Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
www.cochranelibrary.com
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol)
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
TABLE OF CONTENTS
HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) i
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
[Diagnostic Test Accuracy Protocol]
Vijay Tailor1 , Sara Balduzzi2 , Sarah Hull1 , Jugnoo Rahi3 , Christine Schmucker4 , Gianni Virgili5 , Annegret Dahlmann-Noor1
1 NIHR Biomedical Research Centre at Moorfields Eye Hospital and UCL Institute of Ophthalmology, London, UK. 2 Italian Cochrane
Centre, Department of Diagnostic, Clinical and Public Health Medicine, University of Modena and Reggio Emilia, Modena, Italy.
3 Department of Epidemiology, Institute of Child Health, London, UK. 4 German Cochrane Centre, Institute of Medical Biometry and
Medical Informatics, University Medical Center Freiburg, Freiburg, Germany. 5 Department of Translational Surgery and Medicine,
Eye Clinic, University of Florence, Florence, Italy
Contact address: Annegret Dahlmann-Noor, NIHR Biomedical Research Centre at Moorfields Eye Hospital and UCL Institute of
Ophthalmology, 162 City Road, London, EC1V 2PD, UK. annegret.dahlmann-noor@moorfields.nhs.uk.
Citation: Tailor V, Balduzzi S, Hull S, Rahi J, Schmucker C, Virgili G, Dahlmann-Noor A. Tests for detecting strabismus in
children age 1 to 6 years in the community. Cochrane Database of Systematic Reviews 2014, Issue 7. Art. No.: CD011221. DOI:
10.1002/14651858.CD011221.
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ABSTRACT
This is the protocol for a review and there is no abstract. The objectives are as follows:
To assess and compare the accuracy of tests, alone or in combination, for screening for strabismus in children aged one to six years, in
a community setting by lay screeners or primary care professionals.
Other objectives will be to investigate sources of heterogeneity of diagnostic accuracy, including:
• age;
• setting;
• type of professionals performing the test;
• study design;
• study size (< 100 vs. ≥ 100 participants, which may reflect the adoption of different sampling strategies);
• variation in the way a test is carried out;
• type of strabismus (convergent vs. divergent, horizontal/vertical);
• severity of strabismus (amount of misalignment, constant/intermittent/latent);
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 1
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
BACKGROUND mature at five years or 12 years of age, respectively (Simons 1981a;
Walraven 1993). Normal adult stereopsis is 50 to 60 seconds of arc;
some childhood vision screening program have used a threshold of
Target condition being diagnosed 400 seconds of arc for “suspicion of amblyopia” (Traboulsi 2008).
Reduced stereopsis adversely affects motor skills, particularly fine
Strabismus is a condition in which the eyes are not aligned. It is as- motor skills (Grant 2007; Hrisos 2006; O’Connor 2010; Webber
sociated with deficient binocularity, the mechanism that integrates 2008).
visual information from both eyes. Strabismus can be primary, or Significant misalignment can affect development, social interac-
it can be a consequence of poor vision in one eye or of refractive tions, and emotional well-being. In children with infantile es-
errors. Less commonly, strabismus can be caused by lesions affect- otropia, surgical correction of strabismus leads to improvement
ing the oculomotor, trochlear or abducens nerve, or higher neu- in general development as measured by the Bayley scale (Rogers
rological pathways. Rarely, strabismus is caused by developmental 1982). Scores on anxiety and depression scales such as the National
or traumatic defects of the extraocular muscles. Eye Institute Visual Functioning Questionnaire and the Hospital
Prevalence figures for strabismus vary. A screening study in Balti- Anxiety and Depression Scale are significantly different from non-
more, USA, found a prevalence of manifest strabismus of 3.3% in strabismus children, and improve following surgical strabismus
Caucasian and of 2.1% in African American children aged 6 to 71 correction (Bernfeld 1982; Chai 2009). Children with strabismus
months (Friedman 2009). Other population-based studies have may have significantly greater conduct and externalising problems
reported a prevalence of childhood strabismus between 0.01% (Koklanis 2006).
and 3.1%, indicating that prevalence may vary greatly between Strabismus can also be an indicator of severe eye and health prob-
different ethnic groups (Graham 1974; Matsuo 2007a; Matsuo lems. As it can indicate poor vision, it may in rare cases be the first
2007b; Preslan 1996; Traboulsi 2008; Turacli 1995; Wedner 2000) sign of childhood cataract, glaucoma, or tumours of the eye, op-
(Appendix 1). tic nerve, orbit or brain, such as retinoblastoma, glioma, or rhab-
domyosarcoma.
Relevance of strabismus in children
Gross misalignment of the eyes is usually noticed by members of
There are many subtypes of strabismus. In the context of childhood
the family or carers. Small angles of deviation are not necessarily
vision screening programs, the most relevant distinction is between
apparent. In young children, features such as a broad nasal bridge
manifest and latent strabismus. Manifest strabismus is a risk factor
or certain lid positions and shape (epicanthus) can give rise to
for the development of amblyopia, the commonest vision disorder
pseudostrabismus, i.e. a perception of strabismus when in fact the
in children (prevalence 1.6 to 3.6% in Western societies) (Simons
eyes are straight.
1996a).
Amblyopia is a developmental anomaly of spatial vision, usually as-
sociated with strabismus, anisometropia or form deprivation early Diagnosis of strabismus: the cover test
in life (Ciuffreda 1991). Amblyopes have reduced visual acuity in The cover test is based on the observation of a refixation movement
one or both eyes, reduced contrast sensitivity and reduced contour of a deviated eye when the fixing eye is covered (Gamble 1950;
integration. Clinical definitions of amblyopia are based on visual McKean 1976; Romano 1971; Scott 1973). The basic form of
acuity only, taking into consideration the age of the child and pro- the cover test, the cover-uncover test, establishes the diagnosis of
gressive improvement of “normal visual acuity” in the early years. manifest strabismus. An occluder is introduced in front of one eye,
Unilateral amblyopia is often defined as an interocular difference then removed again, re-establishing binocular viewing. If an eye
in best-corrected visual acuity (of two logMAR or Snellen chart moves when the other is covered, this indicates that this eye was
lines) (Friedman 2009), or best-corrected visual acuity of 0.30 log- not fixing before the cover was introduced. Any eye movement is
MAR or worse in either eye (Rahi 2002; Traboulsi 2008). In three- interpreted as “test positive” and “manifest strabismus present”; the
year old children, bilateral amblyopia is suspected if best-corrected magnitude of the movement is often categorised as small, moderate
visual acuity is worse than 0.40 logMAR in one eye and worse or large. This test is used in some screening programs to detect
than 0.3 logMAR in the other eye in the presence of a bilateral strabismus (Fogt 2000). The accuracy of the cover test in detecting
amblyogenic risk factor. In four-year old children, the thresholds amblyopia may be affected by the child’s age at screening, with
are 0.3 and 0.18 logMAR, respectively (Schmidt 2004). better test performance in children over the age of three years
Strabismus has a profound effect on stereopsis or perception of (Williams 2001).
