- Email: [email protected]
Evaluation of a newly developed direct agglutination test (DAT) for serodiagnosis and sero-epidemiological studies of visceral leishmaniasis: comparison with IFAT and ELISA
TF.ANSXTION~ OF THE ROYALSocr~n OF TROPICALhkXINE
603
AND HYGIENE(1987)81, 603-606
Evaluation of a newly developed direct agglutination test (DAT) for serodiagnosis and sero-epidemiological studies of visceral leishmaniasis: comparison with IFAT and ELISA A. E. HARJTH’, A. H. J. KOLK’,
P. A. ICAGER*, J. LEEUWENBIJRG~, F. J. FABER*, R. MUIGAI~, S. KIUGU~ AND J. J. LAARMAN’
‘Swelkxgrebel Laboratory of Tropical Hygiene, Royal Tropical Institute, Amsterdam and Laboratory of Parasitology, University of Amsterdam, Amsterdam, The Netherlands; *Unit of Tropical Medicine and Infectious Diseases, University of Amsterdam, Amsterdam; 3Medkal Research Centre, Kenya Medical Research Institute, Nairobi, Kenya; 4Clinical Research Centre, Kenya Medical Research Institute, Nairobi, Kenya
Abstract
A newly developed direct agglutination test (DAT) for visceral leishmaniasis, IFAT and ELISA were applied to sera of patients with visceral leishmaniasis, African and American trypanosomiasis, other parasitic infections and healthy controls. The sensitivities of the 3 tests were cornJ..arable (96.3% to 100%); excluding patients with African and American trypanosomiasis, the spec cltles of DAT and IFAT were 100% and ELISA 87.3%. When trypanosomiasis serawere included, the specificities were 72*6%, 94.3% and 79.4% in DAT, IFAT and ELISA respectively. In 273 sera from a leishmaniasis endemic area (Baring0 District, Kenya), the sensitivity was 80% in DAT and IFAT and 60% in ELISA, specilicities being 99.6% (DAT), 985% (IFAT) and 625% (ELISA). As the new DAT is economical and easy to perform, it is recommended for sero-epidemiological visceral leishmaniasis.
Introduction The final diarmosis of visceral leishmaniasis. bv demonstration gf the parasite, may not be easily
applicable on a large scale in the field and early infections, due to low numbers of parasites, may escape diagnosis. Serological techniques that aid in
selecting suspectedindividuals help to avoid, asmuch as possible, invasive diagnostic procedures. The indirect immuno&oresc&ce antibody test (IFAT) is one of the most commonlv used technioues (DUX~JRY & SADUN, 1964; SHAW & VOLLER, 1464; BRAY & LAINSON, 1%5; BEHFOROUZ et al., 1976; LATIF et al., 1979). Results are satisfactory but
wide-scale application is hampered by lack of technical facilities in endemic areas. Since its introduction by HOMMEL et al. (1978), the enzyme linked immunosorhent assay (ELISA) has heen used frequently (EDRISSIAN& DARUAN, 1979; RASSAM 81AL-MUDHAFFAR, 1980; SRNASTAVA et al., 1979; HO et al., 1983). To suit routine screening and large scale application in less equipped laboratories and under field circumstances, further simplification of this assaywas achieved (PAPPASet al., 1983, 1985; JAHN & DIESFELD, 1983; MOHAMMED et al., 1985). A simple and economical direct agglutination test (DAT) for the diaanosis of visceral leishmaniasis. iecentiy developed & our laboratory (HARITH et al.; 1986), is compared here with IFAT and ELISA. A?Uigt??l
Materials and Methods
Leishmaniahani strain I-S (kindly providedby Dr J. Schottelius, Hamburg) was maintained in wizrousing NNN
Correspondence: A. E. Ha&h, Swellengrebel Laboratory of Tropical Hygiene, Royal Tropical InSdNte, Meibergdreef 39, 1105 AZ Amsterdam Z-O, The Netherlands.
field work on
medim?. Procedures for mass product@ of promastigotes :;8mgen preparauon have been described (HARITHet al., Test fwocedures Diiect agglutination test (DAT). Antigen preparation has been described @bRtTH et al., 1986). In brief, centrifuged promastigotes were washed in cold Locke’s solution, subjected to trypsin treatment, washed, and fixed in 2% formaldehyde in Locke’s solution at 4°C for 20 h. Promastigotes we& stained with Coomassie brilliant blue and the concentration adjusted to 1 x 10sparasites/ml in 1% formalin in physiological saline. After filtration through nylon gauze of 30 w mesh size, the antigen was stable in the dark at 4°C for 75 d at least. Recent studies showedthat antigen preparations wrapped in album foil were stable for one month at 521°C or 2 weeks at ambient temperatures of 18-37’C (Baring0 District? Kenya; unpublished results). The test was performed m V-shaped micro&e-plate wells with a starting serum dilution of 1 : 100 in 1% of fcetal
bovineserumin physiologicalsaline.After overnightincubation at 21”C, tbe titre in the test serum was read as one dilution beforethat yielding a clear, sharpblue spot similar to the one in the negative control well. Individuals with titres
greater than 1 : 1600 were considered DAT seropositive (HARITH et al., 1986). Immuna~?wescenceantiboa’y te.vt(IFAT). Washed promastigotes were dispensed on small circles left uncovered on glass slides sprayedwith Teflon (KOLK et al., 1984). After drying and fixation in acetone, a previously determined serum dilution of 1 : 100 was used throughout. As conjugate, sheepantihuman IgG(H + L) labelled with FITC (Pasteur Institute, Paris) was used. Fluorescence intensity was arbitrarily assessedas + , 2+, 3+ or 4+. Since a considerable number of individuals without visceral leishmaniasis showed + Buorescence.22+ was considered IFAT seropositive. Enzyme-linked immunosorbentassay(ELBA). Supematant of a son&ted promastigote suspension in 0.5 ELMsodium carbonate (pH 9.7), equivalent to 2 x 10’ parasites/ml, was centrifugedat 10000 x g and used for coating polystyrene
604
SERODIAGNOSIS
AND
SERO-EPIDEMIOLOGY
microtitre ELISA plates. Preliminary results showed that 1 : 500 serum dilution was optimal in discriminating between visceral leishmaniasis p&ients and individuals with other conditions. All sera were thus screenedat this dilution in triplicate. The diluent used was phosphate-buffered saline (PBS) at pH 7.2, containing 05% bovine serum albumen and 0.05% Tween 20. Goat antihuman IgG (H + L) labelled with peroxidase (Pasteur Institute, Paris, France), diluted 1 : 1000,was used as conjugate and the assay was read at 492 nm. Values greater than 0.28 (the mean of 10 African and 4 Dutch healthy control sera plus 3 times the standard deviation) were considered ELISA seropositive. A considerable number of persons without visceral leishmaniasis showed 30.28 absorbance. Taking absorbance = 057 as cut-off point resulted in a higher specificity but lower sensitivity, compared to O-28.
