Are you rK with DAT?

 

A literature review of two diagnostic tests for visceral leishmaniasis

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Stephanie Oberfoell and Claudia Skieller

Parasites and Pestilence

May 23, 2008

 


 

Across 65 countries and four continents, over 200 million people are at risk of contracting Visceral Leishmaniasis (VL), the systemic form of leishmaniasis (OneWorld Health, 2008).  There are a reported 500 000 cases of VL per year (WHO, 2002). Because the disease occurs mostly in rural communities in the developing world, it is likely that the disease prevalence is severely underreported.

To decrease the prevalence of the disease, VL diagnostics must identify infected individuals, assess the effectiveness of treatment, and accurately survey the disease prevalence and distribution. Tests must be easy to use in the field, inexpensive, sensitive and specific. Sensitivity defines how well a test can identify infected people.  Specificity defines how well a test can identify healthy people (Loong, 2003).  For example, a test with a 100% sensitivity but low specificity will all of the infected, but falsely identify many healthy individuals as well.

Developing diagnostics for VL has proven difficult.  Most importantly, the poor and isolated populations most affected cannot afford treatment and diagnosis, limiting funding for technology development.  The areas in which VL is endemic lack health infrastructure, requiring diagnostics that are easy to use and can withstand heat and shaking.  Furthermore, infected people can be asymptomatic for an average of four months in South Asia (Burn and Chowdhury, 2006).  During this time period, the individual can infect many new sandflies, putting family members and neighbors at risk of contracting the disease. Even when patients exhibit clinical symptoms, the symptoms mimic many other illnesses like malaria, tropical splenomegaly, schistosomiasis, African trypanosomiasis, tuberculosis, brucellosis, typhoid fever, bacterial encarditis, histoplasmosis, malnutrition, lymphoma, and leukaemia (Singh, 2006).  Furthermore, VL patients are often co-infected with malaria, HIV, typhoid, and/or tuberculosis, making diagnosis even more difficult.  These factors have lead to improper diagnoses. 

 

Background

            An intramacrophage protozoa parasite causes the vector-born disease, leishmaniasis (Singh, 2006).  There are five known strains of visceral leishmania parasites, L. donovani, L. tropica, L. amazonensis, L. chagasi, and L. infantum (John and Petri, 2006).  The most prevalent form is L. donovani, which is endemic to Asia mainly in India, Myanmar, Thailand; and to Africa in Central African Republic, Sudan, Kenya, Niger, northern Uganda, Dijibouti, Ethiopia.  L. infantum  centers around the Mediterranean.  L. chagasi focuses around Central and South America. Figure 1 shows the overall distribution of the disease.

The disease vector, the Phlebotomus sandfly, breeds in rat burrows, caves, and forest areas (WHO, 2008). While dogs and rats are reservoirs, transmission is primarily from human to human, especially in highly endemic areas (Bern and Chowdhury, 2006).

The lifecycle of VL parasite (Figure 2) begins when a Phlebotomine sandfly ingests amastigotes from the hostŐs blood. In the midgut of the sandfly, the amastigotes develop into stationary metacyclic promastigotes, multiply, then move into the salivary gland (CDC, 2008; Singh, 2006).  During the next blood meal, some metacyclic promastigotes are injected into the host.  Macrophages digest the metacyclic promastigotes that flourish in the low pH environment in the macrophage. Amastigotes develop and multiply by binary fission.  Eventually, these amastigotes lyse the macrophage.  The released amastigotes can either be taken up by another sandfly or re-invade the new host cells (Sing, 2006).

Amastigotes reside in reticuloendothelial cells, concentrating in the liver and spleen.  After 2 weeks to 18 months, acute symptoms may begin including periodic fevers and malaise.  Over the course of 2 years, anemia, hepatosplenomegaly, lymphadenopathy, emaciation, and skin darkening may occur.  If untreated, the disease is fatal in 75-95% of cases within 2 years (John and Petri, 2006; Singh, 2006).

