key: cord-0689715-8l831mre
authors: Brenner, Erica J.; Ungaro, Ryan C.; Gearry, Richard B.; Kaplan, Gilaad G.; Kissous-Hunt, Michele; Lewis, James D.; Ng, Siew C.; Rahier, Jean-Francois; Reinisch, Walter; Ruemmele, Frank M.; Steinwurz, Flavio; Underwood, Fox E.; Zhang, Xian; Colombel, Jean-Frederic; Kappelman, Michael D.
title: Corticosteroids, but not TNF Antagonists, are Associated with Adverse COVID-19 Outcomes in Patients With Inflammatory Bowel Diseases: Results from an International Registry
date: 2020-05-18
journal: Gastroenterology
DOI: 10.1053/j.gastro.2020.05.032
sha: eb98b14541b1182a1b7e34e173a34179af033356
doc_id: 689715
cord_uid: 8l831mre

Background and Aims The impact of Coronavirus disease 2019 (COVID-19) on patients with inflammatory bowel disease (IBD) is unknown. We sought to characterize the clinical course of COVID-19 among IBD patients and evaluate the association between demographics, clinical characteristics, and immunosuppressant treatments on COVID-19 outcomes. Methods Surveillance Epidemiology of Coronavirus Under Research Exclusion for Inflammatory Bowel Disease (SECURE-IBD) is a large, international registry created to monitor outcomes of IBD patients with confirmed COVID-19. We calculated age-standardized mortality ratios (SMRs) and utilized multivariable logistic regression to identify factors associated with severe COVID-19, defined as intensive care unit admission, ventilator use, and/or death. Results 525 cases from 33 countries were reported (Median age 43 years, 53% men). Thirty-seven patients (7%) had severe COVID-19, 161 (31%) were hospitalized, and 16 patients died (3% case fatality rate). SMRs for IBD patients were 1.8 (95% confidence interval [CI] 0.9-2.6), 1.5 (95% CI 0.7-2.2), and 1.7 (95% CI 0.9-2.5) relative to data from China, Italy, and the US, respectively. Risk factors for severe COVID-19 among IBD patients included increasing age (adjusted odds ratio [aOR] 1.04, 95% CI 1.01-1.02), ≥2 comorbidities (aOR 2.9, 95% CI 1.1-7.8), systemic corticosteroids (aOR 6.9, 95% CI 2.3-20.5), and sulfasalazine or 5-aminosalicylate use (aOR 3.1, 95% CI 1.3-7.7). TNF antagonist treatment was not associated with severe COVID-19 (aOR 0.9, 95% CI 0.4-2.2). Conclusions Increasing age, comorbidities, and corticosteroids are associated with severe COVID-19 among IBD patients, although a causal relationship cannot be definitively established. Notably, TNF antagonists do not appear to be associated with severe COVID-19.

Inflammatory Bowel Disease (SECURE-IBD) is a large, international registry created to monitor outcomes of IBD patients with confirmed COVID-19. We calculated age-standardized mortality ratios (SMRs) and utilized multivariable logistic regression to identify factors associated with severe COVID-19, defined as intensive care unit admission, ventilator use, and/or death.

