Indian Journal of Medical Microbiology Home 

[Download PDF]
Year : 2017  |  Volume : 35  |  Issue : 3  |  Page : 376--380

Diagnostic efficacy of microscopy, rapid diagnostic test and polymerase chain reaction for malaria using bayesian latent class analysis

Sreemanti Saha1, Rahul Narang1, Pradeep Deshmukh2, Kiran Pote1, Anup Anvikar3, Pratibha Narang1,  
1 Department of Microbiology, Mahatma Gandhi Institute of Medical Sciences, Wardha, Maharashtra, India
2 Department of Community Medicine, Mahatma Gandhi Institute of Medical Sciences, Wardha, Maharashtra, India
3 National Institute for Malaria Research, New Delhi, India

Correspondence Address:
Rahul Narang
Department of Microbiology, Mahatma Gandhi Institute of Medical Sciences, Sewagram, Wardha, Maharashtra


Introduction: The diagnostic techniques for malaria are undergoing a change depending on the availability of newer diagnostics and annual parasite index of infection in a particular area. At the country level, guidelines are available for selection of diagnostic tests; however, at the local level, this decision is made based on malaria situation in the area. The tests are evaluated against the gold standard, and if that standard has limitations, it becomes difficult to compare other available tests. Bayesian latent class analysis computes its internal standard rather than using the conventional gold standard and helps comparison of various tests including the conventional gold standard. Materials and Methods: In a cross-sectional study conducted in a tertiary care hospital setting, we have evaluated smear microscopy, rapid diagnostic test (RDT), and polymerase chain reaction (PCR) for diagnosis of malaria using Bayesian latent class analysis. Results: We found the magnitude of malaria to be 17.7% (95% confidence interval: 12.5%–23.9%) among the study subjects. In the present study, the sensitivity of microscopy was 63%, but it had very high specificity (99.4%). Sensitivity and specificity of RDT and PCR were high with RDT having a marginally higher sensitivity (94% vs. 90%) and specificity (99% vs. 95%). On comparison of likelihood ratios (LRs), RDT had the highest LR for positive test result (175) and the lowest LR for negative test result (0.058) among the three tests. Conclusion: In settings like ours conventional smear microscopy may be replaced with RDT and as we move toward elimination and facilities become available PCR may be roped into detect cases with lower parasitaemia.

How to cite this article:
Saha S, Narang R, Deshmukh P, Pote K, Anvikar A, Narang P. Diagnostic efficacy of microscopy, rapid diagnostic test and polymerase chain reaction for malaria using bayesian latent class analysis.Indian J Med Microbiol 2017;35:376-380

How to cite this URL:
Saha S, Narang R, Deshmukh P, Pote K, Anvikar A, Narang P. Diagnostic efficacy of microscopy, rapid diagnostic test and polymerase chain reaction for malaria using bayesian latent class analysis. Indian J Med Microbiol [serial online] 2017 [cited 2020 Jul 5 ];35:376-380
Available from:

Full Text


Eliminating the reservoir of infection is an important strategy for malaria elimination. Accurate diagnosis of malaria followed by prompt treatment interrupts transmission and is critical as the Government of India has targeted to eliminate malaria. This can only be achieved through an efficient point-of-care test at a lower cost.

Under the National Vector Borne Disease Control Programme (NVBDCP) of India Microscopy (visualisation of parasites in stained blood smears) is the mainstay of diagnosis while rapid diagnostic test (RDT) is being used in hard to reach areas where workforce is scanty and early diagnosis is an emergency in the wake of falciparum malaria. The limitations of microscopy are well-documented.[1],[2],[3] RDT is used in a limited way because of its perceived high cost and inefficiency in the diagnosis. Polymerase chain reaction (PCR) is presumed to be very efficient in its diagnostic ability, but is costly and needs specialised infrastructure and workforce; hence, its use is limited to research purposes only. At present, PCR is being extensively used as a reference standard in efficacy studies for antimalarial drugs, vaccines and evaluation of other diagnostic agents.[4],[5],[6]

Hence, the present study was undertaken to find the diagnostic efficacy of microscopy, RDT, and PCR using Bayesian latent class analysis. Latent class analysis constructs its gold standard based on the results of all the tests used in analysis and labels a subject as diseased or not diseased and calculates sensitivity and specificity for all the tests used.[7]

 Materials and Methods

Study setting

We conducted a cross-sectional study on patients visiting the outpatient department of Kasturba Hospital, Sewagram, Wardha situated in central India. The hospital mainly caters to a rural population of Eastern Maharashtra and the Northern part of Telangana states. Wardha district has Annual Parasite Incidence (API) of <1, but its neighbouring district reports API >1.[8]

Study population

We included 200 blood samples from malaria suspects in the study. Subjects with clinical symptoms of malaria were selected irrespective of age and sex. Consecutive subjects in whom clinicians suspected malaria and raised investigations during the study were included in the study. This mainly involved subjects having a fever with chills and rigor in the absence of any obvious cause such as upper respiratory tract infection. All patients who had been diagnosed and/or treated with antimalarial drugs within the past 6 months were excluded from the study.

