INTRODUCTION MATERIALS AND METHODS

Transcription

1 Am. J. Trop. Med. Hyg., 76(3), 2007, pp Copyright 2007 by The American Society of Tropical Medicine and Hygiene WHICH MALARIA RAPID TEST FOR MADAGASCAR? FIELD AND LABORATORY EVALUATION OF THREE TESTS AND EXPERT MICROSCOPY OF SAMPLES FROM SUSPECTED MALARIA PATIENTS IN MADAGASCAR ARSÈNE RATSIMBASOA, ARTHUR RANDRIAMANANTENA, ROGELIN RAHERINJAFY, NOÉLINE RASOARILALAO, AND DIDIER MÉNARD* Epidemiology Unit, and Malaria Unit Research, Institut Pasteur de Madagascar, Antananarivo, Madagascar Abstract. Field and laboratory studies were carried out in October and November 2005 to provide a comparative evaluation of the performance of three rapid malaria detection tests, two of which were recently introduced (the CareStart Malaria test and the SD Malaria Antigen Bioline test) and the well-known OptiMAL-IT test. Compared with microscopy, the sensitivity of the three tests to detect Plasmodium falciparum malaria was 97% for the CareStart Malaria test, 89.4% for the SD Malaria Antigen Bioline test, and 92.6% for the OptiMAL-IT test. The three tests were less sensitive in detecting non-p. falciparum infections, and the sensitivity decreased at levels of parasitemia 500 parasites/ L for P. falciparum and 5,000 parasites/ L for other Plasmodium spp. On the basis of World Health Organization recommendations, only the CareStart Malaria test and the OptiMAL-IT test had sensitivities greater than 95% for samples with parasitemias 100 parasites/ L. INTRODUCTION * Address correspondence to Didier Menard, Malaria Research Unit, Institut Pasteur de Madagascar BP 1274, Antananarivo, Madagascar. dmenard@pasteur.mg 481 In Madagascar, malaria remains a major public health problem and is the primary cause of morbidity. In 2003, malaria accounted for 18.8% of all outpatient cases (2,114,400 suspected malaria cases including 740,000 children less than five years of age). 1,2 Health centers in 111 districts are well distributed over the whole island, but we observed that these centers are lacking in material and human resources. The malaria diagnostic approach currently used in these centers is based solely on clinical diagnosis. However, because the symptoms of malaria are very non-specific and overlap with those of other febrile illnesses, 3 a malaria diagnosis based on clinical grounds alone is unreliable. Microscopic examination of a thick blood film is the standard method for malaria diagnosis and it is only one currently used in provincial or district hospitals in Madagascar or in some private laboratories in the capital of Antananarivo. This method is relatively simple and has low direct costs, but it requires microscopes and trained microscopists; its reliability is also questionable, particularly at low levels of parasitemia and in the interpretation of mixed infections. 4,5 The new Madagascar national policy for the fight against malaria was elaborated in 2005, and it will be supported by resources from the Global Fund. Its main goal nationally is to reduce morbidity and mortality caused by malaria, especially by reducing transmission in unstable malaria-endemic areas such as in the central highlands and the semi-arid southern region, by reducing mortality in areas of stable malaria in the coastal zones, by improving malaria case management through the implementation of biologic diagnosis based on rapid diagnosis tests (RDTs), and by rapid treatment of positive cases with combinations of artesunate plus amodiaquine. At the end of 2005 at the request of the National Malaria Control Program (NMCP), the principal recipient of the Global Fund project round 4 for Madagascar launched an invitation for acquisition of an RDT based on the detection of Plasmodium-specific lactate dehydrogenase (pldh), according to the technical choice of the NMCP. Three RDT suppliers responded to the invitation with products that target Plasmodium falciparum pldh and pan-malaria pldh: the CareStart Malaria test, the SD Malaria Antigen Bioline test, and the OptiMAL-IT test. References in peerreviewed journals were not found for the CareStart Malaria test and the SD Malaria Antigen Bioline test. Only the SD Malaria Antigen Bioline test and the OptiMAL-IT test were found on the list of Manufacturers and Distributors of all known commercially available malaria RDTs from the World Health Organization website ( at the beginning of the study in October The Malaria Research Unit of Institut Pasteur de Madagascar