The Relationship between Active Trachoma and Ocular Chlamydia trachomatis Infection before and after Mass Antibiotic Treatment

Abstract

Background: Trachoma is a blinding disease, initiated in early childhood by repeated conjunctival infection with the obligate intracellular bacterium Chlamydia trachomatis. The population prevalence of the clinical signs of active trachoma; ''follicular conjunctivitis" (TF) and/or ''intense papillary inflammation" (TI), guide programmatic decisions regarding the initiation and cessation of mass drug administration (MDA). However, the persistence of TF following resolution of infection at both the individual and population level raises concerns over the suitability of this clinical sign as a marker for C. trachomatis infection.

Methodology/principle findings: We systematically reviewed the literature for population-based studies and those including randomly selected individuals, which reported the prevalence of the clinical signs of active trachoma and ocular C. trachomatis infection by nucleic acid amplification test. We performed a meta-analysis to assess the relationship between active trachoma and C. trachomatis infection before and after MDA. TF and C. trachomatis infection were strongly correlated prior to MDA (r = 0.92, 95%CI 0.83 to 0.96, p<0.0001); the relationship was similar when the analysis was limited to children. A moderate correlation was found between TI and prevalence of infection. Following MDA, the relationship between TF and infection prevalence was weaker (r = 0.60, 95%CI 0.25 to 0.81, p = 0.003) and there was no correlation between TI and C. trachomatis infection.

Conclusions/significance: Prior to MDA, TF is a good indicator of the community prevalence of C. trachomatis infection. Following MDA, the prevalence of TF tends to overestimate the underlying infection prevalence. In order to prevent unnecessary additional rounds of MDA and to accurately ascertain when elimination goals have been reached, a cost-effective test for C. trachomatis that can be administered in low-resource settings remains desirable.

Fig 1. The natural history of an episode of ocular C. trachomatis infection and the associated conjunctival inflammatory response.

Trachomatous Inflammation–Follicular (TF): the presence of five or more follicles in the upper tarsal conjunctiva. Trachomatous Inflammation–Intense (TI): pronounced inflammatory thickening of the upper tarsal conjunctiva that obscures more than half of the normal deep tarsal vessels [2].

Fig 2. The relationship between the prevalence of disease signs and infection before and after the introduction of MDA.

(a) Community prevalence of TF (or TF/TI) vs. the community prevalence of C. trachomatis infection before the introduction of MDA. (b) Community prevalence of TI vs. the community prevalence of C. trachomatis infection before the introduction of MDA. (c) Community prevalence of TF (or TF/TI) vs. the community prevalence of C. trachomatis infection after the introduction of MDA. (d) Community prevalence of TI vs. the community prevalence of C. trachomatis infection after the introduction of MDA. (e) Community prevalence of TF (or TF/TI) vs. the community prevalence of C. trachomatis infection after the introduction of MDA, showing only the communities with less than 15% TF. Data from population-based studies, summarised in Tables 1 and 2. The size of the circles reflects the sample size. Line fitted by linear regression, weighted by the size of the studies.

Fig 3. The relationship between the individual level presence of Active Trachoma (TF or TF/TI) and the detection of C. trachomatis infection by community TF prevalence before the introduction of MDA.

(a) Sensitivity of TF for infection. (b) Specificity of TF for infection. (c) Positive Predictive Value (PPV) of TF for C. trachomatis infection. (d) Negative Predictive Value (NPV) of TF for C. trachomatis infection. Data from population-based studies, summarised in Table 1. The size of the circles reflects the sample size. Line fitted by linear regression, weighted by the size of the studies, except for the PPV which was fitted by polynomial.

Fig 4

A Forest plot showing the relationship between Active Trachoma (TF or TF/TI) and the detection of C. trachomatis infection at the individual level (a) before and (b) after the introduction of MDA, grouped by community TF prevalence level. Studies are ordered by increasing prevalence of TF, the size of the grey boxes represent the how much weight each study contributes to the overall estimate, the blue diamonds represent the subtotal and overall pooled odds ratio estimates. The odds ratios are for ocular C. trachomatis infection as the outcome and TF as the explanatory variable.

Fig 5

Sensitivity versus specificity of each study, using TF to diagnose C. trachomatis infection at the individual level (a) before and (b) after the introduction of MDA. Each circle represents the estimate for a single study, with the size of the circle representing the size of the study. The red square is the estimated pooled sensitivity and specificity for all studies (Pre- or Post-MDA). The orange dashed line represents the 95% CI (or confidence region in 2 dimensions) for the sensitivity and specificity. The grey curve (hierarchical summary receiver operating characteristic (HSROC) curve) represents the estimated relationship between sensitivity and specificity in these studies, with the grey dashed line indicating the region in which we would expect 95% of studies to fall.

Fig 6. The relationship between the individual level presence of Active Trachoma (TF or TF/TI) and the detection of C. trachomatis infection by community TF prevalence after the introduction of MDA.

(a) Sensitivity of TF for infection. (b) Specificity of TF for infection. (c) Positive Predictive Value (PPV) of TF for C. trachomatis infection. (d) Negative Predictive Value (NPV) of TF for C. trachomatis infection. Data from 6 population-based studies, summarised in Table 2. The size of the circles reflects the sample size.