Before the start of study permission was obtained from the Institutional Ethics Committee of Mahatma Gandhi Institute of Medical Sciences, Sevagram. Informed consent was obtained from the participants or their legal guardians.

Peripheral smear examination for malarial parasites

Thick and thin blood films were prepared from collected blood on the new, clean, grease-free glass slide. They were stained using Leishman stain and examined as per the national guidelines by two microscopists having >15 years of experience. Both of them were blind to each other's report. Discordance was resolved by a third reader (microbiologists).

Rapid diagnostic test

The RDT used in the study was One Step Malaria Pf/Pv test (SD BIOLINE Malaria Antigen P.f/P.v test cassette). The test was performed and interpreted as per manufacturer's instructions.

Polymerase chain reaction

For detection and identification of Plasmodium species, we performed nested-PCR targeting the Plasmodium DNA within the highly conserved regions of the small-subunit (SSU) rRNA gene as described by Snounou et al.[9] In brief, for Plasmodium genus detection nest1 PCR reaction was done using rPLU1 and rPLU5 primers followed by nest2 reaction using the primers rPLU 3 and rPLU4. All genus-specific positive results were analysed to species level by species-specific nested PCR using primers rFAL1 and rFAL2 (Plasmodium falciparum) [Table 1]. All oligonucleotide primers were obtained from Sigma (Mumbai, India).{Table 1}

All PCR reactions were performed in 25 μl of reaction volume. For the nest1 genus-specific PCR, 2.5 μl of whole blood was added to 25 μL of reaction mixture including 12.5 μl of Phusion blood PCR mix (Thermofisher scientific, USA) containing 1U Phusion Blood II DNA polymerase, 50 mM MgCl2, 10 mM dNTPs and 10 μM of each primer (rPLU1 and rPLU5). DNA amplification was carried out under following conditions: 98°C for 5 min, 35 cycles of 98°C for 1 min; 65°C for 1 min, 72°C for 1 min followed by final extension at 72°C for 5 min.

The resulting nest 1 PCR product was centrifuged at 1000 × g for 3 min. For the nest 2 genus-specific PCR, 2.5 microliter of nest 1 PCR product was added to 12.5 μl of Toptaq master mix (Qiagen, USA) containing 10 × coral load solution, 5 × Q-solution, 5U TopTaq DNA Polymerase, 15 mM MgCl2, 400 μM each dNTP and 10 μM of each primer (rPLU3 and rPLU4). DNA amplification was carried under following conditions: 94°C for 3 min, 35 cycles of 94°C for 30 s; 58°C for 30 s, 72°C for 30 s followed by final extension at 72°C for 2 min.

Detection of species P. falciparum was performed using species-specific primers (rFAL1 and rFAL2) where conditions and concentrations were identical to those used for genus-specific second amplification except different primers were used.

Each PCR run included known positive-control (previously diagnosed sample) and a negative control (nuclease-free water). Nest 2 PCR products were analysed by gel electrophoresis on 1.5% agarose gel stained with ethidium bromide.

Microscopy, RDT and PCR were done by different technicians, and results of all the three tests were kept blind.

There were certain discordant results obtained by these three tests in our laboratory. The discordant blood samples were submitted to the National Institute for Malaria Research, New Delhi (NIMR) for confirmation. The final results after confirmation from NIMR were included for further analysis.

Statistical analysis

Age- and sex-distribution of study subjects was expressed in percentages. The magnitude of disease, sensitivity and specificity of the test along with their 95% confidence intervals (CIs) were calculated using Bayesian latent class analysis. This technique also gives sensitivity and specificity of the test which otherwise would have been used as a reference standard of which evaluation is difficult (PCR for the present study). Likelihood ratios (LR) of positive and negative test results were calculated manually using point estimates of sensitivity and specificity.


