Onchocerciasis in Yemen: moving forward towards an elimination program

Abstract

The onchocerciasis focus in Yemen has been known for many years as an endemic area with unique characteristics, notably the atypical and most severe form of onchodermatitis, known as sowda or reactive onchodermatitis (ROD). The national effort to control the disease began in 1992 as an individual case treatment program by administering ivermectin to those presenting with ROD. The challenging geography of the endemic area and the current political and military unrest both underscore a need for special approaches when attempting to eliminate onchocerciasis from this country. An assessment of the national situation regarding this disease was carried out in 2011–2013 aimed at defining the best approach for moving from individual clinical case treatment to elimination of transmission. The history of the control efforts and the current status of the disease are reviewed and the essential changes needed to a mass drug administration (MDA) approach are identified as the national program addresses elimination. Yemen, despite the current troubles, has shown that it can successfully implement MDA programs despite many difficulties and therefore should be supported in its efforts towards countrywide elimination of this infection; however, success will need renewed national and international efforts.

Onchocerciasis, Yemen, Sowda, MDA, Onchodermatitis, Elimination

Introduction

Onchodermatitis was first described in the Gold Coast (now Ghana) in 1895 by the British naval surgeon John O’Neill, who identified the parasites in the skin of patients suffering from a severe dermatitis that he termed ‘craw craw’. However, the major focus of concern in much of the past 50 years has been on the ocular pathology and blindness induced by this parasite; this is reflected in the common name for the disease, ‘river blindness’. The very successful program set up in 1987 by MSD, also known as Merck & Co., Inc., Kenilworth, NJ, USA, to donate their drug Mectizan (ivermectin) for onchocerciasis control was initially directed at reducing the onchocercal eye disease, although the focus for its use in Yemen has always been for treating the skin disease occurring in this country. In the past two decades the dermatological aspects of this infection have been better understood across all endemic areas and the use of ivermectin is now firmly focused on reducing skin disease in addition to the ocular form. A major change in the global program in the past few years is the redirection from a disease control program to one now aimed at elimination and the breaking of transmission of infection.¹

Yemen has played an important role in developing the clinical spectrum definition of onchodermatitis, largely through the original description of a severe form of the onchodermatitis (reactive onchodermatitis [ROD]) present in this country. Aside from being the only major onchocerciasis endemic focus present in Asia, the Yemeni focus has always been described as being an anomaly in that there is little evidence of ocular damage and observations show that palpable subcutaneous nodules (containing adult worms) are uncommon; this contrasts with the typical picture in most endemic areas. The unique clinical character of Yemeni patients, together with the fact that it is a relatively small endemic area and that it is located outside Africa, where the majority of the infections are seen, has meant that at the global level Yemeni onchocerciasis has often been ignored. However, this was certainly not the case within the country itself, where in 1992 concerted efforts began to treat and control clinical dermatological disease in the endemic areas of the country using ivermectin²; this was carried out using the unusually energetic approach of administering four rounds a year (compared to the annual or biannual distributions in Africa and Latin America, respectively).

This article describes the history of the national onchocerciasis program’s efforts and the current situation for onchocerciasis in Yemen and addresses the needs for establishing an elimination program. It should be noted that much of the more recent information on this disease in the country is present in a variety of unpublished documents prepared by external consultants and local supporting agencies.

Epidemiology

Sowda is considered to be a public health problem in Yemen that is confined to a portion of the valleys facing the Red Sea that separate the Arabian Peninsula from Africa (Figure 1), and there is no evidence to date to suggest that there are endemic areas outside these known valleys. These westward running wadis contain rapidly flowing water during the rainy season, from April to September, at which time the majority of the fly breeding takes place. Nevertheless, in the dry season there are still a few areas with running water present in these wadis, with active breeding of Simulium blackflies taking place.

Figure 1.

Map of areas endemic for onchocerciasis.

