TMIH645. Summary. keywords Plasmodium falciparum, Plasmodium vivax, epidemic, forest, ICT malaria P.f/P.v-test, Central India

Transcription

1 TMIH645 Tropical Medicine and International Health volume 5 no 11 pp november 2000 Field evaluation of the ICT Malaria P.f/P.v immunochromatographic test for diagnosis of Plasmodium falciparum and P.vivax infection in forest villages of Chhindwara, central India Neeru Singh 1,Ajay Saxena 1 and Neena Valecha 2 1 Malaria Research Centre (Field Station), Jabalpur, India 2 Malaria Research Centre, Delhi, India Summary A rapid new immunochromatographic test (ICT malaria P.f/P.v) for diagnosis of Plasmodium falciparum and P.vivax was evaluated against thick blood smears in forest villages of Chhindwara, Madhya Pradesh, where both Plasmodium falciparum and P.vivax are prevalent. 344 symptomatic patients (Gond ethnic tribe) in five villages were screened by field staff of the Malaria Research Centre in October For P.falciparum, the ICT was 97.5% sensitive and 88% specific, with a positive predictive value (PPV) of 87.6% and a negative predictive value (NPV) of 97.6%. For P.vivax the sensitivity was only 72%, the specificity 99%, with a PPV of 92% and an NPV of 96%. Although a negative test result was inadequate to exclude parasitaemia 300/ l for P.falciparum and 1500/ l for P.vivax, the test is potentially useful in remote areas. keywords Plasmodium falciparum, Plasmodium vivax, epidemic, forest, ICT malaria P.f/P.v-test, Central India correspondence Dr Neeru Singh, Malaria Research Centre (Field Station), Medical College Buildings, Jabalpur , India. oicmrc@bom6.vsnl.net.in Introduction The urgent need for a simple and cost-effective diagnostic test to overcome the deficiencies of light microscopy has been recognized for a long time (WHO 1996). Rapid and specific diagnostic tests to identify individuals infected with malaria are very important for control of the disease. During the past five years several new diagnostic techniques for Plasmodium falciparum malaria have been developed (Bustos et al. 1999). The commercially available methods using HRP-2 antigen detection (ParaSight TM F, Becton Dickinson, Sparks, MD, USA), immunochromatography (ICT malaria P.f., Sydney, Australia) and lactate dehydrogenase enzyme (Opti MAL test, Flow Inc., USA) have a level of sensitivity comparable to or better than conventional microscopy (Kilian et al. 1997; Singh et al. 1997a; Ende et al. 1998). A major limitation of antigen detection tests has been their inability to detect malaria caused by Plasmodium vivax (Singh et al. 1997b), with the exception of Opti MAL (Palmer et al. 1998). Recently, another rapid immunochromatographic test, AMRAD s ICT malaria P.f/P.v, was developed for rapid diagnosis of P.falciparum and P.vivax. The test kit is identical to the P.falciparum-specific HRP-2-test described previously (Garcia et al. 1996), but a second monoclonal antibody directed against a panmalarial antigen is added. To assess the usefulness of this test in field, this kit was used in difficult-to-access forest villages of Chhindwara, where an outbreak of malaria was recorded in 1998 (Singh 1999), to detect and differentiate P.falciparum from P.vivax infections. Materials and methods To evaluate the performance of ICT P.f/P.v-test for P.falciparum and P.vivax infections against thick blood smear, a mobile field lab was established in five epidemicaffected forest villages of Mohkhed PHC, district Chhindwara, during the month of October The study was approved by the Ethics Committee of the Malaria Research Centre, Delhi Blackwell Science Ltd 765

2 Study Area The villages are surrounded by scrub jungle intersected by streams and low rocky hills. The streams have many rocky pools and pits that support heavy breeding of two potential vectors, i.e. Anopheles culicifacies and An. fluviatilis. During the rainy season, most villages remain cut off from the main road and from each other for 2 3 months. Precipitation peaks from mid-june to mid-september (rainfall mm). 