was asked to evaluate the performances of three RDTs for malaria diagnosis. The study was designed to assess sensitivity, specificity, and positive and negative predictive values in field conditions of the three RDTs in symptomatically diagnosed malaria patients compared with microscopy of thick or thin blood films. A laboratory study was also carried out to test absolute parasite detection limits of the RDTs. MATERIALS AND METHODS Rapid diagnostic tests. The following chromatographic tests, which were designed to detect pldh of both P. falciparum and other plasmodia, were included in the study. The OptiMAL-IT test (lots , , and ; DiaMed AG, Cressier sur Morat, Switzerland) were used according to the manufacturer s instructions. Ten microliters of blood was added to one drop (30 L) of the buffer in a sample well and mixed. The OptiMAL-IT test strip was placed in the well, and the blood-buffer mixture was allowed to wick completely up the test strip (up to 10 minutes). The strip was then moved to a wash well filled with buffer and removed after the buffer had cleared the blood color in the background of the strip (approximately 10 minutes). Interpretation of the test results was done immediately. The SD Malaria Antigen Bioline test (lots T5001, T5002, T5003, and T5004; Standard Diagnostics, Suwon City, South Korea) was used according to the manufacturer s instructions. This test contains a membrane strip encased in a flat plastic housing, and the strip is precoated with two polyclonal anti-

2 482 RATSIMBASOA AND OTHERS bodies (one specific to the pldh of P. falciparum and the other to the pan-specific pldh of Plasmodium species). One drop of assay diluent was dispensed into the lysis well. Twenty microliters of whole blood was drawn using a 20- L capillary pipette and mixed with the assay diluent in the lysis well. After one minute, the specimen from the lysis well was transferred to a square sample well using a 10- L capillary pipette. Four drops of assay diluent were added to the well. Interpretation of the test result was done within 20 minutes. The CareStart Malaria test (lots J25IL, J35IL, J45IL, and J55IL; Access Bio Inc., Monmouth Junction, NJ) was used according to the manufacturer s instructions. The test strip was encased in a flat plastic housing. Five microliters of blood were drawn using a capillary pipette and added to the sample well. Four drops of buffer were dispensed into the buffer well. Results from the test were interpreted within 20 minutes. Field study. This study was conducted in primary health centers in rural areas of Mahasolo (western foothill areas of the highlands) and in Saharevo (eastern foothill areas of the highlands) in October and November Malaria transmission in both areas is low and predominantly seasonal. The main vector is Anopheles funestus and the number of infective bites associated with P. falciparum is estimated to be 1 2 per person per year. Febrile patients with typical malaria symptoms coming to health facilities were given complete information concerning the trial and invited to participate in the study. Once informed consent had been given, approximately 3 ml of venous blood was obtained if they had an axillary temperature 37.5 C or history of fever in the past 24 hours. Pregnant women and patients with signs of severe and complicated P. falciparum malaria according to the definition of the World Health Organization (2001) 6 were excluded. The study protocol was reviewed and approved by the expert committee of the NMCP of the Ministry of Health of Madagascar. Freshly collected blood in an EDTA anti-coagulation tube was first used by a technician to prepare thick and a thin blood smears. The blood sample was then given to a second technician who did each of the immunochromatographic tests according to manufacturers instructions. All tests were kept at room temperature and opened just before use to avoid humidity damage. Thick and thin blood smears were made immediately after blood collection and stained with 4% Giemsa for 20 minutes and analyzed by light microscopy by an experimented technician without reference to RDT results. A minimum of 200 consecutive fields was counted in the thick blood film before classifying a slide as negative. Parasites in thick blood films were counted against white blood cells. The parasite density was estimated assuming 8,000 white blood cells/ L of blood. 