It was estimated in the 1980s that some 50 000 people living in these endemic wadis were infected with Onchocerca volvulus and were presenting with clinically obvious sowda, while at that time there were some 200 000 people at risk. However, the population in these valleys has grown considerably and it is expected that the exposed population in these endemic areas is now between 350 000 and 400 000, with many more people (3 million or more) actually residing in the districts where these endemic valleys are located. For the elimination program to be successful it is essential to map out the limits of transmission of the disease; this is one of the most urgent needs for the program to carry out when stability permits. Unlike in Africa, nodules and skin snips will not be a useful indicator for mapping. Recent studies have shown that the most common antibody assays used in other regions of the world to define onchocerciasis (e.g., Ov16 recombinant antigen serology) also highlight Yemeni onchocercal infection and are thus valid tools for use in epidemiological studies in this country (C. Mackenzie, A. Al-Kubati, T. Nutman, personal communication).

Entomologists over the past 40 years have confirmed the first observation in 1969 by Merighi et al.³ that Simulium rasyani is the sole vector of onchocerciasis in the Yemeni endemic valleys. However, it should be noted that there remains a healthy discussion concerning the true extent of the geographic distribution of this vector. Although similar vectors have been described in the southern areas of Saudi Arabia in its border region with Yemen,⁴ to date, human infections of O. volvulus have not been identified in these border locations, nor has the parasite been identified in the Simulium vectors present in these geographic areas. S. rasyani, a member of the Simulium damnosum blackfly complex, has characteristics making it clearly distinct from other blackflies present in the country. Why the infection and this particular blackfly are confined to specific valleys remains a question discussed by entomologists and epidemiologists.

The clinical disease

The first description of clinical cases of onchocerciasis in Yemen was made in 1957 by Fawdry,⁵ who documented 50 cases of a form of dermal onchocerciasis present in the western valleys of the country. Sowda, or sowdah, was the local name for this condition recorded by Fawdry, and this name has been commonly used in the scientific literature for this particularly severe, and often localized, form of onchodermatitis, also known as reactive onchodermatitis (ROD). Sowda in Arabic means ‘black’ and refers to the hyperpigmentation that is present in these cases; in Yemen this condition is most often present on the lower legs (Figures 2 and 3). Another distinctive characteristic of the disease in Yemen is that it is almost exclusively dermal in its clinical presentation and appears to not involve the eye to any significant extent. Büttner et al.⁶ reported that most of the patients with localized onchocerciasis are older children, teenagers or young adults; however, it is the experience of the current onchocerciasis program team that all ages, including elderly people, can be affected by sowda.

Figure 2.

The clinical presentation of onchocerciasis in Yemen. (A) A man suffering from sowda on his left leg, which is darkened by the condition. (B) A subcutaneous onchocercal nodule present on the hip of a Yemeni male, an unusual finding.

Figure 3.

Clinical changes in the skin of Yemeni onchocerciasis patients. (A) Scratch marks on the skin induced by the intense pruritus. (B) A severe case of reactive onchodermatitis with localized areas of hyperpigmentation associated with indurated papular pathology. (C) Loss of pigment typical of chronic onchocerciasis, probably due to a combination of changes due to onchocerciasis and reactions to bites by the vector, Simulium spp.

Intense pruritus is a major presenting symptom of the disease in Yemen,⁵ occurring both by day and night (Figure 3A), and to such a degree that it can interfere with the patient’s ability to sleep. Fawdry⁵ also stated that there can also be a more ‘tolerable’ level of itching, one that patients will accept for many months, even years, before presenting for medical consultation and treatment. It is likely that other reasons for non-attendance at medical clinics by those suffering from ‘tolerable’ itching include the difficulty of transport, a low level of health education or perhaps the inadequacy of medical services at that time. The scratching associated with the severe itch commonly leads to skin abrasion and the development of secondary infections. These sequelae can eventually destroy areas of epidermis and lead to destruction of melanocytes and the consequent loss of skin pigmentation that often brings the patient to the dermatologist complaining of vitiligo (Figure 2).

The most common presentation is that of an asymmetrical, localized, large lesion, usually on one leg, and relatively uncommonly involving both legs, or an arm or a shoulder.⁷ The lesions can consist of papules, macules, erosions, impetiginous areas and crusts, and is usually accompanied by areas of hyper- and hypopigmentation and, as mentioned above, even depigmentation (Figure 2). The skin sometimes becomes thin, inelastic and vulnerable to injury. In the light-skinned people of Yemen, the skin of the affected limb becomes a darkened colour that contrasts dramatically with the skin of the normal limb, hence the name sowda (Figures 2 and 3). The altered skin pigmentation that results from this condition, often presenting as ‘leopard skin’ (Figure 3C), can cause sufferers to be socially excluded in their communities. This social stigma can interfere with their chances of marriage if they are young, single adults.

