Evaluation of four therapeutic regimens for falciparum malaria in Mozambique, 1986

Bulletin of the World Health Organization, 66 (2): 219-226 (1988) © World Health Organization 1988 Evaluation of four therapeutic regimens for falciparum malaria in Mozambique, 1986 A. SCHAPIRA1 & J. F. L. SCHWALBACH2 A randomized study on the effect of the following four treatment regimens on Plasmodium falciparumparasitaemia was carried out on 200 asymptomatic schoolchildren in Maputo, Mozambique: chloroquine (25 mg/kg body weight), amodiaquine (25 mg/kg), sulfadoxine-pyrimethamine (25 mg/kg and 1. 25 mg/kg), or amodiaquine (25 mg/kg) + sulfadoxine-pyrimethamine (25 mg/kg and 1. 25 mg/kg) administered on the third day of the study. The results of in vivo tests indicated that 94% ofthe infections were resistant to chloroquine, 76% to amodiaquine, and 16% to sulfadoxine-pyrimethamine. The cure rate with amodiaquine+sulfadoxine-pyrimethamine was 100%, which was not significantly differentfrom that with sulfadoxine-pyrimethamine alone; the latter regimen was the most rapidly acting of the treatments studied. It is concluded that amodiaquine is not an appropriate substitute for chloroquine, but that the effect of the combination amodia- quine +sulfadoxine-pyrimethamine may be superior to that ofsulfadoxine-pyrimethamine alone, although this requires further study. The spread of chloroquine-resistant malaria in Africa makes it necessary to define alternative treatment regimens for cases that display such resist- ance or for use in areas where chloroquine has lost its therapeutic value. Sulfadoxine-pyrimethamine is the traditional substitute for chloroquine, but resistance to this combination has developed with deplorable speed in south-east Asia (1) and has been reported also in East Africa (2). Amodiaquine has been shown to retain some efficacy in Kenya (3), despite the presence ofchloroquine resistance in the country, and has been recommended by some groups for first-line treatment in cases of chloroquine failure (4). The present study was undertaken to investigate whether the combination of amodiaquine and sulfa- doxine-pyrimethamine might remain effective de- spite the possible presence of resistance to the separate components. The effect of four treatment regimens- chloroquine, amodiaquine, sulfadoxine- pyrimethamine, and amodiaquine+ sulfadoxine-pyri- methamine- was investigated in a randomized trial using an in vivo test methodology supplemented by in vitro drug susceptibility assays. 1 Research Fellow, Malaria Department, Instituto Nacional de Sadde, Maputo, Mozambique. 2 Chief, Malaria Department, and Director, Instituto Nacional de Sadde, C.P. 264, Maputo, Mozambique. Requests for reprints should be sent to this author. MATERIALS AND METHODS Study area The study was carried out from May to September 1986 in Machava district, Maputo, Mozambique, a semi-urban area with perennial transmission of mal- aria that is exacerbated from February to May. In 1985,a 78% of Plasmodium falciparum isolates in Maputo were chloroquine resistant in vitro. The only control measure available against malaria in the area is presumptive treatment with chloroquine through primary health care. Subjects Schoolchildren from three primary schools were screened from May to September 1986 by examin- ation of samples of capillary blood. Those who exhibited more than 800 asexual P. falciparum parasites per Id upon on-the-spot microscopic study of 10 fields were selected. Each carrier was then carefully questioned about whether they had taken chloroquine over the previous 14 days, and, if this was denied, the child was entered in the study. When a INSTITuTo NACIONAL DE SA(JDE. [ 7he susceptibility of P. falciparum to chloroquine in Maputo]. Maputo, Instituto Nacional de Satdde, 1985 (in Portuguese). 