Blastomycosis of bones and joints.

Abstract: A retrospective study of 45 patients hospitalized with blastomycosis of bones or joints revealed 41 cases of osteomyelitis and 12 cases of septic arthritis. The majority were men (35 [78%] patients) and non-Aboriginal (32 [71%] patients). Median time from the onset of symptoms to hospitalization was shorter in women than men (male, 48 d; female, 14 d; P < 0.02), and shorter for Aboriginals than non-Aboriginals (non-Aboriginal, 50 d; Aboriginal, 19 d; P < 0.04). Cutaneous disease was present in 33 (73%) patients, and lung involvement was present in 29 (64%) patients. The most common osseous sites of involvement were the lower limb and axial skeleton. Common orthopaedic symptoms of bone lesions included bone pain in 42 (78%) patients, swelling in 32 (59%) patients, and soft tissue abscesses in 21 (39%) patients. Joint infection (12 patients) manifested as a monoarticular arthropathy presenting with effusion in 9 (75%) patients, pain in 8 (67%) patients, and decreased range of motion in 5 (42%) patients. Osseous blastomycosis can mimic bacterial infection and should be included in the differential diagnosis of bone and joint infection in patients who have visited or who live in geographic regions where B dermatitidis is endemic.

Key Words: Blastomyces dermatitidis, musculoskeletal, osteomyelitis, septic arthritis

**********

Blastomycosis is an uncommon systemic fungal infection caused by the thermally dimorphic fungus Blastomyces dermatitidis, which is endemic in the wooded areas of Manitoba, northwestern Ontario, the Great Lakes, and the Mississippi and Ohio River valleys. (1,2) In the lung, inhaled conidia from the soil convert to the yeast phase, leading to respiratory infection. (3) Subsequent hematogenous dissemination may cause extrapulmonary disease. (4-6)

Bone is the third most common site of blastomycotic lesions, after lung and skin. (5-8) Most available information about skeletal blastomycosis is from case reports or small groups of patients. (7,9) Bones may be involved in 14 to 60% of cases of blastomycosis, and any bone may be involved, including the vertebrae, ribs, facial bones, skull, long bones, small bones, pelvis, and scapulae. (7,10-12) Patients with blastomycotic osteomyelitis have pain and swelling at the site, often accompanied by an overlying skin abscess or ulcer. Treatment consists of antifungal drugs such as intravenous amphotericin B and oral agents such as itraconazole and ketoconazole, but in many instances surgical treatment may be necessary, including debridement and curettage. Synovial joint involvement also has been reported. (13) Delayed diagnosis until late in the natural history is common.

The purpose of this study was to evaluate a large group of patients with this uncommon disease to determine the disease profile, including patient characteristics, clinical manifestations, bone and joint sites involved, method of diagnosis, and the effectiveness of treatment with antifungal agents with or without surgical intervention.

Materials and Methods

Subjects

A retrospective medical record review was done for all patients with proven bone or joint infection with B dermatitidis treated at all seven hospitals with greater than 150 beds in the Canadian province of Manitoba (Brandon General Hospital, Concordia General Hospital, Grace General Hospital, Health Sciences Centre, Saint Boniface General Hospital, Seven Oaks General Hospital, and Victoria General Hospital) and Lake of the Woods District Hospital in Kenora, Ontario. Patient medical records for evaluation were identified using ICD-9 code 116.0 for blastomycosis. (14) Blastomycosis involving the bones or joints was confirmed with a positive culture or smear from the specific bone or joint site, but cases with radiographic evidence of osteomyelitis or septic arthritis that had the pathogen isolated from a separate site such as lung or skin were included. Prior approval for the study was obtained from the Research Ethics Board, Faculty of Medicine, University of Manitoba and the Institutional Review Boards of the facilities where patient records were reviewed.

There were 45 patients identified with blastomycosis of bones and joints. The earliest identified case was in the year 1963 and the most recent in 2005 (1.1 cases per year). However, only one of the eight hospitals (Health Sciences Centre) had medical records that were accessible as early as 1963. Medical records were reviewed and data recorded including demographic characteristics such as gender, ethnicity, age, and place of residence; past medical history including smoking and alcohol use; presenting symptoms including constitutional, pulmonary, cutaneous, and orthopaedic symptoms; diagnostic parameters including method of diagnosis, length of time from appearance of symptoms to admission, and length of time from admission to diagnosis; location of cutaneous lesions; sites of bone or joint involvement; and effectiveness of treatment with antifungal agents with or without surgical intervention. Previous reports from this institution had included some musculoskeletal information from 11 of the patients, (7) nonorthopaedic data of other patients, (1) and three case reports (15-17) of patients who were included in the present study.

