Study reference

Study characteristics

Patient characteristics

Intervention (I)

Control (C)

Follow-up

Outcome measures and effect size

Comments

Hedayatfar, 2014

Type of study: RCT

Country: Iran

Competing interests: none declared

Inclusion criteria:

unilateral acute anterior uveitis (AAU), at least 2+ anterior chamber cells based on the Standardization of Uveitis Nomenclature (SUN) Working Group Classification, age 16 years or older, attack duration less than a 1-week period, intraocular pressure (IOP) ≤ 21 mm Hg at presentation, and negative evidence for infectious (eg, corneal scars, iris atrophy) or granulomatous (eg, granulomatous keratic precipitate

or iris nodule) inflammation

Exclusion criteria:

eyes with hypopyon or florid fibrin formation,

previous unsuccessful topical treatment, known cases of Behçet disease, cases with evidence for other causes of their uveitis (eg, positive skin tuberculin test), any slit-lamp evidence of significant blepharitis or meibomian gland dysfunction, concomitant topical eye drops for chronic eye disease (eg, glaucoma medication, artificial tears), and the recent use of contact lenses

Group A: N=40 eyes

Methylprednisolone group. Nonpreserved methylprednisolone sodium

succinate 1%.

Group B: N=38 eyes

Prednisolone group. Preserved prednisolone acetate suspension 1% (Pred Forte 1%, Allergan, Inc), containing chloride (BAC) 0.004%.

Treatment regimen:

Patients were dispensed bottles containing 10 mL of the steroid solu-

tion and were asked to keep these refrigerated and to discard these after 4 weeks. Patients were instructed to instil 1 drop of study medications hourly in the affected eyes for the first 72 hours and then every 2 hours for the rest of the week. During sleep time, betamethasone 0.1% topical ointment

was applied instead of eye drops.

Homatropine 2% eye drops were prescribed for all the patients every 8 hours.

At the first-week follow-up visit, the patients who did not improve clinically were given an additional regional steroid injection or oral

prednisolone.

Patients with a good response in the first week of treatment were encouraged to continue the eye drops in a tapering regimen according to clinical response.

Length of follow-up:

2 weeks after start treatment

Lost to follow-up:

Intervention: 2/40=5% eyes (due to protocol violation)

Control: 4/38=11% eyes (n=2 due to procotol violation, n=2 lost to follow-up)

Outcome measures:

Inflammation measured as anterior chamber cells and graded according to SUN working group classification. Successful treatment defined as anterior chamber cell grade ≤ 1.

Results:

Anterior chamber cells

Group A: 29/38 eyes (76.3%) successfully treated at day 14

Group B: 24/34 eyes (70.6%) successfully treated at day 14

Difference between groups not statistically significant (p=0.582) (non-inferiority)

Ocular discomfort

At the second-week follow-up visit, preservative-free methylprednisolone-treated

patients reported a significantly lower subjective discomfort compared with that of group B. (0.55 vs.1.43, P = 0.01).

IOP:

The difference in the IOP was not

significant between groups A and B at the follow-up visits

Randomisation by using balanced blocked randomization method.

Investigators were blinded to allocation.

Participants were not blinded to allocation because of ethical and feasibility constraints.

Data of patients lost to follow-up were not analysed (no intention to treat analysis)

Restricted inclusion/exclusion criteria: study only included patients with pure anterior uveitis (idiopathic and HLA B 27 related). Patients with severe inflammation (including patients with hypopyon or fibrin formation) and those who did not respond to topical treatment in previous episodes were excluded.

Small sample size (imprecision)

Co-medication (homatropine) (indirectness of effect)

GRADE: low quality of evidence due to lack of blinding of patients (not possible), no intention to treat analysis, imprecision, and indirectness.

Dunne, 1979

Type of study: RCT

Country: UK

Competing interests: not reported

Inclusion criteria:

cases of acute

unilateral nongranulomatous anterior uveitis presenting to the Casualty Department of St. Paul's Eye Hospital, Liverpool.

Exclusion criteria:

not reported

Intervention/control:

Clobetasone butyrate (N=16) vs. betamethasone phosphate: (N=18) vs. placebo (N=14)

Treatment regimen:

Patients were asked to instil the drops daily into the relevant eye using 2 drops 2-hourly each day for a period of 3 weeks. In addition, atropine 1% drops were instilled twice daily and, when necessary, acetazolamide for secondary glaucoma.

