of the histologic finding related to sex, no sex difference was found in the frequency of lymphocytic infiltration of the lacrimal gland. 17 Henderson and Prough' reported that sex is not a factor in the rate of the flow of tears, even though a small difference was found. Further investigation therefore is needed to clarify the relation between sex difference and lacrimal gland pathology. Interestingly, several histopathologic differences were shown between the palpebral and orbital lobes. The reasons for these differences, however, are not clear. Although the orbital lobe is approximately one and a half times larger than the palpebral lobe, it is not known whether structural and functional differences exist. We speculate that there are both anatomic and immunologic differences in these lobes. In conclusion, we found that the lacrimal gland undergoes major histologic changes with age and sex. Diffuse fibrosis, diffuse atrophy, and periductal fibrosis found more frequently in elderly women suggested a relation with keratoconjunctivitis sicca in postmenopausal women. Ductal pathologic change, such as periductal fibrosis or interlobular ductal dilatation, may be an important factor in decreasing tear fluid outflow and/or lacrimal gland dysfunction. Another interesting finding was that there are histopathologic differences between the palpebral and orbital lobes of the lacrimal gland. We affirm that the results of this study of the human lacrimal gland provide standards for comparison by which changes in diseased lacrimal gland tissues can be recognized.
References I. Henderson JW, Prough WA. Influence of age and sex on flow of tears. Arch Ophthalmol 1950;43:224-31.
685
Ophthalmology
Volume 102, Number 4, April1995
2. de Roetth A Sr. Lacrimation in normal eyes. Arch Ophthalmoll953;49:185-9. 3. Nom MS. Tear secretion in normal eyes-estimated by a new method: the lacrimal streak dilution test. Acta Ophthalmol 1965;43:567-73. 4. McGill JI, Liakos GM, Goulding N, Seal DV. Normal tear protein profiles and age-related changes. Br J Ophthalmol 1984;68:316-20. 5. Seal DV. The effect of ageing and disease on tear constituents. Trans Ophthalmol Soc UK 1985;104:355-62. 6. Prager A. Makroskopische und Mikroskopische Untersuchungen tiber die Altersatrophie der Tranendriise (vorlaufige Mitteilung). Bib! Ophthalmol 1966;69:146-58. 7. Damato BE, Allan D, Murray SB, Lee WR. Senile atrophy of the human lacrimal gland: the contribution of chronic inflammatory disease. Br J Ophthalmol 1984;68: 674-80. 8. Roen JL, Stasior OG, Jakobiec FA. Aging changes in the human lacrimal gland: role of the ducts. CLAO J 1985;11: 237-42. 9. Iwamoto T, Jakobiec FA. Lacrimal glands. In: Jakobiec FA, ed. Ocular Anatomy, Embryology, and Teratology. Philadelphia: Harper & Row, 1982; Chap. 25. 10. Chandler JW, Gillette TE. Immunologic defense mechanisms of the ocular surface. Ophthalmology 1983;90:585-91. 11. Copperman R. Decreased lacrimation in amyotrophic lateral sclerosis [letter]. JAMA 1974;230:536. 12. Scott J. Structure and function in aging human salivary glands. Gerodontology 1986;5: 149-58.
686
13. Waterhouse JP, Chisholm DM, Winter RB, et al. Replacement of functional parenchymal cells by fat and connective tissue in human submandibular salivary glands: an age-related change. J Oral Pathol 1973;2:16-27. 14. Alexander JH, Lennep EWvan, YoungJA. Water and electrolyte secretion by the exorbital lacrimal gland of the rat studied by micropuncture and catheterization techniques. Pflugers Archiv 1972;337:299-309. 15. Dartt DA. Physiology of tear production. In: Lemp MA, Marquardt R, eds. The Dry Eye: A Comprehensive Guide. Berlin: Springer-Verlag, 1992; Chap. 3. 16. Waterhouse JP. Focal adenitis in salivary and lacrimal glands. Proc R Soc Med 1963;56:911-18. 17. Nasu M, Matsubara 0, Yamamoto H. Post-mortem prevalence of lymphocytic infiltration of the lacrymal gland: a comparative study in autoimmune and non-autoimmune diseases. J Pathol 1984;143:11-15. 18. Cotran RS, Kumar V, Robbins SL, eds. Robbins' Pathologic Basis of Disease, 4th ed. Philadelphia: WB Saunders, 1989;202-4. 19. Sullivan DA. Hormonal influence on the secretory immune system of the eye. In: Freier S, ed. The NeuroendocrineImmune Network. Boca Raton: CRC Press, 1990; chap. 15. 20. Cornell-Bell AH, Sullivan DA, Allansmith MR. Genderrelated differences in the morphology of the lacrimal gland. Invest Ophthalmol Vis Sci 1985;26: 1170-5. 21. Holly FJ, Lemp MA. Tear physiology and dry eyes. Surv Ophthalmol 1977;22:69-87.
