Immunopathology and Histopathology of Conjunctival Biopsies in Patients with Presumed Idiopathic Punctal Stenosis

Abstract

Background/Aims

Numerous processes have been implicated as causes of punctal stenosis. Recent studies have highlighted inflammation in punctoplasty specimens in patients with punctal stenosis. Conjunctival biopsy has been suggested as a means to determine underlying etiologies, although little is known regarding what conjunctival pathologies are associated with punctal stenosis. Our objective is to examine the pathologic and immunologic findings in conjunctival biopsy specimens in patients with presumed idiopathic punctal stenosis.

Methods

A retrospective chart review was performed at a single institution over a five year period of patients with presumed idiopathic punctal stenosis who underwent conjunctival biopsy for routine histopathologic and/or direct immunofluorescent (DIF) examination. Patients with known etiologies of punctal stenosis were excluded.

Results

23 patients met inclusion criteria. Conjunctival biopsies (n = 36) from all 23 patients underwent histological examination. 35 of 36 specimens (97.2%) showed a lymphocytic infiltrate. Two patients (8.7%) had findings suggestive of sarcoidosis. Conjunctival biopsies from 18 of the 23 patients were also evaluated by DIF. Nine patients (50.0%) had fibrinogen characteristics suggestive of lichen planus. There were no complications related to the conjunctival biopsy procedures.

Conclusion

Conjunctiva appears to be an excellent proxy in the evaluation of patients with presumed idiopathic punctal stenosis. The biopsy is safe and provides ample tissue. 11 of the 23 patients in this study demonstrated findings consistent with underlying immunologic disorders, suggesting that conjunctival biopsy may play a role in identifying the underlying etiology of punctal stenosis. Knowledge of these underlying conditions impacted medical and surgical treatments for several patients.

Introduction

Despite its rarity in the general population, punctal stenosis may have a significantly higher incidence in specific patient populations. Many conditions and medications have been associated with acquired punctal stenosis. These include infectious causes, such as Actinomyces; autoimmune and inflammatory conditions, such as chronic blepharitis, ectropion, and graft-versus-host disease; and both topical and systemic medications, such as taxanes and topical glaucoma medications.¹

The proposed pathophysiologic mechanism of acquired punctal stenosis involves a chronic inflammatory process that leads to fibrosis and ultimately results in punctal occlusion. Until recently, there was scant histopathologic data in the literature to corroborate this theory. Supporting evidence instead partly came from biopsies of the lower lacrimal drainage system in cases where there were obstructions in this area. The local and systemic associations of lower lacrimal drainage system obstruction, particularly when it occurs bilaterally, has significant overlap with those conditions mentioned previously that are associated with punctal stenosis. It was a logical next step to hypothesize that punctal specimens from cases of stenosis would have similar histopathologic findings as those seen in the lower lacrimal drainage system. More recent literature has supported this theory, finding similar signs of chronic inflammation in the punctal tissue in cases of punctal stenosis.^2,3

These recent papers allow more confident correlations to be made between the etiologic processes and pharmacologic agents associated with punctal stenosis, and the pathophysiologic processes that lead to this stenosis. However, no current studies have examined the utility of histopathologic and direct immunofluoresence (DIF) studies of conjunctival biopsies in patients with punctal stenosis. The conjunctiva is accessible with less clinical consequences of biopsy compared to the canaliculus and punctum, and provides a greater amount of tissue for examination. Further insight into the potential role of conjunctival biopsy in the evaluation of punctal stenosis could thus be quite valuable. Additionally, while there are many proposed etiologies of punctal stenosis, some patients develop this condition despite a lack of any relevant disease or medication history. To date, no study has focused exclusively on patients with presumed idiopathic punctal stenosis.

This study proposes to expand upon the findings mentioned above by examining the histopathology and DIF of conjunctival biopsy specimens in patients with presumed idiopathic punctal stenosis.

