Conjunctival inflammation in patients under topical glaucoma treatment with indication to surgery

PURPOSE: To compare the frequency of conjunctival HLA-DR expression (a surrogate marker for inflammation) in eyes treated with topical prostaglandin analogues versus eyes treated with other topical antiglaucomatous drugs. METHODS: Patients diagnosed with primary open-angle glaucoma presenting indication for trabeculectomy were divided in groups according to the use or not of prostaglandin analogues. All subjects were treated with the maximum tolerated dose of antiglaucomatous drugs until the date of the surgery. At the beginning of the surgical procedure, a 5 x 5 mm biopsy of the bulbar conjunctiva was collected, incubated with monoclonal anti-HLA-DR antibody and processed for histological analysis. RESULTS: Of the 31 eyes included (31 patients), 25 were under topical prostaglandin analogues (Group 1) and six under other topical pharmacological agents (Group 2). Fourteen eyes of Group 1 (56%) and three of Group 2 (50 %) were positive for the inflammatory marker HLA-DR (P=1.0). The percentage of stained cells ranged from 15.49 to 48.09% (median: 27.61) in Group 1, and from 18.35 to 28 (median: 20.71) in Group 2, with no differences statistically significant (p=0.33). CONCLUSION: The use of prostaglandin analogues did not increase conjunctival expression of HLA-DR compared to other topical antiglaucomatous agents.


Introduction
Glaucoma affects up to 60.5 million people worldwide, and more than 8 million of them are blind 1 .Medical therapy is the first choice therapy to reduce intraocular pressure and reduce the risk of visual field loss 2 .Among topical agents, prostaglandin analogues are currently the first-line therapy for glaucoma 3 , and some of the side-effects related to their use are iris pigmentation, transient conjunctival hyperemia 4 , corneal punctate staining 5 , squamous metaplasia 6 and subclinical conjunctival inflammatory reaction that, although subclinical, may be detected by inflammatory markers, such as HLA-DR, expressed on epithelial cells [7][8][9][10] .
There are studies in the literature describing that chronic topical therapy for glaucoma increases the risk of trabeculectomy failure, and that the preoperative conjunctival inflammation caused by the chronic antiglaucomatous medication would be involved in this process 11 .But although most of the studies report conjunctival inflammation in prostaglandin analogues users 7,8,10,12 , this feature may also be present in patients under the use of other antiglaucomatous drugs 9,10,[13][14][15] .
Guenoun et al. 16 studied, in vitro, conjunctiva-derived cells under the action of prostaglandin analogues in commercial solutions, prostaglandin F2 alpha, TNF-alpha, and of benzalkonium chloride (BAK).The results suggested that toxicity is induced by the preservative rather than the medication itself.Uusitalo et al. 17 reported that even patients using preservative-free prostaglandin analog may present conjunctival expression of HLA-DR 17 .
As the success of glaucoma surgical treatment depends on the status of the conjunctival and subconjunctival tissue that will be part of the surgical bleb 11 , it is important to analyze if prostaglandin analogues are able to induce subclinical inflammation, also if this inflammation is higher than the one caused by other topic agents.
In this study, we selected patients with primary openangle glaucoma with indication to trabeculectomy surgery, divided them in groups according to the use or not of topical prostaglandin analogues, and compared the conjunctival expression of HLA-DR.

Methods
The investigation was carried out according to Helsinki Subjects were divided in two groups according to the use or not of topic prostaglandin, and exclusion criteria were: presence of pterygium 18 , signs of clinical inflammation, and any type of previous surgery in the studied eye.All patients were treated with the maximum tolerated dose of antiglaucomatous drugs until the date of the surgery.
At the beginning of the surgical procedure, a 5 x 5 mm biopsy of the inferior bulbar conjunctiva was collected (6 o'clock position), fixed in formol and processed for histological analysis.
The preparation of the slides has been previously described 12 .Briefly, slides were incubated at room temperature in a moist chamber using the primary monoclonal anti-HLA-DR antibody at a 1:50 concentration (Dako™, Glostrup, Denmark).
Histological sections of human lymphoma were used as positive controls to examine the immunologic reaction.Negative controls consisted of the same histological sections but did not include the step of incubation with the primary antibody; this antibody was replaced with PBS.
Quantitative immunohistochemical analysis of HLA-DR expression was performed by image analysis using an Olympus BX41 trinocular microscope coupled with a CCD Samsung™ color camera (SDC-313/243) and PCTV software from the Pinnacle Media Center™.The images were analyzed using the Image J software (version 1.41, National Institutes of Health, USA) 19 .Cells were considered to be positive for the marker if a brown immunoprecipitate was present in the cytoplasm and/or on the cytoplasmic membrane.
The cell counter method was used to count labeled cells directly from the monitor of the computer; marks of different colors were placed on the positive and negative cells with a click of the mouse.After the count, all data were exported to Excel 2007 ® spreadsheets (Microsoft Corporation, USA), which provide the sum of the results of the count on one slide.At least 600 cells per biopsy were counted and the percentage of cells showing immunohistochemical expression of HLA-DR was calculated.
The statistical analysis was performed using the Fisher´s Exact test and ANOVA (for differences between percentages of stained cells).Significant differences were considered when p<0.05.