 

Diagnostics Methods

Invasive Methods

Tissue biopsies can definitively demonstrate active VL.  Healthcare workers search for amastigotes, round or oval protozoa with organelles, nucleus, and kinetoplasts, from the spleen, bone marrow, lymph nodes, or peripheral blood (See Figure 3)(Singh, 2006).  Obtaining a sample from the spleen has the highest sensitivity (95%), as compared bone marrow (60-80%) and lymph node aspiration (40-60%) (Boelart, 2007; Singh, 2006). Spleen samples may also help to determine the severity of the disease by estimating the parasite load. However, this method is invasive and results in a fatal hemorrhage in 2/10 000 patients (Singh, 2006). Bone marrow aspirates are safer but more painful and less specific.  Lymph node aspiration and peripheral blood tests rarely identify amastigotes for VL.  Aspiration procedures also require extensive medical infrastructure that is not available in the poor and rural regions where VL is most prevalent. The lack of widely accessible diagnostic methods leads to delayed diagnosis, untreated cases, and thus the further spread of the disease. 

Noninvasive methods

Although current noninvasive antibody tests cannot distinguish between active and previous VL, they are important rapid diagnostic tools (Sundar, 2007).  One of the first noninvasive antibody tests was the Direct Agglutination Test (DAT) in the 1980Ős.  DAT works with in vitro cultivated promastigotes to create an antigen solution that must be stored at 4ˇC and not frozen (WHO, 1996; Sinha, 2007).  To perform the test, one must perform a series of dilutions of blood serum and mix fixed and stained L. Donovani promastigotes. The solution incubates in V-shaped wells at room temperature overnight. The results are read visually by judging the extent of agglutination. The test was used across hospitals in Africa and India.  However, the test was less useful in the field due to excess heat and shaking of the liquid antigen, and inconsistent readings of the test (Boelaert, 1999). 

To improve field applicability, freeze-dried antigen instead of the aqueous antigen was developed for the DAT test.  Freeze-dried antigen is reconstituted with a saline solution prior to use.  This antigen can withstand more heat and motion than its liquid predecessor, and can be read within three hours (Singh, 2006).

Another form of antibody diagnostic test is the enzyme-linked immunosorbent assay (ELISA) for rK39, an antigen from Leischmania chagasi. Researchers developed other ELISA tests that use antigens from other VL strains including rGBP from L. donovani, rORFF from L. infantum, and rgp63, rk90, and rk26 from L. chagasi (Singh, 2006).  The ELISA test requires expensive laboratory materials like an ELISA plate reader, reagents that must be kept at certain temperatures, and a centrifuge (WHO, 1996).  These materials are difficult to use in the field. 

To improve field utility of diagnostic methods, the rK39 dipstick test was developed.   The test strip is nitrocellulose coated on one end with anti-protein, an antibody for control, and visceral leishmania antigen K39 for a positive result (Iqbal, 2002; Sinha, 2008).  The patientŐs serum (5-10uL) and test buffer solution(three drops) is placed on the other end of the strip.  After the serum diffuses for ten minutes, a color change on the control shows the strip is valid and a change near the K39 antigen means the test is positive (Sundar, 2005). 

There are also other less commonly used antibody diagnostic tools, such as, fluorescent antibody test, Leishmania skin test, immunoblotting, etc (Singh, 2006).  Molecular diagnostic methods have also been developed for VL, but these are expensive and not accessible to most of the resource-poor settings where VL is most prevalent. 

 

Existing Research

Indian Subcontinent

            Many studies have been done on the effectiveness of the FD-DAT and rK39 tests, as well as of other diagnostic tests such as AQ-DAT, IHA (indirect hemagglutination), and KAtex (urine latex agglutination). However, since FD-DAT and rK39 seem to be the most feasible and cost-effective, these two tests in particular have been the focus of studies in Indian states as well as Nepal and Bangladesh.

