Introduction

The Nepal National Trachoma Program (NTP) was initiated in 2002 with the goal of controlling trachoma in Nepal by 2014.[1] The WHO SAFE (Surgery, Antibiotics, Facial Cleanliness, Environmental Improvements) strategy was employed including mass drug administration (MDA) with azithromycin in 17 districts between 2002 and 2012. Population-based surveys of trachoma were performed by Nepal Netra Joyti Sangh (NNJS) workers to evaluate the Nepal program and to identify Districts where further treatment was recommended. Achham District, which was surveyed in 2006, was recognized as an area with the highest prevalence of TF in Nepal (> 20%) among children 1–9 years, triggering 3 MDAs distributed in 2008, 2009 and 2010 with over 500,000 treatments.[1] A follow-up survey in September 2012 determined TF prevalence in Achham was 5.9%, still the highest prevalence among Nepali Districts[1] triggering an additional round of treatment offered in 2014 in accordance with the WHO recommendations prior to our measurements.

We aimed to determine if control had been achieved in the clinical signs of trachoma by examination with certified graders (validated with photographs), in infection with Chlamydia trachomatis by PCR, and in antibody responses to C. trachomatis protein antigens pgp3 and CT694 using serological testing, through population-based cross-sectional measurements in Achham District in early 2014 after repeated rounds of drug administrations had been offered.

Methods

Community randomization

In Achham District, 12 VDCs and 2 communities per VDC were randomly selected (n = 24 communities) for the Examination arm where examination and PCR were performed in January 2014 (Fig 1). For more detailed assessment of disease prevalence, 2 VDCs (18 communities) which were identified as high trachoma-prevalence areas (TF > 20%) by NNJS in the September 2012 survey, were included in the High Prevalence arm where examination and PCR were performed in January 2014. Finally, when serological testing for C. trachomatis became available to us in May 2014, another VDC (with 3 communities) conveniently located was chosen for serology testing in addition to examination and PCR in the Serology arm. Randomization of the 12 VDCs in the Examination arm was performed using Microsoft Excel (Version 2010, RAND and Sort functions).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Participant flow.

https://doi.org/10.1371/journal.pntd.0004462.g001

A current population census was unavailable so household heads and guardians were asked to bring children to a communal space such as a neighborhood school, community center, or other area well known to villagers with open land where examination and sample collection could be performed. For each household included, the name of the household head was recorded along with the age and gender of each child aged 1–9 years. Written informed consent was obtained in the local language by a parent or guardian prior to examination and sample collection.

Results

The participant flow for the study in Achham District is shown in Fig 1. Twelve VDCs that met inclusion criteria were randomly selected and included in the Examination arm, two VDCs were included in the High Prevalence arm, and one VDC was included in the Serology arm. Baseline characteristics were similar in Examination, High-Prevalence and Serology arms (Table 1). The results for clinical trachoma examination, chlamydia trachomatis infection and antibodies against chlamydia trachomatis protein antigens in the Examination, High-Prevalence and Serology arms are shown in Table 2.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Baseline characteristics of communities in examination, high-prevalence and serology arms.

https://doi.org/10.1371/journal.pntd.0004462.t001

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Prevalence of trachoma, infection and antibodies against C. trachomatis protein antigens in examination, high-prevalence and serology arms.

Community means with 95% CIs.

https://doi.org/10.1371/journal.pntd.0004462.t002

Discussion

We found clinical trachoma, ocular chlamydia trachomatis infection by PCR and anti-chlamydia trachomatis antibody responses in 1–9 year olds have been controlled in Achham District as measured in this cross-sectional assessment in January and May 2014.[10,11] Serological testing for chlamydia trachomatis is a new alternative for trachoma surveillance and may be useful tool particularly in areas where chlamydia trachomatis and clinical trachoma are controlled.[8,9]

The First and Second meetings in 1997 and 1998 of the WHO Global Alliance for the Elimination of Trachoma suggested that national programs may be necessary to control trachoma as a public health problem.[12,13] The government of Nepal established the National Trachoma Program in 2002 and implemented MDAs with azithromycin in 17 districts thought to be at high risk for trachoma to achieve this goal. The trachoma prevalence estimate for Achham District was 5.9% in 2012 triggering an additional round of mass antibiotic treatment offered in 2014, prior to our measurements. The WHO recommendation recently updated should trigger reassessment of Achham District following treatment before additional treatments are offered.[14] This will maximize the effect of antibiotics in short supply and minimize adverse events.