Definitions

A patient was confirmed as having a case of blastomycosis when there was a respiratory infection, cutaneous disease, or musculoskeletal syndrome, and B dermatitidis was isolated from a clinical specimen by culture or visualization in a Gomori methenamine silver stain as a characteristic broad-based budding yeast (Fig. 1). (1,18)

The Aboriginal subjects were descendents of the indigenous inhabitants of Canada, including the First Nations (North American Indians) and Metis people of diverse tribal, cultural, and political identities. (19) Current smokers were patients who used tobacco daily, and ex-smokers were defined as "current smokers" who had quit within the past 10 years. Patients were designated as having received care on the ward if all hospital care was delivered on the medical or surgical ward and in the intensive care unit (ICU) if any aspect of their care was delivered in the ICU. Weight loss was defined as a loss of at least 10% of body weight.

[FIGURE 1 OMITTED]

Data Analysis

Data were analyzed with statistical software (SAS version 9.1 SAS Institute, Cary, NC). Mean, median, and standard deviation for continuous variables, and percentage for categorical variables, were determined. Patient characteristics were analyzed for differences between groups with univariate analysis. Population comparisons were made with Poisson distribution. Comparisons were made with Kruskal-Wallis test for continuous variables and [chi square] test for categorical variables. Average results are reported as mean [+ or -] standard deviation. Statistical significance was defined as P [less than or equal to] 0.05.

Results

In total, 45 patients with diagnosed B dermatitidis infection of a bone or joint were identified. There were 41 patients who had bone involved and 12 patients who had a joint affected, including 8 patients in whom both bone and joint were infected. There were 54 bone lesions and 12 septic joints. A single bone lesion was the most common manifestation, occurring in 33 (81%) of the 41 patients with bone disease; 5 (12%) patients presented with two separate foci of bone involvement; and 3 (7%) patients had three or more bone lesions. Joint involvement was present in 4 (12%) of the patients with a single bone lesion, 2 (40%) of those with two bone lesions, and 3 (100%) of those with three or more bone lesions.

The majority of patients were male (35 [78%] male and 10 [22%] female patients). The majority of subjects were non-Aboriginal (32 [71%] non-Aboriginal and 13 [29%] Aboriginal patients), but the percentage of Aboriginal patients in the study population was significantly greater than the Aboriginal population distribution in Manitoba and northwestern Ontario (Aboriginal, 14% of population) (20) (P < 0.005). The mean age of all patients was 34 [+ or -] 18 years. More than one quarter of patients (12 [27%] patients) were less than or equal to 20 years of age, and the fourth decade (31 to 40 years of age) was the most common age of presentation for all patients (13 [29%] patients). A total of 22 (49%) patients were either current or ex-smokers. Sixteen (50%) non-Aboriginal patients lived in Winnipeg, the largest city in the region, and 10 (31%) non-Aboriginal patients lived in the endemic area of northwestern Ontario (rural northwestern Ontario and Kenora). Ten (45%) of the 22 non-Aboriginal patients who lived in Manitoba had a history of travel in northwestern Ontario. Eight (61%) Aboriginal subjects resided in reservations in northwestern Ontario, and only 3 (23%) Aboriginal subjects resided in urban communities. Previously diagnosed infection with B dermatitidis was noted in 4 (9%) patients. Rates of comorbid disease (cardiovascular disease, pulmonary disease, diabetes, cancer, and liver disease) were comparable to population averages (data not shown). Twenty-two (49%) patients were current or former tobacco smokers.

Constitutional symptoms included fever (20 [44%] patients), chills or night sweats (15 [33%] patients), weight loss (13 [29%] patients), anorexia (10 [22%] patients), sore throat or rhinorrhea (3 [7%] patients), and nausea or vomiting (1 [2%] patient); 17 (38%) patients did not have any constitutional symptoms. Pulmonary symptoms included cough (23 [51%] patients), dyspnea (13 [29%] patients), sputum production (11 [24%] patients), chest pain (7 [16%] patients), and hemoptysis (5 [11%] patients). Eighteen (40%) patients did not have any pulmonary symptoms, including 17 patients who did not have pulmonary disease and 1 patient who had no pulmonary symptoms but had pulmonary disease documented by a positive sputum culture. Seven (16%) patients required supplemental oxygen, 5 (11%) patients had respiratory failure and required admission to the ICU, and 4 (9%) patients required a ventilator.

[FIGURE 2 OMITTED]

The majority of patients (32 [71%] patients had skin lesions (total, 52 skin lesions in all patients), and 21 (40%) of the skin lesions appeared to result from underlying bone and joint pathology extending into the subcutaneous tissues as an abscess or draining sinus. The other skin lesions constituted distinct cutaneous blastomycosis, not associated with any contiguous bone or joint disease, and dermatologic manifestations were noted in all regions of the body: head and neck, 14 (27%) lesions; lower extremity, 9 (17%) lesions; thorax, 9 (17%) lesions; foot, 6 (11.5%) lesions; upper extremity, 6 (11.5%) lesions; abdomen, 4 (8%) lesions; and hand, 4 (8%) lesions.

