ABSTRACT
BACKGROUND: Trachoma results from repeated episodes of conjunctival infection with Chlamydia trachomatis and is the leading infectious cause of blindness. To eliminate trachoma, control programmes use the SAFE strategy (Surgery, Antibiotics, Face cleanliness, and Environmental improvement). The A component is designed to treat C trachomatis infection, and is initiated on the basis of the prevalence of the clinical sign trachomatous inflammation-follicular (TF). Unfortunately, TF correlates poorly with C trachomatis infection. We sought to assess a newly developed point-of-care (POC) assay compared with presence of TF for guiding the use of antibiotics for trachoma control. METHODS: We compared performance outcomes of the POC assay and presence of TF using commercial PCR as a comparator in 664 children aged 1-9 years in remote, trachoma-endemic villages in Tanzania. Signs of trachoma were graded according to the WHO simplified trachoma grading system. FINDINGS: Of 664 participants, 128 (19%) were positive for ocular C trachomatis infection by PCR. Presence of TF had a sensitivity of 64.1% (95% CI 55.8-72.4), specificity of 80.2% (76.8-83.6), and positive predictive value of 43.6% (36.5-50.7). By contrast, the POC assay had a sensitivity of 83.6% (77.2-90.0), specificity of 99.4% (98.8-100.0), and positive predictive value of 97.3% (94.2-100.3). Interagreements and intra-agreements between four novice operators were 0.988 (0.973-1.000) and 0.950 (0.894-1.000), respectively. INTERPRETATION: The POC assay is substantially more accurate than TF prevalence in identifying the presence or absence of infection. Additional studies should assess the use of the assay in the planning and monitoring of trachoma control activities.