Materials and Methods

A retrospective chart review was performed of patients who presented to the University of Iowa Hospital and Clinics (UIHC) between January 2010 and January 2015 with presumed idiopathic epiphora, were clinically diagnosed with punctal stenosis, and subsequently underwent conjunctival biopsy for histopathologic and/or DIF examination. The conjunctival biopsies were blind, rather than directed at abnormal-appearing tissue. The biopsies were taken from the palpebral conjunctiva, between the inferior border of the tarsus and the fornix. The following data were collected: age, sex, laterality of punctal stenosis, biopsy date, DIF report, histopathology report, past medical history, and treatments following biopsy. Histopathological examination was performed on 36 conjunctiva specimens using hematoxylin-eosin. Direct immunofluorescence was performed on 27 conjunctiva specimens using antibodies for IgG, IgA, IgM, C3, and fibrinogen. Appropriate controls were utilized for all antibodies. Patients with known systemic or pharmacologic etiologies for punctal stenosis were excluded. These etiologies have been described in the literature previously¹, and include autoimmune disease, infectious agents, eyelid malposition, chronic blepharitis, topical ocular therapy, and systemic chemotherapy. Institutional review board approval was obtained from UIHC with a waiver of consent due to the retrospective nature of the study.

Results

23 patients met inclusion criteria for this study (Table 1). 20 of the 23 patients (87.0%) were female. The mean age was 56.6 years, with a range of 37–85 years. Bilateral disease was present in 21 of the 23 patients (91.3%). Conjunctival biopsies from all 23 patients, representing 36 specimens, underwent histological examination. All patients underwent nasolacrimal duct probing with stent placement and punctoplasty at the time of conjunctival biopsy. There were no complications related to the conjunctival biopsy. The average volume of excised tissue was 30.3 mm³. 35 of 36 specimens (97.2%) showed stromal lymphocytic infiltrates. Three patients (13.0%) displayed evidence of chronic granulomatous inflammation. Two of these three patients (Cases 14, 16) had findings that were suggestive of sarcoidosis. Both patients required multiple surgeries for treatment. Case 14 also presented with chronic cough but further evaluation following the conjunctival biopsy found no evidence of granulomas on lung biopsy. ACE and lysozyme levels were also normal in this patient. Case 16 did not present with any systemic symptoms but was found to have an elevated ACE level.

Table 1.

Histopathologic and Direct Immunoflourescence Details of Presumed Idiopathic Punctal Stenosis Patients