Since many of the studies show similar results, a few indicative studies were chosen to examine the sensitivity and specificity of FD-DAT and rK39 in the Indian subcontinent. Overall, FD-DAT appears to be more sensitive than rK39 in this region, with results varying between 92.6 and 100% sensitive for FD-DAT and 87 and 100% for rK39 (Sundar, 2006; Sinha, 2008; Ritmeijer, 2006; Boelaert, 2008) (Table 1). Specificity varies widely between studies, but is always lower than sensitivity results in the Indian subcontinent.  FD-DATŐs and rK39Ős specificity and sensitivity can vary widely depending on location, implying more diagnostic research must take place. 

            The Indian subcontinent represents a region with high visceral leishmaniasis endemicity. It has been proposed that diagnostic tests such as FD-DAT and rK39 are more sensitive in endemic than in nonendemic areas (Iqbal, 2002). This may be due to lower levels of antibodies in patients from nonendemic areas, who have not been highly exposed to L. donovani. However, in the endemic area of the Indian subcontinent, FD-DAT and rK39 have been found to score highly in sensitivity and specificity analyses, making them comparable in diagnostic effectiveness. The decision of which diagnostic test to utilize in this area then depends on factors such as ease of use, mechanical issues, and cost.

East Africa

            While fewer comprehensive studies have been done examining FD-DAT and rK39 in East Africa, two recent studies have had similar results regarding sensitivity and specificity (Ritmeijer, 2006; Boelaert, 2008). By analyzing results from Sudan, Kenya, and Ethiopia, the average East African sensitivity for FD-DAT was found to be 92.8% and for rK39 was found to be 79.1%, but with some results as high as 90.0% sensitive. As with Indian results, specificity varied widely for FD-DAT and rK39. FD-DAT had an average of 91.2% specificity while rK39 average 84.8% (Table 1). However, in Sudanese studies, which constituted a majority of the East African loci, specificity was usually much greater than sensitivity (Table 2 and 3, Appendix).

            These differences in sensitivity and specificity by region can be due to ethnic background, environment, severity of infection, or differences in L. donovani genotype (Iqbal, 2002). Further, co-infection with different infectious diseases can affect visceral leishmaniasis diagnosis by FD-DAT or rK39 (see Cross Reactivity Problems). Since the environment and disease profile of the Indian subcontinent and East Africa are very different- yet quickly becoming parallel in such statistics as HIV and tuberculosis rates- it is not surprising that sensitivity and specificity analyses should yield very different results. However, these East African studies show that further research must be conducted specifically about the application of these diagnostic tests in East African endemic settings. Although specificity can currently reach 99% in Sudan with rK39, a sensitivity of 90% means that 10% of positive cases are risking being fatally undiagnosed and untreated.

Table 1: Comparison along several parameters of FD-DAT and rK39 for diagnosing visceral leishmaniasis in the Indian subcontinent and East Africa

 

Parameters

Test

FD-DAT

rK39

Sensitivity (%)a

 

 

Indian subcontinent

98.2 (92.6-100)

97.7 (87-100)

East Africa

92.8 (77-99.9)

79.1 (67-90)

Specificity (%)a

 

 

Indian subcontinent

94.6 (83-100)

89.4 (73-100)

East Africa

91.2 (73.2-99.9)

84.8 (46.3-99)

Additional requirementsb

Available antigen, several hours incubation

Test buffer solution

Cost per testc

$2.50

$1-$1.30

Major cross-reactionsd

malaria, TB

malaria, TB, HIV

 

a Sundar S et al, (June 2006); Sinha PK et al. (Mar 2008); Ritmeijer K et al. (2006); Boelaert M et al. (Jan 2008).

b Sundar S et al, (June 2006).

c Chappuis F et al. (Dec 2005); Boelaert M et al. (1999).

d Sundar S et al, (June 2006); Sinha PK et al. (Mar 2008).