Originally received: May 17, 1994. Revision accepted: January 3, 1995. 1 Department of Pathology, Faculty of Medicine, University of Tokyo, Tokyo. 2 Department of Epidemiology and Biostatistics, Faculty of Medicine, University of Tokyo, Tokyo. Presented as a poster at the ARVO Annual Meeting, Sarasota, May 1993. The authors received no specific financial support for this study, and have no financial interest in any company or product discussed in it.
678
Ocular dryness is a common condition found in elderly patients. As reported previously, the quality and quantity of tear fluid change during aging. 1- 5 Histologic studies of the human lacrimal gland and of the cause of these changes have been few, and neither the effect of aging nor Reprint requests to Hiroto Obata, MD, Department of Pathology, Faculty of Medicine, University ofTokyo, Hongo 7-3-1, Bunkyo-ku, Tokyo 113, Japan.
Obata et al
Human Lacrimal Gland
Table 1. Age and Sex Distribution Age (yrs)
0-9 10-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 Total
No. of Cases
2 2 0 0 12 18 28 16 2 80
Males
1 1 0 0 7 12
20 13
1
55
Females
1 1 0 0
5
6 8 3 1 25
sex on the histopathology of the human lacrimal gland has been evaluated fully. 6- 8 The main human lacrimal gland is comprised of the palpebral and orbital lobes which are continuous with each other at the lateral edge of the aponeurosis of the levator palpebrae superioris. 9 The orbital lobe lies in the lacrimal fossa on the anterior lateral part of the orbit. The palpebral lobe lies below the aponeurosis of the levator palpebrae superioris and is in contact with the superior and lateral fornix of the conjunctiva. Because of its anatomic characteristics, the palpebral lobes may be exposed to foreign antigens. In addition to its basic role of the secretion of tear fluid, the lacrimal gland has an important function in the defense system of the ocular surface and in providing the local and systemic immunologic mechanisms of the ocular surface. In addition, it forms part of the conjunctival-associated lymphoid tissue. 10 Previous histopathologic studies of the human lacrimal gland6- 8 did not state which portions of the palpebral or orbital lobes had been removed, nor the size samples of the lacrimal gland. Therefore, differences in the palpebral and orbital lobes were not investigated, and the extents of histologic changes were not evaluated fully. For our study, we completely removed both lobes at autopsy and investigated in detail the extent of the various histopathologic changes in each lobe of the main lacrimal gland. We also made a statistical analysis of the relation between histopathologic change and the patient's age and sex to elucidate the pathogenesis of ocular dryness.