LEISHMANIASIS
(0 28 healthy controls including 10 Ghanaians, 10 Brazilians and 8 Dutch. (g) 273 individuals from the Perkerra area, Baring0 District in Kenya, a known focus of visceral leishmaniasis (LEELJWENBURG er al., 1982). Sera were collected as part of a continuing epidemiological study in March-April 1984. Included in this group were 3 cured visceral leishmaniasis patients, 10 of their household members, and the inhabitants of the wide surroundings of their homesteads. Amastigotes were later demonstrated in splenic aspirates of 2 persons admitted to the district hospital with suspected visceral leishmaniasis. All persons were examined for splenomegaly; thin and thick blood films were examined for malaria parasites from all persons with splenomegaly and 232 sera were examined for Brucelkz antibodies (from the remaining 41 no serum was available for this test). Results
Serum samples
In Table 1 are presented the serological results obtained with DAT, IFAT and ELISA. All recent and treated visceral leishmaniasis patients, except for one in ELISA, had values indicative of the disease in all three tests. Sera from recent and treated casesof
Serawere collected from the following groups of individuals. (a) 27 patients with proven visceral leisbmaniasis (Kenya) including 17 collected before, during or up to one month after treatment, and 10 taken 4-14 months after treatment when the patients were apparently cured. (b) 25 patients with African trypanosomiasis (Zaire), provided by Dr E. Magnus (Antwerp, Belgium). (c) 20 patients with Chagas disease, provided by Dr M. E. Camargo (So Paulo, Brazil). (d) 10 patients with schistosomiasis mansoni (Kenya). (e) 9 patients with giardiasis and 8 with toxoplasmosis (all from The Netherlands).
Table l-Results conditions
Group*
OF VISCERAL
of the newly developed DAT,
IFAT,
but
less so in DAT
and ELISA.
The
sensitivities of 3 tests in this po ulation were 100%for DAT and IFAT, and %-3% Por ELISA. The specificities were 72.6% (DAT), 94.3% (IFAT) and 79.4% (ELISA)
when trypanosomiasis
sera were included;
IFAT and ELISA on sera of patients with visceral leishmaniasis
DAT titre <1600 3200-25 600 a51 200
No.
visceral leishmaniasis showed comparable results in
IFAT intensity of fluorescence >2+ +
Eg;
:3
8
0
17
i
:
t%
El
:;:
i
;
13
12
and other
ELISA absorbance <0.28 0.28-O-56 0.57-2.00
:i 0
iI
1
is
1:
: z :1
i?
10 9
10 9
i
iI
1:9
i
8
;:
i
is:
2:
2;
ii0
i0
2:
:0
80
2;
i4
15
*VL(R): visceral leishmaniasis (recent); VL(T): visceral leishmaniasis (treated); AT: African trypanosomiasis; CD: Chagas disease; SM; scbistosomiasis mansoni; GI: giardiasis; TO: toxoplasmosis; HC: healthy controls.
Table 2-Results Kenya
of DAT,
IFAT
and ELISA
Diagnosis
No.
Brucellosis Malaria Splenomegaly** Apparently healthy Recent VL Treated VL*** Sensitivity Specificity Predictive Predictive
on 273 sera collected from leishmaniasis
% % value of positive test % value of negative test %
it
5
130 2 3
DAT al:3200
endemic area (Baring0 district)
ELISA* 30.28 >0.56
x
i
:
0 1 :
: :
2.6 ii.6
z.5 62.5 98.5
* Serum dilutions in IFAT and ELISA were 1:lOO and 1:500 respectively. ** Brucellosis, malaria and visceral leishmaniasis (VL) patients not included. *** 6 months (one patient) and 36 months (2 patients) after cure.
i
1 1
t: 2 1
::
fS.5 0.29 98.8
it 0.69 90
;
in
A.
by excluding ;bLg;d-100%
E.
