Investigation Into the Effect of COVID-19 Infection on Length of Hospital Stay and Mortality in Patients With Rheumatoid Arthritis

Background SARS-CoV-2 (COVID-19) is a positive-stranded ribonucleic acid (RNA) virus of the coronavirus family, which has resulted in one of the most serious pandemics, with more than 14 million cases confirmed globally. Rheumatoid arthritis (RA) is estimated to be prevalent in 0.5-1% of the U.S. population. So far, there has been little evidence of COVID-19 infection and its propensity to result in increased mortality or length of hospital stay in patients with RA. To contribute to this body of literature, this study will assess the degree to which COVID-19 is associated with increased mortality and length of hospital stay in patients with RA while also taking into account these patients' comorbidities. Methods Our retrospective study included 14,180 patients (age >18, median 58, range 18-90) who tested positive for COVID-19 or were assumed to have COVID-19 infection from January 1st, 2020, through July 31st, 2020. Patients were grouped based on the diagnosis of RA and COVID-19 infection versus those without RA. Patients who were diagnosed with systemic lupus erythematosus (SLE), chronic obstructive pulmonary disease, and hypertension were excluded. Covariates included age, body mass index (BMI), race, sex, maximum C-reactive protein value, maximum D-dimer value, and comorbid diabetes mellitus. Outcome measures were length of hospital stay (LOS), in-hospital mortality, intensive care unit (ICU) admission, ICU LOS, mechanical ventilation, time on mechanical ventilation, and discharge to hospice. The logistic regression model was used to estimate the probability of in-hospital mortality, ICU admission, placement on mechanical ventilation, discharge to hospice, and in-hospital mortality related to home anti-inflammatory use when comparing patients with RA and COVID-19 infection to COVID-19 infected patients without RA. Results Of the total 14,180 patients (males 57.1%, females 42.9%), 159 patients (1.1%), had a diagnosis of RA. There was no significant association between RA and hospital LOS, ICU admission, ICU LOS, LOS on mechanical ventilation, or discharge to hospice among those infected with COVID-19. Yet, RA was associated with higher mortality (OR: 1.65; 95% CI: 1.07-2.53; p=0.02) and placement on mechanical ventilation (OR: 1.82; 95% CI: 1.22-2.71; p<0.01) amidst patients infected with COVID-19. Conclusion This study suggests that patients with RA and COVID-19 have a significantly increased likelihood of in-hospital mortality and placement on mechanical ventilation. While challenging to realize in a pandemic situation, large studies nationwide are necessary to improve our understanding of COVID-19 infection in patients diagnosed with RA.


Introduction
SARS-CoV-2 is a positive-stranded ribonucleic acid (RNA) virus of the coronavirus family, which has resulted in one of the most serious pandemics, with more than 14 million cases confirmed globally. The range of associated symptoms is diverse; the most frequently observed are cough, fever, myalgia, headache, and dyspnea [1]. Some of the most common medical comorbidities associated with severe illness and mortality are cardiovascular disease, diabetes mellitus, hypertension, chronic lung disease, cancer, and chronic kidney disease [2][3][4][5]. Rheumatoid arthritis (RA) is estimated to be prevalent in 0.5-1% of the U.S. population [6]. As seen in the population-based study by Doran [7], there is an increased risk of serious infection from diseases in patients with RA compared to normal subjects (rate ratio [RR]: 1.53, 95% CI: 1.41-1.65). The increased risk of infection was largely attributed to the fact that patients with RA have an immune system that does not function optimally. This increased rate of infection is dependent on the degree of RA disease activity, which is evident in the study by Au and colleagues demonstrating that for each 0.6 unit rise in Disease Activity Score 28, there was a parallel increase in infection requiring hospitalization of 25% (incident rate ratio 1.25, p=0.03) [8].
In addition to the nature of RA's effect on immune function in itself, one must also account for the influence of standard RA treatment such as systemic corticosteroids, biologic disease-modifying antirheumatic drugs (bDMARDs), and conventional synthetic disease-modifying antirheumatic drugs (csDMARDs). The predilection of systemic corticosteroids to increase the risk of viral infection in patients with RA is appreciated in a retrospective-cohort study by Widdifield and associates, which revealed an increased risk of herpes-zoster infection with an incidence rate of 8.54 cases per 1,000 patient-years in patients treated with corticosteroids [9]. The use of csDMARDs in patients with RA may result in a higher chance of contracting a viral infection. The retrospective-population based study by Lacaille et al. demonstrates that when patients with RA are treated with csDMARD therapy, there is a small decrease in risk of mild infection (RR: 0.90, 95% CI: 0.88-0.93) and no significant association of serious infection risk (RR: 0.92, 95% CI: 0.85-1.00) [10]. So far, there has been little evidence of COVID-19 infection and its propensity to result in increased mortality or length of hospital stay in patients with RA. Due to this lack of data, the focus of our study will be to measure the degree to which COVID-19 is associated with increased mortality and length of hospital stay in patients with RA while also taking into account these patients' comorbidities.

