The Brief Case: Delayed Diagnosis of Cutaneous Mycobacterial Infection in Rural Mississippi

CASE

An 85-year-old Caucasian man with no significant past medical history presented to an outside dermatology clinic with a 6-month history of rash on his hands and feet. The patient worked as a farmer for 30 years in rural Mississippi and retired 30 years prior. He denied any known trauma or penetrating injury, exotic exposures, or significant travel except to South Korea 60 years ago. Skin biopsy was performed on the right dorsal hand and revealed granulomatous inflammation on hematoxylin and eosin stain (HE stain). Acid-fast bacillus (AFB) stain was positive for numerous bacilli on both histopathology and on tissue submitted to the clinical microbiology laboratory. However, mycobacterial cultures demonstrated no growth after 8 weeks of incubation. The skin lesions continued to progress despite 6 weeks of clarithromycin, which was stopped due to diarrhea, followed by 4 weeks of ciprofloxacin and minocycline. The lesions were associated with significant pain and swelling, prompting him to present to our medical center for further evaluation.

On physical examination, the patient was in mild distress but afebrile, alert, and oriented. Skin examination revealed tender, erythematous plaques and diffuse edema on bilateral hands extending to the elbows (Fig. 1A) and on feet extending below the knees (Fig. 1B). C-reactive protein was 16.3 mg/dL (reference range, 0 to 0.50 mg/dL), the erythrocyte sedimentation rate was 54 mm/h (reference range, 0 to 29 mm/h), and the white blood cell count was 5.6 cells/μL (reference range, 4.0 to 10.0 cells/μL). A tuberculosis interferon gamma release assay was negative. A computed tomography scan of the chest, abdomen, and pelvis did not show any abnormalities. A repeat skin biopsy was performed on the right forearm. Histopathology showed extensive granulomatous inflammation on HE stain (Fig. 2A), with Ziehl-Neelsen stain showing clusters of acid-fast bacilli throughout the dermis (Fig. 2B). There was no neural tissue involvement. In the microbiology laboratory, the specimen was positive for AFB by auramine-rhodamine fluorochrome stain, and a subsequent Kinyoun stain demonstrated >9 AFB/high-power field (Fig. 2C and D). Biopsy material was inoculated onto various media, including broth (Middlebrook 7H9; MGIT) and a Lowenstein-Jensen (LJ) slant, both incubated at 37°C; additional LJ and chocolate slants were incubated at 30°C to optimize growth of some fastidious cutaneous mycobacteria. Differentials included Mycobacterium abscessus, Mycobacterium chelonae, Mycobacterium haemophilum, and Mycobacterium marinum as potential mycobacteria responsible for causing cutaneous infection.

FIG 1.

(A and B) Erythematous plaques and diffuse edema on bilateral hands (A) and feet (B) on presentation; (C and D) significant improvement in erythema and edema on bilateral hands (C) and feet (D) noted 3 months into treatment.

FIG 2.

(A) Skin biopsy specimen, hematoxylin and eosin stain, ×100 magnification. Shown are superficial and deep dermis with diffuse granulomatous inflammation with multinucleated giant cells and histiocytes with pale cytoplasm admixed with lymphocytes. (B) Skin biopsy specimen, Ziehl-Neelsen stain, ×400 magnification. Shown are abundant acid-fast bacilli, many in small clusters. (C and D) Skin biopsy specimen (ground tissue), Kinyoun stain, ×1,000 magnification. Shown are numerous acid-fast bacilli.

