Introduction

Three years into the Corona Virus Disease-19 (COVID-19) pandemic, caused by Severe Acute Respiratory Syndrome Corona Virus 2 (SARS-CoV-2), has resulted in unprecedented global issues with more than 0.6 billion cases of the disease and 6.7 billion deaths (as of February 1, 2023) [1]. Numerous countries have faced several waves of COVID-19 outbreaks with enormous financial and social implications.

Despite the improved understanding of the acute phase of the disease, its long-term consequences are still poorly explored. One-third of COVID-19 survivors are repeatedly hospitalized over a mean period of 140-day follow-up [2]. And up to one-third of survivors experience a variety of symptoms after the acute phase of COVID-19 [3, 4], raising concerns of uncontrolled complications.

The underlying pathobiology of COVID-19 long-term consequences remains poorly explored while an increasing number of patients with rheumatic diseases (RDs) and COVID-19 complications are encountered in the rheumatologist's practice. The association between SARS-CoV-2 and RDs is seemingly bidirectional. There is a risk of COVID‐19 advancement in subjects with RDs [5]. On the other hand, COVID-19 itself may trigger autoimmunity and result in new rheumatic manifestations [6].

There are scarce data on outcomes of COVID-19 in patients with RDs. Better understanding of prolonged COVID-19 symptoms in patients with RDs may help improve monitoring and treatment. With that in mind, we aimed to analyze pathogenic mechanisms of COVID-19 symptoms and phenotypes in patients with RDs over a long term.

Search strategy

We searched through Medline/PubMed, Scopus, and Directory of Open Access Journals (DOAJ) databases for relevant original articles, case studies, and reviews published by February 1, 2023. The following keywords were employed: post-COVID-19, post-acute COVID-19 syndrome, Long COVID, rheumatic diseases, immune-mediated inflammatory disease. Our search strategy was in line with previously published recommendations [7].

COVID-19 features in patients with RDs

SARS-CoV-2 is an enveloped single-stranded RNA virus with a weight of approximately 29.9 kB and a diameter of 50–200 nm [8]. The virus penetrates human cells by binding to angiotensin-converting enzyme-related carboxypeptidase (ACE2) receptor which is expressed on numerous organs, including the lungs, intestines, brain, heart, kidneys, and liver [9]. It enters its target cells after being cleaved by the transmembrane protease serine 2 (TMPRSS2) via viral spike proteins 1 and 2 [10]. ACE2 is downregulated after binding with virus, significantly contributing to cytokine-mediated hyperinflammatory response [11].

Clinical characteristics of COVID-19 range from asymptomatic or mild course to multiorgan failure (MOF) and death. A Chinese study of 44.000 COVID-19-positive subjects demonstrated that 81% of patients had mild symptoms, 14% developed severe symptoms, and 5% progressed to life-threatening conditions [12]. Timely diagnosis of COVID-19 is critically important in subjects with underlying immune disbalance and high activity of RDs [3, 13]. A US-based retrospective study revealed that patients with RDs have a threefold higher incidence of coronavirus infection than the general population. Moreover, SARS-CoV-2 positivity significantly increases the risk of RD flares (Odds Ratio [OR] 4.6, 95% Confidence Interval [CI] 1.2–17.4) [14]. Likewise, patients with immune-mediated inflammatory diseases (IMIDs) and COVID-19 presented with an increased hospitalization risk compared to non-IMID controls (adjusted OR [aOR] 1.2) [15]. Patients with iritis (OR 1.5), psoriatic arthritis (PsA) (OR 2), rheumatoid arthritis (RA) (OR 1.4), and vasculitis (OR 2.1) were also at risk of hospitalization due to COVID-19 [16]. Severe COVID-19 has been associated with RDs. In fact, an increased frequency of respiratory failure was reported in patients with RDs and COVID-19 as compared to non-RD controls (38% vs 10%, p < 0.001) [17].

