点击显示 收起
Swiss Tropical Institute, Basel
Médécins sans Frontières, Geneva, Switzerland
Programme Nationale de Lutte contre la Trypanosomiase Humaine Africaine, Kinshasa, Democratic Republic of Congo
Programme National de Lutte contre la Trypanosomiase, Brazzaville, Republic of Congo
International Medical Corps, Nairobi, Kenya
Instituto de Combate e de Controlo das Tripanossomíases, Luanda, Angola
Programme National de Lutte contre la Trypanosomiase Humaine Africaine, Bangui, Central African Republic
Projet de Recherche Clinique sur la Trypanosomiase, Daloa, Cte d'Ivoire
Programa Nacional de Control de Tripanosomiasis, Bata, Equatorial Guinea
Fundació Centro de Información y Documentación Internacionales en Barcelona, Barcelona, Spain
Background.
Treatment of late-stage human African trypanosomiasis (HAT) with melarsoprol can be improved by shortening the regimen. A previous trial demonstrated the safety and efficacy of a 10-day treatment schedule. We demonstrate the effectiveness of this schedule in a noncontrolled, multinational drug-utilization study.
Methods.
A total of 2020 patients with late-stage HAT were treated with the 10-day melarsoprol schedule in 16 centers in 7 African countries. We assessed outcome on the basis of major adverse events and the cure rate after treatment and during 2 years of follow-up.
Results.
The cure rate 24 h after treatment was 93.9%; 2 years later, it was 86.2%. However, 49.3% of patients were lost to follow-up. The overall fatality rate was 5.9%. Of treated patients, 8.7% had an encephalopathic syndrome that was fatal 45.5% of the time. The rate of severe bullous and maculopapular eruptions was 0.8% and 6.8%, respectively.
Conclusions.
The 10-day treatment schedule was well implemented in the field and was effective. It reduces treatment duration, drug amount, and hospitalization costs per patient, and it increases treatment-center capacity. The shorter protocol has been recommended by the International Scientific Council for Trypanosomiasis Research and Control for the treatment of late-stage HAT caused by Trypanosoma brucei gambiense.
Human African trypanosomiasis (HAT; also called "sleeping sickness") ranks third among all parasitic diseases in sub-Saharan Africa, behind malaria and filariasis [1], in terms of disease burden as expressed in disability-adjusted life years [2]. Sixty million people in 36 African countries are at risk of becoming infected, and the number of cases is estimated at 350,000. At present, only a fraction of the population at risk is under surveillance; therefore, the 45,000 cases reported and treated per year may be a significant underestimation [3]. HAT is caused by the protozoan parasite Trypanosoma brucei subspecies and is transmitted by the bite of Glossina species tsetse flies [4].
HAT occurs in 2 distinct forms: a chronic form caused by Trypanosoma brucei gambiense and an acute form caused by Trypanosoma brucei rhodesiense; T. brucei gambiense causes 99.5% of cases [1, 4]. During the early hemolymphatic disease stage, the trypanosomes multiply in blood and lymph glands; they then invade the central nervous system, which corresponds to the late or meningoencephalitic disease stage. Without treatment, the disease is invariably fatal.
Today, 2 drugs are available for the treatment of late-stage HATeflornithine and melarsoprol. Eflornithine is difficult to administer, requires good logistics, and is expensive to manufacture. Therefore, it is of very limited use in rural treatment centers. In addition, it is ineffective against the acute form of the disease (caused by T. brucei rhodesiense). For these reasons and the lack of alternatives, melarsoprol remains the drug of choice for the treatment of late-stage HAT, even though it is highly toxic and does not results in a 100% cure rate [5, 6]. A major problem with melarsoprol is its long treatment schedule, which was developed empirically; 55 years after its introduction onto the market, regimens still vary considerably [1]. Generally, 34 series of 34 injections of increasing doses are given, spaced with resting periods of 710 days. To optimize and standardize melarsoprol treatment, an abridged 10-day protocol has been elaborated on the basis of pharmacological investigations [79], animal experiments [10], and pilot testing of patients with HAT in the former Zaire [11]. The effectiveness of this regimen, in terms of safety and efficacy, in comparison with a standard treatment regimen was shown in a large-scale randomized clinical trial in Angola (the Improved Application of Melarsoprol I study) [12, 13]. To further assess the effectiveness of the 10-day regimen under natural conditions, we performed a multinational, multicenter drug-utilization study for the treatment of late-stage T. brucei gambiense HAT.
