More than 200 countries have been affected by the Coronavirus disease 2019 (COVID-19) pandemic that has caused more than 6 million confirmed deaths and 15 million estimated deaths (Wang H., 2022).

Renal abnormalities and gastrointestinal symptoms are common in COVID-19 patients which contributed a lot to the mortality (Austhof et al., 2022; Cheng et al., 2020). SARS-CoV-2 primarily targets the kidneys and gut, which are both important for uric acid excretion. Patients with severe COVID-19 disease had significantly lower serum uric acid concentrations (He et al., 2020; Werion et al., 2020), possibly because inflammation reduced renal tubular uric acid net reabsorption, according to recent studies. However, the retrospective study from Leishenshan Hospital in China demonstrated that the levels of uric acid≥423 μmol/L and ≤278 μmol/L were associated with increased risk of the composite outcome and mechanical ventilation (Chen et al., 2021). Additionally, hyperuricemia was found in many COVID-19 patients during favipiravir therapy (Koseki et al., 2022; Hanai et al., 2022). A population-based study in the UK Biobank cohort demonstrated that gout is a risk factor for COVID-19 incidence and COVID-19-related death (Topless et al., 2022). Several studies have provided evidence of the association between COVID-19 and gout. A large-scale cohort study indicated that gout patients still faced a higher risk of SARS-CoV-2 infection and severe disease progression even after vaccination (Xie et al., 2023). Another study from Mexico revealed that the incidence of gout flares significantly increased from 4.4% before the pandemic to 36% during the pandemic, representing an approximately 8-fold rise (García-Maturano et al., 2022). Data analysis from a global multicenter cohort (including 52,000 COVID-19 patients from China, the United States, and Brazil) revealed that the gout flare rate among patients with mild COVID-19 symptoms was 9.2%, whereas it reached 28.7% in severe cases. Moreover, among severe patients, the risk of gout flare was 3.1 times higher in those who did not regularly take urate-lowering therapy compared with those who did (Jatuworapruk et al., 2023).

Infection with the SARS-CoV-2 virus can trigger a stress response in the immune system, leading to the release of a large number of cytokines and initiating an inflammatory response (Pasrija and Naime, 2021). Gout itself is an inflammatory disease, and persistent immune system stress can amplify the inflammatory response. Concurrently, this stress disrupts the balance between uric acid production and excretion, thereby increasing the risk of gout onset (Wu et al., 2022). Notably, a key cytokine, IL-17A, has been implicated in both conditions. In gout, IL-17A plays a crucial role in mediating the inflammatory response. Intriguingly, research has found that the ORF8 protein expressed by the SARS-CoV-2 virus mimics the function of human IL-17A. It can bind to the IL-17 receptor A, even inducing a more intense inflammatory reaction than human IL-17A (Lin et al., 2023). This suggests a potential common inflammatory pathway between COVID-19 and gout. Furthermore, patients with gout frequently have a high burden of comorbidities linked to worse COVID-19 prognosis, including chronic kidney disease, hypertension, cardiovascular disease, obesity, and diabetes mellitus (Singh and Gaffo, 2020). This comorbid profile places the gout population a priori at a higher risk of complications from respiratory infections.

Recent studies have found shown that febuxostat, a non-purine selective inhibitor of xanthine oxidase (XO), is more effective and had a better safety outcome than allopurinol in lowering the uric acid levels in patients with hyperuricemia and gout (Konishi et al., 2022; Gao et al., 2021; Dehlin et al., 2020; Tsai et al., 2023).

Despite these strong mechanistic and epidemiological premises, and the risk between COVID-19 and gout, no large-scale study has investigated whether Febuxostat could reduce the risk of COVID-19 infections in gout patients. In the present study, we aimed to provide some evidence for Febuxostat use and the risk of COVID-19 infection.