It is also common for the draining inguinal lymph nodes to be greatly enlarged when there is an active onchodermatosis of the associated limb, even when the skin immediately distal to the lymph node is normal. These nodes can increase in size to be up to 2–5 cm long and 2–3 cm wide; they are usually not painful or tender, and are mobile and rubbery in consistency. The local Yemeni people call these lymph nodes Al-Umm, meaning ‘the mother’, believing that it is the source of the disease, and they will often go to a surgeon to have them removed. Unfortunately, due to inadequate knowledge of this disease and its manifestations, some surgeons will remove these lymph nodes, believing them to be adult worm–containing subcutaneous nodules. The only consequence of this surgery is to compromise lymphatic drainage, in fact making the condition worse. Palpable subcutaneous onchocercal nodules are not common in Yemeni patients with localized onchocerciasis (Figure 2B). Fawdry⁵ noted that only five patients had subcutaneous nodules among the 50 sowda patients he examined, and in support, Büttner and Racz⁸ also stated that the prevalence of nodules in Yemeni patients is very low. In the past 25 years the Yemeni Onchocerciasis Program has found very few patients with palpable onchocercal nodules.

It has been thought by many that Yemen has a very unique and exclusive form of onchocercal disease; however, it arguably is more likely that sowda is one component of the clinical spectrum known in onchocerciasis that happens to be very dominant in Yemeni onchocerciasis patients. The different forms of onchodermatitis have been detailed^9–11 previously and range from a generalized distribution on the body to a more localized presentation, the latter often being associated with more severe changes; this clinical spectrum is associated with parallel immunological and parasitological spectra. Nwokolo¹² described two forms and suggested that there were relatively few microfilariae present in the localized form compared with the higher loads in those with the generalized forms of onchocercal disease. This general concept of the existence of a spectral presentation of dermal disease has been described by other investigators in Africa—in Cameroon,¹³ Sudan¹⁴ and Nigeria.¹² The majority of cases seen in Yemen represent one end of this clinicoparasitological spectrum.

Histopathology

A limited amount of histopathology has been carried out on tissues from Yemeni onchocerciasis patients. Gasparini^15–17 described hyperkeratosis, parakeratosis, acanthosis and elongated papillae in untreated onchocercal skin lesions, changes that are generally attributed to this condition. A diffuse inflammatory cellular infiltration, specifically consisting of eosinophils, macrophages and giant cells, was present in the upper dermis around the small blood vessels. The basal layer of epidermis in these biopsies showed uneven pigmentation, and occasionally more or less intact microfilariae were present, as were degenerating parasites surrounded by macrophages and giant cells.^6,8 Lymph nodes taken from a sowda patient presented a reactive profile with enlarged coalescing follicles and the presence of active germinal centres, supporting the classification of this disease as ROD. The capsule of the lymph node was thickened and a cellular infiltration of plasma cells, eosinophils and mast cells was present. This active pathological picture is consistent with an active host response against a parasite being present in Yemeni patients, and this is also consistent with other studies emphasizing the immunologically active nature of localized onchodermatitis.¹⁸

Entomology

As described above, the vector of O. volvulus in the Yemeni endemic foci is the anthrophilic blackfly, S. rasyani.^2,18–20 The breeding site for this fly, as with other Simulium spp., is shallow, rapidly flowing, well-oxygenated water, such as runs through the wadis (Figure 4). This blackfly is the Yemeni representative of the S. damnosum complex and was named after one of the endemic valleys, Wadi Rasyan, which lies between 300 and 900 m above sea level (13°38′ N and 43°47′ E, in the Taiz governorate on the road to Mocha, the ancient Yemeni port on the Red Sea famous for exporting Mocha coffee).

Figure 4.

Yemeni onchocerciasis endemic environment and entomology. (A) River bed in the dry season. (B–D) Simulium larvae and pupae on rocks and vegetation collected in the dry season.