Out of the 200 fever cases studied, the majority (68.5%) were in the age group of 21–60 years followed by 20.5% as adolescents. Mean and median age of the subjects was 34.6 years and 32 years, respectively. Males were 56%, whereas 44% were females [Table 2].{Table 2}

Using Bayesian latent class analysis magnitude of malaria was found to be 17.7% (95% CI: 12.5%–23.9%) among the study subjects. However, it was 11.5% (95% CI: 7.6%–16.5%) for microscopy, 17.0% (95% CI: 12.3%–22.7%) for RDT and 20.5% (95% CI: 15.3%–26.5%) for PCR [Table 3].{Table 3}

For microscopy, the sensitivity and specificity was found to be 63.0% (95% CI: 45.1%–78.5%) and 99.4% (95% CI: 97.1%–100.0%), respectively. LR of positive microscopy result and that of negative microscopy result were 105.0 and 0.372, respectively. Sensitivity and specificity of RDT was found to be 94.2% (95% CI: 77.7%–99.9%) and 99.4% (95% CI: 96.7%–100.0%). The LR of positive RDT result and that of negative RDT result were 175.0 and 0.058, respectively. Sensitivity and specificity of PCR was found to be 89.9% (95% CI: 76.6%–97.9%) and 95.6% (95% CI: 91.4%–98.7%). The LR of positive PCR result and that of negative PCR result were 20.4 and 0.106, respectively [Table 4].{Table 4}


In the present study, the sensitivity of microscopy was 63%, but it had very high specificity (99.4%). Sensitivity and specificity of RDT and PCR were high with RDT having a marginally higher sensitivity (94% vs. 90%) and specificity (99% vs. 95%). On comparison of LRs, RDT had the highest LR for positive test result (175) and the lowest LR for negative test result (0.058) among the three tests.

Early laboratory diagnosis of malaria facilitates the parasite based management of the disease. The NVBDCP uses both microscopy and RDTs for diagnosis of malaria depending on the settings. In expert hands, the sensitivity of microscopy can be excellent, with detection of malarial parasites at densities as low as 4–20 parasites/μL of blood (approximately 0.0001%–0.0005% parasitaemia).[2],[10],[11] Microscopy allows identification of the Plasmodium species as well as quantification of parasitaemia. In the present study, we did not do quantification; however, we observed that the majority of smears had low parasitaemia. Diagnostic errors for malaria occur more commonly in the settings of low-density parasitaemia (10–100 parasites/μL of blood). Although, these errors can also occur with higher densities.[12],[13],[14] For example, in Ethiopia, use of thin films with little quality control resulted in lower sensitivity.[6],[15] The results could affect even further when slides are reused.[16] The misdiagnosis of malaria initiates the vicious cycle of ill-health and poverty.[17] Similar situations could be prevailing elsewhere as well.

RDT with 95% sensitivity or more is considered to be a useful diagnostic tool.[18] Performance of RDT declines when parasite load goes <500/μL blood for P. falciparum and <5000/μL blood for Plasmodium vivax.[19],[20],[21] Despite the lower levels of observed parasitaemia in the present study, performance of RDT was at the desired level.

In India, falciparum malaria and the drug resistance are increasing steadily. Hence, conventional antimalarials are being replaced by artemisinin-based regimen which is costly.[22] Presumptive treatment is less cost-effective than the diagnosis based treatment in such situations.[23] This necessitates the point-of-care diagnosis of malaria. National Drug Policy on Malaria mandates diagnosis to be made within 24 h, especially in falciparum endemic areas.[22] An RDT with the desired performance can reduce the economic burden as well as the burden of morbidity and mortality among the poor malaria-affected population and also has potential to reduce the burden on their health services.[24] The high cost of RDT is one of the limiting factors for its routine use and to shift from symptom-based malaria management to parasite-based management. Microscopy has potential to be used as a point-of-care test for malaria despite its lower sensitivity. Under the programme, the technician visits the primary health centre and processes the slides on a weekly basis. Alternatively, the slides are transported to the designated facility on a predecided periodicity. This is either because of lack of workforce or to minimize the cost. However, these strategies lead to considerable delays in diagnosis. This is, in addition, to delay in accessing health care by patients. On an average, patients access health care after 3–4 days of the onset of fever.[25],[26] This eventually leads to transmission of disease further. In addition, delay in diagnosis is a significant determinant of death due to malaria.[27] The delay in diagnosis incurs an indirect cost to patients. Other costs for microscopy are the cost of organisation, supervision, quality control and skilled personnel, which may balance out the higher cost of RDT. The use of RDT may be most effective in areas where the quality of microscopy is poor and where there is multidrug resistance leading to use of expensive drugs. Use of microscopy will lead to missing on true cases in the community, which will help maintain the chain of transmission, which ultimately incurs additional costs for elimination efforts. Use of RDT as a point-of-care diagnostic tool will address this issue and reduce the net cost of the health system.[17]

