Description of the condition
The phrase allergic conjunctivitis describes a group of clinically distinct inflammatory conditions of the ocular surfaces including seasonal allergic conjunctivitis ('hayfever'), perennial allergic conjunctivitis, atopic keratoconjunctivitis and vernal keratoconjunctivitis. Atopic keratoconjunctivitis (AKC) and vernal keratoconjunctivitis (VKC) are the most serious forms of allergic conjunctivitis. These conditions are usually easy to distinguish, having characteristic epidemiology (AKC has an older age of onset and is seen almost exclusively in patients with atopic dermatitis) and clinical signs (giant papillae seen in VKC). A key distinction is that VKC will usually become quiescent in the third decade whereas AKC will not.
Vernal keratoconjunctivitis (VKC) is a severe form of allergic eye disease that predominantly affects male children (Bonini 2000; Leonardi 2002). The disease is rare in Europe, with an estimated prevalence of 3.2 per 10,000 (Bremond-Gignac 2008). It is more common in warmer climates such as the Mediterranean, Africa, Asia and the Southern United States (Leonardi 2002; Resnikoff 1988). In central Africa the prevalence of VKC in children has been estimated at 4% (Smedt 2011). Onset of disease is typically in the first decade. VKC is characterised by symptoms of itch, photophobia, reduced vision and mucoid discharge. It is usually seasonal, with acute exacerbations in spring and summer, and relative quiescence between acute attacks, but a significant proportion of patients develop year-round disease activity and hence a more chronic disease course.
VKC primarily affects the conjunctiva, the thin smooth layer of tissue covering the white of the eyeball and the inside of the eyelids. It is characterised by conjunctival inflammation with formation of large or giant papillae. It is usually bilateral, although the disease severity may be asymmetric. VKC is classified according to which area of the conjunctiva is most affected. It may affect the palpebral conjunctiva, which is the lining of the inside of the eyelids, or the limbal conjunctiva, which is the covering of the eyeball near the cornea (the clear window at the front of the eyeball). The pattern may also be a mix of limbal and palpebral VKC.
Severe or protracted activity of limbal VKC may cause peripheral corneal scarring. Severe or protracted activity of palpebral VKC may cause central corneal scarring and significantly reduced vision.
VKC is considered a type of atopic disease and patients often have features typical of allergy such as raised serum IgE (antibodies against allergen), the presence of eosinophils (a subtype of white blood cell which is commonly seen in tissues affected by allergic inflammation) in conjunctival biopsy specimens and a personal or family history of other allergic diseases such as asthma or eczema. However, the pathogenesis is incompletely understood. The cellular mechanisms appear to be more complex than just the typical IgE-mediated Type-1 hypersensitivity reaction seen in seasonal allergic conjunctivitis (hayfever) with increased recognition, in recent years, of a role for T lymphocytes in the process (Bonini 2004; Leonardi 1999; Leonardi 2002; Metz 1997).
The condition tends to improve after puberty and to regress as patients enter their twenties (Bonini 2000; Leonardi 2002). If it is not treated appropriately when it is active, VKC has the potential to cause morbidity ranging from chronic ocular discomfort and photophobia, loss of schooling to permanent visual loss from corneal scarring.
Description of the intervention
The spectrum of disease severity is wide. The factors to consider in deciding the treatment strategy for each patient include the severity of the disease at its most active, the severity of the disease at its least active and the frequency of exacerbations. The therapeutic regimen should be tailored for each patient (Sacchetti 2010). A number of topical treatment are used in VKC to try to control disease activity. These include:
- topical mast cell stabilisers;
- topical antihistamines;
- topical dual-acting agents (drugs with both a mast cell stabilising and an antihistamine effect);
- topical steroid;
- topical immunomodulatory agents (e.g., ciclosporin).
Topical steroid has potential side effects including cataract, glaucoma, and enhanced secondary infection, whereas the other treatments listed above have few serious side effects. Therefore the broad therapeutic aim is to control signs and symptoms of VKC while using as little topical steroid as possible (Sacchetti 2010). However, the control of severe exacerbations may require intensive application of steroid drops which, apart from the risk of side effects, may cause practical difficulties for young children and their parents.
