Key points are not available for this paper at this time.
Background: Individuals with inflammatory rheumatic and musculoskeletal diseases (iRMDs) face a significantly higher risk of adverse COVID-19 outcomes, with increased rates of hospitalization, intensive care unit (ICU) admission, and mortality, compared to the general population 1. However, individual factors influencing the specific outcomes for iRMD patients in ICUs are incompletely understood. Objectives: This study aims to identify factors associated with adverse outcomes in iRMD patients with COVID-19 admitted to the ICU. Methods: We conducted a retrospective study at a large tertiary university hospital, analyzing patients admitted from March 2020 to December 2021. The study involved a total of 1562 critically ill patients with COVID-19, including 98 iRMD patients and 1464 non-iRMD patients. Multivariable Cox proportional regression analysis was used to evaluate factors associated with mortality in the iRMD group. We also examined factors linked to prolonged ICU stays (>30 days) among survivors. Furthermore, we compared the differences between factors associated with mortality between the iRMD and the general cohort. Results: Mean age in the iRMD group was 48.1 ± 14.5 years with an average disease duration of 16.3 ± 9.9 years, of which 51% were female. Rheumatoid arthritis was the most frequent diagnosis among iRMD patients (58%), followed by vasculitides (13%) and connective tissue diseases (7%). Within the iRMD cohort 81% required mechanical ventilation, 33% died, and 24% of the survivors experienced prolonged ICU stays. Multivariable analysis based on pre-selected factors (age, body mass index (BMI), sex, presence of chronic kidney disease, diabetes mellitus, vaccination status, virus variant, peak Interleukin (IL)-6 levels, latest glucocorticoid dose and use of csDMARDs) identified BMI (HR 1.11, CI 1.01-1.21, p=0.030), male sex (HR 14.47, CI 1.27-165.26, p=0.031), peak IL-6 levels (HR 1.05, CI 1.02-1.09, p=0.001) as significant factors associated with mortality. Factors such as vaccination status or iRMD specific treatment did not independently affect mortality risk. Prolonged ICU stay was associated with high IL-6 (by 1000, HR 1.37, CI 1.02-1.87, p=0.004), and general ICU complications such as delirium (HR 7, CI 1.89-25.97, p=0.004), pulmonary embolism (HR 6.6, CI 1.55-28.16, p=0.011), and major cardiac events (HR 11.67, CI 2.50-54.54, p=0.002) in univariable regression analysis. Compared with the general ICU cohort, mortality in the iRMD group was more likely to be associated with three or more COVID-19 vaccinations (HR 6.31, CI 1.32-29.65, p=0.020) and peak IL-6 levels (HR 1.03, CI 1.06-25.97, p=0.041), otherwise the factors influencing mortality were similar. Conclusion: In iRMD patients, higher BMI, male sex, and elevated IL-6 levels are key indicators of a poorer prognosis in the ICU setting for COVID-19. Apart from IL-6, these factors and the outcomes for iRMD patients do not significantly differ from the general ICU population, highlighting the impact of hyperinflammation. REFERENCES: 1 Conway R, Grimshaw AA, Konig MF, Putman M, Duarte-García A, Tseng LY, Cabrera DM, Chock YPE, Degirmenci HB, Duff E, Egeli BH, Graef ER, Gupta A, Harkins P, Hoyer BF, Jayatilleke A, Jin S, Kasia C, Khilnani A, Kilian A, Kim AHJ, Lin CMA, Low C, Proulx L, Sattui SE, Singh N, Sparks JA, Tam H, Ugarte-Gil MF, Ung N, Wang K, Wise LM, Yang Z, Young KJ, Liew JW, Grainger R, Wallace ZS, Hsieh E; COVID-19 Global Rheumatology Alliance. SARS-CoV-2 Infection and COVID-19 Outcomes in Rheumatic Diseases: A Systematic Literature Review and Meta-Analysis. Arthritis Rheumatol. 2022 May;74(5):766-775. doi: 10.1002/art.42030. Epub 2022 Mar 28. PMID: 34807517; PMCID: PMC9011807. Acknowledgements: NIL. Disclosure of Interests: Edgar Wiebe honoraria from Novartis, Frédéric Münch: None declared, Nadège Léprêtre: None declared, Jens Klotsche: None declared, Jan-Hendrik Bernhard Hardenberg: None declared, Kerstin Rubarth: None declared, Mirja Mittermaier: None declared, Felix Balzer: None declared, Robert Biesen: None declared, Gerhard Krönke: None declared, Kai-Uwe Eckardt: None declared.
Wiebe et al. (Sat,) studied this question.