Reply

We appreciate Dr Campos and colleagues’ interest in the article “Clinical Presentation and Course of Pulmonary Involvement in Chronic Nonbacterial Osteomyelitis”1 and the critical, helpful discussion on the spectrum of differential diagnoses. We very much agree with the authors’ suggestion that possible differential diagnoses of pulmonary infiltrates in patients with chronic nonbacterial osteomyelitis (CNO) vary between geographic regions, the socioeconomic environment of patients and their families, medication intake, comorbidities, and other factors.2, 3 Indeed, pulmonary histoplasmosis may be an important differential diagnosis in individual cases in countries where Histoplasma species are endemic (such as the here reported) or in patients with a travel history to these areas, especially when receiving immunomodulating treatments.2, 4, 5 As the authors point out correctly, pulmonary nodules caused by Histoplasma species typically present as small, defined lesions that may appear solitary or multifocal. Calcifications are common, particularly in “old” healing lesions, affecting the nodule itself and/or hilar or mediastinal lymph nodes.2, 4, 6 We reported in the original case collection from Germany and Austria that none of the patients with CNO with pulmonary lesions showed either lesional or lymph node calcifications.1 We furthermore suggested additional differential diagnostic steps to be undertaken, specifically in patients with abnormal appearance of hilar/mediastinal lymph nodes (figure 3 in ref 1). A prominent subcarinal lymph node and calcified mediastinal and hilar lymph nodes together with multiple echogenic changes affecting the spleen were observed in the patient reported by Campos et al, findings that were not present in any of the patients reported in the national cohort from Germany and that would have been classified as “red flags” (figure 3 in ref 1), warranting further diagnostic assessment. Campos et al suggest that, especially in patients receiving immunomodulating treatments, the differential diagnosis of pulmonary nodules include infectious etiologies, including histoplasmosis and tuberculosis.7 Indeed, tuberculosis as well as atypical environmental mycobacterial infections had been considered and were excluded in all patients in the cohort published.1, 8, 9 We agree that individual patients in central Europe, where histoplasmosis is not endemic, may experience infections with Histoplasma, especially those with a recent travel history.5 However, we are confident that none of the patients reported in the national cohort experienced histoplasmosis; not only is this infection very uncommon in Germany and Austria, but also none of the reported patients experienced symptoms associated with histoplasmosis or showed abnormal hilar/mediastinal lymph nodes and/or calcifications.1 Campos et al argue that patients with CNO may have an increased susceptibility to fungal infections associated with immunomodulating therapy with tumor necrosis factor (TNF) inhibitors.7, 10 Notably, in the report cohort from Germany and Austria,1 pulmonary lesions associated with CNO affected some patients before treatment initiation, whereas others received treatment when pulmonary lesions were first detected. In these patients, pulmonary involvement was unrelated to treatments used.1 Importantly, none of the reported cohort patients demonstrated progression of lung lesions, and the majority exhibited complete resolution without specific anti-infective therapy and, in some cases, despite use of immunosuppressive therapy, further supporting that these lesions were not likely to be infective in etiology. Although Histoplasma infections in endemic countries may indeed be more common in patients treated with TNF inhibitors and therefore result in pulmonary lesions, osteomyelitis occurrence has nearly exclusively been reported in disseminated histoplasmosis.7, 10 Considering safety of TNF inhibitor (or conventional disease-modifying antirheumatic drugs) use in rheumatology centers across Germany and Austria, atypical and/or treatment-associated fungal infections (including histoplasmosis) have not been recorded in the long-term juvenile idiopathic arthritis biologic register (BiKeR) and the follow-up register Juvenile Arthritis Methotrexate/Biologics Long-Term Observation (JuMBO)11 (https://biker-register.de/). In summary, we thank Campos et al for their thoughtful and important comment, underscoring that infectious differential diagnoses require careful consideration in patients with CNO and pulmonary lesions, especially when receiving immunomodulating treatments (including tuberculosis, histoplasmosis, and others). Assessments should include a thorough exposure history, including travel. We also very much appreciate that infective differential diagnoses vary between geographic regions and socioeconomic groups and with other factors, such that regional diagnostic algorithms may vary to accommodate for these. Disclosure form. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.