depth. Stereopsis normally develops within the first three to four
months of age and reaches adult levels by to the age of 24 to 36 Variations of the cover test are used to diagnose latent strabismus,
months (Braddick 1980; Fawcett 2005; Fox 1980; Petrig 1981; and to measure the magnitude of both manifest and latent stra-
Takai 2005). Two studies reported that stereoacuity continues to bismus. The presence of latent strabismus is assessed by using the
develop beyond the age of three years, and may not yet be fully alternate cover test. The occluder is moved from eye to eye, allow-
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 2
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
ing viewing of the target with one eye only, without permitting crowded Kay picture test, displayed as paper- or computer-based
binocular viewing. The observer notes refixation movements of charts or conventional retroilluminated charts.
either eye as the cover is removed. • Type 4 tests are those automated refraction devices which
Quantitative measurements are obtained by neutralising the stra- are also designed to report ocular misalignment.
bismus with prisms held in front of one eye whilst performing
We will include studies that report combinations of several index
the cover/uncover test (simultaneous prism and cover test) or the
tests.
alternate cover test (prism alternate cover test); the endpoint of
Other tests for strabismus, such as controlled binocular acuity test,
measurements is the prism with which no refixation movement
suppression tests, blur test, and tests designed to detect reduced
is observed when the cover is removed. To trained professionals
fusional reserve (prism reflex test, prism fusion range) are not used
(orthoptists) slow refixation on cover test is a potential indicator of
by lay screeners, but only by trained professionals (orthoptists).
strabismus; however, this has not been used in published screening
Orthoptist-delivered screening is not within the remit of this re-
studies.
view.
All cover tests are carried out with the participant fixing on a target
presented at distances of six, four or three metres, and then at near
(33 or 40 cm). In children, the distance target is often presented at Principles underlying each type of index test
three metres. In very young children the test is often only carried Detailed information about each index test is given in Appendix
out at near fixation. 2.
The cover/uncover test aims to detect strabismus, but not refrac-
tive errors, the other significant group of amblyogenic risk fac-
tors. Its accuracy as a standalone amblyopia screening test is there- Type 1: Tests which directly identify ocular misalignment
fore limited (Schmidt 2004). Conversely, addition of the cover/ In manifest strabismus with childhood onset, information from
uncover test to vision screening tests increases the detection rate the deviating eye is suppressed, so that a person does not perceive
of strabismus (VIP 2007). Vision screening programs for children double vision. The principle of the cover test is that when the
between 4 and 6 years traditionally use optotype testing to deter- fixating eye is covered, the deviating eye will move to primary
mine visual acuity (matching or naming letters or pictures), with position (looking straight ahead position) to take up fixation, as
or without a cover test to detect strabismus. In an effort to screen long as it has some vision and does not have eccentric fixation or
younger children to identify and treat problems early, these “man- severe eye movement deficit. Presence, speed and magnitude of
ual” screening programs are increasingly supplemented or even this refixation movement are the outcomes of the cover test.
replaced by the use of devices such as photorefractors, which also
aim to provide information about refractive amblyogenic risk fac- Type 2: Tests of binocular function - stereopsis
tors. The American Association of Pediatric Ophthalmology and
The visual axes need to be within a certain angle to each other in
Strabismus recently published updated recommendations for au-
order to detect information that is presented in stereotests. Stra-
tomated screening programs (Donahue 2013). Screening methods
bismus may be associated with reduced stereopsis.
were categorised into refractive and non-refractive screening in-
struments. With regards to detection of strabismus, the American
Association of Pediatric Ophthalmology and Strabismus recom- Type 3: Tests designed to detect reduced central
mends that non-refractive screening devices should detect manifest vision/visual acuity
strabismus > 8 prism dioptres (PD) in primary position (Donahue Though not a specific indicator, visual acuity tests may indicate
2013). the presence of strabismus if visual acuity is reduced in a deviating
eye.
Index test(s)
Type 4: Automated refraction devices designed to report
Different tests are in use to detect strabismus in a community or ocular misalignment
primary care setting by lay screeners or primary care professionals. Some autorefractors indicate asymmetry of corneal reflections.
• Type 1 tests directly identify ocular misalignment, for
example corneal (Hirschberg) or fundus reflections test
(Brückner).
• Type 2 tests assess binocular function such as stereoacuity,
Clinical pathway
e.g. contour and random dot stereotests. Childhood vision screening programs vary around the world, and
• Type 3 tests are designed to detect reduced central vision/ may differ between high and low-middle income countries.
visual acuity, e.g. HOTV, LEA symbols, Keeler (previously World Health Organization (WHO) Member States are grouped
Glasgow) crowded logMAR, Sonksen crowded logMAR, into low and middle-income countries (LMIC) by WHO region,
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 3
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
separating out high-income countries within each of these regions, not give specific information on strabismus detection, instead re-
based on the World Bank’s gross national income per capita (World porting overall precision in detecting amblyopia.
Bank; World Health Organization 2014). Low and middle income This review does not propose screening for strabismus that is of
states have a gross national income per capita of less than USD no aesthetic concern or visual consequence, but it is important to
$12,276 whereas high income states have a gross national income summarise current evidence on accuracy of tests in detecting stra-
per capita of USD $12,276 or more. bismus in a screening setting, in order to enable healthcare com-
In high income countries, there tend to be national guidelines missioners to implement the most effective screening program.
for screening, though these are not necessarily matched by an es-
tablished national screening program. In Israel and Sweden, there
are established national screening programs (Schmucker 2009).
