English

Abstract

BACKGROUND A tertiary healthcare set up receives an increasing load of patients with complaints of diminished vision on a daily basis. It can be quite cumbersome for conducting automated refractometry in each of them, on the suspicion of refractive error being the cause of vision loss. Few studies have shown a simple visual acuity comparison of individual eyes with and without pin-hole can determine if the patient has significant refractive error. There are minimal studies done to determine the reliability of such use of pin-hole, and even less of such studies are done in our country. The purpose of this research was to calculate the sensitivity and specificity of significant visual acuity change on using pin-hole in identifying refractive errors. METHODS Patients were selected in the age group of 15 years to 55 years with chief complaints of blurred / diminished / low vision. Patients with known history of diabetes, hypertension, trauma to eyes, and neurological/psychological complications were excluded from the study. Log MAR chart was used to measure the visual acuity in the selected subjects. Visual acuity was recorded for individual eye in each patient; first without, and then with pin-hole. An improvement in vision by at least one Log MAR scale was taken into consideration. Such patients were then subjected to autorefractometry without use of any cycloplegic drugs. A spherical refractive value of ± 1.00 D was considered significant. RESULTS Pin-hole testing has a sensitivity of 94.17 percent (95 % CI: 92.61 to 95.48) and specificity of 89.95 percent (95 % CI: 84.75 to 93.84) when compared with autorefractometry finding for identification of refractive errors (p < 0.0001). The test has an accuracy level of 93.55 per cent (95 % CI: 92.06 to 94.83). CONCLUSIONS The accuracy of pin-hole vision comparison with plain sight in diagnosing refractive errors as the cause of vision loss is reliable enough for it to be used in prompt identification of such cases. KEY WORDS Pin-Hole, Refraction, Vision