Length of follow-up:

21 days

Lost to follow-up:

12/60=20%

Four patients were withdrawn because of worsening of

the condition and they were later found to have

been on placebo eye drops. One patient was given

another medication in error after starting the trial, 1 patient was withdrawn after 7 days due to development of bilateral anterior uveitis, and 6 patients defaulted

Intention to treat analysis:

No

Outcome measures:

Change in signs and symptoms score.

Signs: ciliary injection, oedema, keratic precipitate, flare, cells in aqueous and in vitreous, exudate, posterior synechiae, ptosis, hypopyon.

Symptoms: blurring, redness, watering, photophobia, pain.

Clinically relevant difference in score not defined.

Results:

Signs:

Betamethasone better than placebo, statistically significant after 21 days treatment.

Clobetasone butyrate was not significantly better than placebo.

Symptoms:

Betamethasone phosphate and clobetasone butyrate drops were significantly better than placebo after 3 days' treatment. There was a difference at 7 days, though this just failed to reach a statistical significance, probably owing to the small

numbers of the placebo group. The difference between steroid and placebo groups did not reach a statistical difference after 14 days and 21 days.

IOP:

clobetasone butyrate had significantly

less effect on raising intraocular pressure in known steroid responders and ocular hypertensives than did dexamethasone

Randomisation method: not reported.

Blinding of participants, personnel, outcome assessors: not reported

Data of patients lost to follow-up were not analysed (no intention to treat analysis)

Small sample size per treatment group (imprecision)

Co-medication (atropine) (indirectness of effect)

GRADE: very low quality of evidence due to not reporting of randomisation and blinding method, no intention to treat analysis, imprecision, and indirectness.

Dunne, 1985

Type of study: RCT

Country: UK

Competing interests:

Smith and Nephew Ltd provided tolmetin eyedrops

Inclusion criteria:

consecutive

patients presenting at the Casualty Department of

the Manchester Royal Eye Hospital with acute endogenous non-granulomatous anterior uveitis.

Exclusion criteria:

Patients on topical or systemic

steroids for other conditions and pregnant women

Intervention/control:

Tolmetin sodium dihydrate 5% solution (N= 21) vs. prednisolone

disodium phosphate 0.5% (Predsol) (N=19) vs. betamethasone

disodium phosphate 0.1% (Betnesol)(N=20).

Treatment regimen:

Patients were asked to instil one drop into the affected eye every two hours during the waking period of each day, and this treatment regimen was subsequently reduced by the examining ophthalmologist as

required. Initially all patients also received atropine 1.0% once daily.

Length of follow-up:

21 days

Lost to follow-up:

11/71=15%

Four of the excluded patients did

not return to the clinic for follow-up visits at an early

stage in the trial. Another four did not reattend at all. One patient in the Predsol group was inadvertently

changed to treatment with Betnesol, and these

results were excluded. Two patients in the Betnesol

group developed bilateral uveitis and were removed

from the study.

Intention to treat analysis:

During the course of the trial a number of patients were removed from the study because of lack of response to treatment. Where this occurred the last available signs and symptoms scores

have been carried over into missing assessment periods until the end of the trial (21 days).

Outcome measures:

Change in signs and symptoms score.

Signs: ciliary injection, oedema, keratic precipitate, flare, cells in aqueous and in vitreous, exudate, posterior synechiae, ptosis, hypopyon, mobility of aqueous cells.

Symptoms: blurring, watering, photophobia, pain.

Clinically relevant difference in score not defined.

Results

Symptom scores.

Between-group comparisons

showed no statistically significant difference in either total symptom scores at 1, 3, 7, 14, and 21 days or change in total symptom scores at days 3, 7, 14, and 21.

Sign scores:

Comparisons between treatment groups for either total sign scores at days 1, 3, 7, 14, and 21 or change from baseline total sign scores at days 3, 7, 14, and 21 showed no statistically significant differences between the three groups. Nevertheless the trend

shows that the greatest and most rapid response was seen in the Betnesol treated group. Predsol and tolmetin gave similar results.