85% of the population belong to the Gond tribe. People are poor and live in small, dark, mud-plastered huts without electricity which are scattered in the fields and forests. The main occupation is agriculture, wood cutting and wage labour in forest nurseries and road construction. A malaria epidemic broke out in October November 1998 which was investigated on request of the Government of Madhya Pradesh. Both P.vivax and P.falciparum are prevalent. Chloroquine-resistant P.falciparum is common (Unpublished data). Sample collection, immunochromatographic testing and microscopy All fever cases were screened independently for malaria using both the thick blood smear and the rapid diagnostic method to evaluate the performance of the test under field conditions. Blood was obtained by finger prick for the ICT P.f/P.v and thick smear before patients received treatment. History of fever, symptoms and any drugs taken were recorded in each case. The test procedure and interpretation of the test results was explained by a Research Scientist to two Field Laboratory Assistants (FLAs). All specimens were tested on site with the ICT by the FLAs, who were blinded to the clinical status and blood smear results of the subjects. Both FLAs performing the ICT assay exchanged their test card after reading so a second independent interpretation of this test could be recorded. To assess consistency of interpretation, the ICT result cards were read again at the end of the diagnostic trial. AMRAD s ICT malaria P.f/P.v (Australia) is a rapid, in vitro immunodiagnostic test for the detection of circulating P.falciparum and P.vivax antigens in whole blood. The test uses two antibodies which have been immobilized as two separate lines across a test strip. One antibody (test area 1) is specific for the histidine rich protein 2 antigen of P.falciparum (Pf HRP-2). The other antibody (test area 2) is specific for a malarial antigen which is common to both P.falciparum and P.vivax species. 15 l whole blood is applied to a sample pad impregnated with colloidal goldlabelled antibodies directed against the two malarial antigens. When a malaria-positive sample is applied, malarial antigens bind to the gold-coupled antibodies in the pad, and the immune complexes formed migrate along the test strip where they are captured by the immobilized antibodies. When capture occurs, a pink line forms in area 1 and/or 2 of the test window. When a malaria-negative sample is applied, these lines do not form. A control line will appear, in area 3 of the test window, if the test has been performed correctly. The interpretation of the test strip results is as follows: positive for P.falciparum, one control band plus one/two test bands; positive for P.vivax, one control band plus one test band (bottom line); malaria negative, one control band at the top of the test strip. All patients infected with P.falciparum were given three standard oral doses of 1500 mg sulfadoxine plus 75 mg pyrimethamine, followed by a single dose of 45 mg primaquine. P.vivax cases were given 1500 mg chloroquine in divided doses (600 mg on day 0, 600 mg on day 2 and 300 mg on day 3) followed by 15 mg primaquine daily for 5 days. Infants and children were given proportionally less. Infants and pregnant women were not given primaquine as per NAMP (National Anti-Malaria Programme). Simultaneously, thick blood films were prepared and stained with JSB stain (Jaswant Singh & Bhattacharya 1944). The blood films were examined by an experienced microscopist in the field laboratory without reference to the results of the ICT. The microscopist counted 200 white blood cells (WBC) before classifying a slide as negative. The results of both microscopy and ICT malaria P.f/P.v were matched by an independent expert on the site who was blinded to the patient s clinical status, microscopy and ICT results. The microscopic diagnosis was used as the gold standard to determine the sensitivity and specificity of the rapid diagnostic test. Negative slides were re-examined by counting up to 500 WBC by an expert microscopist in the laboratory of the Malaria Research Centre at Jabalpur if the patient was having severe symptoms, the corresponding ICT results had been positive or a P.vivax infection was diagnosed. This microscopist was also blinded to the previous microscopy and ICT P.f/P.v results. Parasite densities were calculated according to the standard method (parasite/ l no. of asexual parasites WBC count/no. of WBC counted). Data analysis Sample sizes were not calculated as this study provided a preliminary estimate of the ICT test sensitivity and specificity for P.vivax and P.falciparum measured against the standard of thick blood films in an epidemic-affected forested belt. To calculate sensitivity and specificity, the first ICT reader s interpretation was compared with field microscopy. The figures for specificity, sensitivity, predictive values and efficiency were calculated as suggested by Tjitra et al. (1999). Briefly, performance indices were calculated for malaria as a whole (diagnosis of either species), P.falciparum malaria Blackwell Science Ltd