In the UK, the National Screening Committee recommends vi- OBJECTIVES
sion screening of children during the school year which includes To assess and compare the accuracy of tests, alone or in combina-
their 5th birthday, to be delivered under the supervision of or- tion, for screening for strabismus in children aged one to six years,
thoptists (Hall 2003; UK National Screening Committee). De- in a community setting by lay screeners or primary care profes-
spite this recommendation, implementation of childhood vision sionals.
screening continues to show regional variation. In the US, Canada,
Belgium and Germany there are no national screening programs
although regional programs do exist. There are, however, guide- Secondary objectives
lines in the US which recommend vision and alignment screening
between the age of 3 to 3.5 years of age by a suitably trained indi- Other objectives will be to investigate sources of heterogeneity of
vidual (American Academy of Ophthalmology 2012). In Canada, diagnostic accuracy, including:
there are national guidelines for screening visual acuity and ocu- • age;
lar alignment at 3 to 5 years of age, but no established screening • setting;
program (Canadian Paediatric Society 2009). In many countries • type of professionals performing the test;
office-based paediatricians, ophthalmologists and optometrists of- • study design;
fer annual “child health” or “eye health” checks, respectively, but • study size (< 100 vs. ≥ 100 participants, which may reflect
these occur outside of national programs. the adoption of different sampling strategies);
In low and middle income countries, little information on national • variation in the way a test is carried out;
screening programs is available in the literature or online (World • type of strabismus (convergent vs. divergent, horizontal/
Health Organization 2014). One exception is Iran, where a na- vertical);
tional screening program of 3 to 6 year olds performed by kinder- • severity of strabismus (amount of misalignment, constant/
garten teachers assessing visual acuity with illiterate ’E’ Snellen intermittent/latent);
charts has been in place since 1996 with an estimated uptake of
67% of eligible children in 2005 (Khandekar 2009). In India there
is no national screening (Jose 2009).
There are efforts to find cost-effective strategies for screening in METHODS
developing countries such as a remote photo screener system pi-
loted in Brazil and China, and a home-based screening in China
performed by parents (Donahue 2008; Lan 2012). Criteria for considering studies for this review
Types of studies
We plan to include all prospective or retrospective population-
Rationale based test accuracy studies of consecutive participants. For popu-
Strabismus is a risk factor for the development of amblyopia. lation-based we mean not only screening studies, implying sam-
Whilst large deviations may be detected by family, friends or lay pling based on census, but also studies recruiting from commu-
screeners, small deviations may go unnoticed, leading to suppres- nity services such as schools or paediatric health districts. Hospital
sion of visual information from the deviated eye. Childhood vi- cohorts will be excluded, unless the sampling from a community
sion screening programs use varying combinations of tests which, service is clearly described.
depend on the age at which children are tested, and the type of We will include studies that compare a single index test, or a com-
professionals carrying out tests. Strabismus tests as part of a combi- bination of index tests, with a reference standard (cover test, per-
nation of tests may increase the precision of childhood amblyopia formed as a standalone test or as part of a comprehensive eye ex-
screening (VIP 2007). Published vision screening studies often do amination). Case-control studies, in which children are selected
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 4
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
based on their disease status, will be excluded unless they are nested • Type 3: Tests designed to detect reduced central vision/
in large prospective consecutive studies. Studies have to provide visual acuity:
sufficient data to calculate diagnostic accuracy (sensitivity, speci- ◦ e.g. HOTV, LEA symbols, Keeler (previously
ficity). We will include studies in which only a subgroup of partic- Glasgow) crowded logMAR, Sonksen crowded logMAR,
ipants has undergone the reference test; the result from these will crowded Kay picture test, displayed as paper- or computer-based
be considered by subgroup analysis. charts or conventional retroilluminated charts
• Type 4: Automated refraction devices designed report
ocular misalignment.
Participants
Should the search reveal several high-quality studies for each test
Children aged 1 to 6 years will be the population of interest. type for inclusion in this review, we will consider splitting the
Strabismus is a risk factor for the development of amblyopia during review by test type group.
the “sensitive period” of vision development. During this period, Finally, we will not consider tests that require specialist skills, such
neural plasticity is greatest, and it begins to decline around the age as the 4 dioptre prism test, since we are concerned with population
of 6 years; clinical interventions are typically offered to children screening which is typically carried out by lay professionals, not
up to the age of 10 years. Screening programs therefore attempt by orthoptists or ophthalmologists, who would directly use our
to identify children with amblyogenic risk factors before the age reference standard, the cover-uncover test.
of 6 years, to allow remedial treatment. We set the lower limit of
the age range at 1 year to avoid overlap with early postnatal eye
screening programs. Target conditions
In countries where children start school in the academic year of The target condition is constant or intermittent manifest strabis-
their 5th birthday, screening programs aim to capture children mus of any magnitude and type (esotropia, microtropia, exotropia,
age 4 to 5 years, i.e. during their first year at school. In other hyper/hypotropia).
countries, the year of school entry can be earlier or later, and vision
screening programs may be carried out in the first year of school,
Reference standards
or independent from schools. An age range of 1 to 6 years will
allow inclusion of all population-based studies in children at risk The reference standard considered in this review will be the
of developing amblyopia from strabismus. cover-uncover test, whether used alone or within a comprehensive
When studies include children outside the range of 1 to 6 years, ophthalmic examination, or in combination with other tests, by
we will try to obtain subgroup data. If subgroup data are not trained personnel.
obtained, we will include these studies if the proportion of children We will include studies that use the cover/uncover test, regardless
beyond age 6 is less than an agreed threshold, e.g. 20%, and we of the type of professional performing the test. Type of professional
will conduct sensitivity or subgroup analyses as appropriate. (ophthalmologist, orthoptist, trained technician, lay screener) will
We will consider children attending population-based vision be noted and analysed as subgroups.
screening programmes. We will also include opportunistic screen- We will include studies in which the cover/uncover or alternate
ing programmes, such as including children attending schools. We cover test is used as part of a comprehensive eye examination,
will exclude orthoptist-delivered programmes, as these include the which often also includes visual acuity, biomicroscopy and refrac-
reference standard. tion. For the latter scenario there is a risk of incorporation bias.
This bias can be avoided by ensuring that the tests that are part of
the reference standard “comprehensive eye examination” do not
Index tests belong to the same type of test as the index test included in that
analysis. This means we will exclude the whole study if a single
We will include any test used by lay screeners to detect strabismus,
test is assessed and there is incorporation bias, and that we will
either directly by identifying misalignment, or indirectly by iden-
exclude part of the study data if a study comparing several index
tifying a consequence of strabismus such as loss of stereovision.
tests suffers from incorporation bias regarding a specific test.
The participant age range means that different tests may be used,
as appropriate for the age of participants in each particular study.
