Critical Illness Due to Covid-19: A Description of the Surge in a Single Center in Sioux Falls

Background We aim to describe the basic demographics, clinical course and outcomes of critically ill patients with Covid-19 admitted to Avera McKennan Hospital and University Health Center Intensive Care Unit (ICU) between March 20 and May 4, 2020. Methods In this single centered, retrospective, observational study, we enrolled 37 critically ill adults with COVID-19 pneumonia admitted to the (ICU) between March 20 and May 4, 2020. Demographic data, admitting symptoms, laboratory values, co-morbidities, treatments and clinical outcomes were collected. Data was compared between survivors and non-survivors. We aim to describe our data and report the 28-day mortality as of June 1, 2020. Results Of 154 patients admitted with COVID-19 pneumonia during our study period, 37 (24 percent) were critically ill and required an ICU stay. The mean age was 58 years and 76 percent were men. Of these 37 patients, 28 (78 percent) had a chronic illness (diabetes in 43 percent, hypertension in 47 percent). In addition, 54 percent were associated with a local meat packing plant. Most common presenting symptoms were dyspnea (92 percent), cough (70 percent) and fever (68 percent). The mean PaO2/ FiO2 ratio was 143 (67-362). Significant lab findings include the following: 54 percent of patients had lymphocytopenia, the mean ferritin was 850 ng/mL (10-3528), the mean D-Dimer was 4.09 FEU ug/mL and the mean IL-6 was 96.5 pg/mL. At 28 days, 24 percent (nine) had died. Twenty-five (68 percent) patients required mechanical ventilation, with 10 (27 percent) of those patients requiring initiation of neuromuscular blocking agents for ventilator compliance. Of those four (40 percent) did not survive. In addition, 20 patients (54 percent) were proned. Pneumomediastinum or pneumothorax occurred in five of the 37 (14 percent). Renal replacement therapy was required in 6 of the 37 patients, 4 of whom (66 percent) died. Steroids were used in 70 percent of patients, tocilizumab in 59 percent, and hydroxychloroquine in 27 percent. All patients received antibiotics. Convalescent plasma became available for our 5th patient. A total of 29 (78 percent) received convalescent plasma, (86 percent of survivors and 56 percent non-survivors). Median ICU length of stay was 11 days for both survivors (1-49) and non-survivors (1-21). There were no differences in age, body mass index (BMI), or initial PaO2/FiO2 (P/F) among those two groups. Non-survivors (nine) included the two immune compromised patients in our cohort, two patients with pre-existing DNR/DNI status, and one death within two hours of admit. Compared with survivors, more of the non-survivors received vasopressors (78 percent vs 46 percent), dialysis (44 percent vs 7 percent) and hydroxychloroquine (44 percent vs 21 percent). The first 5 patients treated in the ICU did not survive. One month after the initial case was reported in South Dakota, our ICU experienced a six-week surge. At its highest, COVID-19-related census reached 63 percent of the ICU capacity (15/24). Conclusion Mortality of critically ill patients with COVID-19 is high. Multi-organ, advanced and prolonged critical care resources are needed. Interpretation of our data is limited by a higher mortality of the earlier members of the cohort, a change in therapeutic practice over time and institution of social distancing.