Ten (22%) patients had a documented history of trauma at the site of bone or joint involvement that preceded the development of symptoms. There were a total of 54 bone lesions (1.2 bony lesions per patient) (Fig. 2 and Fig. 3), and the most common sites of bone lesions were the lower limb and axial skeleton (Table 1). The tibia and ribs were the most common bones to be affected (Table 1). Of the 19 lesions of the long bones of the lower and upper extremity, 16 (84%) were in the metaphysis and 3 (16%) were in the diaphysis. The most frequent patient complaints at presentation to hospital were pain, soft tissue swelling, and skin abscess (Table 1). Pathologic fractures were uncommon and anatomic location varied (Table 1). In the 12 cases of joint involvement, the pattern of infection was monoarticular, the most frequent sites were the knee and elbow, and the most frequent signs and symptoms were effusion, joint pain, and decreased range of motion (Table 1).

[FIGURE 3 OMITTED]

In the 41 patients with blastomycosis of bone, the microbiologic diagnosis was made from culture of bone fragments and curettings in 28 (68%) patients, debrided soft tissue or biopsy specimens in 7 (17%) patients, abscess aspirate in 2 (5%) patients, lung biopsy tissue in 2 (5%) patients, skin swab in 1 (2.5%) patient, and sputum in 1 (2.5%) patient. In the 12 patients with blastomycosis of a joint, the microbiologic diagnosis was made from culture of synovial fluid in 11 (92%) patients and lung biopsy in 1 patient. All patients had at least one imaging study, including plain x-rays in 42 (93%) patients, bone scans in 18 (40%) patients, and computerized tomography (CT) scans in 13 (29%) patients. Magnetic resonance imaging (MRI) was done in 3 (7%) patients for evaluation of epidural invasion or spinal cord compression.

In the 45 patients with bone or joint blastomycosis, skin involvement was noted in 33 (73%) patients and lung involvement was noted in 29 (64%) patients; 20 (44%) patients had both lung and skin involvement. Only 3 (7%) of the 41 patients who had bone or joint blastomycosis had neither lung nor skin involvement. Despite the pulmonary portal of entry, 16 (36%) of the patients in the study did not have evidence of pulmonary infection at the time of presentation. For the 29 (64%) patients with pulmonary involvement, chest radiographic findings consisted of consolidation in 11 (38%) patients, miliary pattern in 8 (28%) patients, interstitial pattern in 7 (24%) patients, pleural effusion in 5 (17%) patients, pulmonary mass in 1 (3%) patient, and no radiographic abnormality in 1 (3%) patient; multiple radiographic findings were noted in 4 patients. Miliary pattern was associated with widely disseminated disease including 4 (80%) of the 5 cases of concurrent lungs, skin, bone, and joint involvement, and 4 (29%) of the 14 cases with concurrent lung, skin, and bone disease.

Delay between onset of symptoms and admission to hospital was common, on average more than two months (Table 2). The median delay between onset of symptoms and admission to hospital was significantly greater for male than female patients (male, 48 d; female, 14 d; P < 0.02) and for non-Aboriginal than Aboriginal patients (non-Aboriginal, 50 d; Aboriginal, 19 d; P < 0.04). The diagnosis was made in 10 (22%) patients before admission to a healthcare facility. Of the 35 (78%) patients diagnosed in hospital, the median time from admission to establishment of the diagnosis was 7 days. Twenty three (51%) patients were initially misdiagnosed as having bacterial osteomyelitis. Average duration of hospital stay was almost one month (Table 2).

The majority of patients received hospital care in a regular ward, but a significantly greater proportion of Aboriginal than non-Aboriginal patients required admission to the ICU because of more serious pulmonary disease that necessitated close monitoring and treatment (Table 2). In three of the five patients who required admission to the ICU, the time from symptom onset to hospital admission (15, 20, 24, 56, and 85 d) was less than the population median (34 d); in two of these five patients, time from hospital admission to diagnosis (1, 4, 8, 10, and 14 d) was less than then population median (7 d). Therefore, delay in diagnosis did not explain the need for ICU admission. Two (4%) patients died as a result of severe pulmonary blastomycosis; the duration of symptoms before treatment in these patients (25 and 33 d) was less than the population median of time from first symptoms to hospital admission (34 d), suggesting that delay in treatment did not explain these deaths. Recurrent infection at the original or a distant site occurred in 3 (7%) patients.