We will describe the index test by test type rather than enumer-
ated. Search methods for identification of studies
• Type 1: Tests which directly identify ocular misalignment:
◦ corneal reflections tests (Hirschberg);
◦ fundus reflections test (Brückner). Electronic searches
• Type 2: Tests of binocular function: stereoacuity: We will search the Cochrane Database of Systematic Reviews
◦ e.g. contour and random dot stereotests. (CDSR), the Database of Abstracts of Reviews of Effects (DARE),
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 5
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
the Health Technology Assessment Database (HTA) and the NHS • false negatives (FN), i.e. participants categorised as
Economic Evaluation Database (NHSEED) on The Cochrane Li- strabismic by the reference test, but as non-strabismic by the
brary (latest issues), Ovid MEDLINE, Ovid MEDLINE In-Pro- index test;
cess and Other Non-Indexed Citations, Ovid MEDLINE Daily, • true negatives (TN), i.e. participants categorised as non-
Ovid OLDMEDLINE (January 1946 to present), EMBASE (Jan- strabismic by both the reference and index tests;
uary 1980 to present), CINAHL (January 1937 to present), • false positives (FP), i.e. participants categorised as non-
Web of Science Conference Proceedings Citation Index-Science strabismic by the reference test, but as strabismic by the index
(CPCI-S) (January 1990 to present), BIOSIS Previews (January test;
1969 to present), MEDION (to present date), the Aggressive • participants with uninterpretable index test results;
Research Intelligence Facility database (ARIF) (to present date), • missing data, i.e. participants included in the study, but not
the metaRegister of Controlled Trials (mRCT) (www.controlled- in the analyses, by causes of exclusion.
trials.com), ClinicalTrials.gov (www.clinicaltrials.gov) and the
World Health Organization (WHO) International Clinical Trials Uninterpretable test results at individual participant level may be
Registry Platform (ICTRP) (www.who.int/ictrp/search/en). We recorded in primary publications when a child does not comply
will not use any date or language restrictions in the electronic with a test, i.e. refuses to give an answer during assessment of visual
search for trials. acuity or stereovision, or does not fixate on targets for automated
See: Appendices for details of search strategies for The Cochrane devices, or in case of ocular abnormalities affecting the clarity of
Library (Appendix 3), MEDLINE (Appendix 4), EMBASE cornea, lens and/or vitreous. For each study, we will record how
(Appendix 5), CINAHL (Appendix 6), CPCI-S (Appendix such cases were treated in the analyses.
7), BIOSIS Previews (Appendix 8), MEDION (Appendix 9),
ARIF (Appendix 10), mRCT (Appendix 11), ClinicalTrials.gov Two review authors will independently extract the data to ensure
(Appendix 12) and the ICTRP (Appendix 13). consistency and enter these into the Cochrane Collaboration’s sta-
tistical software, Review Manager (RevMan 2012). We will extract
the data shown in Table 1, which we will display in the “Charac-
Searching other resources teristics of included studies” tables.
We will use the following weblink: http://pcwww.liv.ac.uk/
~rowef/index_files/Page646.htm. to search the following orthop- Assessment of methodological quality
tic journals and conference proceedings which are not electron-
ically listed: British and Irish Orthoptic Journal, American Or- We will use the Quality Assessment of Diagnostic Accuracy Stud-
thoptic Journal, Australian Orthoptic Journal, European Strabis- ies (QUADAS)-2 tool to evaluate the risk of bias and applicabil-
mus Association, International Strabismus Association and the In- ity of primary studies (http://www.bris.ac.uk/quadas/quadas-2).
ternational Orthoptic Congress. QUADAS-2 consists of four key domains: patient selection, index
test, reference standard, and flow and timing. The tool is com-
pleted in four phases:
1. the review question is stated;
Data collection and analysis 2. development of review specific guidance;
3. review of the published flow diagram for the primary study
or construction of a flow diagram if none is reported;
Selection of studies 4. judgement of bias and applicability.
Two review authors will independently assess titles and abstracts Each domain is assessed in terms of the risk of bias and the first
for eligibility. We plan to sort abstracts into “definitely exclude” three are also assessed in terms of concerns regarding applicability.
and “possibly include” categories, recognising that sometimes it is To help reach a judgement on the risk of bias, signalling questions
not possible to judge from the abstract whether a reference fulfils are included. These flag aspects of study design related to the
the criteria or not. We will place all abstracts selected by at least one potential for bias and aim to help review authors make risk of
review author in the “possibly include” category. We will resolve bias judgements. Two review authors will independently assess the
disagreements at each step by discussion between the two review methodological quality of the included studies. Disagreements on
authors and a third senior author. study quality will be resolved by a third senior author. Table 2
shows the guidance the review authors will use when judging the
methodological quality of studies.
Data extraction and management We will score risk of bias signalling questions as “yes/no/unclear”
We will extract the number of: as detailed in Table 2. Risk of bias is judged as “low”, “high” or
• true positives (TP), i.e. participants categorised as “unclear”. If all signalling questions for a domain are answered
strabismic by both the reference and index test; “yes” then the risk of bias can be judged as “low”. If any question is
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 6
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
answered “no”, this flags the potential for bias. Should this occur, To analyse different tests within each test type group, we will use
we will follow the guidelines developed in phase 2 of the quality the following strategy. For each study, we will extract data at specific
assessment process to judge the risk of bias. We will use the “un- thresholds if available. Ideally, we would extract cut-offs of 8 PD
clear” category only when insufficient data are reported to permit for horizontal and 1 PD for vertical deviations in test type 1; 400
a judgment. arc seconds for test type 2; visual acuity 0.2 logMAR for test type 3.
