Purpose
To characterize the clinical features of patients with direct immunofluorescence (DIF)-negative mucous membrane pemphigoid (MMP).
Design
Retrospective case series.
Methods
Thirty-six patients who underwent a conjunctival biopsy for suspected MMP were included. Demographic and clinical information was collected. Main outcome measures included visual acuity, Foster stages, presence of extraocular involvement, history of autoimmune disease, and durations of follow-up.
Results
Thirty-two patients had a negative DIF. Of those, 2 had a positive DIF on repeat biopsy. Eleven showed progression of conjunctival scarring during a median follow-up of 42 months (range, 8-100 months) and were diagnosed with biopsy-negative MMP. Another 11 patients with a median follow-up of 54 months (range, 15-138 months) were diagnosed with cicatrizing conjunctivitis of other causes. The median visual acuity of patients with biopsy-negative MMP at presentation was significantly lower compared to patients with cicatrizing conjunctivitis of other causes (20/400 vs 20/40, P = .02). Conjunctival scarring score at presentation in both biopsy-positive and biopsy-negative MMP groups was significantly higher compared to patients with cicatrizing conjunctivitis of other causes (median Foster stage, 3 vs 1, P = .009; and 3 vs 1, P = .01, respectively).
Conclusions
Patients with progressive cicatrizing conjunctivitis likely have MMP in the absence of alternate diagnoses. Our findings emphasize that suspicion for MMP must remain high for patients who have Foster stage 3 conjunctival scarring on presentation or worsening of scarring during follow-up, even in the setting of negative DIF.
Cicatrizing conjunctivitis is an uncommon ocular surface disease characterized by conjunctival scar formation that ranges from subepithelial fibrosis to ankyloblepharon. , Progressive corneal neovascularization and conjunctivalization due to ocular surface inflammation and cicatricial eyelid changes can lead to irreversible vision loss.
Many diseases inciting conjunctival inflammation result in cicatrizing conjunctivitis. Its etiology encompasses chemical injury, infection (trachoma, Corynebacterium diphtheriae conjunctivitis, streptococcal conjunctivitis), inflammatory disorders (ocular rosacea, Sjögren syndrome, atopic keratoconjunctivitis, graft-vs-host disease, drug-induced cicatrizing conjunctivitis), autoimmune disorders (mucous membrane pemphigoid [MMP], lichen planus), malignancy, squamous cell carcinoma, and allergic reactions (Stevens-Johnson syndrome/toxic epidermal necrolysis). Once an acute case of conjunctival scarring is excluded, a conjunctival biopsy is crucial to diagnose the exact underlying etiology in chronic cicatrizing conjunctivitis.
Examination of the conjunctival specimens to determine the cause of cicatrizing conjunctivitis includes a combination of light microscopy (LM), direct immunofluorescence (DIF), and electron microscopy. DIF is a valuable tool to diagnose MMP, which can cause vision- and life-threatening sequelae. The DIF detection of linear deposition of immunoreactants (immunoglobulin [Ig] G, IgA, IgM, or complement 3 [C3]) in the basement membrane zone is the main diagnostic criterion for MMP set by the First International Consensus on MMP. However, the limitation of DIF on the conjunctival tissue has been well recognized, with a reported positive rate ranging from 30% to 80% ( Table 1 ). In our experience, there is a subset of patients with cicatrizing conjunctivitis and negative DIF whose disease course behaves identical to biopsy-positive MMP patients and exhibit aggressive progression of conjunctival scarring. When these patients are managed as non-MMP cicatrizing conjunctivitis without timely intervention, they are at inevitable risk of irreversible blindness. This subset of patients, so-called “biopsy-negative MMP” also has been reported in other studies ( Table 1 ).
| Study | Positive-Biopsy MMP/Total Number of MMP Patients (%) | Negative-Biopsy MMP/Total Number of MMP Patients (%) |
|---|---|---|
| Thorne et al | 74/86 (86) | 12/86 (14) |
| Mehra et al | 17/38 (45) | 21/38 (55) |
| Goldich et al | 8/27 (30) | 17/27 (63) |
| Mudhar et al | 31/41 (76) | 10/41 (24) |
| Labowsky et al | 44/55 (75) | 41/55 (75) |
| Ong et al | 43/73 (59) | 26/73 (36) |
As clinicians rely heavily on clinical judgment to manage those patients with negative conjunctival biopsy, understanding the clinical course of these patients is crucial. In this study, we aimed to characterize the clinical features and outcomes of those patients suspected of MMP on presentation and who had a negative DIF from the diagnostic conjunctival biopsy.
Methods
This retrospective case series was approved by the Johns Hopkins Institutional Review Board and was conducted in accordance with the principles of the Declaration of Helsinki. Informed consent was waived as there was no prospective recruitment.
Billing records were searched electronically using the Current Procedural Terminology code 68100 for conjunctival biopsy to create a list of patients who were evaluated at the Ocular Surface Diseases and Dry Eye Clinic, The Wilmer Eye Institute, The Johns Hopkins University School of Medicine, Baltimore, Maryland, by a single physician (E.K.A.). The medical records of the patients were reviewed retrospectively by 2 ophthalmologists (J.Y. and E.K.A.) to collect information regarding demographics, medical history, clinical presentation, histopathologic findings, and clinical course. Data were entered into an electronic database for analysis.