 

Cross Reactivity Problems

            Both FD-DAT and rK39 present diagnostic problems in patients infected with diseases such as tuberculosis, HIV, and malaria, as examined in two recent studies in India (Sundar, 2006; Sinha, 2008). The results were slightly contradictory with regard to which of the two tests had greater specificity, but both indicated small percentages of false positives in confirmed cases of malaria, HIV, tuberculosis, or HIV/TB co-infection. These statistics ranged from 5% rK39 false positives for patients with only malaria or TB (Sundar, 2006) to 27% rK39 false positives for HIV patients (Sinha, 2008). Interestingly, one patient with visceral leishmaniasis/HIV co-infection in the 2008 Sinha et al. study tested falsely negative in the FD-DAT test but positive in the rK39 test. As the HIV and visceral leishmaniasis endemics spread and overlap in areas such as India and East Africa, problems with cross reactivity will have to be addressed in order to improve the relative effectiveness of FD-DAT and rK39. At the least, health care providers should be wary of results- both negative and positive- in patients with confirmed cases of the above diseases, and should utilize secondary diagnostic techniques such as a second assay or tissue biopsy.

 

Disadvantages of FD-DAT and rK39

            While these two diagnostic tests have been shown to be the most feasible and applicable in rural endemic areas such as parts of the Indian subcontinent and East Africa, there are a few disadvantages to each that must be addressed. FD-DAT is significantly easier to use in rural settings than its cousin, AQ-DAT, which utilizes an aqueous antigen instead of a freeze-dried antigen and therefore requires cold-chain storage. However, even FD-DAT requires incubation before results can be read. Therefore, it is not practical in peripheral settings without appropriate laboratory equipment (Sundar, 2006).

            The rK39 test is more applicable in rural settings since it only requires test flow strips and test buffer solution in order to run the patientŐs serum along the strip. However, there are even variations among rK39 dipstick tests, since some packaged commercial test kits require a simple fingerprick of blood whereas others require venous blood and therefore necessitate basic healthcare materials (Chappuis, 2005). Clearly, those tests that require the fewest excess materials, such as DiaMed-IT LEISH by DiaMed AG in Switzerland, are ideal for rural diagnosis of visceral leishmaniasis. Further, rK39 has been shown to show positive results in patients even three years after successful treatment (Goswami, 2003). This means that clinical presentation in patients must be considered when correctly diagnosing visceral leishmaniasis with rK39.

 

Costs

            Both FD-DAT and rK39 can be offered at very low costs, making feasible their usage in developing countries with endemic visceral leishmaniasis. In one 1999 study, FD-DAT was calculated to cost $2.50/test, including peripheral costs for materials and labor not included in packaged tests (Boelaert, 1999). In a 2005 study in Uganda, rK39 tests were found to cost between $1 and $1.30, depending on the manufacturer (Chappuis, 2005). Therefore, rK39 is clearly more cost-effective in terms of number of patients diagnosed. These costs issues must be considered with costs of feasibility and applicability in rural, both endemic and nonendemic, settings.

Conclusions

            As summarized in Table 1, FD-DAT scores higher in sensitivity and specificity in both the Indian subcontinent and East Africa. However, given the lower costs of rK39 and the fewer additional requirements, rK39 is currently more feasible and easy to use in the field. Further, its sensitivity and specificity results are acceptable given these additional advantages.

            However, there are caveats to utilizing rK39, especially in nonendemic areas. Given the possibility for false negatives or positives, one should be wary of accepting a rK39 result if a clinical presentation indicates otherwise. In a 2002 study of the rK39 strip test in nonendemic Kuwait, sensitivity was found to be 80.0%, even though specificity was high (99.0%) (Iqbal, 2002). Therefore, in areas such as this or if clinical presentation indicates any symptoms of visceral leishmaniasis, secondary diagnostics such as FD-DAT or a liver biopsy should be used if possible.

