Disseminated Primary Mycobacterium bovis Infection in Patient Receiving Adalimumab Therapy: A Case Report

Abstract

Anti‐TNF‐α therapies are associated with an increased risk of tuberculosis, mainly via the reactivation of latent Mycobacterium tuberculosis. Primary zoonotic Mycobacterium bovis infection is rare and diagnostically challenging, particularly in immunosuppressed patients. We report a case of disseminated Mycobacterium bovis infection in a man receiving long‐term adalimumab for psoriasis, complicated by diagnostic delay, suspected hemophagocytic lymphohistiocytosis, and immune reconstitution inflammatory syndrome (IRIS), resulting in a challenging treatment course. Epidemiological history revealed recent travel to Turkey and ingestion of unpasteurized dairy products. This case highlights the need for ongoing exposure assessment, sustained clinical awareness, and individualized therapeutic strategies throughout immunomodulatory treatment.

1. Introduction

Mycobacterium bovis is the most common cause of zoonotic tuberculosis (TB) in humans, usually linked to the consumption of unpasteurized dairy products. Its intrinsic resistance to pyrazinamide and frequent extrapulmonary involvement can delay both diagnosis and appropriate treatment. Patients receiving tumor necrosis factor‐alpha (TNF‐α) inhibitors such as adalimumab are particularly vulnerable to mycobacterial infections due to impaired cell‐mediated immunity. Reactivation of Mycobacterium tuberculosis infection is the most commonly reported mechanism, whereas primary zoonotic Mycobacterium bovis infection during long‐term immunosuppressive therapy is rarely described. This case describes a disseminated primary Mycobacterium bovis infection in a patient treated with adalimumab and highlights the importance of thorough and ongoing exposure assessment, awareness of atypical mycobacterial infections, sustained clinical suspicion, and individualized management in immunocompromised patients throughout the course of immunomodulatory therapy.

2. Case

2.1. Case History and Examination

A male in his mid‐fifties originally from Kayseri, Turkey, living in Denmark since the 1980s, had a history of psoriasis for 14 years and chronic lower back pain but, otherwise, healthy. His psoriasis had been treated with adalimumab (40 mg every two weeks) and local therapies (Daibovet 50 + 500 μg) for the past 14 years. Before starting adalimumab, he tested positive for TB on a Mantoux test, which was attributed to prior BCG vaccination, and no prophylactic treatment was initiated.

The patient was admitted to the emergency department at the end of 2022 after experiencing one month of febrile episodes, mild weight loss (3‐4 kg), and fatigue. Initial blood tests revealed an elevated erythrocyte sedimentation rate (65 mm), CRP (65 mg/L), and LDH (408 U/L), with no notable physical findings. He was discharged with an outpatient follow‐up as the symptoms were thought to be related to an underlying malignancy. A CT scan of the thorax, abdomen, and pelvis revealed small bilateral lung and pleural opacities, confluent mesenteric lymphadenopathy, and mesenteric fat stranding. A PET/CT scan showed further enlargement of lymph nodes in the mesentery, mediastinum, and left neck, along with significant FDG uptake in the left tonsil and terminal ileum (Figure 1A).

FIGURE 1.

(A + B): Fluorine‐18‐flourodeoxyglucose (¹⁸F‐FDG) PET/CT from vertex cranii to proximal femora. Primary scan (A) shows intense FDG uptake in the left tonsil (yellow arrow). Lymph nodes on the neck and mediastinum exhibit moderate to intense FDG uptake (blue arrows). Intense FDG uptake in multiple small and enlarged lymph nodes located in the mesentery and an area stretching toward right iliac fossa with an intense focal uptake in terminal ileum (purple arrows). We see inhomogeneous, moderately intense FDG uptake in a segment of the small intestine situated in the pelvis (green arrows), seen with wall thickening on CT. Follow‐up scan 2 months later (B) shows stationary conditions with moderate to intense FDG uptake in the lymph nodes of the neck, mediastinum, and the abdominal region as well as unchanged intense FDG uptake in the segment of the small intestine located in the pelvis. Accentuated splenic FDG uptake, which is now diffuse and homogeneous compatible with a longer lasting infection. There is no longer uptake in the left tonsil due to tonsillectomy.

