Endobronchial tuberculosis with atypical radiological images and features in association with long-term use of inhaled corticosteroid for bronchial asthma: lessons from the case and literature review

Abstract

A 57-year-old man presented with endobronchial tuberculosis (EBTB). He had a history of using inhaled corticosteroids (ICS) for years since asthma was diagnosed during childhood. The present case was remarkable in that ICS alone may have affected the onset and progression of the EBTB. In addition, diagnosing the patient's condition apparently was challenging because he had atypical epidemiology, diagnostic imaging findings, and sites of occurrence for tuberculosis.

The CT images revealed no cavitation, which is reported to be relatively specific and typical for tuberculosis. And then, it demonstrated mixed shadows indicating consolidation in the lower lobe and tree-in-bud, which are not specific. And the patient had no other risk factors of tuberculosis but prolonged ICS use. Tuberculosis was also cited as the differential diagnosis based on the chronic clinical course, bronchial wall calcification, and the tree-in-bud sign on imaging, not being highly suspected overall. To understand the clinical condition, we performed bronchoscopy. As a result, EBTB and pulmonary tuberculosis were diagnosed.

Even when findings are largely atypical for tuberculosis, it should be strongly considered in the differential diagnosis when any typical elements, which are sometimes not specific, are encountered. It is also because EBTB is reported to be extremely transmissible owing to its high rate of bacterial shedding. Moreover, the recognition of the risk of reactivation especially in patients with asthma using ICS will also be important in the future although the possibility that ICS can affect it has been discussed for a long time.

1. Introduction

Endobronchial tuberculosis (EBTB) occurs in approximately 10 to 39% of patients with pulmonary tuberculosis and has a predilection for the left bronchi and bilateral upper lobes in young women [[1], [2], [3], [4]]. Classical symptoms associated with EBTB, such as cough and sputum production, are nonspecific and often overlap with other respiratory disorders. Bronchoscopy, enabling direct visualization and targeted biopsy lesions, leads to a definitive diagnosis of EBTB [5]. Moreover, the use of high-dose inhaled corticosteroids (ICS) is thought to increase the risk of tuberculosis reactivation [6]. There are few cases of EBTB in which ICS use alone was considered to affect the onset and progression of the disease [7,8]. A rare case of asymptomatic EBTB that was eventually diagnosed by bronchoscopy may have been triggered solely by prolonged ICS use and demonstrated atypical epidemiology, sites of occurrence is presented, and imaging findings.

2. Case report

A 57-year-old man with bronchial asthma was referred to our hospital for testing after an unexplained abnormality was found on chest X-ray performed during a routine physical examination two months before his current presentation in mid-September 2024 (Fig. 1). He'd been receiving fluticasone furoate 800 μg/day as ICS ever since asthma was diagnosed during childhood. He denied fever, dyspnea, and chest pain. Radiography showed reticular shadows mainly in the bilateral, lower lungs. Chest computed tomography (CT) demonstrated calcification along the walls of the segmental bronchi in the lower right lobe, shadows indicating pulmonary consolidation and airway constriction, micronodular opacities, and the tree-in-bud sign extending to the distal, lower right lobe (Fig. 2A–C). A grainy shadow was also seen in the middle lobe. A bacterial sputum culture grew only oral flora, and sputum smears and polymerase chain reaction (PCR) were negative for acid-fast bacilli. The result of inflammation markers, such as white blood cell and C-reactive protein, was consistently normal.

Fig. 1.

Chest X-ray film at the first visit showed diffuse small nodular shadows in both lungs.

Fig. 2.

Chest CT findings at the first visit. A: There is calcification along the wall of the segmental bronchi in the lower right lobe (black arrow). B: There are shadows indicating pulmonary consolidation and atelectasis, and the tree-in-bud sign extending to the distal, lower right lobe. C: The tree-in-bud sign was also confirmed in the middle lobe.

The patient had no history of smoking or alcohol consumption and was working as a janitor. He had no other, significant medical history. There was also no family history of tuberculosis or a history of overseas travel. He had no factors that could potentially cause immunosuppression.