% treatment success (clinically judged):

Tolmetin: 57%

Predsol: 68%

Betnesol: 90%

Side effects (complaints of stinging or irritation):

Tolmetin: 30%

Predsol: 16%

Betnesol: 0%

IOP:

IOP was often lower in the affected eye at the initial visit. In 17% of patients

pressures were the same and in 25% higher than in the non-inflamed eye. The distribution was similar in all three groups. In 40-50% of the patients the pressure rose slightly during the trial and remained

the same or decreased in the remainder.

Randomisation method: randomised treatment code.

Blinding of participants, personnel: yes, Neither patient

nor ophthalmologist was advised of the nature of the

treatment.

Blinding of outcome assessors: not reported

Data of 11 patients lost to follow-up were not analysed, but data of patients removed from trial due to lack of response to treatment were included (last available scores)

Small sample size per treatment group (imprecision)

Co-medication (atropine) (indirectness of effect)

GRADE: low quality of evidence due to not reporting of blinding of outcome assessor, imprecision, and indirectness.

Foster, 1996

Type of study: RCT (two multicenter studies)

Country: USA

Competing interests:

individual interests not stated. Investigation was supported by a grant from Alcon Laboratories.

Inclusion criteria:

participation by males and

nonpregnant or nonlactating females with an ocular

diagnosis of uveitis (acute uveitis, recurrent iridocyclitis, or chronic uveitis) for whom a topical corticosteroid was indicated.

Exclusion criteria:

pregnancy, lactation, or the use of inadequate birth control methods; patients for whom a topical

corticosteroid was contraindicated; contact lens wear during the course of the study; uveitis secondary to a local or systemic infectious disease; concomitant

medications that could interfere with therapeutic

response; corticosterod use (topical or systemic) within 24 hours before study entry; partial or total cataract interfering with the fundus examination;

hypersensitivity to any component of the study medications; known alcohol or drug abuse; or use of any investigational drug within 30 days before receipt of

the study medication.

Intervention/control:

rimexolone 1% ophthalmic suspension (Vexol 1%, Alcon) vs. 1% prednisolone acetate (Pred Forte Allergan)

study 1 (N=181):

rimexolone (N=86) vs. prednisolone (N=95)

study 2 (N=92):

rimexolone (N=46) vs. prednisolone (N=46)

Treatment regimen:

All patients were instructed to instill their assigned study medication in the affected eye or eyes as follows:

one to two drops every hour while awake during the first week, every two hours while awake during the

second week, four times a day during the third week, two times a day for four days, and once a day for the

last three days.

Concomitant therapies permitted during the study were the use of topical ocular medication required to control an increase in IOP or photophobia during the study. The prescription of cyclopentolate hydrochloride to prevent synechiae formation was also permitted.

Length of follow-up:

28 days

Lost to follow-up:

study 1:

21/181=12%

use of prohibited

concomitant medications (N=6), dosing violations (overdose) (N=4), systemic corticosteroid use prior to study (N=11).

study 2:

7/92=7.6%

use of prohibited

concomitant medications (N=2), dosing violations (N=4), lens opacity (N=1).

Intention to treat analysis:

No

Outcome measures:

Change in anterior chamber cell and flare.

Clinically significant difference in cell score (0-5) was a difference between treatments of more than 20% of scale range, or one score unit.

Intraocular pressure (IOP). Clinically significant increase was defined as ≥ 10 mmHg raise.

Results

Anterior chamber cell score:

Rimexolone and prednisolone reduced

cells in the anterior chamber equally effectively in each study (P > 0.05) . Both treatments reduced the number of cells in the anterior chamber by more than one score unit from baseline.

Flare score:

In Study 1, rimexolone was as effective as prednisolone in decreasing flare at all visits except Day 28, when there was a statistically, but not clinically, significant difference favouring prednisolone (P = 0.04)·

No statistically or clinically significant differences were noted between treatments in Study Two.

IOP:
study 1:

 6/88 (6.8%) rimexolone patients vs. 11/93 (12%) prednisolone patients had ≥ 10 mmHg increase

study 2:

1/46 (2.2%) rimexolone patients vs. 8/46 (17%) prednisolone patients had ≥ 10 mmHg increase

Randomisation method: not reported

Blinding of participants and personnel: yes. The study medications were supplied in identical 5-ml masked containers labeled with the patient's identification number.