3 Table 1 Malaria prevalence in forest villages of Chhindwara in October 1999 Patients groups Blood slide Malaria Mixed infection Slide Slide P.falciparum (years) examined positive P.vivax P.falciparum of Pv Pf positivity rate falciparum rate percentage Pregnant women Total (including mixed infection) and P.vivax malaria. The variables measured were number of true positives (TP), number of true negatives (TN), number of false positives (FP) and number of false negatives (FN). Sensitivity was calculated as TP/(TP FN). Specificity was calculated as TN/(TN FP). The positive predictive value (PPV) was calculated as TP/(TP FP) and negative predictive value (NPV) was calculated as TN/(FN TN). Test efficiency, the proportion of all tests that gave a correct result, was defined as (TP TN)/number of all tests. When the data was analysed for malaria as a whole, results were considered false positive if P.falciparum were detected in thick smear and ICT P.f/P.v detected P.vivax and vice versa. As mixed infections are read as P.falciparum alone, when analysing test performance for the detection of P.vivax, mixed infections detected by microscopy were considered true negative if immunochromatographic testing detected P.falciparum and true positive if immunochromatographic testing detected P.vivax. As sexual stages do not cause fever, samples that were HRP-2 or antigen positive by immunochromatographic test but asexual parasite negative and gametocyte positive on microscopy were considered false positive. Results 344 patients with fever were suspected of having malaria (age 6 months to 65 years). 207 (60%) were found to be infected, 47 with P.vivax (13.6%), 152 with P.falciparum (44.2%) and eight (2.3%) with both P.vivax and P.falciparum. Table 1 shows a breakdown of malaria cases in age groups. The results of parasite detection by microscopy and immunochromatographic testing were compared in Table 2. The test was sensitive (97.5%) and specific (88%) for the diagnosis of falciparum malaria with a PPV and an NPV of 87.6 and 97.6%, respectively (Table 3). Efficiency was 92.4%. The corresponding sensitivity for the diagnosis of vivax malaria was 72% which was significantly lower than P.falciparum (P ). The specificity for P.vivax was 99% and also differed significantly from P.falciparum (P ). However, PPV (92%), NPV (96%) and efficiency (95.3%) of P.vivax were not significantly different from P.falciparum. 17 P.falciparum and three P.vivax were false positives by rapid test and microscopy. Most of these cases, which had no parasites on microscopy but an ICT P.f/P.v result indicating P.falciparum or P.vivax, gave a history of chloroquine treatment in the preceding 2 weeks. Re-examination of these cases in the laboratory revealed that Table 2 Comparison of ICT P.f./P.v. and microscopic examination of thick blood smear for malaria for 344 patients with a presumptive clinical diagnosis of malaria ICT P.f./P.v. Microscopic result P.falciparum P.vivax Negative Total P.falciparum, asexual ( sexual)* P.vivax, asexual ( sexual) P.falciparum + P.vivax ( sexual) P.falciparum sexual only Negative Total * sexual with or without sexual stage parasites Blackwell Science Ltd 767

