Post-tuberculosis lung disease and chronic obstructive pulmonary disease

Introduction

Tuberculosis (TB), caused by Mycobacterium tuberculosis (Mtb) infection, is an important global public health problem, especially in low- and middle-income countries (LMICs).^[1] Each year, over 10 million new cases of TB occur worldwide, with 1.5 million deaths.^[1] China is a country with a high TB burden, accounting for 8.4% of all TB cases worldwide, second only to India (26%) and Indonesia (8.5%). With the implementation of standard anti-TB therapy, the majority of the patients have been cured of the disease. However, many patients continue to experience chronic respiratory symptoms (cough, sputum production, and/or dyspnea), thus, affecting their quality of life and increasing the risk of death.^[2] These long-term sequelae of post-tuberculosis lung disease (PTLD) are observed in chest imaging changes and pulmonary function impairments.

Epidemiology of PTLD

PTLD was defined as "evidence of chronic respiratory abnormality, with or without symptoms, attributable at least in part to previous pulmonary tuberculosis."^[3,4] Studies show that up to 50% of patients with TB continue experiencing PTLD-related health problems after the completion of anti-TB treatment.^[5–10] TB infection potentially causes permanent damage to lung anatomy and may be associated with loss of lung function. Recent studies have demonstrated that adult patients with a history of previous TB have a 2–4 times higher probability of developing persistent lung function impairment (airflow obstruction and/or restriction) compared to people without a history of TB.^[8,11] After a year of successful TB treatment, the majority of patients' lung function may be restored; however, high heterogeneity exists. Approximately 12–31% of patients continue to experience an accelerated decline in lung function after the completion of TB therapy.^[12,13] Post-TB pulmonary function abnormalities include airflow obstruction, restriction, or mixed patterns. In a large cross-sectional and sampling survey conducted in China in 2021,^[14] participants aged ≥15 years were included. The post-TB diagnosis was based on radiological evidence and/or medical TB history. In this population sample (n = 8680; average age: 40.1 years), 610 (7.0%) participants had a post-TB diagnosis and experienced more frequent respiratory symptoms (46.8% vs. 28.3%), among whom 130 (21.3%) participants had airflow obstruction. After adjusting for other confounders, airflow obstruction was significantly correlated with post-TB diagnosis (odds ratio [OR] = 1.31). In this study, 297 (48.9%) subjects with post-TB diagnosis also had small airway dysfunction (OR = 1.28). A prospective study on PTLD in Malawi enrolled 405 patients with active TB, 368 of whom completed a 1-year follow-up, and 301 completed a 3-year follow-up. Among these patients, 34% had lung function impairment upon completing TB treatment.^[12] After 3 years of follow-up, approximately 27.9% of patients continued experiencing abnormal lung function, predominantly airflow obstruction (15.8%).^[15]

Imaging Patterns of PTLD

TB infection frequently leaves its footprints of damage in the lungs of survivors. Scarring, fibrosis, and residual "healed" granulomas are common findings frequently associated with calcification, while some patients may show bronchiectatic change or signs of emphysema and gas-trapping [Figure 1]. The imaging patterns of PTLD include five categories: (1) airway lesions: TB-associated obstructive lung disease, and bronchiectasis; (2) parenchymal lesions: calcification, parenchymal destruction, fibrotic change, and Aspergillus-related lung disease; (3) pleural lesions: chronic pleural disease; (4) pulmonary vascular changes: pulmonary hypertension; and (5) others: other pathologies not meeting the above criteria.^[3]

TB-Associated Chronic Obstructive Pulmonary Disease (COPD)

The interplay among TB, chronic airflow obstruction, and smoking is complex with COPD and TB being the two respiratory diseases commonly found in LMICs. Smoking has traditionally been considered the main risk factor for COPD, while in LMICs, TB is now recognized as an important and independent risk factor for chronic airflow obstruction. Smoking is a risk factor for the development of active TB and severe disease progression.^[16] A study in India showed that the death rates from medical causes of ever-smokers were double those of never-smokers.^[17] Of the 1840 deaths from TB among men aged 25–69 years, 79% were smokers.^[17] A study in Hong Kong, China enrolled 16,345 patients with active TB, among whom the number of current smokers and ex-smokers were 3950 (male 91.3%) and 4708 (male 89.9%), respectively.^[18] Both current smokers and ex-smokers were associated with more extensive lung disease, lung cavitation, and positive sputum bacteriology at baseline, and the follow-up results showed that both current smokers and ex-smokers were 1.5–2 times as likely to remain smear-positive and culture-positive after 2 months of treatment, less likely to achieve cure or treatment completion within 2 years, and had a higher relapse risk compared with never-smokers.^[18] Our recent study found that male smoker patients with active pulmonary TB showed a higher chest X-ray (CXR) score, characterized by more cavitary lesions and emphysema, as compared to non-smokers.^[17] Consistently, we also found higher CXR scores at 2 months and 6 months after anti-TB therapy in smokers as compared to non-smokers, suggesting exacerbated lung pathology by cigarette smoke (CS) exposure.^[19]