To the best of our knowledge, this is the first large-scale cohort study to demonstrate that Febuxostat decreases the risk of COVID-19 incidence in gout patients. The present study showed that Febuxostat could significantly reduce the risk of COVID-19 incidence and 3 years follow-up. Furthermore, the consistent findings of significant lower risk of COVID-19 and hospitalization in gout patients were found across four different regression models. We also found that male gout patients in the Febuxostat group exhibited a significant reduction in COVID-19 incidence. In addition, Febuxostat significantly reduced hospitalization and critical care service among patients with serum uric acid < 10 mg/dL. Furthermore, Febuxostat significantly reduced the risk of COVID-19 incidence in the follow-up periods of 7 days to 2 years, and 7 days to 3 years. Moreover, Febuxostat could also reduce the hospitalization among COVID-19 infection in gout patients due to renal impairment or with tophus.

The findings from this study carry substantial clinical implications, particularly in the context of managing gout patients during the COVID-19 pandemic and potentially future viral outbreaks. Given that gout itself is a risk factor for both COVID-19 incidence and severe outcomes, the identification of a commonly used urate-lowering therapy that may also reduce viral infection risk offers a dual therapeutic benefit. Febuxostat’s association with reduced COVID-19 incidence and hospitalization suggests that it may serve as a protective agent beyond its primary indication, especially in high-risk subgroups such as males, the elderly, and those with renal impairment or tophaceous gout.

During the development of gout, monosodium urate crystals are formed around the joints, due to elevated serum uric acid concentrations. As the age increases, men are more likely to suffer from gout than women (Dehlin et al., 2020; Han et al., 2024). In the present study Febuxostat significantly reduced the risks of COVID-19 incidence in male gouts and hospitalization among COVID-19 infection in elder patients (more than 45 years old) compared to the Allopurinol group.

Compared with individuals with a baseline serum urate of <6 mg/dL, the risks of gout in those with a baseline serum urate of ≥10 mg/dL increased to 64 times (Evans et al., 2018). Gout patients with serum uric acid (SUA) concentration of more than 10 mg/dL are likely associated with many of the comorbidities. A Retrospective Study demonstrated that Febuxostat showed efficacy and renal safety in male patients with gout and chronic kidney diseases. Moreover, gout patients with SUA < 10 mg/dL and body weight <70 kg reached more preferrable results (Zhang et al., 2022). In the present study, around 40% of the participants have comorbidities with acute kidney failure and chronic kidney disease. Febuxostat significantly reduced the risks of hospitalization and critical care service in gout patients among COVID-19 infection with SUA <10 mg/dL. Moreover, hospitalization was significantly reduced in gout patients due to renal impairment or with tophus.

COVID-19 vaccination has been demonstrated to be associated with increased odds of gout flare (Lu et al., 2022; He et al., 2024). Similar pathophysiological mechanism of crystal-induced arthritis and the inflammatory response of COVID-19 vaccination might make gout patients more susceptible to vaccine-induced flare than other rheumatic patients (Tai et al., 2022; Perozzo et al., 2022). For gout patients, MSU crystal and prolonged hyperuricemia could increase the inflammatory response to a secondary stimulus such as COVID-19 vaccine (Bodofsky et al., 2020). In the present study, Febuxostat could increase the risk of mechanical ventilation for gout patients with vaccination. In contrast, the risks of COVID-19 incidence and hospitalization were reduced in gout patients without vaccination in the Febuxostat group.

Cytokine storm including inflammatory cytokines IL-1β, IL-6, IL-17, and TNF-α could be provoked in moderate and severe COVID-19 patients (Montazersaheb et al., 2022). From a translational perspective, these results support the reconsideration of Febuxostat as a preferred urate-lowering therapy in gout patients with comorbidities that increase susceptibility to severe viral infections. This is particularly relevant given the drug’s favorable safety profile in patients with renal impairment, a common comorbidity in gout populations. Moreover, the anti-inflammatory properties of Febuxostat, through inhibition of the JAK/STAT and NF-κB pathways and subsequent reduction in pro-inflammatory cytokines such as IL-1β, IL-6, and TNF-α, may underlie its protective effects against COVID-19-related hyperinflammation and organ damage (Hao et al., 2019; Khan et al., 2017; Abo-Youssef et al., 2020). This mechanistic insight not only reinforces the biological plausibility of our findings but also opens avenues for repurposing Febuxostat in other inflammatory or infectious conditions characterized by cytokine dysregulation.