PATIENTS, MATERIALS, AND METHODS
Centers and patients.
The study was implemented in 16 HAT treatment centers suggested by the respective national HAT programs or by nongovernment organization (NGOs), where applicable, in 7 sub-Saharan African countries where T. brucei gambiense is endemic. The minimal conditions for the center selection were reasonable accessibility, the availability of retrospective data on HAT treatment, and the exclusive use of the new treatment schedule for 12 consecutive months.
Study design and implementation.
A very simple study design, without randomization and sample-size calculation, was chosen to monitor the effectiveness of the 10-day treatment schedule for melarsoprol under true field conditions (drug utilization; IMPAMEL II). The enrollment period for each center was 12 months, to balance for seasonal variation. The study was approved by the ethics committee of the 2 cantons of Basel (Ethikkommission beider Basel) and the relevant ethics committees and authorities in the respective countries. In the selected centers, the abridged schedule was introduced as the standard treatment; therefore, no consent was obtained from the patients.
The patient inclusion criterion was confirmed late-stage gambiense HAT. At present, the methodology and criteria for diagnosis and staging of the disease are still not standardized among African countries; patients were included according to the staging criteria of the respective national HAT control programs. Generally, diagnosis of late stage was made by microscopic examination of the cerebrospinal fluid (CSF) for the presence of trypanosomes and/or an increased white blood cell (WBC) count. Different cutoff criteria for the WBC in CSF were used: >5 cells/mm3 in Central African Republic, Democratic Republic of Congo, Republic of Congo, and Sudan; 10 cells/mm3 in Equatorial Guinea; and 20 cells/mm3 in Angola and Cte d'Ivoire.
Treatment.
Patients were treated with 2.2 mg/kg/day of melarsoprol for 10 days, as a 3.6% solution in propylene glycol (Arsobal; Aventis), by slow intravenous (iv) injection. Before melarsoprol treatment, all patients received supplementary medication: antimalarials, mebendazole, multivitamins, and paracetamol (acetaminophen). During melarsoprol treatment, different prophylactic corticosteroid treatments were administered: prednisolone at a dose of 1 mg/kg on days 17, 0.75 mg/kg on day 8, 0.5 mg/kg on day 9, and 0.25 mg/kg on day 10 in Democratic Republic of Congo, Equatorial Guinea, and Sudan; 1 mg/kg on days 19, 0.75 mg/kg on day 10, 0.5 mg/kg on day 11, and 0.25 mg/kg on day 12 in Angola; 1 mg/kg on days 19, 0.75 mg/kg on day 10, and 0.5 mg/kg on day 11 in Central African Republic; 0.75 mg/kg on days 110 in Republic of Congo; and 1 mL of bethamethasone in Cte d'Ivoire. No prednisolone was administered, but, on days 110, promethazine (antihistamine) was given to the patients in 1 center in Democratic Republic of Congo (Centre Neuro-Psycho Pathologique/Cliniques Universitaires de Kinshasa).
When an encephalopathic syndrome developed, treatment with melarsoprol was suspended, and the patient was treated according to national guidelinesfor example, with adrenaline (epinephrine), corticosteroids (usually hydrocortisone), and/or diazepam. If possible, the melarsoprol treatment was resumed after 13 days or was considered to have been completed if 8 doses had already been given. For each patient, a case-report form was filled out that contained information on demographic, diagnostic, and clinical characteristics before and after treatment and an assessment of adverse events on a graded scale from 0 to 2 (no, moderate, or severe reactions).
Outcome measures.
Efficacy of the treatment was demonstrated, by the absence of trypanosomes, by microscopic examination of the blood and/or lymph and CSF and/or a reduction in WBC count. Patients were scheduled for clinical examination, including lumbar puncture (LP), 24 h after treatment and every 6 months for 2 years after treatment, to monitor for treatment failures and relapses. Treatment failures were defined as patients in whom trypanosomes could still be found in any body fluid 24 h after treatment; relapses were defined as patients presenting during follow-up with (1) trypanosomes in any compartment and/or (2) an increase in CSF WBC count to >50 cells/mm3 that was double that seen at the previous examination or a WBC count of 649 cells/mm3 that was accompanied by clear symptoms attributed to relapse (somnolence, long-lasting headache, and recurrent fever). The primary efficacy outcome was parasitological cure 24 h after treatment (treatment failures), and the secondary efficacy outcome was relapse within the follow-up period.
