The early diagnosis of leishmaniasis is important in order to avoid severe damage or even death of the patient. The routine diagnosis of leishmaniasis relies on either the microscopical demonstration of Leishmania amastigotes in aspirates from lymphoid tissue or liver, in slit skin smears or in peripheral blood or culturing. However, the retrieval of samples is uncomfortable for the patient and the isolation of parasites by culturing is time-consuming, difficult and expensive. Because of these limitations, a number of indirect immunological methods, such as enzyme-linked immunosorbent assay (ELISA), dipsticks and direct agglutination test (DAT), have been developed. Despite the large number of serological tests that are available, there is still no gold standard diagnostic test. This is in part because of the fact that none of the tests is 100% sensitive and specific. Moreover, the spread of Leishmania/HIV coinfection complicates the use of the serological techniques as a result of low or lack of antibody responses of these patients (WHO 2000).
Identification of parasite antigens for serodiagnosis
Western blotting techniques have been extensively used to identify antigens for the serodiagnosis of (visceral) leishmaniasis. The most promising candidate so far is a 39 amino-acid repeat from a kinesin-like protein that is predominant in Leishmania chagasi tissue amastigotes (Burns et al. 1993). The recombinant protein rK39 was initially used in the ELISA, after which it was used for the development of a rapid immunochromatographic dipstick test (Jelinek et al. 1999). Both the ELISA and the dipstick test were tested for the diagnosis of visceral leishmaniasis. Initial reports were very promising: rK39-based tests were very sensitive and specific for visceral leishmaniasis, could be used in HIV-positive patients, and antibody levels against rK39 declined rapidly after successful treatment (Burns et al. 1993; Qu et al. 1994; Houghton et al. 1998). However, more recent reports show that the dipstick test lacks sensitivity (Jelinek et al. 1999; Zijlstra et al. 2001) and specificity (Veeken 2001). Currently, there are no other widely available diagnostic tests that are based on defined, cloned Leishmania antigen(s). The DAT remains the serological test of choice, in particular, in many developing countries (Schallig et al. 2001).
Diagnosis by polymerase chain reaction (PCR)
Over the years, a number of different PCR assays has been developed for the detection of Leishmania DNA in a variety of clinical samples such as skin biopsies and smears, bone marrow and lymph node aspirates and peripheral blood. Several target sequences have been used for the PCR. Maximum sensitivity can be achieved by using multicopy sequences as the PCR target (Lachaud et al. 2002). Examples of such targets are ribosomal RNA genes, kinetoplast DNA, mini-exon-derived RNA genes and genomic repeats (Osman 1998). The specificity of the PCR can be adapted to specific needs by targeting conserved or variable regions. In this way it is possible to characterize the parasite to the level of the genus complex, species or even the individual isolate.
For diagnosis of VL, bone marrow and lymph node aspirates as well as blood samples have been evaluated. Bone marrow aspirates from parasitologically confirmed VL patients were always PCR positive in several studies (Mathis & Deplazes 1995; Andresen et al. 1997; Osman et al. 1997a). In another study, comparison of PCR on bone marrow aspirates with microscopic examination and culture for diagnosis of VL in immune-compromised patients showed that PCR exhibited a higher sensitivity (82%) than microscopy (55%) and culture (55%) (Piarroux et al. 1994). In human lymph node aspirates Leishmania DNA was detected by PCR in all six samples in one study (Andresen et al. 1997) and in 33/38 samples in another (Osman et al. 1997a). In general, PCR is a more sensitive method for the detection of Leishmania in lymph node and especially bone marrow aspirates of VL patients than microscopy and is especially useful for the confirmation of cases of suspected VL.
Because lymph node, bone marrow and splenic aspiration is painful and can even be dangerous for the patient, peripheral blood, which is easy to obtain, may be used for the initial PCR screening of people suspected of having VL. The sensitivity of PCR for the detection of Leishmania DNA in blood samples ranges from around 70% (Adhya et al. 1995; Osman et al. 1997a) to 90% (Nuzum et al. 1995) and higher (Andresen et al. 1997; Salotra et al. 2001). It must be emphasized that if PCR on blood is negative, a PCR on lymph node and/or bone marrow material should be performed, because PCR on these materials is more often positive (Osman et al. 1997a).
PCR may also be useful for the confirmation of the diagnosis in HIV/Leishmania coinfected patients. Pizzuto et al. (2001) showed that all 76 HIV/Leishmania coinfected patients were parasitaemic by PCR on peripheral blood before therapy. In another study, 15 of 20 (75%) patients were PCR positive (Campino et al. 2000).