ease. Samples were taken from 55 men and 25 women. Mean age(± standard deviation) was 59± 16 years (range, 1-87 years) (males: 60 ± 14 years; range, 2-82 years; and females: 56± 18 years; range, 1-87 years). There was no significant difference in age between the male and female patients (Table I). The mean postmortem time was 8.9 ± 5.5 hours. In general, the main lacrimal gland consisting of the palpebral and orbital lobes was removed by means of an intracranial approach made through the superior orbital bones (Figs IA and IB), but some glands were removed by means of a transconjunctival approach. Lacrimal gland tissues removed from the right orbit were cut into two portions, the palpebral and orbital lobes, then immediately fixed in 4% paraformaldehyde in 0.1 M phosphate buffer, and kept overnight. The fixed palpebral or orbital lobe tissue was embedded in paraffin then cut into 4-J,Lm-thick sections at the widest cut surface obtainable, after which the sections were deparaffinized then stained with hematoxylin-eosin. The respective palpebral and orbital lobe sections were examined by two observers
Subjects and Methods Tissues Eighty samples of human lacrimal gland that included the palpebral and orbital lobes were obtained from autopsies done at the University of Tokyo Hospital. Autopsies were excluded from the study if the patients had autoimmune diseases (including Sjogren syndrome), rheumatoid arthritis, endocrine diseases, head irradiation, amyotrophic lateral sclerosis, 11 and graft-versus-host dis-
Figure 1. A, human lacrimal gland in the right orbit. The superior orbital bone has been removed. The orbital lobe of the lacrimal gland (arrowheads) lies in the lacrimal fossa on the lateral part of the orbit. SR = superior rectus muscle; MR = medial rectus muscle; SO = superior oblique muscle. B, entire main lacrimal glands from the lefr and right orbits. The palpebral lobes are above and the orbital lobes below the dotted line. Each orbital lobe is approximately one and a half times larger than the palpebral lobe. R = right; L = left.
679
Ophthalmology
Volume 102, Number 4, April 1995
'··
' .· -.;..;
'·· ·-;.~-'
t
••:
Figure 2. Focal atrophy without fibrosis. A palpebral lobe section from a 43-year-old woman. Acinar atrophy is present in part of one lobule. Atrophic acini look like intralobular ducts. Eosinophilic amorphous material, interpreted as the stasis of tear fluid, is present in some lumens (hematoxylin- eosin; original magnification X20).
(HO and HH) under a light microscope in a masked manner without prior knowledge of age and sex.
Histopathologic Findings and Definition We examined histopathologic changes in each lobe of the lacrimal gland samples and established the following changes as the study parameters: l. Atrophy (Figs 2-4) (focal, lobular, and diffuse);
2. Fibrosis (Fig 4) (focal, lobular, and diffuse); 3. Periductal fibrosis (Fig 5); 4. Interlobular ductal changes (interlobular ductal dilatation [Fig 6] and interlobular ductal proliferation [Fig 7]);
Figure 4. Lobular atrophy with lobular fibrosis. A palpebral lobe section from a 74-year-old man. Both acinar atrophy and periacinar fibrosis are present throughout one lobule. Normal acini are present on the left side of the section (hematoxylin- eosin; original magnification, X33).
5. Lymphocytic changes (periductal lymphocytic infiltration [Fig 8]; lymphocytic foci [Fig 9]); and 6. Fatty infiltration (Fig 10). These phenomena often were associated with each other in the same specimen (e.g., lobular atrophy, lobular fibrosis, and peri ductal fibrosis are all present in Fig 11).
Grading The following criteria were used to grade the fibrosis and atrophy: "focal" was defined as change present only within one lobular architecture, "lobular" as change present in less than 50% of the lobules per section, and "diffuse" as more than 50% change. Periductal fibrosis was classified as grade 0 if not present, grade 1 if present in less than 50% of the intralobular and interlobular ducts, and grade 2 if present in more than 50% of them.
..."" - . ,·;~,~ ~
' ;
.·
I
}
~- }
~
, / ".
--~.\ )$
Figure 3. Lobular atrophy without fibrosis. A palpebral lobe section from a 44-year-old man. Acinar atrophy covers the whole of one lobule, but no periacinar fibrosis is present. Eosinophilic amorphous material is present in the lumens of the dilated intralobular ducts (asterisk). Normal acini are present on the left side of the section (hematoxylin- eosin; original magnification, X25).
680
A.
,'~
'"
Figure 5. Periductal fibrosis. An orbital lobe section from a 62-year-old man. H yalinized thick connective tissue is present around the interlobular duct (arrows). The ductal epithelium is atrophic (hematoxylin-eosin; original magnification, X50).
Obata et al · Human Lacrimal Gland
Figure 6. Interlobular ductal dilatation. A palpebral lobe section from a 64-year-old man. Cystic dilatation of the interlobular ducts with atrophic ductal epithelium (arrows) is present. Stasis of the tear fluid is seen as eosinophilic amorphous material in the cystic ducts (asterisk) (hematox· ylin-eosin; original magnification, Xl6).