Case #	Age (yrs), Gender	Laterality	DIF Report	Histopathology Report
1	58, F	Bilateral	Right bx only: No markings for fibrinogen, IgG, IgM, IgA, or C3.	Right bx only: band of lymphocytes in superficial stroma with scattered epithelioid histiocytes and giant cells.
2	47, F	Bilateral	Right bx only: linear fibrinogen marking along the BMZ. Globular markings with IgG, IgA, IgM at the D-E junction. No marking for C3.	Left bx only: mild to moderate band-like infiltrate of lymphocytes and plasma cells in superficial stroma.
3	55, F	Bilateral	Left bx only: thick, shaggy fibrinogen marking along BMZ. Numerous IgA and IgM positive lymphocytes seen below BM. No markings for IgM or C3.	Left bx only: thin band-like infiltrate of lymphocytes and plasma cells in superficial stroma.
4	48, F	Bilateral	Left bx only: linear fibrinogen marking along BMZ. No markings for IgG, IgM, IgA, or C3.	Left bx only: scattered lymphocytes with mild sclerosis of conjunctival stroma
5	67, F	Bilateral	Bilateral: shaggy fibrinogen marking along BMZ. No markings for IgG, IgM, IgA, or C3.	Right bx: mild, patchy lymphocytic infiltrate. Left bx: thin, band-like infiltrate of lymphocytes and plasma cells in superficial stroma.
6	63, F	Bilateral	Inadequate sample.	Bilateral: dense, follicular lymphocytic infiltrate of stroma.
7	73, F	Bilateral	Right bx only: discontinuous, shaggy fibrinogen marking along BMZ. No markings for IgG, IgA, IgM, or C3.	Right bx only: light, patchy lymphocytic infiltrate; band-like in some areas.
8	70, F	Bilateral	Bilateral: No markings for fibrinogen, IgG, IgM, IgA, or C3.	Right bx: light lymphocytic infiltrate in stroma; focal band- like areas. Left bx: minimal lymphocytic infiltrate.
9	83, M	Bilateral	Bilateral: fine linear fibrin marking along BMZ. No markings for IgG, IgM, IgA, or C3.	Right bx: neutrophilic infiltrate of epithelium; mild to moderate band-like lymphoplasmacytic inflitrate. Left bx: mild to moderate lymphocytic infiltrate.
10	40, F	Bilateral	Right bx: focal duplication of fibrinogen along BMZ. No markings for IgG, IgM, IgA, or C3. Left bx: no markings for fibrinogen, IgG, IgM, IgA, or C3.	Bilateral: dense band-like lymphocytic infiltrate.
11	66, F	Right	Not performed.	Right bx only: dense band-like lymphocytic infiltrate with nodular pattern.
12	37, F	Bilateral	Not performed.	Bilateral: moderately dense band-like lymphocytic infiltrate in stroma.
13	37, F	Right	Right bx only: No markings for fibrinogen, IgG, IgM, IgA, or C3.	Right bx only: moderate patchy lymphocytic infiltrate.
14	65, F	Bilateral	Not performed.	Bilateral: moderately dense band-like lymphocytic infiltrate in stroma, some follicles. Epithelioid histiocytes noted in left bx.
15	61, F	Bilateral	Right bx only: linear fibrinogen deposition along BMZ. No markings for IgG, IgM, IgA, or C3.	Right bx only: moderate patchy lymphocytic infiltrate.
16	45, F	Bilateral	Not performed.	Bilateral: moderate band-like lymphocytic infiltrate with focal granulomas.
17	56, F	Bilateral	Left bx only: globular, discontinuous, shaggy marking for fibrinogen along BMZ. No markings for IgG, IgM, IgA, or C3.	Right bx only: light, band-like, nodular lymphocytic infiltrate.
18	47, F	Bilateral	Left bx only: linear marking for fibrinogen along BMZ. No markings for IgG, IgM, IgA, or C3.	Left bx only: minimal patchy lymphocytic infiltrate.
19	59, F	Bilateral	Bilateral: No marking for fibrinogen, IgG, IgM, IgA, or C3.	Bilateral: minimal patchy lymphocytic infiltrate with sclerosis.
20	47, F	Bilateral	Bilateral: focal marking for fibrinogen with thick and splitting pattern at BMZ. No markings for IgG, IgM, IgA, or C3.	Right bx: moderate band-like lymphocytic infiltrate with sclerosis. Left bx: mild patchy lymphocytic infiltrate.
21	46, M	Bilateral	Right: focal marking for fibrinogen with thick and splitting pattern at BMZ. No markings for IgG, IgM, IgA, or C3. Left: No marking for fibrinogen, IgG, IgM, IgA, or C3.	Bilateral: mild to moderate band- like lymphocytic infiltrate.
22	47, M	Bilateral	Bilateral: focal marking for fibrinogen with thick and splitting pattern at BMZ. Nor markings for IgG, IgM, IgA, or C3.	Right bx: unremarkable conjunctiva. Left bx: rare lymphoid aggregates in superficial stroma.
23	85, F	Bilateral	Bilateral: shaggy marking for fibrinogen with splitting pattern at BMZ. No marking for IgG, IgM, IgA, or C3.	Bilateral: lymphocytic infiltrates with follicular hyperplasia throughout.

Abbreviations: DIF, direct immunofluorescence; bx, biopsy; BMZ, basement membrane zone; D-E, dermal-epidermal.

Conjunctival biopsies from 18 of the 23 patients, representing 27 specimens, were also evaluated by DIF. Four of the 18 patients (22.2%) (Cases 1, 8, 13, 19) had no identifiable deposition along the basement membrane zone (BMZ) after staining with the antibodies listed previously. Five patients (27.8%) (Cases 2, 4, 9, 15, 18) displayed fine linear deposition of fibrin or fibrinogen along the BMZ with no deposition of immunoglobulin or complement C3. Nine patients (50.0%) (Cases 3, 5, 7, 10, 17, 20, 21, 22, 23) had depositions of fibrinogen – described as being shaggy, discontinuous, duplicative, globular, and/or split – along the BMZ (Figure 1). These findings were described as being consistent with a lichenoid reactive process, such as lichen planus (LP). Seven of these nine patients (77.8%) were female, and all nine patients (100%) presented with bilateral punctal stenosis.