Results: 525 cases from 33 countries were reported (Median age 43 years, 53% men). Thirtyseven patients (7%) had severe COVID-19, 161 (31%) were hospitalized, and 16 patients died (3% case fatality rate). SMRs for IBD patients were 1.8 (95% confidence interval [CI] 0.9-2.6), 1.5 (95% CI 0.7-2.2), and 1.7 (95% CI 0.9-2.5) relative to data from China, Italy, and the US, respectively. Risk factors for severe COVID-19 among IBD patients included increasing age (adjusted odds ratio [aOR] 1.04, 95% CI 1.01-1.02), ≥2 comorbidities (aOR 2.9, 95% CI 1. 1-7.8) , systemic corticosteroids (aOR 6.9, 95% CI 2.3-20.5), and sulfasalazine or 5aminosalicylate use (aOR 3.1, 95% CI 1.3-7.7). TNF antagonist treatment was not associated with severe COVID-19 (aOR 0.9, 95% CI 0.4-2.2). Inflammatory bowel diseases (IBD), including Crohn's disease (CD) and ulcerative colitis (UC), are chronic inflammatory conditions of the gastrointestinal tract affecting millions of people worldwide. [5] [6] [7] Patients with IBD and related rheumatologic, dermatologic, and neurologic autoinflammatory conditions frequently require treatment with immunosuppressant medications which can increase the risk of infection. [6] [7] [8] [9] [10] Corticosteroids, immunomodulators (thiopurines, methotrexate), biologics, and janus-kinase inhibitors, commonly used to treat chronic autoinflammatory conditions, have been associated with higher rates of serious viral and bacterial infections including influenza and pneumonia. [11] [12] [13] [14] [15] Yet, it is also possible that some forms of immune suppression may blunt the excessive immune response/cytokine storm characteristic of severe COVID-19 infection and consequently reduce mortality, as suggested by emerging case reports of anti-IL-6 therapy. 16, 17 Little is known about the impact of COVID-19 on patients with chronic auto-inflammatory   diseases such as IBD, particularly those who require systemic immunosuppressant medications. An initial report of COVID-19 among 1,099 patients in China included only two persons with immune deficiency. 18 A subsequent report found that cancer patients had a higher risk of severe COVID-19, but this conclusion was based on only 16 patients. 19 In Italy, Mazza et al reported a case of COVID-19 pneumonia leading to death in a patient with severe acute ulcerative colitis treated with systemic corticosteroids. 20 In order to provide better guidance to patients and their health care providers and to inform strategies for prevention of COVID-19 and medication management, more data are urgently needed regarding the impact of IBD and treatments on COVID-19 outcomes. In the present work, we report on the clinical course of COVID-19 and risk factors for adverse outcomes in a large cohort of patients with IBD collected through an international registry. (Table S4) .

Physicians and other health care providers were encouraged to voluntarily report all cases of Polymerase Chain Reaction (PCR)-confirmed COVID-19 occurring in IBD patients, regardless of severity. To foster international collaboration and promote transparency, we developed a project website (www.covidibd.org) to acknowledge the contributions of individual reporters and share crude, aggregate data along with an interactive web-based map displaying the geographic location of reported cases (https://covidibd.org/map/).

We instructed health care providers to report cases after a minimum of 7 days from symptom onset and sufficient time had passed to observe the disease course through resolution of acute illness or death. In the event that a patient's status changed after reporting or if there were concerns about data accuracy, we instructed reporters to re-report and contact the research team to remove their initial entry.

We utilized REDCap (Research Electronic Data Capture), a secure, web-based electronic data capture tool hosted at the University of North Carolina at Chapel Hill to collect and manage study data. Health care providers recorded the following information: age, country of residence, state of residence (if applicable), year of COVID-19 diagnosis, name of center/practice/physician providing care, sex, race, ethnicity, height, weight, patient's diagnosis 

We removed all known duplicate or erroneous reports. We identified additional potential duplicate records based on matching age, sex, IBD disease type, country, and state (U.S. only), and reviewed these manually. Reports from non-valid email addresses were flagged as potential errors and we performed a Google search of reporters and practice locations to confirm legitimacy of reports.

We used descriptive statistics to summarize the basic demographic and clinical characteristics of the study population. We summarized continuous variables using means and standard deviations. We expressed categorical variables as proportions. Comorbidities were collapsed into the following categories: cardiovascular disease, diabetes, hypertension, stroke, lung disease, kidney disease, liver disease, and cancer.

