Onchocerciasis transmission in Ghana: persistence under different control strategies and the role of the simuliid vectors

Abstract

Background: The World Health Organization (WHO) aims at eliminating onchocerciasis by 2020 in selected African countries. Current control focuses on community-directed treatment with ivermectin (CDTI). In Ghana, persistent transmission has been reported despite long-term control. We present spatial and temporal patterns of onchocerciasis transmission in relation to ivermectin treatment history.

Methodology/principal findings: Host-seeking and ovipositing blackflies were collected from seven villages in four regions of Ghana with 3-24 years of CDTI at the time of sampling. A total of 16,443 flies was analysed for infection; 5,812 (35.3%) were dissected for parity (26.9% parous). Heads and thoraces of 12,196 flies were dissected for Onchocerca spp. and DNA from 11,122 abdomens was amplified using Onchocerca primers. A total of 463 larvae (0.03 larvae/fly) from 97 (0.6%) infected and 62 (0.4%) infective flies was recorded; 258 abdomens (2.3%) were positive for Onchocerca DNA. Infections (all were O. volvulus) were more likely to be detected in ovipositing flies. Transmission occurred, mostly in the wet season, at Gyankobaa and Bosomase, with transmission potentials of, respectively, 86 and 422 L3/person/month after 3 and 6 years of CDTI. The numbers of L3/1,000 parous flies at these villages were over 100 times the WHO threshold of one L3/1,000 for transmission control. Vector species influenced transmission parameters. At Asubende, the number of L3/1,000 ovipositing flies (1.4, 95% CI = 0-4) also just exceeded the threshold despite extensive vector control and 24 years of ivermectin distribution, but there were no infective larvae in host-seeking flies.

Conclusions/significance: Despite repeated ivermectin treatment, evidence of O. volvulus transmission was documented in all seven villages and above the WHO threshold in two. Vector species influences transmission through biting and parous rates and vector competence, and should be included in transmission models. Oviposition traps could augment vector collector methods for monitoring and surveillance.

Fig 1. Maps showing the location of Ghana (A), the boundaries and start dates for the Onchocerciasis Control Programme (OCP) phases (B), and the seven Ghanaian study sites (C).

The OCP began vector control operations across West Africa in 1975. Asubende received vector control from 1986, which was interrupted several times during 1987–1989 because of community trials of the impact of ivermectin mass treatment on transmission and microfilarial loads. At the time of closure of the OCP in 2002, the Asubende focus was incorporated into a special intervention zone (SIZ) due to on-going transmission. The breeding sites at Asukawkaw Ferry, Dodi Papase and Pillar 83 were first treated with larviciding insecticides during OCP experimental campaigns (reinvasion studies) in 1981 (see Fig 2 of Cheke & Garms 1983 [94]), before becoming part of the South-eastern extension, which reached these river basins when it became fully operational in February 1988.

Fig 2. Methods used to obtain host-independent (A, B) and host-dependent (C, D) adult female blackfly samples.

(A) Bellec (sticky) trap situated above rapids; (B) Monk’s Wood (light) trap placed near presumed breeding sites; (C) human-baited tent; (D) cow-baited tent. A and B illustrate traps to collect ovipositing flies; C and D depict methods to obtain host-seeking flies.

Fig 3. Proportions of infected and infective flies by season and sampling method for Bosomase and Gyankobaa.

Data and error bars are as in Fig 3, but excluding the 2006 Bellec-caught flies collected at Bosomase, as during the pilot study comparisons with other fly collection methods were not conducted.

Fig 4. Proportions of infected and infective flies (assessed by heads and thoraces dissection) in study villages and years of ivermectin treatment.

Infected flies (green bars) are those with any larval stage of Onchocerca volvulus; infective flies (blue bars) are those harbouring L3 larvae in heads and/or thoraces. Error bars are exact 95% confidence intervals. The results for Bosomase include the Bellec-caught flies obtained during the pilot study conducted at Bosomase in January–February 2006.

Fig 5. Therapeutic coverage of ivermectin treatment in all study villages.

The plots show the percentage of the overall population treated at each ivermectin round since mass ivermectin distribution began: (A) Asubende and (B) Agborlekame in the Brong-Ahafo Region; (C) Asukawkaw Ferry, (D) Dodi Papase and (E) Pillar 83 in the Volta Region; (F) Bosomase in the Western Region, and (G) Gyankobaa in the Ashanti Region. The dashed lines are the best fit least squares polynomial functions to the data, presented to facilitate visual inspection of the coverage trends. Biannual ivermectin distribution started in 2009 in Asubende, Agborlekame and Gyankobaa, whilst annual distribution has continued in the remaining villages.

Fig 6. Numbers of L3 larvae per 1,000 parous (host-seeking) flies or per 1,000 ovipositing flies.

Data are presented for the three villages where infective Onchocerca volvulus larvae were detected during the wet and/or dry seasons of 2009 to 2011. In all cases, except Asubende parous flies, the number of (A) O. volvulus L3 larvae per 1,000 parous or (B) O. volvulus L3 larvae per 1,000 ovipositing flies exceeds the World Health Organization (WHO) threshold for one L3 per 1,000 (parous) flies. In Asubende the value was 1.4 L3 per 1,000 ovipositing flies. Error bars are 95% CI calculated using percentile bootstrapping.