Delirium and neuropsychological outcomes in critically Ill patients with COVID-19: a cohort study.

OBJECTIVE: To characterise the clinical course of delirium for patients with COVID-19 in the intensive care unit, including postdischarge neuropsychological outcomes. DESIGN: Retrospective chart review and prospective survey study. SETTING: Intensive care units, large academic tertiary-care centre (USA). PARTICIPANTS: Patients (n=148) with COVID-19 admitted to an intensive care unit at Michigan Medicine between 1 March 2020 and 31 May 2020 were eligible for inclusion. PRIMARY AND SECONDARY OUTCOME MEASURES: Delirium was the primary outcome, assessed via validated chart review method. Secondary outcomes included measures related to delirium, such as delirium duration, antipsychotic use, length of hospital and intensive care unit stay, inflammatory markers and final disposition. Neuroimaging data were also collected. Finally, a telephone survey was conducted between 1 and 2 months after discharge to determine neuropsychological function via the following tests: Family Confusion Assessment Method, Short Blessed Test, Patient-Reported Outcomes Measurement Information System Cognitive Abilities 4a and Patient-Health Questionnaire-9. RESULTS: Delirium was identified in 108/148 (73%) patients, with median (IQR) duration lasting 10 (4-17) days. In the delirium cohort, 50% (54/108) of patients were African American and delirious patients were more likely to be female (76/108, 70%) (absolute standardised differences >0.30). Sedation regimens, inflammation, delirium prevention protocol deviations and hypoxic-ischaemic injury were likely contributing factors, and the most common disposition for delirious patients was a skilled care facility (41/108, 38%). Among patients who were delirious during hospitalisation, 4/17 (24%) later screened positive for delirium at home based on caretaker assessment, 5/22 (23%) demonstrated signs of questionable cognitive impairment or cognitive impairment consistent with dementia and 3/25 (12%) screened positive for depression within 2 months after discharge. CONCLUSION: Patients with COVID-19 commonly experience a prolonged course of delirium in the intensive care unit, likely with multiple contributing factors. Furthermore, neuropsychological impairment may persist after discharge.

Clinically applicable approach for predicting mechanical ventilation in patients with COVID-19.

BACKGROUND: Patients with coronavirus disease 2019 (COVID-19) requiring mechanical ventilation have high mortality and resource utilisation. The ability to predict which patients may require mechanical ventilation allows increased acuity of care and targeted interventions to potentially mitigate deterioration. METHODS: We included hospitalised patients with COVID-19 in this single-centre retrospective observational study. Our primary outcome was mechanical ventilation or death within 24 h. As clinical decompensation is more recognisable, but less modifiable, as the prediction window shrinks, we also assessed 4, 8, and 48 h prediction windows. Model features included demographic information, laboratory results, comorbidities, medication administration, and vital signs. We created a Random Forest model, and assessed performance using 10-fold cross-validation. The model was compared with models derived from generalised estimating equations using discrimination. RESULTS: Ninety-three (23%) of 398 patients required mechanical ventilation or died within 14 days of admission. The Random Forest model predicted pending mechanical ventilation with good discrimination (C-statistic=0.858; 95% confidence interval, 0.841-0.874), which is comparable with the discrimination of the generalised estimating equation regression. Vitals sign data including SpO2/FiO2 ratio (Random Forest Feature Importance Z-score=8.56), ventilatory frequency (5.97), and heart rate (5.87) had the highest predictive utility. In our highest-risk cohort, the number of patients needed to identify a single new case was 3.2, and for our second quintile it was 5.0. CONCLUSION: Machine learning techniques can be leveraged to improve the ability to predict which patients with COVID-19 are likely to require mechanical ventilation, identifying unrecognised bellwethers and providing insight into the constellation of accompanying signs of respiratory failure in COVID-19.