4,7,8 The results of microscopic examination were reported to the health center physician. If the results were positive for malaria, patients were treated with the first-line treatment according to NMCP recommendations. Laboratory sensitivity study. The EDTA anti-coagulated P. falciparum-infected blood with approximately 10,000 parasites/ L and EDTA anti-coagulated P. vivax-infected blood with approximately 1,000 parasites/ L were collected in the primary health center of Mahasolo during the field study and sent to the Malaria Unit laboratory within eight hours in controlled cool box at 4 C. For both samples, parasitemia was verified by microscopic examination. Plasmodium falciparum infected blood was diluted with uninfected donor blood to produce seven serial dilutions ranging from 10,000 to 150 parasites/ L. Plasmodium vivax infected blood was diluted with uninfected donor blood to produce five serial dilutions ranging from 1,000 to 65 parasites/ L. Mixtures of the two Plasmodium-infected blood samples were also prepared: 1) four serial dilutions containing mostly P. falciparum-infected blood (2,500 parasites/ L ofp. falciparum-infected blood and serial dilutions of P. vivax-infected blood ranging from 1,000 to 125 parasitesd/ L) and 2) four serial dilutions containing mostly P. vivax-infected blood (1,000 parasites/ L ofp.vivax-infected blood and serial dilutions of P. falciparum-infected blood ranging from 1,250 to 150 parasites/ L). Each dilution level was tested in duplicate by two independent technicians with three RDTs. The proportion of serial dilutions testing positive gave the absolute parasite detection limits for each RDT. Data analysis. Data were entered, processed, and analyzed using Epi-Info version software (Centers for Disease Control and Prevention, Atlanta, GA). The chi-square test was used to compare the performance between the three tests and microscopy. P values < 0.05 indicated statistically significant differences. For sensitivity and specificity, RDT results were compared with Giemsa stain microscopy results. Sensitivity was the proportion of positive test results obtained among samples containing malaria parasites by microscopy. Specificity was the proportion of negative test results obtained among samples whose thick blood films were negative. Positive and negative predictive values were the proportion of true-positive results among all positive samples and the proportion of true negative results among all negative samples. RESULTS From October through November 2005, we recruited 194 patients (159 from Mahasolo and 35 from Saharevo) with an age range of 1 57 years (mean ± SD age 15.2 ± 12.8 years); 12.9% were < 5 years of age, 33.5% were 5 14 years of age, and 53.6% were > 15 years of age. The male:female ratio was 0.98:1. The mean ± SD axillary temperature was 37.5 ± 0.9 C (range C) and the mean ± SD parasitemia density was 6,564 ± 26,553 parasites/ L (range ,600 parasites/ L). Seventeen percent of these patients reported that they had had previous anti-malarial therapy before consultation (chloroquine in 57.6%, sulfadoxine-pyremethamine [SP] in 30.3%, tetracycline in 6.1%, chloroquine plus SP in 3%, and quinine in 3%). Microscopy results showed that 90 (46.4%) of 194 had malaria. Plasmodium falciparum was present in 80%, P. vivax in 10%, P. malariae in 6.7%, and mixed infections with P. falciparum and P. vivax in 3.3% of the positive specimens. Details of field study patients positive for Plasmodium spp. by microscopy, the CareStart test, the Malaria, SD Malaria Antigen Bioline test, and the OptiMAL-IT test are shown in Table 1. Diagnostic performances of the three RDTs are shown in Table 2 and sensitivity of the three RDTs for Plasmodium spp. at different levels of parasitemia is summarized in Table 3. On the basis of the World Health Organization recommendation for RDT performance, 7 only two RDTs had a sensi-