Periductal lymphocytic infiltration was defined as lymphocytic infiltration of 50 or more cells found only around the inter- or intralobular ducts (Fig 8). Lymphocytic foci was defined as lymphocytic infiltration regardless of the periductal region (Fig 9). The lymphocytic foci parameter was assessed statistically and included the periductal lymphocytic infiltration findings because most lymphocytic foci were at the edges of the interlobular or intralobular ducts. The criteria used to grade lymphocytic foci and periductallymphocytic infiltration were grade 0 if not present per section, grade 1 if present in one focus per section, and grade 2 if in two or more foci per section. Fatty infiltration was defined as replacement of 30% or more of the parenchyma by adipose tissue. The presence or absence of interlobular ductal dilatation and interlobular ductal proliferation was investigated.
Statistical Analysis All statistical analyses were performed using the SAS version 6 statistical software package (SAS Institute, Inc,
Figure 8. Periductal lymphocytic infiltration. A palpebral lobe section from a 64-year-old woman. Lymphocytic infiltration can be seen around the intralobular duct (hematoxylin-eosin; original magnification XSO).
Cary, NC). McNemar's test was used to compare the histopathologic differences between the palpebral and orbital lobes. Because it is clear that the palpebral and orbital lobes are not independent of each other, we statistically analyzed the parameters in each lobe in the current study. To estimate the relation between the parameters and age and sex, ( 1) Spearman's rank correlation test was used to analyze the relation between the parameters and age in each sex group, and (2) odds ratios (OR) were calculated with 95% confidence interval (95% CI) to examine the relation between the parameters and sex in each age group. On the parameters having grade 1 and 2, the presence or absence was statistically analyzed. An OR of more than 1 means that a parameter is observed more frequently in men than in women. To analyze the relation between the parameters and sex, the samples were divided into two age groups: group 1, 40 to 59 years of age (n = 30); group 2, 60 to 87 years of age (n = 46). Four patients younger than 39 years of age were omitted from the current study
Figure 9. Lymphocytic foci. An orbital lobe section from a 72-year-old man. Lymphocytic infiltration is present around the acinar lobule but not in the periductal region (hematoxylin-eosin; original magnification, X40).
681
Ophthalmology
Volume 102, Number 4, April1995
Figure 10. Fatty infiltration. An orbital lobe section from a 48-year-old woman. Fat cells have infiltrated the acinar lobules (hematoxylin-eosin; original magnification, X16).
because of a possible bias. In addition, for the parameters that have no interaction with age and sex, by using Breslow-Day test for interaction, Cochran-Mantel-Haenszel test was performed to combine the effects over strata.
Results Predominant Histopathologic Parameters The proportions with the histopathologic parameters in the palpebral and orbital lobes are shown in Figure 12. They ranged from 3.8% to 35.0%. The most common parameter was periductal fibrosis (grade 2) in the orbital lobes (28 patients, 35.0%), and the least common was interlobular ductal dilatation in the palpebral lobes (3 patients, 3.8% ). No periductal fibrosis (grade 2) was found in the palpebral lobes. Histopathologic differences between the palpebral and orbital lobes are shown in Figure 12. Lobular fibrosis, lobular atrophy, diffuse fibrosis, diffuse atrophy, periductal fibrosis, lymphocytic foci, and fatty infiltration were more frequent in the orbital lobes with statistical significance (P < 0.05, McNemar's test), whereas interlobular ductal dilatation was more frequent in the palpebral lobes with statistical significance (P < 0.05, McNemar's test).
