High Prevalence of Asymptomatic Leishmania spp. Infection Among Liver Transplant Recipients and Donors From an Endemic Area of Brazil

Authors


Abstract

Visceral leishmaniasis is an uncommon disease in transplant recipients; however, if left untreated, the mortality can be high. If an organ donor or recipient is known to be an asymptomatic Leishmania spp. carrier, monitoring is advised. This study proposes to assess the prevalence of asymptomatic Leishmania spp. infection in liver transplant donors and recipients from an endemic area. A total of 50 liver recipients and 17 liver donors were evaluated by direct parasite search, indirect fluorescent antibody test (IFAT), anti-Leishmania rK39 rapid test and Leishmania spp. DNA detection by polymerase chain reaction (PCR). Leishmania spp. amastigotes were not observed in liver or spleen tissues. Of the 67 serum samples, IFAT was reactive in 1.5% and indeterminate for 17.9%, and the anti-Leishmania rK39 rapid test was negative for all samples. The PCR test was positive for 7.5%, 8.9%, and 5.9% of blood, liver and spleen samples, respectively (accounting for 23.5% of the donors and 8% of the recipients). Leishmania infantum-specific PCR confirmed all positive samples. In conclusion, a high prevalence of asymptomatic L. infantum was observed in donors and recipients from an endemic area, and PCR was the most sensitive method for screening these individuals.

Abbreviations
EIA

enzyme immune assay

HAI

hemagglutination assays

IFAT

indirect fluorescent antibody test

IFI-LH®

indirect immunofluorescence

LT

liver transplant

MELD

Model for End-Stage Liver Disease

PBS

phosphate buffered saline

PCR

polymerase chain reaction

UNG

uracil N-glycosylase

VL

visceral leishmaniasis

Introduction

Visceral leishmaniasis (VL) is endemic in 65 countries on four continents, and 500 000 new cases occur each year. There has been a significant increase in disease incidence in the last two decades, with 90% of all cases concentrated in India, Nepal, Sudan, Bangladesh, and Brazil [1]. In Brazil, approximately 3500 cases are reported yearly, with a steady increase in urban areas [2, 3]; Brazilian cases account for up to 90% of all cases in the Americas. Although VL is not a commonly reported infectious disease in transplant recipients, the risk of infection (especially in endemic areas) and the high lethality rate of this disease if left untreated warrant attention [4]. High proportions of organ donors and recipients live or have lived in endemic areas and could have been exposed to Leishmania spp. The early detection of infection is crucial for adequate follow-up and prompt treatment of organ transplant recipients.

The gold standard for the diagnosis of VL involves the demonstration of the parasite in smears and/or cultures of the spleen, bone marrow, lymph nodes or liver. These techniques are invasive and require clinical and laboratory expertise. Although highly specific, their sensitivities vary depending on the tissue evaluated. Serological tests have been used to replace or complement parasitological diagnosis. In Brazil, two tests are available at the National System of Laboratories of the Brazilian Ministry of Health: indirect immunofluorescence (IFI-LH®) and, since 2010, Kala-Azar Detect® kits (a rapid test using a recombinant K39 antigen-rK39 performed on serum samples). The former test is also available at reference hospitals. Leishmania DNA detection by polymerase chain reaction (PCR) is a highly efficient tool for the diagnosis of VL in different biological samples. The PCR for Leishmania DNA in peripheral blood is also useful for the diagnosis of asymptomatic infection [5].

The aims of this study were to assess the prevalence of asymptomatic Leishmania infantum infection in liver transplant (LT) donors and recipients and to evaluate the risk of transmission through the graft or the development of manifest disease.

Materials and Methods

Study population

This cross-sectional prospective study included patients who received livers from deceased donors at the Organ Transplant Service of Alfa Gastroenterology Institute, University Hospital, Federal University of Minas Gerais, from August 2010 to September 2011. The Model for End-Stage Liver Disease [6] was adopted to select recipients according to the severity of liver disease. The study utilized convenience samples; the sample size was calculated based on 74 transplants (including four re-transplants) during the referred period. With a regional VL frequency in the asymptomatic general population of 2%, 25 individuals were necessary to obtain 5% precision and a confidence level of 95%. Donor samples were included only when the donor hepatectomy and the transplant procedure were performed in the same city (n = 17). Demographic and clinical information on donors and recipients were obtained from the transplant database, which is provided by an electronic medical record of the Transplant Service. The Ethical Review Boards and Ethical Committees approved the study (CAAE 0311.0.203.245-08; ETIC 311/08). Donor samples were obtained following the provision of authorization by family members and transplant recipients who signed Informed Consent Terms. No donor or recipient had a previous history VL disease or treatment.

