MR-proAdrenomedullin as predictor of renal replacement therapy in a cohort of critically ill patients with COVID-19.

Roedl, Kevin; Jarczak, Dominik; Fischer, Marlene; Haddad, Munif; Boenisch, Olaf; de Heer, Geraldine; Burdelski, Christoph; Frings, Daniel; Sensen, Barbara; Karakas, Mahir; Kluge, Stefan; Nierhaus, Axel

BACKGROUND Coronavirus Disease 2019 (COVID-19), caused by the novel coronavirus (SARS-CoV-2), has led to a large number of patients having to be admitted to intensive care because of respiratory failure and multi-organ dysfunction. About 20% of critically ill patients with COVID-19 require renal replacement therapy (RRT). However, data on the prediction of acute kidney injury requiring RRT is lacking. Novel biomarkers, such as mid-regional pro-adrenomedullin (MR-proADM) might be likely candidates for risk assessment due to its involvement during the initial phase of multiple organ dysfunction. Therefore, the aim of this study was to investigate MR-proADM for prediction of RRT in critically ill patients with COVID-19. METHODS We analysed data of all consecutive patients with microbiologically confirmed COVID-19, requiring ICU admission at one of 12 intensive care units at a tertiary care university hospital in Germany between March and September 2020. Clinical characteristics including details on acute kidney injury, RRT, and survival were assessed. MR-proADM was measured as part of routine lab monitoring on admission to ICU. RESULTS 64 patients were included in the study. The median age was 62.5 (54 - 73) years; 49 patients (77%) were male. 47 (73%) patients were mechanically ventilated and 50 (78%) needed vasopressor therapy. 25 (39%) patients suffered from severe ARDS; veno-venous extracorporeal membrane oxygenation was established in 10 patients. 29 (45%) patients required RRT during the ICU-stay; Median time from ICU admission to start of RRT was 2 (1 - 9) days. MR-proADM on ICU admission was significantly higher in patients requiring RRT (2.491 vs. 1.230 nmol/l; p = 0.002) and showed the highest correlation with SOFA renal subscore on day 1. ROC curve analysis showed that MR-proADM predicts RRT with an AUC of 0.685 (95% CI: 0.543 - 0.828; p = 0.019). In multivariable logistic regression MR-proADM was an independent predictor (OR: 3.813, 95% CI 1.110 - 13.102, p < 0.05) for the requirement of RRT. CONCLUSION Acute kidney injury requiring RRT is a frequent finding in critically ill patients with COVID-19. MR-proADM on admission to ICU was able to predict later requirement for RRT, which may be of interest for early risk stratification and patient management.