We analyzed a variety of COVID-19 outcomes, including outpatient care only, hospitalization, ICU or ventilator requirement, and death from COVID-19 or related complications. Crude data are provided for the overall study population, and stratified by a variety of demographic and clinical characteristics. To understand the impact of IBD on case fatality, we computed expected and observed deaths and age-standardized mortality ratios (SMR) utilizing published agestratified COVID-19 case fatality rates from China and Italy 2, 23 and publically available data from the U.S. 24, 25 Multivariable logistic regression estimated the independent effects of age, sex, disease (CD vs UC/IBD-U), disease activity, smoking, BMI ≥30, and number of comorbidities (0, 1, ≥2) on the primary outcome of severe COVID-19, defined as a composite of ICU admission, ventilator use, and/or death, consistent with existing COVID-19 literature. 18 Models also included tumor necrosis factor (TNF) antagonist use (versus not) and sulfasalazine/5-aminosalicylate (5-ASA) use (versus not) as these were the two most commonly reported medication classes and systemic corticosteroid use (versus not) based on increased risk of infectious complications based on prior literature and crude data. A secondary outcome was the composite of any hospitalization and/or death. We also analyzed death as a separate endpoint. We reported adjusted odds ratios (aOR) and 95% confidence intervals (CI) for each demographic or disease characteristic.

We also performed a series of exploratory sub-analyses. We compared TNF antagonist monotherapy versus combination therapy with immunomodulators (6-mercaptopurine [6MP], azathioprine, or methotrexate), controlling for the above demographic and clinical factors as well as the use of systemic corticosteroids and 5-ASA/sulfasalazine. In addition, given the surprising association between 5-ASA/sulfasalazine use and more severe COVID outcomes in our main analyses, we performed a sub-analysis to directly compare the effects of TNF antagonists versus 5-ASA/ sulfasalazine, controlling for the above factors as well as use of immunomodulators. The primary outcome of these exploratory analyses was the composite of any hospitalization and/or death. The number of events was too sparse to evaluate other outcomes. All data were prepared and analyzed using SAS v 9.3 (SAS Institute, Cary, North Carolina). Two-sided p values < 0.05 were considered statistically significant.

Each SECURE-IBD survey item met criteria for de-identified data, in accordance with the HIPAA 

At the time of this writing, a total of 525 cases were reported to the SECURE-IBD database from 33 different countries and 28 states within the United States (Figures 1 and 2 ; Tables S1 and S2). Demographic, clinical, and IBD treatment related characteristics are summarized in Table   1 . The median age was 41 years, with a range from 5 to ≥ 90 years, and there was a slight predominance of males (52.6%). Most cases were reported in whites (84.2%). Ethnicity was reported as Hispanic/Latino in 14.3% of cases ( Table 1) .

The majority of patients had CD (59.4%), and IBD disease activity by PGA was classified as remission in 58.9% of cases. The most common class of IBD treatment was TNF antagonist therapy (43.4% overall, 33.5% monotherapy and 9.9% combination therapy with azathioprine, 6mercaptopurine, or methotrexate). Use of other medications is described in Table 1 . Most patients (63.4%) had no comorbidities other than IBD; 21.0% had one, 6.7% had two, and 5.5% had three or more. Four percent of the cohort reported using tobacco and/or electronic cigarettes (Table 1) .

Crude outcome data are summarized in Table 2 Table 2 .

Sixteen deaths (3% of reported cases) are summarized in Table S3 . Eight deaths (50%) occurred in patients ≥ 70 years of age. No deaths occurred in patients < 30 years of age. Most deaths had comorbidities, including eight with cardiovascular disease. The age-standardized SMRs for the SECURE-IBD population relative to China, Italy, and the U.S. were 1.8 (95% confidence interval [CI] 0.9-2.6), 1.5 (95% CI 0.7-2.2), and 1.7 (95% CI 0.9-2.5) respectively (Tables 3 and 4) .