The safety of treatment was determined by the frequency of adverse events. The primary safety outcomes were death in temporal relation to treatment and the frequency of encephalopathic syndromes. The rate of other severe adverse reactions was defined as a secondary outcome.
Data management and statistical analysis.
All data were double entered and verified by use of EpiData 2.1 [14] software, and analysis was done by use of the statistical software package STATA (version 7.0; Stata). The findings were compared with historical data from the participating centers, literature, and the randomized clinical trial recently conducted in Angola [12, 13]. For calculation of the efficacy, the total number of all treated patients was used as the denominator, to allow comparison with previously reported rates.
RESULTS
Study population and baseline characteristics.
A total of 2800 patients were enrolled between June 1999 and June 2002; 780 patients were not eligible for the analysis for several reasons (figure 1). The final cohort consisted of 2020 patients who had been correctly diagnosed and were treated with the 10-day melarsoprol schedule in 10 HAT treatment centers in 5 countries.
The demographic, diagnostic, and clinical characteristics of the patients are shown in table 1. Distributions by age, sex, and nutritional status at the time of admission were similar among patients at different centers. The diagnostic findings varied from center to center, probably because of different methodologies and cutoff criteria. The majority of patients had lymphadenopathies, headaches, pruritus, general weaknesses, and sleeping disorders; a large variation in clinical manifestations among the centers was evident.
Efficacy.
The parasitological cure rate 24 h after treatment was 93.9% (1897/2020; range among centers, 85.7%100%) (table 2). A total of 119 (5.9%) patients died during treatment, and trypanosomes were detected in 4 (0.2%) patients who were treated for relapse. Two years after treatment, the observed cure rate was 86.2% (1742/2020), with considerable intercenter variation (range, 70.9%100%). Follow-up participation was highly variable (15%100%; data not shown). Many of the patients who were cured at discharge did not attend any of the clinical follow-up examinations (936/1897 [49.3%]) and were considered to have remained cured. Some 50.7% (961/1897) attended at least 1 of 4 prescribed follow-up examinations with LP; 144 of these (7.1%) were diagnosed as having relapses. During follow-up, 18 (0.9%) patients died, 7 (0.3%) after relapsing. In 53 (36.8%) of the patients who relapsed, trypanosomes were found in blood (7 [4.9%]) or CSF (46 [31.9%]). All other relapses were diagnosed by an elevated CSF WBC count of >50 cells/mm3 (61 [42.4%]) or by a CSF WBC count that had at least doubled since discharge from the hospital that was accompanied by clear symptoms of disease (30 [20.8%]). Relapses were not further analyzed, because of the large variation of the follow-up coverage among the different centers.
Safety.
The safety results are presented in table 3 by severity and country. These results reflect the expected large variation observed in the treatment of HAT.
During treatment, 119 patients died, an average after 9 days of treatment (range, 129 days). The major causes were encephalopathic syndromes, which contributed to 67.2% (80) of the fatalities. Other causes were advanced HAT (15 [12.6%]), concomitant diseases (10 [8.4%]), unknown etiology (9 [7.6%]), and bullous skin reactions (5 [4.2%]).
A total of 176 (8.7%) patients had an encephalopathic syndrome; they generally received iv steroids at different doses. The onset of the encephalopathic syndrome was reported after an average of 9 days of treatment (median, 9 days; mean, 9.2 days; range, 128 days). In patients who survived, treatment was resumed after a suspension of 3 days (median, 3 days; mean, 3.2 days; range, 112 days). No significant seasonal variation was observed in any of the centers (data not shown). Headache preceded the onset of the encephalopathic syndrome in 34.1% (60/176) and fever in 54.0% (95/176) of patients. In 22.7% (40/176) of patients with an encephalopathic syndrome, malaria parasites were detected during the syndrome, which probably caused the fever in 20 (11.4%) of them. The effect of the prophylactic use of prednisolone could only be evaluated for 2 southern Sudanese centers that provided reliable information for each patient. In those centers, 163 (26.8%) of 607 patients received prednisolone. Patients who did not receive prednisolone were at a slightly (statistically nonsignificant) higher risk of developing an encephalopathic syndrome (relative risk [RR], 1.3 [95% confidence interval {CI}, 0.82.2]; P = .286) or of dying (RR, 1.4 [95% CI, 0.53.9]; P = .598).