Investigations in the 1980s defined many of the entomological characteristics of S. rasyani. Büttner et al.⁶ collected male and female flies, larvae and pupae from five valleys, Wadis Rasyani, Kabir, Al-Barh, Zabid and Surdud, all valleys with similar vegetation and ecology and all at altitudes between 300 and 900 m above sea level. Garms and colleagues¹⁹ reported on the chromosomal nature of these larvae and their morphology, and compared them to the African species of the S. damnosum complex. He concluded that there are chromosomal differences between African and Yemeni larvae in chromosomes II and III. In addition, the female flies collected by Büttner et al.⁶ appeared to be morphologically uniform and clearly different from the African species of the S. damnosum complex. The males also have a different morphology from the African Simulium. In contrast, the larvae collected have similar morphology to those of West African flies.^20,21 However, as the ecological habitat of S. rasyani is very different from that of Central African members of the S. damnosum complex, it is unlikely that S. rasyani is closely related to these African species. Garms and Kerner²² suggested that as the habitats of the Yemeni wadis are similar to many valleys in Saudi Arabia, it was possible that S. rasyani could also occur in Saudi Arabia, especially near the border between the two countries. Nonetheless, to date S. rasyani has not yet been defined as being present in Saudi Arabia or in other countries in the region, including Eritrea, Djibouti and Ethiopia, nor in the Hanish Islands that are relatively close across the narrow Red Sea.

The female blackflies that can carry the infective larvae in Yemen are known to bite throughout the day, although the main biting periods occur from sunrise to 09:00 h and from 16:00 h to dusk. Fawdry⁵ suggested the fact that the skin lesions were most commonly present on patients’ legs was due to the vector fly being most active at ground level and thus most biting occurs predominately on the lower limbs. Alternatively, however, the distribution of clinical lesions may be more related to the clothing habits of the Yemeni people. Males usually wear a short dress, known as a fotah (Figure 3), which leaves the legs exposed to the biting of the flies. Females more often have lesions on their upper arms, and this may correspond to the fact that it is common dress practice for women to cover their legs and leave their arms exposed.

As these endemic river valleys are subject to marked changes in water flow that depend on the time of the year, it makes the use of alternative control mechanisms, such as breaking transmission by vector control, somewhat challenging. It should be noted that in onchocerciasis programs in Africa and Latin America the role of vector control has been minimal, with vector control being seen as useful only in certain special situations. Given the difficulties of access and the annual variability in water flows in the Yemeni endemic areas, it is most likely that the primary activity for elimination will be ivermectin distribution with high MDA coverage.

Parasitology

The parasitological features of the Onchocerca microfilariae found in humans in Yemen were detailed by Omar et al.²³: the length of microfilariae of 270±18.5 μm (range 230–310 μm) compares well with that of parasites from Upper Volta (Burkina Faso), which are 276 μm (range 220–300 μm) long. There was also reasonable morphological conformity between the microfilariae from these two areas. However, Omar et al.²³ stated that microfilaria from sowda patients had acid phosphatase staining enzyme activity in the anal excretory vesicle area that was unlike that seen with African microfilariae, where there is generally no such staining pattern. Both adult worms and microfilariae from Yemen and West Africa appeared to have similar physical morphology when examined under a scanning electron microscope.²³ Nodules excised and examined by Büttner and Racz⁸ were shown to be similar to those seen in onchocerciasis-endemic areas in Africa.

Approaches to control

Much effort has been made over the past 25 years to treat and reduce the incidence of the dermatologic cases in Yemen, actions that have taken place under difficult geographic circumstances and with very little support from external sources. Nevertheless, MSD has consistently provided Mectizan to Yemen for patient care, with local medical programs (non-governmental development organizations/government health system) as part of their overall activities targeting dermatologic conditions, implementing distribution of this drug.

Following the original identification of the infection the microfilaricidal drug diethylcarbamazine (Hetrazan)²⁴ was used in Yemen as well as Suramin for its macrofilaricidal action. Both these agents caused severe side effects, and antihistamines and systemic steroids were often needed to treat the problems arising from administering these drugs. Ivermectin (Mectizan), a macrocyclic lactone isolated from fermentation broths of the soil fungus Streptomyces ivermitilis, is currently the standard treatment for this disease and was first used in Yemen in 1992 by the National Leprosy Elimination Programme (NLEP) based in Taiz; it was administered for case treatment and not as a community-wide approach. A single oral dose, as determined by body weight (150 μg/kg), was given to sowda patients who were not seriously ill and who were not pregnant, as required for the appropriate administration of ivermectin. As the signs and symptoms of sowda (predominantly pruritus) were often seen to return only 3 months after the initial treatment, it was decided to administer the ivermectin on a quarterly basis rather than the standard annual regimen used in Africa²; this regimen was adopted with approval from the Mectizan Donation Program.²⁵