The following biological samples were collected to investigate Leishmania infection: (a) 5 mL of blood in an EDTA tube; (b) 5 mL of blood collected without anticoagulant (Esterile interior®, Shandong Weigao Medical Polymer Group Co. Ltd-RPC, Weihai, China) from the donor (during the hepatectomy surgery) and from the recipient (during the transplant procedure); (c) a fragment of liver tissue collected with a specific hand-held needle (16 g 10 cm Velox®, Medax, Italy) from the donor and from the recipient (explanted liver); and (d) a fragment of spleen tissue from the donor. Whole blood collected in EDTA was stored at −70°C until Leishmania DNA PCR was performed. Samples were centrifuged and stored at −20°C until serological tests were performed. Liver and spleen fragments were prepared for cytological examination. The biopsy fragment samples for PCR were stored in liquid nitrogen and, subsequently, frozen at −70°C prior to tissue PCR. The serological or molecular tests were performed in the Reference Laboratory Center for Leishmaniasis; this center has specific protocols and follows the recommendations of the Ministry of Health. Direct examination was performed immediately after receiving the samples. The PCR assays were performed once samples collection for the study was complete.

Laboratory tests

Donor and recipient samples were serologically screened for Chagas disease with enzyme immune assay and indirect fluorescent antibody test (IFAT) or hemagglutination assay.

Stained smear

Immediately after biopsy, liver and spleen tissue imprints were performed, followed by fixation on slides. For each fragment, one slide was prepared with eight imprints. The slides were fixed with methanol, stained with Giemsa and observed by oil immersion microscopy (100×). The slides were then analyzed by qualified professionals.

The current reference test for VL diagnosis is the demonstration of Leishmania in bone-marrow aspirate. For ethical reasons, we could not perform any invasive procedures in transplant recipients with no symptoms related to VL infection.

Leishmania culture

Leishmania culture was not performed considering cost, time required, contamination rate and the low sensitivity in an asymptomatic infection setting.

Indirect immunofluorescence test

The IFI-LH® test (IFI-Leishmaniose Humana-Biomanguinhos, Fiocruz, Brazil) was performed according to the manufacturer's instructions. Briefly, slides were coated with 10 µL of antigen at room temperature for 12 h. For each reaction, 10 µL of a twofold serum dilution (1:40 to 1:640) in phosphate buffered saline (PBS) was added over the slide holes. After incubation in a humid oven at 37°C for 30 min, the slides were washed for 3 min in PBS and for 3 min in distilled water. The conjugate was diluted to 1:800 in PBS with 0.025% Evan's Blue, and 15 µL of this solution was placed over the slide holes. The incubation and washing steps were repeated. Slides were mounted with buffered glycerin, covered with a cover slip and read under an Olympus BX FLA fluorescent microscope (Hicksville, NY) equipped with a 100 W mercury lamp with 400× magnifying power.

According to the recommendations of the Brazilian Ministry of Health [2], a result was considered positive when titers ≥1:80; titers equal to 1:40 were considered indeterminate.

Anti-Leishmania rK39 rapid test

The rapid test (Kala-Azar detect®; InBios International, Seattle, WA) was performed according to the manufacturer's instructions. Briefly, 20 µL of serum was added to the dipstick, followed by two or three drops of chase buffer. Results were interpreted in just 10 min and declared positive or negative. Even a faint reactive line was considered positive.

