Re-emergence of epidemic sleeping sickness in southern Sudan

Authors


Anne C. Moore Division of Parasitic Diseases, M/S F-22, Centers for Disease Control and Prevention, 4770 Buford Highway, Atlanta, GA 30341, USA. E-mail: ary2@cdc.gov

Abstract

A resurgence of sleeping sickness developed in southern Sudan during the past decade. Prevalence of confirmed Trypanosoma brucei gambiense infection in humans now exceeds 5% in several foci. From 1997 to 1999, trypanosomiasis control programmes in three counties of Western Equatoria Province detected 3785 new cases among 67 181 persons screened. A major contributing factor in the re-emergence of epidemic sleeping sickness was the lack of active case-finding throughout the 1990s. Although the situation is improving in sites where trypanosomiasis programmes have been recently implemented, co-ordination and additional international assistance are needed to bring sleeping sickness under control in Sudan.

Introduction

Human African trypanosomiasis (HAT), or sleeping sickness, has been a public health problem in Sudan throughout most of the 20th century. Endemic foci of Trypanosoma brucei gambiense are found in southern Sudan (World Health Organization 1998) in a belt bordering Central African Republic (CAR), Democratic Republic of Congo (DRC) and Uganda. Epidemic flare-ups of trypanosomiasis have been observed periodically in this region. After a resurgence of the disease in the late 1970s in the provinces of Western and Eastern Equatoria, a bilateral Belgian–Sudanese trypanosomiasis treatment and control programme was implemented. This effort brought the disease under relatively good control within a few years. However, the trypanosomiasis programme collapsed in 1990 during the civil war. From 1990 to 1995, sleeping sickness treatment in southern Sudan remained available at a single site, the Li Rangu Hospital in Yambio County. Although treatment drugs were in limited supply, staff members were unpaid, and transport to the hospital was not available, 40–50 patients per year were treated. There was virtually no active case detection (ACD) in southern Sudan for more than 8 years, from 1989 until late 1997.

We describe a resurgence of sleeping sickness that occurred during the period when treatment and control were interrupted. Data from the province of Western Equatoria were collected in 1997–1999 and were provided by the non-governmental organizations (NGOs) International Medical Corps (IMC), CARE South Sudan, Médecins Sans Frontières, Holland (MSF-H), and Epicentre, and by the Centers for Disease Control and Prevention (CDC).

Prevalence of HAT in Western Equatoria Province

In 1997, it became evident that sleeping sickness was staging a comeback. After IMC observed a progressive increase in passively detected cases in Tambura County, Sudan, a population-based prevalence survey documented epidemic levels of trypanosomiasis in the south-western part of the county bordering CAR and DRC (Moore et al. 1999). In this focus, 19.4% of the population was seroreactive when screened by Card Agglutination Test for Trypanosomiasis (CATT) (Magnus et al. 1978), 13% of the population had confirmed infection based on a single parasitologic examination.

At the time of the survey in Tambura, general health services in Western Equatoria were fragmentary. Although NGOs, United Nations agencies, and religious groups provided basic services in some areas, passive case detection for sleeping sickness was extremely limited. However, the high infection prevalence found in the south-west Tambura focus, as well as anecdotal reports of sleeping sickness in other communities, prompted the collection of additional prevalence data in 1998 and 1999 (Table 1). The areas investigated are sites where sleeping sickness has been a major public health problem in the past. However, other historically active foci remain unexamined. Because the geographical distribution of trypanosomiasis tends to be uneven, the full extent of the current public health threat in southern Sudan is unknown. Nevertheless, sleeping sickness at epidemic levels (parasitologically confirmed prevalence > 2%) has been documented at sites within three of the four counties investigated in Western Equatoria (Fig. 1). These levels of confirmed trypanosomiasis probably underestimate the true prevalence of infection. Screening for sleeping sickness in Sudan is conducted by serologic testing with the CATT, followed by microscopic examination of blood and lymph node fluid to detect the parasite. In most settings, the CATT specificity is good, and its sensitivity is higher than parasite detection because the level of parasitemia in T. b. gambiense infection can be low and variable. In southern Sudan, seroprevalence using the CATT substantially exceeds confirmed infection prevalence in most foci.

