The management of persistent discomfort after acute coronavirus disease 2019 (COVID-19) has emerged as a primary concern. The resulting pulmonary fibrosis is one of the most common post-COVID-19 diagnoses (Babar et al., 2024), also named post-COVID-19 pulmonary fibrosis (PCPF) (Ambardar et al., 2021; Kostopanagiotou et al., 2022). Extensive clinical research has been performed in the last few years to explore mechanisms or risk factors associated with PCPF. The identification of risk factors for PCPF is consistent; however, the incidence rates have fluctuated as a consequence of the evolution of the SARS-CoV-2 virus and variations among research institutions (Tanni et al., 2021; Bocchino et al., 2023). Therefore, constant attention to the long-term prognosis of patients with PCPF is essential, given the substantial number of affected patients and the perpetual evolution of SARS-CoV-2.

Prior studies have classified post-COVID-19 pulmonary CT results into three categories: (1) predominantly ground glass; (2) mixed ground glass and fibrotic; and (3) predominantly fibrotic (Solomon et al., 2021). The overlap of the pulmonary alveolitis phase and fibrosis phase, along with the lack of direct indicators, complicates diagnosis, unless a patient is monitored with multiple CT scans and a further understanding of PCPF, seriously delaying treatment. Consequently, timely, convenient, and effective treatment strategies are urgently needed.

Anti-inflammatory and antifibrotic therapies have emerged as focal points in PCPF research given the critical roles of inflammation and fibrosis in this process (Myall et al., 2021; George et al., 2020; Kooistra et al., 2023; Chen et al., 2022). However, the clinical efficacy of these therapies remains a subject of debate or is largely theoretical (Kooistra et al., 2023). As investigations into long-term COVID-19 progress, reports have been published on the extensive impacts of the persistent reservoir of SARS-CoV-2 on various organs, highlighting the necessity of excluding residual effects induced by the virus (Yao et al., 2023). The significance of small-molecule antiviral drugs (SMADs) during the viral infection phase in affecting PCPF outcomes has gradually been recognized, although it has yet to be thoroughly documented.

Accordingly, we ensured the inclusion eligibility of all the enrolled patients who were identified as having PCPF following strict inclusion and exclusion criteria. We aimed to further explore the one-year prognosis, especially the readmission rate of survivors, which represents a significant financial burden given the high prevalence of infectious cases. Moreover, we aimed to determine the effect of SMADs on the prognosis of patients with PCPF. This could provide an important contribution to the expanding research on the relationship between antiviral therapy and the prognosis of patients with PCPF.

A total of 11,733 adult patients with SARS-CoV-2 were admitted during the study period. Among these patients, 4,925 patients had severe disease and required admission to the ICU or suffered from severe respiratory failure or shock while hospitalized in a general ward. Based on the comprehensive review of chest CT scans and medical records of all the patients, 372 (3.17%) patients diagnosed with PCPF were included in the final analysis. Ultimately, PCPF was identified in 2.78% (137/4,925) of patients with severe disease (Supplementary Figure 1). 90 patients were lost to follow-up after discharge. Consequently, follow-up records were available for 266 patients. A total of 132 patients were readmitted within one-year after discharge. Sixteen PCPF patients died during the hospital stay (4.30%), whereas an additional nineteen died within one-year following discharge. A total mortality rate of 12.41% (35/282) was observed among patients with observational outcomes throughout the study period (Figure 1).

Our study revealed that pulmonary fibrosis induced by SARS-CoV-2 is not uncommon, affecting at least 3.17% of hospitalized patients with COVID-19. PCPF may have long-term implications; nearly 50% of patients under follow-up experienced critical illness and were readmitted within 1 year postdischarge, with some ultimately succumbing to respiratory complications. These basic findings are consistent with research showing that 4.8% (35/837) of COVID-19 survivors progressed to significant pulmonary fibrosis at 4 weeks after discharge (Myall et al., 2021).

