Advantage of Vital Sign Monitoring Using a Wireless Wearable Device for Predicting Septic Shock in Febrile Patients...

Area of Science:

• Emergency Medicine
• Critical Care
• Biomedical Engineering

Background:

• Intermittent manual vital sign measurements may delay sepsis prediction in febrile emergency department (ED) patients.
• Early detection of clinical deterioration is crucial for managing febrile patients at risk of sepsis.

Purpose of the Study:

• To evaluate the predictive performance of a wireless monitoring device for continuous heart rate (HR) and respiratory rate (RR) measurement.
• To assess the utility of machine learning analysis with continuous vital sign data for predicting clinical deterioration in febrile ED patients.

Main Methods:

• Analysis of 468 febrile patients (temperature >38 °C) admitted to an ED isolation care unit.
• Comparison of a machine learning model using continuous wireless device data versus intermittent manual data.
• Model performance evaluated using Area Under the Receiver Operating Characteristic curve (AUROC).

Main Results:

• Wireless device data models showed high predictive performance (AUROC 0.858-0.861) comparable to manual data models (AUROC 0.841-0.853).
• Continuous monitoring models detected clinical deterioration 9 hours (fragmented model) and 5.5 hours (accumulated model) earlier than manual data models.
• Wireless device monitoring demonstrated accurate prediction of clinical deterioration in stable, febrile ED patients.

Conclusions:

• Continuous vital sign monitoring with a wearable device offers accurate prediction of clinical deterioration in febrile ED patients.
• This technology can significantly reduce the time required to recognize potential clinical deterioration, aiding in timely sepsis management.