Table 1.   Prevalence of HAT Western Equatoria Province, Sudan and adjacent areas 1997–1999* Thumbnail image of
Figure 1.

 Prevalence of HAT in Southern Sudan 1997–1999. Prevalence of parasitologically confirmed infection is shown.

In 1998, the most active focus of trypanosomiasis identified in Western Equatoria was Ibbe, in Maridi County, with a confirmed infection prevalence of 29%. Other areas with epidemic disease were Kajo Keji (5.1%) and several sites in Tambura County (Ezo, 13.6%; Source Yubu, 6.4%; Mupoi, 5.4%). In late 1999, 20 months after a cycle of ACD had been conducted in Tambura County, repeat screening in several areas showed that the prevalence of confirmed infection had declined dramatically. Despite an influx of new infections among Sudanese refugees repatriated from DRC, trypanosomiasis prevalence decreased in Ezo from 13.6 to 3.1% and in Source Yubu from 6.4 to 1.6%. The observation that active surveillance and case treatment produces a rapid and substantial reduction of disease transmission is consistent with previous reports from Sudan and elsewhere (Simarro et al. 1991).

The current resurgence of trypanosomiasis is not confined to Western Equatoria or to southern Sudan. Epidemic levels of the disease are also present in adjacent areas of CAR and DRC (Table 1). Bazigbiri, a village in CAR approximately 75 km from the Sudan border, was nearly depopulated by sleeping sickness in 1995 (J. Jannin, personal communication).

HAT treatment and control

West African trypanosomiasis is transmitted through the bite of the tsetse fly and follows a chronic clinical course. Initially, the parasite is confined to the haemolymphatic system (stage I), however, ultimately, after a period of months, the central nervous system is invaded (stage II). Untreated, the infection tends to progress over several years and is almost invariably fatal. CATT seroreactivity detects infection in both disease stages. The primary method for control of West African trypanosomiasis is serologic screening of the population using CATT, followed by treatment of parasitologically confirmed infections. Because humans constitute the only epidemiologically significant reservoir of T. b. gambiense, this control strategy lowers disease incidence by reducing the reservoir. It also limits mortality. ACD is supplemented by passive surveillance at hospitals and selected health centres. As an adjunct approach, vector control is used to limit tsetse density at targeted sites and to promote community interest and participation in population screening.

At present, health care and public health activities in Western Equatoria are almost entirely dependent on external support. NGOs and United Nations agencies operate most health programmes, with accreditation by the Sudan Relief and Rehabilitation Association (SRRA). In the past decade, funding for sleeping sickness control has been minimal and donor interest has been low. After it became clear that sleeping sickness was re-emerging in multiple foci, limited resources for control became available, and trypanosomiasis programmes were implemented by NGOs in three (Maridi, Tambura and Yambio) of the four Western Equatoria counties where a sleeping sickness problem had been documented. A programme in the fourth county (Kajo Keji) was put in place during 2000. A summary of control activities to date is shown in Table 2.

Table 2.   Active case detection and treatment of HAT in southern Sudan 1997–1999* Thumbnail image of

In Tambura County trypanosomiasis control began through a collaborative IMC/CARE programme in October 1997. In the subsequent 16 months, one mobile team screened the population living in trypanosomiasis-endemic areas of Tambura County. Based on multiple sources (e.g. local authorities, headmen) the 35 724 persons screened during the first round of ACD represented more than 97% of the population in these areas. Overall seroprevalence by CATT was 9.3% (range 0.5–27.0%), and prevalence of parasitologically confirmed infection was 4.8% (range 0–17.4). (Table 1). In 1997–1999, 3237 new case-patients were treated in Tambura (Table 2). Because most patients have been identified through ACD, nearly 60% had stage I infection (cerebrospinal fluid (CSF) containing no detectable trypanosomes and a leucocyte count < 6/μl).