Patients whose VKC is quiescent or minimally active between acute exacerbations are often treated with maintenance therapy using a topical mast cell stabiliser or a dual acting agent. Topical mast cell stabilisers are the most commonly used maintenance therapy. They increase the threshold stimulus for mast cell degranulation in the conjunctiva. Their mechanism of action is not well-understood. Some patients with a more severe or chronic course are treated with low-dose maintenance topical steroid or topical ciclosporin.
How the intervention might work
The symptoms and signs of VKC, like all inflammatory conditions, are caused by release of mediators from resident white blood cells in the tissues or the interaction with the tissue by infiltrating activated white blood cells from the blood, or both. The interventions to be studied in this review all interfere with or inhibit some action of white blood cell function. The types of white blood cell seen in the conjunctiva during active VKC include mast cells, neutrophils, eosinophils, macrophages and monocytes, T lymphocytes and dendritic cells. Some of the interventions are more specific for white blood cell subtypes than others. For example, topical steroids are known to have effects on neutrophils, monocytes, dendritic cells and T lymphocytes whereas mast cell stabilisers specifically inhibit mast cell degranulation in response to allergen (by mechanisms that are, as yet, poorly understood).
Why it is important to do this review
VKC causes significant morbidity, loss of schooling and loss of vision in children. A wide variety of treatments are used to treat exacerbations and prevent recurrences of disease. A systematic review of topical treatments for people with VKC, published in 2007, showed that a variety of available topical treatments were effective in treating acute phases of VKC (Mantelli 2007). This review, which included studies published before 2004, revealed a lack of head to head comparisons between treatments for VKC. The aim of the present review is to assess the most up-to-date available evidence on the use of topical interventions for the treatment of VKC. It is hoped that the results of this review, in conjunction with those of concurrent reviews of systemic treatments and systemic versus topical treatments for VKC respectively, will help to identify areas where there is a particular deficiency of data or which might be rewarding targets for further research.
To assess the benefits and risks of topical interventions in vernal keratoconjunctivitis.
Criteria for considering studies for this review
Types of studies
We will include randomised controlled trials of topical interventions for vernal keratoconjunctivitis.
Types of participants
Patients of any age who have been diagnosed with vernal keratoconjunctivitis by an ophthalmologist. The diagnosis may be clinical, that is without reliance on laboratory supportive investigations. Patients with coexisting ocular morbidity will be excluded.
Types of interventions
We will include all topical pharmacological interventions. These are likely to include:
- topical mast cell stabilisers;
- topical antihistamines;
- topical dual-acting agents;
- topical corticosteroids;
- topical immunomodulatory agents.
Comparisons may be between:
- any topical treatment or combination of topical treatments versus no treatment or placebo;
- any two or more topical treatments or combinations of topical treatments (head to head comparisons).
Types of outcome measures
- Percentage of patients with improvement of ocular surface inflammation based on physician assessment at final assessment at less than or up to four weeks following commencement of treatment
- Percentage of patients with subjective improvement of symptoms of ocular surface inflammation at final assessment at less than or equal to four weeks following commencement of treatment
- Percentage of patients with improvement of ocular surface inflammation based on physician assessment at final assessment at greater than four weeks following commencement of treatment
- Percentage of patients with subjective improvement of symptoms of ocular surface inflammation at final assessment at greater than four weeks following commencement of treatment
- Percentage of patients suffering adverse effects of treatments. Adverse events will be defined by the included studies. Expected adverse events would include ocular hypertension, glaucoma and cataract for topical steroids and hypersensitivity reaction for any topical treatment
- Percentage of patients with recurrence of disease within three months of starting treatment
- Where available, analysis of cost-effectiveness or cost-utility of treatments will be included
Search methods for identification of studies
We will search the Cochrane Central Register of Controlled Trials (CENTRAL) (which contains the Cochrane Eyes and Vision Group Trials Register) (The Cochrane Library), Ovid MEDLINE, Ovid MEDLINE In-Process and Other Non-Indexed Citations, Ovid MEDLINE Daily, Ovid OLDMEDLINE, EMBASE, the metaRegister of Controlled Trials (mRCT) (www.controlled-trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the WHO International Clinical Trials Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We will not use any date or language restrictions in the electronic search for trials.
Searching other resources
We will scrutinise the reference lists of the studies included in the review. We will also use the Science Citation Index to identify reports that have cited the studies included in this review. With both of these strategies we will aim to identify any other relevant reports or trials that have not been identified by the electronic searches. We will not handsearch journals or meeting abstracts.
