PREDICTORS of MORTALITY in SEVERE and CRITICAL COVID-19 PATIENTS WHO RECEIVED ANAKINRA

ABSTRACT

Background: Several predictive factors were described in COVID-19 during the pandemic, however there is limited data in severe/critically ill patients (pts) with COVID-19 who received biologic therapies including anakinra. Objectives: We aimed to evaluate the predictive factors of mortality in pts with severe and critically ill COVID-19 who received anakinra. Methods: Diagnosis of COVID-19 was confrmed by PCR and/or typical CT fnd-ings. Severe and critically ill pts according to NIH severity scale who received anakinra in the ward were evaluated retrospectively (1). Laboratory values at admission, highest levels and the last day of hospitalization were recorded. COVID hyperinfammatory syndrome score (cHIS) was calculated according to the highest levels of laboratory results (2). All pts received background steroid therapy with 80 mg methylprednisolone (or its equivalent). Anakinra was started in pts who did not respond to steroid therapy at least two days or concomitantly with steroids in pts with higher risk and/or critical illness at admission. Average starting dose of anakinra was 600 mg/day intravenously and increased gradually to 1600 mg/day if necessary. Results: Data of 148 pts (53 % male) were analyzed. Of those 57 pts (38.5 %) severe, 91 pts (61.5 %) had critical disease. Overall, 56 pts (37.8 %) died during the follow-up and intensive care unit admission was in 60 pts (40.5 %) and intubation in 54 pts (54.5 %). Diabetes mellitus was in 28 %, hypertension in 59 %, coronary heart disease in 19 %, heart failure in 12.6 %, chronic kidney failure in 21 %, chronic obstructive pulmonary disease in 16 %, dementia in 12.8 %, malignancy in 11 % and rheumatic disease in 5.6 % of pts. Only dementia signifcantly differed between pts had mortality and had not (p=0.005 OR9.8). In univariable analysis;patient age, neutrophil to lymphocyte ratio (NLR), mean cHIS scores were higher in pts had mortality. Higher baseline, maximum and last values of CRP, LDH, ferritin, D-dimer levels were observed in pts had mortality. Mortality was also higher in pts with critically ill compared to severe disease In multivariate analysis;higher age (p=0.01 OR1.05 CI 1.01-1.09), cHIS score (p=0.002 OR2.6 CI1.4-4.9), critical illness compared to severe disease (p=0.02 OR14 CI1.6-122) were associated with mortality. In ROC analysis;cut-of values for cHIS score (3.5), NLR (7.1), CRP (160.5 mg/L), LDH (581 U/L), ferritin (771 ng/mL), D-dimer (7.56 mcg/mL) with 72/72, 69/70, 70/69, 72/75, 70/72, 70/73.4 sensitivity/specifcity were calculated, respectively. Conclusion: In our study mortality was developed in third of anakinra receiving pts. Mortality was independently associated with advanced age, critical illness and higher cHIS score refecting higher infammatory burden. Furthermore, highest levels of CRP, LDH, ferritin, D-dimer and higher cHIS score predict higher mortality in pts receiving anakinra. It is important to identify the pts with higher mortality risk to improve outcome.