./ Culan I'altwl 1993:Printed in Denmark - att righb

Co/itiigtil © MunLigaaitt 1993

Journal of

Cutaneous PathologyN tnm:u,9in

Post-kala-azar dermal leishmaniasis:A light and electron microscopic studyof 18 casesPost-kala-azar dermal leishmaniasis (PKDL) is caused by theprotozoan parasite, Leishmania donovani, and is seen in patientswith history of having been treated earlier for the visceral dis-ease form, kala-azar, caused by the same organism. The findingsfrom 18 patients with PKDL are described in this study. Theskin manifestations ranged from hypopigmented macules toinfiltrated plaques and nodules. Histopathologic examinationrevealed a cellular infiltrate of lymphocytes, plasma cells, andmacrophages. The macrophages were scattered amidst the infil-trate without any localization. In hypopigmented lesions, theinfiltrate was confined to the perivascular region in the super-ficial dermis and was composed mainly of lymphocytes and fewplasma cells. In the nodular lesions, the infiltrate occupied theentire thickness of the dermis. Leishman-Donovan bodies werescarce and identified in 16 cases after a prolonged search ofWeigert's iron hematoxylin-stained sections. In 2 cases, Leish-man-Donovan bodies were not demonstrable. Electronmicro-scopic study revealed parasitized macrophages which showed nostructural evidence of activation despite the active cellular re-sponse around them. The fine structure of the parasites in thehistiocytes was also well maintained. This unusual tropical der-matosis is a unique example of change in organotropism of aparasite associated with a change in the host response.

Mukherjee A, Ramesh V, Misra RS. Post-kala-azar dermalleishmaniasis: A light and electron microscopic study of 18 cases.J Cutan Pathol 1993: 20: 320-325. © Munksgaard 1993.

Ashok Mukherjee*,Venkatesh Ramesh^^ andRadhey Shyam Misra^'Institute of Pathology (ICMR) and•Department of Dermatology and Leprology,Safdarjang Hospital, New Delhi, India

Dr. Asfiok Mukherjee, Deputy Director,Institute of Pathology (ICMR), P.O. Box 4909,NewDelhi-110029, India

Accepted January 16, 1993

Of the unicellular flagellate organisms elassified un-der the genus Eeishmania that are human pathogens,the ones found in the Indian subcontinent are thosebelonging to the E. donovani, E. tropica, and E. majorcomplexes. The L. donovani complex species causeOld World visceral leishmaniasis (VL), also knownas 'kala-azar' (KA), a systemic infection endemic tocertain parts to Asia, Africa, and southern Europe.The species in the other complexes are responsiblefor anthroponotic cutaneous leishmaniasis (ACL), acutaneous infection leading to painless ulceration ofthe skin which usually heals spontaneously, oftenleaving disfiguring scars.

In India, KA is endemic in the eastern part of thecountry in the states of Bihar, West Bengal, and

Uttar Pradesh. The disease presents as a chronicdebilitating condition with extensive parasitizationof the reticulo-endothelial system. Post-kala-azardermal leishmaniasis (PKDL) is an unusual derma-tosis seen in treated VL patients. It occurs in about20% of cases at a variable interval after remission ofthe visceral form (1). The condition was first de-scribed by Brahmachari in 1922. As the name im-plies, the disease remains restricted to the skin.PKDL manifests in a variety of clinical forms fromhypopigmented macules to infiltrated plaques andnodules, and the lesions are quite often confusedwith those of leprosy. Treatment is based on thepentavalent antimony compounds, meglumine anti-moniate and sodium stibogluconate, as first-line

320

LeJshmaniasis

/v". /. Dermal irrfrhr'atc from anodule shovvirri; plasma cell.s,lymphocyies and macropliaB;es.No f J ) B are seen irr this frcid(I-mieron plastic section, toluidinebhie XlOOO).

drugs and amphotericin-B and pentamidine as thesecond-line treatment for resistant cases. Allopuri-nol and, more recently, ketoconazole have also beenused as second-line medicines for the disease (1).The literature on the clinical and histopathologicaspects of PKDL is quite sparse and, to the best ofour knowledge, there are no reports of the electronmicroscopic features of the dermal lesions. We pre-sent here the light and electron microscopic featuresobserved in 18 cases of PKDL.