Figure 1. Abnormal conjunctival basement membrane zone morphology on direct immunofluorescent staining with anti-fibrinogen antibodies.

A, Shaggy deposition of fibrinogen. B, Fibrinogen with reduplication and fragmentation. C, Fibrinogen with spikes. All images at 40x magnification.

Of these nine patients with DIF findings consistent with LP, none had a history of skin or mucosal lesions or previous diagnoses of LP. Seven of the nine patients (Cases 5, 10, 17, 20, 21, 22, 23) were started on cyclosporine A 0.05% eyedrops (Restasis) after the results of the DIF were obtained.

Discussion

Many etiologic processes and pharmacological agents have been associated with the acquisition of punctal stenosis, yet until two recent studies,^2,3 little was known about the underlying pathophysiologic processes. This study further expands upon those previous findings while also examining the utility of conjunctival biopsy in patients with presumed idiopathic punctal stenosis.

Both recent studies examined punctoplasty specimens^2,3 and showed evidence of chronic inflammation in the majority of specimens. Port et al. (2013) examined punctal histopathology in patients with punctal stenosis and found that over 83% of these specimens displayed signs – such as chronic inflammation, fibrosis, and squamous metaplasia – that were consistent with an inflammatory process. Presumed punctal stenosis etiologies in these patients included chronic blepharitis, Meibomian gland dysfunction, and topical and chemotherapeutic medications. Ali et al. (2015) found fibrosis, again suggesting a chronic inflammatory process, in 100% of their punctal specimens from patients with undefined etiologies of punctal stenosis. Our study, in which 97.2% of conjunctival specimens displayed inflammatory infiltrates, further corroborates both of these prior two studies and the current proposed mechanism of acquired punctal stenosis, which involves a chronic inflammatory process leading to fibrosis and punctal occlusion. Importantly, our study population was purely patients with presumed idiopathic punctal stenosis.

The indication for biopsy in these patients was the finding of punctal stenosis without any know etiology. Although this is not routine, the performance of this was based on the hypothesis that the etiology of the punctal stenosis would be reflected in the adjacent conjunctival tissue. Known causes of punctal stenosis that were not elicited in history or examination of the patient are known to have histopathology and immunopathology findings in the conjunctiva, such as ocular cicatricial pemphigoid and sarcoidosis. This was the basis for performing conjunctival biopsies on these patients with presumed idiopathic punctal stenosis.

Two patients (Cases 14, 16) were found to have conjunctival histopathology suggestive of sarcoidosis, an inflammatory disorder that most commonly affects the lungs and lymph nodes. However, it can also affect the eyes and orbit, most commonly manifesting as anterior uveitis. Blind biopsies to screen for sarcoidosis are generally low yield for positive results. One can hypothesize that there is a selection bias in this population of patients with punctal stenosis, 97% of whom were found to have inflammatory changes in the conjunctiva, thus increasing the positive yield for finding changes consistent with sarcoidosis. Involvement of the lacrimal drainage system in sarcoidosis appears to be quite rare, with one paper reporting 1.8% (n=281) of patients with ocular sarcoidosis having lacrimal obstruction, most commonly the nasolacrimal duct.⁴ Patients with obstruction of the lacrimal system associated with sarcoidosis are notoriously difficult to treat, often requiring multiple surgical procedures before symptoms are resolved. Both patients in our study underwent several lacrimal surgeries to treat their epiphora.

Of the 18 patients who underwent conjunctival DIF examination, five patients were found to have linear fibrin or fibrinogen markings along the BMZ, while nine others had BMZ fibrinogen markings described as shaggy, discontinuous, duplicative, and/or split. Unlike in skin and the oral mucosa, where the presence of subepithelial fibrinogen is abnormal and in itself very suggestive of LP, subepithelial fibrinogen in the conjunctiva has been shown to be a normal feature.⁵ With conjunctival specimens, it is vital to not look just for the presence or absence of fibrinogen, but rather its morphology. A fibrinogen layer with spikes, spurs, fragmentation, reduplication, or a shaggy appearance is more likely to be consistent with a diagnosis of LP.^6,7,8 Therefore, the five patients with linear markings most likely display only non-specific inflammation, whereas the other nine patients show morphology that is indicative of a lichenoid process.