The frequency of skin reactions was high (28.3% [571/2020]) but varied between the centers. However, the majority of recorded skin reactions were moderate pruritus; this was consistently not considered to be a significant problem by the treating staff and could be controlled with steroids or promethazine. Patients in south Sudan who received prednisolone as prophylactic treatment were at a lower risk of developing a moderate skin reaction (RR, 0.6 [95% CI, 0.40.8]; P = .0004).
Other adverse reactions that were often reported included fever and headache. However, these are common symptoms and signs of the disease and are not easily discriminated from adverse events caused by treatment. Less frequently reported reactions included polyneuropathies, diarrhea, jaundice, and hypotension.
DISCUSSION
The present work (IMPAMEL II) was a noncontrolled, multinational, multicenter drug-utilization study to evaluate, under true field conditions, the abridged treatment schedule of melarsoprol during late-stage T. brucei gambiense HAT. Because of the very basic equipment available in treatment centers, the often low level of staff qualifications, and the lack of experience in the conduct of clinical trials in most centers, it was not possible to conduct a randomized study. Cluster randomization was also considered to be impossible, because of the inherent differences in the outcome of HAT treatment in different centers and countries and the limited number of centers available. The 12-month enrollment period was chosen to reveal a potential seasonal variation of the outcome [15]; centers that enrolled patients for <12 months were excluded from the formal analysis, and data for a longer period were truncated after the first 12 months. Other patients excluded from the formal analysis were those who received any treatment other than 10-day melarsoprol or who were preselected on the basis of the self-defined criteria of a treatment centerfor example, adults only or patients admitted in good health.
The study population corresponded to the average population of African countries [16], except for the patients treated in Cte d'Ivoire (n = 27), who were much younger (median age, 15 years) than the overall study population (median age, 27 years). The clinical conditions and the diagnostic characteristics of the patients at the time of admission were highly variable among centers, which reflects the different levels of surveillance activities in each country, the diverse qualifications of staff and their perception of illnesses and adverse reactions (nurses vs. medical doctors; national control programs vs. NGOs), and differences in laboratory equipment among the centers. However, there was no difference in the outcome of treatment among the cohorts that were diagnosed according to the different cutoff criteria in use (data not shown).
The average short-term (93.9%) and long-term (86.2%) efficacies, with "all patients treated" as the denominator and considering patients who not seen during the follow-up to be cured, were comparable to published data [5, 12, 1722] and the centers' histories (table 2). This approach is somewhat unsatisfactory, because it leads to an overestimation of effectiveness. The assessment of long-term efficacy is complicated by generally low follow-up rates; thus, no consistent information on follow-up coverage is given in the literature. Our approach yielded the best comparison to recently published studies [12, 1723] and to the centers' histories. The proportion of patients in our study who attended a follow-up examination was highly variable among centers (0%100%). On average, one-half of all patients attended at least 1 follow-up examination, mostly within the first year after treatment; therefore, the failure rates in the present study are difficult to estimate and to compare to reported rates. The overall rate of treatment failures and relapses in our study was 7.1% (144/2020), a value that compares well with those in the literature (5%8% [24]). However, because alarmingly high ratesup to 30%have been documented in some areas (Uganda [23], Angola [25], and Sudan [26]), failure rates by disease focus should be reported more carefully. In our study, most of the relapses (136/144) were reported in southern Sudanese centers, which may be explained (at least partially) by the enhanced follow-up activities conducted by the executing organizations, which resulted in a 12.2% (136/1111) failure rate in this area. However, this is still much lower than reported rates from this area (centers' histories, 15%25%; literature, 16%24% [22]).
The safety and tolerability of the 10-day melarsoprol treatment schedule were similar to those of standard treatment schedules [12, 2731], with the exception of skin reactions, fevers, and headaches (table 4). The variability of the adverse events among the different study centers was expectedly high, and comparable results were found in a separate analysis of data from the patients who had been excluded from the formal analysis. Mild symptoms and signs such as fever (32.3%), headache (29.6%), and pruritus (23.7%), which are also common symptoms and signs of HAT, were more frequently reported during the study. This may be an observation bias prompted by solicitation of information that usually is not recorded. That more treatment interruptions caused by moderate adverse events were reported in centers operated by NGOs supports this reflection.
