The histopathological features of pale grain eumycetoma

839 TRANSACTIONS OF THE ROYALSOCIETY OF T~omc~r MEDICINEAND HYGIENE,VOL. 76, No. 6, 1982

The histopathological

features

of pale grain eumycetoma

R. J. HAY AND D. W. R. MACKENZIE London School of Hygiene and Tropical Medicine, Keppel Street, London WCIE

Summary

Histological findings in 24 cases of pale grain eumycetoma have been compared. The inflammatory response and morphology of grains caused by different organisms are very similar. However the histological appearancesof Neotestudina rosatii infections, although reminiscent of dermatophyte mycetomas, are distinctive. The characteristic features include the presence of cement and the shorr swollen hyphal segments. Petriellidium boydii grains are more likely to contain numerous vesicles or swollen hyphae and be surrounded by a prominent eosinophilic fringe. In contrast, Fusarium and Acremonium grains usually have a minimal fringe and contain a dense mass of intermeshing hyphae. However the features separating P. boydii and FusariumlAcremonium grains are not invariable and can only be used as an approximate guide to the correct identity of the organism. Introduction

Mycetoma (Madura foot) is a chronic infective processof subcutaneous tissue in which the causative organisms are present in the form of aggregates (grains) of actinomycete or fungal filaments. The infection either follows injury to skin and implantation of the organisms which exist as saprophytes in soil or through the prick of parasitised thorns (BASSET et al., 1965). Direct spreadof the mycetoma to involve overlying skin and subcutaneous tissue or underlying bone is usual but metastasesto remote sites are rare. The causative organisms are shown in Table I. Mycetoma may be caused by actinomycetes (actiTable

I-Classification

7HT

nomycetoma) or true fungi (eumycetoma). Actinomicetomamust be distinguished from actinomycosis and botryomycosis. The source of infection in the former is endogenous and invasion normally occurs in a locality where saprophytic carriage of Actinomyces spp. may occur. Rarely primary cutaneous actinomycosis by inoculation has been described (MAHGOUB & YACOUB, 1968). In botryomycosis the causative organisms are bacteria such as Staphylococcus aureus and the infection may localize in internal organs such as lung, as well as in subcutaneous tissue (WINSLOW 1959). The identification of mycetoma organisms depends on their isolation in culture? the determination of the presenceof specific antities by serological methods and histopathology (DESTOMBES, 1964). In darkly pigmented true fungal mycetoma (the dark grain eumycetomas) or the actinomycetomas, the histopathological appearancesof grains in tissue sections are characteristic enough to permit specific diagnosis (MARLAT et al., 1977). ,For instance, Streptomyces somaliensis, an actinomycete, produces large, regular grains which stain homogeneously with eosin and show transverse “fracture lines”. .In Madurella mycetomatis mycetoma there is dense permeation of the grain by a dark brown, cement-like substance. However in the non-pigmented (pale grain) eumycetoma, the morphology of grains is not so distinctive and indeed most have a similar appearance. Histopathological differentiation between different organisms is usually not possible. This study is an attempt to de6ne the histopathological features of pale grain eumycetoma.

of mycetoma

Eumycetoma (infections caused by true fungi)

Dark grain eumycetoma:

Pale grain eumycetoma:

Madurella mycetomatis M. grisea Exophiala jeanselmei Pyretwchaeta romeroi Leptosphaeria senegalensis Cuwulati lunata Petriellidium boydii Acnmwnium spp. (A. kiliense, A. recifer] Fusarium spp. (F. solani, F. oxysporum) Neotestudina rosatii Aspergillus nidulans, A. jlavus Dermatophytes, e.g., Microsporum ferrugineum, M. audouinii

Actinomycetoma (infections caused by actinomycetes) Pale grain actinomycetoma: Actinomadura madurae StreptDmyces somaliensis Red grain actinomycetoma: A. pelletieri Small (microscopic) actinomycetoma: Nocardia spp. (N. bra&en&,

N. asteroides, N. cawiae)

840 Table

HISTOPATHOLOGICAL

II-Histological

features

Diameter (mean p)* Eosinophilic fringe Shape Central invasion by neutrophils Vesicles present Nucleoli visible (HE)

FEATURES

of non-pigmented

OF PALE

eumycetoma

GRAIN

MYCETOMA

grains

P. boy&i

Acremonium spp. Fusarium spp.

