Family Practice Vol. 20, No. 4, 425-427
© Oxford University Press 2003
Clinical Research |
The predictive value of the penlight test for photophobia for serious eye pathology in general practice
Department of Family Medicine, Rabin Medical Centre-Beilinson Campus, Petach Tikvah, Israel 49100 and
a Kidlington Health Centre, Kidlington, Oxon OX5 1AP, UK.
Correspondence to Dr John Yaphe; E-mail: yonahyaphe{at}hotmail.com
Yaphe J and Singh Pandher K. The predictive value of the penlight test for photophobia for serious eye pathology in general practice. Family Practice 2003; 20: 425–427.
Received 21 October 2002; Accepted 28 March 2003.
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Background. Photophobia has been described as a useful sign in distinguishing severe from benign eye pathology.
Objective. This observational study was performed to assess the utility of the penlight test for photophobia in patients presenting with a red eye in general practice.
Methods. Thirty patients presenting with a red eye to one GP in 1 year were studied. All patients were examined with a penlight flashlight for photophobia. This was followed by slit-lamp examination for diagnosis of serious or benign eye pathology.
Results. The penlight test was positive in eight out of 10 (80%) patients with serious pathology (mainly uveitis and keratitis) and was negative in 21 out of 26 (81%) patients with less serious pathology (mainly conjunctivitis). The positive predictive value of the test was 60% and the negative predictive value was 90% (chi-squared = 7, P = 0.001, odds ratio = 17, 95% confidence interval 2–179).
Conclusion. The penlight test for photophobia was found to be a useful test for distinguishing serious from benign eye pathology in a general practice population.
Keywords. Eye diseases, photophobia, primary health care.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Patients often present with a unilateral red eye in general practice, and the challenge lies in distinguishing benign from serious pathology. Although conjunctivitis is a common reason for encounter in general practice, serious pathology is rare, with the annual incidence of uveitis given as 12 per 100 000,1 and of ocular herpes as 20 per 100 000.2 One classification of eye conditions in primary care distinguishes ‘non-vision-threatening’ problems (including blepharitis, lid problems, dry eyes and conjunctivitis) from ‘vision-threatening’ problems (including corneal, uveal tract and anterior chamber problems) and states that photophobia is a symptom of ‘vision-threatening disease’.3 There are other published descriptions of photophobia as a common sign present in iritis and keratitis,4 but these reports do not provide data on the sensitivity and specificity of the sign. A report based on findings from a series of patients presenting to the emergency department of an eye hospital suggested that unilateral photophobia in the red eye is a good predictor of serious pathology.5 This study was conducted to test the value of the penlight test in the detection of serious eye disease in a general practice setting.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
All patients in a suburban community presenting to a GP with a red eye as the chief complaint (and not as an incidental finding) over a 1-year period were included in the study. One physician (KSP) conducted all examinations. The age, sex and chief complaint of each patient were recorded (Table 1
). On physical examination, the side of the red eye was noted, visual acuity was tested in each eye, and the penlight test was performed on each eye. A pocket flashlight (penlight) powered by two 1.5 V batteries was shone directly into one of the eyes from a distance of 15 cm for 2 s (‘the penlight test’). The presence of additional discomfort was recorded as a positive test. The test was then repeated for the other eye. A randomization list was used to decide which of the two eyes would be tested first for each patient. Patients then underwent slit-lamp examination for diagnosis. Diagnosis, medications prescribed and referrals were noted. Physical findings and treatments on follow-up visits were also recorded, but this analysis includes data only from the initial visit for each episode of illness. Benign pathology was defined as conjunctivitis (bacterial, viral or allergic), blepharitis, hordeolum, dry eye and corneal foreign body. Serious pathology was defined as uveitis, iridocyclitis, herpetic keratitis, corneal ulcer and corneal abscess. The association between photophobia and serious pathology was tested using the chi-square statistic. Data were entered and analysed using Epi-Info Version 6 software.
|
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Between June 1992 and July 1993, 2055 patients were registered with the GP conducting the study. The patients made 6576 visits. Thirty-two patients were seen with a chief complaint of ‘red eye’. These patients made 59 visits (0.9% of all visits) with a range of 1–5 visits per patient. The patients with red eye ranged in age from 8 to 90 years, with a mean age of 50 years and a median age of 48 years. Seventy-one percent of patients were female. Complete data for analysis were available for 36 visits made by 30 patients. Two patients with data missing from the medical record on the results of the penlight test or the final diagnosis were excluded from the analysis.
The penlight test was positive in 13 out of 36 examinations (36%) overall and was positive in eight out of 10 examinations (80%) in which serious pathology was later diagnosed [chi-square = 9, P = 0.001, odds ratio = 16.8, 95% confidence interval (CI) 2–171] (Table 2). The sensitivity of the test was 80% (CI = 44–95), the specificity 81% (CI = 60–93), the positive predictive value 62% (CI = 31–91) and the negative predictive value 91% (CI = 71–98). The positive likelihood ratio was 4.16 (CI = 1.78–9.69) and the negative likelihood ratio was 0.2476 (CI = 0.007–0.86). The order in which the eyes were examined did not affect the results. In only one patient was photophobia found in the non-red eye. One patient had bilateral red eyes. Six patients required further referral to ophthalmologists for treatment. There was no seasonal variation in the presentation of patients during the study period.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
The penlight test for photophobia was found to be a useful test for distinguishing serious eye pathology for patients with a red eye in a general practice population, given the finding of a sensitivity of 80% for this test. This study confirms the findings of Chong and Murray,5 who conducted their study in an eye hospital emergency department. A study among referred patients may suffer from selection bias, hence the need to replicate some studies in general practice. However, selection bias may also have affected the results of the current study as the GP conducting this study has special interest and training in ophthalmology and may see more cases of eye disease than do his partners in the practice.
An additional bias may be present because the same examiner conducted and interpreted the penlight test after taking an initial history (as part of normal office procedures) and might have already come to some informed conclusion about the diagnosis. A future study confirming the value of this test might employ an independent examiner to perform the penlight test and a second observer to make the clinical diagnosis as the ‘gold standard’.
The adequacy of training of GPs in the management of common eye diseases is a topic under current debate. In some settings, fundoscopic examination by GPs is rare, though this figure may be higher among those with specialist certification in family medicine.6 In The Netherlands, a continuing medical education programme has been implemented to improve the skills of the GP with the slit-lamp.7 It is hoped that this report will help to generate further study into simple methods that can improve the performance of GPs in the identification and management of eye diseases.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
1 Curl A, Matos K, Pavesia C. Acute anterior uveitis. Clin Evidence 2002; 7: 555–559.
2 Barker N. Ocular herpes simplex. Clin Evidence 2002; 7: 597–604.
3 Hara JH. The red eye: diagnosis and treatment. Am Fam Physician 1996; 54: 2423–2430.[Medline]
4 Michelson PE. Red eye unresponsive to treatment. West J Med 1997; 166: 145–147.[Medline]
5 Chong NV, Murray PI. Pen torch test in patients with unilateral red eye [letter]. Br J Gen Pract 1993; 43: 259.[Medline]
6 Gross R, Yuval D, Yaphe Y, Boerma W. The Role of the Primary Care Physician in Israel: Preliminary Findings from a National Survey. Research Report. JDC-Brookdale Institute; 1994.
7 Baggen JL, van Leeuwen YD. Red eyes. Fam Doctor 2000; 13: 4–6.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||