The frequency of treatment-related death was 5.9% (119/2020), and the most severe reactions were encephalopathic syndromes (8.7% [176/2020]). These rates are comparable to published data [32, 33], although they are at the upper end of the range. In accordance with the findings of previous reports [12, 2731], encephalopathic syndromes occurred between days 1 and 28 after the initial injection of melarsoprol, with most of them occurring between days 9 and 11 (mean, 9.2 days; SD, 4 days), which supports the view that the event is independent of the treatment schedule and the dose applied. Also, the resumption of melarsoprol after interruption and patients' improvement did not result in a recurring reaction. The effect of prednisolone prophylaxis could only be evaluated for 2 treatment centers in Sudan, because other centers used dosage regimens that were slightly aberrant from the study protocol, had interruptions in drug stocks, or provided information for many patients that was not detailed enough. In the analyzed population, there was no statistically significant prophylactic effect of prednisolone on the development of an encephalopathic syndrome or a difference in the case-fatality rate in this cohort, as was suggested by Pepin et al. [30]. However, with prednisolone treatment, we observed clear protection against the development of moderate drug-related skin reactions [33].
The present results corroborate those of a randomized, controlled clinical trial previously conducted in Angola [12] (IMPAMEL I) that demonstrated that a 10-day regimen was not inferior to the lengthy standard treatment schedule. The overall frequency of adverse events was high, but, again, there was no increase, compared with that of standard treatment schedules, except for skin reactions. At the time of discharge from the hospital, symptoms and signssuch as pruritus, fever, headache, tremor, weakness, and unusual behaviorhad substantially improved (data not shown). Also, the long-term efficacy of our approach appeared to be equivalent to that of standard treatment schedules [13] when follow-up activities were purposely not assisted. The 10-day schedule has several advantages over the very lengthy standard schedules: it is favorable in socioeconomic (less drug is used and shorter hospitalizations are required), technical (10 consecutive days and no dosage adjustment), pharmacologic (the dosage used is the basis of all combinations of melarsoprol in the compassionate treatment of refractory cases), and psychological (patient and physician compliance) terms.
Clearly, the tolerability and safety of melarsoprol are inferior to those of eflornithine [34]. However, the administration of eflornithine as slow infusions every 6 h over the course of 14 days is difficult and requires qualified staff and very good logistics. Therefore, eflornithine remains restricted to centers that receive substantial and consistent support from NGOs, and the vast majority of patients are still treated with melarsoprol.
Also, melarsoprol is still is the only treatment for T. brucei rhodesiense, because of the inconsistent efficacy of eflornithine against this form of the parasite. However, the use of the 10-day melarsoprol schedule against T. brucei rhodesiense is strongly discouraged. The clinical nature of this form is very different, and high parasitemia levels are observed. Confirmation of the clinical evaluation of the 10-day schedule for rhodesiense HAT has yet to be performed.
On the basis of the results of IMPAMEL I and II and the experiences in each country, in September 2003 [35], the 10-day melarsoprol schedule was discussed, at the request of the World Health Organization, by the International Scientific Council for Trypanosomiasis Research and Control and was recommended as the standard schedule for the treatment of late-stage T. brucei gambiense HAT with melarsoprol. Its use was continued after the study in several countries (Central African Republic, Cte d'Ivoire, Democratic Republic of Congo, Equatorial Guinea, and Republic of Congo), and it is currently being implemented by the national HAT programs.
IMPAMEL II STUDY
In addition to the authors, the IMPAMEL II study was conducted by investigators in the following countries. Angola: Dr. Gedeo Vatunga, Dr. Francisco Manuel, Inacio Zua Antonio, and Dr. Andre Jose Ribeiro; Central African Republic: Dr. Sylvestre Mbadingai and Dr. André Sandoka; Cte d'Ivoire: Dr. Norbert Dje N'goran; Democratic Republic of Congo: Dr. Pascal Lutumba, Jean Kwete, Mandefu, Landu Rando Malu, Dr. Leon Kazumba, and Bonga Nsangu; Equatorial Guinea: Dr. Mario Sarsa, Dr. Jose Ramon Franco, and Eustaquio Nguema Ndong; Republic of Congo: Ngondongo Philippe, Dr. Sonja van Osch, Dr. Genevieve Kabonga, Dr. Unni Karunakara, and Dr. Diakite Drissa; and Sudan: Dr. Cedric Yashimoto, Dr. Mario Enrile, Dr. Anne Pittet, and Dr. Luca Flamingui.
