Abstract
Purpose.:
The aim of the present randomized, double-blind, placebo-controlled clinical trial was to assess the efficacy and tolerability of 0.5% indomethacin (INDOM) eye drops in adult patients suffering from macular edema (ME) related to different etiology uveitis.
Methods.:
Forty-six eyes of 31 adult patients (20 females and 11 males) mean age 39 years, affected by inflammatory ME, were randomized to receive a dose of commercial 0.5% INDOM eye-drops four times per day (16 subjects = 23 eyes) or placebo (the vehicle of INDOM, 15 subjects = 23 eyes) during a 6-month active therapy follow-up. Study assessment at each visit included visual acuity testing (VA), slit-lamp examination, IOP evaluation, and Heidelberg Spectralis optical coherence tomography (OCT) central foveal thickness (CFT) measurement. Any variation in subjective symptoms and tolerability was also detected.
Results.:
Statistical analysis showed, from baseline to the 6-month visit, a significant reduction in CFT (P < 0.0001) and a significant improvement in VA only in the 0.5% INDOM-treated group; a global reduction of discomfort symptoms was present in both groups (P < 0.001).
Conclusions.:
The four times per day administration of 0.5% INDOM eye drops in eyes affected with uveitic ME from different etiologies, compared with placebo, is associated with a significant reduction in ME at the 6-month follow-up visit, as measured by spectral-domain optical coherence tomography (SD-OCT). However, not all eyes showed a complete resolution of ME because of vitreoretinal traction. ( https://eudract.ema.europa.eu/index.html number, EUDRACT 2011-001522-20.)
The present work is a prospective, randomized, double-blind, with paired groups, controlled with placebo, study (EUDRACT 2011-001522-20), including 46 eyes of 31 patients (20 females and 11 males), affected by inflammatory uveitic ME in one or both eyes. These numbers are related to the eyes/subjects who fully completed the 6-month follow-up.
The total number of included subjects was 43 (68 eyes), but 12 patients were lost during the follow-up period; five patients (eight eyes) were not compliant to the scheduled protocol, and seven patients (14 eyes) were excluded because during the follow-up period they required a rescue treatment in relation to the worsening of visual acuity and/or of inflammatory CME (four subjects underwent systemic immunosuppressive treatment with mycofenolate mofetil, in association with oral steroids, and three patients underwent interferon α-2a in association with systemic steroids and immune suppressants).
Patients were randomized to continuously receive 0.5% INDOM eye drops (INDOM; Intes Industria Terapeutica Splendore, Naples, Italy) four times per day (QID) (Group B), or placebo (the vehicle of INDOM, Group A) during a 6-month (+20 days) period treatment.
Table 1 shows the baseline characteristic of both groups.
Table 1 Demographic and Clinical Characteristics
Table 1 Demographic and Clinical Characteristics
| | Group A, n = 16 (23 Eyes) | Group B, n = 15 (23 Eyes) | P Value |
| Age, y | 59 ± 19.03 | 65.69 ± 11.27 | 0.10 |
| Sex, male/female | 6/9 | 5/11 | 0.61 |
| Central foveal thickness at baseline, μm | 390.6 ± 162.3 | 446.0 ± 149.4 | 0.11 |
| Global discomfort index at baseline | 1.08 ± 0.90 | 1.04 ± 0.7 | 0.81 |
The study inclusion criteria was age over 18 years, both sexes, any race, any kind of uveitis complicated by OCT-detected ME, subjects able to give informed consent, to actively participate in the scheduled visits of the study, and not included in other clinical studies.
The exclusion criteria included age under 18 years, participation in another study, positive history for recent (6 months previously) ocular trauma, ocular surgery, ocular herpes or other kinds of inflammatory or infective disease of the eye and its adnexa, diabetic retinopathy, glaucoma, recent (3 months before) intravitreal treatment, exfoliative syndrome, active peptic ulcer, severe liver or renal impairment, uncontrolled asthma, chronic systemic treatment with NSAIDs, mono-ocularity, pregnancy or breast feeding, and any intolerance or known hypersensitivity to the study drug or to other NSAIDs. In the end, 31 subjects were selected to participate in this study.