Blinding of outcome assessors: not reported.

No intention to treat analysis

concomitant therapies permitted (imprecision of effect)

GRADE: low quality of evidence due to not reporting of randomisation method and blinding of outcome assessor, no intention to treat analysis, and indirectness of effect

Loteprednol Etabonate US Uveitis Study Group, 1999

Type of study: RCT (2 parallel, randomized, double-masked, active-controlled comparisons conducted at ac-

ademic or private practice clinics and conducted in sequence)

Country: USA

Competing interests: Supported by Pharmos Corporation and Bausch and Lomb Pharma-

ceuticals, Inc.

At the time that this study was conducted, Dr Howes and Ms Coultas were employees of Pharmos Corporation, the developers of loteprednol

etabonate. Dr Strahlman is an employee of Bausch and Lomb Pharma-

ceuticals, the United States distributor of loteprednol etabonate. All

three have had stock relationships within the past three years. Drs Crockett and Novack serve as consultants to Pharmos Corporation.

Inclusion criteria:

Patients aged 18 to 75 years with a clinical diagnosis of acute anterior uveitis. A key requirement for study entry was a cell score of greater than 1.5. Hypopyon was not allowed at entry.

Also required for entry was ocular pain (≥ 1 on a scale of 0 to 3), anterior chamber flare (≥ 1 on a scale of 0 to 4), photophobia (≥ 1 on a scale of 0 to 3), and presence of ciliary flush. If the patient had previous episodes of acute anterior uveitis, the most recent attack was to have been

more than 6 weeks before enrollment.

Exclusion criteria:

Patients with known sensitivity to the study medications, uncontrolled glaucoma or ocular hypertension (intraocular pressure ≥ 25 mm Hg), hyphema, patients in whom uveitis was clearly secondary to systemic infectious or inflamma-

tory disease, posterior uveitis, pars planitis, choroiditis, significant macular edema, or history of

severe / serious ocular pathology or other medical condition that could result in the patient’s ability to complete the study safely. If a patient was using a prescription medication before study entry, the dosing was not to have

changed within the previous month and was to remain stable throughout the study. Pregnant or lactating women were excluded from the study, and women of childbearing

potential were required to be using adequate birth control and test negative to a blood or urine pregnancy test.

Intervention/control:

Study 1:

loteprednol etabonate ophthalmic suspension 0.5% (N=36) vs. prednisolone acetate ophthalmic suspension 1.0% (N=34)

Treatment regimen:

days 0 to 7, 8 times per day; days 8 to 14, 6 times per day; days 15 to 21,

4 times per day; after day 21 the rate of dosing was tapered depending on the severity of the uveitis. If the

anterior chamber cell rating was 0, treatment was to be concluded by tapering the dosing regimen from 4 times per day to none over a 14-day period. The on-treatment phase was from baseline through the visit at which the cell rating reached 0, or week 7 occurred, or the patient

discontinued the treatment prematurely. Although patients with bilateral disease were allowed entry and bilateral treatment was employed, only one eye qualified for all

study patients.

Except for the assigned masked test medications, cycloplegic agents to control posterior synechiae, and ocular hypertensives, no other concomitant ocular therapy was to

be used during the study.

Concomitant systemic medications used occasionally (such as over-the-counter pain relievers) were allowed at the discretion of the investigator.

Intervention/control

Study 2:

loteprednol etabonate ophthalmic suspension 0.5% (N=84) vs. prednisolone acetate ophthalmic suspension 1.0% (N=91)

Because the reduced sample size considered in study 1 was less than that typically required for adequate power to detect meaningful differences, a second larger study was

conducted. It was similar in design to study 1, with the addition of an enhanced dosing regimen more aligned with standard clinical practice and a shorter treatment duration (28 days). The minimum anterior chamber cell score required for entry was 1.0.

Treatment regimen:

days 0 to 7, every hour, up to 16 times per day; days 8 to 14, every 2 hours, up to 8 times per day; days 15

to 21, 4 times per day; days 22 to 25, twice daily; and days 26 to 28, once daily (morning).