Bronchoscopy, which was performed after all other tests failed to clarify the pathology of the present case, showed edema, redness, and ulceration of the tracheal and bronchial lining in the right middle and lower lobes, which were absent in bilateral upper lobes, the lingular segment, and the left lower lobe (Fig. 3A–E). Transbronchial lung biopsies of the right medial basal bronchus (B7) showed epithelioid granulomas without caseous necrosis and infiltration of macrophage and lymphocytes. No malignancy was found, and Ziehl-Nielsen staining was negative for any pathogens.

Fig. 3.

The findings of bronchial lumen by a bronchoscopy. A: Edema, redness, and ulceration were confirmed in the trachea. B: The same findings are confirmed in the entrance of middle and lower lobe. C: right upper lobe, D: left upper lobe and lingular segment, E: left lower lobe.

Smears of bronchoalveolar lavage fluid (B5, 10) and sputum performed immediately after bronchoscopy returned positive for acid-fast bacilli, and PCR was positive for Mycobacterium tuberculosis. Based on these findings, EBTB was clinically diagnosed. Treatment with isoniazid 300 mg/day, rifampicin 600 mg/day, ethambutol 1000 mg/day, and pyrazinamide 1500 mg/day was begun. About one month later, the pathogen was found to be pan-drug-sensitive. Thus, ethambutol and pyrazinamide were stopped at the end of November. Eight-week sputum cultures consistently returned negative results three times, and there were no remarkable adverse events or complications during the clinical course. Therefore, the patient was discharged in December 2024.

3. Discussion

EBTB usually occurs in young women. CT findings of secondary tuberculosis include shadows in the upper lobes [2,3]. Some previous studies have reported that EBTB lesions occur mostly in the bilateral upper lobes, bilateral main bronchus, and left lung, unlike in the present case, which may consequently be considered atypical [4].

The radiological and pathological findings of the present case should be considered based on the pathogenesis of tuberculosis. In the early stages, exudative lesions are usually formed, which are composed of lymphocytes and macrophages. Subsequently, macrophages differentiate into epithelioid cells 5-7 weeks after the development of disease, and they form granulomas; this is called a ‘proliferative reaction’. At the same time, the pathogens themselves remain in the lesion of caseous necrosis, but this reaction specifically suggests the establishment of a specific cellular immune response for tuberculosis [9]. Then, a ‘productive reaction’ occurs, in which epithelioid cells undergo atrophy, and collagenous fibers proliferate. CT demonstrates consolidation reflecting the exudative reaction, which is, however, hard to distinguish from other diseases. In the proliferative reaction stage, we are able to confirm cavities that are considered to reflect caseous necrosis, poorly marginated nodules, and tree-in-bud, the latter two of which reflect granuloma formation, but are not specific for tuberculosis. Moreover, CT shows well-marginated nodules in the productive reaction stage, reflecting the decrease of inflammatory cell components [10]. Based on these facts, the radiological images in the proliferative period are most suitable to be defined as ‘typical’.

The present patient's CT findings included mixed shadows indicating consolidation without cavitation in the right lower lobe, which were defined as ‘atypical’ and nonspecific for tuberculosis, whereas the tree-in-bud sign indicated bronchial involvement, which was considered ‘typical’ for tuberculosis, but not specific. Therefore, diagnosing the patient's condition on the strength of the imaging findings alone was challenging, although tuberculosis had also been considered in the differential diagnosis based on the chronic clinical course and bronchial wall calcification in the right segmental bronchi. It was considered possible that the bronchial calcification stemmed from some other disease, particularly lymphoproliferative disorders and allergic bronchopulmonary aspergillosis, from the imaging findings and the patient's history of asthma. The coexistence of multiple pathophysiologies was also considered. These findings rendered the diagnosis difficult and prompted the performance of bronchoscopy, which resulted in the diagnosis of EBTB and pulmonary tuberculosis, which in turn enabled the pathogen and its susceptibilities to be determined.

The pathological findings in the present case showed mainly exudative lesions and a few granulomas without caseous necrosis. Considering that the disease had developed more than two months earlier, it was presumed that the cellular immune response was delayed [9,11]. These pathological findings are considered comparable to the consolidation in the radiological findings described above. In this context, it is considered that ICS could have caused a localized cellular immune deficiency. It is then difficult to rule out the hypothesis that ICS was the sole cause, because the patient had no other risk factors for tuberculosis reactivation.