HARITH
them, higher specificities were in DAT and IFAT and 87.3% in
In DAT none of the 45 trypanosomiasis sera cross-reacted at 3 1 : 51 200 dilution, while 6 in IFAT and 4 in ELISA did so at 32+ fluorescenceand 30.57 absorbance. African trypanosomiasis sera showed relativelv more cross-reactivitv in ELISA than in DAT and IFAT, while chagasic sera were more cross-reactive in IFAT than DAT and ELISA. In DAT, 8 African trypanosomiasis sera had titres up to 1:25 600 and all 20 American trypanosomiasis sera scored values inseparable from those of healthy controls (
The newly developed DAT was sensitive and specific even when African and American trypanosomiasis serawere included. All seraof chagasicpatients showed titres inseparable from those of healthy controls and only 8 out of 25 African trypanosomiasis sera had titres in the range seen in treated visceral leishmaniasis patients; none had titres as high asthose seen in recent visceral leishmaniasis infections (351 200). When applied to sera from an endemic area, the sensitivity and specificity remained high; a negative test had a very high predictive value and that of a positive test was only slightly reduced when treated visceral leishmaniasis patients (up to 36 months after cure) were included. Relatively fewer cross-reactions were found with the IFAT than with ELISA in seraof healthy controls and patients with conditions other than visceral leishmaniasis. Malaria sera, in particular, crossreacted significantly with i. do&vani antigen in IFAT (HOMMEL et al., 1978). In our results such a degree of cross-reactivity was not found in any of 36 malarious sera. In 10 children suswcted for L. d. infanturn infection, and with posit&e IFAT results, LATIF et al. (1979) were not able to demonstrate Lei.hania narasites. Our resulk in ELISA agreewith those of HOMMEL et al. (1978). who detected L. donavant’ antibodies in 35 ouibf 36visceral leishmaniasis patients, 20 being parasitologically confirmed. Higher ELISA specifkitv. however. was found in sera collected from the s&e area (Baringo, Kenya) using sonicated (HO et al.. 1983: JAHN 81 DIESFELD 1983) and intact L. dotkani biomastigote antigens (MOHAMMED et al., 1985). In a study including L. donovani and Trypanasoma cruai antigens tested against homologous and heterologous sera, GUIMARAES et al. (1981) reported discrepancies in results with ELISA and not with IFAT. Our results with African trypanosomiasis sera with the 3 tests agree with those of VOLLER et al. (1975),
et
d.
605
who reported
“strong cross-reactions” in T. and T. crwi infections. Such cross-reactions were not observed by HOMMEL et al. (1978) or Ho et al. (1983). It remains inexnkable why such discrenancies exist in results obtained by various investigators with IFAT and ELISA. Thoueh HOMMEL et al. (1978) doubted that the reason was the choice of conjugate; in our experience variations in ELISA readings using the same antigen and serum samples were due to differences in batches of plates, conjugate, or both. Standardixation of these factors was tedious and time consuming. Because of its high sensitivity and speciflcity, &d the ease and low-cost of its application cornoared to IFAT and ELISA. the newly develowd DAT is recommended for mass screening programmes and sero-epidemiological studies of visceral leishmaniasis. rhodesiense, L. abnovani
Acknowledgements We thank Dr.E. Magnus and Prof. D. le Ray (Institute of Tropical Medicine, Antwerp) for supplying African trypanosomiasis sera and laboratory facilities; Dr M. E. Camargo (Instituto Medicina Tropical de Sio Paulo), for American trypanosomiasis sera; Dr J. Schottelius (Hamburg) for L. donmuni I-S strain. The technical assistance of Mrs S. Kuijper and Mr H. Gilis, and the secretarial assistanceof Mrs A. Gilaard and Mrs A. Heibloem, are highly appreciated. We thank Prof. M. Mugambi, Dr D. K. Koech and Dr S. N. Kinoti (Kenya Medical Research Institute, Nairobi) for their support of the leishmaniasis project m Baring0 District and for providing staff and facilities. This investigation received 5nancial support from the UNDP/World Bank/WHO Swcial Programme for Research and Training in Tropical Diseases. References Behforouz, N., Remi, H. R. & Gettner, S. (1976). Application of immunofluorescence to detection of antibody in Leishmania infections. Annals of Tropical Medicine and Parasiwlagy, 70, 293-301. Bray, R. S. & Lainson, R. (1965). The immunology and serology of leishmankis. 1. The fluorescent antibody staining technique. Transactions of the Royal So&@ of Tropical Medicine and Hygbne, 59, 535-544.
Duxbury, R. E. & Sadun, E. H. (1964). Fluorescent antibody test for the diagnosis of visceral leishtmmiasis. f2E &umal of Tropical Medicine and Hygme, 13,
Edrissian, Ch. H. & Darabisn, P. (1979). A comparison of enzyme-linked immunosorbent assay and fluorescent antibody tests in the serodiagnosis of cutaneous and visceral leishmaniasis in Iran. Transactions of the Royal Socie~ of Tropical Medicine and Hygiene, 73, 289-292. Guimaraes, M. C. S., Celeste, B. J., De Castilho, E. A., Mince, J. R. & Din&J. M. P. (1981). Immunoenzymatic assay (ELISA) in mucocutaneous leishmania& kala-azar and Chagas’ disease.An epimastigote Trypam somu cruzi antigen able to distinguish between antiTtypammma and anti-Lkskmania antibodies. American Jounual of Tropkal Medicine and Hygiene, 30, 942-947. Harith, A. E., Kolk, A. H. J., Kager, P. A., Leeuwenburg, J:, Muigai, R., Kiugu, S. & Laarman, J. J. (1986). A simple and economical direct agglutination test for serodiagnosisand sero-epidemiological studies of visceral leishmaniasis. Transactions of the Royal Sake@ of Tropical Medicine and Hygiene, 80, 583-587.
Ho, M., Leeuwenburg, J., Mbugua? A., Wamachi, A. & Voller, A. (1983). An enzyme-lmked immunosorbent assay (ELISA) for field diagnosis of visceral leishmaniasis. American Journal of Tropical Medicine and Hygiene, 32, 943-946.