Material and methods

The 18 patients included in the study were seen atthe Department of Dermatology, Safdarjang Hospi-tal, a large general hospital in New Delhi. All wereresidents of Bihar, Eastern Uttar Pradesh, or WestBengal who had migrated to New Delhi recently.They ranged from 10 to 50 years of age, and therewere 1 7 men and 1 woman in the group. Previoushistory of KA was available in 14 patients, and theinterval between remission of KA and onset ofPKDL varied from 2 to 4 yr in most, with intervalsof 1 yr and 32 yr being seen in individual patients.In 4 patients, there was no previous history of KA.None of the patients had received any treatment forPKDL prior to being seen at the center.

Skin biopsies taken from the lesions were fixed ina mixture of 4% formaldehyde and 1% glutaralde-hyde in phosphate buffer (pH 7.2) and processed forboth light and electron microscopy using standardprotocols. Sections for light microscopy were cut at5 microns and stained with H&E, Fite-Faraco,Giemsa, and Weigert iron hematoxylin stains. Forelectron microscopy, ultrathin sections were con-trasted with lead and uranium salts and examined

under a Jeol Temscan 100 CX II electron micro-scope.

ResultsClinical features

Three types of lesion, i.e., papulo-nodules, infil-trated plaques and hypopigmented macules wereseen. Most patients presented with two or all threetypes of lesions simultaneously. There were 2 pa-tients with hypopigmented lesions only. The centralpart of the face was invariably affected, with lesionsalso being seen on the trunk and extremities inpatients with generalized disease. Involvement ofmucous membranes was seen in 5 patients, with thelips and glans penis being the sites most commonlyaffected. Detailed clitiical descriptions and responseto therapy Iiave been recently reported in a separatearticle (3).

Histopathological features

All biopsies showed a dermal inflammatory infil-trate comprised of lymphocytes, plasma cells, andmacrophages (Fig. 1). The amount of infiltrateranged from focal perivascular collections localizedto the superficial dermis in the hypopigmented le-sions, to diffuse infiltration of the entire dermis inthe papulo-nodular lesions. Lymphocytes andplasma cells were the predominant cell types seen.Macrophages were found scattered amidst the infil-trate without any evidence of clustering or granu-loma formation. Epithelioid cells, giant cells, eosi-nophils, and basopliils were conspicuous by their

21 Culaiiroiis P.Ttllology 321

Mukherjee et al.

Fig. 2. Phototnicrograpit ol asection of another case showittg asparse ittftlttatc with twonuictopliages containing l.DB(arrows) (1-micion plastic section,toluidine l)hu- X lOOO).

^^^'WmKl^inmmSk^'.

absence. Neutrophils formed a significant compo-nent ofthe infiltrate in 3 cases but were scarce in theotheis.

The epidermis was normal in biopsies with focalitifiltrates and atrophic in most of the cases withdillusc infiltration. Two cases showed, in addition toa dillltse innaminatory infiltrate occupying morethan 90% of the dermis, epidermal hyperplasia,follicular plugging, and hyperkeratosis. All caseswith dillusc itifiltrate also showed edema, vasculardilatation, and local hemorrhages in the sub-epider-mal zone, with the inflammatory infiltrate reachingand encroaching upon the basal layer.

Leishman-Donovati bodies were found in 16 biop-sies. These were located within the macrophagecyloplastn and were always multiple in each cell(Fig. 2). Parasitized macrophages were scarce and

only found after prolonged search. They were foundmainly in the inllammatory cell collections in thesuperficial dermis. The organisms were visualizedbetter in sections stained with Weigert iron hema-toxylin than in those stained with H&E or Giemsastains.

Eiectronmicroscopic examination

Electronmicroscopy confirmed the presence of a richcellular infiltrate around the parasitized macro-phages. Plastna cells with abundant rough endo-plasmic reticulum and nutnerous Golgi bodies wereinterspersed with lymphocytes showing hypercoti-voluted nuclei and sparse organelles (Fig. 3). Themacrophages did not exhibit any structural evi-

Fig. 3. Elcctrontnicrograplt of aportion of the dermal ittfiltratc. Alytnphocytc with a cotivolutcdnucleus (L), pUistiia cells (PL)and a mactopliagc with smoothcontouts (Mo) arc si-oti, indicatinglack of any evidence of activationofthe macrophage in s])itc of thepresence of tiumerousimmunocompctctit cells (X4800).