test). The correlation coefficients of these parameters were 0.66, 0.58, and 0.49, respectively, indicating that these were age-related changes. The relation between the parameters in the palpebral lobes and sex in each age group is shown in Table 2. The parameters whose 95% confidence interval for odds ratio does not include one indicate that there is an elevated risk of sex difference. In the palpebral lobes, no sex difference was found. The relation between the parameters in the orbital lobes and sex by age group is shown in Table 2. In the orbital lobes, the parameters in which the 95% CI does not include one were as follows: lobular fibrosis, diffuse fibrosis, and focal atrophy in age group 2, and fatty infiltration in age group 1. Of these parameters, lobular fibrosis and focal atrophy were observed more frequently in men than in women (indicated by OR > 1), whereas diffuse fibrosis and fatty infiltration were observed more frequently in women than in men (OR < 1). In addition, it was suggested that diffuse atrophy in the orbital lobes in age group 2 also was observed more frequently in women than in men, as shown by the OR and its 95% CI. The parameters that have not shown interaction effect were (I) in the palpebral lobes, focal, lobular, and periductal fibrosis; lymphocytic foci; periductal lymphocytic infiltration; and interlobular ductal proliferation, and (2) in the orbital lobes, focal, lobular, and periductal fibrosis; lymphocytic foci; and periductal lymphocytic infiltration (P > 0.20; because the power of Breslow-Day test is not large, we use alpha = 0.20 as the rejection criteria). As a result ofCochran-MantelHaenszel test for these parameters, there were statistically significant associations between age and interlobular ductal proliferation in the palpebral lobes and periductal fibrosis in the orbital lobes, whereas there was statistically significant association between sex and focal fibrosis in the orbital lobes, which was more frequently observed in men.
Relation between Parameters and Age and Sex The relation between the parameters in the palpebral lobes and age in each sex group is shown in Table 2. Statistically significant correlations existed for age and periductal fibrosis in men (P < 0.05, Spearman's rank correlation test). The positive correlation coefficient (0.29) means agerelated change. Similarly, the relation between the parameters in the orbital lobes and age by sex is shown in Table 2. Statistically significant correlations existed for age and diffuse fibrosis, diffuse atrophy, and periductal fibrosis in women (P < 0.05, Spearman's rank correlation
682
Figure 11. Lobular atrophy, lobular fibrosis, and periductalfibrosis. An orbital lobe section from a 68-year-old man. Acinar atrophy, periacinar fibrosis, and periductal fibrosis are present (hematoxylin-eosin; original magnification, X10).
Obata et al · Human Lacrimal Gland Focal fi brosis (p=0.34) Lobular fibrosis (p=0.004)* Diffuse fibrosis (p=0.003)* Focal atro phy (p=0.51) Lobular atrophy (p=0.007)* Diffuse atrophy (p=0.003)* Periductal fibrosis (p
G1 Figure 12. Prevalence of the h istopathologic parameters in the palpebral and orbital lobes. Numerals on the bars give the number of subjects (percentage of total patients). For the statistical analysis, the McNemar's test was used (*P < 0.05). Gl = grade 1; G2 = grade 2.
G2
Lymphocytic G 1 ~~=~~::=~::!__-, foci (p=0.008)* G2 G1 Periductal lymphocytic infiltratio n G2 (p=0.35) Interlobular ductal dilatation (p=0.008)* Interlobular ductal proliferation (p=0.046)*
~~::=:::=:::=::::~~
1----,
Fatty infiltration (p
0
10
20
30
c=J :Palpebral lobe, IM8:Orbital lobe Discussion Our study showed various histopathologic changes in human lacrimal glands obtained postmortem, some of which were age related, including (I) diffuse fibrosis, diffuse atrophy, and periductal fibrosis in the orbital lobes of female patients, and (2) periductal fibrosis in the palpebral lobes of male patients. Examples of age-related changes are the histologic findings of a 17-year-old boy and an 87year-old woman shown in Figure 13. Histologic studies of the salivary glands of the elderly show that, in general, acinar elements are decreased and replaced with adipose and fibrotic tissue. 12•13 Prager 6 pointed out that a decrease in lacrimal gland weight, fibrosis, and acinar atrophy are changes that occur with
%
senility, but they did not do a statistical study. Damato et al 7 reported that periductal and periacinar fibrosis, acinar atrophy, and dilated and tortuous ducts are age-related changes based on results of histologic examinations of 99 human lacrimal glands. Moreover, Roen et al8 reported that chronic inflammation, periductal fibrosis, and enlarged ducts are aging changes based on results of histologic examinations of 33 lacrimal glands. Neither report, however, compared the incidences of individual histologic changes in the two lobes. We removed the lacrimal gland completely, including both lobes, and examined the histopathologic findings statistically in detail. Our results showed that diffuse fibrosis and diffuse atrophy of women in the orbital lobes are age-related changes. The atrophic acinar cells showed scanty cy-
683
Volume 102, Number 4, April1995
Ophthalmology
Table 2. Relationship between the Parameters and Age and Sex Age
Sex
Female
Male
cc
Histologic Parameters
cc
p
Group 2t (n = 46)
Group 1* (n = 30)
(n = 25)
(n =55)
p
OR
95% CI
OR
95% CI
Palpebral Lobes Focal fibrosis Lobular fibrosis Diffuse fibrosis Focal atrophy Lobular atrophy Diffuse atrophy Periductal fibrosis Lymphocytic foci Periductallymphocytic infiltration Interlobular ductal dilatation Interlobular ductal proliferation Fatty infiltration Focal fibrosis Lobular fibrosis Diffuse fibrosis Focal atrophy Lobular atrophy Diffuse atrophy Periductal fibrosis Lymphocytic foci Periductallymphocytic infiltration Interlobular ductal dilatation Interlobular ductal proliferation Fatty infiltration CC = correlation coefficient; OR • Group 1 = 40-59 yrs of age. t Group 2 = 60-87 yrs of age.
