Patients on a tumor necrosis factor inhibitor for their rheumatic disease when they became infected with COVID-19 were markedly less likely to subsequently require hospitalization, according to intriguing early evidence from the COVID-19 Global Rheumatology Alliance Registry.
Dr. Jinoos Yazdany
On the other hand, those registry patients who were on 10 mg of prednisone or more daily when they got infected were more than twice as likely to be hospitalized than were those who were not on corticosteroids, even after controlling for the severity of their rheumatic disease and other potential confounders, Jinoos Yazdany, MD, reported at the virtual edition of the American College of Rheumatology’s 2020 State-of-the-Art Clinical Symposium.
“We saw a signal with moderate to high-dose steroids. I think it’s something we’re going to have to keep an eye out on as more data come in,” said Dr. Yazdany, professor of medicine at the University of California, San Francisco, and chief of rheumatology at San Francisco General Hospital.
The global registry launched on March 24, 2020, and was quickly embraced by rheumatologists from around the world. By May 12, the registry included more than 1,300 patients with a range of rheumatic diseases, all with confirmed COVID-19 infection as a requisite for enrollment; the cases were submitted by more than 300 rheumatologists in 40 countries. The registry is supported by the ACR and European League Against Rheumatism.
Dr. Yazdany, a member of the registry steering committee, described the project’s two main goals: To learn the outcomes of COVID-19–infected patients with various rheumatic diseases and to make inferences regarding the impact of the immunosuppressive and antimalarial medications widely prescribed by rheumatologists.
She presented soon-to-be-published data on the characteristics and disposition of the first 600 patients, 46% of whom were hospitalized and 9% died. A caveat regarding the registry, she noted, is that these are observational data and thus potentially subject to unrecognized confounders. Also, the registry population is skewed toward the sicker end of the COVID-19 disease spectrum because while all participants have confirmed infection, testing for the infection has been notoriously uneven. Many people are infected asymptomatically and thus may not undergo testing even where readily available.
Early key findings from registry
The risk factors for more severe infection resulting in hospitalization in patients with rheumatic diseases are by and large the same drivers described in the general population: older age and comorbid conditions including diabetes, hypertension, cardiovascular disease, obesity, chronic kidney disease, and lung disease. Notably, however, patients on the equivalent of 10 mg/day of prednisone or more were at a 105% increased risk for hospitalization, compared with those not on corticosteroids after adjustment for age, comorbid conditions, and rheumatic disease severity.
Patients on a background tumor necrosis factor (TNF) inhibitor had an adjusted 60% reduction in risk of hospitalization. This apparent protective effect against more severe COVID-19 disease is mechanistically plausible: In animal studies, being on a TNF inhibitor has been associated with less severe infection following exposure to influenza virus, Dr. Yazdany observed.
COVID-infected patients on any biologic disease-modifying antirheumatic drug had a 54% decreased risk of hospitalization. However, in this early analysis, the study was sufficiently powered only to specifically assess the impact of TNF inhibitors, since those agents were by far the most commonly used biologics. As the registry grows, it will be possible to analyze the impact of other antirheumatic medications.
Being on hydroxychloroquine or other antimalarials at the time of COVID-19 infection had no impact on hospitalization.
The only rheumatic disease diagnosis with an odds of hospitalization significantly different from that of RA patients was systemic lupus erythematosus (SLE). Lupus patients were at 80% increased risk of hospitalization. Although this was a statistically significant difference, Dr. Yazdany cautioned against making too much of it because of the strong potential for unmeasured confounding. In particular, lupus patients as a group are known to rate on the lower end of measures of social determinants of health, a status that is an established major risk factor for COVID-19 disease.
“A strength of the global registry has been that it provides timely data that’s been very helpful for rheumatologists to rapidly dispel misinformation that has been spread about hydroxychloroquine, especially statements about lupus patients not getting COVID-19. We know from these data that’s not true,” she said.
Being on background NSAIDs at the time of SARS-CoV-2 infection was not associated with increased risk of hospitalization; in fact, NSAID users were 36% less likely to be hospitalized for their COVID-19 disease, although this difference didn’t reach statistical significance.
Dr. Yazdany urged her fellow rheumatologists to enter their cases on the registry website: rheum-covid.org. There they can also join the registry mailing list and receive weekly updates.