BMC Medicine BioMed Central

Background: Onchocerciasis is a leading cause of blindness worldwide, hence elimination of the infection is an important health priority. Community-based treatment programs with ivermectin form the basis of control programs for the disease in Latin America. The long-term administration of ivermectin could eliminate Onchocerca volvulus infection from endemic areas in Latin America.


Background
Onchocerciasis is an important blindness-causing disease resulting from infection by the filarial parasite Onchocerca volvulus, transmitted by Simulium blackflies. An estimated 18 million persons are infected worldwide [1], and approximately 770000 are blind or visually impaired as a consequence of the infection [1]. The infection also causes severe and troublesome skin disease [2] and is associated directly with increased mortality [3].
Ivermectin (Mectizan ® ) is an oral microfilaricidal drug that is safe and effective for mass treatment of onchocerciasis [4] and is provided free to endemic countries by Merck & Co. The Onchocerciasis Elimination Program for the Americas (OEPA) has coordinated the distribution of ivermectin since 1991 in the six endemic countries in Latin America (Brazil, Colombia, Ecuador, Guatemala, Mexico and Venezuela) with two principal goals: to eliminate morbidity caused by onchocerciasis by 2007 and to eliminate the infection where possible using a strategy of mass treatment of endemic communities with ivermectin [5].
Elimination is a high priority to prevent new morbidity, to allow the reallocation of limited health and financial resources to other health priorities, and because of the potential for the development of parasite resistance to ivermectin in the long term [6]. The development of resistance to ivermectin by O. volvulus is a serious concern because there are no safe or practical alternative treatments currently available for mass distribution in human populations. It has been suggested that the infection could be eliminated from the Americas, through twiceannual treatments with ivermectin to greater than 85% of the eligible population at risk for onchocerciasis over a period of 10-15 years, via the interruption of the transmission of infection and death of adult parasites by senescence [7].
In Ecuador, the National Program for the Elimination of Onchocerciasis has pursued a program of mass treatments with ivermectin in endemic communities over the past 14 years (1990-2003) and has achieved high rates of coverage. The Ecuadorian Program provides an important model to evaluate whether elimination of O. volvulus infection is a realistic aim for an endemic country in the Americas using an ivermectin-based treatment strategy.

Study area
The province of Esmeraldas, latitude 1°96'S to 1°43'N and longitude 78°48'W to 80°9'W, is situated west of the Andes in North-Western Ecuador [8]. The principal endemic area of onchocerciasis is located in the District of Eloy Alfaro (Figure 1). The Santiago basin focus is formed Geographic locations Figure 1 Geographic locations. (A) Geographic location of endemic area for onchocerciasis in Ecuador. The endemic focus is located in the Northern coastal area of Ecuador adjacent to Colombia. There are a total of 119 endemic communities for onchocerciasis located in Esmeraldas Province (117 communities) and Pichincha Province (two communities). The endemic area consists of hyperendemic (red circles), mesoendemic (yellow squares), and hypoendemic (green triangles) communities. There are six satellite foci (Tuluví, Verde, Viche, Canandé, Sucio and Santo Domingo de Los Colorados). The areas in green represent the foci of onchocerciasis. (B) Geographic location of the seven sentinel communities in the District of Eloy Alfaro in Esmeraldas Province. Hyperendemic (red circles), and mesoendemic (yellow square) sentinel communities are shown. The areas in green represent the foci of onchocerciasis.
by the union of three rivers, the Río Santiago, Río Cayapas and Río Onzole and their respective tributaries, each forming a separate sub-focus. Satellite endemic foci are located on separate rivers caused by the migration of microfilariae-positive individuals from the principal endemic focus [9]. Five satellite foci are found in the province of Esmeraldas, located on the rivers Río Canandé, Río Verde, Río Viche, Río Sucio, and Río Tuluví. A small focus is also found in the neighboring province of Pichincha near Santo Domingo de los Colorados. A total of 119 communities are considered endemic for onchocerciasis.

