The pathogenesis, epidemiology and biomarkers of susceptibility of pulmonary fibrosis in COVID-19 survivors

Introduction: the long-term health consequences of COVID-19

On December 31st, 2019 the WHO China Country Office was informed of cases of pneumonia of unknown etiology detected in Wuhan City; the report signaled the beginning of the coronavirus disease 2019 (COVID-19) emergency and pandemic, with currently over 183 million confirmed cases and four million deaths recorded [1]. As the pandemic evolves, evidence-based medical treatments including anticoagulants, corticosteroids, oxygen therapy and mechanical ventilation (MV) have succeeded in improving patients’ outcomes, and the majority are now expected to recover spontaneously or after acute phase management [2]. In particular, a prolonged administration of low-dose methylprednisolone or dexamethasone has been found to reduce mortality in hospitalized patients with severe disease [3, 4]; accordingly, a WHO guideline panel has recommended systemic corticosteroids in patients with severe or critical COVID-19 [5].

While effective treatments for the acute phase were feverishly being developed, the scientific community also set out to investigate the clinical and functional long-term consequences of COVID-19, which include a complexity of symptoms and organ-related injuries referred to as “long COVID” or “post-acute COVID-19 syndrome” (PACS) [6]. The term “post-acute COVID-19 syndrome” has been proposed to refer to the symptoms and/or abnormalities not attributable to another cause persisting or present beyond four weeks from the time the disease was diagnosed [7]. The severity and extent of the long-term effects of COVID-19 are even now not entirely clear: indeed, on April 1, 2021, the Office for National Statistics estimated that 1.1 million people in the UK were experiencing PACS [8].

According to the scientific literature, PACS is a heterogeneous condition, consisting of a constellation of approximately 50 symptoms, including fatigue, headache, breathlessness, chest discomfort, attention disorders, neurological symptoms such as dementia, depression, anxiety, as well as sleep disturbances and organ lesions affecting the respiratory, cardiovascular, renal, endocrine, ear-nose-throat, gastrointestinal or dermatological systems, and many patients have reported experiencing multiple, overlapping symptoms (Figure 1) [9]. Considered a key component of PACS, pulmonary fibrosis (PF), can severely complicate the clinical course of COVID-19 survivors. Indeed, almost half of patients with moderate to severe COVID-19 pneumonia may show impaired pulmonary diffusion after discharge from hospital. Moreover, lung fibrotic changes on chest CT scan have been reported in approximately one third of patients who recovered from severe or critical disease. It appears that residual fibrotic lesions could be prevented by the prolonged use of low-dose corticosteroids [10, 11].

Figure 1:

Clinical manifestations of patients with “post-acute COVID-19 syndrome”, indicating the possible involvement of the respiratory, cardiovascular, nervous, renal, endocrine, ear-nose-throat, gastrointestinal, musculoskeletal and dermatological system.

The current review focuses on the epidemiology and pathogenesis of post-COVID-19 PF and the biomarkers that may be able to identify the patients who have an increased likelihood of developing the disease. In particular, this study aims to describe different pathogenic hypotheses formulated to explain the development of PF and investigate both the acute phase and follow-up biomarkers that may predict the onset of pulmonary fibrotic lesions.

Pulmonary fibrosis consequent to COVID-19: epidemiology and risk factors

A pathological outcome of acute/chronic interstitial lung diseases associated with abnormal wound healing, PF may contribute to alterations in lung mechanics and pulmonary gas exchange. The most important histopathological characteristics of the disease are: impaired reconstruction of the damaged alveolar epithelium, persistence of fibroblasts, excessive deposition of extracellular matrix (ECM) components such as collagen, and the destruction of normal lung architecture [12]. Notably, clinical, radiographic, and autopsy findings of PF were commonly noted in patients with severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS), conditions caused by a coronavirus similar to COVID-19.

As the number of patients recovering from severe COVID-19 disease rises, it is becoming increasingly apparent that a substantial number show signs of post-inflammatory PF, in particular those patients overcoming COVID-19 related Acute Respiratory Distress Syndrome (ARDS) requiring treatment with high flow nasal oxygen (HFNO) and/or MV [13, 14]. Indeed, PF subsequent to ARDS is a well-recognized phenomenon. In fact, when Thille et al. [15] analyzed 169 autopsies of patients with ARDS linked to a variety of etiologies, they reported that PF was present in three (4%) out of 82 patients with disease duration of less than 1 week, 13 (24%) out of 54 patients with 1 to 3-week disease duration, and 14 (61%) of 23 patients with disease duration longer than three weeks. Moreover, when Desai et al. analyzed 27 patients who underwent MV to treat ARDS, they reported that 23 (85%) showed signs of PF 110–267 days after extubation, and a close association with the duration of the pressure-controlled inverse-ratio ventilation was noted [16].