The detected causes of inflammatory ME are summarized in
Table 2. In particular, we found in 23 eyes (50%) of 15 patients, intermediate uveitis of sarcoid, idiopathic, demyelinating origin; in 17 eyes (37%) of 10 subjects, autoimmune retinal vasculitis derived from rheumatoid arthritis or related conditions, Crohn's disease and sarcoidosis; in three eyes (6.5%) of three patients, post infective (one Lyme disease, one Varicella Zoster Virus infection, one Toxoplasmosis) retinal vasculitis; in three eyes (6.5%) of three different patients, HLA B
27 + related anterior uveitis.
Table 2 Main Causes of Inflammatory Macular Edema
Table 2 Main Causes of Inflammatory Macular Edema
| Type of Uveitic CME | Eyes N (%) | Patients N (%) |
| Intermediate uveitis | 23 (50) | 15 (48.4) |
| Retinal vasculitis | 17 (37) | 10 (32) |
| Postinfective uveitis | 3 (6.5) | 3 (9.8) |
| HLA B27/anterior uveitis | 3 (6.5) | 3 (9.8) |
| Total | 46 (100) | 31 (100) |
In all the included eyes, we ruled out the presence of an active (Herpetic, Toxoplasmic, or Borrelia Burgdoferi–related) infection by means of aqueous PCR examination; the three postinfective CME cases were PCR-negative, and were considered as immune-related uveitis.
Acute (not present at previous check-ups) OCT-detected inflammatory ME was the only selection criteria.
Concomitant ongoing systemic treatments with steroids (prednisone), immune-suppressants (azathioprine, methotrexate, cyclosporine, and mycophenolate mofetil), and biologicals (infliximab, adalimumab, and rituximab) were recorded at each visit and, during the 6-month period study, were not modified.
Local treatment with mydriatic (1% tropicamide or 0.5% atropine) or steroid (0.2% dexamethasone) eye drops to treat occurring acute anterior uveitis was permitted only for 2 weeks at a 5-day tapered dose (from QID) and only twice during the 6-month protocol.
Patients were selected from November 2011 until July 2012 in Rapallo Hospital, Tertiary Referral Uveitis Centre (Geneva, Italy) and over a period of approximately 180 days, subjects included into the study were scheduled to undergo four study visits: at baseline (V0), at 30 + 3 days (V1), at 90 + 10 days (V2), and at 180 + 20 days (V3).
Study assessments at each visit included accurate history and evaluation of inclusion/exclusion criteria if any change had occurred during the trial, visual acuity (VA) testing (Sloan decimal scale), anterior segment slit-lamp examination, fluorescein staining for corneal epithelial damage or dry eye evidence, IOP measurement (Goldmann applanation tonometry), pupil dilated fundus examination, OCT (HRA+OCT Heidelberg Spectralis; Heidelberg Engineering, Heidelberg, Germany) to assess CFT (μm).
To test VA, we used the VA electronic optotype (Vista Vision Wide Iris; DMD MED TECH, Villarbasse, Italy) and to uniform results we used the Sloan 5 letters optotype (the registered decimal value matches to the line where at least three letters were correctly read).
Central macular thickness (fovea center) was appraised by Spectralis HRA+OCT instrument (Heidelberg Engineering) and the T
0 value (in μ) was compared with the following measurements by using the “follow-up” program of the instrument. We took as normal CFT, the value of 280 μ + 20 μ, as written in recent literature.
28–30
General history investigation, included local or systemic allergy, systemic hypertension or cardiovascular diseases, respiratory system, liver, metabolism diseases, and recent surgery (up to 1 year before). Ocular history survey included dry-eye syndrome, glaucoma, recent trauma or surgery, lid, conjunctiva, and cornea diseases.
Before beginning the study, all patients were required to provide written informed consent and the study was approved without any change by the Rapallo Hospital ethical committee and was conducted following Helsinki Declaration guidelines.