4 Table 3 Performance of ICT P.f./P.v. relative to those of microscopy for 344 patients with a presumptive clinical diagnosis of malaria Microscopic result Sensitivity % Specificity % PPV % NPV % Efficiency (CI 95%) (CI 95%) (CI 95%) (CI 95%) % Total ( ) ( ) ( ) ( ) P.falciparum, asexual ( sexual)* ( ) ( ) ( ) ( ) P.vivax, asexual ( sexual) ( ) ( ) ( ) ( ) * sexual, with or without sexual stage parasites. there was no disagreement in the results from field microscopy. Five cases were positive for P.vivax by microscopy (Table 2) but tested positive for P.falciparum by ICT P.f/P.v-test. Upon re-examination, they revealed scanty rings in two cases and only gametocytes of P.falciparum in one case along with P.vivax. Thus three of these five cases were mixed infections. There was no difference between the performance of the FLAs with regard to the specificity, sensitivity or positive/negative predictive values of the ICT assay. The agreement between readings of the ICT test results by both FLAs was excellent. Only three of 147 P.falciparum (2%) readings were discordant. These were declared negative by one FLA but were, in fact, faintly positive. There was no disagreement for P.vivax cases or mixed infections of P. vivax and P.falciparum. Consistency of interpretation was also excellent when the same person re-read the card at the end of the study, with 99.3% (146/147) agreement with the original score and only one outlier. Reading accuracy remained unchanged and the high level of inter-reader agreement was sustained for more than 6 months. Figure 1 summarizes the data on the threshold of parasite detection by the ICT malaria P.f/P.v-test relative to thick blood smear. With parasitaemia of 40/ l, the standard blood film examination was 100% sensitive for both P.vivax and P.falciparum. However, the ICT malaria P.f/P.v-test was less sensitive and specific for both P.vivax and P.falciparum. The threshold of detection of P.falciparum using this test was estimated to be 300/ l, as the ICT was 100% sensitive for P.falciparum with parasitaemia of 300/ l, but sensitivity dropped to 89.8% (P ) for lower parasitaemia levels. For P.vivax, the ICT malaria P.f/P.v-test did not identify eight patients out of 47 with a level of parasitaemia 1500 parasites/ l (four patients had 500/ l, two had 750/ l Sensitivity % Parasites/µl 6000 Figure 1 Sensitivity and specificity of ICT malaria P.f/P.v test for diagnosing P.falciparum, P.vivax and overall infections relative to standard thick blood film examination (parasites/ l) Blackwell Science Ltd

5 and two had 1500/ l). Thus, for P.vivax the sensitivity of the ICT test was only 60.3% with a parasite density of 1500/ l and up to 100% when parasitaemia was 1500/ l (P 0.05). We tried to determine the correlation between the depth of the colour reaction in the ICT P.f/P.v and the density of parasitaemia of P.falciparum and P.vivax as determined by the thick blood smear. Generally, no agreement was found, except when parasitaemia was very low, i.e. 100 parasites/ l for both P.falciparum and P.vivax, when the reaction was represented by a much fainter test line. Five ICT P.f/P.v malaria tests were invalidated, as they showed no band formation on the positive control of test card. Discussion In remote forest villages control of malaria is a great challenge. Strengthening national capabilities to provide early diagnosis and treatment both within and outside the health services is of highest priority in WHO s action plan for malaria control (WHO 1996). In view of this we tested the recently developed ICT P.f/P.v at village level and compared the results with traditional blood film examination. Field teams were able to carry out the test without any difficulty and achieved excellent results. Currently, management of malaria by the Indian National Anti-Malaria Programme (NAMP) is based on presumptive treatment of fever cases. Because symptoms lack specificity, most diagnoses are inaccurate, resulting in both overtreatment with antimalarial agents and undertreatment of those with other illnesses (Taylor & Mtambu 1986). This places the population at undue risk of side-effects with no resulting benefit from the treatment and wastes drug supplies. In this study 117 microscopy-negative patients out of 344 were saved from unnecessary antimalarial therapy as a result of the ICT P.f/P.v malaria test and referred to the Government Hospital to obtain appropriate treatment for their illness. The ICT s decline in sensitivity at parasitaemia levels 300/ l for P.falciparum is of concern, although this is well below the danger range for complicated or severe malaria (Wongsrichanalai et al. 1999). However, it does prevent timely diagnosis of patients with low parasitaemia. This cut-off point in sensitivity is consistent with published ICT malaria P.f test studies (Kilian et al. 1997; Pieroni et al. 1998) and other Pf HRP-2 based test studies (Humar et al. 1997; Kilian