Various RD phenotypes may confound clinical outcomes of COVID-19. Notably, the Global Rheumatology Alliance reported a greater risk of worse COVID-19 outcomes in patients with giant-cell arteritis at older age (OR 1.9, 95% CI 1.3–2.8) and those with obesity (OR 3, 95% CI 1.2–7.6) [15]. Similarly, older age (OR 1.6, 95% CI 1.3–1.9), rituximab (RTX) use (OR 2.15, 95% CI 1.15–4.01), cyclophosphamide use (OR 4.3, 95% CI 1.1–16.8) were associated with severe COVID-19 in subjects with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) [15]. Therefore, there is a need for RD phenotype stratification to adjust treatment strategies and avoid worse COVID-19 outcomes [15].

Hyperinflammatory response in COVID-19

Immune dysregulation merits special attention in subjects with COVID-19 [18]. Available evidence points to an interaction of SARS-CoV-2 with innate immunity via pattern recognition receptors (PRRs) such as Toll-Like Receptors (TLRs) or Retinoic acid-inducible gene I (RIG-I)-like from the components receptors (RLRs) [19]. The interaction is followed by downstream signaling with secretion of various cytokines, including type I/III interferons (IFNs), tumor necrosis factor-alpha (TNF-alpha), interleukin-1 (IL-1), and IL-6, resulting in adaptive immune responses [20]. An augmented inflammatory response with an increase in IL-6/IL-10 ratio may lead to adverse outcomes [21]. However, regulated production of IFN-I counteracts SARS-CoV-2 infection [20]. Overall, an uncontrolled increase in cytokines during the acute period of COVID-19 manifests as cytokine release syndrome (CRS) and MOF and heightens the risk of autoimmune complications at a later stage [22].

Immune mechanisms in Long COVID

More than 20% of subjects surviving acute COVID-19 may suffer from some persisting symptoms of infection and develop new ones after one month [23]. This clinical condition is known as post-COVID-19 syndrome (PCS) or Long COVID that frequently presents with fatigue, musculoskeletal signs, cognitive impairment, and sleep disorders [23]. Although here is no universally accepted definition of Long COVID-19, it can be characterized as a systemic inflammatory response develo** during or after SARS-CoV-2 infection, lasting from 3 to 12 months, and not attributable to any alternative diagnosis [24].

Long COVID manifests as a clinically variable condition. A sizeable proportion of Long COVID patients develop rheumatic symptoms [25]. Older age (OR 1.2, 95% CI 1.1–1.4, per 10-year increase) and female gender (OR 1.6, 95% CI 1.1–2.2) increase the risk of rheumatic symptoms [25]. Emerging evidence considers immune dysregulation as the key pathogenic feature of Long COVID. A case series of Long COVID-19 revealed significantly elevated TNF-α, IL-1β, and IL-13 and significantly decreased interferon-γ-induced protein 10 (IP-10) as compared to the onset of COVID-19 [26]. In view of immunological shifts, IgG anti-SARS-CoV-2 S1 antibodies remained significantly elevated from the acute phase of COVID-19 and persisted in Long COVID [26]. Lymphocytes, including Th9, CD8 + effector T cells, B cells, and CD4 + effector memory cells also remained increased compared to pre-pandemic controls and to levels measured at day 28 of COVID-19 [26]. Upregulation of proinflammatory mediators appeared to be maintained in COVID survivors even 8 months after the infection [27]. Persistent overexpression of type I IFN (IFN-β) and type III IFN (IFN-λ1) and high proinflammatory cytokines such as IFN-β, IFN-γ, and IL-6 linked to Long COVID with 78.5–81.6% accuracy [27]. This chronic inflammatory response with elevated B cells, known as the primary source of autoantibodies accounts for the likelihood of autoimmune mechanisms in long-term COVID complications [26].