Length of follow-up:

Study 1:

Up to 35 days

Study 2:

Up to 28 days

Lost to follow-up:

Study 1:

4/70=6%

2 patients, one in each group, were observed only at baseline. 2 patients, one in each group, had already enrolled for their fellow eye.

Study 2:

5/175=3%

Failed to return for any on-treatment observations (3 in intervention and 2 in control group)

Intention to treat analysis:

modified intent-to-treat population that included only patients with at least one on-treatment

evaluation.

Study 1: N=66

Study 2: N=170

Outcome measures:

- Distribution of time for anterior chamber cell count to return to a rating of 0.

- Primary efficacy:

achievement of 0 cell score by

day 35 (study 1) or by day 28 (study 2).

- Secondary efficacy:

0 flare score by day 35 or day 28 as well as resolution of pain and photophobia.

Results

Median time to score 0:

Study 1: days 11-17 for both groups (p=0.290)

Study 2: days 6-10 for both groups (p=0.105)

% patients achieving 0 score:

Study 1: 74% loteprednol vs. 88% prednisolone (p=0.194)

Study 2: 72% loteprednol vs. 87% prednisolone (p=0.015)

% patients achieving 0 flare:

Study 1: 71% loteprednol vs. 81% prednisolone (p=0.330)

Study 2: 66% loteprednol vs. 82% prednisolone (p=0.017)

Pain or photophobia:

No statistically significant differences between treatment groups in pain of photophobia in both studies.

Adverse events:

In study 2, two patients in the loteprednol group (cystoid macular edema, decreased visual acuity and various ocular symptoms, and two patients in the prednisolone group (interstitial keratitis and increase in age-related macular degeneration).

Intraocular pressure:

An increase in IOP of > 10 mmHg was observed more frequently in patients receiving prednisolone acetate (N=7) than those receiving loteprednol etabonate (N=1).

Randomisation method: computer generated randomisation. Medications were supplied in opaque plastic containers

with opaque dropper tips, and their identity was not

disclosed.

Blinding of participants, personnel: likely, double-masked

Blinding of outcome assessors: not reported

Publication bias: Study 1 consisted of two almost identical protocols, one carried out in the United States and the other in the United Kingdom. Because of differences in the standard of

uveitis care between the two countries and because the

second trial was later conducted at US sites only, the data

of the UK sites in the first trial have been omitted from this

article.

Study and authors supported by Pharmos.

Co-medication (cycloplegics) (indirectness of effect)

GRADE: low quality of evidence due to not reporting outcome assessor blinding, possible publication bias, and indirectness.

Sand, 1991

Type of study: RCT

Country: Denmark

Competing interests:

Dumex Copenhagen supplied test solution

Inclusion criteria:

age 18-80, mild to moderate acute anterior uveitis (AAU), no attack of AAU 3 months before inclusion,

no other eye disease, no severe systemic disease, no other anti-inflammatory treatment

Exclusion criteria:

concurrent intraocular affection apart from incipient cataract and senile macula degeneration

Intervention/control:

1% indometacin in ricinus oil (N=25) vs. 0.1% dexametason in water with addition of hydrocypropylmelthylcellulose and benzalkonium chloride (N=24)

Treatment regimen:

The drops were administered six times daily. Treatment was in most cases (18/24 indomethacin and 17/24 dexametason) supported by administration of a cycloplegic three times a day.

Patients were withdrawn from study in case of aggravation or lack of improvement at day 7.

Length of follow-up:

14 days

Lost to follow-up:

3/25=12% in indometacin and 1/24=4.2% in dexametason group.

Patients were withdrawn from study due to aggravation or lack of improvement at day 7.

Intention to treat analysis:

No

Outcome measures:

Inflammation: scoring ciliary injection, aqueous flare and iris hyperemia each on a 0-3 scale.

Symptoms: patients were asked for subjective symptoms (graded 0-3)

Cure rate: defined as no clinical signs nor symptoms of inflammation)

IOP

Results

Inflammation:

No statistically significant difference in inflammatory score at day 14 between the groups.

Symptoms:

No statistically significant difference in symptom score at day 14 between the groups.

Cure rate:

Statistically significant higher cure rate in dexametason group at day 7, but this difference was not statistically significant at day 14.

IOP:

No statistically significant difference in IOP between groups. IOP ranged from 9-22 mmHg.