A second cycle of population screening was conducted in south-western Tambura County in the fall of 1999. Despite the nearly 2-year interval between cycles of case detection, and the introduction of new cases from DRC during that time, the overall prevalence of confirmed infection decreased nearly threefold, from 7.0 to 2.4%.

Control activities in Maridi County are based at Ibbe, the focus with the highest sleeping sickness prevalence found in recent surveys. The programme, operated by MSF-H, began treating passively detected case-patients in February 1999 and subsequently implemented targeted ACD with one mobile team. By the end of 1999, 940 new case patients were treated. The proportion of cases detected passively was high, and 82% of patients treated in the Ibbe programme had evidence of central nervous system involvement (stage II, CSF containing trypanosomes and/or a leucocyte count of 6/μl or more). Médecins Sans Frontières, France, will conduct future control activities in Maridi County.

A control programme in Yambio County, under the direction of IMC and based at the Li Rangu Sleeping Sickness Hospital, began passive case detection in September 1999. ACD in Yambio and implementation of a treatment and control programme in Kajo Keji by Médecins Sans Frontières, Switzerland (MSF-CH) were initiated during 2000.

The Tambura County programme includes a component of community-based vector control, implemented in 1998 with assistance from Medical Emergency Relief International (MERLIN). Control of Glossina fuscipes fuscipes, the vector species for T. b. gambiense in southern Sudan, is targeted on potential transmission sites (e.g. water sources) within or near villages in which prevalence of sleeping sickness is high. Locally made pyramidal traps are maintained by 350 community volunteers trained in vector recognition. Supervision and reporting are integrated with the primary health system. Although the impact of vector control on disease transmission in this setting is unknown, Tambura County sites observed a decline of 70–80% in the number of trapped flies within 3 months of trap placement.

Current and future issues

Treatment failure.

The choice of therapy for HAT is determined by the presence or absence of CNS involvement. Stage I T. b. gambiense infection is treated with pentamidine. However, the CNS penetration of the drug is poor, and therefore, stage II disease is treated with melarsoprol, an organoarsenic compound. Melarsoprol is a powerful trypanocide used exclusively for stage II disease because of its toxicity. Since the introduction of melarsoprol in the 1940s, treatment failure has been noted in a small proportion of patients, but the failure rate has remained below 10% for 5 decades (Pepin & Milord 1994). However, high rates of treatment failure recently have been observed in discrete foci in Uganda (Legros et al. 1999), and there are anecdotal reports of similar failure rates in Angola and other disease-endemic areas. Interpretation of these findings is complicated by the lack of standard melarsoprol treatment protocols and by the use of different denominators in calculating treatment failure rates. The underlying cause of melarsoprol-refractory infection is unknown but could be drug resistance, host factors, or both. To date, few T. b. gambiense isolates have been tested for melarsoprol susceptibility.

In Sudan, data about melarsoprol treatment failures are limited. The trypanosomiasis programmes have not been operating long enough to follow most treated patients for the 2-year period generally recommended to assess cure rates for sleeping sickness. No isolates from Sudan have been tested for melarsoprol resistance. However, existing data indicate that patients should be monitored carefully and that treatment centres should watch treatment failure rates closely. In Tambura, among 1670 hospital inpatients admitted for stage II sleeping sickness, 271 (16%) had prior therapy with melarsoprol. In Ibbe, where the treatment programme had operated for just 11 months, 87 (11%) of 791 stage II inpatients were admitted for relapse. Furthermore, post-treatment follow-up of stage II patients treated in Ibbe between February and July 1999 found that more than 20% had relapsed by the end of the year. This number of treatment failures detected within a short period is relatively high, and additional data are needed over a longer duration of follow-up to determine whether melarsoprol efficacy has declined in these areas. Because the case-fatality rate for sleeping sickness is 100% without appropriate treatment, the problem of melarsoprol failure may be compounded by lack of availability of eflornithine and other alternative therapies (Pécoul et al. 1999).