4875 -219- 220 A. SCHAPIRA & J. F. L. SCHWALBACH possible, the information given by a child was checked by questioning a parent. The Dill-Glazko test for the presence of aminoquinolines in the urine was not used, because we have previously found it to be unreliable (5). Most of the parasite carriers were asymptomatic, but some had fever and headache. Severely ill chil- dren did not participate in the study, but were given routine treatment and followed up. Before the start of the investigation, parents were informed of the trial and its objectives and those who did not wish their child to participate were asked to advise the head- master. Children were removed from the study if they did not complete their treatment or if they were lost from follow-up before day 14 of the trial. A total of 200 children took part in the study. Treatment Randomization was performed reiteratively in order to create four groups of 50 children who were assigned to one of the following treatments, which were written on cards kept in envelopes numbered from 1 to 200 and opened as the children entered the trial: chloroquine diphosphate,b 25-30 mg base per kg body weight over 3 days (10+10+5 mg/kg); amodiaquine hydrochloride,' same dosage as chloroquine diphosphate; -sulfadoxine-pyrimethamine,d 25-30 mg sulfa- doxine and 1.25-1.5 mg pyrimethamine per kg as a single dose; and amodiaquine (25-30 mg base per kg over 3 days (10+ 10+5 mg/kg)) supplemented on the third day of the trial (day 2) with sulfadoxine-pyrimethamine (25-30 mg sulfadoxine and 1.25-1.5 mg pyri- methamine per kg) together with 5 mg/kg amodia- quine. In the last-mentioned treatment regimen, sulfa- doxine-pyrimethamine was given only on the third day, because it was assumed a priori that amodia- quine acted more rapidly than sulfadoxine-pyri- methamine; also, especially in symptomatic patients, it was feared that giving both amodiaquine and sulfadoxine-pyrimethamine at the start of the treatment might cause vomiting. No placebo was given. Treatment was given under the authors' direct supervision, and the children were kept under observation for 30 minutes after each dose. The quality of the amodiaquine tablets was con- trolled by the Ministry of Health's Laboratory for b Avloclor, ICI, lot PA 51/7/1. " Pharmnachemic, batch 83 J 09/1. d Fansidar, Hoffmann-La Roche, lot B 4093 MFD 0583. Control of Quality of Drugs (B.P. 82, Mr J. Neves de Figueiredo). The WHO 28-day test (6) was carried out with the following modifications: thick fims of blood were prepared daily from day 0 to day 9 (except Sundays) and on days 14, 21, 28, and 35. For all films 100 microscope fields were examined and the asexual and sexual plasmodia present counted against 500 leuko- cytes, or, if very rare, in 100 fields, corresponding to approximately 2000 leukocytes. The results of the in vivo tests were classified using standard WHO criteria (6) and only cases that had been followed for 28-35 days or those proven to be resistant earlier were included. Recrudescences in RI cases were classified as early, if present before or on day 14, otherwise as late. The study was not blind, but the microscopists were not aware of the treatment regi- men that had been assigned to individual children. In vitro tests Before the start of treatment, 200 1l of capillary blood was drawn into heparinized capillary tubes and transferred immediately to 1.8 ml of RPMI 1640 mediume containing 25 mmol/l HEPES buffer, 25 mmol/l sodium bicarbonate, and 40 mg/l genta- micin. This suspension was used for standard WHO "micro" in vitro schizont maturation testsf with chloroquine (plate batches C-66, C-79, C-80, C-83, and C-84), amodiaquine (plate batches 11 February 1985, A-3 and A-9). The results with amodiaquine plates A-3 and A-9 were later found to be invalid and are not reported. The residual blood and medium suspension was used to carry out a merozoite re- invasion test for pyrimethamine susceptibility (7). After centrifugation for 5 minutes at 114 g, the supernatant was carefully withdrawn with an Eppen- dorf pipette, replaced with RPMI 1640 medium sup- plemented with 10% non-immune-type AB serum, and the erythrocyte volume fraction adjusted to 2%. Aliquots of 100 dl were placed in the wells of microtitration plates that were pre-dosed with pyri- methamineg to yield final concentrations of the drug of0, 0.03, 0.1, 0.3, 1, 3, 10, and 30 smol/l. The tests were harvested after 48 hours' incubation at 37- 38 °C, thin films being made from the cells in each well. These were examined by counting the number of young rings in either 4000 erythrocytes per fim or in 100 fields