Curiously, no growth was seen on mycobacterial cultures after 6 weeks. Since the skin biopsy specimen showed abundant amount of organism, the formalin-fixed paraffin-embedded tissue block was sent for molecular identification to a reference laboratory (Molecular Microbiology Laboratory, Department of Laboratory Medicine and Pathology, University of Washington Medical Center, Seattle, WA, USA). A mycobacterial rpoB primer set yielded an amplicon that had a 259-bp alignment with 100% identity to multiple Mycobacterium leprae sequences in the NCBI database (1). Mycobacterium lepromatosis sequences had only 96% identity. The diagnosis of borderline lepromatous leprosy with reversal reaction was established. Working with the National Hansen’s Disease Program (NHDP), prednisone and methotrexate were started for treatment of reversal reaction, and 10 days later, rifampin, minocycline, and moxifloxacin were added for treatment of leprosy. Three months into therapy, there was a significant improvement in patient’s debilitating symptoms and skin examination (Fig. 1C and D), and he gradually returned to his baseline functional status.

DISCUSSION

Leprosy, or Hansen’s disease, caused by Mycobacterium leprae or Mycobacterium lepromatosis, remains an important neglected tropical disease in many parts of the world. According to the World Health Organization (WHO), 127,558 new cases of leprosy were diagnosed globally in 2020, with the highest numbers of cases reported in India, Brazil, and Indonesia (2, 3). One hundred fifty-nine new cases of leprosy were diagnosed in the United States in 2020, the majority (69%) of which were reported in Florida, California, Louisiana, Hawaii, New York, and Texas (4). M. leprae is a Gram-positive, acid-fast, intracellular pathogen that most frequently invades Schwann cells in peripheral nerves and macrophages in the skin (5). It was the first bacterium found to cause disease in humans, and M. leprae DNA has been detected from skeletal remains from over 4,000 years ago (5). Leprosy is often underdiagnosed due to several factors, including chronicity of the disease, inability to grow on routine media, fear of stigma, and failure of clinicians to recognize the disease. Nine-banded armadillos (Dasypus novemcinctus) serve as a zoonotic reservoir for human infection and are considered responsible for some transmission of indigenous cases of leprosy in the United States as the disease has been found to infect armadillos throughout the southeastern region of the country (6). An estimated 95% of the world’s population is not genetically susceptible to the disease; however, age, genetic factors, and ancestral exposure to the bacillus can lead to certain variations (7). Our patient was elderly, had no travel to countries with high rates of leprosy (3), and was not aware of any direct physical contact with armadillos but possibly had some indirect exposure given his occupation as a farmer for several decades in the southeastern United States.

The Ridley and Jopling classification system divides leprosy, based on a spectrum of cell-mediated responses to the disease, into five subtypes: these include, from the most robust to the least immune response, tuberculoid, borderline tuberculoid, mid-borderline, borderline lepromatous, and lepromatous. A simpler WHO classification system exists that divides leprosy, based on number of lesions, into paucibacillary (5 lesions or less) and multibacillary (more than 5 lesions) leprosy and can be helpful for determining treatment duration (5, 7).

M. leprae is a weakly acid-fast organism and may be negative by routine AFB stains, such as Kinyoun and Ziehl-Neelsen. If Hansen’s disease is suspected clinically, notification of the microbiology and histopathology laboratories is helpful, so a modified acid-fast stain, such as Fite-Faraco, can be used, as this is the stain of choice (8). The Fite-Faraco staining method involves adding groundnut oil (e.g., peanut) to xylene as the deparaffinizing agent, which helps the carbol fuchsin dye bind to mycolic acids. Some methods also use a weaker decolorizer, such as 1% sulfuric acid or acid alcohol (9). A major clue that M. leprae is the causative pathogen is failure to grow in culture, even on media and conditions optimized for cutaneous mycobacterial pathogens. Although M. leprae does not grow on artificial solid or liquid culture media, it has been propagated in the research lab environment in armadillos and gene knockout mice, including on mouse footpads (first described by Shepard in 1960) (8). Slit skin smears or histopathology of a skin biopsy is most helpful to detect M. leprae, and the infection is most pronounced at the border of the lesion (9). HE staining is typically necessary for characterizing type of immune infiltrate and involvement of cutaneous nerves. Most full-thickness skin biopsy specimens note granulomatous infiltrate, with features differing depending on the subtype of leprosy. Histopathology of tuberculoid infections shows well-formed granulomas with epithelioid cells, nerve infiltration, and few bacilli. On the lepromatous side of the spectrum, histopathology notes disorganized granulomas and foamy macrophages with many bacilli, also known as lepra cells, which appear as circular aggregates of bacilli on AFB stains. The borderline forms may have features of both tuberculoid and lepromatous leprosy. The presence of AFB within a nerve is diagnostic of leprosy (10). Histopathology in our patient’s case showed poorly defined, granulomatous inflammation with evidence of high bacillary load, suggestive of disease on the lepromatous side of the spectrum.

