key: cord-0938518-3pmus4vq
authors: Bournia, V.-K.; Fragoulis, G.; Mitrou, P.; Mathioudakis, K.; Tsolakidis, A.; Konstantonis, G.; Tseti, I.; Vourli, G.; Tektonidou, M. G.; Paraskevis, D.; Sfikakis, P. P.
title: Different Covid-19 Outcomes Among Systemic Rheumatic Diseases: A Nation-wide Cohort Study
date: 2022-03-12
journal: nan
DOI: 10.1101/2022.03.11.22271887
sha: 5636f8092d5ef6f78bf14cc8a236b328f593343a
doc_id: 938518
cord_uid: 3pmus4vq

Background: Nationwide data at a country level on Covid-19 in unvaccinated patients with rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), systemic lupus erythematosus (SLE) and systemic sclerosis (SSc) are scarce. Methods: By interlinking data from national electronic registries, covering nearly 99% of the Greek population (approximately 11,000,000), between March 2020 and February 2021, when vaccination became available, we recorded confirmed infections and Covid-19-associated hospitalizations and deaths in essentially all adult patients with RA, AS, PsA, SLE, and SSc under treatment (n=74,970, median age of 67.5, 51.2, 58.1, 56.2, 62.2 years, respectively) and in individually matched (1:5) on age, sex, and region of domicile random comparators from the general population. Results: Binary logistic regression analysis after adjusting for age, sex and biologic agents, revealed that RA, PsA, SLE and SSc, but not AS patients, had significantly higher risk of infection (by 43%, 25%, 20% and 49%, respectively), and hospitalization for Covid-19 (by 81%, 56%, 94%, and 111%, respectively), possibly due, at least in part, to increased testing and lower threshold for admission. Patients with RA and SSc had indeed higher Covid-19 associated mortality rates [OR:1.86 (95% CI 1.37 to 2.52) and OR:2.90 (95% CI 0.97 to 8.67), respectively] compared to the general population. Each additional year of age increased the risk of hospitalization for Covid-19 by 3% (OR 1.030, 95% CI: 1.028 to 1.034) and the risk of Covid-19 related death by 8% (OR 1.08, 95% CI: 1.07 to 1.09), independently of gender, systemic rheumatic disease, and biologic agents. A further analysis using AS patients as the reference category, adjusting again for age, sex and use of biologic agents showed that patients with SSc had increased mortality (OR: 6.90, 95% CI: 1.41 to 33.72), followed by SLE (OR: 4.05 95% CI: 0.96 to 17.12) and RA patients (OR: 3.65, 95% CI: 1.06 to 12.54), whereas PsA patients had comparable mortality risk with AS patients. Conclusion: Comparing to the general population, Covid-19 may have a more severe impact in real-world patients with systemic rheumatic disease. Covid-19 related mortality is increased in subgroups of patients with specific rheumatic diseases, especially in older ones, underscoring the need for priority vaccination policies and access to targeted treatments.

The first confirmed case of Covid-19 in Greece was identified on 26 February 2020. 

In this nationwide, population-based cohort study we identified all adult patients with RA, AS, PsA, SLE, and SSc alive on 1-March-2020 and matched each patient to 5 random referents from the general population for gender, age and region of domicile.

For this purpose, we used our published data [13] derived from the electronic prescription database for social security services (IDIKA) including all adult (aged ≥ 18years old) patients who had filled at least one prescription for corticosteroids, conventional synthetic disease modifying anti-rheumatic drugs (DMARDs), immunosuppressants, biologic DMARDs, targeted synthetic DMARDs, advanced vasodilatory medications or antifibrotic agents between 1-January-2015 and 31-December-2019 with a diagnosis of either RA, AS, PsA, SLE or SSc, based on prespecified for each disease ICD-10 codes. To avoid selection bias, we limited our requirements for inclusion in the cohort to one filled prescription with any of the . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022. ; https://doi.org/10.1101/2022.03.11.22271887 doi: medRxiv preprint above-mentioned medications of interest between 2015-2019. Therefore, only a very small minority of patients consistently buying their rheumatology drugs over the counter for 5 consecutive years would not have been captured.

For all subjects in our cohort, we retrieved data on age, gender, use of biologic agents, and by crosslinking with the national Covid-19 registry, data on SARS-Cov-2 infection confirmed by reverse-transcriptase-polymerase-chain-reaction or antigendetecting rapid diagnostic testing, as well as Covid-19-associated hospitalization and death, during the study period. Following approval of our formal request, the Greek February 2021 are also shown in Table 1 . As depicted in Figure 1 , binary logistic regression analysis performed in the entire cohort revealed that patients collectively, had 34% higher risk to get infected (OR 1.34, 95% CI: 1.24 to 1.44), 79% higher risk to get hospitalized (OR 1.79, 95% CI:

1.59 to 2.02) and 77% higher risk to die (OR 1.77, 95% CI: 1.36 to 2.30), compared to their matched referents from the general population.

Each additional year of age increased the risk of hospitalization for Covid-19 by 3% (OR 1.030, 95% CI: 1.028 to 1.034) and the risk of Covid-19 related death by 8% (OR 1.08, 95% CI: 1.07 to 1.09), independently of gender, systemic rheumatic disease, and biologic agents.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022. (Figure 1 ).