We will judge applicability of primary studies to the review ques- UK screening recommendations specify “less than 0.2 logMAR” as
tion in a similar manner. referral threshold (UK National Screening Committee); guidelines
We will also record study sponsorship. from the American Association of Pediatric Ophthalmology and
Strabismus specify that optotype-based screening (which covers
test type 3) should detect visual acuity of less than 0.176 logMAR
Statistical analysis and data synthesis (Snellen 20/30) at all ages. Threshold values for test type 4 have
We will use two-by-two data of index and reference test results to not been published; we will therefore use “any asymmetry, in mm
calculate the sensitivities and specificities, with their 95% confi- or as ratio” as threshold. Thresholds are summarised in Table 3.
dence intervals. We will use the RevMan software for these de- We expect that studies assess several tests for use in screening pro-
scriptive analyses, and to plot individual studies in forest plots, as grammes, and report their sensitivity as a fixed, high specificity,
well as in the receiver operating characteristic (ROC) space, and such as 90% or 95%. In this case, we will extract sensitivities at
to produce summary ROC curves. 95% specificity with a 5% tolerance, which means at least 90%
Because several tests are grouped in test types, a two-step analytic specificity and analyses these using univariate meta-analysis.
procedure will be used. First, individual tests will be compared A final scenario will be that most studies report sparse data at a
within each test type, because they are based on the same construct single and heterogenous threshold cut-off, and numerically iden-
and a common output measure may be available. Then, we will tical thresholds may yield different sensitivity/specificity pairs. In
compare all test types with each other using all available studies, this case we will pool data as available, regardless of the cut-off
considering individual tests as equivalent in each type group, unless used, using the HSROC model.
differently advised by analyses in the first step. To compare different testing devices within test type, we will in-
The analytic plan presented hereafter shows that we will attempt corporate categorised covariates coding for individual test in the
to analyse the full impact of cut-off values on test performance, but bivariate or HSROC model as outlined in Chapter 10 of the
are prepared to ultimately analyse sparse and heterogenous data. Cochrane Handbook for Systematic Reviews of Diagnostic Test Accu-
As a consequence of this potentially complex analytic framework, racy (Macaskill 2010).
naturally arising from the inclusion of multiple tests, we plan to
use a bivariate (Reitsma 2005) and/or a hierarchical summary re-
ceiver-operator curves (HSROC) (Rutter 2001) model, using the
Comparison of test types
’METADAS’ macro in the SAS analytical software, version 9.2, or
carry out univariate analyses, depending on the analytic step and We will use all tests that were reported in all studies for this analysis,
the data available. and we will compare test types using a covariate coding for each
Considering test threshold across different test types will be the test type in the bivariate model. We will analyse studies reporting
most important analytic issue in this review. In fact, a continu- on multiple tests as if the tests were independent.
ous output measure should be available for most tests: ocular mis- If index test combinations are found, we will extract the most
alignment as PD or degrees for test type 1, stereoscopic acuity as frequent combinations and analyse them as if each combination
seconds of arc for test type 2, visual acuity in logMAR for test type were a unique testing criterion.
3 (acknowledging that comparison of values may be hampered by For these analyses, we will give preference to the inclusion of a
use of charts with different optotype size steps, and that simple common threshold within each test type. If this is not available,
mathematical conversion from Snellen to logMAR may be inac- we will follow the procedure described above.
curate), millimetres or a ratio for test type 4. Other tests listed
in Appendix 2 and used in the diagnosis of, but not in screen-
ing for, strabismus are not based on an explicit common measure.
Direct comparisons between tests and test types
However, in practice the heterogeneous execution and technical
characteristics of the tests will make it difficult to consider using We will report comparisons made between specific tests in a single
an explicit threshold in statistical analyses, and implicit threshold study and analyse them separately both at the test and at type
effects are more likely. level. Any differences between tests found by means of indirect
comparisons will in fact be treated with caution as compared to
such direct comparisons, since the former have been shown to be
Analyses within each test type at risk of biased conclusions (Takwoingi 2013).
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 7
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Investigations of heterogeneity studies of low quality (high likelihood of bias) to determine if the
The framework for likely sources of heterogeneity was described results are influenced by inclusion of the lower quality studies.
previously and mainly includes setting and study population, par- We will also carry out sensitivity analyses assuming that any un-
ticularly regarding referral method and inclusion criteria, type of interpretable index test results are all correct or incorrect.
professional executing the reference standard, and study quality
assessment. Assessment of reporting bias
We will investigate heterogeneity win the first instance through We will not assess publication bias since there is no standard
visual examination of forest plots of sensitivities and specificities method to achieve this in diagnostic test accuracy reviews (Deeks
and through visual examination of the ROC plot of the raw data. 2005). For selective outcome reporting issues, such as the use of
We will initially investigate the effect of sources of heterogeneity a specific cut-off of ocular misalignment, we will not search for a
by including each of these as covariates in the meta-regression protocol to assess within-study reporting bias, since protocols of
analyses. diagnostic accuracy studies are not routinely reported.
Sensitivity analyses
Where appropriate (i.e. if not already explored in our analyses
ACKNOWLEDGEMENTS
of heterogeneity) and if sufficient data are available, we will ex-
plore the sensitivity of any summary accuracy estimates to aspects The Cochrane Eyes and Vision Group will create and execute
of study quality such as nature of blinding and type of reference the search strategies. We wish to thank Mrs Angela Coleman,
standard, guided by the anchoring statements developed in our Head Orthoptist at Moorfields at Bedford Hospital, for the critical
QUADAS-2 exercise. Primary analysis will include all studies (un- review of the protocol, and Tess Garritty, Helen Griffiths and Fiona
less incorporation bias is evident); sensitivity analysis will exclude Rowe for external peer review comments to the protocol.
REFERENCES
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 12
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
diseases in young children: the first four years. Transactions Wedner 2000
of the American Ophthalmological Society 2008;106:179–85. Wedner SH, Ross DA, Balira R, Kaji L, Foster A. Prevalence
Turacli 1995 of eye diseases in primary school children in a rural area of
Turacli ME, Aktan SG, Duruk K. Ophthalmic screening Tanzania. British Journal of Ophthalmology 2000;84(11):
of school children in Ankara. European Journal of 1291–7.
Ophthalmology 1995;5(3):181–6.
Weinand 1998
UK National Screening Committee Weinand F, Graf M, Demming K. Sensitivity of the MTI
UK National Screening Committee. The UK NSC policy on photoscreener for amblyogenic factors in infancy and early
vision defects screening in children. www.screening.nhs.uk/ childhood. Graefe’s Archive for Clinical and Experimental
vision-child (accessed 30 April 2013). Ophthalmology 1998;236(11):801–5.
VanEenwyk 2008
VanEenwyk J, Agah A, Giangiacomo J, Cibis G. Wick 1980
Artificial intelligence techniques for automatic screening Wick B, London R. The Hirschberg test: analysis from
of amblyogenic factors. Transactions of the American birth to age 5. Journal of the American Optometric Association
Ophthalmological Society 2008;106:64–73. 1980;51(11):1009–10.