2.2. Diagnosis, Investigations, and Treatment

On Day 20, repeat blood tests showed a severe increase in CRP (287 mg/L), LDH (1360 U/L), and anemia (hemoglobin 6.3 g/dL), and the patient experience worsening of symptoms and persisting fever, leading to readmission. Adalimumab was discontinued upon admission (Day 20) due to suspected bacterial infection. Despite broad‐spectrum antibiotics (initially piperacillin/tazobactam 4 g/6/h and later meropenem 1 g/8/h and metronidazole 500 mg/8/h), fever persisted. Tests revealed severely increased ferritin > 16,500 ng/mL, hypertriglyceridemia (4.5 mmol/L), and IL‐2 receptor (16,000 kU/L), raising suspicion for underlying lymphoma, hemophagocytic lymphohistiocytosis (HLH), or sarcoidosis. Blood cultures, viral tests, and serology (CMV, HIV, EBV, hepatitis B/C, and Coxiella burnetii) were negative as was rheumatic factor and ANA. ACE was slightly elevated (72 U/L).

Without a definitive diagnosis, tonsillectomy and colonoscopy were performed. Histopathology of the terminal ileum revealed large nodular granulomas with central necrosis, surrounded by a rim of histiocytes and lymphocytes, a characteristic of caseating granulomas, which are a hallmark of mycobacterial disease (Figure 2). Following this finding, Ziehl–Neelsen staining was performed—a technique used to demonstrate acid‐fast bacteria such as Mycobacterium species. The result was positive. Laboratory investigations revealed a positive interferon gamma release array (IGRA) test, and PCR analysis of tracheal secretions, urine, and tonsillar samples confirmed infection with Mycobacterium tuberculosis complex. Antituberculous therapy was initiated with Rimstar 5 tablets (275‐mg ethambutol hydrochloride, 75‐mg isoniazid (later 225 mg), 400‐mg pyrazinamide, and 150‐mg rifampicin) alongside corticosteroids due to systemic inflammation (prednisolone 20 mg daily initially and increased to 30 mg after only 4 days due to limited clinical effect). Symptoms improved, with resolution of fever, indicating treatment response. Thirty‐seven days after the initial confirmatory PCR, Mycobacterium bovis was identified in all submitted specimens. A detailed exposure history obtained from the patient’s relatives revealed travel to Kayseri, Turkey, 1.5 years prior to symptom onset, as well as consumption of soft cheese brought home from Kayseri 7 months before symptom onset.

FIGURE 2.

(A, B, C): Histological examination of biopsy from terminal ileum revealed submucosal nodular granulomatous structures with necrotic tissue; hematoxylin and eosin stain (A). CD68 immunostaining for histiocytes showed numerous histiocytes; 6 × magnification (B). Ziehl‐Neelsen stain showed a small amount of positive acid‐fast bacilli within the granulomatous foci (C).

At the time of PCR confirmation of Mycobacterium tuberculosis complex, the patient had developed pancytopenia. To exclude an underlying hematologic malignancy and assess complications such as HLH, a bone marrow aspiration was performed. This revealed no evidence of malignant hematologic disease but showed activated macrophages with phagocytosed erythrocytes. These findings alongside persisting fever, pancytopenia, increased ferritin, hypertriglyceridemia, and soluble IL‐2 supported the suspicion of secondary HLH complicating an established disseminated Mycobacterium bovis infections (Figure 3A, B).

FIGURE 3.

(A + B): Bone marrow biopsy with large macrophages containing engulfed erythrocytes consistent with hemophagocytosis (chloroacetate esterase stain) (A). Immunohistochemistry for CD163 on bone marrow biopsy highlighting numerous activated macrophages (CD163) (B).