=
0.085 0.063 0.12 0.14 0.057 0.099 0.29 0.086
0.54 0.16 0.37 0.32 0.68 0.47 0.03'1' 0.53
0.31 -0.085 -0.028 0.27 -0.085 -0.028 0.34 -0.026
0.13 0.36 0.89 0.20 0.69 0.89 0.098 0.90
6.7 1.2 0.56 2.7 1.2 0.56 0.53 0.43
0.33-137 0.69-14.7 0.03-9.87 0.26-27.5 0.09-14.7 0.03-9.87 0.06-4.41 0.09-2.05
1.9 5.7 1.9 0.92 4.7 2.8 0.52 0.83
0.20-18.1 0.30-109 0.09-42.9 0.20-4.26 0.24-90.8 0.13-57.8 0.12-2.23 0.20-3.41
-0.004 -0.018 0.25 -0.01
0.98 0.90 0.062 0.94
-0.003 0.24 0.17 -0.14
0.99 0.26 0.41 0.50
0.43 0.84 1.8 0.56
0.09-2.05 0.12-6.03 0.07-49.8 0.03-0.72
0.52 0.52 0.40 1.9
0.12-2.23 0.10-2.60 0.08-2.13 0.09-42.9
0.004 0.19 0.05 0.13 0.22 0.05 0.17 0.14
0.98 0.16 0.67 0.34 0.11 0.71 0.20 0.29
-0.20 -0.19 0.66 -0.20 -0.05 0.58 0.49 0.36
0.34 0.36 0.0003'1' 0.34 0.82 0.0026'1' 0.013'1' 0.073
5.8 0.71 1.9 3.6 0.50 1.9 1.9 1.9
0.61-55.7 0.13-3.99 0.17-20.6 0.36-35.5 0.08-3.06 0.17-20.6 0.39-9.70 0.42-8.92
5.7 8.7 0.12 6.8 3.1 0.26 0.64
0.30-109 1.01-75.0 0.03-0.55 2.77-129 0.58-16.4 0.06-1.06 0.27-4.39 0.17-2.40
0.07 0.09 0.22 0.24
0.63 0.53 0.11 0.08
0.21 -0.20 0.35 -0.35
0.32 0.34 0.08 0.08
3.8 0.72 1.9 0.13
0.79-18.1 0.03-19.2 0.17-20.6 0.03-0.72
1.3 0.65 0.62 1.4
0.34-4.74 0.09-42.9 0.15-2.59 0.32-6.37
odds ratio; CI
=
Orbital Lobes
1.1
confidence interval.
i' P < 0.05: Spearman's rank correlation test. toplasm. There is strong evidence that these atrophic changes are related to the decrease of tear proteins with age4 •5 (i.e., the decrease of quality of tear fluid, but the mechanisms of fibrosis and acinar atrophy are not clear). The most interesting results found were ductal changes. Moreover, periductal fibrosis in the palpebral lobes of men and in the orbital lobes of women were related to advanced age. We consider periductal fibrosis to be an important factor related to the decrease in the outflow of tear fluids with age. Because both ductal epithelial cells and acinar cells are responsible for electrolyte and water secretion, 14•15 periductal fibrosis and the atrophy of ductal epithelial cells may interfere with the normal physiologic function of the ducts.