Eye drops were administered to patients in the single dose formulation (0.5 mL) in a strip containing 10 single unit dose preservative-free eye drops, again contained in a large box of 10 blisters (for a total of 100 single-use eye drop bottles). Each single dose bottle contained 0.5% INDOM for Group B patients or an artificial tear (vehicle of INDOM for Group A). The latter product was a methyl-hydroxy-propyl-cellulose “4000,” monobasic and dibasic sodium phosphate, chloride sodium, edetate sodium, and purified water mixture, used also as an artificial tear substitute.
Patients who met the inclusion/exclusion criteria, were randomized in a 1:1 ratio to instill one drop QID either 0.5% INDOM or vehicle eye drops for a period of approximately 6 months. Allocation of therapy was determined by a unified randomization table provided by the company (Intes Industria Terapeutica Splendore) to the investigator. Both kinds of eye drops were labeled identically and divided into group A and group B boxes.
A questionnaire about treatment tolerance and agreement was given to all the subjects and at each control visit any variation was recorded. The questionnaire, included in a range from 0 to 3 (0 none; 1 mild; 2 moderate; 3 severe), main symptoms such as tearing, itching, photophobia, burning, and global discomfort sensation.
Primary efficacy end-point was CFT reduction, detected by testing the differential value between day 0 and 180 days after treatment.
Statistical analysis on differences between baseline and 30, 90, and 180 days after treatment for CFT and VA was carried out with repeated measures analysis with clustered subject (or nested model procedure) because it is useful to correlate subjects in which only one or both eyes were treated.
To evaluate the primary efficacy end-point (change in foveal thickness during the treatment time) the Bonferroni correction has been used and χ2 test was used to test secondary parameters of efficacy.
Variability in adherence across age quartiles (
Table 3) were evaluated using a Fisher's exact test. A
P value less than or equal to 0.05 was considered as statistically significant.
Table 3 Age Categorized by Quartiles
Table 3 Age Categorized by Quartiles
| | Min | First Quartile (Q1) | Median (Q2) | Third Quartile (Q3) | Max |
| Group (A+B) | 22 | 53 | 62.35 ± 15.83 | 75 | 83 |
| Group A | 22 | 38 | 59 ± 19.03 | 75 | 83 |
| Group B | 38 | 60 | 65.7 ± 11.27 | 75 | 80 |
In summary, this clinical trial is designed to detect a difference between randomly assigned treatment groups considered during the time as primary parameter of efficacy CFT from baseline to the end of follow-up, 6 months after treatment. Sample size calculations estimated that enrollment of less than 50 patients would provide 80% power to detect this difference, allowing for a type one error probability of 0.05 (α = 0.05), assuming 10% losses to follow-up.
Statistical analysis was carried out using Fisher's exact test to evaluate other data such as connection between age and treatment, different diagnosis, and treatment, sex- and treatment-related differential diagnosis.
Our study was aimed at assessing the efficacy of 0.5% INDOM single dose eye drops on inflammatory ME complicating different forms of uveitis and one of the common causes of visual impairment.
We included in our study a homogeneous group of eyes affected with CME of immune-mediated or auto-immune origin and excluded, by means of aqueous PCR, any active infection.
The current normal standard of care for ME as a severe uveitic complication includes the use, often in association, of topical and systemic steroids and NSAIDs, in order to take advantage of their not proven, but evidence-based synergistic effect sometimes combined with the anti-inflammatory action of mydriatics.
20,23–25
Nonsteroidal anti-inflammatory drugs are widely used agents, which despite their chemical heterogeneity, share similar therapeutic properties and adverse effects.
31–36
Nonsteroidal anti-inflammatory drugs are cyclo-oxygenase (COX
1-COX
2), pro-inflammatory prostaglandins (PGs), and tromboxanes potent inhibitors.
18,19 In the current ophthalmologic practice, they are mainly used to maintain pupil dilation during intraocular surgery,
31 to give relief to ocular allergic diseases symptoms,
31–33 to reduce discomfort after refractive surgery,
34 to treat postoperative pain and/or inflammation, and to prevent postsurgical CME.