SERODIAGNOSIS
606
AND
SERO-EPIDEMIOLOGY
Hommel, M., Peters, W., Ranque, J., Quilici, M. & Lanotte, G. (1978). The micro-ELISA technique in the serodiagnosisof visceral le.&man&is. Annals of Tropical Medicine and Parasitolo~, 72, 213-218. Jahn, A. & Diesfeld, H. J. (1983). Evaluation of visually read ELISA for semdiagnosis and sero-epidemiological studies of kala-azar in the Baring0 District, Kenya. Transactionsof the Royal Society of Tropical Medicine and Hygiene, 77, 451-454. Kolk, A. H. J., Ho, M. L., Klatser, P. R., Eggelte, T. A., Kuijper, S., De Jonge, S. & Van Leeuwen, J. (1984). Production and characterization of monoclonal antibodies to MycobEtetium tubercu@is,M. bovis (BCG) and Ei-gae. Climcal and Exp~mental Immunology, 58, Latif, B. k. A., Al-Shenawi, F. A. & Al-Alousi, T. I. (1979). The mdire-ctfluorescent antibody test for diagnosis of kala-azar infection in Iraq. Annals of Tropical Medicine and Parasitology, 73, 31-35. Leeuwenburg, J., Bryceson, A. D. M., Mbugua, G. G. & Siongok, T. K. (1982). The use of the leishmanin skin test to define transmission of leishmaniasis in Baring0 District, Kenya. East Ajiican MedicalJoumal, 60,81-84. Mohammed, E. A. E. R., Wright, E. P., Kager, P. A., Laarman, J. J. & Pondman, K. W. (1985). ELISA using intact promastigotes for immunodiagnosis of kala-azar. Transactionsof the Royal So&y of Tropical Medicine and Hygiene, 79, 344-350.
TRANSACTIONS OF THH
1Short
ROYALSOCIETY OF TROPICAL.
LEISHMANIASIS
Pappas, M. G., Hajkowski, R. & Hockmeyer, W. T. (1983). Dot enzyme-linked immunosorbent assay (dot-ELISA): a micro technique for the rapid diagnosis of visceral Iee;y. &wrnal of Immunologwal Methods, 64, Pappas, M. G., Hajkowski, R., Diggs, C. L. & Hockmeyer, W. T. (1985). Disposable nitrocellulose filtration plates simplify the dot-ELISA for serodiagnosis of visceral leishmaniasis. Transactionsof theRoyal Societyof Tropical Medicine and Hygiene, 79, 136. Rassam, M. B. & Al-Mudhaffar, S. A. (1980). The micro-ELISA sandwich technique for the quantitation of Leishmania donovani soluble antigen. Annals of Tropical Medicine and Parasitolo~, 74, 591-595. Shaw, J. J. & Voller, A. (1964). The detection of circulating antibodies to kala-azar by means of immunofluorescent techniques. Transactionsof the Royal Sociery of Tropical Medicine and Hygiene, 58, 349-352. Srivastava, L., Suri, J. C. & Sanyal, R. K. (1979). ELISA test in diagnosis of kala-azar in current epidemic areasin Bihar. 3ournal of CommunicableDiseases, 11, 183-187. Voller, A., Bidwell, D. 81Bartlett, A. (1975). A serological study on human Trypanosomarhodesiense infections using a micro-scale enzyme linked immunosorbent assay. Tropenmedizin und Parasitologic, 26, 247-251. Accepted
for publication I8 November 1986
MEDICINE AND HYG~WE (1987) 81, 606
Report1
Spontaneous healing of cutaneous Leiskwuznia braziliensis braziliensis ulcers J. M. L. COSTA, E. M. NETTO, K. C. VALE, N. K. OSAKI, M. S. TADA AND P. D. MARSDEN Nticleo de Medicina Tropical e NutriGao, Universida& de Brasilia, Brazil
We have previously reported spontaneous healing of Leiskmania braziliensis braziliensis (Lbbj skin ulcers over a variable time scale (MARSDEN‘et al:, 1984). No one has much experience of untreated L. brazihensis type infections as patients normally accept treatment advice. A recent epidemic in our field area of endemic Lbb transmission enabled us to extend our observations to a larger number of untreated patients. Due to the great nu;nber of patients requirini help diagnosis mainly relied on the leishmanin skin test and the indir&t immunofluorescent reaction coupled with the suggestive clinical picture and histopathology on biopsy. Although parasitological diagnosis is desirable such diaPnostic Drocedures are adeauate for Lbb (CUBA et-&., 1984). 8 cases(40%) heaed in under 6 months and 17 (85%) in under 12 months. Another 3 healed after 14, 99’and 128 months. One patient relapsed. These observations agree with our previous report of a variable evolution of these l&ions witliout treatment UV~RSDEN et al., 1984). Thev also ap;ree closely wit& the retrospectiv;
OF VISCERAL
survey by Dk BELD~R et
al. (1978) of 48 patients whose ulcers had healed without injection therapy. As spontaneousresolution of skin leishmaniasis can occur, it is necessary to compare the results of treatment with any candidate drug with results in a group receiving standard antimonial therapy (MARSDEN
et al., 1979).