et al. 1997; Pieroni et al. 1998), which detected low parasitaemias through microscopic or molecular means. Improvement in the sensitivity of Pf HRP-2 based detection of parasitaemia using next-day repeated testing has been reported by Pieroni et al. (1998) and Karbwang et al. (1996). Unfortunately, in this study follow-up of patients was not done. For P.vivax it was not possible to define the detection threshold with certainty because of the relatively few patients with vivax malaria and low parasitaemia, and thus the sensitivity in detecting suspected cases of vivax malaria by the ICT assay was lower than that of a microscopist, but results were obtained in a shorter time and with less training or expertise. A second generation of ICT malaria P.f/P.v is required to detect parasitaemia at lower parasite density. From a clinical perspective, failure to diagnose P. vivax infection when parasitaemia is 1500 parasites/ l in this study and other reports (Tjitra et al. 1999) is important. The reasons are unknown. Unless a solution is implemented for these rare but recurrent diagnostic failures which are reported for P.vivax in this study as well as for P.falciparum in others (Beadle et al. 1994; Wongsrichanalai et al. 1999), programme managers using antigen-detection tests must bear in mind that a negative test greatly reduces the probability, but does not rule out, a parasitaemia of 1500/ l. Furthermore, the reasons for the false positivity of ICT P.f/P.v for P.vivax in three cases should be the subject for further studies such as molecular testing or careful follow-up of patients. The ICT P.f/P.v-test appears to show little cross-reactivity with infections caused by other species of human malaria parasite, although the data to support this are limited (WHO 1996). The two microscopically detected P.vivax cases (five by field microscopy) shown as P.falciparum by ICT P.f/P.v could have been due to mixed infection with P.falciparum (as reexamination revealed three cases were infected with scanty P.falciparum) because low-level P.falciparum infections appear to have been obscured or overlooked in mixed infections dominated by P.vivax. Alternately, P.falciparum parasites can be sequestered and not present in the peripheral blood (Singh et al. 1997b). Currently, multicentre controlled trials in various epidemiological settings are underway to accurately establish performance characteristics and to generate data to assess the relative utility of the available combined antigen-detection tests in areas of different geographical terrain. Despite these limitations, this test can be used as an epidemiological tool because, in areas where both P.vivax and P.falciparum are prevalent, it can be used to identify the Plasmodium species infecting the patients in each village quickly and it allows public health workers to deliver the appropriate chemotherapy rather than just give chloroquine to everyone with malarial symptoms. This is particularly important where the preferred therapies for the two infections are different. However, despite its advantages over microscopy and clinical diagnosis, the cost of the ICT malaria P.f/P.v (US$ 1.2 per test) is too high (Tjitra et al. 1999) and prevents its widespread use in malaria-endemic areas of developing countries where many patients need a fever screen. Further use of the test will result in savings in time and expensive 2000 Blackwell Science Ltd 769

6 drugs. Moreover, in a field situation like ours, the reliability of each test, the ease with which it is performed and the rapidity of yielding results in a large number of sick patients are all relevant factors for consideration apart from cost. The stability of the test card has been good as these test strips have not lost their pink colour after 6 months at an average temperature of 38 C and an average humidity of 70%. In such areas microscopy is not readily accessible and it can take 4 6 weeks before slide results are available (Singh et al. 1996). Hence, the delay in the diagnosis and treatment of cases contributes to the continuing transmission. On the contrary, this test can be used for on-the-spot diagnosis and treatment of both P.vivax and P.falciparum infection by nonmedical staff in urgent or epidemic situations and thus could help to attain the goals of the Roll Back Malaria