Case Presentation: A 77-year-old Russian-speaking woman recently recovered from severe COVID-19 infection presented for incision and drainage of a right thumb abscess complicated by synovitis and osteomyelitis. Four months prior, she injured her right thumb after twisting a jar lid. X-rays showed soft tissue swelling with cystic changes and crystals in the scaphoid and radial styloid bones. She was diagnosed with synovitis and prescribed celecoxib for presumed soft tissue injury. After two months of progressive pain and swelling, MRI demonstrated an abscess of the right first interphalangeal (IP) joint with associated cellulitis and osteomyelitis. Incision and drainage yielded cloudy yellow gelatinous fluid with negative gram stain and culture. She was admitted for further operative exploration with biopsy and extended cultures. Laboratories including CBC and CRP were within normal limits on admission. Bone biopsy revealed chronic necrotizing granulomatous inflammation. While awaiting final cultures, she was discharged home on an empiric 6 week course of daptomycin and ceftriaxone. On day 25, acid-fast culture grew mycobacterium tuberculosis (TB). She was transitioned to rifampin, isoniazid, pyrazinamide, and ethambutol (RIPE) therapy in consultation with the local Health Department. X-ray at this time revealed progressive bone lysis and fragmentation of the first proximal phalanx head and subluxation at first IP joint.

Discussion: Osteoarticular TB represents 5% of global TB cases, and of these cases tubercular dactylitis is rare (1). Even in endemic countries, tubercular dactylitis is difficult to diagnose because it presents as the only active site of tubercular infection. Furthermore, clinical and radiographic findings are generally non-specific, making it difficult to differentiate from common monoarticular inflammatory or infectious conditions. Bone biopsy demonstrates characteristic granulomas with epithelioid and Langhans giant cells and associated caseous necrosis. Standard RIPE therapy is effective and should be initiated while waiting for acid-fast culture to result. TB dactylitis occurs in immunosuppressed patients and occasionally after trauma. This case is timely because the osteomyelitis arose 3 to 4 months after the patient received high dose corticosteroids and tocilizumab for severe COVID-19 infection. Her possible TB exposure includes residing in an endemic country and occupational contact while working in a Belarusian microbiology lab. Notably, her T-spot TB test (an Interferon-gamma Release Assay [IGRA]) was negative prior to starting immunosuppression. Though IGRAs have specificity >90% for diagnosis of latent TB infection (2), these screening tests cannot be used to diagnose active TB. In active TB, the diminished cell-mediated immune response may affect the performance of an IGRA test (3). In this case, physicians did not suspect extra-pulmonary TB and this patient was not risk-stratified to airborne plus contact isolation during the abscess incision and drainage.

Conclusions: Hospitalists should consider extra-pulmonary tuberculosis infection in patients with TB risk factors and history of severe COVID-19 infection treated with immunosuppressive therapies. A high index of suspicion for TB is necessary to activate proper isolation protocols on hospital admission. Additional research is needed to determine the performance of IGRA testing in populations at high risk for active TB infection.