 

 

 

 

 

 

 

 

 

 

 


Appendix

Table 2: Overview of rK39 antigen-based test results for visceral leishmaniasis (VL)

Location/Name

Year

Country

Total sample

VL cases

Sensitivity (%)

Specificity (%)

South Asia

 

 

 

 

 

 

Sundar S

1998

India

344

127

100

98

Bern C

2000

Nepal

127

14

100

100

Qu JQ

2000

China

13

13

100

 

Singh S

2002

India

308

228

100

100

Sarker CB

2003

Bangladesh

180

180

97

98

Chappuis F

2003

Nepal

184

139

97

71

Boelaert M

2004

Nepal

310

 

87

93

Latin America

 

 

 

 

 

 

Delgado O

2001

Venezuela

117

41

88

100

Braz RF

2002

Brazil

208

120

93

99

Carvalho SG

2003

Brazil

188

128

90

100

Mediterranean

 

 

 

 

 

 

Ozensoy S

1998

Turkey

71

24

96

94

Maalef IA

2003

Tunisia

146

38

100

97

Sudan

 

 

 

 

 

 

Zijlstra EE

1998

Sudan

15

15

93

 

Zijlstra EE

2001

Sudan

116

55

67

97

Veeken H

2003

Sudan

77

50

92

59

MSF unpublished

1999

Sudan

91

60

67

87

Present study

2004

Sudan

338

201

90

99

 

Adapted from Ritmeijer, 2006.

Table 3: Prevalence in the test sample and sensitivity and specificity of freeze-dried direct agglutination test (FD-DAT), rK39 dipstickÉ used for the diagnosis of visceral leishmaniasis (VL) in [three] countries in East Africa and the Indian subcontinent

Parameters

East Africa

Indian subcontinent

 

Ethiopia estimate

Kenya estimate

Sudan estimate

India estimate

Nepal estimate

Prevalence

57.2 (40.5-73.4)

60.9 (54.7-66.7)

37.0 (31.0-43.2)

79.6 (75.1-83.7)

71.0 (63.5-77.9)

FD-DAT

 

 

 

 

 

Sensitivity

94.0 (80.0-99.8)

98.8 (96.6-99.9)

85.7 (77.0-92.7)

98.3 (96.2-99.6)

98.5 (94.8-100)

Specificity

93.6 (77.4-99.8)

81.9 (73.2-89.8)

98.2 (94.8-99.9)

91.0 (83.0-96.6)

95.4 (87.1-99.6)

rK39 dipstick

 

 

 

 

 

Sensitivity

75.4 (55.9-90.5)

84.7 (78.6-89.8)

77.9 (69.2-85.6)

99.6 (98.4-100)

96.5 (92.1-99.2)

Specificity

70.0 (46.3-88.9)

89.9 (83.2-95.1)

91.8 (86.7-96.2)

90.0 (81.2-96.4)

90.9 (80.8-97.5)

 

Adapted from Boelaert, 2008.


Works Cited

 

1. 21 May 2008 <http://www.who.int/leishmaniasis/leishmaniasis_maps/en/index.html>.

 

2. 21 May 2008 <http://www.who.int/topics/leishmaniasis/en/>. 

 

3. Bern, C, and R Chowdhury. (2006). The Epidemiology of Visceral Leishmaniasis in Bangladesh: Prospects for Improved Control. Indian Journal of Medicine, 123, 275-788.

 

4. Boelaert, M, S El-Safi, A Hailu, M Mukhtar, S Rijal, S Sundar, M Wasunna, A Aseffa, J Mbui, J Menten, P Desjeux, RW Peeling. (Jan 2008). Diagnostic tests for kala-azar: a multi centre study of the freeze-dried DAT, rK39 strip test and KAtex in East Africa and the Indian subcontinent. Trans R Soc Trop Med Hyg., 102(1), 32-40.

5. Boelaert, M, S El Safi, H Moussa,  and J Githure. (Jan 1999). Multi-Centre Evaluation of Repeatability and Reproducibility of the Direct Agglutination Test for Visceral Leishmaniasis. Tropical Medicine & International Health, 4(1), 34-37.