Autoimmunity in COVID-19 and Long COVID

Some of the reported autoimmune manifestations of Long COVID can be induced by viral persistence, endothelial dysfunction, and altered levels of inflammatory biomarkers after the acute phase of COVID-19 [28]. Numerous autoantibodies, including antinuclear antibodies (ANA) and ANCA, have been found in patients infected with SARS-CoV-2 [29,30,31]. Chang et al. reported that approximately 50% of subjects hospitalized with SARS-CoV-2 positivity had autoantibodies linked with various RDs such as myositis and systemic sclerosis [32]. Another study reported the frequency of anti-52 kDa SSA/Ro, anti-60 kDa SSA/Ro, and ANA antibodies (20%, 25%, and 50%, respectively) in patients with COVID-19, suggesting the activation of the autoimmunity [33].

One of the proposed mechanisms of Long COVID-19 involves the formation of neutrophil extracellular traps (NETs) that may create the environment for autoantibody production with the exposure of intranuclear components of cells [34]. Another mechanism of Long COVID associates autoimmunity with pathobiology of B cells. In severe COVID-19 extrafollicular B-cell activation strongly correlated with poor disease prognosis and elevated SARS-CoV-2 specific neutralizing antibodies [35]. A case of post-COVID-19 systemic lupus erythematosus (SLE) with accompanying B-cell activation was reported [36]. Another study revealed that 83% of patients with Long COVID develop latent autoimmunity and 62% present with polyautoimmunity. More than 85% of patients had anti-SARS-CoV-2 IgG antibodies that positively correlated with autoantibodies, age, and body mass index (BMI) [37]. Prevention of Long COVID triggering mechanisms remains poorly understood (Fig. 1).

Fig. 1
figure 1

Pathogenic mechanisms of Long COVID

Full size image

Long COVID symptoms over time

Long COVID clinical features are variable over time (Table 1). A meta-analysis by Lopez-Leon et al. reported tiredness, headache, hair loss, and dyspnea as the most frequently seen symptoms in Long COVID. However, this report did not distinguish sets of symptoms in hospitalized and non-hospitalized subjects and did not relate to follow-up periods [38]. Based on the COVID-19 Global Rheumatology Alliance Vaccine Survey of subjects with systemic autoimmune rheumatic diseases (SARDs), 1 in 4 had COVID-19 symptoms lasting 28 days or longer and 1 in 10 experienced the same 90 days or longer. The following factors were associated with prolonged symptom complexes: hospitalization for COVID-19 (age-aOR 6.5, 95% CI 3.0–14.1), comorbidities (aOR 1.1 per comorbidity, 95% CI 1.0–1.2), and osteoarthritis (aOR 2.1, 95% CI 1.0–4.3) [39].

Table 1 Follow-up period of patients recovering from SARS-CoV-2
Full size table

Two-month follow-up of patients recovering from SARS-CoV-2 infection

Two months after the recovery from COVID-19 pneumonia, almost half of survivors demonstrated restrictive lung disease while a reduced lung diffusion capacity (DLCO) was reported in one-third of cases [40].

Three-month follow-up of patients recovering from SARS-CoV-2 infection

Three months after the acute phase, 64% (n = 35) of patients still suffered from COVID-19-related symptoms and 71% (n = 39) showed various degrees of radiological and physiological pulmonary abnormalities [64].

Endotheliopathy and Long COVID

Persisting endotheliopathy is a typical feature of Long COVID [65]. SARS-CoV-2 damages the endothelium by targeting ACE2 receptor [66]. The resultant endotheliopathy underlies multisystemic features in COVID-19 [67].

Markers of endothelial cell (EC) activation, including von Willebrand factor antigen (VWF: Ag), VWF pro-peptide (VWFpp), and factor VIII significantly increase in convalescent COVID-19 subjects compared to healthy controls [65]. Thrombogenic factors associated with increased COVID-19-related mortality include elevated D-dimer at hospital admission with an OR of 18.4 (95% CI 2.6–128.5) [68]. However, Moreno-Perez et al. reported that CRP and D-dimer were not predictive of Long COVID [69].

Pulmonary impairment and Long COVID

A meta-analysis of 16 cohort studies with hospitalized patients followed more than 1 month post-discharge or more than 2 months post-admission reported 20% of pulmonary disturbances. Diffusion impairment was the most frequent feature, followed by restrictive ventilatory defects [70]. Acute respiratory distress syndrome (ARDS) associated with COVID-19 may contribute to the lasting damage of alveoli with irreversible fibrosis [71]. Cytokine storm as the trigger of immunopathological pathways may be associated with pulmonary fibrosis in Long COVID [72].

COVID-19 may contribute to thromboembolic microangiopathy with subsequent immunomodulatory reactions in the pulmonary vascular bed. These mechanisms may result in chest pain in post-COVID-19 periods [73]. Overall, individuals who recovered from COVID-19 with residual pulmonary injury should be monitored for at least 36 months [74].

Potential biomarkers of Long COVID

Increasing cases of Long COVID draw the public health system's attention to seek diagnostic markers identifying illness progression and timely follow-up after hospital discharge. SARS-CoV-2 spike protein might justify its diagnostic value. S1 subunit of SARS-CoV-2 spike protein is detected in about 65% of patients with Long COVID during a one-year follow-up period [75]. Dysregulation of the neuropilin-1 (NRP-1)/vascular endothelial growth factor (VEGF)-A pathway may also underly the course of Long COVID. In a cohort study, about 50% (n = 48) of individuals with COVID complications had significantly increased serum levels of VEGF compared to fully recovered subjects [76]. More studies are warranted to correctly screen, diagnose, and manage Long COVID.

Treatment of Long COVID

RTX therapy has been associated with the risk of severe COVID-19 in patients with RDs; subjects on RTX therapy have frequently presented with critical conditions (effect size 3.3, 95% CI 1.7–6.4) and experienced a longer hospital stay (0.6, 95% CI 0.5–0.9) in comparison with subjects not treated with RTX [77]. Long-term RTX therapy reduces levels of B-cells with subsequent low antibody production. This effect may cause the impairment of viral clearance by antibodies [78].

HCQ was another widely used DMARD for COVID-19 at the beginning of the pandemic. Initial reports were optimistic over the use of HCQ to treat SARS-CoV-2 infection [79] suggesting that low doses of HCQ were relatively safe for patients with RDs infected with SARS-CoV-2 [80]. A clinical trial investigating HCQ with and without azithromycin versus placebo reported reduced viral load/disappearance in patients with severe disease [81]. However, side effects of high doses of HCQ were described in COVID-19 patients [82]. Cardiotoxicity with arrhythmias reported in the elderly with pre-existing cardiac history questioned the safety and efficacy of HCQ therapy [83]. Another trial demonstrated that hospitalized COVID-19 subjects on HCQ did not show a lower incidence of death at 28 days compared to non-HCQ subjects, confirming the lack of HCQ efficacy in COVID-19 [84]. The role of HCQ in the context of COVID-19 long-term complications has not been examined. The potential utility of HCQ in Long COVID could be due to its inhibition of TLR signaling, binding to their ligands, and dampening proinflammatory cytokine production [85].

GCT has attracted more attention in the COVID-19 pandemic. Patients with RDs on ≥ 10 mg per day systemic GCT presented with increased odds of positive SARS-CoV-2 test (aOR 1.47, 95% CI 1.05–2.03), severe COVID-19 (OR 1.76, 95% CI 1.06–2.96), and COVID-19-related death (OR 3.34, 95% CI 1.23–8.9) [86]. Careful dose adjustment may prevent poor clinical outcomes in COVID-19 due to the immune-modulating effect [86]. One study identified the effects of long-term GCT on the incidence and outcomes of COVID-19 in patients with RDs [13]. A univariate analysis revealed that moderate doses of glucocorticoids (GCs) (7.5–20 mg) conferred a higher risk of COVID-19 (RR = 1.7, 95% CI 1.1- 2.6) that persisted in a multivariate analysis (RR 1.6, 95% CI 1.0- 2.5) [13].

A large case series of the COVID-19 Global Rheumatology Alliance reported an association between GC dose ≥ 10 mg/daily and hospitalization (OR 2.05, 95% CI 1.06–3.96) [5]. However, there was no association with hospitalization for DMARD alone or in combination with biologics/Janus Kinase inhibitors [5]. Likewise, there was no association for non-steroidal anti-inflammatory drugs (NSAID) [5].

Notably, corticosteroid therapies in the pandemic may increase the frequency of avascular necrosis (AVN), particularly with high doses and longer duration of therapies. A recent case series showed the presence of AVN in patients with COVID-19 treated with prednisolone at mean dose of 758 mg (400 mg–1250 mg) [87] that was less than the mean cumulative dose reported in the literature [88]. Careful monitoring of patients with hip and thigh pain and timely bisphosphonate therapy after acute COVID-19 may help avoid severity complications through [87]. Finally, GC use may increase the risk of thrombosis in patients exposed to SARS-CoV-2 [89].

Discontinuation of treatment during the COVID-19 pandemic

Many patients with RDs discontinued their therapies due to the fear of SARS-CoV-2 infection [90, 91]. There was a significant difference in disease activity in those who discontinued their treatment and those who continued (p = 0.001) [28].

In a case report of a 37-year-old female with history of RA, antiviral therapy with nirmatrelvir/ritonavir reduced Long COVID symptoms; meanwhile, discontinuation of tocilizumab therapy triggered new-onset brain damage [92]. Rational and informed discussions about the risks of withdrawal effects may help to optimize the recovery in Long COVID.

Vaccination and COVID-19

An Italian multicentric cohort study has shown that vaccinated patients were less likely to suffer from Long COVID [93]. Patient education was shown to increase awareness of vaccine safety and efficiency and to reduce related fears [94].

A UK-based study analyzed the association between COVID-19 vaccination and post-COVID-19 symptoms in patients previously infected with SARS-CoV-2 [95]. The first vaccine dose reduced the risk of post-COVID-19 symptoms by 13% within a median follow-up of 141 days. The second dose decreased the same risk by 9% within a median follow-up of 67 days [95]. Overall, COVID-19 vaccination decreases the risk of Long COVID.

Common side effects of COVID-19 vaccination include pain at the injection site, redness, paresthesia, and fatigue; more serious adverse effects are vaccine-induced immune thrombotic thrombocytopenia (VITT) and cerebral venous sinus thrombosis (CVST) [96]. The mechanism of VITT could be due to the interaction between free DNA in the vaccines and platelet factor 4 (PF4), leading to PF4-reactive autoantibodies production [96]. In addition, post-COVID-19 vaccination was associated with a sixfold greater risk of gout flare within three months (aOR 6.02, 95% CI 3.00–12.08); colchicine prophylaxis showed its effectiveness due to lowering the risk of gout flare by 47% after vaccination [97]. Another research reported the possible association between the development of lupus anticoagulant-associated venous thromboembolism and Pfizer mRNA COVID-19 vaccination [98].One study reported 5 cases of RDs after the second dose of Pfizer-BioNTech and Moderna COVID-19 vaccinations [99]. Further studies are required to elucidate the pathogenesis and possible connection between COVID-19 mRNA vaccination and the likelihood of RDs occurrence.

Conclusions

Considering a high frequency of Long COVID, more efforts are needed to raise public awareness of mitigating COVID-19 lasting consequences in RDs. Numerous questions remain unanswered regarding the prevention of immune complications over a prolonged period in patients with RDs.

Dampening prolonged inflammatory reactions and persistent autoantibody production may help avoid COVID-19 lasting complications. Rheumatologists should closely monitor COVID-19 survivors with lasting symptoms and personalize diagnostic and therapeutic approaches.