corresponding to 10 000 -15 000 erythrocytes, if only a few rings were visible in the ' Gibco, Paisley, Scotland. f PAYNE, D. Practical aspects ofthe use ofthe standard WHO in vitro macro- and microtest systems for the determination of the sensitvity of Plasmodium falciparum to chloroquine, mefloquine, amodiaqsune, and quinine. Unpublished WHO document MAP/ 84.2, 1984. ' Weilcome, Research Triangle Park, NC, USA. EVALUATION OF FOUR THERAPEUTIC REGIMENS FOR FALCIPARUM MALARIA Table 1. Some characteristics of the children in the four treatment groups Amodiaquine Sulfadoxine- + sulfadoxine- Chloroquine Amodiaquine pyrimethamine pyrimethamine No. of children 33 35 35 28 Mean age (years) 9.7 (2.5)' 9.8 (2.2) 10.6 (2.2) 9.9 (2.2) Geometric mean parasitaemia level on day 0 (asexual Pf. /41) 6150 (1.72) 5330 (3.87) 5480 (3.88) 3980 (3.88) 26.0 (1.0)b Mean dose 29.0 (3.7)C + 26.7 (2.8)c (mg/kg body weight) 26.2 (1.8) 25.9 (0.90) + 1.45 (0.19)d + 1.34 (0.14)d Figures in parentheses are standard deviations. b Values are for amodiaquine. c Values are for sulfadoxine. d Values are for pyrimethamine. control. A test was rejected if 10 young rings could not be found in the control. For both types of in vitro tests, minimal inhibitory concentrations (MIC) were defined as the lowest con- centration that caused complete inhibition of schizont maturation or merozoite re-invasion. The concen- trations that resulted in 50% inhibition of parasite development (ICso) were calculated by regression analysis.* h GRAB, B. & WEsNSDORFER, W. H. Evaluation ofin vitro tests fordrug sensitivity in Plasmodium falciparum: probit analysis oflog dose/response test from 3-8 points assay. Unpublished document WHO/MAL/83.990, 1983. RESULTS In vivo tests During the survey, the P. falciparum infection rate in the screened children declined from 37.1% in May to 14.2% in September 1986. Some characteristics ofthe children in the four treat- ment groups are shown in Table 1. There were no significant differences between the groups as far as age, initial parasitaemia level, and dosage of amodia- quine or of sulfadoxine-pyrimethamine were con- cerned (single classification ANOVA, P>0.05). Of Table 2. Results of the 28-day in vivo test with four treatments for Plasmodium falciparum malaria in Maputo, Mozambique, 1986a Amodiaquine + Sulfadoxine- sulfadoxine- Chloroquine Amodiaquine pyrimethamine pyrimethamine n % n % n % n % S 2 6.1 8 24.2 21 84.0 22 100 RI, late recrudescence 7 21.2 17 51.5 3 12.0 0 0 RI, early recrudescence 20 60.6 8 24.3 0 0 0 0 RII 3 9.1 0 0 1 4.0 0 0 Rill 1 3.0 0 0 0 0 0 0 Total 33 33 25 22 Mean parasite clearance time (days) 2.48 (0.21)b 2.23 (0.59) 1.85 (0.33) 2.04 (0.68) ' Cases were followed for 28-35 days or diagnosed as resistant before this time. b Figures in parentheses are standard deviations. 221 A. SCHAPIRA & J. F. L. SCHWALBACH the 200 children who entered the study, 69 were later excluded, because the initial parasitaemia, when re- checked, was too low, because of non-compliance with treatment or because they were lost from follow- up before day 14. Non-compliance was less frequent for the sulfadoxine-pyrimethamine group than for the others. The results of the in vivo test shown in Table 2 clearly separate the treatments into two groups: those including sulfadoxine-pyrimethamine, with an over- all cure rate of 91%, and those including only a 4-aminoquinoline, with an overall cure rate of 15 %. Within these groups, there is a significant difference between the outcomes ofthe amodiaquine and chloro- quine treatments (Mann-Whitney test, P<0.05), but not between those of sulfadoxine-pyrimethamine with and without amodiaquine (P=0. 