Finally, -treated or untreated, patients with Lbb must be advised of the possibihty of future mucosal metastasis. They should be instructed that if they develop nasal inflammation with blockage, epistaxis, persistent secretion, hoarseness of the voice, or dysphagia they should seek medical advice. References Cuba, C. C., Llanos Cuentas, E. A., Barreto, A. C., Ma-es, A. V., Lago, E. L., Reed, S. G. & Marsden, P. D. (1984). Human mucocutaneous leishmaniasis in T&s Bracoos,Bahia-Brasil. An area of Leishmania braziliensis braziliensis. I. Laboratory diagnosis. Revista da So&dude deBrasileira deMedicina Tropical, 17,161-167. De Belder, M., Belby, M. J. & Akers, J. E. (1978). Report of the Cambridge Medical Ex dition to Brazil, 1977. Mimeographed document, 13F pp. Marsden, P. D., Cuba, C. C., Barreto, A. C., Sampaio, R. N. & Rocha, R. A. A. (1979). Nifurtimox in the treatment of South American leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 73, 391-394. Marsden, P. D., Tada, M. S., Barreto, A. C. & Cuba, C. C. (1984). Spontaneous healing of Leishmania braziliensis braziliensisskin ulcers. Transactionsof theRoyal Societyof Tropical Medicine and Hygiene, 78, 561-562. Accepted
for publication 2 February 1987
603
AND HYGIENE(1987)81, 603-606
Evaluation of a newly developed direct agglutination test (DAT) for serodiagnosis and sero-epidemiological studies of visceral leishmaniasis: comparison with IFAT and ELISA A. E. HARJTH’, A. H. J. KOLK’,
P. A. ICAGER*, J. LEEUWENBIJRG~, F. J. FABER*, R. MUIGAI~, S. KIUGU~ AND J. J. LAARMAN’
‘Swelkxgrebel Laboratory of Tropical Hygiene, Royal Tropical Institute, Amsterdam and Laboratory of Parasitology, University of Amsterdam, Amsterdam, The Netherlands; *Unit of Tropical Medicine and Infectious Diseases, University of Amsterdam, Amsterdam; 3Medkal Research Centre, Kenya Medical Research Institute, Nairobi, Kenya; 4Clinical Research Centre, Kenya Medical Research Institute, Nairobi, Kenya
Abstract
A newly developed direct agglutination test (DAT) for visceral leishmaniasis, IFAT and ELISA were applied to sera of patients with visceral leishmaniasis, African and American trypanosomiasis, other parasitic infections and healthy controls. The sensitivities of the 3 tests were cornJ..arable (96.3% to 100%); excluding patients with African and American trypanosomiasis, the spec cltles of DAT and IFAT were 100% and ELISA 87.3%. When trypanosomiasis serawere included, the specificities were 72*6%, 94.3% and 79.4% in DAT, IFAT and ELISA respectively. In 273 sera from a leishmaniasis endemic area (Baring0 District, Kenya), the sensitivity was 80% in DAT and IFAT and 60% in ELISA, specilicities being 99.6% (DAT), 985% (IFAT) and 625% (ELISA). As the new DAT is economical and easy to perform, it is recommended for sero-epidemiological visceral leishmaniasis.
Introduction The final diarmosis of visceral leishmaniasis. bv demonstration gf the parasite, may not be easily
applicable on a large scale in the field and early infections, due to low numbers of parasites, may escape diagnosis. Serological techniques that aid in
selecting suspectedindividuals help to avoid, asmuch as possible, invasive diagnostic procedures. The indirect immuno&oresc&ce antibody test (IFAT) is one of the most commonlv used technioues (DUX~JRY & SADUN, 1964; SHAW & VOLLER, 1464; BRAY & LAINSON, 1%5; BEHFOROUZ et al., 1976; LATIF et al., 1979). Results are satisfactory but
wide-scale application is hampered by lack of technical facilities in endemic areas. Since its introduction by HOMMEL et al. (1978), the enzyme linked immunosorhent assay (ELISA) has heen used frequently (EDRISSIAN& DARUAN, 1979; RASSAM 81AL-MUDHAFFAR, 1980; SRNASTAVA et al., 1979; HO et al., 1983). To suit routine screening and large scale application in less equipped laboratories and under field circumstances, further simplification of this assaywas achieved (PAPPASet al., 1983, 1985; JAHN & DIESFELD, 1983; MOHAMMED et al., 1985). A simple and economical direct agglutination test (DAT) for the diaanosis of visceral leishmaniasis. iecentiy developed & our laboratory (HARITH et al.; 1986), is compared here with IFAT and ELISA. A?Uigt??l
Materials and Methods
Leishmaniahani strain I-S (kindly providedby Dr J. Schottelius, Hamburg) was maintained in wizrousing NNN
Correspondence: A. E. Ha&h, Swellengrebel Laboratory of Tropical Hygiene, Royal Tropical InSdNte, Meibergdreef 39, 1105 AZ Amsterdam Z-O, The Netherlands.
field work on
medim?. Procedures for mass product@ of promastigotes :;8mgen preparauon have been described (HARITHet al., Test fwocedures Diiect agglutination test (DAT). Antigen preparation has been described @bRtTH et al., 1986). In brief, centrifuged promastigotes were washed in cold Locke’s solution, subjected to trypsin treatment, washed, and fixed in 2% formaldehyde in Locke’s solution at 4°C for 20 h. Promastigotes we& stained with Coomassie brilliant blue and the concentration adjusted to 1 x 10sparasites/ml in 1% formalin in physiological saline. After filtration through nylon gauze of 30 w mesh size, the antigen was stable in the dark at 4°C for 75 d at least. Recent studies showedthat antigen preparations wrapped in album foil were stable for one month at 521°C or 2 weeks at ambient temperatures of 18-37’C (Baring0 District? Kenya; unpublished results). The test was performed m V-shaped micro&e-plate wells with a starting serum dilution of 1 : 100 in 1% of fcetal
bovineserumin physiologicalsaline.After overnightincubation at 21”C, tbe titre in the test serum was read as one dilution beforethat yielding a clear, sharpblue spot similar to the one in the negative control well. Individuals with titres
greater than 1 : 1600 were considered DAT seropositive (HARITH et al., 1986). Immuna~?wescenceantiboa’y te.vt(IFAT). Washed promastigotes were dispensed on small circles left uncovered on glass slides sprayedwith Teflon (KOLK et al., 1984). After drying and fixation in acetone, a previously determined serum dilution of 1 : 100 was used throughout. As conjugate, sheepantihuman IgG(H + L) labelled with FITC (Pasteur Institute, Paris) was used. Fluorescence intensity was arbitrarily assessedas + , 2+, 3+ or 4+. Since a considerable number of individuals without visceral leishmaniasis showed + Buorescence.22+ was considered IFAT seropositive. Enzyme-linked immunosorbentassay(ELBA). Supematant of a son&ted promastigote suspension in 0.5 ELMsodium carbonate (pH 9.7), equivalent to 2 x 10’ parasites/ml, was centrifugedat 10000 x g and used for coating polystyrene
604
SERODIAGNOSIS
AND
SERO-EPIDEMIOLOGY
microtitre ELISA plates. Preliminary results showed that 1 : 500 serum dilution was optimal in discriminating between visceral leishmaniasis p&ients and individuals with other conditions. All sera were thus screenedat this dilution in triplicate. The diluent used was phosphate-buffered saline (PBS) at pH 7.2, containing 05% bovine serum albumen and 0.05% Tween 20. Goat antihuman IgG (H + L) labelled with peroxidase (Pasteur Institute, Paris, France), diluted 1 : 1000,was used as conjugate and the assay was read at 492 nm. Values greater than 0.28 (the mean of 10 African and 4 Dutch healthy control sera plus 3 times the standard deviation) were considered ELISA seropositive. A considerable number of persons without visceral leishmaniasis showed 30.28 absorbance. Taking absorbance = 057 as cut-off point resulted in a higher specificity but lower sensitivity, compared to O-28.