Treatment consisted of antifungal drugs in all patients and surgery in the majority (Table 3). Before the late 1980s, parenteral amphotericin B (1.5-2.5 g per day) was used exclusively, and none of these patients died or had recurrent disease. After newer, less toxic antifungal agents became available, therapy was changed to a brief initial course of amphotericin B in hospital followed by an outpatient course of an oral antifungal agent such as ketoconazole or itraconazole. Clinically milder cases were treated with six months of oral itraconazole (200 mg per day) or ketoconazole (400 mg per day) (Table 3). Amphotericin B was used in 24 patients and was discontinued in 7 (29%) patients because of increased serum creatinine levels and acute renal failure (six patients) and severe cutaneous eruptions (one patient). An additional six patients experienced renal toxicity on amphotericin B at the above dosage but tolerated therapy at a lower dosage. There were no documented cases of chronic renal failure arising from amphotericin B therapy, and no patients in the study suffered bone marrow suppression. Oral drugs were discontinued because of hepatotoxicity in three patients from itraconazole and one patient from ketoconazole. There was a mild increase in liver enzymes in two patients, and more serious hepatic disease in one patient, but liver function returned to normal in all these cases after the antifungal drugs were discontinued. Of the three patients who had recurrence of blastomycosis, one had been treated with a small dose of amphotericin B (330 mg) followed by oral ketoconazole, one had received ketoconazole only, and one had received itraconazole only. Surgical treatment included 40 operations in 29 (64%) patients (1.4 operations per patient who had surgery). The 76 procedures (1.9 procedures per operation) included debridement (31 procedures; 24 [53%] patients), bone curettage (20 procedures; 17 [38%] patients), arthrotomy (6 procedures; 5 [11%] patients), ribresection (4 procedures), arthroscopic lavage (3 procedures), bone biopsy (3 procedures), closure of defect (3 procedures), bone grafting (2 procedures), craniotomy (2 procedures), sternum excision (1 procedure), and synovectomy (1 procedure).

Residual bone or joint symptoms, including impaired function, were present in 11 (24%) patients. The most common residual symptoms consisted of pain and decreased range of motion (Fig. 2, Fig. 4, and Fig. 5). Five (63%) of the 8 cases of pathologic fracture resulted in residual symptoms.

Discussion

Although blastomycosis is an uncommon cause of osteomyelitis or septic arthritis, the data suggest that this disease should be considered in the differential diagnosis of lung, skin, and musculoskeletal symptoms, especially in people living or traveling in regions where B dermatitidis is endemic. When the diagnosis was made promptly, bone and joint blastomycosis responded to antifungal drugs with or without surgical debridement. The key to the diagnosis was an index of suspicion and a low threshold to obtain fungal cultures, particularly if treatment for bacterial infection had been ineffective. It is possible that more cases may have occurred in this region but were not included in the present study because only one of the study hospitals had accessible records as early as the year 1963.

[FIGURE 4 OMITTED]

The observed predominance of males also has been demonstrated in other studies and has been attributed to an increased risk of exposure in men through outdoor work or recreational activities. (8,21,22) The predominance of Aboriginal subjects in the present study compared with that in the general population suggests that there may be a greater predilection of blastomycosis for the Aboriginal than non-Aboriginal population, as has been reported in other studies. (1,21) The reason that Aboriginal, more frequently than non-Aboriginal, patients had severe pulmonary disease that necessitated treatment in an ICU, and the greater prevalence of disease in the Aboriginal population, may reflect compromised nutritional and immune factors associated with socioeconomic status and genetic factors. (23,24) Furthermore, the finding that the majority of Aboriginal patients lived on a reservation suggests that increased environmental exposure to B dermatitidis may have occurred in this group as a result of greater contact with soil-containing conidia related to either residential, recreational, or employment activities. In the non-Aboriginal population in Winnipeg, a major urban center, exposure may have been from the common local practice of vacationing in northwestern Ontario, and history of recent travel to the endemic area may be underreported in medical records. (1)

[FIGURE 5 OMITTED]

Smoking may impair immune resistance and increase susceptibility to pulmonary infection. Tobacco use was prominent in patients with bone or joint blastomycosis and was more frequent than in the general population of Manitoba and Ontario (20%) in recent years. (25) The finding that many patients developed bone or joint disease without pulmonary symptoms is consistent with reports of mild or subclinical, asymptomatic B dermatitidis pulmonary infection. (26,27) Hematogenous spread of the pathogen from a subclinical pulmonary infection may cause extrapulmonary infection analogous to that of infection with Mycobacterium tuberculosis. (28-30) Furthermore, reactivation of blastomycosis may occur as late as forty years after original exposure or after treatment with a full course of amphotericin B. (31,32) The three patients in the present study who had a history of blastomycosis before the current bone and joint infection may have had cases of reactivation and not necessarily de novo exposure.