Ivermectin distribution
The National Program for Onchocerciasis Elimination in Ecuador chose to use ivermectin as the disease control strategy through mass drug distribution in endemic communities since 1990. Before 1990, control relied largely on periodic nodulectomy campaigns in the endemic areas [10]. The distribution of ivermectin was started in endemic communities over the period 1990 to 1993 although treatment was delayed in a few until 1997. All consenting eligible inhabitants (approximately 19420) of the 119 communities have been treated for up to 14 years (1990-2003). Eligibility criteria for treatment were: weight greater than 15 kg, women not pregnant or nursing infants up to 3 months of age, and free of serious illness (e.g. active tuberculosis, terminal cancer etc). Ivermectin is distributed by specially-trained primary health care workers in each community at a dose of 150 µg/mg using standard exclusion criteria [11]. The population census is used as the basis for drug distribution. Health workers are responsible for health education, updating the census annually, identifying those not eligible for treatment and treating all eligible individuals.
Ivermectin has been distributed annually to endemic communities in the Río Cayapas since 1991-1992 except for seven communities in the hyperendemic area of the Río Cayapas sub-focus that received their first mass treatment between 1995 and 1996. A second annual dose of ivermectin was provided to hyperendemic communities in the Río Cayapas over the periods 1992-1994 and 1998-2003. A second dose was not given in the years 1995-1997 due to financial restraints on the program and a decision to fall in line with the strategy of annual treatments adopted by the African Program for Onchocerciasis Control (APOC) [7]. However, following the finding that twice-annual treatments could suppress onchocerciasis transmission in the Santiago sub-focus [11], the strategy of twice-annual treatments in hyperendemic communities was reinitiated from 1998. All endemic communities have been treated twice annually since 2001. The number of treatments and the start of ivermectin mass distribution provided to each river sub-focus and the satellite foci according to endemicity of infection are provided in Table  1.

Sentinel communities
The use of sentinel communities has been recommended for the in-depth evaluation of ivermectin distribution programs [1]. To evaluate the impact of mass distribution of ivermectin, seven communities were selected for periodic prospective in-depth epidemiological surveys. The seven sentinel communities were located in the following areas (3) mesoendemic area -satellite focus in Río Canandé: Naranjal de los Chachis community. Sentinel communities in the principal hyper-endemic sub-foci of the Río Santiago and Río Cayapas were chosen to evaluate the potential for reduction in transmission of infection through ivermectin treatment. A meso-endemic community in the Río Canandé was included because of large changes in prevalence that occurred over the period 1980-1986 and the high infectivity and human bite rates in this satellite focus [9]. Evaluations were performed before the start of ivermectin treatment in 1989 and in 2000 and 2004. The period from the last treatment with ivermectin provided in 2003 to the evaluations performed in 2004 was approximately 12 months. Parasitological ophthalmological, and entomological parameters were evaluated using standardized methodology [12]. Selection for inclusion in the parasitological and opthalmological assessments were: (1) parasitological -all individuals in the community census aged 1 year or greater, and (2) ophthalmological -all individuals in the census aged 10 years or greater. The study was approved by the Ethics Committee of the Hospital Vozandes, Quito, Ecuador.
Informed verbal consent to participate in the study was obtained through meetings of community elders and the individual participants.

Prevalence of infection and microfilarial load
Skin snips were taken from both iliac crests using a Stolz corneoscleral punch and weighed. The biopsies were placed in micro-well plates with phosphate buffered saline, and emergent microfilariae were counted microscopically after 24 hours. Microfilarial densities are expressed as the number of microfilariae per milligram of skin (mf/mg). Mean microfilarial load (MMFL) was calculated as the geometric mean of infection intensities of all members of the community including skin-snip negatives (i.e. e Sum(loge[x+1])/N-1 , where x is individual microfilarial loads and N is the number of individuals examined).

Ophthalmology
Ocular examinations included assessment of visual acuity, slit lamp examination of the anterior segment, and direct and indirect ophthalmoscopy of the posterior segment [13]. All examinations were performed by two experienced ophthalmologists (JR and RP) using the modified rapid ocular evaluation criteria recommended by the Pan American Health Organization [14].