684
Of the various ductal pathologic changes found, interlobular ductal dilatation in the palpebral lobes is of great importance. It suggests that stenosis or obstruction of the orifices of the excretory duct in the fornix of the conjunctiva causes cystic dilatation of the interlobular ducts in the palpebral lobes, an indication of the obstruction of tear fluid outflow. Moreover, stenotic or obstructive changes in the ducts may be caused by various types of subclinical conjunctivitis because the palpebral lobe is in anatomic contact with the fornix of the conjunctiva. Damato et aC reported that patients with cystic dilatation of the ducts were significantly older than patients without it. In contrast, our results showed that interlobular ductal dilatation was not related to age.
Obata et al · Human Lacrimal Gland
Figure 13. Histologic findings of age-related changes. A, normal histologic finding of a 17-year-old boy. B, histologic finding of an 87-year-old woman shows the acinar atrophy and periductal fibrosis (arrows) at the same magnification (hematoxylin-eosin; original magnification, XSO).
Interlobular ductal proliferation in the palpebral lobes was shown to be an age-related ductal change in our study. It probably is the same as the tortuous abnormal ducts described in Damato's7 report. The meaning of these findings, however, is not clear. In histologic examinations of tissues from 239 autopsies, Waterhouse 16 found focal lymphocytic infiltration in the lacrimal and salivary glands of elderly patients who had no other evidence of glandular disease. Nasu et al 17 reported focal lymphocytic infiltration of the lacrimal glands in 63.5% of 115 patients who underwent autopsy who had no autoimmune diseases. In our study, lymphocytic foci (grade 2) was the predominant parameter of lymphocytic infiltration, the incidence being 31.3%. Waterhouse 16 defined focal adenitis of the submandibular, parotid, and lacrimal glands as occurring when two or more lymphocytic foci are present. In Sjogren syndrome, the earliest histologic finding in the salivary glands is described as periductallymphocytic infiltration.18 Therefore, we consider focal lymphocytic infiltration of the lacrimal gland an important finding because it suggests subclinical dacryoadenitis. The fatty infiltration that replaces the parenchyma in the salivary glands generally is believed to accompany advanced age. 13 Our results do not agree with this belief. Fatty infiltration was more frequent in the orbital lobes
than in the palpebral lobes. Fatty infiltration in the orbital lobes may be related to the lobe's anatomic characteristics. The lacrimal gland has no definite capsule, and the orbital lobe is surrounded by orbital adipose tissue. Moreover, we found that fatty infiltration in middle-aged group, which was more frequently observed in women than in men, was a sex-associated finding. The lacrimal gland is considered to be responsible for various sex-associated differences, such as mucosal immunity and morphology, based on a study that showed that the lacrimal glands of male rats produce significantly more IgA and secretory component than do the glands of female rats. 19 Moreover, a morphometric study has shown that the acinar area in the lacrimal glands of males is larger than that of females in several species, including humans. 20 We found that diffuse fibrosis and diffuse atrophy were observed more frequently in elderly women than in elderly men. This result is of great importance because it suggests a relation with keratoconjunctivitis sicca encountered more frequently in postmenopausal women. 21 In contrast, we found that lobular fibrosis and focal atrophy were observed more frequently in elderly men than in elderly women. This sex difference was considered to be a result from the relation with diffuse fibrosis and diffuse atrophy more frequently observed in elderly women. In the only report of the histologic finding related to sex, no sex difference was found in the frequency of lymphocytic infiltration of the lacrimal gland. 17 Henderson and Prough' reported that sex is not a factor in the rate of the flow of tears, even though a small difference was found. Further investigation therefore is needed to clarify the relation between sex difference and lacrimal gland pathology. Interestingly, several histopathologic differences were shown between the palpebral and orbital lobes. The reasons for these differences, however, are not clear. Although the orbital lobe is approximately one and a half times larger than the palpebral lobe, it is not known whether structural and functional differences exist. We speculate that there are both anatomic and immunologic differences in these lobes. In conclusion, we found that the lacrimal gland undergoes major histologic changes with age and sex. Diffuse fibrosis, diffuse atrophy, and periductal fibrosis found more frequently in elderly women suggested a relation with keratoconjunctivitis sicca in postmenopausal women. Ductal pathologic change, such as periductal fibrosis or interlobular ductal dilatation, may be an important factor in decreasing tear fluid outflow and/or lacrimal gland dysfunction. Another interesting finding was that there are histopathologic differences between the palpebral and orbital lobes of the lacrimal gland. We affirm that the results of this study of the human lacrimal gland provide standards for comparison by which changes in diseased lacrimal gland tissues can be recognized.