18). Mean parasite clearance times were significantly lower for monotherapy with sulfadoxine-pyrimethamine than with chloroquine or amodiaquine (Student's t-test, P< 0.05), but the difference between the two sulfa- doxine-pyrimethamine regimens was not significant. Fig. 1, which shows the mean parasite densities as a proportion of their initial values, indicates that the most rapid clearance was obtained with sulfadoxine- pyrimethamine. Also, it should be noted that the therapy with amodiaquine differs from that with chloroquine mainly in terms of late recrudescences. The frequency of gametocyte patency in the four groups is shown in Fig. 2. All the treatment regimens were associated with an increase in the gametocyte index, which reached a maximum on day 7-8 and ,w 25 Chloroquine25 - 2m ---Amodiaquine \\-Suffadoxine-pyrimethamine 10 ---Amodiequine + suffadoxine-pyrimethamine Day w a X Fig. 1. Geometric mean and standard error of the mean for density of asexual Plasmodium faciparum parasites expressed as a proportion of that on day zero+0.1 during and after treatment with chloroquine, amodia- quine, sulfadoxine-pyrimethamine, or amodiaquine + sulfadoxine-pyrimethamine. 0 40 CD 0 20- 'E a *o- Chloroquine Amodiequine .------ Suadoxinepyrimethamine --Amodisquine + suffadoxine-pyrimethamine ... ..... ... 4 5 6 Day 9 14 21 29 Fig. 2. Frequency of Plasmodium falciparum gametocyte patency during and after treatment with chloroquine, amodiaquine, sulfadoxine-pyrimethamine, or amodia- quine + sulfadoxine-pyrimethamine. declined towards day 21-28. Gametocytaemia was less prolonged for those treated with amodiaquine+ sulfadoxine-pyrimethamine than for those who received sulfadoxine-pyrimethamine, but the differ- ence between the number of person-days with gametocytaemia in the two groups was not significant (X2 test, P>0.05). With the exception of one patient who was treated with chloroquine and became posi- tive on day 35 of the study, the results of the in vivo tests on day 35 did not alter the in vitro test classi- fications obtained on day 28. In vitro tests The distributions of the MIC and IC50 values for chloroquine and amodiaquine are shown in Table 3. There is no clear correlation between the in vivo and in vitro test results. According to criteria used by WHO, the MIC values for chloroquine in the in vitro tests indicate that 92.8% of cases were resistant to the drug, which is almost identical to the level of resistance determined from the results of the in vivo tests (Table 2). For amodiaquine, where an MIC > 4 umol/l is considered to be indicative of resistance (W. H. Wernsdorfer, personal communication, 1985), the frequency of resistance in vitro was 35.3%, which differs significantly from that in vivo of 75.8% (X2 test, P< 0.05). The IC5o and MIC values for pyrimethamine are compared with the in vivo results in Table 4. The one case that was resistant to sulfadoxine-pyrimethamine, and which had a successful in vitro test, had an IC5o value > 3 limol/l, which previous studies have shown to be associated with in vivo resistance to this com- bination (2). A similar in vitro result was found in a case which was cured by amodiaquine+ sulfadoxine- pyrimethamine. The frequency of isolates with ICso u t f f i' 5 5 222 6 1, ... IW0 EVALUATION OF FOUR THERAPEUTIC REGIMENS FOR FALCIPARUM MALARIA Table 3. Correlation of the results of the in vitro tests for chloroquine and amodiaquine and in vivo tests for all treatments In vivo test response Amodiaquine + Sulfadoxine- sulfadoxine- Chloroquine Amodiaquine pyrimethamine pyrimethamine In vitro test result S RI Rul S RI S RI S Totala Chloroquine b n 2 12 2 2 13 8 1 6 69 MIC >1.6pmolI 1 11 1 2 12 8 1 6 64(92.8%) MIC 41.14,smol/A 1 1 1 - 1 - - - 5 (7.2%) Mean IC5 (jsmolfl) 0.66 (0.27)c 1.52 (0.26) 0.44(0.17) 1.79(0.10) 1.51 (0.30) 0.91 (0.31) 0.68 1.50(0.19) 1.38 (0.11) Amodiaquined n MIC >0.4 umolI MIC <0.2 pmol/V Mean IC5o (jmol/A) 2 1 3 - - 5 _ _ - 3 1 3 - 0.21 0.10 (0.01) - 2 - 0.08 (0.01) 1 - 4 17 - - 1 6 (35.3%) 1 - 3 11(64.7%) - - 0.18 (0.09) 0.12 (0.02) ' The total number of in vitro tests conducted was greater than the sum of the numbers in the groups, because some were not complemented by in vivo tests. b A minimal inhibitory concentration (MIC) A 1.6 gsmol/l was taken to be indicative of drug resistance. ' Figures in parentheses are the standard error of the mean. d A minimal inhibitory concentration (MIC) ;0.4 jmolA was taken to be indicative of drug resistance. values > 3 gmol/l was 6.3%, which is not signifi- candy different from that of the in vivo resistance to sulfadoxine-pyrimethamine. Of the MIC values, 31.7% were > 3 imol/l, which suggests the presence of a low proportion of highly resistant parasites in a high proportion of the infections. There is a marginal correlation between the degree of in vivo susceptibility to chloroquine and the pyrimethamine ICso value (Spearman's e =0.87; 0.1 >P>0.05), but not between amodiaquine in vivo susceptibility and the pyrimethamine IC5o value (e =0. 