VIP 2005 Williams 2001
Anonymous. Preschool vision screening tests administered Williams C, Harrad RA, Harvey I, Sparrow JM, ALSPAC
by nurse screeners compared with lay screeners in the vision Study Team. Screening for amblyopia in preschool children:
in preschoolers study. Investigative Ophthalmology and results of a population-based, randomised controlled
Visual Science 2005;46(8):2639–48. trial. ALSPAC Study Team. Avon Longitudinal Study of
VIP 2007 Pregnancy and Childhood. Ophthalmic Epidemiology 2001;
Anonymous. Does assessing eye alignment along with 8(5):279–95.
refractive error or visual acuity increase sensitivity for
detection of strabismus in preschool vision screening?. World Bank
Investigative Ophthalmology and Visual Science 2007;48(7): World Bank. How we classify countries.
3115–25. data.worldbank.org/about/country-classifications (accessed
Walraven 1993 30 April 2013).
Walraven J, Janzen P. TNO stereopsis test as an aid to the
prevention of amblyopia. Ophthalmic and Physiological World Health Organization 2014
Optics 1993;13(4):350–6. World Health Organization. Definition of region
groupings. www.who.int/healthinfo/global_burden_
Walsh 2000
disease/definition_regions/en/ (accessed 9 June 2014).
Walsh LA, Laroche GR, Tremblay F. The use of binocular
visual acuity in the assessment of intermittent exotropia. Yang, 2012
Journal of AAPOS 2000;4(3):154–7. Yang HK, Han SB, Hwang JM, Kim YJ, Jeong CB, Kim
Webber 2008 KG. Assessment of binocular alignment using the three-
Webber AL, Wood JM, Gole GA, Brown B. The effect of dimensional Strabismus Photo Analyzer. British Journal of
amblyopia on fine motor skills in children. Investigative Ophthalmology 2012;96(1):78–82.
∗
Ophthalmology and Visual Science 2008;49(2):594–603. Indicates the major publication for the study
ADDITIONAL TABLES
Table 1. Data extraction from included studies
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 13
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 1. Data extraction from included studies (Continued)
Study design Selection: as single group/as separate group with/without target condition
Enrolment: Consecutive series.
Identification: prospective/retrospective
If more than one test: how were tests allocated to individuals, did each individual receive all tests?
Target condition Constant and intermittent manifest strabismus (esotropia, exotropia, vertical tropia, microtropia),
including the prevalence of the target condition in the sample
Reference standard Test definition and description, i.e. cover test; “comprehensive eye examination” (visual acuity, cover
test, cycloplegic refraction)
Test operator(s)
Timing of reference standard
PATIENT SELECTION
Describe methods of patient selection: Describe included patients (prior testing, presentation, intended use of index test and
setting)
Was a consecutive or random Consecutive sampling or ran- Non random sampling or sam- Unclear whether consecutive or
sample of patients enrolled? dom sampling of children ac- pling based on volunteering or random sampling used
cording to inclusion criteria referral
Was a case-control design Yes for all studies since case- N/A N/A
avoided? control studies are excluded un-
less nested in cohort studies
Did the study avoid inappro- • Exclusions are detailed Inappropriate exclusions are re- Exclusions are not detailed
priate exclusions? and felt to be appropriate ported e.g. of children in whom (pending contact with study au-
(systemic disease causing strabismus has been suspected thors)
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 14
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 2. QUADAS-2 assessment guidance (Continued)
Concerns regarding applica- Inclusion of children in com- Inclusion of children over the Unclear inclusion criteria
bility: Are there concerns that munity settings, such as school age of 6 years, referred to clini-
the included patients do not or screening settings, with no cal settings, referred to eye pro-
match the review question? previous diagnosis of any eye fessionals for suspect eye dis-
disease ease, or assessed in commercial
settings on a volunteer basis;
or previous diagnosis of failed
screening test or strabismus
INDEX TEST
Describe the index test and how it was conducted and interpreted
Were the index test results in- Test performed “blinded” or Reference standard results avail- Unclear whether results are in-
terpreted without knowledge “independently and without able to those who conducted or terpreted independently
of the results of the reference knowledge of ” reference stan- interpreted the index tests
standard? dard results are sufficient and
full details of the blinding pro-
cedure are not required; or clear
temporal pattern to the order of
testing that precludes the need
for formal blinding
If a threshold was used, was it Many included index tests are A study is classified at higher No information on pre-selec-
pre-specified? based on continuous measures risk of bias if the authors de- tion of index test cut-off values
(e.g. eye deviation, stereopsis, fine the optimal cut-off post-
refractive error, visual acuity); hoc based on their own study
the study authors declare that data
the selected cut-off used to di-
chotomise data was specified a
priori, or a protocol is available
with this information
Concerns regarding applica- Tests used and testing proce- Tests used are not validated or Unclear tests (e.g. stereopsis-
bility: Are there concerns that dure clearly reported and tests study personnel is insufficiently based tests but does not men-
the index test, its conduct, or executed by personnel with suf- trained tion if a validated test is used) or
interpretation differ from the ficient training unclear study personnel profile,
review question?
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 15
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 2. QUADAS-2 assessment guidance (Continued)
REFERENCE STANDARD
Describe the reference standard and how it was conducted and interpreted
Is the reference standard Cover-un- Complete eye examination with Complete eye examination used
likely to correctly classify the cover test performed by trained cover-uncover test used as refer- but unclear whether cover-un-
target condition? professionals, e.g. ophthalmol- ence standard but not only the cover test used
ogists, optometrists, orthoptists cover-uncover test used to judge
on strabismus (e.g. visual acuity
measure also used)
Were the reference standard Reference standard performed Index test results available to Unclear whether results are in-
results interpreted without “blinded” or “independently those who conducted the ref- terpreted independently
knowledge of the results of and without knowledge of ” in- erence standard; or the index
the index test? dex test results are sufficient and test is part of the reference stan-
full details of the blinding pro- dard (e.g. visual acuity within a
cedure are not required; or clear compete ophthalmic examina-
temporal pattern to the order of tion used as reference standard
testing that precludes the need and visual acuity is also the in-
for formal blinding dex test analysed- this will be
specific of each analysis)
Concerns regarding applica- Cover-uncover test used and Cover-uncover test used by per- Unclear study personnel profile,
bility: Are there concerns that testing procedure executed by sonnel with inappropriate pro- background and training
the target condition as de- personnel with sufficient train- file or insufficient training
fined by the reference stan- ing
dard does not match the re-
view question?