In addition, we observed that the CT scores decreased at 6 months but increased at 12 months after discharge. Similarly, a previous study reported that the proportion of patients with post-COVID-19 respiratory symptoms slightly increased from 26% (313/1185) at the 6-month visit to 30% (380/1271) at the 12-month visit (p = 0.014) (Huang et al., 2021). The above evidence suggests that patients with PCPF require effective treatment to avoid these long-term problems.

Notably, the delayed and atypical nature of diagnostic indicators for PCPF may complicate treatment. The treatment of PCPF has focused mainly on the use of antifibrotic therapy and anti-inflammatory therapy. Antifibrotic drugs, such as nintedanib and pirfenidone, may be effective. However, the application of these drugs is limited by diagnostic delays, management challenges, and the high cost of the medications (George et al., 2020). To date, most relevant studies have been limited by small sample sizes (Kerget et al., 2023; Sansores et al., 2023), indicating the need for further investigations on the effectiveness of antifibrotic therapies for treating PCPF. Anti-inflammatory therapies can be initiated promptly during the acute infectious phase, but their use in clinical practice remains controversial. Early administration of corticosteroids was well tolerated and associated with rapid and significant improvement (Myall et al., 2021). Conversely, early treatment with dexamethasone did not appear to affect the incidence or clinical progression of PCPF, nor did it significantly influence clinical outcomes (Kooistra et al., 2023), which is similar to the results of this study (Supplementary Figure 4; Supplementary Table 3). Furthermore, there is an ambiguous distinction between the acute phase and fibrotic phase of COVID-19 infection (Solomon et al., 2021). However, research has shown that patients receiving high-dose corticosteroids during the fibrotic phase exhibited markedly reduced survival times; additionally, no significant differences in survival durations were observed across various dosage groups in the acute alveolitis phase (Chen et al., 2022). These findings complicate decisions regarding the optimal timing for initiating glucocorticoid therapy. Consequently, three critical dilemmas about PCPF treatment must be addressed: timeliness, convenience, and effectiveness. Therefore, SMADs may be the choice.

In accordance with established guidelines and policies, the application of SMADs during the acute phase of COVID-19 has proven to be both accessible and prompt. Fortunately, our study confirmed the effectiveness of SMADs for the treatment of PCPF via the correlation between SMAD treatment and low in-hospital mortality and readmission rates. Previous studies have shown a strong correlation between persistent reservoirs of SARS-CoV-2 and post-acute-COVID-19 symptoms. Chinese scholars reported that the detection of persistent SARS-CoV-2 RNA was significantly correlated with the occurrence of long COVID-19 symptoms in the convalescent population (Zuo et al., 2024). Relevant direct evidence has also been provided. For example, persistent reservoirs of SARS-CoV-2 (Yao et al., 2023) contribute to long-term dysfunction of taste papillae after COVID-19. The same phenomenon was observed in lung tissue. A study revealed SARS-CoV-2 RNA in 80% of lung tissue samples obtained from individuals up to 174 days after COVID-19, whose lung tissue samples were derived mainly from autopsies, explants, and surgical lung biopsies. In these lung samples, fibrotic changes were observed in 75% of the specimens (Roden et al., 2022). Therefore, the recovery, elimination of infectious virus, and development of immunity to acute nonretroviral RNA viruses do not necessarily indicate simultaneous elimination of the viral RNA (Griffin, 2022). In addition to the persistent reservoirs of SARS-CoV-2, related clinical manifestations in individuals infected with SARS-CoV-2 are caused not only by the direct toxicity of the virus in the acute phase but also by virus-triggered immunopathologic effects (Tanni et al., 2021; Lee et al., 2023). A randomized, controlled trial indicated that early antiviral therapy with SMADs could reduce the viral load quickly (Hammond et al., 2022). Data from the health care databases of the US Department of Veterans Affairs revealed that early treatment with SMADs was associated with a reduced risk of post-COVID-19 conditions at 180 days (Xie et al., 2023). These findings are consistent with our results, showing the significance of viral clearance in the acute phase of infection for impacting persistent reservoirs, alleviating the progression of PCPF.