LP, a mucocutaneous auto-inflammatory disease, most commonly affects the skin and oral mucosa. Conjunctival involvement with LP is a relatively rare feature of the disease, with 64 cases described in the literature and the vast majority of patients presenting with cicatricial changes.^{5,9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24} Even more uncommonly reported is LP with lacrimal drainage system involvement. Only 11 such patients have been described in various case reports and series^9,10 before Durrani et al. (2008) performed a retrospective review of 184 patients with canalicular obstruction. The authors found that eight of these patients had histories of systemic LP, with canalicular tissue biopsy showing changes also consistent with LP. These results led the authors to advise clinicians to be suspicious for LP in cases of bilateral canalicular obstruction without the presence of other risk factors. However, all these previous reports of lacrimal obstruction secondary to LP involved patients with some kind of systemic manifestations. The one exception is the study by Brewer et al. (2011) in which three of eleven patients had LP isolated to the conjunctiva, only one of which was also described to have punctal stenosis. The nine relevant patients in our study did not have any other associated mucocutaneous or skin findings. These cases seem to be examples of LP isolated to the conjunctiva and lacrimal drainage system leading to punctal stenosis without other systemic systemic or clinical conjunctival findings. While there are three published reports describing six patients with isolated LP leading to cicatricial conjunctivitis^14,15,25, these cases centered on the conjunctival findings without description of the examination of the lacrimal system.

The scarcity of similar cases in the existing literature raises the question of whether this entity is an under-diagnosed disease or if other classes of conjunctival inflammation can also appear similar under DIF. If these cases do represent true instances of LP, they suggest that isolated LP to the conjunctival may be a relatively common cause of punctal stenosis in patients without other risk factors. Thus, patients meeting these criteria may benefit from prompt histopathological and DIF examination.

Treatment for patients with systemic LP with ocular symptoms has primarily consisted of oral cyclosporine.⁹ In a report describing two patients with lichen planus isolated only to the conjunctiva, the patients were started on topical cyclosporine A 0.05%.²² The authors found that this topical medication adequately controlled inflammation and the patients’ symptoms. Given the lack of systemic symptoms in our patients, systemic cyclosporine use was not indicated; seven of our patients (Cases 5, 10, 17, 20, 21, 22, 23) were started on topical ophthalmic cyclosporine A 0.05%. Cases 10, 17, 22, 23 had marked improvements in epiphora symptoms at follow-up time ranging from 4–16 months. The effects of topical cyclosporine on Cases 5, 20, and 21 are still pending at this time.

In total, of the 23 patients in this study with presumed idiopathic punctal stenosis and epiphora, 11 (47.8%) were found to have underlying etiologies for their presentation. This information suggests the importance of performing a thorough work up, including both histopathologic and DIF examination, in these patients with presumed idiopathic punctal stenosis. Diagnosis may impact the implementation of medical therapies or the success rate of surgical therapies.

The utility of conjunctival biopsy in patients with punctal stenosis was also examined in this study. The previous studies finding histopathological evidence of chronic inflammation in these patients^2,3 included punctal specimens. That conjunctival biopsies in this study displayed similar findings with regards to inflammation as those seen in the previous punctal specimens suggests that conjunctival biopsies may indeed be a safe and effective proxy in patients with punctal stenosis. The use of adjacent tissue to the puncta rather than the punctal tissue itself is a limitation to the study. The main advantage of using the conjunctiva is the volume of tissue available for histopathology and immunohistochemistry, compared to the relatively scarce amount of tissue available from the puncta. Whether these pathological changes observed in this population with presumed idiopathic punctal stenosis would be observed in tissue from the canaliculus, lacrimal sac, nasolacrimal duct, or nasal mucosa is speculation, and additional studies would be needed to examine this.

Additional limitations to this study include the retrospective method and the lack of a control group, consisting of histopathologic and DIF examination of conjunctiva from patients without punctal stenosis, although the literature has already described the findings in normal and inflamed conjunctiva. Future studies are needed to determine the characteristic DIF findings in other etiologies of conjunctival inflammation, which will allow us to more confidently diagnose ocular adnexal LP via conjunctival DIF.