On multivariable analysis, increasing age (aOR 1.04, 95% CI 1.01-1.06), ≥ 2 comorbidities (aOR 2.9, 95% CI 1.1-7.8), systemic corticosteroids (aOR 6.9, 95% CI 2.3-20.5), and 5-ASA/sulfasalazine use (aOR 3.1, 95% CI 1.3-7.7) were positively associated with the primary endpoint after controlling for all other covariates listed in Table 5 . No significant association was seen between TNF antagonist use and the primary endpoint (aOR 0.9, 95% CI 0.4-2.2). Similar associations were observed for our secondary outcomes, although TNF antagonist use was inversely associated with the outcome of hospitalization or death while only age and systemic corticosteroid use were positively associated with the outcome of death.

In our exploratory analyses, we found that TNF antagonist combination therapy, compared to monotherapy, was positively associated with the outcome of hospitalization or death (aOR 5.0, 95% CI 2.0-12.3), after adjusting for clinical and demographic variables and use of systemic corticosteroids and 5-ASA/sulfasalazine. Compared to TNF antagonists, 5-ASA/sulfasalazine was positively associated with the outcome of hospitalization or death (aOR 3.8, 95% CI 1.7-8.5).

We report the development of an international, physician-driven, reporting system to study the was also associated with more severe COVID-19. Reassuringly, TNF antagonist biologic therapy was not an independent risk factor for more severe COVID-19.

In this international IBD population, we observed an age-standardized mortality ratio of approximately 1.5 to 1.8, as compared to the general populations of China, Italy, and the U.S. with confidence intervals crossing the null. We note no deaths occurred in the 29 reported cases occurring in patients <20 years of age, extending the findings of an earlier case series suggesting a milder course of COVID-19 in pediatric patients. 22 In contrast, 50% of deaths occurred in patients over 70 years of age and 50% of patients who died had cardiovascular comorbidities.

The strong positive association between systemic corticosteroid use and our primary and secondary outcomes is consistent with extensive prior literature in IBD and other autoinflammatory conditions describing the infectious complications of corticosteroid use as well as more recent data indicating that corticosteroids are not beneficial, and may even be harmful, in the treatment of coronavirus and similar viruses (MERS, SARS, etc.). 26 Forty-three percent of our cohort was exposed to TNF antagonist medications. In the adjusted analysis of our primary outcome, we observed no association between TNF antagonist use and severe COVID-19. As TNF antagonists are the most commonly prescribed biologic therapy for patients with IBD, these initial findings should be reassuring to the large number of patients receiving TNF antagonist therapy and support their continued use during this current pandemic. In our exploratory subgroup analysis, we observed a higher risk of hospitalization and/or death with TNF antagonist combination therapy versus monotherapy, consistent with prior studies of other infectious complications. 12 Given the overall effect estimate of TNF antagonists (combination and monotherapy combined) in our primary model was 0.9, one can hypothesize that TNF antagonist monotherapy may have a protective effect against severe COVID-19, as suggested in a recent commentary. 27 We observed a higher risk of our primary outcome in patients exposed to 5-ASA/sulfasalazine. This finding persisted after controlling for age, comorbidities, IBD disease characteristics, corticosteroid use, and other factors. Furthermore, in a direct comparison, we observed that 5-ASA/sulfasalazine treated patients fared worse than those treated with TNF inhibitors. Although we cannot exclude unmeasured confounding, further exploration of biological mechanisms is warranted. Conversely, although the number of reported cases exposed to other IBD treatments is currently small, it is worth noting that 51/55 (93%) patients treated with anti-IL12/23 required outpatient care only and none died.