By Day 39 of treatment, the patient experienced recurrence of fever and malaise and a worsening of his known psoriasis after a minor prednisolone taper from 25 mg to 20. Given the confirmed Mycobacterium bovis infection, pyrazinamide was discontinued and the regimen therefore adjusted to isoniazid, rifampicin, and ethambutol. Persistent symptoms by Day 50 led to suspicion of immune reconstitution inflammatory syndrome (IRIS), readmission, and an increase in prednisolone to 30 mg daily. A new PET/CT scan showed consistent but not increased FDG uptake in previously affected regions, supporting the IRIS diagnosis rather than treatment failure (Figure 1B). On admission, laboratory analysis revealed hyperglycemia (34.8 mmol/L) but absence of ketonemia. Although late‐onset Type 1 diabetes mellitus (DM1) could not be entirely excluded, the pronounced hyperglycemic state was likely secondary to prolonged high‐dose corticosteroid therapy. Relevant treatments were effectuated, and the patient improved clinically and was discharged.

After 2 months of treatment (56 days), the patient was switched to continuation therapy with isoniazid and rifampicin; however, elevated liver enzymes by Day 75 required switching to moxifloxacin and rifampicin. By Day 103, worsening of the patient’s psoriasis and presumed corticosteroid‐induced DM led to initiating TNF‐α inhibition (Etanercept) under tapered prednisolone coverage—improving patient’s skin symptoms, glycemic control, and inflammatory markers at 4‐month follow‐up.

A summary of the clinical course until reintroduction of etanercept is presented in Figure 4. As the patient remained clinically stable without evidence of relapse during subsequent follow‐up after immunomodulatory therapy was reintroduced (initially etanercept later adalimumab), this period was not included in the timeline.

FIGURE 4.

Timeline of clinical course from first medical evaluation until the reintroduction of etanercept. Symptoms had been present for approximately 1 month prior to the start of the illustrated clinical course. The timeline includes key diagnostic investigations, treatment interventions, inflammatory markers (CRP and ferritin), body temperature, and major clinical events. Day 0 represents initiation of antimycobacterial therapy. Prednisolone dosing is shown throughout the treatment course, illustrating the relationship between corticosteroid tapering and recurrence of inflammatory activity. (1) Initial CT (thorax, abdomen, and pelvis) showing bilateral pulmonary opacities, mesenteric lymphadenopathy, and mesenteric fat stranding. (2) PET/CT showing mesenteric, mediastinal, and cervical lymphadenopathy with increased FDG uptake including left tonsil and terminal ileum (Figure 1A). (3) Colonoscopy with biopsies. (4) Tonsillectomy and positive IGRA. (5) Positive Mycobacterium tuberculosis complex PCR in tracheal secretions, urine, and tonsillar biopsy; bone marrow aspiration as part of HLH or lymphoma evaluation. (6) HLH diagnosis. (7) Culture confirmation of Mycobacterium bovis from tracheal secretions. (8) Culture confirmation of Mycobacterium bovis from urine. Clinical deterioration with fatigue. (9) Mycobacterium bovis identified in all submitted specimens. (10) PET/CT findings consistent with IRIS.

2.3. Outcome and Follow‐Up

After 10 months, antimycobacterial therapy was terminated, with the patient remaining biochemically stable, mild psoriasis activity, and controlled blood glucose levels with no need of continuing antidiabetic medication, supporting a corticosteroid‐induced DM rather than DM1. Five months after completion of treatment, psoriasis worsened, necessitating the reintroduction of adalimumab. One year into immune therapy (initially with etanercept and later adalimumab), no relapse of Mycobacterium bovis infection was observed. However, due to the prolonged disease, the patient remained unable to return to his previous physically demanding job.