Results
Biopsy samples were obtained from 31 eyes, from 31 patients (22 male), with ages ranging from 50 to 76 years (median=66 years).Twenty five eyes were under prostaglandin analogue treatment (Group 1) and six were under the use of other pharmacological agents (Group 2) during the 30 days prior to the surgical procedure.The number of antiglaucoma eye drop preparations in use ranged from 1 to 4 (median 3) per eye, and the most frequently used medication was 0.5% timolol maleate (27 eyes).

Discussion
In the present study, the use of prostaglandin analogues was not a risk factor for conjunctival HLA-DR expression.In healthy subjects, conjunctival expression of Human Leucocitary Antigens (HLA) is low.But in patients under antiglaucomatous topical medication, HLA-DR and other inflammatory markers are detected even in the absence of clinical signs of inflammation, such as conjunctival hyperemia [6][7][8][9][10][12][13][14][15] .
Although most of the similar studies found in the literature were performed with patients in use of prostaglandin analogues, it is currently accepted that BAK, a preservative widely used in ophthalmic preparations, play an important role in conjunctival inflammation 20 .
Recently, it was described that patients presenting ocular side-effects due to topic prostaglandin analogues had a better tolerability when switching to a preservative-free prostaglandin solution (tafluprost) 17 .But although the percentage of patients presenting HLA-DR expression decreased, 87% of them were still expressing abnormal levels of HLA-DR after 12 weeks under preservative-free medication, which suggests that BAK is not the only factor involved in subclinical conjunctival inflammation.
In a previous publication, our group demonstrated a higher expression of conjunctival HLA-DR in eyes from patients in use of topic prostaglandin analogues for thirty or more days (94% of all eyes), although patients remained without any clinical sign of conjunctival inflammation 12 .In the current study, although most of the subjects under chronic medication presented conjunctival HLA-DR expression, the prevalence was lower (54.8%including all eyes, 56% considering only the group under prostaglandin analogues medication).This frequency could be explained by the differences in methodology (impression cytology versus biopsy).Another potential explanation could be that subclinical inflammation is more frequent in early stages of the treatment.Rodrigues et al. 12 observed HLA-DR expression in conjunctival cells 94% of the patients followed-up, after one month of treatment with prostaglandin analogues.
We decided to select patients with indication of trabeculectomy, so a conjunctival biopsy could be ethically justified at the moment of the surgery.That way, more cells can be analyzed than when an imprint is obtained.
One of the limitations of this study is the reduced number of patients, especially in Group 2. For ethical reasons, the patients remained with the maximum tolerated treatment until the day of the surgery, when the specimens were collected.
As prostaglandin analogues currently are the first choice drugs in antiglaucoma therapy 3 , most of the patients were under this type of medication.Thus, the present study was performed before the preservative-free prostaglandin tafluprost was commercially available, so the effect of this drug on conjunctival HLA-DR expression could not be evaluated.
The frequency of subclinical inflammation in both groups suggests that this condition may be attributed to both, prostaglandin analogues and the vehicle of ophthalmic solutions used in glaucoma treatment (BAK).Further studies analyzing the expression of inflammatory markers in asymptomatic patients under preservative-free antiglaucoma medication should be performed in order to investigate if these medications, although causing less ocular symptoms, are able or not to produce subconjunctival inflammation and also to increase the risk of trabeculectomy failure.

Conclusion
HLA-DR conjunctival expression in eyes treated with prostaglandin analogs was not significantly different of the expression in eyes treated with other topical agents, used for glaucoma treatment.
Declaration (Medical School of Ribeirao Preto University Hospital Ethics Committee protocol number 15582/2005).The expression of the inflammatory marker HLA-DR was analyzed in biopsies of the bulbar conjunctiva from patients diagnosed with primary open-angle glaucoma presenting indication for trabeculectomy, followed at the Glaucoma Outpatient Clinic of the University Hospital, Faculty of Medicine of Ribeirao Preto-University of Sao Paulo.