Although the utilization of extracorporeal membrane oxygenation (ECMO) is increasing and its technology is evolving, only a few epidemiologic reports have described the uses and outcomes of ECMO. The aim of this study was to investigate the changes in utilization and survival rate in patients supported with ECMO for severe respiratory failure in Korea. This was a multicenter study on consecutive patients who underwent ECMO across 16 hospitals in Korea. The records of all patients who required ECMO for acute respiratory failure between 2012 and 2015 were retrospectively reviewed, and the utilization of ECMO was analyzed over time. During the study period, 5552 patients received ECMO in Korea as a whole, and a total of 2472 patients received ECMO at the participating 16 hospitals. We analyzed 487 (19.7%) patients who received ECMO for respiratory failure. The number of ECMO procedures provided for respiratory failure increased from 104 to 153 during the study period. The in-hospital survival rate increased from 30.8% to 35.9%. The use of prone positioning increased from 6.8% to 49.0% (p < 0.001), and the use of neuromuscular blockers also increased from 28.2% to 58.2% (p < 0.001). Multiple regression analysis showed that old age (OR 1.038 (95% CI 1.022, 1.054)), use of corticosteroid (OR 2.251 (95% CI 1.153, 4.397)), continuous renal replacement therapy (OR 2.196 (95% CI 1.135, 4.247)), driving pressure (OR 1.072 (95% CI 1.031, 1.114)), and prolonged ECMO duration (OR 1.020 (95% CI 1.003, 1.038)) were associated with increased odds of mortality. Utilization of ECMO and survival rates of patients who received ECMO for respiratory failure increased over time in Korea. The use of pre-ECMO prone positioning and neuromuscular blockers also increased during the same period.

Physical therapy (PT) and early mobilization for critically ill patients have been popularized to decrease the length of hospital stay and to improve the quality of life after discharge. We reviewed our experience of PT and active mobilization for patients on extracorporeal membrane oxygenation (ECMO) in terms of its technical feasibility and safety. Study endpoints were safety events during PT and PT interruptions due to unstable vital signs. Of the eight patients, one patient (12.5%) had venoarterial ECMO, seven patients (87.5%) had venovenous ECMO. Among total of 62 sessions including 31 sessions (50%) of passive range of motion and electrical muscle stimulation, 17 sessions (27.4%) were performed for patients who were sitting in bed or on the edge of bed, two sessions (3.2%) were for strengthening in sitting, 11 sessions (18%) were for standing or marching in place, one session (2%) was for walking. Eight sessions (13%) of sitting were supported with invasive mechanical ventilation. Three sessions (5%) were stopped due to tachycardia (n = 1) and tachypnea (n = 2). There was no clinically significant adverse event in patients. Thus, early PT and mobilization for patients on ECMO might be feasible and safe at an experienced ECMO center.

Sepsis is a major cause of mortality and morbidity in critically ill patients. Procalcitonin (PCT) and C-reactive protein (CRP) are the most frequently used biomarkers in sepsis. We investigated changes in PCT and CRP concentrations in critically ill patients with sepsis to determine which biochemical marker better predicts outcome. We retrospectively analyzed 171 episodes in 157 patients with severe sepsis and septic shock who were admitted to the Samsung Medical Center intensive care unit from March 2013 to February 2014. The primary endpoint was patient outcome within 7 days from ICU admission (treatment failure). The secondary endpoint was 28-day mortality. Severe sepsis was observed in 42 (25%) episodes from 41 patients, and septic shock was observed in 129 (75%) episodes from 120 patients. Fifty-five (32%) episodes from 42 patients had clinically-documented infection, and 116 (68%) episodes from 99 patients had microbiologically-documented infection. Initial peak PCT and CRP levels were not associated with treatment failure and 28-day mortality. However, PCT clearance (PCTc) and CRP (CRPc) clearance were significantly associated with treatment failure (p = 0.027 and p = 0.030, respectively) and marginally significant with 28-day mortality (p = 0.064 and p = 0.062, respectively). The AUC for prediction of treatment success was 0.71 (95% CI, 0.61–0.82) for PCTc and 0.71 (95% CI, 0.61–0.81) for CRPc. The AUC for survival prediction was 0.77 (95% CI, 0.66–0.88) for PCTc and 0.77 (95% CI, 0.67–0.88) for CRPc. Changes in PCT and CRP concentrations were associated with outcomes of critically ill septic patients. CRP may not be inferior to PCT in predicting outcome in these patients.