LEISHMANIASIS
(0 28 healthy controls including 10 Ghanaians, 10 Brazilians and 8 Dutch. (g) 273 individuals from the Perkerra area, Baring0 District in Kenya, a known focus of visceral leishmaniasis (LEELJWENBURG er al., 1982). Sera were collected as part of a continuing epidemiological study in March-April 1984. Included in this group were 3 cured visceral leishmaniasis patients, 10 of their household members, and the inhabitants of the wide surroundings of their homesteads. Amastigotes were later demonstrated in splenic aspirates of 2 persons admitted to the district hospital with suspected visceral leishmaniasis. All persons were examined for splenomegaly; thin and thick blood films were examined for malaria parasites from all persons with splenomegaly and 232 sera were examined for Brucelkz antibodies (from the remaining 41 no serum was available for this test). Results
Serum samples
In Table 1 are presented the serological results obtained with DAT, IFAT and ELISA. All recent and treated visceral leishmaniasis patients, except for one in ELISA, had values indicative of the disease in all three tests. Sera from recent and treated casesof
Serawere collected from the following groups of individuals. (a) 27 patients with proven visceral leisbmaniasis (Kenya) including 17 collected before, during or up to one month after treatment, and 10 taken 4-14 months after treatment when the patients were apparently cured. (b) 25 patients with African trypanosomiasis (Zaire), provided by Dr E. Magnus (Antwerp, Belgium). (c) 20 patients with Chagas disease, provided by Dr M. E. Camargo (So Paulo, Brazil). (d) 10 patients with schistosomiasis mansoni (Kenya). (e) 9 patients with giardiasis and 8 with toxoplasmosis (all from The Netherlands).
Table l-Results conditions
Group*
OF VISCERAL
of the newly developed DAT,
IFAT,
but
less so in DAT
and ELISA.
The
sensitivities of 3 tests in this po ulation were 100%for DAT and IFAT, and %-3% Por ELISA. The specificities were 72.6% (DAT), 94.3% (IFAT) and 79.4% (ELISA)
when trypanosomiasis
sera were included;
IFAT and ELISA on sera of patients with visceral leishmaniasis
DAT titre <1600 3200-25 600 a51 200
No.
visceral leishmaniasis showed comparable results in
IFAT intensity of fluorescence >2+ +
Eg;
:3
8
0
17
i
:
t%
El
:;:
i
;
13
12
and other
ELISA absorbance <0.28 0.28-O-56 0.57-2.00
:i 0
iI
1
is
1:
: z :1
i?
10 9
10 9
i
iI
1:9
i
8
;:
i
is:
2:
2;
ii0
i0
2:
:0
80
2;
i4
15
*VL(R): visceral leishmaniasis (recent); VL(T): visceral leishmaniasis (treated); AT: African trypanosomiasis; CD: Chagas disease; SM; scbistosomiasis mansoni; GI: giardiasis; TO: toxoplasmosis; HC: healthy controls.
Table 2-Results Kenya
of DAT,
IFAT
and ELISA
Diagnosis
No.
Brucellosis Malaria Splenomegaly** Apparently healthy Recent VL Treated VL*** Sensitivity Specificity Predictive Predictive
on 273 sera collected from leishmaniasis
% % value of positive test % value of negative test %
it
5
130 2 3
DAT al:3200
endemic area (Baring0 district)
ELISA* 30.28 >0.56
x
i
:
0 1 :
: :
2.6 ii.6
z.5 62.5 98.5
* Serum dilutions in IFAT and ELISA were 1:lOO and 1:500 respectively. ** Brucellosis, malaria and visceral leishmaniasis (VL) patients not included. *** 6 months (one patient) and 36 months (2 patients) after cure.
i
1 1
t: 2 1
::
fS.5 0.29 98.8
it 0.69 90
;
in
A.
by excluding ;bLg;d-100%
E.