Randomisation method: not reported.

Blinding of participants, personnel: not reported.

Blinding of outcome assessors: Yes

No intention to treat analysis.

Small sample size per treatment group (imprecision)

GRADE: low quality of evidence due to not reporting of randomisation method, blinding of participants and personnel, and imprecision.

Young, 1982

Type of study: RCT

Country: UK

Competing interests:

Smith and Nephew Ltd provided tolmetin (McNeil

Pharmaceuticals), D. Gilbert, and K. M. Shawley, both of Smith

and Nephew Research Ltd, provided personal assistance.

Inclusion criteria:

patients who presented consecutively in the Casualty Department of St Paul's Eye Hospital, Liverpool,

during the period September 1980 to February 1981

with unilateral acute endogenous non-granulomatous anterior uveitis

Exclusion criteria:

Pregnant women, patients under the age of 16,

patients with one eye, and those on systemic or topical

steroids.

Intervention/control:

Unpreserved tolmetin 5% (N=32) vs. prednisolone (Predsol) 0.5% (N=32) vs. preserved sterile saline 0.9% (N=30)

Treatment regimen:

Patients were instructed to instil the trial drops every 2 hours during

the waking period and also atropine 1% drops once a day.

Length of follow-up:

21 days

Lost to follow-up:

6/100=6%

Six patients were excluded from analysis, 5 failed to reattend, and one developed bilateral disease. Not reported to which treatment groups these 6 were allocated.

Intention to treat analysis:

No

Outcome measures:

Symptoms and signs on a 0-3 severity scale. Symptoms and signs and scale not defined/reported. Clinically relevant improvement not defined.

Results

Treatment altered:

8 prednisolone, 10 tolmetin, and 12 saline patients were judged to have failed to respond within a 15-day period and had

their treatment altered.

% treatment failures:

Tolmetin: 53%

Prednisolone: 31%

Saline: 47%

% cure rate:

Tolmetin: 47%

Prednisolone: 69%

Saline: 53%

The difference between the groups was

not statistically significant.

Improvement in symptoms and signs scores:

All the treatment groups showed statistically significant improvement in total symptom scores at 3, 7, 14, and 21 days.

The prednisolone group showed statistically

significant improvement in total clinical sign

scores at 3, 7, 14, and 21 days, while the tolmetin group did not show a significant reduction in total sign scores until the 14th day, and the saline group showed a significant reduction only at 21 days.

Randomisation method: not reported.

Blinding of participants, personnel: likely, double-masked

Blinding of outcome assessors: not reported

Data of patients lost to follow-up were not analysed (no intention to treat analysis)

Small sample size per treatment group (imprecision)

Co-medication (atropine) (indirectness of effect)

GRADE: very low quality of evidence due to not reporting of randomisation and blinding of outcome assessor, no intention to treat analysis, imprecision, and indirectness.

Mercieca, 2010

Type of study: Cross-sectional survey (retrospective study)

Country: UK

Competing interests: none declared

Inclusion criteria:

patients with chronic

uveitis affecting the anterior segment, who had used topical steroids continuously for 2 years or more

Exclusion criteria:

not stated

Intervention: N=248

Use of long-term topical steroids

Comparison:

A comparison group of 113 patients with chronic glaucoma (but not uveitis) using long-term (more than 2 years) prostaglandin analogue (72), beta-blocker (19) or both (22) was

also analysed prospectively.

Length of follow-up:

not reported (more than 2 years on treatment)

Primary outcome measures:

Incidence of sub-conjunctival haemorrhage (SCH)

Results:

SCH:

24/248 (9.7%) experienced single or recurrent SCH. Of these, 17 (71%) used prednisolone acetate 1%, 4 (17%) used rimexolone 1%, and 3 (12%) used dexamethasone 0.1%.

None of comparison group reported SCH over the previous 2 years

No statistical analysis to adjust for potential confounders (e.g. comedication, duration and dose of steroids, differences in duration and type of uveitis)

SCH occurs more frequently in patients with long-term topical steroid use than in patients using long-term topical antiglaucoma medication. The most commonly associated steroid (71%) is prednisolone acetate 1%, but this eye drop was also more frequently used in the study population, so the result does not imply, that this steroid in particular is more likely to lead to SCH.