Abstract
Purpose: :
Trachoma is the leading cause of infectious blindness worldwide. Current guidelines recommend that areas with >10% clinically active trachoma receive three annual mass azithromycin treatments before reassessment for further treatment. However, the relationship between clinical activity and laboratory evidence of chlamydial infection is unclear. This study evaluates the relationship between the prevalence of clinical activity and chlamydial infection in a community, and assesses the time course of clinical resolution after antibiotic treatment.
Methods: :
Four biannual mass azithromycin treatments were administered to 24 randomly selected villages in a region of Ethiopia with hyperendemic trachoma. Before each treatment, the prevalence of clinical activity and infection was calculated for each village. Clinical activity was defined as TF or TI in the WHO Simplified Grading Scale, and infection was determined by chlamydial PCR of conjunctival swabbings. The correlation coefficient between the prevalence of clinical activity and infection in each village was calculated before and after antibiotic treatment. A generalized estimating equation (GEE) with an autoregressive error structure was used to predict the prevalence of clinical activity over time, in order to determine the rate of clinical resolution.
Results: :
The mean prevalence of clinical activity was 86.0% before antibiotic treatment, which decreased gradually to 39.2% at 24 months after the initial treatment. The mean prevalence of chlamydial infection was 52.9% before antibiotic treatment, which decreased to 5.6% by two months after the first treatment, and remained <8.4% for the remainder of the study (although individual villages ranged from 0% to 42% after treatment). The pre-treatment correlation coefficient of clinical activity and infection was r=0.75, which decreased to r=0.15 at 24 months (i.e., 6 months after the last treatment). Additional follow-up was completed for 16 villages, and by 42 months (i.e., 24 months after the last treatment), the correlation coefficient was r=0.75. We estimate that the prevalence of clinical activity decreased 33.0% per year (95% confidence interval 29.3 to 36.5% per year) in the presence of minimal amounts of infection (and in some cases unidentifiable infection).
Conclusions: :
Clinical activity is correlated with ocular chlamydial infection prior to mass antibiotic treatment, but is less strongly correlated after mass treatment. After mass treatments have been discontinued, clinical activity and infection again become correlated.
Clinical Trial: :
www.clinicaltrials.gov NCT00221364
Keywords: trachoma • clinical (human) or epidemiologic studies: outcomes/complications • antibiotics/antifungals/antiparasitics