In all cases, slit lamp photographs were used to designate disease staging based on Foster’s classification system as follows : stage 1, subconjunctival fibrosis; stage 2, inferior fornix foreshortening; stage 3, symblepharon; and stage 4, ankyloblepharon and surface keratinization. There were 4 cases (1 with biopsy-negative MMP and 3 with cicatrizing conjunctivitis of other causes) where slit lamp photographs taken on presentation were not available for a review as the photographs were taken prior to electronic medical records. In these cases, reviewers assessed progression or nonprogression based on documented interpretation of nondigital photographs, clinical examination description and drawings documented in the medical records of the patients.
Conjunctival biopsy was performed in the clinic after informed consent was obtained. The ocular surface was anesthetized with topical proparacaine and prepared using 5% betadine solution. Subconjunctival 1% lidocaine with 1:100,000 epinephrine was injected into the lower bulbar and forniceal conjunctiva via a 30-gauge needle. A sterile fine smooth forceps and Westcott scissors were then used to obtain a 2×6-mm piece of inferior forniceal or bulbar conjunctiva from the site of active inflammation. The tissue was then trisected and placed in formaldehyde, glutaraldehyde, and Michel solution and submitted for LM, electron microscopy, and DIF, respectively. The biopsied eye was dressed with an antibiotic-steroid combination ointment and patched overnight.
Biopsy-positive MMP was diagnosed based on the presence of typical clinical findings of cicatrizing conjunctivitis and DIF showing the linear deposition of IgA, IgG, IgM, or C3 immunoreactants in the basement membrane zone of the conjunctival specimen. Biopsy-negative MMP was defined as a case with clinically documented progression of conjunctival scarring in the absence of immunoglobulins and/or complement deposition at the basement membrane and any alternate etiology based on LM findings and clinical grounds.
Because of the small sample size, median values and ranges were reported for continuous variables. The Kruskal-Wallis test or Mann-Whitney U test was used to compare the median values of the continuous variables. A χ 2 test or a Fisher exact test was used to compare categorical variables. P values of <.05 were considered statistically significant. All statistical analyses were performed using IBM SPSS Statistics version 23 (IBM Corp, Armonk, New York, USA).
Results
A total of 118 patients underwent a diagnostic conjunctival biopsy during the 8-year study period between January 1, 2007, and October 31, 2015. Thirty-six of those patients were evaluated specifically for suspected MMP as they had clinical evidence of conjunctival scarring without any known or anticipated ocular or systemic diagnoses ( Figure 1 ). Patients who underwent a confirmatory conjunctival biopsy for a suspected condition (eg, a patient with suspected lung sarcoidosis and conjunctival nodules or a patient with known graft-vs-host disease) were excluded. Patients who underwent the biopsy for other reasons, such as chronic follicular or papillary conjunctivitis or excisional biopsies for neoplastic lesions, were also excluded.
Overall, the mean age of the patients at presentation was 69 years. There was a female preponderance, with 29 women and 7 men in the cohort. The average follow-up period for all patients was 40 months (range, 1-138 months).
Among all 36 patients, a total of 5 patients had a positive DIF biopsy with a linear deposition of immunoreactants to epithelial basement membrane on biopsy. Only 2 of these patients (2/36, 6%) showed a positive DIF on initial biopsy. The other 3 patients had a positive DIF on a repeat biopsy. One of these patients had an absent epithelium on initial biopsy and the specimen could not be interpreted. In other 2 patients, clinical suspicion was high with active conjunctival inflammation and discomfort, and repeat biopsies were performed, which were read positive.
Thirty-one patients (31/36, 86%) had negative DIF on conjunctival biopsy. Eleven patients (11/31, 35%) had progression of cicatrizing conjunctivitis based on slit lamp photographs and were diagnosed with biopsy-negative MMP and treated accordingly.
Another 11 patients (11/31, 35%) with a negative DIF and at least 1 year of follow-up were diagnosed with various other conditions as a cause of their cicatrizing conjunctivitis.
Nine patients with less than 1 year of follow-up and without a definitive diagnosis were excluded from further analyses.
The site of conjunctival biopsy developed further conjunctival scarring visible at the slit lamp in a few of the patients; however, there was no advancement of Foster staging at the site of biopsy.
Patients With Biopsy-Positive MMP
In this study, MMP was confirmed by DIF only in 5 patients (5/16, 31%). Two patients showed positive DIF on the initial biopsy. Two patients with negative initial DIF had a positive repeat biopsy using avidin-biotin complex immunoperoxidase techniques. One patient with denuded epithelium on the initial biopsy demonstrated IgA deposition in the basement membrane by DIF on the repeat biopsy. The median Foster clinical conjunctival scarring stage in this group was 3 (range, 2-3), with only 1 patient having conjunctival scarring less than stage 3. Only 1 patient in this group had extraocular mucocutaneous manifestation (oral mucosal involvement). There was no patient on glaucoma drops. The median follow-up duration was 31 months (range, 4-72 months). All 5 patients were treated with systemic immunosuppression. Two patients required fornix reconstruction during the follow-up.
Patients with Biopsy-Negative MMP
Eleven patients with negative DIF demonstrated a progression of cicatrizing conjunctivitis and were diagnosed with biopsy-negative MMP in the absence of alternative diagnoses on clinical and LM examination ( Figure 2 ). One patient underwent 2 additional biopsies during follow-up, both of which were also negative. The median Foster clinical conjunctival scarring stage was 3 (range, 1-4), with 4 patients having less than stage 3. The median follow-up duration was 42 months (range, 8-100 months). None of the patients in this group had extraocular mucosal inflammation. Four patients had a history of glaucoma on intraocular pressure-lowering drops.