3 MALARIA RAPID TESTS FOR MADAGASCAR 483 TABLE 1 Field study patients positive for Plasmodium spp. by microscopy, CareStart Malaria, SD Malaria Antigen Bioline, or OptiMAL-IT, Madagascar, October November 2005* Microscopy No. (%) Negative Pf Non-Pf Negative Pf Non-Pf Negative Pf Non-Pf Negative 104 (53.6) Plasmodium falciparum 68 (75.6) P. vivax 9 (10.0) 1 8 1# 8 1** 8 P. malariae 6 (6.7) P. falciparum plus P. vivax 3 (3.3) P. falciparum gametocytes 4 (4.4) * Pf P. falciparum; Non-Pf non-p. falciparum. SD Malaria Antigen Bioline, 0/2 positive; OptiMAL-IT, 0/2 positive. CareStart Malaria, 5/7 positive; OptiMAL-IT, 1/7 positive. CareStart Malaria, 3/5 positive; SD Malaria Antigen Bioline, 0/5 positive. SD Malaria Antigen Bioline, 1/1 positive; OptiMAL-IT, 0/1 positive. # CareStart Malaria, 1/1 positive; OptiMAL-IT, 1/1 positive. ** CareStart Malaria, 0/1 positive; SD Malaria Antigen Bioline, 1/1 positive. CareStart Malaria, 1/3 positive; OptiMAL-IT, 2/3 positive. CareStart Malaria, 1/1 Pf positive; SD Malaria Antigen Bioline, 0/1 positive. CareStart Malaria, 1/1 Pf positive; OptiMAL-IT, 1/1 Non-Pf positive. tivity greater than 95% for samples with parasitemias 100 parasites/ L: the CareStart Malaria test had a sensitivity of 96.8% (95% confidence interval [CI] %) and the OptiMAL-IT test had a sensitivity of 95.2% (95% CI %). As expected, the three test kits had lower sensitivity for lower parasitemias. For detection of P. falciparum, only the CareStart Malaria test and the OptiMAL-IT test had a sensitivity of 100% for samples with parasitemias > 500 parasites/ L (the sensitivity of the SD Malaria Antigen Bioline test decreased to 90%). For samples with parasitemias 500 parasites/ L, the sensitivity of the three tests decreased to 57.1% for the SD Malaria Antigen Bioline test and to 60% for the CareStart Malaria test and the OptiMAL-IT test. For detection of non-p. falciparum spp., the sensitivity of all three RDTs decreased at parasitemias 5,000 parasites/ L, which was a more rapid decrease in sensitivity than for detection of P. falciparum. Results of pldh threshold detection levels with the three RDTs at different levels of parasitemia in a laboratory study are shown in Table 4. For serial dilutions of P. falciparuminfected blood, the three RDTs gave excellent results with a minimum threshold detection level of 150 parasites/ L, but surprisingly, for these dilutions, no non-p. falciparum bands were observed at all parasitemia levels with the SD Malaria Antigen Bioline test and above 1,250 parasites/ L for the CareStart Malaria test. DISCUSSION In Madagascar, policy makers have decided to implement rapid diagnostic tests to allow a biologic diagnosis where conventional microscopy is not possible. The Malaria Research Unit of Institut Pasteur de Madagascar, with support from the Global Fund project, investigated the field and laboratory performances of three rapid diagnostic tests, two of which were recently introduced, the CareStart Malaria test and the SD Malaria Antigen Bioline test, and the well-known OptiMAL-IT test. Field studies were designed to test the performance of the three RDTs relative to microscopy with samples from symptomatically diagnosed malaria patients at the level of primary health centers. On the basis of recommendations of the expert committee of the NMCP, we excluded pregnant women and patients with signs of severe and complicated P. falciparum malaria. The main limitation of the study was to use microscopy as the gold standard method. It is well known that microscopy is imperfect and its sensitivity decrease with parasite density. 9,10 In practice, microscopy varies a great deal in different settings because of variable techniques of blood film preparation, staining, malaria film reading standards, and the level of expertise of the examining microscopists. 11,12 Therefore, to limit these difficulties, microscopy in our study was performed by experimented technicians who analyzed films independently and were blinded to the RDT results. Moreover, at the end of the study, 10% of the slides were examined by another welltrained technician and no discordant results were found. The three tests detected P. falciparum malaria with high sensitivity relative to microscopy: 89.4% for the SD Malaria Antigen Bioline test, 92.6% for the OptiMAL-IT test, and 97% for the CareStart Malaria test. However, we observed a significant difference between the results of the SD TABLE 2 Diagnostic performance of CareStart Malaria, SD Malaria Antigen Bioline, and OptiMAL-IT tests in detecting Plasmodium spp. in field study patients, Madagascar, October November 2005* Pf (n 172) Non-Pf (n 119) Pf (n 172) Non-Pf (n 119) Pf (n 172) Non-Pf (n 119) Sensitivity (95% CI) 97.0 ( ) 92.3 ( ) 89.4 ( ) 73.3 ( ) 92.6 ( ) 86.7 ( ) Specificity (95% CI) 94.1 ( ) 96.9 ( ) 97.1 ( ) 98.0 ( ) 100 ( ) 98.1 ( ) Positive predictive value (95% CI) 91.4 ( ) 80.0 ( ) 95.1 ( ) 84.6 ( ) 100 ( ) 86.7 ( ) Negative predictive value (95% CI) 97.9 ( ) 99.0 ( ) 93.4 ( ) 96.1 ( ) 95.3 ( ) 98.1 ( ) * Pf Plasmodium falciparum; Non-Pf non-p. falciparum; CI confidence interval.

