Al-Shifa Journal of Ophthalmology (ASJO)
|Aims and Scope||
Sensitivity of diagnostic tests used in kerato-conjunctivitis sicca
Madiha Durrani, MBBS, MCPS, FRCS
Purpose of the study: To identify the appropriate and sensitive methods for diagnosis of keratoconjunctivitis sicca.
Design: Cross sectional prevalence study starting from July 2002 till Dec. 2002.
Participants: This study included 100 patients, who had symptoms associated with dry eye recruited from the out patient department of the Al-Shifa Trust Eye Hospital, Rawalpindi, by convenient sampling.
Outcome measures: The sensitivity of three diagnostic tests; Tear film break-up time, Schirmer test and Rose Bengal staining in correctly diagnosing patients with dry eye.
The mean sensitivity of these tests in
diagnosis of dry eyes was 51.5 % for Rose Bengal staining, 50.3% for
TBUT and 47.5% for Schirmerís.
Conclusions: There is no single test that is absolutely diagnostic of dry eyes and performing all three tests in all patients is not cost effective. Depending upon the sensitivity of each test in different groups of patients it is recommended that Rose Bengal stain should be preformed on patients with chemical burns Steven Johnson and Sjogren syndrome. Schirmer test should be performed on patients with trachomatous scarring and postmenopausal females. TBUT should be performed on contact lens wearers and patients with ocular allergic diseases. Al-Shifa Journal of Ophthalmology 2005; 1: 25-29 © Al-Shifa Trust Eye Hospital, Rawalpindi, Pakistan.
eye syndrome or Keratoconjunctivitis sicca is one of the most
misdiagnosed ocular conditions. Complaints of patients range from
burning, redness, foreign body sensations and photophobia, to stringy
mucus, sandy or gritty sensations and tearing. All these complaints are
also characteristic of other ocular ailments, so raising doubts about
the diagnosis. Tear film stability can be threatened by many eye
disorders that disturb tear volume, composition, and hydrodynamic
received: May 15, 2004
Participants and methods
Study Design: This was a cross sectional prevalence study starting from July 2002 till Dec. 2002.
Subjects: This study included 100 patients, who had symptoms associated with dry eye recruited from the out patient department of the Al-Shifa Trust Eye Hospital, Rawalpindi, by convenient sampling. On the basis of presenting complaints and provisional diagnosis the patients were divided into the following nine groups:
Group 1: Patients with lid-anomalies e.g. blepharitis, history of lid surgery (Ptosis surgery)
Group 2: History of systemic diseases e.g. Sjogrenís or Steven-Johnson Syndrome.
Group 3: Patients with corneal and conjunctival conditions like epithelial defects and pterygium.
Group 4: Patients on drugs like systemic anti-hypertensives and topical Timolol maleate.
Group 5: Patients with trachomatous scarring.
Group 6: Post-menopausal females.
Group 7: Patients using contact lenses.
Group 8: Patients with allergic diseases.
Group 9: Patients with chemical burns.
∑ Tear film break-up time: fluorescein strip was installed in the lower fornix to stain the tear film. The tear film was examined with Cobalt Blue filter. The time taken till the appearance of first dry spot was noted. The test was repeated three times and the average time taken was noted down. It was taken as positive if the TBUT was less than 9 seconds.
∑ Schirmer test with out topical anesthesia, using strips of Whatman filter paper in the inferior fornix, at the junction of lateral one third and medial two thirds was performed. Patients were instructed to blink normally. It was taken as positive if wetting was less than 10mm in five minutesRose Bengal staining was carried out using Rose Bengal impregnated strips. It was done after Schirmer test to avoid false results. The degree of staining was recorded in each zone on a scale of 0 to 3. It was taken as positive when more than 3 zones were stained. These three zones are, nasal and temporal bulbar conjunctiva and cornea, whereas inferior fornix staining was ignored.