The NLEP first began this treatment program by using mobile teams to distribute ivermectin to sowda patients in Wadi Al-Ghayl, an endemic area previously identified by several authors^19,20,26 and where, at that time, more than 50% of inhabitants were suffering from this disease. In 1993 the sowda control efforts in Yemen continued to be integrated with the ongoing national efforts to control leprosy and the treatment program was expanded to the western running valleys in the Tihama region. After 8 years of distributing ivermectin every 3 months, a total of some 116 900 tablets had been given to 16 778 patients. In 2001 the Ministry of Public Health delegated the responsibility of sowda control to the NLEP as the National Onchocerciasis Control Program (NOCP), located in Taiz. In the same year, a local non-governmental organization, the Charitable Society for Social Welfare (CSSW), joined in the effort to take the lead for drug distribution in the northern endemic valleys. Thus the country’s program was divided into two subprograms: the northwestern area, supervised by the CSSW, and the remainder, supervised by the NLEP. In 2002 the NOCP in Taiz also began an MDA program to eliminate lymphatic filariasis (LF) in six wadis that coincidently were co-endemic with onchocerciasis; this combined effort came to an end with the elimination of LF in these wadis in 2007.

Success to date

The Yemini program has been successful to date in a number of areas. First is the achievement of having reached the majority of individuals suffering from the sowda form of onchodermatitis with four rounds of treatment per year—a difficult task in the best of field situations, but especially in a country with so many logistical challenges. This success is due to the efforts of the healthcare workers and supporting agencies, and the NELP and CSSW staff, as well as the success of the surveillance system established as part of the countrywide healthcare worker training programs. By 2007 more than 250 000 ivermectin tablets had been distributed in the southern areas of the onchocerciasis endemic valleys covered by the NLEP centred in Taiz. In the partner program run by the CSSW, serving the northern endemic valleys, more than 750 000 ivermectin tablets were distributed in 2001–2006.

Importantly, maintenance of this difficult drug administration schedule has had a dramatic effect on disease prevalence. By the beginning of 2007, an assessment of the treatment program distribution of ivermectin to residents living in the northern endemic valleys under the care of the CSSW showed a marked decrease in the occurrence of new cases of sowda, with the prevalence of cases needing treatment decreasing from >50% before drug distribution to around 6% in the worst affected area of Wadi Surdud. Currently, in most areas it is uncommon to find new cases of sowda, even in the valleys that were formerly the most endemic, such as Wadis Al-Ghayle, Khozigah, Mulhem and Rasyan. It should also be noted that the NLEP team has had success in eliminating LF through the distribution of around 1.3 million tablets of ivermectin (together with albendazole) in wadis that were co-endemic for these two filariae; the MDA activities carried out for LF are likely to have also had an effect on the prevalence of onchocerciasis in the co-endemic areas.

As the country moved towards establishing an MDA program for elimination of onchocerciasis transmission, it was important to better understand the epidemiology of the infection in those endemic area residents who did not present with sowda. It has been observed during ivermectin distribution that many of the non-sowda residents experienced the acute dermal reactions that are known to occur in onchocerciasis after the administration of microfilaricidal drugs (known as Mazzotti reactions²⁷); this indicated that there was a high likelihood that these people were in fact infected but not presenting with sowda. Recent unpublished studies have shown that, using current onchocerciasis serological tests, up to 25% of non-sowda individuals in the major endemic wadis carry evidence of being or having been infected with O. volvulus (C. Mackenzie, A. Abdul-Samid, T. Nutman, personal communication).