To validate the robustness of our study and reduce the bias, we performed four models of Cox regressions, which all revealed that Febuxostat could reduce the risk of COVID-19 incidence and hospitalization of gout patients infected with COVID-19. Furthermore, after dealing with the mortality competing risk, the results consistent with the former ones.

However, our study has several limitations. First, even though treatment start time of Febuxostat was evaluated, Febuxostat and Allopurinol dose data could not be obtained from the database. We were unable to evaluate dose–response or persistence/time on treatment because the data platform lacks standardized dosing and days-supply/refill information and does not support patient-level time-varying dose construction. As a result, unmeasured differences in titration or persistence between groups could remain and may have influenced effect estimates; findings should be interpreted with this limitation in mind. In addition, there were relatively few Asian people, American Indian or Alaska Native, or Native Hawaiian or Other Pacific Islander participants in our study, which could have caused results to be biased based on race. Collaborative electronic health records was unable to obtain healthcare insurance status during the COVID-19 pandemic, which could confound the results. Second, because index dates were distributed across 2020–2022, many patients had index dates before vaccines became available, resulting in low observed vaccination rates. Additionally, vaccination administered outside of the network may not have been captured, potentially leading to misclassification. Furthermore, our composite outcome definition may have introduced misclassification, as some non-COVID pneumonia cases could have been included, and some infections may have been missed if patients did not seek care or were tested outside the network. Such misclassification is likely non-differential, which would bias effect estimates toward the null. PSM was performed to minimize bias, but misclassification bias and residual confounding could not be completely eliminated due to some disadvantages of an electronic health record database. Moreover, the dataset is not a population-based dataset. Therefore, patients may receive medical treatment at hospitals outside of the collaborative electronic health records platform, resulting in a loss to follow up. Third, a potential limitation of this study is that exposure index dates spanned multiple pandemic waves with evolving testing availability, treatment approaches, and circulating variants. Although the index date distributions overlapped across treatment groups, residual calendar-time confounding cannot be completely excluded. Besides, although we attempted to minimize immortal-time bias by applying a day-7 landmark after the index date and excluding patients who switched between febuxostat and allopurinol during follow-up, residual immortal-time bias cannot be completely ruled out. Readers should interpret the observed associations with caution, recognizing the possibility that any remaining immortal-time bias may have attenuated or inflated the observed effect estimates. Fourth, some combined treatments such as NSAIDs, corticosteroids, and other anti-gout drugs such as colchicine may have effects on the COVID-19 risk of gout patients and some drugs may have complex drug-drug interactions with Febuxostat or Allopurinol. Fifth, we reported exploratory subgroup results for SUA, vaccination record, sex, and age. Because interaction tests were approximated from subgroup HRs rather than estimated within a single model, and no multiplicity control could be implemented, type-I error may be inflated and interaction findings should be interpreted cautiously. Notably, point estimates were directionally consistent (febuxostat associated with lower COVID-19 risk) in SUA >10 mg/dL, vaccinated, female, and ≥65-year strata. Finally, despite propensity score adjustment, residual confounding by indication may persist because febuxostat is often selected for patients with allopurinol intolerance/contraindications or chronic kidney disease, and we could not fully capture gout severity (e.g., tophi, flare frequency, serum urate trajectories), provider preference, or access to care. The primary estimate for COVID-19 infection (HR = 0.878; 95% CI, 0.801–0.963) corresponds to E-values of 1.54 for the point estimate and 1.24 for the upper CI bound, indicating that an unmeasured confounder associated with both treatment assignment and infection risk by a risk ratio of ≥1.54 could fully explain the observed association, whereas a confounder with associations of ∼1.24 could render the result non-significant. Accordingly, our findings should be interpreted with caution in light of potential residual bias.

Future research should include randomized controlled trials to confirm these observational findings and elucidate the optimal dosing and timing of Febuxostat administration relative to vaccination and infection exposure. Additionally, studies exploring the effect of Febuxostat on other respiratory viruses or post-acute sequelae of COVID-19 could further expand its clinical utility.

Taken together, in this retrospective cohort study, Febuxostat was associated with lower risks of COVID-19 incidence in gout patients.