Data collection and analysis
Selection of studies
Study selection will be carried out by two authors independently (TF and ST) to reduce the possibility that relevant reports will be discarded. Review authors will not be masked with respect to study authors, institution or journal. First, titles and abstracts will be examined and reports that are obviously irrelevant will be excluded. The full-text of potentially relevant reports will be retrieved. Care will be taken to identify multiple reports of the same study and, where identified, these will be linked together. Full-text reports will be examined for compliance with eligibility criteria. Where appropriate, investigators will be contacted to clarify study eligibility. Disagreements about whether a study should be included will be resolved by discussion. Where the disagreement is due to a difference in interpretation that cannot be resolved, a third author (AD) will arbitrate. Studies that are excluded after receiving the full-text will be listed in an excluded studies table and reasons will be given for exclusion.
Data extraction and management
A data collection form based on the checklist in Table 7.3a of the Cochrane Handbook for Systematic Reviews of Interventions will be created. Data will be extracted from the reports by two authors independently (TF and ST). Data will be entered into the RevMan software (RevMan 2012) by one author and the data will be checked for errors or inconsistencies by a second author.
The following data will be extracted.
- Methods (study design, duration etc.).
- Risk of bias.
- Participants (number in each group, age, gender, comparability at baseline etc).
- Outcomes (primary and secondary outcomes collected and reported; for each outcome the unit of measurement will be recorded or, if a scale is used, the upper and lower limits of the scale of measurement and whether a high or low score is good, etc.).
- Results (sample size, missing participants, summary data etc.).
- Miscellaneous (funding, whether correspondence is required, miscellaneous comments by the review authors etc.).
Assessment of risk of bias in included studies
The methodological quality of each study will be assessed independently by two authors (TF and ST) who will use The Cochrane Collaboration’s 'Risk of bias' tool to assess risk of bias as described in Chapter 8 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). This tool focuses on the following five domains of bias: selection, performance, detection, attrition and reporting. It also includes an 'other bias category'. Review authors will not be masked with respect to study authors, institution or journal. Disagreements about risk of bias will be resolved by discussion and, if necessary, in consultation with a third review author.
Random sequence generation (selection bias)
The method used to generate the allocation sequence will be described and assessment will be made by the authors to categorise the risk of selection bias, which is due to inadequate generation of a randomised sequence, as 'Low risk', 'High risk' or 'Unclear risk'. Methods such as computerised randomisation and minimisation will be viewed as adequate. Systematic methods such as assignment according to date of birth or date at presentation would be viewed as inadequate.
Allocation concealment (selection bias)
The method used to conceal the allocation sequence will be described and assessment will be made by the authors to categorise the risk of selection bias, which is due to inadequate concealment of allocations prior to assignment, as 'Low risk', 'High risk' or 'Unclear risk'. Studies that have included central randomisation by a third party or the use of sequentially numbered opaque sealed envelopes will be viewed as adequate. Use of sealed envelopes without mention of methods used to ensure that the allocation process preserved concealment would be viewed as inadequate.
Masking of participants and personnel (performance bias)
The methods used to mask study participants and personnel from knowledge of which intervention a participant received will be described. Assessment will be made by the authors to categorise the risk of performance bias due to knowledge of the allocated interventions by participants and personnel as 'Low risk', 'High risk' or 'Unclear risk'.
Assessment will be made for each primary outcome.
Masking of outcome assessment (detection bias)
The methods used to mask outcome assessors from knowledge of which intervention a participant received will be described. Assessment will be made by the authors to categorise the risk of detection bias due to knowledge of the allocated interventions by outcome assessors as 'Low risk', 'High risk' or “\'Unclear risk'.
Assessment will be made for each primary outcome.
Incomplete outcome data (attrition bias)
The completeness of data for each main outcome will be described in detail and an assessment will be made by the authors to categorise the risk of attrition bias due to amount, nature or handling of incomplete outcome data as 'Low risk', 'High risk' or 'Unclear risk'.