4 484 RATSIMBASOA AND OTHERS TABLE 3 Sensitivity of CareStart Malaria, SD Malaria Antigen Bioline, and OptiMAL-IT in detecting Plasmodium spp. at different levels of parasitemia in field study patients, Madagascar, October November 2005* Parasitemia/ L of blood Sensitivity (95% CI) Pf Non-Pf Pf Non-Pf Pf Non-Pf > 50, ( ) ND 100 ( ) ND 100 ( ) ND 5,001 50, ( ) 100 ( ) 100 ( ) 100 ( ) 100 ( ) 100 ( ) 501 5, ( ) 77.8 ( ) 90.0 ( ) 77.8 ( ) 100 ( ) 88.9 ( ) ( ) 100 ( ) 57.1 ( ) 50.0 ( ) 78.6 ( ) 100 ( ) < ( ) 50.0 ( ) 80.0 ( ) 50.0 ( ) 60.0 ( ) 50.0 ( ) *CI confidence interval; Pf Plasmodium falciparum; Non-Pf non-p. falciparum; ND no data. Malaria Antigen Bioline test and the CareStart Malaria test (P = 0.005). Detection of P. falciparum antigen by the three tests was also specific: 94.1% for the CareStart Malaria test, 97.1% for the SD Malaria Antigen Bioline test, and 100% for the OptiMAL-IT test. However, we also found a significant difference between the CareStart Malaria test and the OptiMAL-IT test (P =0.002). The OptiMAL-IT test positive predictive value (100%) was significantly higher than the CareStart Malaria test positive predictive value (91.4%; P=0.001) and the SD Malaria Antigen Bioline test positive predictive value (95.1%; P=0.005). We observed no significant differences in the negative predictive values between the three RDTs. The three tests were more sensitive in detecting P. falciparum infections than non-p. falciparum infections. This trend and values for sensitivity, specificity, positive predictive value, and negative predictive value of the OptiMAL-IT TABLE 4 pldh threshold detection level with CareStart Malaria, SD Malaria Antigen Bioline, and OptiMAL-IT at different levels of parasitemia in a laboratory study, Madagascar, October November 2005* Parasitemia/ L of infected blood CareStart Malaria pldh threshold detection level Malaria Antigen Bioline OptiMAL-IT Pf Pv Pf Pan malaria Pf Pan malaria Pf Pan malaria 10, , , , , ,500 1, , , , ,250 1, , , , * No discrepancies among the two technicians was observed. pldh Plasmodium falciparum lactate dehydrogenase; Pf P. falciparum; Pv P. vivax. test were consistent with the published results of similar studies We also found variable sensitivities at different levels of parasitemia, which were similar to results of previous studies. 14,16 19 For P. falciparum, the sensitivity of the three tests decreased at levels of parasitemia 500 parasites/ L. For other Plasmodium spp., the sensitivity of the three RDTs decreased at levels of parasitemia 5,000 parasites/ L. Although samples used in the laboratory study were not true population samples, but artificial samples taken from serial dilutions, our findings supported the field study data. The three tests detected pldh of P. falciparum at low parasitemias, for all three RDTs, there was a higher threshold detection level for non-p. falciparum pldh than for P. falciparum pldh, and the CareStart Malaria test and the OptiMAL-IT tests were more sensitive than the SD Malaria Antigen Bioline test. All three tests showed several false-positive results for detection of P. falciparum and non-p. falciparum infections in samples that were negative by microscopic examination. False-positive results were reported in earlier investigations with RDTs based on pldh detection, but these false-positive results did not appear to be detrimental to the usefulness of malaria tests for clinical settings. It is likely that false-positive results were actually true-positive results that were not detected by microscopy because of sequestration that limited the number of circulating parasites at the time of blood collection, 20 because parasitemias were below the threshold detection level of approximately 50 parasites/ L by microscopy, or because samples were from patients with circulating rheumatoid factors. 