Acknowledgments
We are indebted to our patients and their families; to the members of the data and safety monitoring board, Pierre Cattand, Lars Rombo, Dr. Blaise Genton, and Dr. Steve Bennett; to Dr. Jean Jannin (World Health Organization), for logistical support and drug supply; to the medical advisors, Drs. Johannes Blum and Jorge Seixas; to the statistical consultants, Dr. Penelope Vounatsou and Thomas Smith; to the project assistant, Doris Magdalinski; to the data entry clerks, Cedric Wasser, Kirsten Gillingwater, Katrin Bettge, and Verena Schfer; to the scientific advisors, Dr. Anne Moore, Reto Brun, and Philippe Buscher; and to Dan Anderegg, for manuscript revision. Logistic and technical support was provided by the World Health Organization, International Medical Corps, Médécins sans Frontières, the Fundació Centro de Información y Documentación Internacionales en Barcelona, and the Ministries of Health of the participating countries.
References
1. World Health Organization (WHO). Control and surveillance of African trypanosomiasis. WHO technical report series. Vol 881. Geneva: WHO, 1998. First citation in article
2. World Health Organization (WHO.) World health report 2004: changing history. Geneva, WHO: 2004. First citation in article
3. World Health Organization. African trypanosomiasis or sleeping sicknessfact-sheet. Wkly Epidemiol Rec 2004; 79:297300. First citation in article
4. Burri C, Brun R. Human African trypanosomiasis. In: Cook G, Zumla A, eds. Manson's tropical diseases. 21st ed. London: WB Saunders, 2002:130323. First citation in article
5. Legros D, Fournier C, Etchegorry MG, Maiso F, Szumilin E. Therapeutic failure of melarsoprol among patients treated for late-stage of T. b. gambiense human African trypanosomiasis in Uganda [in French]. Bull Soc Pathol Exot 1999; 92:1712. First citation in article
6. Stanghellini A. African human trypanosomiasis: therapeutic strategies [in French]. Bull Soc Pathol Exot 2000; 93:313. First citation in article
7. Burri C, Baltz T, Giroud C, Doua F, Welker HA, Brun R. Pharmacokinetic properties of the trypanocidal drug melarsoprol. Chemotherapy 1993; 39:22534. First citation in article
8. Burri C. Pharmacological aspects of the trypanocidal drug melarsoprol [Ph.D. thesis]. Basel, Switzerland: Swiss Tropical Institute, University of Basel, 1994. First citation in article
9. Burri C, Brun R. An in vitro bioassay for quantification of melarsoprol in serum and cerebrospinal fluid. Trop Med Parasitol 1992; 43:2235. First citation in article
10. Burri C, Onyango JD, Auma JE, Burudi EM, Brun R. Pharmacokinetics of melarsoprol in uninfected vervet monkeys. Acta Trop 1994; 58:3549. First citation in article
11. Burri C, Blum J, Brun R. Alternative application of melarsoprol for treatment of T. b. gambiense sleeping sickness: preliminary results. Ann Soc Belg Med Trop 1995; 75:6571. First citation in article
12. Burri C, Nkunku S, Merolle A, Smith T, Blum J, Brun R. Efficacy of new, concise schedule for melarsoprol in treatment of sleeping sickness caused by Trypanosoma brucei gambiense: a randomised trial. Lancet 2000; 355:141925. First citation in article
13. Schmid C, Nkunku S, Merolle A, Vounatsou P, Burri C. Efficacy of 10-day melarsoprol schedule 2 years after treatment for late-stage gambiense sleeping sickness. Lancet 2004; 364:78990. First citation in article
14. Lauritsen J, Bruus M. EpiData (version 2.1): a comprehensive tool for validated entry and documentation of data. Odense, Denmark: EpiData Association, 20002005. Available at: http://www.epidata.dk. Accessed 21 April 2005. First citation in article
15. Ancelle T, Barret B, Flachet L, Moren A. 2 epidemics of arsenical encephalopathy in the treatment of trypanosomiasis, Uganda, 19921993 [in French]. Bull Soc Pathol Exot 1994; 87:3416. First citation in article
