CASE REPORT

Top Abstract Case Report References

In February 1998, a 52-year-old female with a 4-year history of dermatomyositis presented to the emergency department of the St-Boniface General Hospital, Winnipeg, Canada, with pain and swelling of the right fifth digit and wrist. Early in the course of her disease, she had been treated with azathioprine (Imuran) and chloroquine (Plaquenil). These medications had been discontinued due to side effects. She worked as a bank teller until 1996, and was presently unemployed, living on a farm. Autoimmune tenosynovitis was diagnosed, and she was started on prednisone (20 mg a day). In July of the same year, she developed an ulcer on the fifth digit, which was swabbed and cultured. There was no history of trauma. Gram stain, fungal stain, viral cultures, fungal cultures, and aerobic and anaerobic bacterial cultures were all negative. An X ray of the hand was normal. The white blood cell count was 11.8 × 10⁹/liter, with 89% neutrophils. She was initially started on metronidazole and cefazolin, but there was poor clinical response to this therapy. A month later, she developed ulcerative subcutaneous nodules, and on examination, the patient had evidence of tenosynovitis of the right fifth digit and wrist, chronic induration of the wrist, and subcutaneous white nodules along the volar aspect of the forearm. There was nontender axillary adenopathy. Surgical debridement of the right palm and wrist was performed. Histopathology of this tissue revealed caseating granulomas. Cultures of the drainage were negative for bacterial and fungal culture. Acid-fast bacillus smears were positive, and she was started on isoniazid, rifampin, ethambutol, and pyrazinamide for presumptive M. tuberculosis infection. It subsequently grew a pure culture of acid-fast nontuberculous mycobacteria. The provincial Mycobacteriology Laboratory at the Health Sciences Centre, Winnipeg, Canada, identified the isolate as M. branderi. Following definitive identification, the antituberculous drugs were discontinued and the patient received empiric antimicrobial therapy with clarithromycin (1,000 mg twice a day [b.i.d.]) and trimethoprim-sulfamethoxazole (TMP-SMX) (one double-strength tablet b.i.d.). Two weeks later, the patient continued to have yellow, odorless discharge from the incision sites on the flexor aspect of the right palm and the right wrist. There were persistent erythema and induration surrounding the draining sinuses on the wrist. Ciprofloxacin (750 mg b.i.d.) was added to the treatment regimen, and the clarithromycin dose was decreased to 500 mg b.i.d. Six weeks later, there had been a dramatic improvement in her right hand and wrist, with resolution of the draining sinuses in the right palm and a significant decrease in the open area in the right wrist. The induration was mostly resolved, although she continued to have some erythema surrounding a 2.5-by-4-cm ulcer in the right wrist area, which was gradually resolving. She received a total of 19 months of antibiotics.

In January of 1999, an additional case of M. branderi was detected from a 74-year-old female with shortness of breath together with back and chest pain who was referred to a respiratory clinic at the Health Sciences Centre, Winnipeg. Chest X ray revealed middle lobe collapse and peripheral pneumonic infiltrates. Clinical symptoms persisted after empiric ciprofloxacin treatment for 2 weeks. Three sputum samples were submitted to the same laboratory where the M. branderi isolate from the first case were identified. One of the three submitted sputum cultures grew M. avium complex. A bronchoalveolar lavage specimen was also submitted, which was negative for routine bacteriology but grew a pure culture of acid-fast nontuberculous mycobacteria, subsequently identified as M. branderi. One year later, the patient remained symptomatic and the underlying cause of disease remained unclear.

M. branderi is a newly described species of mycobacterium (12), and its role as a human pathogen is not well defined. The previously described isolates of M. branderi were respiratory tract isolates obtained from nine patients, some of whom had cavitary mycobacteriosis of the lungs (12). Repeat samples presented M. branderi as the only cultivable organism, suggesting its potential pathogenic role in humans (12).