Joint disease may occur either by direct extension of juxta-articular osteomyelitis or through hematogenous dissemination from the lungs, (33,34) and both mechanisms were observed in the current study. Those patients with multiple bone involvement more frequently had concurrent septic arthritis than those with a single bone lesion. Involvement of the joints of the hands or feet, noted in only one patient (Table 1), rarely has been reported. (35)

Although there is no typical radiographic appearance of blastomycosis osteomyelitis, lesions of long and short bones are classified as either focal or diffuse. (36,37) The focal pattern may have a sclerotic margin and may expand slowly. Periosteal reaction is generally present in the long bones but absent from the short bones. Diffuse osteomyelitis is characterized by rapid bone destruction (moth-eaten pattern) and vigorous periosteal reaction, and the infection may invade adjacent joints or produce draining sinuses. In the spine, commonly observed features include destruction of the disk spaces, anterior destruction of the vertebral body causing collapse, paraspinal masses, and skipped spinal segments, similar to features observed in M tuberculosis infection. (36,38,39) Rib lesions may be produced by invasion from adjacent foci such as the pleura or pulmonary masses or from hematogenous spread. (36) Plain x-rays were used extensively in the clinical evaluation of the present patient population because of availability and low cost. However, sensitivity may be limited because extensive bone loss may be required to produce radio-graphic changes. Bone scintigraphy using technetium-99 or gallium-67 may have greater sensitivity than plain film radiography and may provide a whole body image to detect occult lesions. (38,40)

Blastomycosis may affect almost any bone in the skeleton. (7,10-12) The observation of metaphyseal involvement of long bones has been noted in previous reports and may be attributed to sluggish circulation in metaphyseal vascular loops. (7,9,36) The skeletal distribution of observed lesions was similar to that reported in a recent review of several studies, with lower extremity more prevalent than upper extremity lesions, except that there was a lower prevalence of vertebral osteomyelitis in the present study (Table 1). (12,36,41) Trauma at the site of bone or joint preceding approximately one fifth of blastomycosis cases may have been coincidental or may have caused a local inflammatory reaction that attracted neutrophils carrying B dermatitidis to the site, leading to seeding and infection. (32)

Limitations of the study include those inherent with a retrospective study with information based primarily from medical records of varied quality. Numerous x-rays were not available because of limited retention of x-rays at the study hospitals, precluding a comparative radiographic analysis. Furthermore, it was unclear from the available records the deciding reasons for proceeding with or deferring surgical debridement, and some patients healed with nonoperative treatment; therefore, specific guidelines for the indications and timing of surgical treatment could not be established. Surgical intervention usually was undertaken to obtain deep cultures for diagnosis, drain an abscess, and debride of infected soft tissue and bone to achieve a more expedient healing of the affected area.

Acknowledgments

The authors thank the Medical Records Departments at Brandon General Hospital, Concordia General Hospital, Grace General Hospital, Health Sciences Centre, Lake of the Woods District Hospital, Saint Boniface General Hospital, Seven Oaks General Hospital, and Victoria General Hospital. We appreciate expert statistical analysis from Mary Cheang, M. Math. The authors gratefully acknowledge the assistance of Peter MacDonald, MD, Tracy Crampton, MD, Gerhard Bock, MD, Pete Sarsfield, MD, Kerry MacDonald, MD, Carolyn Garlinski, Debbie Ormiston, Maryanne Kennedy, RN, Lorraine Kuzyk, RN, and Pam Schille, RN.

References

1. Crampton TL, Light RB, Berg GM, et al. Epidemiology and clinical spectrum of blastomycosis diagnosed at Manitoba hospitals. Clin Infect Dis 2002;34:1310-1316.

2. Furcolow ML, Balows A, Menges RW, et al. Blastomycosis: an important medical problem in the Central United States. JAMA 1966;198:529-532.

3. Schwarz J, Baum GL. Blastomycosis. Am J Clin Pathol 1951;21:999-1029.

4. Riegler HF, Goldstein LA, Betts RF. Blastomycosis osteomyelitis. Clin Orthop Relat Res 1974;100:225-231.

5. Chapman SW. Blastomyces dermatitidis. Bennett's Principles and Practice of Infectious Diseases. 6th edition. Mandell GL, Bennett JE, Dolin R, (eds). Philadelphia, Elsevier, 2005, pp 3026-3040.

6. Cherniss EI, Waisbren BA. North American blastomycosis: a clinical study of 40 cases. Ann Intern Med 1956;44:105-123.

7. MacDonald PB, Black GB, MacKenzie R. Orthopaedic manifestations of blastomycosis. J Bone Joint Surg Am 1990;72:860-864.

8. Chapman SW, Lin AC, Hendricks KA, et al. Endemic blastomycosis in Mississippi: epidemiological and clinical studies. Semin Respir Infect 1997;12:219-228.