Since 7.2–31% of COVID-19 patients suffer from viral induced ARDS [17], PF can be considered an important long-term adverse health outcome of the disease [18]. Indeed, bilateral, extensive lung involvement and histopathological findings of diffuse alveolar damage and fibromyxoid cell exudates have been considered structural factors predisposing to impaired lung reconstruction in patients who have overcome a severe case of COVID-19 [19]. Of interest, patients with severe COVID-19 leading to ARDS may also show full thickness tracheal/large airway cartilage lesions with fibrous-hyaline degeneration: this pathologic feature is uncommon in individuals with non-COVID ARDS and could have contributed to the unexpectedly high prevalence of pneumomediastinum reported during the second wave of COVID-19 pandemic [20, 21].

A diagnosis of post-COVID-19 PF should be based on clinical, radiologic, and pathologic findings. While the appropriate timing for a diagnosis of irreversible PF has not yet been established, many specialists recommend serial lung function tests, chest CT scans and exercise tests 3, 6, and 12 months after the onset of an acute COVID-19 episode [12]. Due to the diagnostic challenges linked to the current pandemic, there are no reliable data on the frequency of PF complicating COVID-19. Nevertheless, data collected at the time of hospital discharge suggest that more than a third of COVID-19 patients may develop fibrotic abnormalities [22]. Hu et al. reported that reticular changes, which may represent a sign of irreversible fibrosis, were evident on chest CT scan in 23 out of 46 patients two weeks after the onset of the disease and persisted at the four-week follow-up [23]. Similarly, Pan et al. reported fibrotic alterations on the chest CT scans of 11 out of 63 patients during the acute phase [24]. When Huang et al. carried out a longitudinal analysis of CT findings in patients hospitalized for COVID-19 two months after discharge, they found that 52% (42/81) showed extensive fibrosis [25]. Finally, Wallis et al. reported that 32 out of 101 patients had persistent chest X-ray (CXR) abnormalities at their 12-week post hospitalization follow-up. Notably, the length of the patients’ hospital stay, smoking-status and obesity were found to be independent risk factors for persistent abnormalities. Moreover, serum lactate dehydrogenase (LDH) levels were significantly higher both at baseline and follow-up testing times in the patients with abnormalities compared to those who had recovered [26].

It is important that clinicians are able to recognize potential contributing risk factors for PF. Depending on their origin, these can be classified as patient-related or disease-related (Table 1). Some studies have demonstrated that patients at higher risk of PF tend to be older, male, active smokers and have underlying disorders, such as diabetes and lung and cardiovascular diseases. Indeed, lung fibrosis due to a variety of etiologies is more frequently found in elderly adults. Accordingly, a 230-day follow-up study uncovered that older patients were more likely to develop PF following MERS [27]. Active smoking has been associated to the pathogenesis of various lung diseases such as emphysema, chronic bronchitis, and PF. Epidemiological studies have shown a higher incidence of familial and sporadic Idiopathic PF (IPF) in smokers with respect to their nonsmoking counterparts. A multivariable logistic analysis by Liu et al. uncovered that current smoking was a factor determining progression of lung fibrosis in a group of 78 COVID-19 patients [28]. It has also been suggested that chronic alcoholism increases the risk of PF in COVID-19 patients [29].

Comorbidities also seem to play a decisive role in the development of post COVID-19 PF. Huang et al. used an observational cohort study design to investigate potential indicators of post-COVID-19 PF. Eighty-one COVID-19 survivors who had undergone at least three follow-up CT scans after discharge from hospital were divided into two groups: a fibrosis (n=42) one and non-fibrosis (n=39) one depending on the presence/absence of extensive, persistent fibrotic changes on the CT scans. Compared to the non-fibrotic patients, the fibrotic ones were older (p<0.001), disproportionately male (p=0.036), had more underlying diseases (78 vs. 41%), including diabetes (31 vs. 23%), hypertension (40 vs. 23%), chronic pulmonary disease (21 vs. 7%), chronic liver disease (19 vs. 13%), cardiovascular and cerebrovascular disease (29 vs. 15%) [25].

It has also been noted that the length of stay in a normal ward and in an Intensive Care Unit (ICU), the use of HFNO, the need for MV, and the presence of ARDS seem to be associated with a higher risk of PF [30, 31]. Several clinicians have postulated that prolonged MV-induced lung injury (VILI), which consists in an acute lung injury arising from or exacerbated by MV and is associated with increased ICU mortality in ARDS, may play a role in the development of PF [32]. Abnormalities in ventilator pressure and/or volume settings may underlie the injury leading to a release of pro-inflammatory modulators, worsening acute lung injury and increased mortality or PF in survivors [33].