To investigate for potential differences between systemic rheumatic diseases we performed a subgroup analysis within patients infected with SARS-Cov-2, using AS 3.65, 95% CI: 1.06 to 12.54 for death), whereas PsA patients did not have significantly higher risk of hospitalization and death in comparison with AS patients (Figure 2 ).

Our 12-month study, analysing data derived from the inter-linkage of large nationwide databases covering the entire Greek population, shows that unvaccinated patients with RA, PsA, SLE and SSc, but not AS, have a higher risk of SARS-Cov-2 infection and hospitalization due to Covid-19 compared to the general population.

Patients with RA and SSc had also higher Covid-19 associated mortality rates compared to the general population. The analysis of the entire cohort revealed an inverse correlation between age and the risk of SARS-Cov-2 infection, which is not surprising because of the increased protection measures taken by older subjects.

Notably, in this same analysis each additional year of age increased the risk of Covid-19-related death by 8%, irrespective of gender and disease. Therefore, since we were . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022. ; https://doi.org/10.1101/2022.03.11.22271887 doi: medRxiv preprint not able to correct for comorbidities which are known to be increased in any systemic rheumatic disease, the higher mortality in RA and SSc patients, who are older than patients with AS, PsA and SLE, could be partly explained by the fact that the burden of comorbidities could have a synergistic effect on Covid-19 outcomes with advanced age.

To the best of our knowledge only one previous nation-wide study assessing mortality risk of patients with chronic inflammatory arthritis versus the general population has been published [14] . This was a 6-month study in Sweden, a unique country where public measures were far less strict, especially during the first pandemic wave, compared to the rest of the world. The authors reported that death rates were increased when adjusting only for age, sex and region of domicile, but this risk was mitigated when additionally adjusting for comorbidities and socioeconomic factors. Regarding the risk of Covid-19 infection, with the exception of AS patients, all other patient subgroups were at higher risk, in agreement also with a general populationbased cohort study [18] and two recent-metanalyses [4, 5] . However, the possibility that these patients were more susceptible to increased rates of testing cannot be excluded. Therefore, whether a tendency to contract the infection more easily, together with the increased comorbidity burden, can explain the higher Covid-19 associated death risk found collectively in systemic rheumatic disease patients compared to the general population needs further study [19] .

As regards to hospitalization rates, there are three nation-wide studies from . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022.

In general, findings about Covid-19 related outcomes in systemic rheumatic disease patients should be interpreted with caution. Inconsistencies between various studies may pertain to ethnic or racial differences [28] , to the heterogeneity of systemic rheumatic disease and their treatment, differences in the intensity of pandemic waves and the capacity of health care system, but also to the different study designs or timing of data acquisition with regard to the implementation of patient vaccination policies. That said, we should note that our study was performed before vaccination was available among systemic rheumatic disease patients in Greece.

A recently published study also raised some concerns about high rate of biases (e.g participation and ascertainment bias) occurring in the studies being published about this topic [29] . This was also noted in a recent systematic literature review performed to inform respective EULAR recommendations [3] . In the strengths of our study, one can include the following: firstly, it is a nationwide study examining the whole population of our country (approximately 11 million people) during the whole first

year of the pandemic. To further strengthen our methodology, matching with the general population was adjusted for area of residence, limiting biases concerning access to healthcare facilities and regional differences in Covid-19 incidence rate.

Thus, selection bias occurring in registries is likely eliminated. Also, in contrast to other studies, we have examined simultaneously five major systemic rheumatic disease in the same population which also allowed us to make comparison between diseases. Indeed, our data show that patients with SSc have the worse Covid-19outcomes, followed by SLE and RA, PsA and AS.

One major weakness of our study, as mentioned above, is the inability to adjust our findings for disease duration, disease activity and the presence of comorbidities.

However, in the sense that comorbidities now tend to be seen as an inherent component of systemic rheumatic disease, we believe that it is hard to decipher whether the higher Covid-19 associated death rate observed in certain of these patient' subgroups should be attributed to the systemic rheumatic disease per se or to the concomitant presence of other diseases.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022. ; Table 1 . Demographics and number of SARS-CoV-2 infections, number of Covid-19-associated hospitalizations and deaths, and number of deaths from all causes recorded between 1-3-2020 and 28-2-2021 among patients with rheumatoid arthritis (RA), ankylosing spondylitis (AS), psoriatic arthritis (PsA), systemic lupus erythematosus (SLE), systemic sclerosis (SSc) and their matched population referents. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted March 12, 2022. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 12, 2022. ; https://doi.org/10.1101/2022.03.11.22271887 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted March 12, 2022. ;

Comparison of the characteristics, morbidity, and mortality of COVID-19 and seasonal influenza: a nationwide, population-based retrospective cohort study

Geoepidemiology of autoimmune rheumatic diseases

Risk and prognosis of SARS-CoV-2 infection and vaccination against SARS-CoV-2 in rheumatic and musculoskeletal diseases: a systematic literature review to inform EULAR recommendations