Abstract
There has been discussion regarding the definition of the clinical sign trachomatous trichiasis (TT) for the purposes of determining elimination of trachoma as a public health problem, and whether the definition should include the presence of trachomatous scarring (TS). A multi-country observational study was conducted in Ethiopia, Uganda and Nigeria to assess whether TS grading by field graders using the WHO simplified system in trachoma surveys are comparable with expert grading of tarsal conjunctival scarring (TCS) using a detailed system. The primary outcome was the proportion of eyes graded as "No TS" in the surveys but with TCS from expert photographic grading (the negative predictive value, NPV). In Ethiopia, Uganda and Nigeria, 545 (438 trichiasis cases and 107 comparisons), 256 (156 trichiasis cases and 100 comparisons), and 468 (352 trichiasis cases and 116 comparisons) participants, respectively, were enrolled. In Ethiopia, among 111 trichiatic eyes graded "No TS" in the surveys, 103 (92.8%) had TCS in expert photo grading, NPV 7.2% (95% CI 3.2%-13.7%). In Uganda, among 28 trichiatic eyes graded "No TS" in the surveys, 19 (67.9%) had TCS in expert photo grading, NPV 32.1% (95% CI 15.9%-52.4%). In Nigeria, among 111 trichiatic eyes graded "No TS" in the surveys, 100 (90.1%) had TCS in expert photo grading, NPV 9.9% (95% CI 5.0%-17.0%). Across settings, among eyes misdiagnosed as "No TS" in the survey, 174/250 (69.6%) had extensive TCS (patches of scarring occupying ≥1/3 of the upper tarsal conjunctiva). Trichiatic eyes with TCS had more severe entropion, trichiasis, conjunctival inflammation, and corneal opacity than those without TCS. In all three settings, including TS to define a trichiasis "trachomatous" in a survey could result in the underestimation of the burden of TT. However, TCS can be effectively used to determine TT severity and management.