Was there an appropriate in- No more than three months be- More than three months be- Unclear whether test results are
terval between index test(s) tween index and reference test tween index and reference test executed within three months
and reference standard? execution, and no corrective in- execution
tervention between assessments
Did all patients receive a ref- The verification rate of index All children receiving the index Unclear whether all children re-
erence standard? test positive children is defi- test are verified with the refer- ceiving the index test are veri-
nitely higher than that of neg- ence standard fied with the reference standard
ative children (the opposite is
unlikely)
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 16
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 2. QUADAS-2 assessment guidance (Continued)
Did all patients receive the All children are verified with the Some children, i.e. positive chil- Unclear whether all children are
same reference standard? cover-uncover test by trained dren, are verified with the cover- verified with the cover-uncover
professionals uncover test by specialised per- test by trained professionals
sonnel, while the others are ver-
ified by personnel with lower
level of training
Were all patients included in The number of children in- The number of children in- The number of children anal-
the analysis? cluded in the study does not cluded in the study does not ysed, but not that included in
match the number in analyses match the number in analyses the study are reported, or un-
or children with undefined or and children with undefined or clear if there were inappropriate
borderline test results are ex- borderline test results are ex- exclusions
cluded. However, children in cluded from the analyses
whom one or more index tests
are not performed because they
are poorly cooperative can be
excluded
Were all tests performed on All children received all index Not all children received all in- Not all children received all in-
all patients, or randomly as- tests, or tests were randomly as- dex tests and the assignment dex tests and the assignment cri-
signed? signed criterion was opportunistic or terion was unclear
non-random (e.g. depending
on test availability or type of
professional)
Could the order in which the The order of presentation of the Several tests are delivered in Unclear order of test presenta-
index tests were used affect index test was random or alter- a fixed order which can cause tion
the target condition or the in- nate to avoid fatigue effects; or children to be less compliant
terpretation of the alternative clear that no fatigue effect can with the second or later test
tests? arise
Test type categories Tests included Output measure Threshold to extract data
1) Tests which identify ocular 1.1) Corneal reflections tests: Prism dioptres (PD) 8 PD for horizontal deviations;
misalignment Hirschberg, Krimsky (prism re- 1 PD for vertical deviations (no
flection test) published threshold identified)
1.2) Fundus reflections test:
Brückner
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 17
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Table 3. Thresholds for analysis (Continued)
2) Test of binocular function: Stereoacuity tests such as con- Seconds of arc 400 seconds of arc
stereopsis tour and random dot stereotests
3) Tests designed to detect re- 3.1): visual acuity tests, e.g. LogMAR or 0.2 logMAR
duced ventral vision HOTV, LEA symbols, Keeler logMAR equivalent
(previously Glasgow) crowded
logMAR, Sonksen crowded
logMAR, crowded Kay picture
test
3.2) Suppression tests
3.3) blur test
APPENDICES
Friedman 2009 USA, Baltimore Any constant or intermittent Age 6 to 71 months, white 3.3%
heterotropia at near or dis-
tance fixation
Preslan 1996 USA, Baltimore Manifest strabismic devia- All children attending one 3.1%
tion with or without fixa- school: preschool (125 chil-
tion preference, using alter- dren), kindergarten (213),
nate cover testing with the first grade (165) and second
child fixating on an accom- grade (177); 75% African-
modative target at 33 cm American, 22% white, 3%
other
Traboulsi 2008 USA, Cleveland Manifest ocular deviation of Age 4 to 8 years 2.1%
any type or magnitude with General population of Cleve-
or without fixation prefer- land: 70.3% African Amer-
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 18
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)
Turacli 1995 Turkey, Ankara Not defined, but results re- Children attending 2.5%
port manifest and latent de- randomly selected school in
viations, with manifest in the Ankara urban area, age 5
93% of all those classified as to 12 years
“strabismic”
Wedner 2000 Tanzania, rural area Not defined Tanzanian children 0.5%
and young people age 7 to 19
years
Matsuo 2007a Japan, Okyama Japanese children age 1.5 0.01 to 0.12%
Not defined in abstract; full years 0.99%
article not available
Japanese children age 3 years 0.2 to 0.34%
Matsuo 2007b Japan, Okyama Not defined in abstract; full Japanese children age 6 to 12 0.99% in 2005, 1.28% in
article not available years 2003
Graham 1974 UK Manifest deviation, or ex- All children born in Cardiff manifest or large latent: 7%;
ophoria of 9 PD or more, or between Jan 1 and Dec 31 in manifest only: 5%
esophoria of 7 PD or more, one year (not specified), age large phoria: 1.3%
any hyperphoria 5 to 6 years
Use of the Hirschberg test as a screening test for ocular alignment has been recommended in young preverbal and also in pre-school
children (daSilva 1991; Sansonetti 2004). To allow standardisation, videographic techniques have been proposed (Miller 1993). These
can be applied when using video refractors or photoscreeners developed for the automated assessment of refractive errors (Griffin 1989;
Hasebe 1995; Moghaddam 2012; Schaeffel 2002; Weinand 1998).
Automated assessment of CR on digital photographs and videographs is currently in development (Almeida 2012; Model 2012; Yang,
2012).
Accuracy of the Hirschberg test may be in the range of +/- 9 to 10 PD, which would make it unsuitable to detect or exclude microtropia.
In orthoptic practice, Hirschberg and Krimsky tests are reserved for very young, preverbal patients or those with profound visual
impairment which prevents fixation with the affected eye(s). The Hirschberg test is useful to demonstrate pseudo-strabismus in young
children with a broad nasal bridge and epicanthal folds or in individuals with wide interpupillary distance.