322

Fig. 4. liU'Ctronmicrograpli of aportioti of the dcrttial infiltratefrom another case. A lymphocyte(L) flattkcd by two tnacrophagesis seen. A LDB (arrow) is seettlyitig iti tlic cytoplasm of attothcrmactopliagc. The parasitized celldocs not show any ultrastrticttiralevidence ofactivatioit (X4800).

dence of activation, i.e., the cells had smooth con-tours and the cytoplasm contained the usual com-plement of organelles, with only a few lysosomesbeing seen (Figs 3 & 4). Macrophages parasitized byL. donovani did not demonstrate any significantchange in structure (Fig. 4). The LDB were en-closed by a trilaminar structure whose outer layerwas fortned by the limiting tnembrane ofthe macro-phage vacuole; the two inner layers were of parasitieorigin (Fig. 5). LDB contained nuclei with largenucleoli, flagellar bodies, vacuoles, and multicellu-lar bodies. No evidence of phagolysosomal fusionwas seen. The dermal blood vessels had large hyper-trophied endothelial cells. No LDB were found invascular endothelium or in nerve twigs.

The diagtiosis of PKDL was made in 16 casesafter detection of LDB in the biopsies, whereas inthe other 2 cases diagnosis was made on clinicalgrounds. All 18 patients teceived sodium antimonygluconate in a dosage of 20 mg/kg/day by intra-

Fig. 5. Electronmicrograph ftomthe same lesion as in the previousillusttatioti showing two LDB ittthe cytoplastn of a macrophage.The ttticleus (N), vacuole (V),lipid hody (L), kinetoplast (K)and trilaminar menihraiu; (arrow)are clearly seen (x 14000).

muscular injection. The maximum dosage, how-ever, did not exceed 1 g/d. A good response tochemotherapy was seen, and the skin lesions sub-sided within a period of 3..5 to 5 tnonths (3).

Discussion

PKDL is a unique exatnple of a disease with exten-sive visceral involvetnetit that is followed by a lo-calized dertnal form after treafment. This change inorganottopism has been the subject of earlierstudies. Initially, workers were ofthe view that KAand PKDL were caused by two diiferent strains ofthe parasite. Studies on the biochetnical, anfigenic,and cultutal characteristics ofthe organisms grownftotn the forms ofthe disease could not identify anydifferences (4, 5). The fine structute ofthe LDB inPKDL skin lesions, as shown in this and a previousstudy (6), is identical with the fine structure de-

21* 323

Mukherjee et al.

scribed for the LDB in bone marrow and splenicaspirates in KA (7). In addition, there are rarereports ofthe recurrence of KA along with PKDL inpatients who had been treated for the initial attackof KA (8). It thus seems clear that both diseaseforms are caused by the same organism.

Another explanation put forward is that there is achange in the organotropism of the parasite alongwith a change in the cell-mediated immunity (CMl)status ofthe patient. Haldar et al. (9) have studiedthe CMl status in KA and PKDL and concludethat, while there is a lack of both in vivo and in vitroCMl responses to leishmanial antigens in untreatedKA, PKDL patients show a variable picture, withabout 60% showing positive skin tests to leishma-nin. Neogy et al. (10) report a consistent lack ofleishmanin positivity in PKDL patients even aftertreatment. On the other hand, the histological reac-don to the parasite in the skin in PKDL is a pro-liferative cellular response, with plasma cells,lymphocytes, and macrophages. The electron mi-croscopic findings of numerous lymphocytes andplasma cells in close association with macrophagesis further evidence ofthe active CMl response ofthehost to the parasite. This contrasts with the pictureseen in KA, where a massive reticuloendothelialhyperplasia, with a distinct lack of lymphocytic in-filtration, is seen in the spleen, liver, and bone mar-row along with atropy ofthe paracortical zone in thelymph nodes (11). Thus there appears to be a dis-tinct upgrading of the cellular immune response tothe parasite in PKDL.

