Mycobacterium ulcerans infection (buruli ulcer): A case report of the disseminated nonulcerative form

FROM THE FILES OF THE AFIP

Mycobacterium ulcerans Infection (Buruli Ulcer): A Case Report of the Disseminated Nonulcerative Form Fides Marie V. Abalos, MD, Sr Julia Aguiar, Augustin Gue´de´non, MD, Franc¸oise Portaels, PhD, and Wayne M. Meyers, MD, PhD The World Health Organization recognizes Mycobacterium ulcerans infection (Buruli ulcer) as a reemerging disease. Classically, lesions are indolent, undermined ulcers of the skin. The characteristic histopathologic changes are provoked by a soluble toxin of M ulcerans that is necrotizing and immunosuppressive. After tuberculosis and leprosy, Buruli ulcer is the third most common mycobacterial disease in humans. We report Buruli ulcer in a patient in Benin, West Africa, with widespread edema and diffuse induration of approximately one half of the skin of the trunk. There was no ulceration. The tissue studied was a 16-cm portion excised from the center of the large surgical specimen. Histopathologic analysis showed massive contiguous necrosis of the dermis and subcutis in sections of biopsy specimens from the center, at 2-cm intervals in two radii from the center to the periphery, and at 5-cm intervals around the margin. Acid-fast bacilli infiltrated all specimens except at one peripheral site. Samples of the entire surgical specimen taken from seven sites before fixation were polymerase chain reaction and culture positive for M ulcerans. The disseminated nonulcerative form of M ulcerans infection is well known, but is now increasingly frequent in some highly endemic areas, especially in West Africa. This patient died within 48 hours postsurgery, but cause of death was not established. The only regularly effective treatment for advanced lesions is surgical excision of all infected tissue. Estimation of the lateral limits of invasion by M ulcerans may help the surgeon establish the optimal extent of excision. Refinement of the basic concept we used and adaptation to preoperative assessment of the limit of bacterial invasion are urgently needed, especially for massive lesions. Ann Diagn Pathol 4: 386-390, 2000. This is a US government work. There are no restrictions on its use. Index Words: Benin, Buruli ulcer, mycobacteria, mycobacterial infection, Mycobacterium ulcerans infection, skin

C

LASSICALLY, Mycobacterium ulcerans causes indolent, necrotizing, nonpainful ulcers of the skin (Buruli ulcers).1 The etiologic agent is an acid-fast bacillus (AFB) with an optimal growth temperature of 32° C on routine mycobacteriologic media. M ulcerans infection prevails most com-

From the Department of Infectious and Parasitic Disease Pathology, Armed Forces Institute of Pathology, Washington, DC; Centre Sanitaire et Nutritionnel Gbemoten, Zangnanado, Benin; Programme de Lutte Contre Ulceres de Buruli, Ministere de la Sante, Cotonou, Benin; and the Department of Microbiology, Institute of Tropical Medicine, Antwerp, Belgium. Address reprint requests to Wayne M. Meyers, MD, PhD, Department of Infectious and Parasitic Disease Pathology, Armed Forces Institute of Pathology, 6825 16th St NW, Bldg 54, Washington, DC 20306-6000. This is a US government work. There are no restrictions on its use. 1092-9134/00/0406-0006$0.00/0 doi:10.1053/adpa.2000.19372

386

monly in permanent wetlands of tropical countries. Molecular biologic studies have identified M ulcerans-specific sequences in water from swamps, and in water bugs that dwell among the roots of water plants, suggesting that insects may play a role in the transmission of Buruli ulcer.2 Laboratory diagnosis of M ulcerans infection ideally involves the histopathologic demonstration of typical necrotizing panniculitis with AFB and identification of the etiologic agent by culture and molecular biologic studies. Methods for the direct and rapid specific identification of M ulcerans are available,3,4 and are especially useful in confirming the diagnosis in paucibacillary specimens from lesions. Inoculation into the skin by trauma is probably a common mode of transmission.5 Patients of all ages acquire M ulcerans infection, but occurs most frequently in the first and second decades of