The WHO advises hygiene and environmental improvements to prevent trachoma from returning after antibiotics are discontinued, but there is no evidence any particular intervention is beneficial. In this cross-sectional assessment of Achham District in 2014, we found clinical trachoma, chlamydia trachomatis infection, and antibodies against chlamydia trachomatis protein antigens were undetectable or present at only low levels. Trachoma is controlled in this district and even the loftier goal of local elimination of C. trachomatis infection has been achieved.

This study has limitations which may affect generalizability. The 66 VDCs were randomly selected in the Examination arm, but 2 VDCs in the High-Prevalence arm were selected based on a government survey from September 2012 introducing potential selection bias overestimating prevalence. Inaccurate grading is always a concern with clinical trachoma work. The graders in Nepal were probably different in the 2006 and 2012 assessments introducing variation. To maximize consistency, all graders who took part were certified with k ≥ 0.6 agreement with experienced graders prior to participating and all individuals examined in 2014 had a photograph taken that was subsequently audited by experienced graders. The study only included 1–9 year olds and adults may harbor disease that was missed. However, this is unlikely because the reservoir of trachoma is almost exclusively in preschool children shown in numerous studies.[9,15,16] The negative serology for chlamydia trachomatis was surprising in this previously endemic region, but chlamydia trachomatis antibodies might be detected in adults previously infected who were not sampled here.[9] Note the small serological sample size of 68 may limit generalizability.

Conjunctival samples were pooled into pools of five for PCR to save time and cost as previously described.[6,17] Serological samples were not pooled and were tested as individual samples. The pooling process may have interfered with detection of positive chlamydia trachomatis samples but any equivocal pools were separated and tested as individual samples making this less likely Pooling could in theory reduce the concentration below a detectability threshold, although we have unpooled 100s of pools from previous studies and found that a negative pool with a positive individual sample is no more common than an individual test being positive and then negative when retested—that is, the discrepancy is within the repeatability limit of the test, rather than a dilutional effect.[3,4,18,19,20,21] Also, we’ve been just as likely to find a positive pool and a negative individual test.

Some researchers and program managers believe that children in disadvantaged families are more likely to harbor infection but less likely to be brought for examination and treatment, resulting in an underestimation of trachoma prevalence. However, in a study in Niger where trachoma is highly endemic, we found preschool children who have ocular C. trachomatis have higher odds of presenting early for examination than those who are not infected.[7] Inclusion criteria for VDCs was accessibility less than 2 hours walk from a vehicle-accessible road. This may have introduced bias if less accessible areas are more or less likely to harbor trachoma.

There are various reasons which explain the marked drop in trachoma in Achham District from 2006 to 2014. The Nepal Trachoma Program distributed mass antibiotics in 2008, 2009, and 2010 and implemented face washing and environmental improvements during that time as well. We are cautious in attributing the reduction in trachoma in western Nepal solely to the National Program and must consider other possibilities. In a previous study in 52 villages in Kailali and Konchapur Districts adjacent to Achham District in Western Nepal from 1998 to 2001, we observed that 21% of the reduction in trachoma could be attributed to secular trends rather than programmatic antibiotic distributions.[22] Secular trends in trachoma have also been seen in the Gambia, Malawi, and European countries even without antibiotic distributions.[23,24,25] Much of the reduction of trachoma in Nepal may be due to a secular trend in the area rather than a specific effect of the trachoma control program.

Control of trachoma has been the WHO goal, as the prevalence of blindness is probably negligible when the clinical signs of infection are brought to a low level. Currently, the WHO aim for control is less than 5% district prevalence of TF in children.[26] Achham Nepal has achieved this target. In fact, we found no evidence of current infection and minimal evidence of recent infection by serology. Thus, true elimination of infection may have been achieved here and might be a reasonable goal in similar areas. Other studies have shown significant reduction and elimination of clinical trachoma and chlamydia trachomatis infection in one or even in multiple communities.[27,28,29,30] However, trachoma may return to previously affected communities by contamination from adjacent areas or from migration when antibiotics are discontinued.[3,31]. Here we demonstrate results consistent with absence of signs of clinical trachoma (verified with photographs), chlamydia trachomatis infection (Amplicor PCR), and antibodies against chlamydia trachomatis protein antigens (Luminex) in an entire district. In once severely affected districts like this, elimination of infection may be a realistic goal—not merely control of the clinical signs of trachoma to a low level.

Supporting Information

Acknowledgments

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention. We thank Seth Bloomberg, MD, PhD who randomized the 12 VDCs in the Examination arm and the following field workers who helped with data collection: Bhaloni Singh Bist, Binod Kunwar, Birkha Bogati, Byashmani Rana, Dev Raj Paneru, Janak Kumar Bist, Siddha Raj Bist, Suraj Baldar.