Screening for sight-threatening diabetic retinopathy: comparison of fundus photography with automated color contrast threshold test

Purpose: To compare tritan contrast threshold (TCT) with fundus photography in screening for sight-threatening diabetic retinopathy (STDR), before significant visual loss. Design Prospective, comparative study. 

Methods: A total of 510 consenting diabetic patients attending a hospital-based photographic screening clinic were recruited over a 2-year period. Exclusion criteria included visual acuity of worse than 6/9, previous photocoagulation, and a history of previous eye disease known to affect color vision. The automated TCT test was performed using a computerized cathode ray tube-based technique. Retinal photography was performed using a Polaroid mydriatic fundus camera with a 45 degrees field. Grading of diabetic retinopathy was carried out by an ophthalmologist using slit-lamp biomicroscopy and a 78-diopters lens. Assessments of sensitivity, specificity, and predictive values for both fundus photography and the TCT test were made. 

Results: Both the fundus photography and TCT test correlated significantly with the presence of STDR (P < .0001, χ² test). The TCT test yielded a sensitivity of 94% (95% confidence interval [CI], 73%- 100%) and a specificity of 95% (95% CI, 92%- 96%) for detection of STDR compared with a sensitivity of 88% (95% CI, 66%-97%) and a specificity of 95% (95% CI, 93%-97%) with fundus photography. Combined modality improved overall screening performance. 

Conclusion: The TCT assessment is an effective and clinically viable technique, in comparison with fundus photography, to screen for STDR among a diabetic population. Additionally, our results also showed that combining the TCT test with fundus photography greatly increases the performance of screening for STDR.