Annals of Diagnostic Pathology, Vol 4, No 6 (December), 2000: pp 386-390

Mycobacterium ulcerans Infection

387

Figure 1. A 9-year-old boy was admitted to a hospital in Benin, West Africa, with generalized edema and a massive indurated area of abdominal skin. The lesion is believed to have started in the discolored area of the left midabdominal region. He died 2 days after surgery.

life. The disease is endemic in West and Central Africa, Australia, New Guinea, Southeast Asia, and South America. Many infections begin as a solitary nodule in the skin, and early excision of the nodule with primary closure is curative. If not treated at this stage, most infections progress to local ulceration. There is, however, increasing evidence that the disease may bypass the nodular stage and disseminate contiguously in the skin. Such lesions are usually advanced when detected and require extensive excision and skin grafting, often leading to cosmetic disfigurement and disabling complications. Frequently, inadequate primary excision requires repeated revisions, prolonging hospitalization and increasing deformity. Case Report A 9-year-old boy from a rural village in Benin, West Africa, presented to the Centre Sanitaire et Nutritionnel Gbemoten, Department of Zou. Approximately 2 months prior to admission, induration of the skin began on the left midabdominal area. There was no history of a localized nodule in this area of skin. The induration gradually spread over the abdomen and axilla, with recent generalized edema (Fig 1). On admission, there was mild discoloration and loss of topographic markings of the skin at the presumed site of onset, but no ulceration. The skin was indurated on the left side of the abdomen and axilla from the iliac crest to the armpit. Wide excision of the lesion included the entire thickness of the skin and subcutaneous tissue (Fig 2). A 16-cm portion from the

presumed center of the lesion (the discolored portion) was removed for study. Prior to formalin fixation, seven samples were taken for culture and molecular biologic studies. The patient died 2 days after surgery, but there was no autopsy, and the cause of death was not established.

Results The 16-cm slab of excised abdominal skin was fixed in 10% formalin. Thickness of the skin varied from 1.5 cm to 1.7 cm. There was no ulceration, but in the center the surface of the skin was smooth and discolored. An excisional specimen was removed from the center of this tissue (Fig 3). Punch biopsy specimens (6 mm) were taken at 2-cm intervals along two radii between the center and periphery, and at 5-cm intervals around the entire margin. All sections were embedded in paraffin, sectioned at 4 ␮m, and stained by the hematoxylin-eosin and Ziehl-Neelsen methods. Microscopic sections of all punch specimens showed typical changes of active Buruli ulcer. There was massive contiguous coagulation necrosis of the lower dermis and subcutaneous tissue with minimal inflammation. The interlobular septae of the panniculus were necrotic and thickened (Fig 3). Vasculitis, although present, was not a prominent feature. Ziehl-Neelsen stained sections revealed clumps of AFB distributed throughout the necrotic fat, including the septae. Samples from the center of the specimen contained the largest

388

Abalos et al

Figure 2. The incision demonstrates the margin of excision of the skin and subcutis from the iliac crest to the armpit level. Approximately 50% of the skin of the trunk was resected.

numbers of AFB (Figs 4, 5). The remaining tissue specimens showed AFB deep in the necrotic subcutis, with the exception of one specimen from the periphery which did not contain AFB. All areas sampled for culture and molecular biologic study were positive for M ulcerans. Techniques for culture of M ulcerans and molecular biologic studies were carried out as previously described.3 Discussion The earliest description of ulcers consistent with M ulcerans infection was first recorded by Sir Albert Cook in 1897 in Uganda.6 The lesions were named “Buruli ulcers” by Clancey et al in 1962 because the

Figure 3. This scanning magnification of the center of the lesion shows massive contiguous coagulation necrosis of the lower dermis and subcutaneous tissue. Note that the epidermis is intact but ironed out, and that cellular exudates are minimal in the upper dermis. The interlobular septae are thickened.

