Case Presentation: A 77-year-old man from the Dominican Republic was admitted to an affiliate hospital with two months of low back pain. He had a history of hypertension and COVID-19 pneumonia three months prior to this presentation. MRI at that time showed osteomyelitis (OM), and biopsy of the L5-S1 intervertebral space revealed dense fibrous and granulation tissue. Fungal, bacterial, and acid-fast bacilli (AFB) cultures and stains from the biopsy were negative. A QuantiFERON gold was positive. A computed-tomography (CT) chest showed innumerable pulmonary nodules, thought then to represent residua of COVID-19. As there were no systemic signs of infection, and the only symptom remained low back pain, the patient was started on vancomycin and ceftriaxone and discharged with follow-up. He was evaluated in infectious diseases clinic one month later. His back pain had progressed and he was now requiring a walker. The decision was made to directly admit under airborne precautions. Three sputum samples were negative for AFB. MRI of the lumbar spine showed worsening L5-S1 OM with new intervertebral and bilateral psoas abscesses. Guided drainage of the left psoas abscess revealed AFB, and nucleic-acid-amplification-test confirmed Mycobacterium tuberculosis (MTB). The Department of Health was notified, he was started on four-drug standard RIPE therapy, and was discharged home two weeks later. AFB sputum samples subsequently grew Mycobacterium avium complex (MAC), which was attributed to colonization in the setting of disseminated MTB. Infectious diseases declined to treat.
Discussion: We present a case of disseminated MTB infection with spinal OM and paravertebral pelvic abscesses in an immunocompetent male with epidemiologic risk factors. Disseminated MTB, where successful invasion of the bloodstream is followed by hematogenous spread, occurs in up to 2% of all MTB cases. Usually, disseminated MTB is observed in immunodeficient patients and remains difficult to diagnose given its vague presentations. The unremarkable stains and cultures from the initial bone biopsy led to a diagnosis of vertebral OM and subsequent administration of broad-spectrum antibiotics. Diagnostic momentum allowed the pulmonary nodules to be explained by COVID-19. In this case, the lack of growth from the original bone biopsy was a clue that further investigation was warranted. The differential diagnosis for OM on imaging not responsive to appropriate antibiotics includes fungal infections, lymphoma, Ewing and Kaposi’s sarcoma, multiple myeloma, degenerative disc disease, infiltrative diseases such as sarcoidosis, and tuberculosis. Based on history, the patient had no known tuberculosis exposures and had no symptoms aside from back pain. Studies revealed no reason for any immunocompromise. The onset of back pain was temporally associated with COVID-19 infection, but causal association remains unknown. We were surprised only MAC grew from the lungs, a finding confirmed three times. On last follow-up, he no longer requires a walker.
Conclusions: Disseminated MTB remains a rare yet important etiology when vague signs or symptoms are present across different organ systems. While normally associated with immunocompromise, our case demonstrates that a healthy person with some risk factors can develop disseminated MTB, a diagnosis which should be considered in cases where presumed osteomyelitis doesn’t clear after appropriate treatment.