3. Discussion

TB remains one of the leading causes of death globally, despite extensive efforts to control its transmission and treat affected individuals [1]. Mycobacterium bovis, a zoonotic member of the Mycobacterium tuberculosis complex, poses distinct diagnostic and therapeutic challenges. Radiologically, Mycobacterium bovis is indistinguishable from Mycobacterium tuberculosis; however, Mycobacterium bovis shows a predilection for extrapulmonary manifestations [2–4] possibly reflecting transmission through ingestion of contaminated unpasteurized dairy products rather than inhalation of infectious aerosols [2, 5]. Furthermore, Mycobacterium bovis is intrinsically resistant to pyrazinamide, a standard first‐line component of Mycobacterium tuberculosis treatment, complicating treatment and requiring regimen adjustments once the species is identified [4]. Cases of dual infections have been previously reported, supporting delayed therapeutic adjustment until laboratory confirmation that Mycobacterium bovis was the sole infecting mycobacterial species [2].

Diagnostic delay of extrapulmonary Mycobacterium tuberculosis remains a problem in both low and high endemic settings with additional barriers among migrant populations [6–8] and is multifactorial. A Danish study by Holden et al. [9] highlights that patient delays are notably longer in people of non‐Danish origin compared to those of Danish origin when diagnosed with Mycobacterium tuberculosis. Such delays, which have also been reported for other mycobacterial infections [10], may result in more advanced disease at the time of treatment initiation and should be taken into consideration when evaluating patients without a clear diagnosis. In the present case, extensive lymphadenopathy and gastrointestinal involvement initially mimicked malignancy or sarcoidosis, contributing to diagnostic delay. The final diagnosis of zoonotic infection with Mycobacterium bovis relied on the histopathological findings, positive Ziehl–Neelsen staining, confirmatory PCR identification with Mycobacterium bovis, detailed exposure history, and consumption of unpasteurized dairy products. In this case, the patient had a positive IGRA test which was compatible with the final diagnosis of Mycobacterium bovis infection as current IGRA assay detects antigens expressed by members of the Mycobacterium tuberculosis complex, including Mycobacterium bovis. Notably, IGRA testing cannot distinguish between Mycobacterium tuberculosis and Mycobacterium bovis [11]. However, IGRA results should be interpreted with caution in immunocompromised patients since a false‐negative result can occur in individuals receiving immunosuppressive agents such as TNF‐α inhibitors as well as in patients with severe immunosuppression, such as those with low peripheral lymphocyte levels or HIV infection [12, 13]. Therefore, a negative result cannot rule out active infection caused by members of the Mycobacterium tuberculosis complex.

Immunosuppressive agents such as adalimumab are widely used [14]. These agents are well recognized to increase susceptibility to mycobacterial infections by impairing granuloma formation and intracellular containment of mycobacteria [14–17]. Reactivation of Mycobacterium tuberculosis often occurs within 15–18.5 months of initiating TNF‐α inhibitors [18]. Similarly, the reactivation of Mycobacterium bovis infection has only been described in a few cases, all of which developed symptoms within a few years after initiation of immunomodulatory therapy, and were interpreted as reactivation because the patients had a history of prior exposure but no evidence of any recent or newly acquired exposure [19–21]. Although reactivation cannot be completely excluded in this case, the patient developed disseminated disease after more than a decade of continuous adalimumab therapy, making recent primary infection more likely than reactivation. Furthermore, a plausible recent exposure history was identified through the consumption of unpasteurized dairy products imported from an endemic region several months prior to symptom onset. This interpretation was further supported by microbiological isolates linked to Turkish strains [22]. This underlines the importance of not only performing TB risk assessment at treatment initiation but also revisiting it throughout long‐term immunosuppressive therapy and particularly following new exposures.

This case was further complicated by secondary HLH prior to the initiation of antituberculous therapy, a rare but severe hyperinflammatory syndrome associated with malignancies, autoimmune diseases, and severe infections, including TB, particularly extrapulmonary disease [23]. The presence of 5 of the 7 diagnostic criteria for HLH (persisting fever, pancytopenia, extreme hyperferritinemia, hypertriglyceridemia, and hemophagocytosis in the bone marrow) confirmed the diagnosis of HLH [24]. Secondary HLH due to Mycobacterium tuberculosis is associated with high mortality, and early initiation of antimicrobial therapy is, therefore, linked to improved survival in these patients [23].