majority of their patients came from the Buruli area of Uganda.7 Other investigators have described characteristic Buruli ulcers in many countries including one who recently reported Buruli ulcer in a Canadian man who traveled extensively in Africa.8 There are no known predisposing host factors; however, some observations in highly endemic areas suggest that large families with multiple parentage reveal genetic susceptibility factors (personal communication, F. Portaels, March 1997). HLADR3 antigen was more common in one population.9 M ulcerans infection can be transmitted experimentally to animals including guinea pigs, rats, mice, and the nine-banded armadillo.10

Figure 4. Ziehl-Neelsen stain reveals large numbers of acidfast bacilli distributed throughout the necrotic subcutis. This medium power magnification of a specimen from the center of the lesion shows large masses of acid-fast bacilli.

Mycobacterium ulcerans Infection

389

Figure 5. This high power magnification shows acid-fast bacilli in the interstitium of the adipose tissue of the necrotic subcutis. (Ziehl-Neelsen stain.)

The pathogenesis of M ulcerans infection remains speculative, but most likely proceeds as follows: When M ulcerans bacilli are traumatically introduced into the skin or subcutaneous tissue, the organisms produce a small amount of a necrotizing and immunosuppressive toxin, identified as a lipidlike molecule called mycolactone with an affinity for adipocytes in subcutaneous tissue.11-16 The necrotic hypoxic tissue seems to provide an ideal milieu for the proliferation of M ulcerans, thus promoting advancing disease in the subcutaneous tissue.17 Proliferation of the etiologic agent is suppressed at normal core body temperature, limiting the disease largely to skin. Metastatic spread of M ulcerans, most likely by the hematogenous route can, however, cause cutaneous and osteomyelitic lesions at distant sites. Following inoculation of the etiologic agent, the incubation period can vary from weeks to years, but the median is approximately 2 to 3 months. The typical initial lesion is a slowly enlarging, firm, well-circumscribed pruritic nodule in the dermis and subcutaneous tissue. Radiographically, there often is mineralization in the nodule. As the nodule enlarges, the underlying necrosis erodes the epidermis, discharging an oily liquid. A necrotic slough develops in the base of the ulcer and the surrounding skin is undermined, indurated, and

hyperpigmented.18-20 Ulcers often involve massive areas of skin. There is no clinical or histopathologic evidence that the disease in the patient we are reporting began with the classic nodule of Buruli ulcer. We believe this patient is representative of the rapidly disseminating form of Buruli ulcer which bypasses the nodular stage. Such patients usually develop serious disease that causes the loss of massive areas of skin. Whether or not death of this patient was directly related to systemic effects of the toxin of M ulcerans is unknown; however, the generalized edema suggests this possibility. Connor and Lunn in 1966 developed a histopathologic classification of the disease involving 3 stages: active, organizing (granulomatous), and healing.11 The active stage, which our patient represents, has the most striking and diagnostic feature, ie, contiguous coagulation necrosis of the lower dermis and subcutaneous fat with many acidfast bacilli. Often there is dystrophic calcification and vasculitis. The organizing stage, which is usually evident in longstanding lesions, is characterized by the formation of delayed-type hypersensitivity granulomas and large numbers of foamy histiocytes, lymphocytes, and plasma cells at the margin of the necrotic fat. Healing then ensues

390

Abalos et al

with the formation of granulation tissue, fibrosis, and scarring. Many lesions of M ulcerans infection require excision of massive areas of skin, as was the case in this patient. Ideally, excision should remove a minimum amount of skin and subcutaneous tissue with an optimal result. At the present time there are no criteria other than the acumen of the surgeon to establish limits for excision. Sampling multiple sites of lesions preoperatively for AFB may provide valuable information that will be useful to the surgeon. This will require refined but simplified techniques and extensive observations to be applicable to the medical technological capabilities of most endemic areas of Buruli ulcer. Acknowledgments The authors thank Mr Luther Duckett for photographic assistance, and the American Registry of Pathology for support.