Entomology
Simulium blackflies were collected in the seven sentinel communities during the seasonal period of maximum biting in 1995 in the sentinel communities of Tigre and San Miguel and in all sentinel communities in 2000 and 2004. Standard procedures were followed for blackfly collections [15]. Blackflies were collected in 99% alcohol and identified using a taxonomic key of Simulium [16]. O. volvulus DNA was amplified by PCR using primers for the O. volvulus-specific DNA sequence O-150 [17] in pools of 50 blackflies, and the O. volvulus-specific PCR products were detected by ELISA as reported previously [17]. The prevalence of infection was calculated using the algorithms contained in Poolscreen 2.0 (The University of Alabama, Birmingham, Alabama, USA), a statistical program that estimates the prevalence of infection in the vector population based upon the proportion of positive pools [18].

Ivermectin treatments and treatment coverage in all endemic communities
The  (Table 1). Average coverage per treatment round of ivermectin was high (>83%) in all endemic foci ( Table 1).

Characteristics of sentinel communities
The population eligible for treatment with ivermectin, number of mass treatments, treatment strategy by year (single vs. twice annual), and treatment coverage are shown in Table 2. The age and sex distribution of the sentinel communities were similar between the three surveys. The number of treatment rounds was not uniform for all study communities. Corriente Grande, Río Cayapas, received a total of 21 ivermectin treatments while El Tigre and San Miguel on the same river received 13 treatments. The average treatment coverage was ≥ 85% except for Guayabal (Río Santiago) and Naranjal (Río Canandé).   The rationale for the current ivermectin control strategy is derived from studies conducted in Guatemala that indicated that twice-annual distribution of ivermectin to all eligible persons could suppress transmission [19,20], and the results of simulation studies with a mathematical model that indicated that suppression of transmission  The prevalence of ocular onchocerciasis of the anterior segment is provided by punctate keratitis (PK) and presence of microfilariae in the anterior chamber (MfAC).  could be achieved by treatment of at least 85% of the eligible population [7]. If suppression of transmission can be maintained in an isolated focus for the reproductive life span of adult females, estimated at 13-14 years [21], the adult parasite population would die by senescence [22] and the infection would be eliminated.