PCR assay

Total DNA from blood and biopsy samples was extracted using the QIAamp® DNA MiniKit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's recommendations. The primers used were previously described by Degrave et al [7] and were designed to amplify a 120-bp sequence from the conserved region of the kDNA minicircle (150-GGGG/TAGGGGCGTTCTC/GCGAA and 152C/GC/GC/GA/TCTATA/TTTACACCAACCCC). The detection limit of this reaction was 0.1 fg, as previously determined using eight 10-fold serial dilutions of genomic DNA from the reference strain L. infantum (MHOM/BR/74/PP75). For PCR amplification, blood DNA samples were diluted fivefold, organ DNA samples were diluted 10-fold, and 1 µL was used as a template. PCR was performed in a final volume of 10 µL containing 1.5 U Platinum Taq DNA Polymerase (Invitrogen, São Paulo, Brazil); 1× PCR buffer (200 mM Tris–HCl, pH 8.4, 500 mM KCL; Invitrogen); 0.6 µM each primer; 2.0 mM MgCl2; 1 U uracil N-glycosylase (UNG; Applied Biosystems, Life Technologies, Carlsbad, CA); 400 µM dATP, dCTP and dGTP; and 800 µM dUTP (Promega, Fitchburg, WI). The cycling program was preceded by 10 min of incubation at 37°C (UNG digestion period) and 5 min of incubation at 95°C (Platinum Taq activation and UNG inactivation), followed by 37 cycles of 95°C for 30 s, 60°C for 30 s, and, finally, an elongation step at 72°C for 5 min. The samples were kept at 72°C or stored immediately at −20°C to prevent residual UNG catalytic activity. A positive control based on DNA from the reference strain MHOM/BR/74/PP75 was included in all tests. Negative controls containing all of the elements of the reaction mixture except DNA were also included in each PCR assay to assess contamination. As an internal quality control measure for the DNA isolation procedure and the possible presence of PCR reaction inhibitors, the human beta-actin gene was PCR amplified in all clinical samples with the primers Aco1 and Aco2 [8]. Five microliters of the PCR-amplified material was subjected to electrophoresis on a 6% polyacrylamide gel and analyzed by silver staining. Strict physical methods (separation of rooms and materials and the use of bleach and a laminar flux chamber with ultra violet light) were employed during the procedure to minimize sample contamination. A carryover prevention method was also incorporated using UNG, an enzyme that cleaves the uracil base from the phosphodiester backbone of uracil-containing PCR products prior to amplification, which does not affect natural (i.e. thymine containing) DNA.

A PCR assay was performed to confirm infection by L. infantum. The primers were the following: sense RV1-CTTTTCTGGTCCCGCGGGTAGG and antisense RV2-CCACCTGGCCTATTTTACACCA. These primers amplify a 145-bp fragment of kDNA from L. infantum. The protocol used in the reaction was proposed by Le Fichoux et al [9].

Data analysis

Data were analyzed using the SPSS Base 19.0 for Macintosh (IBM, Chicago, IL).

Results

Demographic and clinical characteristics of liver donors and recipients in Belo Horizonte, MG, Brazil

During the study period, 70 patients underwent 74 transplants (53 male and 17 female; median age 53 years, range 10–70 years). Samples from 50 recipients (35 male and 15 female; median age 52 years, range 10–70 years) and from 17 corresponding liver donors (8 male and 9 female; median age 36 years, range 5–62 years) were included. The remaining samples were not available due to operational difficulties, mainly communication failure. Nevertheless, it is unlikely that selection bias has occurred, since patients were randomly included. The majority of LT recipients (60%) lived in Minas Gerais state. The main indications for LT were ethanolic cirrhosis in 16 patients (32%), hepatitis C cirrhosis in 11 patients (22%) and cirrhosis due to autoimmune hepatitis or cryptogenic and primary sclerosing cholangitis in four patients (8%). Hepatocarcinoma was diagnosed in 13 (26%) cases.

Parasitological and serological tests

Amastigote forms were not detected by direct microscopy in either the liver (n = 67) or spleen (n = 17) fragments (Table 1).

Table 1. Frequency of asymptomatic Leishmania infantum infection in liver transplant donors and recipients as assessed by invasive and noninvasive laboratory tests
Laboratory testRecipients (%)Donors (%)Total (%)
  • IFI-LH®, indirect immunofluorescence test; Kala-Azar detect®, anti-Leishmania rK39 rapid test; NA, not available; PCR, polymerase chain reaction.
  • 1Concordant positive donor samples: three individuals showed PCR-positive results in their blood and liver samples and one of them showed PCR-positive results in the blood and spleen samples.
Direct parasite search
Liver samples
Positive000
Negative50 (100)17 (100)67 (100)
Spleen samples
Positive000
NegativeNA17 (100)17 (100)
IFI-LH®
Positive (≥1:80)1 (2)01 (1.5)
Indeterminate (1:40)10 (20)2 (11.8)12 (17.9)
Negative (<1:40)39 (78)15 (88.2)54 (80.6)
Kala-Azar detect®
Positive000
Negative50 (100)17 (100)67 (100)
PCR
Blood samples
Positive1 (2)4 (23.5)15 (7.5)
Negative49 (98)13 (76.5)62 (92.3)
Liver samples
Positive3 (6)3 (17.7)16 (8.9)
Negative47 (94)14 (82.3)61 (91.1)
Spleen samples
PositiveNA1 (5.9)11 (5.9)
NegativeNA16 (94.1)16 (94.1)

IFI-LH® was performed on 67 samples. Only one recipient had positive results, with a titer of 1:80. Samples from 2 donors and 10 recipients showed titers of 1:40 (considered as indeterminate). The rapid test with the rK39 antigen was negative for all samples.