Clinical specimens consisting of hand drainage from the first patient and bronchoalveolar lavage fluid from the second patient were submitted for mycobacteriology culture. Middlebrook 12B liquid medium (Becton-Dickinson, Sparks, Md.) grew acid-fast bacilli that were subcultured onto Middlebrook 7H10 agar. Each specimen grew a pure culture of M. branderi. On Middlebrook 7H10 agar medium, each clinical isolate of M. branderi showed two colony types: one white and the other opaque. Both colony types were nonchromogenic, raised, smooth edged, and domed. The Kinyoun acid fast stain demonstrated pleomorphic, beaded, slightly curved acid-fast bacilli. The AccuProbe test (Gen-Probe, San Diego, Calif.) was performed according to manufacturer's instruction and was negative for M. tuberculosis complex and M. avium complex.

Biochemical testing of the specimens was performed by conventional methods as previously described (11, 13, 16) and gave identical results for the two organisms isolated (Table 1). Both showed growth from 25 to 42°C and were nonchromogenic. Niacin, nitrate reductase, Tween 80 hydrolysis, urease, tellurite reduction, and iron uptake tests were all negative; as well, there was no acid production from mannitol, sorbitol, and inositol. Both organisms could not utilize sodium citrate as a sole source of carbon. Both organisms were positive for heat-stable catalase, arylsulfatase activity, and pyrazinamidase.

Mycolic acid analysis by high-performance liquid chromatography (HPLC) was performed according to the standardized method (5). The HPLC analysis of mycolic acids of both isolates produced a chromatographic pattern very similar to the patterns produced by M. celatum (6), with the same retention times, but with thicker-looking peaks, different peak height ratios, and less separation within the peaks of the second cluster. Both species have HPLC chromatographic patterns occurring as double clusters not unlike those of M. xenopi and the M. avium complex. The patterns of both isolates also showed additional small peaks at the front of the second cluster.

Sequence-based species identification using the 16S rRNA gene (14) was performed for both specimens. With the first specimen, both colony types were sequenced to confirm culture purity. Sequencing reactions were performed with the ABI PRISM BigDye Terminator Cycle Sequencing Ready Reaction kit (PE Biosystems, Foster City, Calif.) and run on an ABI PRISM 310 Genetic Analyzer (PE Biosystems) according to the manufacturer's instructions. Resulting sequences were assembled and analyzed using Lasergene software (DNASTAR, Inc. Madison, Wis.), resulting in a 1,458-bp fragment of the 16S rRNA gene, equivalent to positions 28 to 1490 of the Escherichia coli 16S rRNA gene. The sequences of both organisms isolated, which were identical, showed highest percent similarity (99.7%) to that of M. branderi ATCC 51789 (GenBank accession no. X82234).

BACTEC 12B radiometric broth macrodilution sensitivity testing was performed on the clinical isolates according to the method used for M. avium complex strains (8, 15). The following drugs were tested, and their MIC results are indicated in Table 2: amikacin, capreomycin, clarithromycin, clofazamine, ciprofloxacin, ethambutol, ethionamide, kanamycin, ofloxacin, rifabutin, rifampin, sparfloxacin, streptomycin, and thiacetazone.

Discussion. Nontuberculosis mycobacterium (NTM) species are becoming increasingly important in the clinical setting, causing nosocomial outbreaks or pseudo-outbreaks; pulmonary disease; lymphadenitis; skin, soft tissue, or skeletal infections; and AIDS-related and -nonrelated disseminated infections, among others (1). Although it is generally believed that the environment is the source of most NTM infections, their pathogenesis remains irresolute and a continuous provision of studies regarding NTM-related infections is required for further knowledge and understanding. This particular study describes two cases implicating M. branderi.

View larger version (4K): [in this window] [in a new window]   FIG. 1.   Clarification of discrepancies and ambiguity detected with the only available nucleotide entry of the 16S rRNA gene sequence of M. branderi in the GenBank database. The top row shows the sequence of ATCC 51789 (accession no. X82234); the bottom row shows the sequence obtained for M. branderi ATCC 51789, ATCC 51788, isolate 1, and isolate 2 from our institution. E. coli positions within 16S rRNA gene are nucleotides 1016 to 1024 (A), nucleotides 1141 to 1149 (B), and nucleotides 1163 to 1171 (C). Dashes indicate identical nucleotides.