<strong class="sub-title"> Backgroun<em>d</em>: </strong> Coronavirus <em>d</em>isease <em>2</em>019 (COVID-19) is associate<em>d</em> with a high <em>d</em>isease bur<em>d</em>en with 10% of confirme<em>d</em> cases progressing towar<em>d</em>s critical illness. Nevertheless, the <em>d</em>isease course an<em>d</em> pre<em>d</em>ictors of mortality in critically ill patients are poorly un<em>d</em>erstoo<em>d</em>.
Methods: Following the critical developments in ICUs in regions experiencing early inception of the pandemic, the European-based, international RIsk Stratification in COVID-19 patients in the Intensive Care Unit (RISC-19-ICU) registry was created to provide near real-time assessment of patients developing critical illness due to COVID-19.
<strong class="sub-title"> Fin<em>d</em>ings: </strong> As of April <em>2</em><em>2</em>, <em>2</em>0<em>2</em>0, 639 critically ill patients with confirme<em>d</em> SARS-CoV-<em>2</em> infection were inclu<em>d</em>e<em>d</em> in the RISC-19-ICU registry. Of these, 398 ha<em>d</em> <em>d</em>ecease<em>d</em> or been <em>d</em>ischarge<em>d</em> from the ICU. ICU-mortality was <em>2</em>4%, me<em>d</em>ian length of stay 1<em>2</em> (IQR, 5-<em>2</em>1) <em>d</em>ays. ARDS was <em>d</em>iagnose<em>d</em> in 74%, with a minimum P/F-ratio of 110 (IQR, 80-148). Prone positioning, ECCO<em>2</em>R, or ECMO were applie<em>d</em> in 57%. Off-label therapies were prescribe<em>d</em> in <em>2</em>65 (67%) patients, an<em>d</em> 89% of all bloo<em>d</em>stream infections were observe<em>d</em> in this subgroup (<i>n</i> = 66; RR=3·<em>2</em>, 95% CI [1·7-6·0]). While PCT an<em>d</em> IL-6 levels remaine<em>d</em> similar in ICU survivors an<em>d</em> non-survivors throughout the ICU stay (<i>p</i> = 0·35, 0·34), CRP, creatinine, troponin, <em>d</em>-<em>dimer</em>, lactate, neutrophil count, P/F-ratio <em>d</em>iverge<em>d</em> within the first seven <em>d</em>ays (<i>p</i><0·01). On a multivariable Cox proportional-hazar<em>d</em> regression mo<em>d</em>el at a<em>d</em>mission, creatinine, <em>d</em>-<em>dimer</em>, lactate, potassium, P/F-ratio, alveolar-arterial gra<em>d</em>ient, an<em>d</em> ischemic heart <em>d</em>isease were in<em>d</em>epen<em>d</em>ently associate<em>d</em> with ICU-mortality.
<strong class="sub-title"> Interpretation: </strong> The European RISC-19-ICU cohort <em>d</em>emonstrates a mo<em>d</em>erate mortality of <em>2</em>4% in critically ill patients with COVID-19. Despite high ARDS severity, mechanical ventilation inci<em>d</em>ence was low an<em>d</em> associate<em>d</em> with more rescue therapies. In contrast to risk factors in hospitalize<em>d</em> patients reporte<em>d</em> in other stu<em>d</em>ies, the main mortality pre<em>d</em>ictors in these critically ill patients were markers of oxygenation <em>d</em>eficit, renal an<em>d</em> microvascular <em>d</em>ysfunction, an<em>d</em> coagulatory activation. Elevate<em>d</em> risk of bloo<em>d</em>stream infections un<em>d</em>erscores the nee<em>d</em> to exercise caution with off-label therapies.
Keywords: Acute respiratory distress syndrome; COVID-19; Coronavirus; Pandemic; Public health.