Effect of early enteral nutrition on critical care outcomes in patients with acute ischemic stroke

Objective Stroke-associated pneumonia (SAP) is a comorbidity of ischemic stroke related to clinical outcomes. Early enteral nutrition (EEN; within 48 hours) reduces the incidence of infection and length of intensive care unit (ICU)/hospital stay. The relationship between EEN and critical care outcomes, including SAP, in patients with ischemic stroke has been insufficiently studied. Methods We recruited 499 patients in this retrospective observational study. We evaluated SAP incidence within 14 days from admission. Patients were divided into an EEN group and a late EN group (LEN; start later than EEN). We compared groups regarding background and length of ICU/hospital stay. Results EN was started within 48 hours in 236 patients. SAP was diagnosed in 94 patients (18.8%), with most in the LEN group (28.1% vs. 8.5%). Median [interquartile range] lengths of hospitalization (22 [12–30] days vs. 35 [20–45] days) and ICU stay (4 [2–5] days vs. 6 [3–8] days) were longer in the LEN group. EEN reduced the incidence of SAP. By contrast, consciousness disturbance and worsening consciousness level increased the SAP incidence. Increased age and National Institutes of Health Stroke Scale score were associated with start of prolonged EN. Conclusions We found that EEN may reduce SAP risk.

A machine learning model to predict critical care outcomes in patient with chest pain visiting the emergency department

Background Currently, the risk stratification of critically ill patient with chest pain is a challenge. We aimed to use machine learning approach to predict the critical care outcomes in patients with chest pain, and simultaneously compare its performance with HEART, GRACE, and TIMI scores. Methods This was a retrospective, case-control study in patients with acute non-traumatic chest pain who presented to the emergency department (ED) between January 2017 and December 2019. The outcomes included cardiac arrest, transfer to ICU, and death during treatment in ED. In the randomly sampled training set (70%), a LASSO regression model was developed, and presented with nomogram. The performance was measured in both training set (70% participants) and testing set (30% participants), and findings were compared with the three widely used scores. Results We proposed a LASSO regression model incorporating mode of arrival, reperfusion therapy, Killip class, systolic BP, serum creatinine, creatine kinase-MB, and brain natriuretic peptide as independent predictors of critical care outcomes in patients with chest pain. Our model significantly outperformed the HEART, GRACE, TIMI score with AUC of 0.953 (95%CI: 0.922–0.984), 0.754 (95%CI: 0.675–0.832), 0.747 (95%CI: 0.664–0.829), 0.735 (95%CI: 0.655–0.815), respectively. Consistently, our model demonstrated better outcomes regarding the metrics of accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. Similarly, the decision curve analysis elucidated a greater net benefit of our model over the full ranges of clinical thresholds. Conclusion We present an accurate model for predicting the critical care outcomes in patients with chest pain, and provide substantial support to its application as a decision-making tool in ED.

SP4.2.10 COVID-19 and General Surgery in a District General; Safety and Quality data

Aim The Covid-19 pandemic forced departments to change standard modes of delivery of care. Within our unit, reductions in junior workforce and changed operating protocols resulted in greater consultant presence on in-patient wards and the ambulatory unit. This study aimed to determine the effect on patient outcomes by interrogation of data collected from weekly Safety and Quality Clinical Governance meetings. Methods Patients admitted between December 2019 and February 2020 were compared to those admitted between April 2020 and June 2020. The weekly meeting mandates consultant discussion of all readmissions, all patients who had a length of stay (LoS) of > 7 days and all admissions to critical care. Outcomes between the two time periods were compared. Results There was a marked reduction in admissions during the second study period. However, the proportion of patients discharged from ambulatory care increased as did the proportion of readmissions; in the pre-Covid period, there were 429 readmissions of which 188 (44%) were unplanned but in the post-Covid period, there were 311 readmissions. There were no serious adverse events from discharged patients or readmissions. There were markedly fewer patients who had a LoS > 7days (179 patients versus 87) and a greater number of unplanned admissions to critical care (44% versus 64%). Conclusions Increased consultant presence may explain the reduced LoS and increased readmissions due to the greater ‘risk’ that senior clinicians are prepared to take. Enhanced consultant presence should be a permanent change, even after the pandemic is over.

A machine learning model to predict critical care outcomes in patient with chest pain visiting the emergency department

Background: Currently, the risk stratification of critically ill patient with chest pain is a challenge. We aimed to use machine learning approach to predict the critical care outcomes in patients with chest pain, and simultaneously compared its performance with the HEART score. Methods: This was a retrospective, case-control study in patients with acute non-traumatic chest pain who presented to the emergency department (ED) between January 2017 and December 2019. The outcomes included cardiac arrest, transfer to ICU, and death during treatment in ED. In the randomly sampled training set (70%), a LASSO regression model was developed, and presented with nomogram. The performance was measured in both training set (70% participants) and testing set (30% participants), and findings were compared with the HEART score.Results: We proposed a LASSO regression model incorporating mode of arrival, reperfusion therapy, Killip class, systolic BP, SCr, CKMB, and BNP as independent predictors of critical care outcomes in patients with chest pain. Our model significantly outperformed the HEART score with AUC of 0.953 (95%CI: 0.922 - 0.984) and 0.754 (95%CI: 0.675 - 0.832), respectively. Consistently, our model demonstrated better outcomes regarding the metrics of accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. Similarly, the decision curve analysis elucidated a greater net benefit of our model over the full ranges of clinical thresholds.Conclusion: We present a promising model for predicting the critical care outcomes in patients with chest pain, and provide substantial support to its application as a decision-making tool in ED.