16. United Nations Development Programme. Human development report 2003, millennium development goals: a compact among nations to end human poverty. Available at: http://hdr.undp.org/reports/global/2003/pdf/hdr03_HDI.pdf. Accessed 21 April 2005. First citation in article
17. Pepin J, Mpia B, Iloasebe M. Trypanosoma brucei gambiense African trypanosomiasis: differences between men and women in severity of disease and response to treatment. Trans R Soc Trop Med Hyg 2002; 96:4216. First citation in article
18. Atouguia JLM, Kennedy PGE. Neurological aspects of human African trypanosomiasis. In: Davis LE, Kennedy PGE, ed. Infectious diseases of the nervous system. 1st ed. Oxford: Reed Educational and Professional Publishing, 2000. First citation in article
19. Richet P, Lotte M, Foucher G. Résultat des traitements de la trypanosomiase humaine a Trypanosoma gambiense par le Mel B ou l'Arsobal. Med Trop (Mars) 1959; 19:25365. First citation in article
20. Dutertre J, Labusquiere R. La thérapeutique de la trypanosomiase. Med Trop (Mars) 1966; 26:34256. First citation in article
21. Bertrand E, Rive J, Serie F, Kone I. Encéphalopathie arsenicale et traitement de la trypanosomiase. Med Trop (Mars) 1973; 33:38590. First citation in article
22. Brun R, Schumacher R, Schmid C, Kunz C, Burri C. The phenomenon of treatment failures in human African trypanosomiasis. Trop Med Int Health 2001; 6:90614. First citation in article
23. Legros D, Evans S, Maiso F, Enyaru JC, Mbulamberi D. Risk factors for treatment failure after melarsoprol for Trypanosoma brucei gambiense trypanosomiasis in Uganda. Trans R Soc Trop Med Hyg 1999; 93:43942. First citation in article
24. Pepin J, Milord F, Khonde A, Niyonsenga T, Loko L, Mpia B. Gambiense trypanosomiasis: frequency of, and risk factors for, failure of melarsoprol therapy. Trans R Soc Trop Med Hyg 1994; 88:44752. First citation in article
25. Stanghellini A, Josenando T. The situation of sleeping sickness in Angola: a calamity. Trop Med Int Health 2001; 6:3304. First citation in article
26. Brun R, Baeriswyl S, Kunz C. In vitro drug sensitivity of Trypanosoma gambiense isolates. Acta Trop 1989; 46:36976. First citation in article
27. Blum J, Nkunku S, Burri C. Clinical description of encephalopathic syndromes and risk factors for their occurrence and outcome during melarsoprol treatment of human African trypanosomiasis. Trop Med Int Health 2001; 6:390400. First citation in article
28. Doua F, Yapo FB. Human trypanosomiasis in the Ivory Coasttherapy and problems. Acta Trop 1993; 54:1638. First citation in article
29. Nkanga NG, Mutombo L, Kazadi K, Kazyumba GL. Neuropathies arsenicales apres traitement de la trypanosomiase humaine au melarsoprol. Med Afr Noire 1988; 35:736. First citation in article
30. Pepin J, Milord F, Guern C, Mpia B, Ethier L, Mansinsa D. Trial of prednisolone for prevention of melarsoprol-induced encephalopathy in gambiense sleeping sickness. Lancet 1989; 1:124650. First citation in article
31. Pepin J, Milord F, Khonde AN, et al. Risk factors for encephalopathy and mortality during melarsoprol treatment of Trypanosoma brucei gambiense sleeping sickness. Trans R Soc Trop Med Hyg 1995; 89:927. First citation in article
32. Kuzoe FA. Current situation of African trypanosomiasis. Acta Trop 1993; 54:15362. First citation in article
33. Pepin J, Milord F. The treatment of human African trypanosomiasis. Adv Parasitol 1994; 33:147. First citation in article
34. Burri C, Brun R. Eflornithine for treatment of human African trypanosomiasis. Parasitol Res 2003; 90(Supp 1):S4952. First citation in article
35. Anonymous. Recommendations of the 27th International Scientific Council for Trypanosomiasis Research and Control (ISCTRC) at Pretoria. Pretoria, South Africa: ISCTRC, 2003. First citation in article