35,36 For decades, their therapeutic efficacy for these conditions has been well established. New ongoing studies are trying to detect their efficacy in the anti-inflammatory treatment of diabetic retinopathy and AMD, which are actually recognized conditions related to the pathogenic role of pro-inflammatory PGs.
35–37 Topical NSAIDs are the lowest risk alternative therapy to steroids; they avoid side effects related to prolonged topical administration of steroids, including glaucoma, cataract, increased infection susceptibility, and delayed wound healing.
38–43 Furthermore, though studies comparing NSAIDs with corticosteroids showed no significant difference between results of these treatments, NSAIDs treatment appears to be more effective than topical steroids in re-establishing the blood–aqueous barrier, as quantitatively measured with anterior ocular fluorophotometry.
39,40 Our experience showed that NSAIDs are an intermediate treatment step to these conditions.
In the last few decades, double-masked, randomized, active, and placebo-controlled studies, including patients undergoing cataract surgery, have reported anti-inflammatory effects from topically applied 0.1% INDOM, 0.03% flurbiprofen, 0.5% ketorolac, and 0.1% diclofenac ophthalmic preparations. The correlations between slit-lamp observations and anterior ocular fluorophotometry or laser cell flare meter methodology are consistent on their efficacy (Rho DS, et al.
IOVS 2006;47:ARVO E-Abstract 5211).
44–48
A number of studies have established the safety and efficacy of 0.1% INDOM in the treatment of postoperative inflammation and pain and in the prevention and treatment of CME,
46–51 but no studies are available on the more concentrated 0.5% INDOM eye drops, which seem to be more effective in relationship with a higher concentration rate in the vitreous and in the retina, though without any improvement in side-effects.
Related to the availability of 0.5% INDOM single-dose eye drops in our country, and following a recent published work of Bucolo and coworkers,
52 we decided to study and compare efficacy and tolerance of this drug versus a placebo formulation (vehicle of the drops) for the treatment of uveitic ME Bucolo evaluated the ocular pharmacokinetics of two different INDOM formulations (0.1% and 0.5%, respectively), in rabbit eyes and the conclusions were that 0.5% INDOM eye drops showed better ocular distribution and relevant drug concentration levels in the posterior pole of the eye, useful to manage retinal inflammatory conditions, when compared with 0.1% concentrations.
Indomethacin eye drops are more soluble and penetrating into eye tissues when they are complexated with cyclodextrins or formulated in ophthalmic suspension and added to artificial tears,
33,37,39 such as hydroxy-propyl-methil-cellulose (which is also present in the formulation we used), thus preserving corneal epithelium integrity.
Some works report that after topical NSAID use, corneal complications like superficial punctate keratitis, corneal infiltrates, epithelial defects (which are also potential secondary risk factors in affected or predisposed dry-eye subjects) and corneal melting, may be detected, but they are mostly in relationship with preserved eye drops.
53,54
Thanks to the characteristics of INDOM (preservative-free solution, tear substitute vehicle of 0.5% INDOM) we have not observed, despite high daily dosage (QID) and long-term treatment, any problem of corneal toxicity or corneal damage or melting as reported sometimes to other drugs (also in the new generation NSAIDs group) (Chu YR, et al.
IOVS 2006;47;E-Abstract 74).
54 –57
We did not take into account the recently introduced NSAIDs bromfenac and nepafenac, because of, until the beginning of our study (November 2011), lack of studies on the retinal concentration of these drugs and their efficacy on the management of uveitic CME, besides the nonavailability of randomized studies and the emerging evidence of potential sight-threatening effects, such as corneal melting in predisposed subjects (Rho DS, et al.
IOVS 2006;47:ARVO E-Abstract 5211).
40–48
In fact, after topical NSAID use, corneal complications are uncommon, but superficial punctate keratitis, corneal infiltrates, epithelial defects (which are also potential secondary risk factors in affected or predisposed dry-eye subjects), and corneal melting have been sometimes reported, mostly in relationship with preserved eye drops.