Table: Percentage Sensitivity of diagnostic tests used in evaluation of clinical groups
In this hospital- based study an attempt was made to find out the sensitivity of various tests performed, in each etiological group of patients.
In the available literature, there is controversy regarding the criteria for diagnosis of dry eyes based on clinical diagnostic tests1. A number of diagnostic tests have been described in the literature for evaluation of dry eyes. All clinical tests have limited diagnostic value if performed individually or in the absence of severe symptoms. It is not uncommon to find an abnormality of a single tear function test2. Thus for diagnosis of dry eyes more then one test should be abnormal.
In this study three tests were performed on the patients. These included Schirmer testing, tear film break up time and Rose Bengal staining of ocular surface. The mean sensitivity of these tests in diagnosis of dry eyes was 51.5 % for Rose Bengal staining, 50.3% for TBUT and 47.5% for Schirmerís.
Conflicting information commonly results from the Schirmerís test .It has been reported in the past that Schirmerís test has a low sensitivity but high specificity3. In my study, the Schirmer test was having an overall sensitivity of 47.5% but it was a highly sensitive test in patients group II of systemic illness i.e. RA, SJ Syndrome and Sjogren (82%), group IX chemical burns (75%), 70% in patients group V of trachoma and 60% in post-menopausal females (group VI). Schirmer test proved to be the most sensitive test in the postmenopausal females indicating a possible relationship with hormonal change. Schirmer Test was performed with out installation of topical anesthesia as the cornea and conjunctiva was preserved from the damaging effects of topical anesthesia and more reliable results for Rose Bengal staining later on.
Tear film breakup time is the only direct evidence of stability of tear film. But the test has its own limitations. The test was therefore conducted carefully, taking average of three readings under same clinical conditions. The study results showed the test to have more than 50% sensitivity for dry eye patients, with highest sensitivity in patient groups of trachomatous complications (group V 70%), contact lens wearers (group XII 50%) and allergic conditions (group VIII 50%).
Previous studies have shown the Rose Bengal test to have high sensitivity and specificity4. The results of Rose Bengal, however, can be fallacious in Schirmer and BUT are done before hand, as the contact of filter paper and fluorescein strip respectively, with the conjunctiva and cornea can cause staining of those sites5. When interpreting the results of Rose Bengal testing, staining in the lower fornix was ignored for these reasons and only the staining in the inter-palpabral area was considered
The study showed Rose Bengal staining, in spite of its limitations, to be the most sensitive test in-patient with dry eyes (51.5%). It was most sensitive in patients with epithelial defects due to chemical burns (group IX 100%), systemic disorders (group II 82%) lid diseases (group I 70%) and corneal and conjunctival diseases (group III 54%).
In short, this study shows that though Rose Bengal testing is the most sensitive test for KCS, no single test is sensitive enough to predict the disease on its own and more than one test must be performed to confirm the diagnosis.
∑ Rose Bengal stain should be preformed on patients with chemical burns Steven Johnson and Sjogren syndrome.
∑ Schirmer test should be performed on patients with trachomatous scarring and postmenopausal females.
∑ TBUT should be performed on contact lens wearers and patients with ocular allergic diseases.
these tests are highly sensitive for the disease conditions, they can be
performed singly to reach a diagnosis.
1. Taylor HR, Louis WJ. Significance of tear film test abnormalities. Ann Ophthalmol 1980; 12: 531-5
2. Taylor HR, Louis WJ. Significance of tear film test abnormalities. Ann Ophthalmol 1980; 12: 531-5
3. Farris Rl, Gilbard JP, Mandel ID. Diagnostic tests in KCS. CLAO J 1983; 9: 23-8
Goren Mb, Goren SB. Diagnostic tests in -patients with symptoms of
dry eye. Am J Ophthalmol 1988; 106: 570-4
5. Bron JA. Duke-Elder Lecture. Prospects for the dry eye. Trans Ophthalmol Soc UK 1985; 104: 801-26