HARITH
them, higher specificities were in DAT and IFAT and 87.3% in
In DAT none of the 45 trypanosomiasis sera cross-reacted at 3 1 : 51 200 dilution, while 6 in IFAT and 4 in ELISA did so at 32+ fluorescenceand 30.57 absorbance. African trypanosomiasis sera showed relativelv more cross-reactivitv in ELISA than in DAT and IFAT, while chagasic sera were more cross-reactive in IFAT than DAT and ELISA. In DAT, 8 African trypanosomiasis sera had titres up to 1:25 600 and all 20 American trypanosomiasis sera scored values inseparable from those of healthy controls (
The newly developed DAT was sensitive and specific even when African and American trypanosomiasis serawere included. All seraof chagasicpatients showed titres inseparable from those of healthy controls and only 8 out of 25 African trypanosomiasis sera had titres in the range seen in treated visceral leishmaniasis patients; none had titres as high asthose seen in recent visceral leishmaniasis infections (351 200). When applied to sera from an endemic area, the sensitivity and specificity remained high; a negative test had a very high predictive value and that of a positive test was only slightly reduced when treated visceral leishmaniasis patients (up to 36 months after cure) were included. Relatively fewer cross-reactions were found with the IFAT than with ELISA in seraof healthy controls and patients with conditions other than visceral leishmaniasis. Malaria sera, in particular, crossreacted significantly with i. do&vani antigen in IFAT (HOMMEL et al., 1978). In our results such a degree of cross-reactivity was not found in any of 36 malarious sera. In 10 children suswcted for L. d. infanturn infection, and with posit&e IFAT results, LATIF et al. (1979) were not able to demonstrate Lei.hania narasites. Our resulk in ELISA agreewith those of HOMMEL et al. (1978). who detected L. donavant’ antibodies in 35 ouibf 36visceral leishmaniasis patients, 20 being parasitologically confirmed. Higher ELISA specifkitv. however. was found in sera collected from the s&e area (Baringo, Kenya) using sonicated (HO et al.. 1983: JAHN 81 DIESFELD 1983) and intact L. dotkani biomastigote antigens (MOHAMMED et al., 1985). In a study including L. donovani and Trypanasoma cruai antigens tested against homologous and heterologous sera, GUIMARAES et al. (1981) reported discrepancies in results with ELISA and not with IFAT. Our results with African trypanosomiasis sera with the 3 tests agree with those of VOLLER et al. (1975),
et
d.
605
who reported
“strong cross-reactions” in T. and T. crwi infections. Such cross-reactions were not observed by HOMMEL et al. (1978) or Ho et al. (1983). It remains inexnkable why such discrenancies exist in results obtained by various investigators with IFAT and ELISA. Thoueh HOMMEL et al. (1978) doubted that the reason was the choice of conjugate; in our experience variations in ELISA readings using the same antigen and serum samples were due to differences in batches of plates, conjugate, or both. Standardixation of these factors was tedious and time consuming. Because of its high sensitivity and speciflcity, &d the ease and low-cost of its application cornoared to IFAT and ELISA. the newly develowd DAT is recommended for mass screening programmes and sero-epidemiological studies of visceral leishmaniasis. rhodesiense, L. abnovani
Acknowledgements We thank Dr.E. Magnus and Prof. D. le Ray (Institute of Tropical Medicine, Antwerp) for supplying African trypanosomiasis sera and laboratory facilities; Dr M. E. Camargo (Instituto Medicina Tropical de Sio Paulo), for American trypanosomiasis sera; Dr J. Schottelius (Hamburg) for L. donmuni I-S strain. The technical assistance of Mrs S. Kuijper and Mr H. Gilis, and the secretarial assistanceof Mrs A. Gilaard and Mrs A. Heibloem, are highly appreciated. We thank Prof. M. Mugambi, Dr D. K. Koech and Dr S. N. Kinoti (Kenya Medical Research Institute, Nairobi) for their support of the leishmaniasis project m Baring0 District and for providing staff and facilities. This investigation received 5nancial support from the UNDP/World Bank/WHO Swcial Programme for Research and Training in Tropical Diseases. References Behforouz, N., Remi, H. R. & Gettner, S. (1976). Application of immunofluorescence to detection of antibody in Leishmania infections. Annals of Tropical Medicine and Parasiwlagy, 70, 293-301. Bray, R. S. & Lainson, R. (1965). The immunology and serology of leishmankis. 1. The fluorescent antibody staining technique. Transactions of the Royal So&@ of Tropical Medicine and Hygbne, 59, 535-544.
Duxbury, R. E. & Sadun, E. H. (1964). Fluorescent antibody test for the diagnosis of visceral leishtmmiasis. f2E &umal of Tropical Medicine and Hygme, 13,
Edrissian, Ch. H. & Darabisn, P. (1979). A comparison of enzyme-linked immunosorbent assay and fluorescent antibody tests in the serodiagnosis of cutaneous and visceral leishmaniasis in Iran. Transactions of the Royal Socie~ of Tropical Medicine and Hygiene, 73, 289-292. Guimaraes, M. C. S., Celeste, B. J., De Castilho, E. A., Mince, J. R. & Din&J. M. P. (1981). Immunoenzymatic assay (ELISA) in mucocutaneous leishmania& kala-azar and Chagas’ disease.An epimastigote Trypam somu cruzi antigen able to distinguish between antiTtypammma and anti-Lkskmania antibodies. American Jounual of Tropkal Medicine and Hygiene, 30, 942-947. Harith, A. E., Kolk, A. H. J., Kager, P. A., Leeuwenburg, J:, Muigai, R., Kiugu, S. & Laarman, J. J. (1986). A simple and economical direct agglutination test for serodiagnosisand sero-epidemiological studies of visceral leishmaniasis. Transactions of the Royal Sake@ of Tropical Medicine and Hygiene, 80, 583-587.
Ho, M., Leeuwenburg, J., Mbugua? A., Wamachi, A. & Voller, A. (1983). An enzyme-lmked immunosorbent assay (ELISA) for field diagnosis of visceral leishmaniasis. American Journal of Tropical Medicine and Hygiene, 32, 943-946.