N. rosatii

480 x 350 (80-940) 5-15 rnp (7) Variable

490 x 366 (130-890) 5-10 rnp (2) Variable

Variable

380 x 320 (30-720) Variable

E?Y’ No

Infrequent Sometimes Uncomm. (1)

Infrequent Rare Yes

Infrequent Yes Yes

560 x 483 (100 x 1100)

*Mean diameter is expressedas the mean of the maximal diameters of each grain measuredat right angles in mp. The range is the minimum and maximum diameters measured. All measurements were made using tissue sections.

Methods

The tissue sections had been collected at the Mycological Reference Laboratory, London School of Hygiene and Tropical Medicine between 1962 and 1981.All sections were stained with haematoxylin and eosin (HE) and either periodic acid Schiff (PAS) or methenamine silver (GMS). 24 cases of pale grain eumycetoma were reviewed. The causative organisms had been identified by culture in 20 casesas follows: Petillidium boydii (nine), Acremonium spp. (two), A. kiliense (two), Neotestudina rosatii (three), Fusarium spp. (two), F. solani (one) and F. oxysporum (one). In four casesthe identity of the causative organism was not known. Individual slides were coded and examined after removal of identifying labels. The following criteria were used to assessthe histopathological appearances and grain morphology: composition of the in&mmatory infiltrate, morphological features of grain structure such as size, pigmentation, presence of cement, distribution of hyphae and invasion by inflammatory cells. Grain diameters and surface areain each section were calculated by planimetry from scale drawings of grains made using a microscope drawing attachment (Carl Zeiss,.Germany).Maximum diameters measured at right angles were calculated for each grain. The mean surface area for grains causedby each organism were compared to the mean area of P. boy&i grains taken as 1. Results InfIammatoy

infiltrate

Grains were usually identified in the centre of a dense inflammatory infiltrate whose composition varied little between individual cases. The most common cells were neutrophils, lymphocytes and foamy macrophages. Occasional foreign body giant cells were seen, although these were strikingly more numerous in the N. rosatii infections. In four cases (one P. boydii, one Acremonium sp., one Fusarium sp., dne unidentified organism) dense accumulations of eosinophils were noted (X 40). Grain morphology

The following aspectsof grain morphology in tissue were compared (Table II). Size and shape:In all forms of pale grain eumycetoma there was considerable variation in the shape of

grains, ranging from cuboid to cylindrical or crescentic. There was also great variation in size. For example, the diameter of grains in two directions ranged from 50 to 75 rnp to 700 to 1000 my. Their mean surface arearatios compared to P. boy&i were as follows: P. boydii 1.0, Acremonium 1.2, Fusarium 1.8, N. rosattii 0.6. The smallest grains were seen in N. rosatii infections. These ranged from 30-45 rnv in diameter although larger grains 300 rnp or more in diameter were common. Some of the smaller grains had been engulfed by giant cells. Eosinophilic fringe (Splendore-Hoeppli phenomenon): A refractile eosinophilic fringe surrounding the perimeter of grains was a feature of many of these mycetomas (P. boydii seven, Acremonium two, N. rosatii one) although it was not seen in the Fusarium cases.The size of the fringe varied but in somecasesit was over 10 rnp in width and very prominent (P. boydii five, Acremonium one). In these cases the refractile material was usually condensed into clublike structures around the periphery of the grain. Cement: Cement seen as homogeneously staining pink material lying between individual hyphae occurred in and was a characteristic feature of N. rosatii infections. It was not seen with other grains. Arrangement of hyphae: Hyphae 3-8 rnp in diameter were present at the periphery of all pale grains and could be distinguished in HE stained sections. In addition expanded hyphae or vesicles (up to 15 rnp in &ameter) were also present at this site. In most of the P. boydii cases (six) and one Acremonium case the centre of the grain was permeated with neutrophils amongst necrotic hyphae (Fig. 1). Vesicles were always most prominent and larger in P. boydii infections (Fig. 2) and were largest in grains where there was also a thick eosinophilic fringe. In two cases of P. boydii infection, where the latter was absent or minimal, large vesicles could be seen scattered throughout the grains (Fig. 3). The arrangement of the hyphae in the Fusarium (four) and most of the Acremonium (three) caseswas different. Grains were characteristically composed of interweaving skeins composed of 5 to 40 regular hyphae to give a dense swirling pattern of mycelium (Figs. 4 and 5). In such cases the cytoplasm and even nucleoli were often prominently stained in HE sections. Dilatation of hyphae was not a prominent feature. By contrast, in N. rosatii grains, short irregular hyphal segmentswere seen. There were usually oval or round, 5 to 10 rnp in

R. J. HAY

AND

D.