Serological tests for Chagas disease were negative for all donors and recipients.

PCR for Leishmania spp. and for L. infantum infection in liver transplantation donors and recipients

Blood PCR was positive for parasitic DNA in 4/17 (23.5%) donors and 1/50 recipients (2%). For three (17.6%) of the four positive donor peripheral blood samples, the liver samples were also positive. For the remaining positive donor peripheral blood sample (5.9%), the spleen sample was also positive (Tables 1-3). The species-specific PCR amplified the DNA of L. infantum in all specimens with positive kDNA PCR.

Table 2. Distribution of PCR-positive results in spleen, liver and blood samples from organ donors and recipients
PatientPCR+Total
SpleenLiverBlood
  • NA, not available; PCR, polymerase chain reaction.
  • *Total of donors included (17).
  • **Total of recipients included (50).
  • ***Total of patients included: 67 (donor and recipients).
Donor1012
Donor0112
Donor0112
Donor0112
Positive donors   4/17*: 23.5%
RecipientNA101
RecipientNA101
RecipientNA011
RecipientNA101
Positive recipients   4/50**: 8%
Total1658/67***: 12%
Table 3. PCR, IFAT and rK39 results of 17 liver donors and their corresponding recipients
No. patientDonorRecipient
PCRIFATrK39PCRIFATrK39
BloodLiverSpleenBloodLiver
  1. IFAT, indirect fluorescence antibody test; Ind, indeterminate; PCR, polymerase chain reaction; rK39, anti-Leishmania rK39 rapid test.
1Ind
2++Ind
3++
4Ind
5+
6
7Ind
8
9Ind
10++
11++
12
13+
14
15
16
17+

None of the recipient or donor samples yielded concordant results in the IFI-LH® test (1:40) and PCR identification in blood or tissues. The recipient who had a positive IFI-LH® test (1:80) had negative PCR results in both blood and liver.

The results of PCR, anti-rK39 and IFI-LH® of the 17 donors and the corresponding recipients are presented in Table 3. All recipients who received an organ from positive-PCR donor (blood, spleen or liver) were negative for all tests (by direct examination, serological or molecular methods). No recipient developed any symptoms of VL within a median follow-up of 24 months (0–34 months) and they did not receive prophylaxis with anti-Leishmania drugs.

Discussion

VL is an uncommon infectious complication in transplant recipients. Nevertheless, it is important to be aware of the risk, since a high proportion of donors and recipients live or have lived in endemic areas and thus could have been exposed to Leishmania protozoa. In these patients, the early diagnosis of symptomatic VL may be crucial for successful disease treatment. In the present study, the prevalence of Leishmania infection in asymptomatic liver donor and recipients was assessed. The serological methods (IFAT and rK39) yielded discrepant results, and PCR identified eight cases of VL infection in 67 individuals (11.9%).

In Brazil, there is a high VL infection rate and a large public organ transplant program [10]. Minas Gerais state has one of the highest VL rates [11], and its capital, Belo Horizonte city, is considered an endemic area [12]. The prevalence of asymptomatic infection in Belo Horizonte is as high as 2% [13]. Leishmaniasis is an uncommon disease among transplant recipients, but the number of published cases has increased in recent years [14-21]. More than 100 VL cases following transplantation have been reported worldwide [14-22], and 3 of the 11 cases after LT were described by our group [14].

The IFI-LH® is the laboratory test recommended by the Brazilian Ministry of Health for the diagnosis of symptomatic infection [2]. DNA detection and anti-Leishmania rK39 tests are new tools for the VL diagnosis. However, their appropriateness for infection screening remains unclear [3, 14, 22]. International guidelines for the management of transplant recipients recommend specific serology in donors and recipients from endemic areas, regardless of laboratory tests limitations and availability [23]. The screening of transplant candidates and donors could reduce mortality by the identification of high-risk patients, allowing prophylaxis or early diagnosis and therapy when symptomatic disease occurs.