References 1. Thangaraj HS, Evans MRW, Wansbrough-Jones MH: Mycobacterium ulcerans disease: Buruli ulcer. Trans R Soc Trop Med Hyg 1999;93:337-340 2. Portaels F, Elsen P, Guimaraes-Peres A, et al: Insects in the transmission of Mycobacterium ulcerans infection. Lancet 1999; 353:986 3. Portaels F, Aguiar J, Fissette K, et al: Direct detection and identification of Mycobacterium ulcerans in clinical specimens by PCR and oligonucleotide-specific capture plate hybridization. J Clin Microbiol 1997;35:1097-1100 4. Ross BC, Marino L, Oppedisano F, et al: Development of a PCR assay for rapid diagnosis of Mycobacterium ulcerans infection. J Clin Microbiol 1997;35:1696-1700 5. Meyers WM, Shelly WM, Connor DH, et al: Human Mycobacterium ulcerans infections developing at sites of trauma to skin. Am J Trop Med Hyg 1974;23:919-923 6. Cook A: Mengo Hospital notes. Makerere Medical School Library, Kampala, Uganda. BMJ 1970;2:378-379

7. Clancey J, Dodge R, Lunn HF: Study of a mycobacterium causing skin ulceration in Uganda. Ann Soc Belge Med Trop 1962;4:585-590 8. Semret M, Koromihis G, MacLean JD, et al: Mycobacterium ulcerans infection (Buruli ulcer): First reported case in a traveler. Am J Trop Med Hyg 1999;61:689-693 9. Evans M, Wansbrough-Jones M: Immunology, in Asiedu K, Scherpbier R, Raviglione M (eds): Buruli Ulcer: Mycobacterium ulcerans Infection. Geneva, Switzerland, World Health Organization, 2000, pp 19-21 (WHO/CDS/CPE/GBUI/2000.1) 10. Walsh DS, Meyers WM, Krieg RE, et al: Transmission of Mycobacterium ulcerans to the nine-banded armadillo. Am J Trop Med Hyg 1999;61:694-697 11. Connor DH, Lunn H: Buruli ulceration. Arch Pathol 1966;81:183-189 12. Read JK, Heggie CM, Meyers WM, et al: Cytotoxic activity of Mycobacterium ulcerans. Infect Immun 1974;9:1114-1122 13. Pimsler M, Sponsler TA, Meyers WM: Immunosuppressive properties of the soluble toxin from Mycobacterium ulcerans. J Infect Dis 1988;157:577-580 14. Pahlevan AA, Wright DJM, Andrews C, et al: The inhibitory action of Mycobacterium ulcerans soluble factor on monocyte/T cell cytokine production and NF-␬B function. J Immunol 1999;163:3928-3935 15. George KM, Barker LP, Welty DM, et al: Partial purification and characterization of biological effects of a lipid toxin produced by Mycobacterium ulcerans. Infect Immun 1998;66:587593 16. George KM, Chatterjee D, Gunawardana G, et al: Mycolactone: A polyketide toxin from Mycobacterium ulcerans required for virulence. Science 1999;283:854-857 17. Palomino JC, Obiang AM, Realini L, et al: Effect of oxygen on Mycobacterium ulcerans growth in the BACTEC system. J Clin Microbiol 1998;36:3420-3422 18. Josse R, Gue´de´non A, Darie H, et al: Les infections cutane´es a` Mycobacterium ulcerans: Ulce`res de Buruli. Revue Ge´ne´rale. [Mycobacterium ulcerans cutaneous infections: Buruli ulcers. General review.] Me´decine Trop 1995;55:363373 19. Hayman J: Out of Africa: Observations on the histopathology of Mycobacterium ulcerans infection. J Clin Pathol 1993;46:5-9 20. Meyers WM: Mycobacterial infections of the skin, in Doerr W, Seifert G (eds): Tropical Pathology (vol 8). Berlin, Germany, Springer-Verlag, 1995, pp 291-377