Prognostic value of estimated glomerular filtration rate in hospitalised older patients (over 65) with COVID-19: a multicentre, European, observational cohort study
Ben Carter, Euan A. Ramsay, Roxanna Short, Sarah Goodison, Jane Lumsden, Amarah Khan, Philip Braude, Arturo Vilches‐Moraga, Terence J. Quinn, Kathryn McCarthy, Jonathan Hewitt, Phyo Kyaw Myint, On Behalf of COPE Study, Eilidh Bruce, Alice Einarsson, Kirsty McCrorie, Ken Aggrey, Jimmy Bilan, Kerr Hartrop, Caitlin C. Murphy, Aine McGovern, Enrico Clini, Giovanni Guaraldi, Alessia Verduri, Carly Bisset, Ross Alexander, J. Daniel Kelly, Caroline Murphy, Tarik El Jichi Mutasem, Sandeep Singh, Dolcie Paxton, Will Harris, Susan Moug, Norman Galbraith, Emma Bhatti, Jenny Edwards, Siobhan Duffy, Maria Fernanda Ramon Espinoza, Thomas Kneen, Anna Dafnis, Hala Allafi, Maria Narro Vidal, Angeline Price, Lyndsay Pearce, Thomas Lee, Shefali Sangani, Madeline Garcia, Charlotte Davey, S Jones, Kiah Lunstone, A. J. M. Cavenagh, Charlotte Silver, Thomas Telford, Rebecca K. Simmons
Abstract
BACKGROUND: The reduced renal function has prognostic significance in COVID-19 and it has been linked to mortality in the general population. Reduced renal function is prevalent in older age and thus we set out to better understand its effect on mortality. METHODS: Patient clinical and demographic data was taken from the COVID-19 in Older People (COPE) study during two periods (February-June 2020 and October 2020-March 2021, respectively). Kidney function on admission was measured using estimated glomerular filtration rate (eGFR). The primary outcomes were time to mortality and 28-day mortality. Secondary outcome was length of hospital stay. Data were analysed with multilevel Cox proportional hazards regression, and multilevel logistic regression and adjusted for individual patient clinical and demographic characteristics. RESULTS: One thousand eight hundred two patients (55.0% male; median [IQR] 80 [73-86] years) were included in the study. 28-day mortality was 42.3% (n = 742). 48% (n = 801) had evidence of renal impairment on admission. Using a time-to-event analysis, reduced renal function was associated with increased in-hospital mortality (compared to eGFR ≥ 60 [Stage 1&2]): eGFR 45-59 [Stage 3a] aHR = 1.26 (95%CI 1.02-1.55); eGFR 30-44 [Stage 3b] aHR = 1.41 (95%CI 1.14-1.73); eGFR 1-29 [Stage 4&5] aHR = 1.42 (95%CI 1.13-1.80). In the co-primary outcome of 28-day mortality, mortality was associated with: Stage 3a adjusted odds ratio (aOR) = 1.18 (95%CI 0.88-1.58), Stage 3b aOR = 1.40 (95%CI 1.03-1.89); and Stage 4&5 aOR = 1.65 (95%CI 1.16-2.35). CONCLUSION: eGFR on admission is a good independent predictor of mortality in hospitalised older patients with COVID-19 population. We found evidence of a dose-response between reduced renal function and increased mortality.