References I. Henderson JW, Prough WA. Influence of age and sex on flow of tears. Arch Ophthalmol 1950;43:224-31.
685
Ophthalmology
Volume 102, Number 4, April1995
2. de Roetth A Sr. Lacrimation in normal eyes. Arch Ophthalmoll953;49:185-9. 3. Nom MS. Tear secretion in normal eyes-estimated by a new method: the lacrimal streak dilution test. Acta Ophthalmol 1965;43:567-73. 4. McGill JI, Liakos GM, Goulding N, Seal DV. Normal tear protein profiles and age-related changes. Br J Ophthalmol 1984;68:316-20. 5. Seal DV. The effect of ageing and disease on tear constituents. Trans Ophthalmol Soc UK 1985;104:355-62. 6. Prager A. Makroskopische und Mikroskopische Untersuchungen tiber die Altersatrophie der Tranendriise (vorlaufige Mitteilung). Bib! Ophthalmol 1966;69:146-58. 7. Damato BE, Allan D, Murray SB, Lee WR. Senile atrophy of the human lacrimal gland: the contribution of chronic inflammatory disease. Br J Ophthalmol 1984;68: 674-80. 8. Roen JL, Stasior OG, Jakobiec FA. Aging changes in the human lacrimal gland: role of the ducts. CLAO J 1985;11: 237-42. 9. Iwamoto T, Jakobiec FA. Lacrimal glands. In: Jakobiec FA, ed. Ocular Anatomy, Embryology, and Teratology. Philadelphia: Harper & Row, 1982; Chap. 25. 10. Chandler JW, Gillette TE. Immunologic defense mechanisms of the ocular surface. Ophthalmology 1983;90:585-91. 11. Copperman R. Decreased lacrimation in amyotrophic lateral sclerosis [letter]. JAMA 1974;230:536. 12. Scott J. Structure and function in aging human salivary glands. Gerodontology 1986;5: 149-58.
686
13. Waterhouse JP, Chisholm DM, Winter RB, et al. Replacement of functional parenchymal cells by fat and connective tissue in human submandibular salivary glands: an age-related change. J Oral Pathol 1973;2:16-27. 14. Alexander JH, Lennep EWvan, YoungJA. Water and electrolyte secretion by the exorbital lacrimal gland of the rat studied by micropuncture and catheterization techniques. Pflugers Archiv 1972;337:299-309. 15. Dartt DA. Physiology of tear production. In: Lemp MA, Marquardt R, eds. The Dry Eye: A Comprehensive Guide. Berlin: Springer-Verlag, 1992; Chap. 3. 16. Waterhouse JP. Focal adenitis in salivary and lacrimal glands. Proc R Soc Med 1963;56:911-18. 17. Nasu M, Matsubara 0, Yamamoto H. Post-mortem prevalence of lymphocytic infiltration of the lacrymal gland: a comparative study in autoimmune and non-autoimmune diseases. J Pathol 1984;143:11-15. 18. Cotran RS, Kumar V, Robbins SL, eds. Robbins' Pathologic Basis of Disease, 4th ed. Philadelphia: WB Saunders, 1989;202-4. 19. Sullivan DA. Hormonal influence on the secretory immune system of the eye. In: Freier S, ed. The NeuroendocrineImmune Network. Boca Raton: CRC Press, 1990; chap. 15. 20. Cornell-Bell AH, Sullivan DA, Allansmith MR. Genderrelated differences in the morphology of the lacrimal gland. Invest Ophthalmol Vis Sci 1985;26: 1170-5. 21. Holly FJ, Lemp MA. Tear physiology and dry eyes. Surv Ophthalmol 1977;22:69-87.