9. Saiz P, Gitelis S, Virkus W, et al. Blastomycosis of long bones. Clin Orthop Relat Res 2004;421:255-259.

10. Jones RR, Martin DS. Blastomycosis of bone. Surgery 1941;10:931-938.

11. Bassett FH, Tindall JP. Blastomycosis of bone. South Med J 1972;65:547-555.

12. Colonna PC, Gucker T. Blastomycosis of the skeletal system. J Bone Joint Surg 1944;26:322-328.

13. Albert MC, Zachary SV, Alter S. Blastomycosis of the forearm synovium in a child. Clin Orthop Relat Res 1995;317:223-226.

14. International Classification of Diseases, 9th Revision. Clinical Modification, 4th ed. Los Angeles, Practice Management Information Corporation (PMIC), 1994.

15. Cherneski C, Bartley D, Al-Thaquafi H, et al. Infection with Blastomyces dermatitidis in a Saudi Arabian man. Am J Orthop, in press.

16. Embil JM, Wiens JL, Oppenheimer M, et al. Foot ulcer and osteomyelitis. CMAJ 2006;174:35-36.

17. Embil JM, Aoki FY, Hershfield ES, et al. Respiratory failure in a man with bone and lung lesions. Hosp Pract (Minneap) 1998;33:211-219.

18. Centers for Disease Control and Prevention (CDC). Blastomycosis--Wisconsin, 1986-1995. MMWR Morb Mortal Wkly Rep 1996;45:601-603.

19. Royal Commission on Aboriginal Peoples. Report of the Royal Commission on Aboriginal Peoples. Volume 1. Looking Forward, Looking Back. Ottawa: Canada Communication Group, 1996.

20. Statistics Canada. Census of Population, May 15, 2001. Available at http://www12.statcanca. Accessed May 22, 2004.

21. Kepron MW, Schoemperlen CB, Hershfield ES, et al. North American blastomycosis in Central Canada. Can Med Assoc J 1972;106:243-246.

22. Lowry PW, Kelso KY, McFarland LM. Blastomycosis in Washington Parish, Louisiana, 1976-1985. Am J Epidemiol 1989;130:151-159.

23. Trovato F. Aboriginal mortality in Canada, the United States and New Zealand. J Biosoc Sci 2001;33:67-86.

24. Larcombe L, Rempel JD, Dembinski I, et al. Differential cytokine genotype frequencies among Canadian Aboriginal and Caucasian populations. Genes Immun 2005;6:140-144.

25. Statistics Canada. Percentage of smokers in the population, by province, 2000-2001. Available at http://www40statcanca/101/cst01/health07b.htm. Accessed February 3, 2005.

26. Davies SF, Sarosi GA. Epidemiological and clinical features of pulmonary blastomycosis. Semin Respir Infect 1997;12:206-218.

27. Vaaler AK, Bradsher RW, Davies SF. Evidence of subclinical blastomycosis in forestry workers in northern Minnesota and northern Wisconsin. Am J Med 1990;89:470-476.

28. Stelling CB, Woodring JH, Rehm SR, et al. Miliary pulmonary blastomycosis. Radiology 1984;150:7-13.

29. Sarosi GA, Davies SF. Blastomycosis. Am Rev Respir Dis 1979;120:911-938.

30. Laskey W, Sarosi GA. Endogenous activation in blastomycosis. Ann Intern Med 1978;88:50-52.

31. Landis FB, Varkey B. Late relapse of pulmonary blastomycosis after adequate treatment with amphotericin B. Am Rev Respir Dis 1976;113:77-81.

32. Ehni W. Endogenous reactivation in blastomycosis. Am J Med 1989;86:831-832.

33. Sanders LL. Blastomycosis arthritis. Arthritis Rheum 1967;10:91-98.

34. Bayer AS, Scott VJ, Guze LB. Fungal arthritis, IV: blastomycotic arthritis. Semin Arthritis Rheum 1979;9:145-151.

35. Claflin K, Milbauer J, Sullivan B. Ketoconazole and blastomycotic osteomyelitis. Ann Intern Med 1983;98:260-261.

36. Gehweiler JA, Capp MP, Chick EW. Observations on the roentgen patterns in blastomycosis of bone: a review of cases from the Blastomycosis Cooperative Study of the Veterans Administration and Duke University Medical Center. Am J Roentgenol Radium Ther Nucl Med 1970;108:497-510.

37. Kalbhen C. Blastomycotic osteomyelitis. Orthopedics 1991;14:722-728.

38. Detrisac DA, Harding WG, Greiner AL, et al. Vertebral North American blastomycosis. Surg Neurol 1980;13:311-312.

39. Guler N, Palanduz A, Ones U, et al. Progressive vertebral blastomycosis mimicking tuberculosis. Pediatr Infect Dis J 1995;14:816-818.

40. Daniel BL, Crabbe JP, Gritters L, et al. Bone scintigraphy in blastomycotic osteomyelitis. Clin Nucl Med 1993;18:203-207.

41. Hershkovitz I, Rothschild BM, Dutour O, et al. Clues to recognition of fungal origin of lytic skeletal lesions. Am J Phys Anthropol 1998;106:47-60.