Type 1 tests:
Direct identification of ocular
alignment
Cover-uncover test Dichotomous: refixation move- Any refixation movement Traboulsi 2008: manifest ocular
ment present/absent deviation of any type or mag-
nitude with or without fixation
preference
Alternate cover test Dichotomous: refixation move- Demonstrates manifest plus la- Traboulsi 2008: Any vertical
ment present/absent tent strabismus phoria; any esophoria more
than 8 PD or exophoria more
than 15 PD at near or distance
Prism and alternate cover test Categorical: Quantifies manifest plus latent Traboulsi 2008: Any vertical
Prisms used to neutralise devia- strabismus phoria; any esophoria more
tion than 8 PD or exophoria more
than 15 PD at near or distance
Contour tests: Titmus Fly 3,600, Animals 400, 200, Fly 3,600 sec of arc; animals Traboulsi 2008: less than 400
100, Wirt circles 800, 400, 200, 400 , 200, 100; Wirt rings 800 seconds of arc
140, 100, 80, 60, 50, 40 sec- to 40 seconds of arc
onds of arc at 40 cm
TNO 480, 240, 120, 60, 30, 15 sec- At age 5 years: greater than 60 Manufac-
onds of arc seconds of arc turer recommends 240 seconds
of arc as “fail” threshold, as 95%
of amblyopes are unable to see
this figure
Random dot E at 0.5, 1 and 1. Schmidt 2004: non-stereocard Schmidt 2004: non-stereocard
5m only; 504; 252; 168 arc seconds only; 504; 252; 168 seconds of
arc
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 23
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)
Stereo Smile II at 40 cm Schmidt 2004: nonstereo card Schmidt 2004: non-stereo card
only; 480; 240; 120 arc seconds only; 480; 240; 120 seconds of
arc
Type 3 tests:
Visual acuity tests
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 24
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
(Continued)
Sonksen
Type 4 tests:
Refraction devices that report
ocular misalignment
Other tests used in the diagnosis of strabismus, but not in primary care or community screening settings delivered by lay
screeners or primary care professionals:
Krimsky test
The prism reflection test is an modification of the Hirschberg test. The patient fixes a spotlight at a near position (33 cm). Prisms are
placed in front of the fixing eye, with the apex pointing in the direction of the deviation. This shifts the CR towards the centre of the
pupil. The prism which positions the CR in the centre of both pupils indicates the angle of deviation (Krimsky 1951). This test is
typically used in children or adults who are not cooperative enough to undertake a Prism Cover Test or in cases when the visual acuity
is poor and the patient is unable to move the eye to take up fixation.The 95% limit of agreement of inter-observer variability for the
Krimsky test has been reported as 6.1 PD (Yang, 2012). Neutralisation of the deviation with prisms requires orthoptic expertise, and
this test is not used by lay screeners or non-ophthalmic professionals in community screening programs.
Prism reflection test
The prism reflection test is similar to Krimsky, but the prism is placed in front of the deviating eye. Some authors found the prism
reflection test to have low accuracy (Choi 1998).
Controlled binocular acuity (CBA) test of strabismus
Some types of strabismus can be controlled by increased accommodation, i.e. “over-focussing”. In intermittent distance exotropia
(IDEX), individuals may use accommodative convergence to control the exodeviation; hence over-accommodating and losing clarity
of vision in the distance (Walsh 2000). To measure controlled binocular acuity, the patient reads an optotype chart whilst the examiner
observes the patient’s ocular alignment by corneal reflections and noting at which optotype size one eye deviates.
Suppression tests
Visual information from the deviating eye is suppressed at the level of the visual cortex of the brain (Smith 1985). Binocular suppression
tests may have high specificity and may be useful for strabismus and amblyopia screening (Pott 1998; Simons 1996b). Tests are based
on dissociation of the images seen by each eye, for example by using polarised lenses (Nuzzi 1986; Pott 1998; Pott 2003; Prakash
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 25
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
1996). Worth’s four light test, Bagolini striated lenses or the synoptophore are used to detect the presence of suppression. The 4 prism
dioptre (PD) prism test is used to detect central suppression. The area of suppression within the field of vision can be approximated
by the synoptophore. Density of suppression can be quantified by using neutral density filters or the Sbisa bar (Bagolini filter bar)
(McCormick 2002). Whilst used in orthoptic practice to determine the risk of developing double vision in adult patients undergoing
amblyopia treatment, suppression measurement is not used in paediatric screening.
Fusion tests
Orthoptists may test children’s ability to overcome a 4, 15 or 20 PD prism to maintain binocular single vision (prism reflex test); failure
do overcome the prism can indicate weakness of fusional control. The prism fusion range uses the same principle, but uses a prism bar,
allowing a more detailed measurement of prisms that can be overcome. Measurements are typically in steps as dictated by prism bar or
loose prism used (2 to 20 PD usually in 2 PD increments, then 20 to 45 PD in 5 PD increments).
Blur test
The blur test aims to detect low hypermetropia, which may be associated with potentially decompensating strabismus.
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 29
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Appendix 7. Web of Science Conference Proceedings Citation Index - Science (CPCI-S) search
strategy
#20 #15 AND #17 AND #19
#19 TS= (infan* or child* or toddler* or boy* or girl* or paediatric* or pediatric* or nursery or nurseries or kindergarten* or pre school*
or preschool* or schoolage* or school age*)
#18 #16 OR #17
#17 TS= (esotrop* or exotrop* or hypertrop* or hypotrop* or microtrop*)
#16 TS= (strabimus or strabismic or squint*)
#15 #1 OR #2 OR #3 OR #4 OR #5 OR #6 OR #7 OR #8 OR #9 OR #10 OR #11 OR #12 OR #13 OR #14
#14 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 optometrist*)
#13 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 ophthalmologist*)
#12 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 orthoptist*)
#11 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 monitor*)
#10 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 referral*)
#9 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 referred*)
#8 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 hospital*)
#7 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 population*)
#6 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 communit*)
#5 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 program*)
#4 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 visual*)
#3 TS= ((test* or screen* or diagnos* or assess*) NEAR/4 vision)
#2 TS=mass screening
#1 TS= vision test
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 30
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
Appendix 9. MEDION search strategy
Database was searched on ICPC code field. Using code “f ” for ophthalmology.
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 31
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.
CONTRIBUTIONS OF AUTHORS
All authors helped to draft the protocol.
DECLARATIONS OF INTEREST
The authors have no interest to declare.
SOURCES OF SUPPORT
Internal sources
• NIHR Biomedical Research Centre, UK.
The contributions by Annegret Dahlmann-Noor to this review were supported by the National Institute for Health Research (NIHR)
Biomedical Research Centre based at Moorfields Eye Hospital NHS Foundation Trust and UCL Institute of Ophthalmology. The
views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health.
External sources
• NIHR Biomedical Research Centre, UK.
Richard Wormald (CEVG Co-ordinating Editor) acknowledges financial support for his CEVG research sessions from the National
Institute for Health Research (NIHR) Biomedical Research Centre based at Moorfields Eye Hospital NHS Foundation Trust and
UCL Institute of Ophthalmology. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or
the Department of Health.
Tests for detecting strabismus in children age 1 to 6 years in the community (Protocol) 32
Copyright © 2014 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.