A Novel External Ventricular Drain Sensor to Improve Acquired Brain Injury Monitoring

The insufficiency of current methods to capture the context and environment of neurocritical care can negatively impact patient outcomes. Insertion of an external ventricular drain (EVD) into the ventricles to monitor intracranial pressure (ICP) is a common lifesaving procedure for acquired brain injury patients. Yet, nursing interventions that significantly affect the measured ICP value, such as changing the EVD stopcock position, are poorly documented. Environmental factors like light and noise levels are not monitored as standard of care despite worse outcomes in patients affiliated with sensory sensitivities. Capturing these missing data is an essential first step toward quantifying their effects.

Our entry point was the development of a stopcock position sensor (SPS) that attaches to the EVD stopcock and time-synchronously annotates the recorded ICP data with its position. A two-phase, prospective, nonrandomized observational study was conducted to evaluate the efficacy of the SPS. Phase I assessed the SPS using an ex vivo simulation of ICP management. Phase II involved human subjects with the SPS attached to the EVD stopcock while patients were managed per standard of care.

The SPS accurately annotated the ICP data and identified that the EVD drained the cerebrospinal fluid for 94.52% of total patient monitoring time (16.98 h). For only 3.54% of the time, the stopcock directed the cerebrospinal fluid into the pressure transducer for accurate ICP measurement. For the remaining 1.94% of the time, the stopcock was positioned off: No cerebrospinal fluid drainage and no ICP monitoring.

We successfully captured an important aspect of the ICP monitoring context, the EVD stopcock position, and time-synchronized it with the recorded physiology. Our system enables future investigations into the impact that a broad contextual data environment has on physiological measurements and acquired brain injury patient outcomes. In the future, we aim to capture additional contextual data sources and expand the scope to battlefield environments.