As mentioned previously, M. leprae is underdiagnosed, and in our case, there were several factors that initially led us away from identification of M. leprae. First, the skin biopsy specimen did not demonstrate nerve inflammation or findings of bacilli within a nerve, which is diagnostic of M. leprae. Furthermore, the organisms stained robustly with Kinyoun and Ziehl-Neelsen stains, which is not typical for M. leprae. Our cultures also failed to grow mycobacteria twice. This inability to isolate on routine cultures, on the other hand, is a characteristic typical of M. leprae that often leads to difficulty in diagnosis. In this case, there was suspicion for mycobacteria given abundant organisms noted on AFB stains, and the tissue block was sent for molecular identification by sequencing at a reference lab. The lab used for this case reports using a multilocus approach to nontuberculous mycobacteria and ultimately a primer set against the rpoB gene that successfully identified the species (1).

Clinical manifestations of leprosy primarily include skin lesions and nerve dysfunction—particularly loss of sensation. However, musculoskeletal involvement, including inflammatory arthritis, though underreported, is quite common and has been identified as the third most common manifestation of Hansen’s disease (11). Immunologic reactions in leprosy can occur during treatment or spontaneously during the disease course and are associated with increased morbidity. There are two types of immune reactions. The type 1 reaction, or reversal reaction, is a delayed-type hypersensitivity reaction associated with increased erythema of lesions, edema of the hands and feet, neuritis, and nerve impairment, and most often occurs in borderline forms of leprosy. The type 2 reaction, or erythema nodosum leprosum, is understood to be an immune complex-mediated reaction causing systemic symptoms, including fevers with eruption of multiple, painful subcutaneous nodules, and tends to occur in borderline or lepromatous leprosy (12). Arthritis secondary to immunologic reactions is symmetrical, affecting small joints of the hands and feet, leading to misdiagnosis as various rheumatologic disorders (11). Our patient’s overall clinical presentation, including increased pain and edema in his extremities, was felt to be consistent with immunologic reaction-associated arthritis and neuritis. The histopathologic findings, including high bacillary load, lymphocytes, and disorganized granulomas, favored disease on the lepromatous side of the spectrum. Management was aimed at controlling the inflammatory component of the disease using immunomodulation first followed by antimycobacterial therapy.

Due to the highly varied presentation of leprosy clinically and histologically, a diagnosis of leprosy should be suspected in a patient presenting with rash and biopsy findings positive for AFB with no growth on culture. Molecular testing using a multilocus approach to mycobacteria can be used to confirm diagnosis. Symptoms of worsening pain, edema, paralysis, and neuropathy in the extremities should additionally prompt concern for possible immunologic reaction to guide initiation of immunosuppressive regimen.

SELF-ASSESSMENT QUESTIONS

1. What is the stain of choice for M. leprae?

• a.Ziehl-Neelsen stain
• b.Fite-Faraco stain
• c.Kinyoun stain
• d.Hematoxylin and eosin stain

2. How long does M. leprae usually take to grow on routine mycobacterial cultures?

• a.6 weeks
• b.8 weeks
• c.12 weeks
• d.Does not grow on routine mycobacterial cultures

3. What type of immunologic reaction is a reversal reaction?

• a.Immune complex-mediated reaction
• b.Delayed-type hypersensitivity reaction
• c.Cytotoxic reaction
• d.Immediate hypersensitivity reaction