In this study, IFI-LH® identified 1 positive recipient and indeterminate titers in 12 other samples (2 donors and 10 liver recipients). In Brazilian patients with clinical disease, IFI-LH® has marginal sensitivity and specificity of 81% and 88%, respectively [24]. These percentages are even smaller for asymptomatic infection, as was the case in the investigated population. IFI-LH® is considered positive with promastigotes fluorescence in a cutoff dilution of 1:80. Lower cutoff (1:40 indeterminate result) is considered positive, when the patient is from an endemic area and presents clinical manifestations. In asymptomatic individuals the positive predictive value of IFAT is lower and indeterminate results may be in fact false positive. Often, when a more specific method is used (as the rK39) and compared with IFAT, there is no agreement, as in this present study. Furthermore, given the potential for cross reactivity in serological assays (especially with Chagas disease), it is important to assess the prevalence of this infection. In our study, none of the samples were positive for Chagas disease (data not shown).

Other more specific techniques are currently recommended for the serological diagnosis of VL. The rapid test for the K39 antigen has a sensitivity of 84–92% and a specificity of 97% in symptomatic patients [24, 25]. In our study, none of the samples were positive by the rapid test. It is possible that the rK39 assay is more useful to diagnose VL disease than to detect VL infection. It could have a different performance in immunocompromised individuals (as recipient candidates). The lower sensitivity of the rK39 test for asymptomatic infection has also been reported in other studies [25-27].

The PCR assay, which generally has high sensitivity and specificity, was performed in blood, liver and spleen to increase the screening power [28-32]. In this study, eight asymptomatic individuals were identified by PCR. Four asymptomatic donors (23.5%) showed concomitant presence of Leishmania DNA in their peripheral blood, liver or spleen. The presence of parasite DNA in the reticuloendothelial system was then confirmed, suggesting that the transmission of Leishmania spp. by liver donors in areas of high disease prevalence may be underestimated. Notably, in those 17 paired donors and recipients, all of the four positive PCR donors had negative corresponding recipients (by PCR and serology), which provides a very interesting group for monitoring. Up to the present day, no recipient participating in this study developed any VL sign or symptom within a median follow-up time of 24 months, particularly considering that the onset of posttransplant VL occurs with a median delay of 18 months.

This study suggests that VL screening by PCR could be performed to identify high-risk patients for VL reactivation (recipient) or transmission (donor), in the transplant scenario. Unfortunately, our study was not large enough to define a management protocol, however, we demonstrated that serological methods might not be sufficiently sensitive or specific for screening. PCR (in blood or tissue samples) is certainly more sensitive, nevertheless a positive result does not guarantee disease development. In general, if the donor is infected, Leishmania transmission is expected, but disease development does not always occur. Prophylaxis after transplantation may be of some benefit, but this issue remains unresolved because we do not know the impact of this approach, since the incidence of VL disease in transplanted patients is low, even in endemic areas [14], and disease development can occur late after transplantation. Another issue is that Leishmania PCR may present a high cost or not be available, and this could vary according to the geographical area, depending on leishmaniasis endemicity, financial and scientific resources.

The greatest problem associated with VL in transplanted patients is the undiagnosed patients with symptomatic infection. These patients might be more difficult to identify in nonendemic areas than in endemic areas, where VL disease is more likely to be suspected and the risk of infection remains high even after transplantation. Therefore, screening to establish risk, perform monitoring or clinical follow-up in order to allow early diagnosis and prompt treatment may be of interest in high-risk patients of nonendemic areas.

To our knowledge, this is the first study to evaluate the serological, molecular and histological profiles of LT donors and recipients for VL. In our opinion, VL serology should not be used for screening, since they often present no agreement in infected asymptomatic individuals [25-27, 33-36]. In this context, PCR assay could be applied with higher sensitivity and specificity. However, PCR availability and cost may vary according to the concerned area. We consider that it can be used if there is an epidemiological risk of VL transmission or reactivation in nonendemic areas, since the recognition of the disease might be more difficult. However, prospective studies with a longer observation period are necessary to determine which tool may better identify an increased risk of developing VL.

Acknowledgments

The authors acknowledge the professionals of the Transplant Team of IAG HC/UFMG, who are dedicated to patient care and made this study possible. This work was supported by “Fundação de Amparo a Pesquisa do Estado de Minas Gerais” (FAPEMIG), Centro de Pesquisas René Rachou, Fiocruz and Pró-reitoria de Pesquisa da Universidade Federal de Minas Gerais. Ana Rabello is a research fellow of the Brazilian National Counsel of Technological and Scientific Development (CNPq).

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary