Early warning scores in the perioperative period

Postoperative Critical Care Admission Was Not Associated with Improved Postoperative Outcomes in Elective Colorectal Surgery: Secondary Analysis Of POWER Trial

BACKGROUND

The efficacy of routine admission of high-risk patients to a critical care unit after surgery is not clear. The aim of our study was to investigate the association between critical care admission after scheduled colorectal surgery and postoperative complications, 30-day mortality, and length of stay in hospital.

METHODS

A pre-defined secondary substudy of POWER study was performed. POWER study was a prospective multicenter observational study of patients undergoing elective primary colorectal surgery during a single period of two months of recruitment between September and December 2017.

RESULTS

A total of 2084 patients from 80 Spanish hospitals were included, of which 722 (34.6%) were admitted to critical care unit (CCU) after elective surgery. After adjusting for confounding factors in the multivariate analysis, postoperative CCU admission was independently associated with a higher incidence of moderate-to-severe postoperative complications (adjusted OR 1.951, 95% CI 1.570, 2.425; p < 0.001). Regarding secondary outcomes, postoperative critical care admission was independently associated with higher 30-day mortality (adjusted OR 6.736; 95% CI 2.507, 18.101; p < 0.001) and independently associated with an increased hospital length of stay (adjusted OR 1.143, 95% CI 1.112, 1.175; p < 0.001).

CONCLUSIONS

Direct admission to CCU after scheduled colorectal surgery was not associated with a reduction in moderate-to-severe postoperative complications.

Using machine learning to improve the accuracy of patient deterioration predictions: Mayo Clinic Early Warning Score (MC-EWS)

OBJECTIVE

We aimed to develop a model for accurate prediction of general care inpatient deterioration.

MATERIALS AND METHODS

Training and internal validation datasets were built using 2-year data from a quaternary hospital in the Midwest. Model training used gradient boosting and feature engineering (clinically relevant interactions, time-series information) to predict general care inpatient deterioration (resuscitation call, intensive care unit transfer, or rapid response team call) in 24 hours. Data from a tertiary care hospital in the Southwest were used for external validation. C-statistic, sensitivity, positive predictive value, and alert rate were calculated for different cutoffs and compared with the National Early Warning Score. Sensitivity analysis evaluated prediction of intensive care unit transfer or resuscitation call.

RESULTS

Training, internal validation, and external validation datasets included 24 500, 25 784 and 53 956 hospitalizations, respectively. The Mayo Clinic Early Warning Score (MC-EWS) demonstrated excellent discrimination in both the internal and external validation datasets (C-statistic = 0.913, 0.937, respectively), and results were consistent in the sensitivity analysis (C-statistic = 0.932 in external validation). At a sensitivity of 73%, MC-EWS would generate 0.7 alerts per day per 10 patients, 45% less than the National Early Warning Score.

DISCUSSION

Low alert rates are important for implementation of an alert system. Other early warning scores developed for the general care ward have achieved lower discrimination overall compared with MC-EWS, likely because MC-EWS includes both nursing assessments and extensive feature engineering.

CONCLUSIONS

MC-EWS achieved superior prediction of general care inpatient deterioration using sophisticated feature engineering and a machine learning approach, reducing alert rate.

Dynamic individual vital sign trajectory early warning score (DyniEWS) versus snapshot national early warning score (NEWS) for predicting postoperative deterioration

Aims

International early warning scores (EWS) including the additive National Early Warning Score (NEWS) and logistic EWS currently utilise physiological snapshots to predict clinical deterioration. We hypothesised that a dynamic score including vital sign trajectory would improve discriminatory power.

Methods

Multicentre retrospective analysis of electronic health record data from postoperative patients admitted to cardiac surgical wards in four UK hospitals. Least absolute shrinkage and selection operator-type regression (LASSO) was used to develop a dynamic model (DyniEWS) to predict a composite adverse event of cardiac arrest, unplanned intensive care re-admission or in-hospital death within 24 h.

Results

A total of 13,319 postoperative adult cardiac patients contributed 442,461 observations of which 4234 (0.96%) adverse events in 24 h were recorded. The new dynamic model (AUC = 0.80 [95% CI 0.78−0.83], AUPRC = 0.12 [0.10−0.14]) outperforms both an updated snapshot logistic model (AUC = 0.76 [0.73−0.79], AUPRC = 0.08 [0.60−0.10]) and the additive National Early Warning Score (AUC = 0.73 [0.70−0.76], AUPRC = 0.05 [0.02−0.08]). Controlling for the false alarm rates to be at current levels using NEWS cut-offs of 5 and 7, DyniEWS delivers a 7% improvement in balanced accuracy and increased sensitivities from 41% to 54% at NEWS 5 and 18% to –30% at NEWS 7.

Conclusions

Using an advanced statistical approach, we created a model that can detect dynamic changes in risk of unplanned readmission to intensive care, cardiac arrest or in-hospital mortality and can be used in real time to risk-prioritise clinical workload.

Perioperative intensive care medicine

Surgery represents one of the main therapeutic references in the world, affording greater survival and life expectancy for many patients. In general, the estimated postoperative mortality is low (around 1-4%). Thirteen percent of the surgical procedures have a high risk of complications, accounting for 80% of all postoperative deaths. Recently, there have been significant advances related to organizational aspects, new anesthetic and surgical techniques, prognostic scales, perioperative management and greater participation and involvement of the patient. This new series of Medicina Intensiva will address fundamental aspects of how Departments of Intensive Care Medicine can add value to the surgical process, in a coordinated manner with other services. Institutional policies are required to ensure the detection of patients at risk in hospitalization wards, with early admission to the ICU of those patients in whom admission is indicated, adapting the treatment in the ICU and optimizing the criteria for discharge. The detection and prevention of post-ICU syndrome in patients and relatives, and the follow-up of ICU discharge and hospitalization in a multidisciplinary manner can reduce the sequelae among critical surgical patients, improving the outcomes and quality of life, and restoring the patient to society. In future publications of this series directed to the surgical patient, updates will be provided on the perioperative management of some of the most complex surgeries.