Selective reporting (reporting bias)
If a study protocol is available, the review authors will ascertain whether all of the study’s pre-specified outcomes that are of interest in the review have been reported. If the study protocol is not available, the authors will assess the full-text reports of the study to ascertain whether the published reports include all expected outcomes including those that were pre-specified. An assessment will be made to categorise the risk of reporting bias due to selective reporting of outcome data as 'Low risk', 'High risk' or 'Unclear risk'.
An assessment will be made as to whether the study appears to be free of any other sources of bias.
Measures of treatment effect
Our primary outcomes are dichotomous. We will use the risk ratio with 95% confidence interval (CI). Secondary outcomes are mostly dichotomous also and will be reported similarly. We will report economic outcomes using the mean difference with a 95% CI.
Unit of analysis issues
We will document whether studies have analysed the results at the level of the patient or at the level of the eye. We will document whether studies have included both eyes of each patient or one eye only, and, if the second, whether the method of selecting one eye has been clearly stated. We will document whether or not studies have given a rationale for studying one or both eyes. Our primary outcome is measured at the patient level and thus our meta-analysis will be conducted at this level. Studies that have used a paired design would be included in the review but would not contribute to the meta-analysis of our primary outcome.
Dealing with missing data
For each outcome we will attempt to carry out 'intention-to-treat' analysis. For outcomes where there are missing data, the original investigators will be asked, where possible, to provide the missing data. Where the missing data are not forthcoming, we will carry out an available case analysis. We will document whether the original studies stated that they compared the characteristics of participants with complete data to those without, and whether they provide any information about possible effects of missing data. If the number of patients missing in any study is small, we may impute data in a sensitivity analysis by taking the best case and worse case scenarios, but we would compare results with imputed data to those with available case analyses.
Assessment of heterogeneity
We will examine studies for sources of methodological and clinical heterogeneity. We will then assess statistical heterogeneity using the Chi
Assessment of reporting biases
Outcome reporting bias will be assessed as outlined in the 'Assessment of risk of bias in included studies' section. A contour-enhanced funnel plot will be plotted if 10 or more studies are included in the analysis. The distribution of the plot will be examined for any asymmetry. The distribution of any observed asymmetry will be examined to help decide whether it is likely to be due to publication bias or other effects (such as small study effects).
We will meta-analyse our primary outcomes only. Some studies may report composite scores for signs and symptoms and others may report individual scores for signs and symptoms. We will perform meta-analysis on studies reporting composite scores of signs and symptoms. The results of studies reporting only individual scores for signs and symptoms will be collated but will not be included in a meta-analysis. Data will be collated for secondary outcomes but to avoid any multiplicity issues and for review clarity, we will not meta-analyse these data but will report summary statistics in tables.
We will use a random-effects model unless there are a very small number of studies (less than three), in which case we will use a fixed-effect model. If we detect substantial heterogeneity, either methodological (by review of studies) or by large values of I
Subgroup analysis and investigation of heterogeneity
In the event that the results of the included studies for the proposed outcomes are very different, we will conduct subgroup analyses based on the following.
- Patient age (less than seven years versus seven plus; this age is suggested based on the authors' clinical experience). Instilling drops in younger children can be difficult and therefore compliance with treatment may be poorer in the younger age group.
- Pattern of disease (limbal versus palpebral versus mixed).
- Geographical location of the study or population (categorised by continent).
Sensitivity analysis will be conducted to establish the effect on our results of:
- assumptions made when dealing with missing data;
- excluding studies at high risk of bias;
- the funding source.
The Cochrane Eyes and Vision Group (CEVG) will create and execute the electronic search strategies. We thank Flavio Mantelli, Stefan de Smedt and Kristina Lindsley for their comments on this protocol and Anupa Shah, Managing Editor for CEVG for her assistance throughout the review process.
Richard Wormald (Co-ordinating Editor for CEVG) acknowledges financial support for his CEVG research sessions from the Department of Health through the award made by the National Institute for Health Research to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health.