initiative. Acknowledgements We thank Becton Dickinson, particularly Dr Jo. Perrone and Sh. K. Goswami, for the test kits and financial support. Dr S.K. Subbarao, Director of the Malaria Research Centre, Delhi, helped in various ways. Dr S.S. Mishra, Asst. Prof. Dept. of Physiology, Medical College Jabalpur, helped with the statistical analysis. We are grateful to the team of the Malaria Research Centre (Field Station), Jabalpur, for their hard work on difficult terrain. References Beadle C, Long GW, Weiss WR et al. (1994) Diagnosis of malaria by detection of Plasmodium falciparum HRP-2 antigen with a rapid dipstick antigen-capture assay. Lancet 343, Bustos DG, Olveda RM, Negishi M & Kurimura T (1999) Evaluation of a new rapid diagnostic test Determine TM Malaria P.falciparum against standard blood film, ICT malaria Pf TM and Parasight TM F. Japanese Journal of Tropical Medicine and Hygiene 27, Ende JVD, Vervoort T, Gompel AV & Lynen L (1998) Evaluation of two tests based on the detection of histidine rich protein-2 for the diagnosis of imported Plasmodium falciparum malaria. Transactions of the Royal Society of Tropical Medicine and Hygiene 92, Garcia M, Kirimoama D, Marlborough D, Leafaria J & Rieckman KH (1996) Immunochromatographic test for malaria diagnosis. Lancet 347, Humar A, Ohrt C, Harrington MA, Pillai D & Kain KC (1997) ParaSight-F Test compared with the polymerase chain reaction and microscopy for the diagnosis of Plasmodium falciparum malaria in travelers. American Journal of Tropical Medicine and Hygiene 56, Karbwang J, Tasanor O, Kanda T et al. (1996) ParaSight-F test for the detection of treatment failure in multidrug resistant Plasmodium falciparum malaria. Transactions of Royal Society of Tropical Medicine and Hygiene 90, Kilian AHD, Mughusu EB, Kabagambe G & Sonnenburg FV (1997) Comparison of two rapid, HRP-2 based diagnostic tests for Plasmodium falciparum. Transactions of the Royal Society of Tropical Medicine and Hygiene 91, Palmer CJ, Lindo JF, Klaskala WI et al. (1998) Evaluation of the Opti MAL test for rapid diagnosis of Plasmodium vivax and Plasmodium falciparum malaria. Journal of Clinical Microbiology 36, Pieroni P, Mills CD, Ohrt C, Harrington MA & Kain KC (1998) Comparison of the ParaSight TM F test and the ICT malaria Pf tests TM with the polymerase chain reaction for the diagnosis of Plasmodium falciparum malaria in travellers. Transactions of the Royal Society of Tropical Medicine and Hygiene 92, Singh J & Bhattacharya LM (1944) Rapid staining of malarial parasites by a water soluble stain. Indian Medical Gazette 79, Singh N, Singh OP & Sharma VP (1996) Dynamics of malaria transmission in forested and deforested regions of Mandla district, Central India. Madhya Pradesh. Journal of the American Mosquito Control Association 12, Singh N, Valecha N & Sharma VP (1997a) Malaria diagnosis by field workers using an immunochromatographic test. Transactions of the Royal Society of Tropical Medicine and Hygiene 91, Singh N, Singh MP & Sharma VP (1997b) The use of a dipstick antigen-capture assay for the diagnosis of Plasmodium falciparum infection in a remote forested area of central India. American Journal of Tropical Medicine and Hygiene 56, Singh N (1999) Usefulness of Dipstick Test (Parasight TM F) for High Risk Groups for Plasmodium Falciparum in India. Paper presented in Informal consultation on Malaria Diagnostics at the Turn of the Century October, Geneva, Switzerland. Taylor P & Mtambu SL (1986) A review of the malaria situation in Zimbabwe with special reference to the period Transactions of Royal Society of Tropical Medicine and Hygiene 80, Tjitra E, Suprianto S, Dyer M, Currie BJ & Anstey NM (1999) Field evaluation of the ICT malaria P.f / P.v immunochromatographic test for detection of Plasmodium falciparum and Plasmodium vivax in patients with a presumptive clinical diagnosis of Malaria in Eastern Indonesia. Journal of Clinical Microbiology 37, Wongsrichanalai C, Chuanak N, Tulyayon S et al. (1999) Comparison of a rapid field immunochromatographic test to expert microscopy for the detection of Plasmodium falciparum asexual parasitaemia in Thailand. Acta Tropica 73, World Health Organization (1996) 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. Bulletin of the World Health Organization 74, Blackwell Science Ltd