Background: Survival rates of critically ill COVID-19 patients are affected by various clinical features and laboratory parameters at ICU admission. Some of these predictors are universal but others may be population specific.Objective: To determine utility of baseline clinical and laboratory parameters in a multivariate regression model to predict outcomes in critically ill COVID-19 patients in a tertiary hospital in Croatia. Methods: 692 critically ill COVID-19 patients treated during a 10-month period were included in this retrospective observational trial to assess the risk factors determining mortality rates. Various anthropometric features, comorbidities, laboratory parameters, clinical features and therapeutic interventions were included in the analysis. ICU mortality rates and length of ICU stay were primary endpoints analyzed in this study. Results: After multivariate adjustment, only the SOFA score, PaO 2 /FiO 2 and history of arterial hypertension had an effect on ICU mortality, as well as the need to initiate invasive mechanical ventilation. Increase in PaO 2 /FiO 2 over the first 7 days was present in survivors, while reverse applied to SOFA. Length of ICU stay was 9 (4À14) days. Factors affecting survival times were admission from wards, congestive heart failure, invasive mechanical ventilation, bacterial superinfections, age > 75 years, SOFA score, and serum ferritin, CRP and IL-6 values at ICU admission. Conclusion: Elevated inflammatory biomarkers and SOFA score at ICU admission were detected as significant predictors of ICU mortality in this cohort, while initiation of invasive mechanical ventilation is the most relevant interventional mortality risk factor in critically ill COVID-19 patients.

Aim To describe epidemiological characteristics and baseline clinical features, laboratory findings at intensive care unit (ICU) admission, and survival rates of critically ill coronavirus disease 2019 (COVID-19) patients treated at a tertiary institution specialized for COVID-19 patients. Methods This retrospective study recruited 692 patients (67.1% men). Baseline demographic data, major comorbidities, anthropometric measurements, clinical features, and laboratory findings at admission were compared between survivors and non-survivors. Results The median age was 72 (64-78) years. The median body mass index was 29.1 kg/m 2 . The most relevant comorbidities were diabetes mellitus (32.6%), arterial hypertension (71.2%), congestive heart failure (19.1%), chronic kidney disease (12.6%), and hematological disorders (10.3%). The median number of comorbidities was 3 and median Charlson Comorbidity Index (CCI) was 5. A total of 61.8% patients received high-flow nasal oxygen therapy (HFNO) and 80.5% received mechanical ventilation (MV). Median duration of HFNO was 3, and that of MV was 7 days. ICU mortality rate was 72.7%. Survivors had significantly lower age, number of comorbidities, CCI, sequential organ failure assessment score, serum ferritin, C-reactive protein, D-dimer, and procalcitonin, interleukin-6, lactate, white blood cell, and neutrophil counts. They also had higher lymphocyte counts, PaO 2 /FiO 2 ratio, and glomerular filtration rate at admission. Length of ICU stay was 9 days. The median survival was 11 days for mechanically ventilated patients, and 24 days for patients who were not mechanically ventilated. Conclusion The parameters that differentiate survivors from non-survivors are in agreement with published data. Further multivariate analyses are warranted to identify individual mortality risk factors.

Background: Superinfections contribute to mortality and length of stay in critically ill COVID-19 patients. The aim of this study was to determine the incidence and pathogen distribution of bacterial and fungal superinfections of the lower respiratory tract (LRTI), urinary tract (UTI) and bloodstream (BSI) and to determine the predictive value of biomarkers of inflammatory response on their ICU survival rates. Methods: A retrospective observational study that included critically ill COVID-19 patients treated during an 11-month period in a Croatian national COVID-19 hospital was performed. Clinical and diagnostic data were analyzed according to the origin of superinfection, and multivariate regression analysis was performed to determine the predictive values of biomarkers of inflammation on their survival rates. Results: 55.3% critically ill COVID-19 patients developed bacterial or fungal superinfections, and LRTI were most common, followed by BSI and UTI. Multidrug-resistant pathogens were the most common causes of LRTI and BSI, while Enterococcus faecalis was the most common pathogen causing UTI. Serum ferritin and neutrophil count were associated with decreased chances of survival in patients with LRTI, and patients with multidrug-resistant isolates had significantly higher mortality rates, coupled with longer ICU stays. Conclusion: The incidence of superinfections in critically ill COVID-19 patients was 55.3%, and multidrug-resistant pathogens were dominant. Elevated ferritin levels and neutrophilia at ICU admission were associated with increased ICU mortality in patients with positive LRTI.

Healthcare-associated infections are an emerging cause of morbidity and mortality in COVID-19 intensive care units (ICUs) worldwide, especially those caused by multidrug-resistant (MDR) pathogens. The objectives of this study were to assess the incidence of bloodstream infections (BSIs) among critically ill COVID-19 patients and to analyze the characteristics of healthcare-associated BSIs due to MDR Acinetobacter baumannii in an COVID-19 ICU. A single-center retrospective study was conducted at a tertiary hospital during a 5-month period. The detection of carbapenemase genes was performed by PCR and genetic relatedness by pulsed-field gel electrophoresis (PFGE) and multilocus-sequence typing. A total of 193 episodes were registered in 176 COVID-19 ICU patients, with an incidence of 25/1000 patient-days at risk. A. baumannii was the most common etiological agent (40.3%), with a resistance to carbapenems of 100%. The blaOXA-23 gene was detected in ST2 isolates while the blaOXA-24 was ST636-specific. PFGE revealed a homogeneous genetic background of the isolates. The clonal spread of OXA-23-positive A. baumannii is responsible for the high prevalence of MDR A. baumannii BSIs in our COVID-19 ICU. Further surveillance of resistance trends and mechanisms is needed along with changes in behavior to improve the implementation of infection control and the rational use of antibiotics.