16; P>0. 1). Table 4. Correlation of the in vitro tests with pyrimethamine and in vivo test results for all treatments In vivo test response" Amodiaquine+ Sulfadoxine- Sulfadoxine- Chloroquine Amodiaquine pyrimethamine pyrimethamine Concentration of pyrimetha- S/RI RIl S RI S RI S Totalb mine in vitro fi%mol/l} IC50 MIC ICws MIC IC50 MIC ICs MIC IC50 MIC IC50 MIC IC50 MIC IC50 MIC >30 - 1 - - - - - - - 3 -1 - 1 - 8 3-29.9 - 1 - 2 - 1 1 2 - 1 1 - 1 2 4 12 0.3-2.99 1 - 1 - - - 1 1 - - - - - 5 5 12 0.03-0.29 1 4 1 - 1 1 - 3 3 4 - - 3 6 11 31 <0.03 4 - - - 1 - 4 - 5 - - - 10 - 43 - Total 6 2 2 6 8 1 14 63 ' An IC54 ;3 pmolIl was considered to be indicative of resistance. b The total number of in vitro tests carried out was greater than the sum of the numbers in the groups, because some were not complemented by in vivo tests. 223 A. SCHAPIRA & J. F. L. SCHWALBACH DISCUSSION Both in vivo and in vitro results indicate a high fre- quency of chloroquine resistance in P. falciparum infections in the study area, and the level of in vitro resistance is probably associated with a higher degree of in vivo resistance among non-immunes. With amodiaquine, the in vivo results were only marginally better than with chloroquine, a finding that is consistent also with the outcome of a 28-day in vivo study of amodiaquine in Rwanda (8). Interest- ingly, had in vivo tests been performed only as far as day 7 or 14 of the study, amodiaquine would have appeared markedly superior to chloroquine. Late recrudescence is indistinguishable from reinfection, but, even with very high inoculation rates, reinfection could only account for a small proportion of the 51.5% of cases treated with amodiaquine that were classified as RI late recrudescence (9). Also, the declining parasite rate during the study implies that the inoculation rate was low. The difference between in vivo and in vitro results with amodiaquine may possibly have arisen because of reinfection; however, amodiaquine is a pro-drug, whose active metabolite, monodesethylamodiaquine, is less active than the parent compound (10). It still seems uncertain whether in vitro tests with amodia- quine are an adequate epidemiological tool. The high degree of susceptibility to pyrimethamine and the corresponding high cure rate and rapid effect of sulfadoxine-pyrimethamine is encouraging, but the in vitro results suggest that there is the potential for rapid evolution of sulfadoxine-pyrimethamine resistance in the study area. Our findings indicate that there was a combination of a high degree and fre- quency of resistance to 4-aminoquinolines and a high degree of antifolate susceptibility in the study area. This contrasts with results from south-east Asia (1) and with those from a study of highly selected East African material (11), but resembles the findings in Kenya reported in 1983 (12). Antifolate resistance in chloroquine-resistant areas is therefore probably mainly conditioned by specific drug pressure. The results of the in vivo tests do not conclusively establish whether the combination of sulfadoxine- pyrimethamine with amodiaquine offers distinct therapeutic advantages, but the lack of correlation between in vivo susceptibility to amodiaquine and in vitro susceptibility to pyrimethamine is supportive evidence in favour of the superiority of the combi- nation. It has previously been pointed out that, even in the absence of cross-resistance, combination treat- ment may only delay the appearance of resistance to the component drugs if the resistance to each separate constituent is relatively rare (13). Although most infections among the children in the present study were resistant to amodiaquine, the proportion ofamo- diaquine-resistant parasites could not be assessed. It must also be taken into account, that in an area where malaria is endemic, the proportion of infections exposed to treatment is low, and that as long as com- bination therapy can be reserved for cases that are both clinically