Appendix 1. CENTRAL search strategy
#1 MeSH descriptor: [Keratoconjunctivitis] explode all trees
#2 MeSH descriptor: [Conjunctivitis, Allergic] explode all trees
#3 vernal near/2 keratoconjunctiv*
#5 #1 or #2 or #3 or #4
#6 MeSH descriptor: [Mast Cells] explode all trees
#7 mast near/2 cell*
#8 MeSH descriptor: [Cromolyn Sodium] explode all trees
#9 MeSH descriptor: [Chromones] explode all trees
#10 MeSH descriptor: [Nedocromil] explode all trees
#11 (c?romo* or nedoc?rom*) near/2 sodium
#12 (c?romo* or nedoc?rom*) near/2 disodium
#14 MeSH descriptor: [Ketotifen] explode all trees
#20 MeSH descriptor: [Histamine H1 Antagonists] explode all trees
#21 antihistamin* or anti-histamin*
#23 bepotastin* or Senju
#27 MeSH descriptor: [Glucocorticoids] explode all trees
#28 corticosteroid* near/3 topical
#34 MeSH descriptor: [Cyclosporine] explode all trees
#36 MeSH descriptor: [Tacrolimus] explode all trees
#37 tacrolimus* or FK506*
#38 #6 or #7 or #8 or #9 or #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21 or #22 or #23 or #24 or #25 or #26 or #27 or #28 or #29 or #30 or #31 or #32 or #33 or #34 or #35 or #36 or #37
#39 #5 and #38
Appendix 2. MEDLINE (OvidSP) search strategy
1. randomized controlled trial.pt.
2. (randomized or randomised).ab,ti.
9. exp animals/
10. exp humans/
11. 9 not (9 and 10)
12. 8 not 11
14. conjunctivitis, allergic/
15. (vernal adj2 keratoconjunctiv$).tw.
18. Mast Cells/
19. (mast adj2 cell$).tw.
20. Cromolyn Sodium/
23. ((c?romo$ or nedoc?rom$) adj2 sodium).tw.
24. ((c?romo$ or nedoc?rom$) adj2 disodium).tw.
32. exp histamine h1 antagonists/
33. (antihistamin$ or anti-histamin$).tw.
35. (bepotastin$ or Senju).tw.
39. exp Glucocorticoids/
40. (corticosteroid$ adj3 topical$).tw.
49. (tacrolimus$ or FK506$).tw.
51. 17 and 50
52. 12 and 51
The search filter for trials at the beginning of the MEDLINE strategy is from the published paper by Glanville et al (Glanville 2006).
Appendix 3. EMBASE (OvidSP) search strategy
1. exp randomized controlled trial/
2. exp randomization/
3. exp double blind procedure/
4. exp single blind procedure/
7. (animal or animal experiment).sh.
9. 7 and 8
10. 7 not 9
11. 6 not 10
12. exp clinical trial/
13. (clin$ adj3 trial$).tw.
14. ((singl$ or doubl$ or trebl$ or tripl$) adj3 (blind$ or mask$)).tw.
15. exp placebo/
18. exp experimental design/
19. exp crossover procedure/
20. exp control group/
21. exp latin square design/
23. 22 not 10
24. 23 not 11
25. exp comparative study/
26. exp evaluation/
27. exp prospective study/
28. (control$ or prospectiv$ or volunteer$).tw.
30. 29 not 10
31. 30 not (11 or 23)
32. 11 or 24 or 31
34. atopic keratoconjunctivitis/
35. allergic conjunctivitis/
36. (vernal adj2 keratoconjunctiv$).tw.
39. mast cell/
40. (mast adj2 cell$).tw.
41. cromoglycate disodium/
42. chromone derivative/
43. nedocromil sodium/
45. ((c?romo$ or nedoc?rom$) adj2 sodium).tw.
46. ((c?romo$ or nedoc?rom$) adj2 disodium).tw.
54. lodoxamide ethyl/
58. exp histamine H1 receptor antagonist/
59. (antihistamin$ or anti-histamin$).tw.
61. (bepotastin$ or Senju).tw.
66. (corticosteroid$ adj3 topical$).tw.
80. (tacrolimus$ or FK506$).tw.
82. 38 and 81
83. 32 and 82
Appendix 4. metaRegister of Controlled Trials search strategy
Appendix 5. ClinicalTrials.gov search strategy
Appendix 6. ICTRP search strategy
Contributions of authors
Review conceived by ST.
Protocol drafted by TF, ST, AD, CB.
Declarations of interest
Sources of support
- NIHR, UK.CB acknowledges financial support from the Department of Health through the award made by the National Institute for Health Research to Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology for a Specialist Biomedical Research Centre for Ophthalmology. The views expressed in this publication are those of the authors and not necessarily those of the Department of Health.
- No sources of support supplied