A known paradoxical reaction when initiating antituberculous treatment regimens, particularly in patients with extrapulmonary disease, is IRIS [25]. Reduction of the bacterial load leads to the release of antigens into the surrounding tissues, resulting in an increased immune response and cytokine storm, all of which may mimic treatment failure or disease progression. Distinguishing IRIS from treatment failure or disease progression is challenging but critical, as management strategies differ substantially. In this case, stable imaging findings on repeat PET/CT and the clinical response to increased corticosteroid dosing supported the diagnosis of IRIS. Reintroduction of TNF‐α inhibitors to manage IRIS, psoriasis, or similar conditions requiring immunomodulating therapies remains controversial, as no formal guidelines exist. Consequently, decisions regarding reintroduction are largely based on expert opinion and must be individualized [26]. The favorable outcome in this case suggests that, with close monitoring and adjunctive corticosteroid coverage, immunomodulating agents such as etanercept or adalimumab can be safely resumed in selected cases. However, further research is needed to establish evidence‐based protocols for the management of these complex infections.

In conclusion, this case highlights the need to consider atypical infections, such as Mycobacterium bovis, in patients receiving long‐term immunosuppressive therapy. Although Mycobacterium bovis is rarely encountered in routine diagnostics in high‐income settings, this case illustrates that it may cause disseminated infection in immunocompromised patients with relevant exposure histories and underscores the importance of repeated assessment of exposure risks, as ongoing or newly acquired exposures during treatment may influence disease progression. Furthermore, this case demonstrates the need for continued vigilance regarding rare but severe complications, such as HLH and IRIS, throughout the entire treatment course.

Nomenclature

TNF‐α

Tumor necrosis factor‐alpha inhibitor

IRIS

Immune reconstitution inflammatory syndrome

HLH

Hemophagocytic lymphohistiocytosis

TB

Tuberculosis

PET/CT

Positron emission tomography–computed tomography

FDG

Fluorodeoxyglucose

CRP

C‐reactive protein

LDH

Lactate dehydrogenase

IL‐2

Interleukin‐2

ANA

Antinuclear antibody

ACE

Angiotensin‐converting enzyme inhibitors

CMV

Cytomegalovirus

EBV

Epstein–Barr virus

IGRA

Interferon‐gamma release assay

DM

Diabetes mellitus

Author Contributions

Michele Bjerre Kirk drafted the manuscript, including timeline and performed literature review. Frauke Rudolf, MD, PhD, contributed to patient care, collected and interpreted clinical and laboratory data, critically revised the manuscript, and gave final approval. Kristina Høeg Rasmussen, MD, reviewed histopathological slides of terminal ileum and drafted the pathological description of this manuscript and reviewed and approved the manuscript. Andreas Gotthelf, MD, contributed with the PET images and their descriptions for the manuscript and reviewed and approved the manuscript. Lisbeth Lund Jensen, MD, PhD, initially reviewed the histopathological bone‐marrow slides and drafted descriptions of these and reviewed and approved the manuscript.

Funding

The authors received no specific funding for this work.

Consent

Written informed consent was obtained from the patient for publication of this case report and any accompanying images.

Conflicts of Interest

The authors declare no conflicts of interest.

Kirk, Michele Bjerre , Rasmussen, Kristina Høeg , Gotthelf, Andreas , Jensen, Lisbeth Lund , Rudolf, Frauke , Disseminated Primary Mycobacterium bovis Infection in Patient Receiving Adalimumab Therapy: A Case Report, Case Reports in Infectious Diseases, 2026, 2363582, 7 pages, 2026. 10.1155/crdi/2363582

Data Availability Statement

All relevant data are included within the article.