and parasitologically chloroquine resistant, only a very small proportion of the parasite pool will be exposed to such therapy. A controlled trial of sulfadoxine-pyrimethamine versus amodiaquine+sulfadoxine-pyrimethamine in a sufficiently large series of symptomatic patients is warranted. If the triple combination proves superior, it would be more rational to administer sulfadoxine- pyrimethamine on the first day instead of on the third day, as here, since it acts more rapidly than amodia- quine. Recent reports suggest, however, that pro- phylaxis with amodiaquine+ sulfadoxine-pyrimetha- mine may be associated with severe side-effects (14). Nevertheless, in Maputo, more than 1000 patients have been treated with this combination and followed up for 28 days, and, apart from pruritus, dermatologi- cal or haematological side-effects have never been observed. However, in holoendemic areas it may often be necessary to treat children for malaria at intervals of 1-2 months, and determination of the risk of severe side-effects associated with repeated amodiaquine + sulfadoxine-pyrimethamine treatment must be considered a research priority. ACKNOWLEDGEMENTS This study was supported by a grant from the WHO/UNDP/World Bank Special Programme for Research and Training in Tropical Diseases. The Department of Parasitology, Statens Seruminstitut, Copenhagen, Denmark, provided some reagents and pyrimethamine pre-dosed plates. Avloclor tablets were donated by ICI Pharmaceuticals and Fansidar tablets by Hoffmann-La Roche. Excellent technical assistance was given by the field workers and microscopists of the Malaria Department, Instituto Nacional de Sadde, especially Mr F. Matsinhe, Mr S. Leio, and Mr F. Capitine, who performed the in vitro tests. Mr S. da Costa executed the drawings, and Dr J. Le Bras and Dr L. T. Almeida Franco gave valuable comments on the manuscript. We owe them all our best thanks. 224 EVALUATION OF FOUR THERAPEUTIC REGIMENS FOR FALCIPARUM MALARIA 225 RESUME EVALUATION DE QUATRE SCHEMAS THERAPEUTIQUES DANS LE PALUDISME A FALCIPARUM AU MOZAMBIQUE, 1986 Pour verifier si l'association amodiaquine plus sulfa- doxine/pyrimethamine conserve son efficacit6 thera- peutique malgre une r6sistance possible de Plasmodium falciparumn chacun de ses constituants, on a effectu6 une 6tude randomis&e de l'effet de quatre traitements sur la parasitnmie a P. falciparum. La population a l'etude consistait en 200 ecoliers asymptomatiques du district de Machava, Maputo, Mozambique, secteur oti la chloroquino- resistance a 6t6 fr6quemment rapportee en 1985. De mai a septembre 1986, alors que la transmission du paludisme etait I un niveau tres faible, 50 enfants asymptomatiques pr6sentant une parasitemie simple a P. falciparum asexue d'au moins 800 parasites par ul de sang ont ete repartis dans chacun des quatre groupes th6rapeutiques suivants: 25 mg/ kg de poids corporel de chloroquine sur trois jours, 25 mg/ kg d'amodiaquine sur troisjours, (25 + 1 ,25)mg/kg de sulfa- doxine/pyrimethamine en une seule dose, et 25 mg/kg d'amodiaquine sur trois jours +(25 + 1 ,25)mg/kg de sulfa- doxine/pyrimethamine le troisieme jour. Au moment de commencer le traitement, des prelevements de sang ont ete realises et les microtests normalis6s OMS de maturation des schizontes pour la determination de la sensibilite a la chloroquine et a l'amodiaquine ont ete effectues ainsi qu'une epreuve de 48 heures de r6invasion par les mero- zoites pour la determination de la sensibilite a la pyri- methamine. Dans la regle, les enfants ont ete suivis au moyen d'epreuves in vivo pendant 35 jours i compter du debut du traitement. Chez les enfants etudies, 94% des infestations etaient resistantes a la chloroquine in vivo, 76% a l'amodiaquine et 16% a la sulfadoxine/pyrimethamine. Dans tous les groupes, la r6sistance etait surtout de degre RI. Le taux de gu6rison etait de 100% avec l'amodiaquine+sulfadoxine/ pyrimethamine (22 cas) mais ne differait pas sensiblement du taux obtenu avec la sulfadoxine/pyrimethamine seule. Parmi les traitements 6tudies, la sulfadoxine/pyrimethamine avait l'effet le plus rapide (temps moyen de disparition des parasites: 1,85 jour), et tous les traitements entrainaient une augmentation