The comparative efficacy of chloroquine and sulfadoxine-pyrimethamine for the treatment of uncomplicated falciparum malaria in Kampala, Uganda Moses R. Kamya’*, Grant DorseyZ, Anne Gasasira I, and Philip J. Rosenthal’ Grace Ndeezil, Juliet N. Babirye’, Sarah G. Staedke’ ‘Makerere University Medical School, Kampala, Uganda; ‘Department of Medicine, San Francisco General Hospital and The University of California, San Franc&o, USA Abstract Chloroquine (CQ) remains the first-line treatment for uncomplicated malaria in much ofAfrica despite the growing problem of resistance to this drug. Sulfadoxine-pyrimethamine (SP) is often used after CQ treatment failure and has replaced CQ as the first-line treatment in parts ofAfrica. To compare the efficacy of these 2 regimens, we evaluated, in March-August 1999, clinical and parasitological responses over 28 days in 214 children and adults from Kampala, Uganda, with uncomplicated falciparum malaria. Compared to SP, significantly more patients treated with CQ developed early or late clinical failure (54% vs 1 l%, P C 0.001) andparasitoIogica1 failure (72% vs 30%, P < 0.00 1) during 14 days of follow-up. The risk of treatment failure occurring after day 14 was similar between the 2 treatment groups. Among those treated with CQ, children aged < 5 years were at higher risk of clinical failure than older individuals (76% vs 28%, P < O.OOl), anassociationnot seenwithSP (11% vs lo%, P = 0.91). Ahhoughearlyparasite clearance was significantly better in the SP group (P = O.OOl), fever clearance at day 3 was the same (CQ 85%, SP 86%). These and other recent findings suggest that consideration be given to replacing CQ as the first-line therapy for uncomplicated malaria in Uganda, particularly in young children. Keywords: malaria, Plumodium fakipamm, chemotherapy, chloroquine, sulfadoxine-pyrimethamine, clinical trial, drug resistance, Uganda Introduction Antimalarial drug resistance is a large and growing problem. Several East tican studies have reported incidences of parasitological resistance of Piasmodium fukipamm to chloroquine (CQ) ranging from 50% to over80% (BRANDLING-BENNETT~~ aE.,1988;Sm~0~ et al., 1988; WATKINS et al., 1988; BAYOUMI eial., 1989; FOWLER et al.. 1993: PREMII et al.. 1993: WOLDAY etal., 1995). .The kmergknce df CQ’ resistance has been associated with increased malaria specific mortality, especially in children (TRAPE et aZ., 1998). However, most African countries continue to use CQ as the first- line agent to treat uncomplicated malaria (BIGLAND et aE., 1998). The basis for a programmatic decision to disc&tinue the use of CQ as &e first-line drug for malaria remains unclear (BARAT et al.. 1998). Althourrh CQ treatment failures &e common; CQ is safe Gd inexpensive, and it continues to be at least partially effective in semi-immune African populations (HOFF- MAN et al, 1984). It is often argued that to limit cost and toxicity, and to forestall the development of drug tesis- tance, alternatives to CQ should be withheld until they are absolutely required (WHO, 1994). Recent studies in Uganda have identified high rates of resistance of malaria parasites to CQ. A study from western Uganda reported 58% clinical and 77% pata- sitological failure among children aged < 5 years (Ugan- dan Ministry of Health, 1997, unpublished data). A study in Kampala revealed 62% clinical and 86% para- sitological failure in children aged < 5 years, although failure rates were significantly lower in older individuals (DORSEY et al., 2000). Sulfadoxine-pyrimethamine (SP) is currently the second-line agent for uncomplicated~malaria in U.&da. Resistance to SP in East Africa has eenerallv been found to be uncommon (NWANYANWU et>l., 199k; FALASCHI & ANSALONI, 1997; VERHOEFF et aZ., 1997) but has begun to emerge in areas with increased drug pressure (WOLDAY et al.. 1995; BONN et al., 1996). In Uganda, limited data primarily- from asymptomatic children re- vealed uarasitoloeical failure rates iti viwo of O-5% during ;be late lu98Os and 1990s (KAMUGISHA et al., 1994;NEvILL et al., ~~~~;NDYOMUGYENYI & IMAG Address for correspondence: Moses R. Kamya, Department of Medicine, Makerere University, P.O. Box 7072, Kampala, Uganda; hone +256 041 541188; fax $256 041 533531, e-mait maParia@infocom.co.ug NUSSEN, 1997). A more recent study in western Uganda reported a 12% clinical failure rate of SP for 122 children with symptomatic malaria (JELINEK et al., 1999). In summary, recent data suggest that resistance to CQ in Uganda is unacceptably high. Similar findings in other East African countries have led to recent decisions to change the recommendation for first-line therapy of uncomulicated malaria from CO to SP (BLOLAND et al., l-993). However, such a chaige may increase cost and toxicity, and may contribute to the selection of multidrug-resistant parasites. In addition, CQ therapy may offer clinical benefits beyond its antiparasitic activ- ity, due to a postulated antipyretic effect (BOIANG et al., 1998). In 2 studies from West Africa, children treated with SP were more likelv than those treated with CO to develbp early clinical filure despite improved parasite clearance in those treated with SP, suggesting that an antipyretic effect of CQ was clinically relevant (MILLER et al., 1996; ONYIORAH e6 aZ., 1996). To compare the practical utility of CQ and SP in an urban population in Uganda, we evaluated clinical and parasitological out- comes following treatment with the 2 agents in patients presenting with uncomplicated falciparum malaria in Kampala. Materials and Methods Study site The study was conducted between March and August 1999 at the Old Mulago Hill Dispensary in Kampala, Uganda. Malaria is meso-endemic in Kampala, occur- ring perennially with peaks during the 2 rainy seasons (Ugandan Ministry of Health, unpublished data). Old Mulago Hill Dispensary is an outpatient clinic that provides primary care for patients from lower socio- economic areas of Kampala. Services at the dispensary were provided free of charge. Smdypartica’pants andprotocol Consecutive satients with svmntoms of acute uncom- plicated malarii and a posi&e*screening thick blood smear (stained with 10% Lcishman’s stain for 10 mini were referred for study enrolment. Patients were enrolled if they met the following inclusion criteria: (i) age 3 6 months, (ii) an elevated temperature at presentation (2 38.O”C tympanic or 375°C axillary) ot a history of fever in the previous 48 h, (iii) P. fczlcipamm mono- infection with 2 2000 asexual parasites/&L measured by thick blood smear, (iv) absence of other causes of fever (based on the clinical judgement of the study physician),