21,22 Most of the false-negative test results observed with the three RDTs were in samples with parasitemias < 150 parasites/ L. However, two samples with high parasitemias (500 and 2,520 parasites/ L) showed falsenegative results with the OptiMAL-IT test and the SD Malaria Antigen Bioline test. These findings were also observed in earlier studies without underlying reasons. 18,19 Besides the performance of the three RDTs and recommendations of the World Health Organization, 23 policy makers in Madagascar must also take into account other important considerations such as stability, ease of use, and cost of an RDT. There were no large differences between the tests in terms of ease of use. They were all fairly easy to use and interpret, had stable results, and were simple to store with no cold chain requirement for any tests. For all three tests, the device was individually wrapped in a packet with desiccant. Unfortunately, the stability and shelf-life of the three RDTs were not analyzed in this study. Prices provided by the manu-

5 MALARIA RAPID TESTS FOR MADAGASCAR 485 facturers to the principal recipient of the Global Fund project round 4 for Madagascar were $1.35 (US dollars) for the SD Malaria Antigen Bioline test, $1.29 for the OptiMAL-IT test, and $1.11 for the CareStart Malaria test. Our results support the idea that under certain circumstances, non-microscopic RDTs for detection of plasmodial antigens may become important diagnostic tools and may be a valuable adjunct to clinical assessment of the patient and blood film microscopy. The performance of the newly developed CareStart Malaria test was comparable with the that of the well-established OptiMAL-IT test in detecting P. falciparum and P. vivax infection in symptomatically diagnosed patients. The performance of these two RDTs was significantly better than that of the SD Malaria Antigen Bioline test. Our results raised concerns over the high number of false-negative results and the sensitivity of the SD Malaria Antigen Bioline test for detecting P. falciparum and non- P. falciparum infections. On the basis of our results, recommendations of the World Health Organization, 23 and a report by Ochola and others, 24 further studies should be done in Madagascar to evaluate the performance of other immunochromatographic tests, especially RDTs based on the detection of histidine-rich protein 2 and pan-malarial antigens, which are RDTs more reliable in terms of sensitivity and stability in remote area field conditions. Received May 24, Accepted for publication October 24, Acknowledgments: We thank the patients for participating in the study, and the Ministry of Health of Madagascar for allowing access to malaria patients. We also thank the staff of the principal recipient of the Global Fund project round 4 for Madagascar and the National Malaria Control program, in particular Dr. Paul Richard Ralainirina and Dr. Philémon Tafangy, for their cooperation. Financial support: This work was supported by the three test suppliers and the Global Fund Project for Madagascar round 3 (Community Action to Roll Back Malaria, Grant number: MDG-304-G05-M). Authors addresses: Arsène Ratsimbasoa and Arthur Randriamanantena, Epidemiology Unit, Institut Pasteur de Madagascar, BP 1274, Antananarivo 101, Madagascar. Rogelin Raherinjafy, Noéline Rasoarilalao, and Didier Ménard, Malaria Unit Research, Institut Pasteur de Madagascar, BP 1274, Antananarivo 101, Madagascar, Telephone: , Fax: , dmenard@pasteur.mg. REFERENCES 1. Institut National de la Statistique (INSTAT) (Madagascar) et ORC Macro, Enquête Démographique et de Santé, Madagascar : Rapport de Synthèse. Calverton, MD: IN- STAT et ORC Macro. 