49–54
Concerning our study, we did not take into account the recently introduced NSAIDs bromfenac and nepafenac, because these formulations were available in the Italian market only some months before the beginning of our study (November 2011) and differently from INDOM eye drops, at this time, we had only some literature indications on their efficacy in the treatment of uveitic CME (Ogawa T, et al.
IOVS 2006;47:ARVO E-Abstract 687).
58 –71
Recent works showed that bromfenac eye drops alone are ineffective for uveitic ME treatment, but may have a synergistic effect with intravitreal steroids in reducing CME in a follow-up period of more than 3 months of treatment.
68
As Cable
69 presented, bromfenac 0.09% QID compared with nepafenac 0.1% three times a day after cataract surgery resulted in positive clinical outcomes of VA, macular volume, and retinal thickness.
In contrast to aqueous drug levels, there is a paucity of human studies measuring NSAID levels in the vitreous after topical application. A single study measured vitreous drug levels in patients who received ketorolac 0.4% four times per day, bromfenac 0.09% twice a day, or nepafenac 0.1% three times a day for 3 days before vitrectomy surgery; only ketorolac resulted in significantly lower vitreous PGE
2 levels compared with placebo.
64
For some authors, steroids have to be used in combination with NSAIDs for CME treatment to obtain a synergistic effect.
40–43 It is important to stress that in the 31 patients included in our study over the full period of follow-up, we never changed ongoing systemic steroid nor immunosuppressive biological treatment.
Moreover, the patients were permitted only twice during the follow-up period, and only in the case of acute relapse of uveitis, to undertake a pulsed period of steroids (0.2% dexamethasone eye drops 3 times a day for 5 days then tapering to 2 weeks), and mydriatics (1% tropicamide eye drops once or twice a day for 10 days) while continuing 0.5% INDOM eye drops.
Only 17 eyes (36.9%) needed this rescue treatment because of anterior acute uveitis relapses mainly related to HLA B27 positivity or autoimmune uveitis.
Primary end points of more recent studies on uveitic CME take into account VA, degree of ME, and a combination of both. New tridimensional Spectralis OCT instruments can help in diagnosis by also showing the hidden forms of uveitis-related ME and provide quantitative and repeated appraisals of CFT.
11–13,72–74
A work by Sivaprasad
49 showed that inner-retinal uveitic CME is more resistant to therapy than other patterns (diffuse or CME), so because of OCT we shall achieve more useful information on its prognosis and on the usefulness of treatment for it.
A study of Markomichelakis and coworkers
75 detected three different OCT patterns of uveitic ME, including diffuse ME, CME, and exudative retinal detachment. Epiretinal membrane coexisted in a significant percentage of patients.
We included in our study OCT-detected diffuse ME and CME and, after 6 months, CFT showed a significant decrease in treated patients by an average of 150 μm (P < 0.0001, 180 days versus baseline) compared with the placebo group in which no significant improvement of ME was seen (P = 1, 180 days versus baseline). In more than 50% of eyes in the placebo group, their OCT-detected ME worsened, as compared with less than 10% in the treated group.
Secondary study endpoints were VA and subjective symptoms' (itching, burning, photophobia, and watering; 0–3 points score) improvement detection.
Visual acuity showed a statistically significant improvement in the treated-eyes group (P < 0.0001, 180 days versus baseline) and significant time-treatment ratio in the treated group only (P < 0.0001); only in Group B (0.5% indomethacin-treated eyes) a significant correlation between retinal thickness decrease and VA increase (R = −0.35022, P = 0.0170) was found after 6 months.
No significant difference in subjective symptoms was observed, no adverse event occurred, except an INDOM-treated group patient who complained about eye burning at instillation during the first treatment month.
The epiretinal membrane was detected by OCT in 17 eyes (36.9%). Seven eyes showed vitreomacular traction (15.2%).
Vision recovery thanks to INDOM treatment was observed in patients suffering from CME, but it was limited in subjects presenting an OCT-detected epiretinal traction membrane.