Nucleic acid tests (e.g. PCR) generally serve as a gold standard in efficacy studies.[4],[5],[6] The theoretical limit of detection for PCR has been estimated at 0.02–1 parasite/μL.[28] Nested PCR is the most sensitive nucleic acid amplification technology; its sensitivity is 400 parasites/mL (0.4 parasites/μL).[29],[30] Commonly used PCR assays target genus-specific and species-specific sequences of 18S SSU ribosomal RNA, circumsporozoite surface protein, and the cytochrome b gene.[29] In this study, the sensitivity and specificity of PCR were very high.

Accurate detection of low-density malaria infection is of increasing importance as some malaria-endemic areas move towards elimination[30] with surveillance and screening playing larger roles in programme management. However, the infrastructure and training required for the use of PCR limit its utility for these applications. These techniques are costly, need specialised workforce along with specialised infrastructure and are far from using them at point-of-care.


The results of the present study support the use of RDT in our area. The method is simple, rapid, and can be performed at any place as a point-of-care test. The study also provides evidence that PCR may also be used in our district if facilities are available especially when parasitaemia goes down towards the end of elimination phase. However, its routine use in the clinical laboratory in the present form does not appear to be feasible due to its high cost, chances for easy contamination, long processing time and requirement to run several PCR cycles for each sample.

The weakness of the study is that it was a hospital-based study and might have spectrum bias.


We are thankful to Mr Baba Jinde, Lab Technician ICMR Zoonosis Project for his technical assistance.

Financial support and sponsorship

Funds were received by Dr Sreemanti D Saha from Kasturba Health Society for her MD dissertation.

Conflicts of interest

There are no conflicts of interest.