2. UNICEF, Multiple Indicator Cluster Survey (MICS) available from madagascar/madagascar.pdf 3. World Health Organization, New Perspectives: Malaria Diagnosis. Report of a Joint WHO/USAID Informal Consultation. Geneva: World Health Organization. WHO./MAL/ World Health Organization, Management of Uncomplicated Malaria and the Use of Antimalarial Drugs for the Protection of Travellers. Report of an Informal Consultation. WHO/MAL/ Molyneux M, Fox R, Diagnosis and treatment of malaria in Britain. BMJ 306: World Health Organization, Monitoring Antimalarial Drug Resistance. Report of WHO Consultation. Geneva: World Health Organization. WHO/CDS/CSR/EPH/ WHO, Malaria Rapid Diagnosis: Making it Work. Meeting report January Manila: World Health Organization. 8. Anonymous, A rapid dipstick antigen capture assay for the diagnosis of falciparum malaria. WHO informal consultation on recent advances in diagnostic techniques and vaccines for malaria. Bull World Health Organ 74: McKenzie FE, Sirichaisinthop J, Miller RS, Gasser RA, Wongsrichanalai C, Dependence of malaria detection and species diagnosis by microscopy on parasite density. Am J Trop Med Hyg 69: Raghavan K, Statistical considerations in the microscopical diagnosis of malaria, with special reference to the role of crosschecking. Bull World Health Organ 34: Kachur SP, Nicolas E, Jean-Francois V, Benitez A, Bloland PB, Saint Jean Y, Mount DL, Ruebush TK, Nguyen-Dinh P, Prevalence of malaria parasitemia and accuracy of microscopic diagnosis in Haiti, October Rev Panam Salud Publica 3: Durrheim DN, Becker PJ, Billinghurst K, Diagnostic disagreement: the lessons learnt from malaria diagnosis in Mpumalanga. S Afr Med J 87: Huong NM, Davis TME, Hewitt S, Huong NV, Uyen TT, Nhan DH, Cong LD, Comparison of three antigen detection methods for diagnosis and therapeutic monitoring of malaria: a field study from southern Vietnam. Trop Med Int Health 7: Iqbal J, Sher A, Hira PR, Malaria in non-endemic Kuwait: detection of very low level of Plasmodium falciparum infections using polymerase chain reaction. Med Princ Pract 7: Moody A, Hunt-Cooke A, Gabbett E, Chiodini P, Performance of the OptiMAL malaria antigen capture dipstick for malaria diagnosis and treatment monitoring at the Hospital for Tropical Diseases, London. Br J Haematol 109: Iqbal J, Khalid N, Hira PR, Comparison of two commercial assays with expert microscopy for confirmation of symptomatically diagnosed malaria. J Clin Microbiol 68: Pattanasin S, Proux S, Chompasuk D, Luwiradaj K, Jacquier P, Looareesuwan S, Nosten F, Evaluation of a new Plasmodium lactate dehydrogenase assay (OptiMAL-IT ) for the detection of malaria. Trans R Soc Trop Med Hyg 97: Palmer CJ, Lindo JF, Klaskala WI, Quesada JA, Kaminsky R, Baum MK, Ager AL, Evaluation of the OptiMAL test for rapid diagnosis of Plasmodium vivax and Plasmodium falciparum malaria. J Clin Microbiol 36: Hunt Cooke A, Chiodini PL, Doherty T, Moody AH, Ries J, Pinder M, Comparison of a parasite lactate dehydrogenase-based immunochromatographic antigen detection assay (OPTIMAL ) with microscopy for the detection of malaria parasites in human blood samples. Am J Trop Med Hyg 60: Bell D, Wongsrichanalai C, Barnwell JW, Ensuring quality and access for malaria diagnosis: how can it be achieved? Nat Rev Microbiol 4: Moody AH, Chiodini PL, Non-microscopic method for malaria diagnosis using OptiMAL IT, a secondgeneration dipstick for malaria pldh antigen detection. Br J Biomed Sci 59: Grobusch MP, Alpermann U, Schwenkl S, Jelinek T, Warhurst DC, False-positive rapid tests for malaria in patients with rheumatoid factor. Lancet 353: WHO with the collaboration of the regional offices for Africa and the Western Pacific, Interim Notes on Selection of Type of Malaria Rapid Diagnostic Test in Relation to the Occurrence of Different Parasite Species. Guidance for National Malaria Control Programmes. Geneva: Roll Back Malaria Department, World Health Organization. WHO. 24. Ochola LB, Vounatsou P, Smith T, Mabaso ML, Newton CR, The reliability of diagnostic techniques in the diagnosis and management of malaria in the absence of a gold standard. Lancet Infect Dis 6: