Increased incidence of giant cell arteritis and associated stroke during the COVID-19 pandemic in Spain: A nation-wide population study

Introduction SARS-CoV-2 infection and COVID-19 vaccines might have increased the incidence of giant-cell arteritis (GCA) and the risk of associated stroke in Spain. Methods Retrospective nation-wide observational analysis of all adults hospitalized with GCA in Spain during 5 years (Jan-2016 and Dec-2021). The incidence and proportion of admissions with or because of GCA and GCA-associated stroke were compared between pre-pandemic (2016–2019) and pandemic (2020 and 2021) years. Sensitivity analyses were conducted for the different COVID-19 waves and vaccine timing schedules. Results A total of 17,268 hospital admissions in patients diagnosed with GCA were identified. During 2020 there were 79.3 and 8.1 per 100,000 admissions of GCA and GCA-associated stroke, respectively. During 2021 these figures were 80.8 and 7.7 per 100,00 admissions, respectively. As comparison, yearly admissions due to GCA and GCA-associated stroke were 72.4 and 5.7 per 100,00, respectively, during the pre-pandemic period (p < 0.05). Coincident with the third wave of COVID-19 (and first vaccine dosing), the rate of GCA-associated stroke admissions increased significantly (from 6.7 to 12%; p < 0.001). Likewise, there was an increase in GCA-associated stroke (6.6% vs 4.1%, p = 0.016) coincident with the third dose vaccination (booster) in patients older than 70 at the end of 2021. In multivariate analysis, only patients admitted during the third COVID-19 wave (and first vaccine dosing) (OR = 1.89, 95% CI 1.22–2.93), and during the third vaccination dosing in patients older than 70 (booster) (OR = 1.66, CI 1.11–2.49), presented a higher GCA-associated stroke risk than the same months of previous years after adjustment by age, sex, classical cardiovascular risk factors and COVID-19 diagnosis. Conclusions The COVID-19 pandemic led to an increased incidence of GCA during 2020 and 2021. Moreover, the risk of associated stroke significantly risen accompanying times of COVID-19 vaccine dosing, hypothetically linked to an increased thrombotic risk of mRNA-SARS-CoV-2 vaccines. Hence, forthcoming vaccine policies and indications must weigh the risk of severe COVID-19 with the risk of flare or stroke in patients with GCA.


INTRODUCTION
Giant-cell arteritis (GCA) is a chronic granulomatous vasculitis involving large and medium sized arteries [1]. Although GCA can potentially occur any time after 50 years of age, the disease typically affects elderly patients, with the peak incidence being between 70 and 80 years [2,3]. Clinically, GCA usually presents with an insidious course with general manifestations including fever, asthenia or weight-loss, along with local symptoms due to arterial inflammation and vascular deficits such as headache or jaw claudication [2]. When the arterial compromise is severe enough, GCA might lead to ischemic phenomena, such as acute ischemic optic neuropathy (AION), transient ischemic attack (TIA) or even established stroke, frequently in the carotid or vertebrobasilar territories [3,4]. These severe complications can occur in up to 20-50% of patients with GCA and usually appear within the period of active disease [4]. GCA pathogenesis is yet practically unknown [5]. However, some reports have pointed out the potential role of certain viruses, such as cytomegalovirus or varicella zoster, as well as influenza vaccines, as triggers for GCA in predisposed individuals [6][7][8].
The surge of SARS-CoV-2 infection by the end of 2019 in China and its rapid spread worldwide is an unprecedented medical phenomenon. Millions of deaths have occurred during the three years since the pandemic onset [9]. However, COVID-19 is clinically very heterogeneous [10]. On the one hand, a majority of patients present an asymptomatic or oligo-symptomatic illness whilst other might suffer an interstitial pneumonia and even an acute respiratory distress syndrome (ARDS), the main cause of death in these patients. In addition, a subset of patients develops what has been called a 'cytokine storm', a syndrome of systemic hyperinflammatory dysregulation with coagulation disorders, thromboembolic events, myocarditis, acute kidney injury, hepatitis and multi-organ failure [11]. The cytokine storm has been compared to catastrophic antiphospholipid syndrome (CAPS), Still's disease or hemophagocytic lymphohistiocytosis, all severe hyperinflammatory conditions strongly associated with autoimmunity [12]. Altogether, the similarities between severe COVID-19 J o u r n a l P r e -p r o o f and certain autoimmune diseases could explain why both SARS-CoV2 infection and COVID-19 vaccines have been related to autoimmune disease onset, disease flares and more severe inflammatory activity during the pandemic [13][14][15]. Indeed, small-size and/or monocenter reports have claimed an increase of cases of GCA and/or associated ischemic events following COVID-19 disease and/or vaccination, although this observation has not been confirmed by others [16][17][18][19][20][21][22][23][24][25][26][27][28][29][30].
In the light of the aforementioned considerations, the aim of the present study was to assess the impact of COVID-19 pandemic and SARS-CoV-2 vaccines on the incidence of GCA and the development of GCA-associated stroke in a nation-wide analysis conducted in Spain, a 47 million population country.

Study population
We performed an analysis of data extracted from the Spanish Hospital Discharge Database Besides, considering that the main diagnosis was the defining reason for admission, all main diagnoses were decodified and four primary outcomes were analyzed: total number of admissions in patients with GCA, admissions attributable to GCA (being GCA the main diagnosis), admissions with GCA-associated stroke and admissions attributable to GCAassociated stroke (being stroke the main diagnosis). GCA-associated stroke was defined by both the presence of stroke (codes I61, I62.9 and I63), AION (code H47.01) or TIA (code G45) in GCA patients.
These outcomes were compared between the years of the pandemic (2020 and 2021) and Finally, the ICD-10 coding was also used to analyze CRF.  All statistical analyses were performed using IBM SPSS for Windows (IBM Corp, Armonk, NY). All tests were two-tailed and only p values <0.05 were considered as significant.

Population characteristics
J o u r n a l P r e -p r o o f were consistent with stroke (ischemic stroke 3.6%, TIA 1.7%, AION 2.8% and hemorrhagic stroke 0.5%). Stroke was the cause of admission in 4.8% of the cases overall. The mean average in-hospital stay was 9.3 days and the global mortality was 7% (1,206 deaths).

3.2.
Differences in the incidence of GCA and GCA-associated stroke during the pandemic. Table 2   In addition, several differences with the pre-pandemic period were found when the rates of admissions attributable to GCA and the proportions of admissions with or because of stroke were compared (table 3,

DISCUSSION
This nation-wide epidemiological study explores the incidence of GCA and the risk of GCAassociated stroke during the COVID-19 pandemic. Our results suggest that COVID-19 disease and, especially, SARS-CoV-2 vaccine, might be responsible for the higher rate of GCA-associated stroke reported in this period, leading to the increased incidence of admissions among patients with GCA during 2020 and 2021.
Previous works have already identified that COVID-19 disease and immunization, especially with mRNA-vaccines, can trigger the onset of GCA and GCA flares [16,17,19,24,25,27,28]. Overall, our results show that the incidence of GCA rose during 2020 and 2021, as previously noticed (16,24,27,28). These figures seem to be mostly determined by admissions associated with COVID-19 and by the higher incidence and proportion of admissions with or because ischemic events, supporting the hypothesized effect of both, the virus and the vaccine, on vascular events in GCA patients (19,25,26,28). Besides, this effect has not been uniform in our population during the different periods of the pandemic. In the first place, the dramatic reduction of GCA admissions during March and April 2020, when COVID- 19 was not yet included in the ICD-10 coding, was followed by an increase of cerebrovascular events in the post-lockdown period. Therefore, the diagnostic and J o u r n a l P r e -p r o o f therapeutic delay during the first wave, with a collapsed emergency system nationwide, probably led to a higher rate of subsequent complications, as other authors have already described (30,34). On the other hand, the higher incidence of GCA admissions identified in our study during the second COVID-19 outbreak following summer 2020 was mostly related to COVID-19, and not to GCA or GCA-CVA themselves.

J o u r n a l P r e -p r o o f
In Spain, the national SARS-CoV-2 vaccination campaign began at the end of December of 2020 [31]. The vaccine policies were initially focused on the elderly and on those living in nursery homes, who were particularly vulnerable to COVID-19 during the first two waves [10,32,33]. Accordingly, the protective effect was not evident until the fourth and fifth waves, as seen in our population and as reported by Barandalla et al [31]. However, a significant increase in ischemic events was identified in January-February 2021 and October-November 2021, in close temporal coincidence with the first two doses of the vaccine in patients older than 80 and the third dose in patients older than 70, respectively. Therefore, the rising rates of GCA-associated stroke identified in patients with GCA during these periods, as confirmed in the multivariate analysis, supports the role of mRNA-vaccine on the subsequent risk of ischemic phenomena in GCA [5,17,19,35].
Despite the many uncertainties in the pathophysiology of GCA, the association of the disease with certain infectious agents has been previously reported [5][6][7][8]. Analogous to other autoimmune or rheumatic diseases, the current pathogenic model accepts that GCA develops in genetically predisposed individuals exposed to a number of triggers, including infections. This hypothesis is well-accepted in other conditions such as CAPS or polyarteritis nodosa, where infections have shown to be strongly associated with the disease onset [6,36]. Both bacteria and virus has been implicated. Escherichia coli, Propionibacterium acnes, Coxiella burnetti, Parvovirus B19, Cytomegalovirus and Varicella zoster, have all been associated with GCA or granulomatous angiitis of the CNS [5][6][7][8]. The association between SARS-CoV-2 infection and/or COVID-19 vaccines and GCA or subsequent CVA, J o u r n a l P r e -p r o o f as seen in our population, seems also plausible given the proinflammatory and prothrombotic character of both the infection and the vaccine [16,17,19,37] .
We should acknowledge several limitations of the present study. Due to the database structure, essential information such as antiplatelet, anticoagulant, tocilizumab or glucocorticoid treatment before and during admission, and more detailed data about the diagnosis and clinical course of GCA, including the extent of the disease and the number or type of vessels involved, was lacking. Secondly, we were not able to retrieve information about COVID-19 infection prior to vaccination, immunological status and, importantly, about the specific vaccine given. However, despite these limitations, we believe that the size of the study population, the nationwide spectrum of the study and the statistical power of the analysis make our data clinically relevant. Moreover, the proportion of individuals effectively vaccinated within the analyzed periods, particularly within older people, was very high in Spain, most of them receiving mRNA-based vaccines [31,38]. Therefore, the assumption that most of the population of the study received such vaccines during the presumed periods of vaccination analyzed is likely to be correct. Thirdly, given the need to fit yearly time periods, the first two months of 2020 preceded the major COVID-19 outbreak in Spain.
Therefore, counting these two months within the pandemic period should have provided an underestimation of the proportion of cases of GCA and GCA-associated stroke collected in this timeframe. Trends in Figure 2 support this observation. Another caveat is the complex interaction over two years between older age, COVID-19 severity, COVID-19 vaccination, COVID-19 mortality, and GCA higher incidence. In Spain, 98,900 deaths due to COVID-19 were reported during 2020 and 2021. However, estimates derived from the excess mortality data during the same period are of 162,000 (1.64-fold greater) [39]. This huge and disproportionate high mortality mostly occurred among the elderly, the population at higher risk for of developing GCA.

J o u r n a l P r e -p r o o f
In conclusion, this nationwide epidemiological study confirms a higher risk of CVA in patients with GCA during the pandemic, probably related, at least in part, to mRNA-SARS-CoV-2 vaccines. Therefore, our findings rise concerns about the potential risk of vascular complications after SARS-CoV-2 vaccination in patients with GCA, as already suggested [16,18]. Whilst the initial indication and benefit of SARS-CoV-2 immunization in this population was obvious, given the favorable effect in terms of COVID-19 severity among elderly patients, the need for repeated doses does not seem that clear [10,31,33]. A third dose of the vaccine has been advocated to confer adequate immunization in GCA patients, many of them under methotrexate and glucocorticoid treatment which may results in lower neutralizing activity and cellular immune protection than healthy controls [40]. However, the inclusion of GCA among the potential conditions associated with the Autoimmune/inflammatory syndrome induced by adjuvants (ASIA), and, particularly, with post-covid-19 vaccination ASIA [41,42], should be also taken into account. Moreover,   (7) Admission-length stay (days) mean (SD) 9.3 (9.5) ICU admission-length stay (days) mean (SD   J o u r n a l P r e -p r o o f  Footnote: The figure represents the monthly incidence of GCA (A) and GCA-associated stroke (B) admissions during the COVID-19 pandemic in Spain. Incidence is expressed by 100,000 annual national admissions.

Figure 2. Factors related to GCA-associated stroke during the COVID-19 pandemic in Spain.
Footnote: The figure represents the binary logistic regression analysis to determine factors related to GCA -associated stroke risk during the pandemic, including the seven periods of the COVID-19 pandemic. For adjustment, age, male sex, baseline CRF and COVID-19 diagnosis during admission were considered.