PKDL generally occurs at a variable time periodfollowing apparent cure of KA. There are occasionalreports, however, of patients having both dermaland visceral lesions, simultaneously (5, 8). Earlyhistologic studies also record the fact that, in someinstances in the apparently normal skin of KA, LDBare detectable within macrophages around dermalappendages (11, 12). It is possible that the skin isone of the sites of localization for the parasite fromthe beginning ofthe disease process. Activated cell-mediated immune response against the parasitesremaining after therapy could lead to enhanced cel-lular infiltration into the areas of the dermis whereorganisms are present, thus giving rise to the lesionsof PKDL. Another possibility is that PKDL arisesafter inoculation of fresh organisms from infectedsandflies into the skin of a patient who has acquiredsome itTtmunity by virtue of a past history of treatedKA. This is consistent with epidemiologic observa-tions indicating reduction ofthe incidence of PKDLin the community after achievement of good vectorcontrol (13).

Whereas the exact pathogenic mechanism re-mains to be elucidated, it is obvious that, in spite ofthe host being able to mount an immune response

against the parasite, the organism is still able toevade the parasiticidal mechanisms of the host.Data from experimental animal studies and in vitromacrophage cultures (1) indicate that the leishtiia-nia in the macrophages are killed by production ofoxygen metabolites induced by lymphokines se-creted from activated lymphocytes. The persistenceof intact and complete organisms in the macrophagecytoplasm, and the absence of lysosomes and phago-lysosomal fusion despite the presence of numerousimmunocompetent cells as observed in fhe presentstudy, indicate a defect in the effector limb of theimmune system, either in lymphokine production orin macrophage response. Thus, while the imtnuneresponse is able to limit the dissemination of theparasite to the skin, it is unable to check growtliand multiplication of the organisms. This leads tochronicity and extension of the lesions within theskin.

The histological features observed in the 18 casesin this study did show a variation in the amount ofinOammatory cell infiltrate which correlated withthe clinical type of lesion, but no significant differen-ces were seen in the composition of the infiltrate.Earlier reports on the histology of PKDL (14—19)describe som lesions with macrophage granulomasand heavy parasitization and others with well-formed epithelioid cell granulomas, indicating aspectrum in PKDL similar to that described forACL by Kubba (20). None of the biopsies in thepresent series had either of these features. It is prob-able that the predominant cellular pattern seen inPKDL is that of a lymphocyte, plasma cell, andmacrophage infiltrate with sparse LDB with theother patterns being uncommon variants whichcould develop in advanced nodular lesions (21).The variations observed in the in vivo and in vitroCMl response and in the histological picture couldprobably be better understood by studies whereboth parameters are correlated along with theclinical picture.

The clinical similarity between the skin lesions ofPKDL and leprosy (22, 23) often leads to a diag-nostic problem, especially in areas with a high in-cidence ofthe latter disease. The lack of any sensoryloss in the lesions is a very i^mportant clinical signwhich can differentiate the two conditions. The hist-ological picture can also cause confusion and a Fite-Faraco stain to detect acid-fast organisms should bedone as a matter of routine in all biopsies receivedwith a clinical diagnosis of PKDL. The finding ofLDB within the fine dermal nerve twigs as reportedin a recent study from the Sudan (17) was not seenin any ofthe biopsies in this study. The iron-hema-toxylin stain was found to be more useful in detect-ing LDB in tissue sections as the organisms werebetter contrasted and thus easier to pick up. Earlier

324

Leishmaniasis

reports on the use of this staining method (15, 24)have also fbund it to be a useful technique.

From the epidemiologic point of view, PKDLrepresents the ability of the L. donovani organism toadapt to the altered immune mechanisms of the hostand thus perpetuate itself in the absence of a zoo-notic reservoir. Further study of this unusual der-matosis could help to provide a better understand-ing of the host-parasite relationships in humanleishmaniasis.

References1. Control of the LcishrTraniasis. Report of a WHO Expert

Committee. Tech Rep Ser No. 793, 1990.2. Brahmachari UN. A new form of Cutaneous Leishmaniasis

- Dermal Leishmanoid. Ind Med Gaz 1922: 37: 125.3. Ramesh V, Misra RS, Sa.xcna U, Mukherjee A. Post-kala-

azar dermal leishmaniasis. A clinical and therapeutic study.Int J Dermatol (in press).