Fig. 1. Grain of P. bcydii showing central invasion by neutmphils.

Fig. 2. P. boydii grain. The vesicles are large and rno~e prominent

W.

841

R. MACKENZIE

x 37 HE

at the periphery

x 150 HE

842

HISTOPATHOLOGICAL

FEATURES

OF PALE

GRAIN

MYCETOMA

Fig. 3. p. boy&i grain. There is a less prominent

eosinophik

fringe but many vesicles are present. x 37 HE

Fig. 4. Grain of a Furariun!

arrangement

of hyphae x 37 HE

sp. Note the swirling

R. J. HAY

Fig. 5. Grain from a mycetoma caused by an AN-ium

AND

D.

W.

R. MACKENZIE

843

sp. x 187 HE

diameter and resembled chlamydospores. In PAS or GMS sections short chains of cells could also be distinguished. Pigmentation: Brown pigmentation of hyphae was not recognized in any of these cases. Discussion

This study confirms the widely held opinion that grains formed by the main causative organisms of pale grain eumycetoma (P. boy&i, Acremonium spp. and Fusarium sup.) are similar histopathologically. However our observations support-the view of CHANDLERet aZ. (1980) who suggested that vesicles were a urominent and characteristic feature of P. boy&i g&m. Even so, such criteria do not appear to be absolutely reliable as in one A. kiliense casevesicles were equally prominent. In this casea wide eosinophilit fringe with clubs was present, a feature not common in Acremonium or Fusarium mycetoma. In the latter infections the hyphae usually form a solid grain of intermeshing hyphal strands. In the two cases of P. boy&i infection where the eosinophilic fringe was not prominent vesicles were present throughout the grain. Therefore we believe that the following features-size of eosinophilic fringe, presence of vesicles or cement and arrangement of hyphae-are helpful in separating the different agents of pale grain eumycetoma. However there may be exceptions. For instance grain structures observed in a case of systemic P. boy&i infection of the paranasal sinuses (BRYAN et al., 1980) were reminiscent of typical Fusarium or Acremonium granules. Although Acnmumium species (A. kiliense and A. recifei) are well recognized as causes of mycetoma

et al., 1961; MURRAY & HOLT, 1964; et al., 1976), Fusarium species (F. solani, F. oxysporum)are less common causesof this condition @MUAT et al., 1977). The cultures of both these genera may be difficult to separateon morphological grounds and it is perhaps not surprising that their xrains are histoloaicallv similar. It is also possible that some confusion m theWidentificationof c&ures taken from such casesmay occur. By contrast, the grains of N. rosatii are auite distinct. The formation of cement and the presence of rounded hyphal fragments is characteristic (ROSATI et al., 1961). They closely resemble the grains seen in mycetoma caused by dermatophytes (AJELLOet al., 1981), although in the latter Splendore-Hoeppli-like material rather than cement surrounds the organism. A further rare cause of pale grain eumycetoma which we have not been able to study is Aspergilhs nidulans (MAHGGUB, 1971). The grains of the latter are oval or round and formed of large compact hyphae. Irregular or round spores occur at the periphery (MAHGOUB & MURRAY, 1973). There were four mycetomas in this study from which no organisms were isolated. In two of these culture was not attempted. The grain in each resembled the P. boy&i and the AcwmoniumlFusarium type respectively. It is possible that these organisms may have been responsible. In the two other cases,both of which originated from Jamaica, a dematiaceous or dark pigmented fungus was isolated repeatedly from surgical material but could not be identified through lack of distinguishing morphological features. A similar occurrence has been recorded in a patient from the .Caribbean island of Guadeloupe (ESCUDIEet al., 1967). It is possible that these are new mycetoma agents. (BAYLET HALDE