Imagination is the beginning of creation. You imagine what you desire,
you will what you imagine and at last you create what you will.
--George Bernard Shaw

Mark Oppenheimer, BSc, John M. Embil, MD, FRCPC, Brian Black, MD, FRCSC, Lyle Wiebe, CPHI(C), Bill Limerick, CPHI(C), Kerry MacDonald, MD, and Elly Trepman, MD

From the Faculty of Medicine, the Department of Medicine, Section of Infectious Diseases, the Department of Medical Microbiology, and the Department of Surgery, University of Manitoba, Winnipeg, Manitoba, Canada; Northwestern Health Unit, Kenora, Ontario, Canada; Lake of the Woods District Hospital, Kenora, Ontario, Canada; and the Department of Orthopaedic Surgery, Grand Itasca Clinic and Hospital, Grand Rapids, MN.

Reprint requests to Dr. John Embil, Infection Control Unit, Health Sciences Centre, MS 673-820 Sherbrook Street, Winnipeg, MB R3A 1R9. Email: [e-mail]

Mark Oppenheimer received a stipend from the Canadian Institutes of Health Research.

The authors have no conflict of interest, commercial interest, or proprietary interest in any drug, device, or equipment mentioned in the article.

Prior approval for the study was obtained from the Research Ethics Board, Faculty of Medicine, University of Manitoba and the Institutional Review Boards of the facilities where patient records were reviewed.

Accepted January 17, 2007.

RELATED ARTICLE: Key Points

* A retrospective study of 45 patients hospitalized with blastomycosis of bones or joints revealed 41 cases of osteomyelitis and 12 cases of septic arthritis.

* Cutaneous disease was present in 33 (73%) patients and lung involvement was present in 29 (64%) patients.

* The most common osseous sites of involvement were the lower limb and axial skeleton.

* Joint infection manifested as a monoarticular arthropathy presenting with effusion, pain, and decreased range of motion.

* Osseous blastomycosis can mimic bacterial infection and should be included in the differential diagnosis of bone and joint infection in patients who have visited or who live in geographic regions where B dermatitidis is endemic.

Table 1. Location and associated signs and symptoms of blastomycosis
lesions of bones and joints

                                      No. (%) lesions with clinical
                                      signs or symptoms
                                                               Decreased
                          No. (%) of  Bone or     Swelling or  range of
Location of lesion        lesions     joint pain  effusion     motion

Bone lesions (b)
  Axial skeleton (c)      22 (41)     15 (68)      7 (32)      0 (0)
  Long bones              19 (35)     16 (84)     14 (74)      2 (11)
    Lower extremity (d)   15 (28)     12 (80)     10 (67)      2 (13)
    Upper extremity (e)    4 (7)       4 (100)     4 (100)     0 (0)
  Small bones             12 (22)     10 (83)     10 (83)      2 (17)
    Feet (d)               8 (15)      7 (88)      8 (100)     1 (13)
    Hands (e)              4 (7)       3 (75)      2 (50)      1 (25)
    Patella (d)            1 (2)       1 (100)     1 (100)     1 (100)
Total (bone lesions) (b)  54 (100)    42 (78)     32 (59)      5 (9)
Joint lesions
  Knee                     6 (50)      2 (33)      4 (67)      2 (33)
  Elbow                    4 (33)      4 (100)     4 (100)     3 (75)
  Ankle                    1 (8)       1 (100)     1 (100)     0 (0)
  Tarsometatarsal          1 (8)       1 (100)     0 (0)       0 (0)
Total (joint lesions)     12 (100)     8 (67)      9 (75)      5 (42)

                                No. (%) lesions with clinical
                                signs or symptoms
                                               Skin
Location of lesion        Pathologic fracture  abscess (a)  Asymptomatic

Bone lesions (b)
  Axial skeleton (c)      3 (14)               10 (45)      6 (27)
  Long bones              3 (16)                6 (32)      0 (0)
    Lower extremity (d)   1 (7)                 6 (40)      0 (0)
    Upper extremity (e)   2 (50)                0 (0)       0 (0)
  Small bones             2 (17)                5 (42)      1 (8)
    Feet (d)              1 (13)                5 (63)      0 (0)
    Hands (e)             1 (25)                0 (0)       1 (25)
    Patella (d)           0 (0)                 0 (0)       0 (0)
Total (bone lesions) (b)  8 (15) (f)           21 (39)      7 (13)
Joint lesions
  Knee                    0 (0)                 1 (17)      0 (0)
  Elbow                   0 (0)                 2 (50)      0 (0)
  Ankle                   0 (0)                 0 (0)       0 (0)
  Tarsometatarsal         0 (0)                 1 (100)     0 (0)
Total (joint lesions)     0 (0)                 4 (33)      0 (0)