The strengths of this study include the robust, worldwide collaboration that enabled us to assemble clinical data on a large, geographically diverse sample of pediatric and adult IBD patients and rapidly define the course of COVID-19 in this population. The reporting directly by physicians or their trained medical staff strengthens the validity of these data. Although our study sample is diverse in terms of age, geography, race, and other factors, we acknowledge the possibility of reporting bias. Reported cases may over-represent more severe COVID-19 patients who come to the attention of their provider and patients in areas with readily available COVID-19 testing. Conversely, our sample may under-represent those severely ill patients who may be hospitalized at an outside hospital or die without their physician's awareness. The registry includes only confirmed cases of COVID-19 in accordance with other reporting initiatives from national authorities and the World Health Organization, 2, 4, 18 though we recognize many patients with suspected infection are never tested. Although we adjusted for many factors such as age, comorbidities, and IBD disease type and severity, we acknowledge the possibility of unmeasured confounding. Additional research is needed to further evaluate causality between the use of corticosteroids and other medications and COVID-19 outcomes.

Finally, we computed age-standardized mortality ratios using case fatality rates reported from China, Italy, and the U.S., yet our study sample arose from 31 different countries. Given the profound effects of age on COVID-19 related mortality, we believe it was useful to standardize to existing data. That our SMR estimates were roughly equivalent when standardizing to Chinese, Italian, or U.S. data suggests the overall validity of this approach.

In summary, older age, increased number of comorbidities and systemic corticosteroid use among patients with IBD are strong risk factors for adverse COVID-19 outcomes. Maintaining remission with steroid-sparing treatments will be important in managing patients with IBD through this pandemic. It appears that TNF antagonist therapy is not associated with severe COVID-19, providing reassurance that patients can continue TNF antagonist therapy.

Turner D, Huang Y, Author names in bold designate shared co-first authorship. Table 1 

Escaping Pandora's Box -Another Novel Coronavirus

Case-Fatality Rate and Characteristics of Patients Dying in Relation to COVID-19 in Italy

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72314 Cases From the Chinese Center for Disease Control and Prevention

Preliminary Estimates of the Prevalence of Selected Underlying Health Conditions Among Patients with Coronavirus Disease 2019 -United States

Worldwide incidence and prevalence of inflammatory bowel disease in the 21st century: a systematic review of population-based studies

Crohn's disease

Ulcerative colitis

ACG Clinical Guideline: Management of Crohn's Disease in Adults

Evidence-based clinical practice guidelines for inflammatory bowel disease

Second European evidence-based consensus on the prevention, diagnosis and management of opportunistic infections in inflammatory bowel disease

Infection-related hospitalizations are associated with increased mortality in patients with inflammatory bowel diseases

Risk of Serious and Opportunistic Infections Associated With Treatment of Inflammatory Bowel Diseases

Increased risk of pneumonia among patients with inflammatory bowel disease

Systematic review with meta-analysis: efficacy and safety of oral Janus kinase inhibitors for inflammatory bowel disease

The Incidence of Influenza and Influenza-related Complications in Inflammatory Bowel Disease Patients across the United States

Tocilizumab, an anti-IL6 receptor antibody, to treat Covid-19-related respiratory failure: a case report

First case of COVID-19 in a patient with multiple myeloma successfully treated with tocilizumab

Clinical Characteristics of Coronavirus Disease 2019 in China

Cancer patients in SARS-CoV-2 infection: a nationwide analysis in China

A fatal case of COVID-19 pneumonia occurring in a patient with severe acute ulcerative colitis

The data model concept in statistical mapping

A fatal case of COVID-19 pneumonia occurring in a patient with severe acute ulcerative colitis

We acknowledge the physicians and other healthcare providers worldwide who have reported cases to the SECURE-IBD database and the organizations who supported or promoted the SECURE-IBD database (Reporter names available at www.covidibd.org/reporteracknowledgment/. See Table S4 for organization names).

LIMITATIONS: Possibility of reporting bias and unmeasured confounding.

Maintaining remission with steroid-sparing treatments is important in managing IBD patients through this pandemic. TNF antagonist therapy does not appear to be a risk factor for severe COVID-19.

We created an international registry of IBD patients who developed COVID-19. Corticosteroids, but not TNF antagonists, were associated with adverse outcomes. Other risk factors were similar to the general population.