 

6. Boelaert, M, L Lynen, P Desjeux, P Van der Stuyft. (1999). Cost-effectiveness of competing        diagnostic- therapeutic strategies for visceral leishmaniasis. Bulletin of the World Health      Organization, 77 (8).

7. Chappuis, F, Y Mueller, A Nguimfack, JB Rwakimari, S Couffignal, M Boelaert, P Cavailler, L Loutan, P Piola. (Dec 2005). Diagnostic Accuracy of Two rK39 Antigen-Based Dipsticks and the Formol Gel Test for Rapid Diagnosis of Visceral Leishmaniasis in Northeastern Uganda. Journal of Clinical Microbiology, 43(12), 5973-5977.

8. Chappuis, F, S Sundar, A Hailu, H Ghalib, S Rijal, RW Peeling, J Alvar, M Boelaert. (Nov 2007). Visceral leishmaniasis: what are the needs for diagnosis, treatment and control?. Nature Reviews Microbiology, 5, S7-S16

9. Goswami, RP, B Bairagi, PK Kundu. (Aug 2003). K39 strip test--easy, reliable and cost-effective field diagnosis for visceral leishmaniasis in India. J Assoc Physicians India, 51, 759-61.

10. Harith, AH, PA Kolk, J Kager, R Leeuwenburg,  and S Muigai. (1986). A Simple and Economical Direct Agglutination Test for Serodiagnosis and Sero-Epidemiological Studies of Visceral Leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygeine, 80(4), 583-6. 

 

11. Iqbal, J, P Hira, G Saroj,  and R Philip. (Feb 2002). Imported Visceral Leishmaniasis: Diagnostic Dilemmas and Comparative Analysis of Three Assays. Journal of Clinical Microbiology, 40(2), 475-9.

 

12. Loong, Tze-Wey. (Sep 2003). Understanding Sensitivity and Specificity with the Right Side of the Brain. British Medical Journal, 327, 716-719. 

 

 

13. OneWorld Health. 21 Apr. 2008. <http://www.oneworldhealth.org/diseases/leishmaniasis.php>. 

 

14. Pedras, MJ, and EJ Oliveira. (Feb 2008). Comparative Evaluation of Direct Agglutination Test, RK39 and Soluble Antigen ELISA and IFAT for the Diagnosis of Visceral Leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 102(2), 172-178.

15. Ritmeijer, K, Y Melaku, M Mueller, S Kipngetich, C OŐKeeffe, RN Davidson. (2006).Evaluation of a new recombinant K39 rapid diagnostic test for Sudanese Visceral Leishmaniasis, Am. J. Trop. Med. Hyg., 74(1), 76-80

16. Singh, RK, HP Pandey,  and S Sundar. (Mar 2006). Visceral Leishmaniasis (Kala-Azar): Challenges Ahead. Indian Journal of Medicine, 123, 331-344.

 

17. Singh, S. (Mar 2006). New Developments in Diagnosis of Leishmaniasis. Indian Journal of Medicine, 123, 311-330.

 

18. Sinha, PK, S Bimal, K Pandey,  and S K. Singh. (2008). A community-based, comparative evaluation of direct agglutination and rK39 strip tests in the early detection of subclinical Leishmania donovani infection.  Annals of Tropical Medicine and Parasitology, 102(2), 119-25.

 

19. Sundar, S, RK Singh, R Maruya,  and B Kumar. (2006). Serological Diagnosis of Indian Visceral Leishmaniasis: Direct Agglutination Test Versus RK39 Strip Test. Transactions of the Royal Society of Tropical Medicine and Hygiene, 100 (6), 533-537.

 

20. Sundar, S, RK Singh, S K. Bimal,  and K Gidwani. (Feb 2007). Comparative evaluation of parasitology and serological tests in the diagnosis of visceral leishmaniasis in India: a phase III diagnostic accuracy study. Tropical Medicine and International Health, 12(2), 284-289.