HISTOPATHOLOGICAL

844

FEATURES

The precise identification of all pale grain eumycetoma organisms in tissue remains an elusive goal although we have been able to describe some helpful features for recognition of individual organisms in this study. The application of immunological methods such as immunofluorescence (KAPLAN & KIWFT, 1969) or immunoperoxidase (SAEED & HAY, 1981) staining to this group of infections may well provide a more objective method of classifying grains although there are some technical difficulties such as auto-fluorescence which have to be overcome. Ackaowledeements

We would like to acknowlgdge the help of the large number of medical colleagueswho have referred histological material from mycetoma cases over the years. References

Ajello, L., Kaplan, W. & Chandler, F. W. (1980). Dermatophyte mycetomas: fact or fiction. Proceedings of the Fifth International Conference on Mycoses: PAHO Sci. Pub., 396, 135-140. Basset, A., Camain, R., Baylet, R. & Lambert, D. (1965). Role des epines de mimosac6es dam l’inoculation des mycetomas (a propos de deux observations). Bulletin de la Societe de Pathologic Exotique, 58, 22-24.

Baylet, R., &main, R., Bezes, H. & Rey, M. (1961). Mycetomas ?i Cephalosporium au Senegal. Bulletin de la Socike’ de Pathologic Exotique, 54, 802-810. Bryan, C. S., Di Salvo, A. F., Kaufman, L., Kaplan, W., Brill, A. H. & Abbott, D. C. (1980). Petriellidium boydii infections of the sphenoid sinus. American Journal of Clinical Pathology, 74, 846-85 1. Chandler, F. W., Kaplan, W. & Ajello, L. (1980). Mycetomas. In: A colour atlas and textbook of the histopathology of mycotic diseases. London: Wolfe Medical Publications Ltd., pp. 76-82. Destombes, P. (1964). Structure bistologique des mycetomes. Annales de la Societe’ Belge de Medicine Tropicale, 44, 897-908.

OF PALE

GRAIN

MYCETOMA

Escudie, A., Segretain, G., Destombes, P., Proye, G., Chatillon, M. & Courmes, E. (1967). Premier cas de mycetoma fongique a grain blancs en Guadeloupe. Probabilite dun nouvel agent de Maduromycose. Bulletin de la Societe de Pathologie Exotique, 60, 13-20.

Halde, C., Padbye, A. A., Haley, L. D., Rinaldi, M. G., Kay, D. & Leeper, R. (1976). Acremonium falciforrne as a cause of mycetoma in California. Sabouraudia, 14, 319-326. Kaplan, W. & Kraft, D. E. (1969). Demonstration of pathogenic fungi in formalin-fixed tissues by immunofluorescence.American .7ournal of Clinical Patkology, 52, 420-437.

Mahgoub, E. S. (1971). Maduromycetoma caused by Aspergillus nidulans. Journal Hygiene, 74, 60-61.

of Tropical Medicine and

Mahgoub, E. S. & Murray, I. G. (1973). In: Mycetoma. William Heinemann Medical Books Ltd.. DD. 30-46. Mabgoub, E. S. & Yacoub, A. A. A. (1968). Primary actinomycosis of the foot and leg: report of a case. 3ournd of Tropical Medicine and Hygiene, 71, 256-258.

Meriat, F., Destombes, P. & Segretain, G. (1977). The mycetomas: clinical features, pathology, etiology and epidemiology. Contributions to Microbiology and Immunology, 4, l-39.

Murray, I. G. & Holt, H. D. (1964). Is Cephalosporium acremonium capable of producing maduromycosis?Mycopathologia et Mycologia Applicata,

22, 335-338.

Rosa& L., Destombes, P., Segretain, G., Nazimoff, 0. & Arcouteil, A. (1961). Sur un nouvel agent de mycetome isole en Somalia. Bulletin de la Soci& de Pathologic Exotique, 54, 1265-1271. Saeed, E. N. & Hay, R. J. (1981). Immunoperoxidase staining in the recogition of Aspergillus infections. Histopathology,

5, 437-444.

Winslow, D. J. (1959). Botryomycosis. AmericanJournal Pathology, 35, 153-167.

Accepted

for

publication

4th June,

1982.

of