4. Bray RS, Ashford RW, Mukherjee AM, Sen Cupla PC.Studies on Immunology and Serology of Leishmaniasis. Ser-ological investigation of the parasites of Indian kala-azarand Indian post-kala-azar dermal leishmaniasis. Trans RoySoc 'Irop Med Hyg 1973: 67: 125.

5. Das Gupta BM. A note on the parasite of dermal leishma-noid. Ind Med Gaz 1927: 62: 11.

6. Sanyal AB, Sen Gupta PC. Fine structure of Leishmania indermal leishmanoid. Trans Roy Soc Trop Med Hyg 1967:61: 211.

7. Wickramasinghc SN, Ahdalla SH, Kasil EG. Ultraslruclurcof bone marrow in patients with visceral leishnianiasis. JClin Pathol 1987: 40: 267.

8. Sen Gupta PC, Mukherjee AM. Recurrence of kala-azarassociated wilh pcjsl-kala-azar dermal leishmaniasis. J IndMc;d Assoc 1968: 50: 1.

9. Haldar JP, Ghosc S, Saha KC, Ghosc AC. Cell mediatedimmune response in Indian kala-azar and post-kala-azardermal leishmaniasis. Infccl Iirrmun 1983: 42: 702.

10. Ncogy AB, Nandy A, Chovvdhury AB. Lcishmanin test inpost-kala-azar dermal lcishmarriasis. Trans Roy Soc lVopMed Hyg 1990: 84: 58.

11. Chatlcrjce KD. Leishmania donovani. In: Human Parasitesand Par'asilic Diseases (published by the author). Calcutta,India, 1952: 180.

12. Napier LE. Leishmaniasis. In: The Principles and Pracliccof Tropical Medicine. New York: Macmillan, 1946: 149.

13. Sanyal RK, Alam SN, Kaul SM, Wallal BL. Sonic ohscr-va-lions on epidemiology of current outbreak of kala-azar inBihar. JComm Dis 1979: 11: 170.

14. Girgla HS, Marsden RA, Singh CM, Ryan TJ. Post-kaia-azar dermal leishmaniasis. BrJ Dermatol 1977: 97: 307.

15. Sen Gupta PC, Bhaliacharjcc B. Histopalhology of post-kala-azar dermal tcishmaniasis. J Trop Med Hyg 19.')3: 56:110.

16. Ycsudian P, Thanibiah AS. Amphotcricin B therapy in der-mal leishmanoid. Arch Dermatol 1974: 109: 720.

17. El Hasan AM, Ali MS, Zijlstra E, EITouni IA, Ghalib HW,Ahmed HMA. Posl-kala-azar dermal leishmaniasis in theSudan: peripheral neural involvcmcnl. IniJ Dcrnialol 1992:31: 400.

18. Al-Awkali N, Khairy N. Post-kala-azar dermal leishmanoidin an Iraqi child. J Fac Med Baghdad 1980: 22: 305.

19. Yawalkar SJ, Mardhckar BV, Mahabir BS. Post-kala-azardermal leishmaniasis. J Trop Med Hyg 1966: 69: 140.

20. Kubba R, Al-Gindan Y. Lcishmarriasis. Dermalol Clin1989: 7: 331.

21. Ridley DS. Pathology. In: Pctc-rs W, KiUick-Kciidrick R,eds. The Lcishmaniases in Biology and Medicine. Vol. 2.London: Academic Prc-ss, 1987: pp. 665-701.

22. Ramesh V, Saxcna U, Misra RS, Mukherjee A. Posl-kala-azar der'mal leishmaniasis: a case report strikingly resem-bling lepror'natous leprosy. Lepr Rev 1991: 62: 217.

23. Ekar'r'rbaram V, Naidu B, Rao VP. Dilfcrcntial diagnosisbetween leprosy arrd posl-kala-azar dermal leishmaniasis.Ind J Lcpr 1984: 56: 641.

24. Sen Gupta PC, Basu Mallick KC, Bhatlacharya B. A rapidmethod iron hacrnatoxylin stain for pr'otozoa in tissue sec-tions and si'nc-ars. Ind Med Gaz 1950: 85: 400.

325