1Dowling MA, Shute GT. A comparative study of thick and thin blood films in the diagnosis of scanty malaria parasitaemia. Bull World Health Organ 1966;34:249-67.
2Bruce-Chwatt L. DNA probes for malaria diagnosis. Lancet 1984;323:795.
3Payne D. Use and limitations of light microscopy for diagnosing malaria at the primary health care level. Bull World Health Organ 1988;66:621-6.
4Milne LM, Kyi MS, Chiodini PL, Warhurst DC. Accuracy of routine laboratory diagnosis of malaria in the United Kingdom. J Clin Pathol 1994;47:740-2.
5Kilian AH, Metzger WG, Mutschelknauss EJ, Kabagambe G, Langi P, Korte R, et al. Reliability of malaria microscopy in epidemiological studies: Results of quality control. Trop Med Int Health 2000;5:3-8.
6Okell LC, Ghani AC, Lyons E, Drakeley CJ. Submicroscopic infection in Plasmodium falciparum-endemic populations: A systematic review and meta-analysis. J Infect Dis 2009;200:1509-17.
7Deshmukh P, Narang R, Dongre A, Behere P. Evaluation of diagnostic tests in absence of gold standard - A special reference to GMHAT (PC Tool). In: Behere PB, Shrama VK, Kumar V, Shah VA, editors. Mental Health Training for Health Professionals: Global Mental Health Assessment Tool (GMHAT). New Delhi: Indian Psychiatric Society; 2017. p. 75-92.
8Government of India. National Framework for Malaria Elimination in India. New Delhi: Government of India; 2016-2013. Available from: [Last accessed on 2017 Feb 06].
9Snounou G, Viriyakosol S, Jarra W, Thaithong S, Brown KN. Identification of the four human malaria parasite species in field samples by the polymerase chain reaction and detection of a high prevalence of mixed infections. Mol Biochem Parasitol 1993;58:283-92.
10Biadglegne F, Belyhun Y, Ali J, Walle F, Gudeta N, Kassu A, et al. Does the practice of blood film microscopy for detection and quantification of malaria parasites in Northwest Ethiopia fit the standard? BMC Health Serv Res 2014;14:529.
11Abreha T, Alemayehu B, Tadesse Y, Gebresillassie S, Tadesse A, Demeke L, et al. Malaria diagnostic capacity in health facilities in Ethiopia. Malar J 2014;13:292.
12Forney JR, Magill AJ, Wongsrichanalai C, Sirichaisinthop J, Bautista CT, Heppner DG, et al. Malaria rapid diagnostic devices: Performance characteristics of the paraSight F device determined in a multisite field study. J Clin Microbiol 2001;39:2884-90.
13Forney JR, Wongsrichanalai C, Magill AJ, Craig LG, Sirichaisinthop J, Bautista CT, et al. Devices for rapid diagnosis of malaria: Evaluation of prototype assays that detect Plasmodium falciparum histidine-rich protein 2 and a Plasmodium vivax-specific antigen. J Clin Microbiol 2003;41:2358-66.
14Fernando SD, Karunaweera ND, Fernando WP. Evaluation of a rapid whole blood immunochromatographic assay for the diagnosis of Plasmodium falciparum and Plasmodium vivax malaria. Ceylon Med J 2004;49:7-11.
15Mixson-Hayden T, Lucchi NW, Udhayakumar V. Evaluation of three PCR-based diagnostic assays for detecting mixed Plasmodium infection. BMC Res Notes 2010;3:88.
16Roper C, Elhassan IM, Hviid L, Giha H, Richardson W, Babiker H, et al. Detection of Very Low Level Plasmodium falciparum Infections Using the Nested Polymerase Chain Reaction and a Reassessment of the Epidemiology of Unstable Malaria in Sudan; 2015. Available from: [Last accessed on 2017 Feb 15].
17Amexo M, Tolhurst R, Barnish G, Bates I. Malaria misdiagnosis: Effects on the poor and vulnerable. Lancet 2004;364:1896-8.
18Toma H, Kobayashi J, Vannachone B, Arakawa T, Sato Y, Nambanya S, et al. Afield study on malaria prevalence in Southeastern Laos by polymerase chain reaction assay. Am J Trop Med Hyg 2001;64:257-61.
19Zakeri S, Najafabadi ST, Zare A, Djadid ND. Detection of malaria parasites by nested PCR in South-Eastern, Iran: Evidence of highly mixed infections in Chahbahar district. Malar J 2002;1:2.
20Alves FP, Durlacher RR, Menezes MJ, Krieger H, Silva LH, Camargo EP, et al. High prevalence of asymptomatic Plasmodium vivax and Plasmodium falciparum infections in native Amazonian populations. Am J Trop Med Hyg 2002;66:641-8.
21Steenkeste N, Incardona S, Chy S, Duval L, Ekala MT, Lim P, et al. Towards high-throughput molecular detection of Plasmodium: New approaches and molecular markers. Malar J 2009;8:86.
22National Vector Borne Diseases Control Program. National Drug Policy on Malaria. New Delhi: National Vector Borne Diseases Control Program; 2013. Available from: [Last accessed on 2017 Feb 16].
23Uzochukwu BS, Obikeze EN, Onwujekwe OE, Onoka CA, Griffiths UK. Cost-effectiveness analysis of rapid diagnostic test, microscopy and syndromic approach in the diagnosis of malaria in Nigeria: Implications for scaling-up deployment of ACT. Malar J 2009;8:265.
24Bell D, Wongsrichanalai C, Barnwell JW. Ensuring quality and access for malaria diagnosis: How can it be achieved? Nat Rev Microbiol 2006;4:682-95.
25Yadav SP. A study of treatment seeking behaviour for malaria and its management in febrile children in rural part of desert, Rajasthan, India. J Vector Borne Dis 2010;47:235-42.
26Chaturvedi HK, Mahanta J, Bajpai RC, Pandey A. Risk of malaria among febrile patients: Retrospective analysis of a hospital-based study in an endemic area of Northeast India. Int Health 2014;6:144-51.
27Estimation of True Malaria Burden in India. Available from: [Last accessed on 2017 Feb 14].
28Schneider P, Bousema JT, Gouagna LC, Otieno S, van de Vegte-Bolmer M, Omar SA, et al. Submicroscopic Plasmodium falciparum gametocyte densities frequently result in mosquito infection. Am J Trop Med Hyg 2007;76:470-4.
29Woyessa A, Deressa W, Ali A, Lindtjørn B. Evaluation of CareStart™ malaria Pf/Pv combo test for Plasmodium falciparum and Plasmodium vivax malaria diagnosis in Butajira area, South-Central Ethiopia. Malar J 2013;12:218.
30Ouédraogo AL, Bousema T, Schneider P, de Vlas SJ, Ilboudo-Sanogo E, Cuzin-Ouattara N, et al. Substantial contribution of submicroscopical Plasmodium falciparum gametocyte carriage to the infectious reservoir in an area of seasonal transmission. PLoS One 2009;4:e8410.