Although SMADs have shown benefits in reducing readmission rates, their effect on one-year mortality remains limited. To further investigate this, we analyzed the causes of death in 35 patients with PCPF (Supplementary Table 4). Respiratory failure was the most frequent direct cause of death (20/35, 57.1%). including direct lung injury from severe SARS-CoV-2 infection which typically occurred during the acute or early recovery phase, and the subsequent severe pulmonary infections secondary to persistent architectural distortion of the lungs. The second most frequent cause of death was septic shock (8/35, 22.9%). This result suggests that patients may exhibit persistent immune dysfunction post-COVID-19. Additionally, corticosteroids or immunosuppressants used to manage fibrotic inflammation may further increase susceptibility to bacterial or fungal infections. It is noteworthy that a subset of patients ultimately succumbed to cardiovascular and thrombotic events. This observation underscores that the impact of the virus on the cardiovascular system cannot be overlooked (7/35, 20%). This risk may be amplified in elderly patients with pre-existing cardiovascular conditions. Therefore, even after recovery from the initial pulmonary infection, patients remain at significantly elevated risk for cardiovascular events, a risk beyond the protective scope of antiviral therapy alone.

Inevitably, there are several limitations because of the real-world, single-center study design. First, the incidence of PCPF in severe patients was 2.78% in this study. However, these results do not negate the high risk of pulmonary fibrosis in patients with severe disease (Tanni et al., 2021). This inaccuracy tied well with the strict inclusion criteria based on CT examinations. Owing to the particularity of the illness, patients in the ICU lacked multiple chest CT scans and were instead given single CT scans or multiple bedside X-ray examinations. Second, not all of the remaining patients completed chest CT examinations at this medical institution for reasons related to patients’ willingness to participate and the accessibility of follow-up visits in this real-world study. Additionally, pulmonary function tests were performed in very few patients, consequently, our study findings cannot reflect the functional pulmonary changes in PCPF patients. Third, the SMADs treatment was initiated during the acute phase in this study. However, the potential need for extended antiviral treatment into the fibrotic phase, as well as the optimal treatment duration, were not thoroughly explored in our research. We acknowledge that the viral persistent reservoir may act as a sustained antigenic stimulus, potentially driving chronic inflammation and fibrosis, a phenomenon observed with other viral infections. Whether sustained antiviral therapy is beneficial in this context remains an open question. Future studies are warranted to further investigate this issue. Finally, 90 survivors were lost to follow-up in this study. Fortunately, there was no significant difference in the 1-year follow-up rate between NO-SMAD group and the SMAD group (69.62% vs. 72.01%, p = 0.715), which did little affect the conclusion of the effect of SMADs on the long-term progression of PCPF patients. We conducted a comparative analysis of baseline characteristics between these lost-to-follow-up patients and those who completed follow-up (Supplementary Table 5). While most baseline variables showed no significant statistical differences between the two groups, the rate of glucocorticoid or immunosuppressive medication use was significantly higher among patients who remained in follow-up. Given the established role of immune dysregulation in the progression of PCPF, this observation raises the question of whether patients with pre-existing immune abnormalities are more susceptible to disease progression. Future prospective studies incorporating more rigorous immunophenotyping are warranted to further explore this hypothesis.

Despite these limitations, several advantages of our study need to be emphasized. The PCPF dataset included data from patients with acute SARS-CoV-2 infection who had reliable evidence and who were continuously observed on the basis of real-world data. This large sample size is beneficial for the generalizability of the study results. Furthermore, bacterial superinfection and mechanical ventilation are important risk factors during acute COVID-19 (Tanni et al., 2021). In this study, most PCPF patients were treated with antibiotics to control secondary infections (94.62%), and few patients were treated with invasive oxygen therapy (4.84%). These results effectively reduce the interference of the above risk factors. Finally, we are the first to confirm a significant association between SMAD treatment and 1-year readmission rates, providing a reliable explanation for the existing intractable treatment of PCPF.

Our findings will hopefully inform future investigations on the characteristics and progression of pulmonary fibrosis following COVID-19 and identify potential effective treatments. Promisingly, with respect to both short-term and one-year prognosis, SMADs may confer benefits to PCPF patients as a convenient and readily available treatment.