Impact of ivermectin: O. volvulus infection prevalence and MMFL
We have demonstrated previously that an operational program based on ivermectin distribution twice annually over a period of 5 years in the Río Santiago sub-focus appeared to completely suppress O. volvulus transmission [11]. We have now extended these findings to show that transmission of O. volvulus in the Río Santiago might have been interrupted for at least 10 years (1995)(1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004). There is no evidence of new onchocercal ocular morbidity in the Río Santiago and dermal microfilariae have been absent in the three sentinel communities since 1996 [11]. Further, studies to detect IgG4 antibodies against the O. volvulus antigen OV16 using immunochromatic cards that are highly sensitive and specific for O. volvulus infection [23] were negative in 112 children aged less than 11 years in 2001 (i.e. children born since 1990 when the first dose of ivermectin was administered; Rumbea Guzman et al, unpublished data) providing further evidence for an interruption of transmission. These data together indicate that O. volvulus infection, after at least 13 years of continuous annual or twice-annual mass treatments with ivermectin, has now been eliminated from the Río Santiago subfocus.
O. volvulus infection remains active in the hyperendemic communities in the Río Cayapas. Individuals with dermal microfilariae were detected in three sentinel communities (total of 5 individuals) during the 2004 survey and infected blackflies were detected in two sentinel communities. Two of the three communities had received fewer ivermectin treatments than other sentinel communities having started to receive treatment in 1996 (Table 2) All of the hyperendemic communities in the Río Cayapas have received at least 5 years of twice-annual treatments with ivermectin (Table 1) and the finding of active infection is in contrast to our previous findings of the possible complete suppression of infection in the Río Santiago after 5 years of twice-annual treatments [11] indicating possible differences in the minimum effective duration of ivermectin mass treatments required to suppress transmission between the two infection sub-foci. There are three possible explanations for this difference: (1) the predominant vector in the Río Cayapas sentinel communities, S. exiguum, has a higher vectorial capacity than S. quadrivittatum that predominates in the Río Santiago [24,25]; (2) the same infected individuals might refuse treatment repeatedly, or are ineligible for long periods (e.g. through repeated pregnancies), and remain an active source of transmission of O. volvulus or temporary migrations of infected individuals in and out the Río Cayapas and the failure of them to receive repeated treatments could maintain the infection reservoir (this appears an unlikely explanation as all these individuals found positive had received significant numbers of ivermectin treatments -10/13, 6/ 13, 12/13, 13/13, and 11/21); and (3) a sub-population of adult female O. volvulus worms appear to respond poorly to the fertility suppressant effects of ivermectin, and this effect is observed after repeated doses [26,27].
The failure to suppress infection in the Río Cayapas might indicate the need for a change in treatment strategy. Studies conducted in Cameroon and Guatemala show that treatment every 3 months with standard doses of ivermectin is safe and well-tolerated and has significant macrofilaricidal effects against adult O. volvulus [28,29]. In fact, histological studies of onchocercal nodules from Ecuador from individuals with a mean of only 8.2 treatments with ivermectin have shown dramatic effects on adult female viability and fertility and also marked reductions in the frequencies of male worms [30], and provided a histological picture similar to that reported in nodules collected in the Onchocerciasis Control Program area of West Africa after 9-10 years of vector control [30]. These observations indicate that the period for which treatment must be maintained after suppression of infection in a community is achieved could be less than previously estimated if adult worms die or cannot maintain sexual reproduction earlier than previously estimated. It is possible that an intensive program of increased frequency of ivermectin treatment (i.e. 3-monthly) in combination with a course of doxycycline that can sterilize adult females worms [31] in skin test positive individuals and the treatment of repeated treatment refusers and migrants could yield rapid results in the suppression of transmission of O. volvulus.
The plan for certification of the elimination of onchocerciasis envisaged by OEPA comprises four phases [15]: phase I -treatment coverage of greater than 85% of the population with twice-annual ivermectin treatments for a period of 2 Our previous observations in the Río Santiago sub-focus [11] would support the time periods provided for phases I and II, but to achieve phase I in the hyperendemic area of the Río Cayapas sub-focus is likely to require much longer with the current twice-annual treatment strategy. It is likely that the periods required for each phase will vary according to the efficiency of the vector species and other factors unique to each geographic sub-focus (e.g. treatment coverage, migration patterns). Our data indicate that the Río Santiago sub-focus might be ready for the third phase of pre-certification. Pre-certification and suspension of ivermectin treatment could be considered also in Río Onzole sub-focus and satellite foci where annual or semi-annual treatments have been provided continuously for at least 13 years.
Before suspension of treatment, a survey of all endemic communities within each sub-focus should be conducted to ensure that the findings in the sentinel communities are generalizeable to non-sentinel communities. These surveys should include more sensitive diagnostic methods including the detection of O. volvulus DNA in skin snips [32] and specific antibodies using the OV16 immunochromatic cards [23,33] to exclude very low-level infections and histologic analysis of the viability and fertility of female worms in onchocercal nodules (if present). Detailed vectors studies might be required to examine the flight range and dispersal patterns of Simulium blackflies from communities in the Río Cayapas where transmission persists because of the potential risk of dispersal of infected Simulium. Also required will be a study of migration to determine if there are significant migrations of potentially infected individuals from the hyperendemic communities in the Río Cayapas to the Río Santiago and vice versa. Pre-certification of these foci of infection could reduce the cost of current control activities by approximately 75% (from US $250000 to just over US $60000) and release resources for an intensive elimination strategy in the Río Cayapas sub-focus.

Conclusion
The National Program for Onchocerciasis Elimination in Ecuador has pioneered the use of community health workers to distribute ivermectin and the integration of ivermectin distribution within primary health care activities. The diversification of the activities of the control team and the integration of control within primary health care services has become increasingly important in maintaining high rates of coverage with ivermectin as the public health threat posed by onchocerciasis is perceived to be less and less important by the inhabitants of endemic communities. The data from this study provide evidence that an operational program in Ecuador might have eliminated ocular onchocerciasis as a public health problem and has made significant progress towards the goal of the elimination of the infection from Ecuador, and provides a model control program for the evaluation of the potential for the elimination of onchocerciasis from the Americas. The elimination of onchocerciasis from Latin America and Africa is an important health priority and the data from the current study provides evidence that an ivermectinbased treatment program could be a feasible strategy for the elimination of the infection from other foci of infection.