Integration of trypanosomiasis control

with the health system.

Integration of trypanosomiasis surveillance with general health services can be helpful in expanding passive case detection (Pepin et al. 1989) and reducing costs. The NGO trypanosomiasis programmes have relied primarily on local staff and have trained approximately 150 local health care workers in HAT diagnosis, clinical management, and follow-up. However, successful integration depends on a functional primary care system with adequate supplies, record keeping and personnel training. The current health care system in Western Equatoria consists of an assortment of services provided by various NGOs and religious groups. Although there is some coordination of activities by UNICEF and by the southern Sudanese health authorities, and although communication among organizations is generally good, these organizations work largely independently. In a given area, the organization responsible for a sleeping sickness programme may not be the provider of general health care. In many areas, primary health services are limited by lack of resources. Because the sleeping sickness programme and the primary health care centres in Tambura County are currently operated under the same NGO umbrella (IMC/CARE), integration has been achieved there to some degree. Microscopic diagnosis, staging of sleeping sickness, treatment of stage I infection, patient follow-up, and vector control surveillance are performed at some Tambura health centres. However, in other areas of southern Sudan, involvement of the general health care system with trypanosomiasis control is minimal.

Greater coordination of trypanosomiasis efforts in southern Sudan is needed so that a comprehensive control strategy can be implemented. Serologic methods, treatment protocols, and reporting are not standardized. Operation Lifeline Sudan (OLS), a consortium of NGOs and United Nations agencies operating in southern Sudan, convened a task force on sleeping sickness to address some of these issues, but not all organizations that conduct sleeping sickness programmes in southern Sudan are members of OLS.

Discussion

A resurgence of sleeping sickness has been documented in four counties of Western Equatoria, however, information about the severity in many other sites in southern Sudan is limited. Because clinical symptoms of the disease are non-specific, particularly in the early stages, epidemic sleeping sickness can remain unrecognized for extended periods, even in settings where health care facilities are operational. In some trypanosomiasis-endemic foci of southern Sudan, not even passive case detection for sleeping sickness is in place. The number of people living in areas at risk for sleeping sickness in southern Sudan can be estimated at 1–2 million, but reliable data are not available. Census data acquired by NGOs for specific sites have found that the current population is substantially less than it was a decade ago.

The re-emergence of epidemic sleeping sickness is a result of the disruption of an ongoing control programme brought about by civil war at the beginning of the decade. This led to a prolonged period in which no ACD and almost no treatment were available. Population movements and other war-related risk factors may have contributed to increased transmission of sleeping sickness in some areas, although they were not epidemiologically significant in the Tambura County focus (Moore et al. 1999). However, the major cause of the resurgence appears to be the lack of case finding and treatment. The current high prevalence foci (e.g. Ibbe, Ezo) were residual hot spots in 1990 when trypanosomiasis control ceased, and prevalence rose 50-fold in these sites during the 8-year interval when case-finding and treatment availability were limited (Moore et al. 1999).

Considerable progress has been made in documenting the trypanosomiasis resurgence and in implementing active surveillance and treatment in several areas of Western Equatoria. However, identifying adequate resources to control sleeping sickness throughout southern Sudan remains a fundamental problem. External funding for trypanosomiasis control is essential for virtually every country coping with this re-emerging infection. However, while the civil war in Sudan continues, some international donors are unwilling to provide development assistance for improving general health services and disease control. Security remains an issue in parts of southern Sudan. In addition, permission for future operation of existing NGO-lead trypanosomiasis control and other health programmes is uncertain and is under negotiation with local authorities.

Acknowledgements

We thank Jim Sanders and Bronwen Blake of Médecins Sans Frontières (Holland) and Greg Brady of CARE South Sudan for generously sharing data. We thank Mary Bartlett for editorial assistance.

Ancillary