Restructuring for elimination

The success of the patient-directed campaign by the national program to reduce sowda catalyzed the possibility of achieving elimination of onchocerciasis in Yemen. Given that the infection and its transmission appear to be restricted to eight major valleys (Figure 1), and that these valleys have had an ongoing drug distribution system for case treatment in place on a quarterly basis for many years, this goal appeared achievable. Consequently, the National Onchocerciasis Programme developed a plan for countrywide elimination in 2010–2012 with assistance from the Mectizan Donation Programme and Michigan State University.²⁷

The fundamental element in the 2012 National Onchocerciasis Elimination Plan was to implement a change from a case-based program to a mass drug distribution program that treated the whole community. As described above, there is post-treatment clinical evidence that many of the non-sowda residents in the Yemeni endemic areas also carry dermal microfilariae and have not been a focus of treated efforts in the past. They are therefore a major source of the parasites that are maintaining transmission, albeit perhaps at a low level. The new national plan is therefore aimed at including these parasitized non-sowda individuals through the MDA approach. As previously mentioned, an MDA approach has been successfully implemented for the LF endemic wadis in Yemen, achieving elimination of Wuchereria bancrofti from these areas. This positive experience of the LF elimination program indicates that an MDA for onchocerciasis can also be implemented and be successful.

A number of challenges face the restructured Yemeni program as it addresses elimination of onchocerciasis. These include adequate financing for the new program, which, unlike most onchocerciasis country programs in Africa and Latin America, has not been regularly funded in the past; the difficulties of drug supply, distribution and administration in areas of civil disturbance; as well as establishing a practical monitoring and evaluation system to assess success. The detection of infection is a challenge in areas where the parasite levels are naturally low, as is the case in Yemen, and this will be even more difficult as the MDA approach takes effect. It is likely that serological approaches together with careful patient-based observations will need to be included in the monitoring methodologies. Mapping the disease distribution using skin snips or assessing the presence of nodules, as in Africa, will not be appropriate due to the relatively low parasite numbers in this endemic area; antibody tests are now the preferred mapping tool.

Current elimination programs in Africa have, where feasible, used vector control approaches to reduce and destroy the vector Simulium breeding activity using larvacidal agents. Whether this approach is a valuable addition to the new Yemen program is unclear. It is possible that the apparent restriction of the vector to the endemic valleys and the limited number of breeding sites in the dry season may present an opportunity to add a vector component to the overall elimination plans. However, at this time, especially given the general challenges facing the new program, it would seem advisable that the program focus on achieving high-coverage MDA using ivermectin.

Comment

Successful elimination in Yemen, although seemingly achievable given the experiences in other endemic countries, e.g., in Northern Sudan,²⁸ is still faced with a number of challenges due to local factors. Yemen is a country that, at the time of writing, is still faced with serious political and financial challenges, together with extreme civil disruption due to serious factional fighting. Currently, widespread and serious food shortages, famine and outbreaks of cholera add to the civil disturbances that plague the Yemeni people, and external financial and technical support for the elimination efforts are obviously needed. These challenges suggest that moving to an MDA will be difficult, although it should also be noted that MDA for onchocerciasis has in fact already begun in the northern areas of the country (in Al Mahwit Governorate) and is moving to a third year of treatment. The extraordinarily active and successful locally driven initiatives, together with dedicated fieldwork by local teams, which that have achieved high levels of drug distribution for clinical disease over the past 25 years, also suggest that elimination can be achieved despite the challenges. However, new initiatives from medical institutions within Yemen are also important.²⁹

The conceptual change in approach from only treating those individuals with visible skin disease to including the whole population through MDA for onchocerciasis is likely to be successful in both treating cases and preventing infections. This approach, accepted by the Yemeni government in 2012 in the proposed National Elimination Plan for Yemeni Onchocerciasis,²⁸ has been supported by internal and external scientists and partners,²⁹ and appropriate funding is being sought for implementation of this plan.

Conclusion

It is clear from the Yemeni program activities over the past 25 years that the administration of ivermectin on a quarterly basis has had a significant controlling effect on onchocerciasis case prevalence. However, to shift to eradicating the disease entirely from the country it is necessary to move to an MDA program, as was used successfully in the LF program in mainland Yemen. It is now a real possibility that the Yemen onchocerciasis program can address the possibility of eliminating O. volvulus as an infection from the country. Such a move from a disease control program to an elimination program will require renewed efforts by the program staff and will necessitate increased support.

Authors' contributions: AA-S and CM wrote the manuscript; AA-S, CM, DB, YA, AA-S and IA carried out the program activities; BT and AH reviewed and edited the manuscript.

Acknowledgements: None.

Funding: None.

Competing interests: None declared.

Ethical approval: Not required.