The possibility that ICS can affect the endogenous reactivation of tuberculosis has been discussed for a long time. One of the previous studies suggested that the population attributable fraction of tuberculosis due to ICS use was as low as 0.49%. Therefore, patients should not be reluctant to use ICS for obstructive airway diseases due to fears of the elevated risk [12]. Another study reported that the relative risk would double with ICS use equivalent to fluticasone 1000 μg/day [6]. Overall, although ICS use alone clinically increases the risk of endogenous reactivation, it was recognized that we rarely prescribe additional treatment in practice [13]. However, many of these reports were unable to completely rule out confounders or some kind of bias and targeted all obstructive airway diseases. Few reports focused on patients with asthma. Rather, one of them suggested that further analysis of data is needed to assess the risk ratio for adverse effects, including active tuberculosis, when patients with asthma use ICS in the appropriate setting [14]. Moreover, one of them presented a detailed subgroup analysis by COPD or asthma [15]. That report suggested that the adjusted odds ratio for tuberculosis development was as high as 8.03 (95%CI 4.23 to 15.23) in patients with asthma, whereas it was 2.14 (95%Cl 1.71 to 2.66) in patients with COPD when they used ICS with a cumulative dose greater than 75,000 μg fluticasone equivalent. Furthermore, the ratio was as low as 0.82 when those with asthma used less than 15,000 μg. It is possible that some unknown factor that patients with asthma have causes the observed higher susceptibility to tuberculosis development from ICS use, although the report also cautioned that we should wait until other studies confirm this observation. Further accumulation of cases similar to the present one and analysis may help us to identify populations of patients with asthma using ICS who are more likely to develop reactivation of tuberculosis. If so, we might have to take some steps, such as treatment of latent tuberculosis infection, without hesitation.

There have been few reports of EBTB caused by ICS so far, most of which suggested that the patients had other risk factors, such as diabetes mellitus, recent use of systemic corticosteroids, and pneumoconiosis [7,8]. In one case, the patient, who had been receiving the equivalent of over 15 mg/day of prednisone for 2 months, was 7.7 times more likely to develop reactivation than a normal individual [16]. In the other case, the patient had diabetes mellitus and pneumoconiosis; the relative risk is considered to increase 30-fold with pneumoconiosis. It is not easy to rule out these risk factors because the relative risk increases about two-fold with ICS. In the present case, there were no other factors that were considered to have a higher probability of reactivating tuberculosis; therefore, it was considered likely that ICS alone affected the pathogenesis more than in the two cases described above. However, considering that some populations of patients with asthma using ICS may more easily develop reactivation, as mentioned above, we will be able to analyze these previous cases in more detail to determine how much ICS had contributed in future research. The perspective on this might evolve.

Despite the recognition that ICS use clinically increases the risk of endogenous reactivation, the degree of suspicion of tuberculosis was initially not high in the present case, especially since there have been no other documented cases of EBTB that may have been the sole result of prolonged ICS use. The imaging findings, location of the lesion, and its epidemiology were also remarkable in that they were not typical or specific for EBTB. The disease is extremely transmissible owing to its high rate of bacterial shedding [17]. Thus, even though the clinical findings are largely atypical for tuberculosis, EBTB with pulmonary tuberculosis should be considered in the differential diagnosis if any typical elements that are not necessarily specific are encountered. The recognition of the risk of reactivation in patients with asthma using ICS will also be important in the future. Clinicians should be aware of the possibility of atypical cases of EBTB similar to the present one, which cannot be diagnosed solely based on imaging findings. If such a case is suspected, appropriate procedures should be performed for timely diagnosis and treatment.

CRediT authorship contribution statement

Saori Ikeda: Writing – original draft, Conceptualization. Yukio Hosomi: Writing – review & editing, Supervision.

Ethical approval statement

This study was conducted in accordance with the principles of the Declaration of Helsinki.

Consent

Written consent was obtained from the patient.

Funding

This research received no specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests:

Yukio Hosomi has received lecture fees and research funding from AstraZeneca K.K. (Osaka, Japan). Saori Ikeda declares no conflicts of interest.