Severe systemic inflammation may also predict the onset of post-COVID-19 PF. Huang et al. demonstrated that neutrophil, the neutrophil-lymphocyte ratio (NLR), C-reactive protein (CRP) and LDH levels were markedly above the normal range during the first four weeks following hospital admission in patients showing extensive fibrosis at the two month follow-up. Conversely, CRP and LDH increased within two weeks, but then sharply dropped to normal levels in the non-fibrosis group [25].

The pathogenesis of post-COVID-19 pulmonary fibrosis

Following an initial phase of lung injury causing acute inflammation, repair mechanisms can elicit the restoration of normal pulmonary architecture or they may lead to PF with architectural distortion and irreversible lung dysfunction. In normal conditions, during the early stages of lung injury, acute damage to the endothelial cells leads to the release of inflammatory mediators and triggers an antifibrinolytic coagulation cascade that temporarily plugs the damaged vessel with a platelet and fibrin-rich clot [34]. Platelet recruitment, degranulation and clot formation rapidly lead to vasodilation with increased permeability, allowing extravasation and the direct recruitment of leukocytes to the injured site. Subsequently, access to the site of tissue damage is facilitated by the action of zinc-dependent endopeptidases, also called matrixmetalloproteinases (MMPs), which may disrupt the basement membrane by cleaving one or more ECM constituents. As a consequence, chemokine gradients may recruit inflammatory cells. Neutrophils, eosinophils, and lymphocytes are present at the sites of acute lung injury with cell debris and areas of necrosis cleared by alveolar macrophages [35]. The nature of the inflammatory response may influence the resident tissue cells and the ensuing inflammatory cells, since the latter additionally exacerbate inflammation by secreting chemokines, cytokines, and growth factors. Many cytokines are involved throughout the wound-healing response, with specific groups of genes activated in different conditions. Interleukin (IL)-4 (IL-4), IL-13, and transforming growth factor-beta (TGF-β) are cytokines that have received attention with regard to various pulmonary fibrotic conditions since each can exhibit pro-fibrotic activity by promoting the recruitment, activation and proliferation of fibroblasts, macrophages, and myofibroblasts [36]. The final phase of wound healing consists of tissue repair, involving angiogenesis, fibroblast activation, and collagen deposition. In the presence of alveolar exudates, organization occurs which consists in the fibroblastic invasion of the alveoli and transformation into myofibroblasts leading to the deposition of the ECM. Epidermal growth factor (EGF) and transforming growth factor-alpha (TGF-α) stimulate proliferation of bronchiolar stem cells to replace damaged alveolar epithelium. Finally, vascular endothelial growth factor (VEGF) and fibroblast growth factor (FGF) stimulate the migration and proliferation of uninjured endothelial cells leading to pulmonary capillary angiogenesis [37]. In the event of ARDS, a condition characterized by persistent injury with damage to the basement membranes, fibroblastic activities may instead persist, converting organizing into fixed and/or progressive fibroblastic tissue. As a consequence, the formation of either focal or diffuse scar tissue may result in disorganized alveolar architecture, and excessive deposition of ECM is a central aspect of the process [38].The pathogenesis of PF in patients overcoming a severe case of COVID-19 is complex, and several hypotheses (listed below) have been formulated to explain its development (Figure 2).

Figure 2:

Pathophysiological pathways underlying pulmonary fibrosis in COVID-19 survivors.

Multiple cellular and molecular mechanisms may contribute to the pathogenesis of the disease, including direct viral and indirect patient-specific immune-mediated damage (ACE2, angiotensin-converting enzyme 2; AEC, alveolar epithelial cells; ANG II, angiotensin II; ARDS, acute respiratory distress syndrome; CRP, C-reactive protein; DAMPs, damage-associated molecular patterns; ERK, extracellular signal-regulated kinase; MAPK, mitogen-activated protein kinase; TGF-β1, transforming growth factor-beta 1; TNF-α, tumor necrosis factor-alpha).

Susceptibility biomarkers to identify patients at risk of developing pulmonary fibrosis

Since it has been demonstrated that PF consequent to ARDS may lead to high levels of morbidity and mortality, it is crucial to identify the COVID-19 patients at risk of developing fibrotic damage permitting timely action to minimize morpho-functional damage to the lungs. Since no reliable clinical or laboratory indicators are as yet available during the early phases of disease progression, some have proposed using susceptibility biomarkers to identify subjects with an increased probability of developing PF. The acute and follow-up phase biomarkers that have been identified until now are listed in Table 2.

Conclusions

Much has yet to learn about the long term effects of COVID-19, including post-inflammatory fibrosis. Our review highlights the importance of both patient-related and disease-related contributing risk factors for PF in COVID-19 survivors and makes it definitely clear the possible use of acute phase and follow-up biomarkers for identifying subjects with an increased probability of developing the disease. We expect to see the number of efficient biomarkers to increase as we improve our understanding of the disease’s pathophysiology. Biomarkers can be used not only to predict the development of PF and other COVID-19 complications, but can also be utilized to personalize treatment.