Patients with cutaneous or muco-cutaneous leishmaniasis (CL or MCL) often have low or no Leishmania antibodies, because of the localized character of the disease, and thus serological tests are mostly negative. Therefore, PCR is an important tool for the diagnosis of CL and MCL. Six of 10 skin biopsies from patients with CL were positive in PCR (Mathis & Deplazes 1995) and Andresen et al. (1996) found that 24 of 28 (86%) samples taken from CL patients were PCR positive, whereas microscopy detected parasites in only 55% of the samples that for reasons of epidemiology, clinical aspect, LST-status and pathology were considered to be CL. Later studies essentially confirmed these early findings with PCR being positive in more than 90% of CL cases (Aviles et al. 1999; Pirmez et al. 1999). In MCL, PCR was capable of detecting parasites in 17 of 24 (71%) patients, whereas diagnosis by conventional techniques could only confirm the disease in four of 24 (17%) patients (Pirmez et al. 1999).
Also in post-kala-azar dermal leishmaniasis (PKDL) PCR has proved its value: when using slit skin smears from PKDL patients, 19 of 23 (83%) samples were PCR positive compared with only seven of 23 (30%) positive samples in microscopy (Osman et al. 1998a). In another study, 45 of 48 PKDL patients were PCR positive (Salotra et al. 2001).
Finally, PCR detection of parasite DNA in either lymph node aspirates or peripheral blood can be used as a prognostic marker for the development of relapse or PKDL after apparently successful treatment. Parasite DNA could still be detected by PCR in 80 and 40%, respectively, of the lymph node aspirates that were obtained from Sudanese VL patients either immediately after treatment (Osman et al. 1998) or at least 3 months after treatment (Osman et al. 1997b). While patients who were PCR negative remained free of signs and symptoms and were apparently cured, 36% of the patients with a positive PCR after treatment developed PKDL and 23% showed recurrence of VL symptoms with microscopic reappearance of parasites in the aspirates (Osman et al. 1998). In VL patients that relapsed after treatment, positive PCR results on peripheral blood almost always appeared before the clinical onset of disease (Lachaud et al. 2000; Pizzuto et al. 2001). Persistent infection in apparently healed scars has been reported for MCL (Delgado et al. 1996; Schubach et al. 1998). Guevara et al. (1994) reported the presence of Leishmania braziliensis in patients cured by immunotherapy or at different stages of treatment. They also found the parasite in subjects who had never suffered from leishmaniasis but who had lived in endemic areas and migrated to non-endemic regions many years earlier.
So far, only one report was published on the use of real-time PCR, namely for the quantification of Leishmania parasites in mouse liver (Bretagne et al. 2001). In the near future, this technique may play an increasingly important role in the quantification of promastigotes or amastigotes present in clinical samples and in the (simultaneous) identification of the infecting species or strain.
It can be concluded that PCR is capable of detecting the Leishmania parasite in a variety of clinical samples and for all clinical manifestations of the disease. PCR has caused a revolution in the diagnosis of leishmaniasis. However, one has to keep in mind that the execution of this very sensitive technique requires precautions:
• the risk of contamination necessitates dedicated laboratory areas for mix preparation, sample preparation and amplification/detection;
• false positive results can be further prevented by using the uracil nucleotide glycosylase/dUridine tri-phosphate (UNG/dUTP system) (Longo et al. 1990);
• appropriate and sufficient positive and negative controls need to be included in each experiment to confirm the sensitivity and specificity of the technique.
Another important issue that, so far, has hardly been addressed is the standardization of PCR technology for the diagnosis of leishmaniasis. Most laboratories use `in house' PCR methods that are based on different primer pairs and DNA targets; the Leishmania PCR does not exist (Lachaud et al. 2002). There are only few comparative studies available in the literature (Meredith et al. 1993; Reithinger et al. 2000; Lachaud et al. 2002) and it would be very valuable to establish a universal PCR for the diagnosis of leishmaniasis.
Nucleic acid sequence-based amplification (NASBA)
Although PCR certainly has proved its merit in detecting Leishmania parasites and the diagnosis of leishmaniasis, a disadvantage of this technique is the fact that it is based on the detection of parasite DNA, which may be present a long time after the parasite has been cleared. NASBA technology, for the amplification of specific RNA sequences, has proven to be a very sensitive and specific assay in diagnostic microbiology (Compton 1991). Tests have been developed for HIV, human papillomavirus, mycobacteria and Plasmodium falciparum. The technique has not yet been developed for leishmaniasis. NASBA has several advantages over PCR: it detects RNA in a background of DNA and may thus serve to measure viable parasites; the NASBA reaction is isothermal (a thermo-cycler is not required) and quick (90 min); it is specific and sensitive, as little as 10–100 target molecules in a sample can be amplified. Moreover, NASBA can be used for the accurate quantification of RNA levels, which allows the accurate quantification (i.e. determining the actual number) of the infectious agent (Schoone et al. 2000). Quantitative analysis of RNA levels after drug treatment could be a useful method to assess the efficacy of anti-Leishmania treatment.
Despite these advantages, NASBA is not widely used. This not only holds true for leishmaniasis, but with the exception of HIV/AIDS, for most other diseases where its value was proved (such as malaria, tuberculosis and leprosy) as well. This is probably because of the fact that PCR and RT-PCR already fill the niche where NASBA could be of value.