According to an analysis of 14,050 people aged >40 years from 19 centers worldwide who participated in the Burden of Obstructive Pulmonary Disease Study, the risk of airflow obstruction in people with self-reported TB history was more than twice that in those without TB history (adjusted OR = 2.51, 95% confidence interval [CI]: 1.83–3.42).^[8] A meta-analysis revealed that the pooled prevalence of COPD in patients with previous pulmonary TB was 21% (95% CI: 16–25%).^[8] In our previous cross-sectional study on COPD in the Tibet autonomous region (altitude ≥3000 m) of China, 45.1% of the patients had signs of previous pulmonary TB as seen in the computed tomography (CT).^[20] Our earlier study on COPD (82.1% had a smoking history) in Beijing revealed that a similar percentage of patients had lesions of previous TB on chest CT, and those with TB lesions also showed a higher prevalence of bronchiectasis and more severe emphysema.^[21]

It has been debated whether chronic airflow obstruction and TB-COPD are different or the same entities. Purists prefer to reserve the term of COPD for smoking-related airflow limitation, with other forms being termed "chronic airflow limitation." Allwood et al^[22] were the first to name the condition of tuberculosis-associated obstructive pulmonary disease as TB-associated COPD, based on distinctive physiological and radiological differences observed between smoking- and TB-associated airflow obstruction. In addition, residual post-TB pathology affecting the small airways and vessels can be seen in lung biopsies of patients with TB-COPD, which is distinct from both smoking-related COPD and bronchiectasis.^[23] Moreover, TB-associated COPD was proposed recently as an endotype of COPD,^[24] and the latest iteration of the Global Initiative for Chronic Obstructive Lung Disease (GOLD) Guidelines has proposed a new taxonomy called "Etiotypes" for COPD, which includes COPD-C (related to cigarette smoking) and COPD-I (due to infections, including TB-associated COPD).

The chest images of patients with TB-COPD and smoking-associated COPD (S-COPD) were significantly different. Post-TB lesions, such as bronchiectasis, are a common manifestation of treated TB, which is distinct from those in S-COPD.^[22] In addition, gas-trapping due to small airway involvement appears more common in expiratory CT imaging of TB-COPD than S-COPD.^[22] Therefore, therapies for TB-COPD and S-COPD are somewhat different from each other.

Long-acting bronchodilator therapy (such as cholinergic receptor antagonists and β2-agonists), individually or in combination with inhaled glucocorticosteroid, is the main therapy for the management of COPD.^[25] Bronchodilator therapy has been shown effective to improve lung function and quality of life.^[26] However, the lung function of TB-COPD has heterogeneity,^[27,28] and there is lack of evidence about the effect of bronchodilators in this population.

The similarities and key differences, including clinical features, chest imaging, mechanisms, and therapies, between TB-PTLD and S-COPD are listed in Table 1.^[29,30]

The Immune Mechanism of PTLD and Effect of CS Exposure

The immune mechanisms that drive PTLD are complex and multifaceted.^[25] Upon inhalation of Mtb, the immune system initiates a response to eliminate the bacteria and to prevent the dissemination. The innate immune system acts as the first line of defense against Mtb. Alveolar macrophages, through phagocytosis, recognize and engulf Mtb, thereby playing a crucial role in the formation of granulomas and tissue remodeling.^[31] Neutrophils^[32], dendritic cells, and natural killer (NK) cells^[33] actively participate in the innate immune response against TB. They release pro-inflammatory cytokines and chemokines, recruiting immune cells to the infection site and enhancing the overall immune response against TB. The adaptive immune response in PTLD involves T and B lymphocytes. T cells, specifically T helper 1 cell (Th1) subtype, produce essential cytokines like interferon-gamma (IFN-γ) and interleukin (IL)-2, contributing to the protective immune response against TB. Conversely, Th2 cytokines such as IL-4, IL-5, IL-10, and transforming growth factor-β (TGF-β) can hinder the Th1-mediated response, promoting lesion progression, necrosis, cavity formation, and dissemination. B cells generate antibodies that aid in opsonization and neutralization of Mtb. Cluster of differentiation 4 positive (CD4+) T cells, despite their protective role, can also mediate excessive inflammation, resulting in fibrosis and granuloma formation. Fibrotic granulomas favor the development of PTLD. The complex mix of caseous pneumonia, cavity formation, and fibrosis further exacerbate PTLD.^[25]