SERODIAGNOSIS
606
AND
SERO-EPIDEMIOLOGY
Hommel, M., Peters, W., Ranque, J., Quilici, M. & Lanotte, G. (1978). The micro-ELISA technique in the serodiagnosisof visceral le.&man&is. Annals of Tropical Medicine and Parasitolo~, 72, 213-218. Jahn, A. & Diesfeld, H. J. (1983). Evaluation of visually read ELISA for semdiagnosis and sero-epidemiological studies of kala-azar in the Baring0 District, Kenya. Transactionsof the Royal Society of Tropical Medicine and Hygiene, 77, 451-454. Kolk, A. H. J., Ho, M. L., Klatser, P. R., Eggelte, T. A., Kuijper, S., De Jonge, S. & Van Leeuwen, J. (1984). Production and characterization of monoclonal antibodies to MycobEtetium tubercu@is,M. bovis (BCG) and Ei-gae. Climcal and Exp~mental Immunology, 58, Latif, B. k. A., Al-Shenawi, F. A. & Al-Alousi, T. I. (1979). The mdire-ctfluorescent antibody test for diagnosis of kala-azar infection in Iraq. Annals of Tropical Medicine and Parasitology, 73, 31-35. Leeuwenburg, J., Bryceson, A. D. M., Mbugua, G. G. & Siongok, T. K. (1982). The use of the leishmanin skin test to define transmission of leishmaniasis in Baring0 District, Kenya. East Ajiican MedicalJoumal, 60,81-84. Mohammed, E. A. E. R., Wright, E. P., Kager, P. A., Laarman, J. J. & Pondman, K. W. (1985). ELISA using intact promastigotes for immunodiagnosis of kala-azar. Transactionsof the Royal So&y of Tropical Medicine and Hygiene, 79, 344-350.
TRANSACTIONS OF THH
1Short
ROYALSOCIETY OF TROPICAL.
LEISHMANIASIS
Pappas, M. G., Hajkowski, R. & Hockmeyer, W. T. (1983). Dot enzyme-linked immunosorbent assay (dot-ELISA): a micro technique for the rapid diagnosis of visceral Iee;y. &wrnal of Immunologwal Methods, 64, Pappas, M. G., Hajkowski, R., Diggs, C. L. & Hockmeyer, W. T. (1985). Disposable nitrocellulose filtration plates simplify the dot-ELISA for serodiagnosis of visceral leishmaniasis. Transactionsof theRoyal Societyof Tropical Medicine and Hygiene, 79, 136. Rassam, M. B. & Al-Mudhaffar, S. A. (1980). The micro-ELISA sandwich technique for the quantitation of Leishmania donovani soluble antigen. Annals of Tropical Medicine and Parasitolo~, 74, 591-595. Shaw, J. J. & Voller, A. (1964). The detection of circulating antibodies to kala-azar by means of immunofluorescent techniques. Transactionsof the Royal Sociery of Tropical Medicine and Hygiene, 58, 349-352. Srivastava, L., Suri, J. C. & Sanyal, R. K. (1979). ELISA test in diagnosis of kala-azar in current epidemic areasin Bihar. 3ournal of CommunicableDiseases, 11, 183-187. Voller, A., Bidwell, D. 81Bartlett, A. (1975). A serological study on human Trypanosomarhodesiense infections using a micro-scale enzyme linked immunosorbent assay. Tropenmedizin und Parasitologic, 26, 247-251. Accepted
for publication I8 November 1986
MEDICINE AND HYG~WE (1987) 81, 606
Report1
Spontaneous healing of cutaneous Leiskwuznia braziliensis braziliensis ulcers J. M. L. COSTA, E. M. NETTO, K. C. VALE, N. K. OSAKI, M. S. TADA AND P. D. MARSDEN Nticleo de Medicina Tropical e NutriGao, Universida& de Brasilia, Brazil
We have previously reported spontaneous healing of Leiskmania braziliensis braziliensis (Lbbj skin ulcers over a variable time scale (MARSDEN‘et al:, 1984). No one has much experience of untreated L. brazihensis type infections as patients normally accept treatment advice. A recent epidemic in our field area of endemic Lbb transmission enabled us to extend our observations to a larger number of untreated patients. Due to the great nu;nber of patients requirini help diagnosis mainly relied on the leishmanin skin test and the indir&t immunofluorescent reaction coupled with the suggestive clinical picture and histopathology on biopsy. Although parasitological diagnosis is desirable such diaPnostic Drocedures are adeauate for Lbb (CUBA et-&., 1984). 8 cases(40%) heaed in under 6 months and 17 (85%) in under 12 months. Another 3 healed after 14, 99’and 128 months. One patient relapsed. These observations agree with our previous report of a variable evolution of these l&ions witliout treatment UV~RSDEN et al., 1984). Thev also ap;ree closely wit& the retrospectiv;
OF VISCERAL
survey by Dk BELD~R et
al. (1978) of 48 patients whose ulcers had healed without injection therapy. As spontaneousresolution of skin leishmaniasis can occur, it is necessary to compare the results of treatment with any candidate drug with results in a group receiving standard antimonial therapy (MARSDEN
et al., 1979).
Finally, -treated or untreated, patients with Lbb must be advised of the possibihty of future mucosal metastasis. They should be instructed that if they develop nasal inflammation with blockage, epistaxis, persistent secretion, hoarseness of the voice, or dysphagia they should seek medical advice. References Cuba, C. C., Llanos Cuentas, E. A., Barreto, A. C., Ma-es, A. V., Lago, E. L., Reed, S. G. & Marsden, P. D. (1984). Human mucocutaneous leishmaniasis in T&s Bracoos,Bahia-Brasil. An area of Leishmania braziliensis braziliensis. I. Laboratory diagnosis. Revista da So&dude deBrasileira deMedicina Tropical, 17,161-167. De Belder, M., Belby, M. J. & Akers, J. E. (1978). Report of the Cambridge Medical Ex dition to Brazil, 1977. Mimeographed document, 13F pp. Marsden, P. D., Cuba, C. C., Barreto, A. C., Sampaio, R. N. & Rocha, R. A. A. (1979). Nifurtimox in the treatment of South American leishmaniasis. Transactions of the Royal Society of Tropical Medicine and Hygiene, 73, 391-394. Marsden, P. D., Tada, M. S., Barreto, A. C. & Cuba, C. C. (1984). Spontaneous healing of Leishmania braziliensis braziliensisskin ulcers. Transactionsof theRoyal Societyof Tropical Medicine and Hygiene, 78, 561-562. Accepted
for publication 2 February 1987