de l'indice gam6tocytaire, qui atteignait son maximum autour desjours 7 et 8. Les r6sultats des epreuves in vitro ont montre que 93% des enfants traites presentaient une chloroquinor6sistance (concentration minimale inhibi- trice > 1,6 gmol/l). En revanche, seuls 35% des enfants traites par 1'amodiaquine presentaient une resistance in vitro; la diff6rence entre les resultats des 6preuves in vivo et in vitro (75,8%) pourrait etre due a une r6infestation ou a des facteurs pharmacocinetiques. Comme dans les etudes prec6dentes, une CIso de pyrimethamine dans du RMPI 1640 superieure ou egale a 3 jsmol/l etait associee a une resistance in vivo a la sulfadoxine/pyrim6thamine. nI n'y avait pas de correlation entre la sensibilit a la pyrimethamine in vitro et la sensibilit6 a I'amodiaquine in vivo. Comme la resistance a l'amodiaquine est fr6quente dans la r6gion 6tudiee, ce medicament ne semble pas convenir pour remplacer la chloroquine. I1 n'est cependant pas in- vraisemblable que 1'effet therapeutique de l'association amodiaquine+sulfadoxine/pyrimethamine soit superieur a celui de la sulfadoxine/pyrimethamine seule, mais cette hypothese devra etre v6rifiee lors d'un essai contr6l6 randomise portant sur des malades symptomatiques. La possibilite d'utiliser l'association pour retarder I'apparition d'une resistance plus frequente aux deux constituants est examinee. L'article conclut que la sulfadoxine/pyri- m6thamine doit etre administr6e au debut du traitement car elle agit plus rapidement que l'amodiaquine. On souligne egalement la necessite d'etudier d'eventuels effets secondaires associ6s a la r6p6tition de traitements par l'association triple. REFERENCES 1. PENICHPONGSE ET AL. An evaluation of five regimens for the outpatient therapy of falciparum malaria in Thailand 1980-81. Bulletin of the World Health Organization, 60: 907-912 (1982). 2. ScHAPIRA, A. ET AL. The susceptibility ofPlasmodium falciparum to sulfadoxine and pyrimethamine: cor- relation of in vivo and in vitro results. Americanjournal oftropical medicine and hygiene, 35: 239-245 (1986). 3. WATKINS, W. M. ET AL. Effectiveness of amodiaquine as treatment for chloroquine-resistant Plasmodiumfal- ciparum infections in Kenya. Lancet, 1: 357-359 (1984). 4. EDITORIAL. Chloroquine-resistant malaria in Africa. Lancet, 1: 1487-1488 (1985). 5. FERNANDES, A. ET AL. [Sensitivity and specificity of the Dill-Glazko test for the detection of 4-amino- quinolines in urine]. Revista mgdica de Mogambique(in Portuguese) (In press). 6. WHO Technical Report Series No. 529, 1973 (Chemo- therapy of malaria and resistance to antimalarials: report of a WHO Scientific Group). 7. SMALLEY, M. E. & BROWN, J. In vitro demonstration of pyrimethamine resistance of "wild" Plasmodium falciparum in the Gambia. Transactions of the Royal Society of Tropical Medicine and Hygiene, 76: 324-328 (1982). 8. GASC6N, J. ET AL. Chloroquine and amodiaquine resistant falciparum malaria in Rwanda. Lancet, 1: 1072 (1985). 226 A. SCHAPIRA & J. F. L. SCHWALBACH 9. MOLINEAUX, L. & GRAMICCIA, G. The Garki Project: research on the epidemiology and control ofmalaria in the Sudan savannah of West Africa. Geneva, World Health Organization, 1980, pp. 125-131. 10. CHURCHILL, F. C. ET AL. Amodiaquine as a prodrug: importance of metabolite(s) in the antimalarial effect of amodiaquine in humans. Life sciences, 36: 53-62 (1985). 11. SCHAPIRA, A. Concomitant resistance to pyrimetha- mine and cycloguanil of chloroquine-resistant falci- parum malaria from East Africa: an in vitro study of 12 isolates. Transactions of the Royal Society of Tropical Medicine and Hygiene, 78: 359-362 (1984). 12. SPENCER, H. C. ET AL. A new in vitro test for pyrimethamine/sulfadoxine susceptibility of Plas- modium falciparum and its correlation with in vivo resistance in Kenya. Bulletin of the World Health Organization, 62: 615-621 (1984). 13. CURTIS, C. F. & OTOO, L. N. A simple model of the build-up of resistance to mixtures of antimalarial drugs. Transactions of the Royal Society of Tropical Medicine and Hygiene, 80: 889-892 (1986). 14. HATTON, C. S. R. ET AL. Frequency of severe neutro- penia associated with amodiaquine prophylaxis against malaria. Lancet, 1: 41 1-414 (1986).