In the 45 patients, 33 (73%) patients had involvement of bone only, 8
(18%) patients bone and joint, and 4 (9%) patients joint only.
(a) Skin abscess directly overlying area of bone or joint blastomycosis.
(b) 54 bones were involved in 41 patients with blastomycosis
osteomyelitis.
(c) Bone lesion sites in axial skeleton: total, 22 (41%) lesions; ribs,
9 (16.5%) lesions; vertebrae, 4 (7%) lesions; skull, 4 (7%) lesions;
sternum, 2 (4%) lesions; innominate bone, 2 (4%) lesions; and scapula, 1
(2%) lesion.
(d) Bone lesion sites in lower limb: total, 24 (44%); tibia, 9 (16.5%)
lesions; tarsal bones, 4 (7%) lesions; femur, 3 (5.5%) lesions; fibula,
3 (5.5%) lesions; metatarsal, 3 (5.5%) lesions; phalanx (foot), 1 (2%)
lesion; patella, 1 (2%) lesion.
(e) Bone lesion sites in upper limb: total, 8 (15%); humerus, 2 (4%)
lesions; radius/ulna, 2 (4%) lesions; carpal/metacarpal, 2 (4%) lesions;
and phalanx (hand), 2 (4%) lesions.
(f) Sites of pathologic fracture: 2 fractures at T11; 1 fracture each at
C4, medial femoral condyle, distal humeral metaphysis, humeral
capitellum, fifth metatarsal, and second proximal phalanx (hand).

Table 2. Time to diagnosis, hospital treatment, and outcomes of patients
with bone and joint blastomycosis (a)

                                  Non-Aboriginal
Parameters               Male          Female       Total

Number of patients        23            9           32
Time from first          120 [+ or -]  25 [+ or -]  95 [+ or -]
  symptoms to admission    160           26           150
  (days)
Time from admission to     9 [+ or -]  24 [+ or -]  15 [+ or -]
  diagnosis (days) (c)       7           50           30
Duration of hospital      24 [+ or -]  20 [+ or -]  22 [+ or -]
  stay (days)               30           28           30
Admission only to ward    21 (91)       8 (89)      29 (91)
Admission to ICU           1 (4)        1 (11)       2 (6)
  during stay (d)
Died                       1 (4)        0 (0)        1 (3)
Residual bone or           5 (22)       2 (22)       7 (22)
  joint symptoms
Recurrence of B            1 (4)        1 (11)       2 (6)
  dermatiditis
  infection

                                  Aboriginal                 Total
Parameters               Male         Female    Total        population

Number of patients       12            1        13           45
Time from first          33 [+ or -]  12        31 [+ or -]  75 [+ or -]
  symptoms to admission    32                     30           130 (b)
  (days)
Time from admission to    8 [+ or -]  30        10 [+ or -]  13 [+ or -]
  diagnosis (days) (c)      7                     9            30
Duration of hospital     29 [+ or -]  38        30 [+ or -]  25 [+ or -]
  stay (days)              20                     20           30
Admission only to ward    8 (67)       1 (100)   9 (69)      38 (84)
Admission to ICU          3 (25)       0 (0)     3 (23)       5 (11)
  during stay (d)
Died                      1 (8)        0 (0)     1 (8)        2 (4)
Residual bone or          4 (33)       0 (0)     4 (31)      11 (24)
  joint symptoms
Recurrence of B           1 (8)        0 (0)     1 (8)        3 (7)
  dermatiditis
  infection

(a) Data reported as mean [+ or -] standard deviation or number (%) of
patients.
(b) Median time from first symptoms to admission, 34 days.
(c) N = 35 patients; in the other 10 patients, diagnosis was made prior
to admission.
(d) ICU, intensive care unit; difference between total non-Aboriginal
and total Aboriginal, P < 0.02.

Table 3. Pharmacological and surgical options for treatment of bone and
joint blastomycosis

                                    No. (%) of patients who had
                             Antifungal            Drug      Treatment
Antifungal treatment         drug        Surgery   toxicity  failure

Amphotericin B               14 (31)      9 (64)    7 (50)   0 (0)
Amphotericin B followed by   10 (22)      7 (70)    3 (30)   1 (10)
  oral antifungal
Itraconazole                 10 (22)      5 (50)    1 (10)   1 (10)
Ketoconazole                  6 (13)      5 (83)    0 (0)    1 (17)
Fluconazole                   1 (2.5)     1 (100)   0 (0)    0 (0)
Treatment not completed (a)   3 (7)       1 (33)    0 (0)    2 (67)
No data (b)                   1 (2.5)     1 (100)   0 (0)    0 (0)
Total                        45 (100)    29 (64)   11 (24)   5 (11)

(a) Treatment not completed in 3 patients: 2 patients died soon after
starting treatment and 1 patient was non-compliant.
(b) Antifungal treatment is unknown in this patient because of
incomplete records.