Materials And Methods
Our retrospective study included 14,180 patients (age >18, median 58, range 18-90) who tested positive for COVID-19 or were assumed to have COVID-19 infection on admission to Hospital Corporation of America (HCA) hospitals from January 1, 2020, through July 31, 2020. Of the total 14,180 patients (males 57.1%, females 42.9%), 159 patients (1.1%), had a diagnosis of RA. Patients were grouped based on the diagnosis of RA and COVID-19 infection versus those without RA. Patients who were diagnosed with systemic lupus erythematosus (SLE), chronic obstructive pulmonary disease, and hypertension were excluded. Pregnant patients and those with no C-reactive protein (CRP) or D-dimer laboratory results were also excluded from the population. Covariates included age, body mass index (BMI), race, sex, maximum C-reactive protein value, maximum D-dimer value, and comorbid diabetes mellitus (DM). Patients recorded the use of home medications which included corticosteroids, csDMARDs, and biologic disease-modifying antirheumatic drugs (bDMARD). Outcome measures were length of hospital stay (LOS), in-hospital mortality, intensive care unit (ICU) admission, ICU LOS, mechanical ventilation, time on mechanical ventilation, and discharge to hospice. Patients' demographic and clinical characteristics are presented in Table 1.  Linear regression was used to determine the difference in hospital, ICU, and mechanical ventilation LOS between patients with COVID-19 infection and COVID-19 infection with RA while controlling for age, sex, race, BMI, maximum D-dimer lab value, maximum CRP lab value, and reported use of home medications. The logistic regression model was used to estimate the probability of in-hospital mortality, ICU admission, placement on mechanical ventilation, discharge to hospice, and in-hospital mortality related to home antiinflammatory use when comparing patients with RA and COVID-19 infection to COVID-19 infected patients without RA. Continuous variables were reported as means and standard deviation (SD) or as median and range values. Categorical variables were expressed as counts and percentages. We used a chi-squared test to compare variables between COVID-19 positive patients with or without rheumatoid arthritis for categorical data. In-hospital mortality, hospital length of stay, ICU admission, ICU length of stay, mechanical ventilator usage, time on a ventilator, and discharge to hospice were calculated and adjusted for patient demographics and comorbidities through multivariate regression analysis. All statistical analyses were performed using SAS software (version 9.4; SAS Institute Inc., Cary, USA), with p<0.05 considered as statistically significant.

Mechanical ventilation placement and LOS
The results illustrate that 14.1% of hospitalized patients with COVID-19 and no RA were placed on mechanical ventilation, while 22.6% of those with COVID-19 and RA required ventilation. There is statistically significant evidence that patients with RA are more likely to be placed on a ventilator than patients without RA Results of the statistical analysis are presented in Table 2.  It is important to note that patients with reported use of home medications, such as DMARDs and corticosteroids, were less likely to experience hospital mortality. Our study demonstrated that patients with reported use of anti-inflammatory medications had a statistically significantly decreased risk of in-hospital mortality compared to patients who did not report using those drugs. The decreased risk of mortality in this group of patients could be due to superior control of the destructive inflammatory response often seen in patients with severe COVID-19 infection. An example of anti-inflammatory control of the COVID-19 inflammatory response is illustrated in the RECOVERY group study, which tests the use of dexamethasone in COVID-19 infected patients requiring oxygen or mechanical ventilation. The incidence of death in COVID-19 infected patients on mechanical ventilation was much lower in patients receiving dexamethasone compared to the usual care (29.3% vs. 41.4%; RR: 0.64; 95% CI: 0.51-0.81), as well as infected patients requiring oxygen 23.3% vs. 26.2%; RR: 0.82; 95% CI: 0.72-0.94) [11]. However, when answering the question of whether RA patients being preemptively treated with DMARDs will result in less incidence of COVID-19 infection, we see in the study by Favalli et al. that the incidence is consistent with the general population (0.62% vs. 0.66%; p=0.92) [12].
In the peer-reviewed literature review, few studies have explored the relationship between patients with RA and COVID-19 infection and their propensity to result in suboptimal hospital outcomes. One study in particular that came close to describing this relationship was the OpenSAFELY study by Williamson and colleagues [5]. The study utilized the OpenSAFELY electronic medical record (EMR), which contained the electronic record of over 17 million patients in England. They analyzed the data to determine factors related to COVID-19 mortality. Some of our study limitations include a relatively small sample size of our population of patients with RA and COVID-19, resulting in 159 patients. Secondly, another potential drawback was that although rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) were obtained, these values were not systematically checked for every patient admitted with RA. It was difficult to distinguish whether these patients were seronegative or seropositive as no outpatient data was obtained. Finally, given that our data was acquired during a relatively early period of the pandemic (01/01/20 -07/31/20) when vaccines were not available, we did not take into account the protective effect of COVID-19 vaccination on patients with RA.
Despite our study limitations, a potential implication would be to include RA as established comorbidity and an independent risk factor that could result in a higher likelihood of COVID-19 related patient morbidity or mortality. Therefore, it could be even more vital for this patient population to receive vaccination, as the phase III placebo-controlled trial for Pfizer vaccine by Polak et al. illustrates that the vaccine provides 95% efficacy when preventing COVID-19 symptoms at least seven days after the second dose [13].

Conclusions
This study suggests that patients with RA and COVID-19 have a significantly increased likelihood of inhospital mortality and placement on mechanical ventilation. While challenging to determine in a pandemic situation, large nationwide studies are necessary to improve our understanding of COVID-19 infection in patients diagnosed with RA. A potential follow-up study could be to investigate how the prophylactic use of anti-inflammatory medications can possibly reduce in-hospital mortality of patients who eventually become infected with COVID-19.