Background: COVID-19, which is accompanied by acute respiratory distress, multiple organ failure, and death, has spread worldwide much faster than previously thought. However, at present, it has limited treatments. Objective: To overcome this issue, we developed an artificial intelligence (AI) model of COVID-19, named EDRnet (ensemble learning model based on deep neural network and random forest models), to predict in-hospital mortality using a routine blood sample at the time of hospital admission. Methods: We selected 28 blood biomarkers and used the age and gender information of patients as model inputs. To improve the mortality prediction, we adopted an ensemble approach combining deep neural network and random forest models. We trained our model with a database of blood samples from 361 COVID-19 patients in Wuhan, China, and applied it to 106 COVID-19 patients in three Korean medical institutions. Results: In the testing data sets, EDRnet provided high sensitivity (100%), specificity (91%), and accuracy (92%). To extend the number of patient data points, we developed a web application (BeatCOVID19) where anyone can access the model to predict mortality and can register his or her own blood laboratory results. Conclusions: Our new AI model, EDRnet, accurately predicts the mortality rate for COVID-19. It is publicly available and aims to help health care providers fight COVID-19 and improve patients’ outcomes.

Acute respiratory distress syndrome (ARDS) is potentially underrecognized by clinicians. Early recognition and subsequent optimal treatment of patients with ARDS may be facilitated by usage of biomarkers. Surfactant protein D (SP-D), a marker of alveolar epithelial injury, has been proposed as a potentially useful biomarker for diagnosis of ARDS in a few studies. We tried to validate the performance of plasma SP-D levels for diagnosis of ARDS. We conducted a retrospective analysis using data from three (two in USA and one in Korea) prospective biobank cohorts involving 407 critically ill patients admitted to medical intensive care unit (ICU). A propensity score matched analysis (patients with versus without ARDS, matched 1:1) was carried out using significant variables from multiple logistic regression. The diagnostic accuracy of plasma SP-D as a diagnostic marker of ARDS was assessed by receiver operating characteristic curve analysis. Out of the 407 subjects included in this study, 39 (10%) patients fulfilled ARDS criteria. Patients with ARDS had higher SP-D levels in plasma (p < 0.01) and higher hospital-mortality (p < 0.001) than those without ARDS. Thirty eight subjects with ARDS (cases) were successfully matched for propensity for ARDS with 38 subjects without ARDS (controls). Plasma levels of SP-D were higher in cases with ARDS compared to their matched controls without ARDS [median 20.8 ng/mL (interquartile range, 12.7–38.4) versus 7.9 (4.1–17.0); p = 0.001]. The area under the receiver operating characteristic curve for SP-D for the diagnosis of ARDS was 0.71 (95% confidence intervals, 0.60–0.83). A cut-off point of 12.7 ng/mL for SP-D yielded sensitivity of 74% and specificity of 63%. High levels of SP-D within 48 h after ICU admission might serve as a diagnostic marker for ARDS in patients hospitalized in medical ICU. Further prospective trials are required to validate the diagnostic role of SP-D in ARDS, and if its usefulness is greater in direct than in indirect ARDS, as well as across different strata of severity of ARDS.

Limited data are available on imaging predictors of neurological outcomes after extracorporeal cardiopulmonary resuscitation (ECPR). We investigated the association of initial brain computed tomography (CT) findings with neurological outcomes following ECPR. Between February 2005 and December 2015, a total of 42 patients who underwent brain CT scans within 48 h after ECPR were analyzed. Loss of the boundary between gray matter and white matter (LOB) or cortical sulcal effacement (SE), gray-to-white matter ratio (GWR), and optic nerve sheath diameter (ONSD) were measured on initial brain CT. The primary outcome was the Cerebral Performance Categories (CPC) scale at discharge. Of the 42 adult ECPR patients, 23 (54.8%) patients survived to discharge and 19 (45.2%) patients had good neurological outcomes (CPC 1 and 2). The area under the curve (AUC) of GWR in the basal ganglia (GWR-BG) was 0.792 (95% confidence interval (CI), 0.639–0.901, p = 0.001). ONSD (AUC 0.745; 95% CI, 0.587 – 0.867, p = 0.007) was 5.57 (interquartile range (IQR) 5.14 – 5.98) mm in the good neurological outcome group versus 6.07 (IQR 5.71 – 6.64) mm in the poor outcome group. LOB or SE were more often detected in the poor neurological outcome group (AUC 0.817; 95% CI, 0.682–0.952, p <0.001). The predictive performance of poor neurological outcomes of a composite of GWR-BG, ONSD, and LOB/SE was significantly improved (AUC 0.904; 95% CI, 0.773–0.973) compared to when each brain CT marker was considered separately (GWR-BG, p = 0.048; ONSD, p = 0.026; LOB/SE, p = 0.028). GWR, ONSD, and LOB/SE on initial brain CT scans are associated with neurological prognosis in patients who underwent ECPR. The new risk prediction model, which uses a composite of GWR, ONCD, and LOB/SE, could provide better information on neurologic outcomes in patients underwent ECPR.