CS contains numerous toxic components that can profoundly influence the immune response and impair the host defense mechanisms against TB. This dysregulated immune response not only fails to effectively control the growth of Mtb, but also contributes to the development and deterioration of PTLD [Figure 2].^[16] CS can directly inhibit the function of immune cells, including macrophages and neutrophils, impairing their ability to phagocytose and eliminate Mtb. Our study has observed increased polarization of M1 and M2 macrophage subtypes in CS-exposed patients with pulmonary TB, indicating an altered immune response.^[34] CS-induced neutrophilic inflammation and increased necrosis-related proteins and inflammatory factors exacerbate airway inflammation. Furthermore, CS exposure leads to an increased proportion of natural killer (NK) cells in smoking patients with pulmonary TB, along with enhanced pro-inflammatory cytokine production and cytotoxicity in specific subsets of NK cells.^[35] CS exposure also negatively impacts T-cell function and proliferation, reducing their ability to respond to TB infection and produce key immune mediators such as IFN-γ.^[36–38] CS exposure further exacerbates the progression and severity of PTLD and worsens respiratory symptoms and functional impairment, especially when combined with smoking-related lung damage such as emphysema and chronic bronchitis.

Serum Biomarkers of TB-Associated Lung Injury

After Mtb invades the host, the Mtb antigen will be recognized by the immune system, leading to complex and multifaceted innate and adaptive immune responses. The inflammatory markers and cytokines that are released in response to active TB infection may cause severe damage and remodeling of the airways as well as lung parenchymal abnormalities (thin-walled cavities, lung fibrosis). Therefore, it is essential to identify serum inflammatory biomarkers that would help in predicting PTLD.

Matrix metalloproteinases (MMPs) are a family with 25 potent proteases mainly released by macrophages, neutrophils, and stromal cells, and are involved in extracellular matrix degradation, which is associated with lung damage in patients with TB, especially those with cavitary lesions.^[30,39] MMP-1, -3, -7, -9, and -10 have been reported to be associated with active TB. Serum MMP-1 levels were observed to be significantly increased in patients with PTLD compared to healthy controls.^[40] It drives the degradation of fibril type I, type III, and type IV collagen and is a key pulmonary collagenase. Neutrophil-derived MMP-8 is known to contribute to collagen degradation. MMP-9 and tumor necrosis factor α (TNF-α) are involved in extracellular matrix degradation.^[30] Our study found that CS and Mtb have a synergistic effect on the generation of MMP. MMP-9 and MMP-12 levels in bronchoalveolar lavage fluid (BALF) of smoking patients with pulmonary TB were significantly increased.^[34]

IL-6 promotes lung injury by boosting the recruitment and survival of neutrophils and macrophages, stimulating protease secretion and matrix deposition. Studies have shown that IL-6 concentration is correlated with sputum bacterial load and the severity of lung disease.^[41,42] A recent multi-center TB cohort study (CTRIUMPH) in India and South Africa revealed that high levels of IL-6 at baseline in plasma were associated with subsequent anti-TB treatment failure, relapse, and death.^[43] Incorporating baseline serum IL-6 levels into the prediction model of adverse treatment outcomes (low body mass index, high smear grade, and cavity formation) could improve the prediction efficiency by 15%, suggesting that IL-6 could serve as a potential biomarker for the adverse outcomes of PTLD.

The metabolomic study showed distinctive inflammatory responses between TB-COPD and S-COPD. The levels of IL-6 and C-reactive protein (CRP) and the St George's Respiratory Questionnaire (SGRQ) scores were significantly higher in patients with TB-COPD than in those with S-COPD. The metabolism of fatty acids, especially arachidonic acid and eicosanoic acid, and tryptophan catabolism were correlated, particularly in the TB-COPD group.^[44] Future studies identifying the underlying immune mechanisms for TB-associated lung injury will help in developing prognostic and therapeutic strategies for PTLD.

Unmet Needs

At present, the definition and diagnostic criteria for PTLD demonstrate sensitivity but not specificity, by design. This can lead to over diagnosis of PTLD in asymptomatic individuals. However, in the absence of longitudinal data, it is not known whether asymptomatic abnormalities in radiology and physiology may impact long-term lung health or mortality. Longitudinal data are also needed to determine whether proposed clinical patterns of PTLD have meaningful clinical relevance in terms of outcomes (e.g., lung function decline) and/or treatment.

Conclusions

Given the high burden of the disease globally, there is an urgent need for further research regarding the pathogenesis and identification of biomarkers for PTLD. We need prediction models to identify patients at high risk of developing severe PTLD early in the disease course, and those who are likely to exhibit accelerated lung function decline. Most importantly, we need data on interventions both to prevent and treat PTLD, and pulmonary rehabilitation to improve long-term outcomes and reduce premature mortality.

Funding

This work was supported by grants from the National Natural Science Foundation (Nos. 81400041 and 8217010845), the Capital's Funds for Health Improvement and Research (No. 2022-2G-40910), and the Key Clinical Projects of Peking University Third Hospital (No. BYSYZD2022014). The funding bodies had no role in the design and writing of the manuscript.

Conflicts of interest

None.