Noninvasive measurement of cerebrospinal fluid flow using an ultrasonic transit time flow sensor: a preliminary study

Abstract

OBJECT Mechanical failure—which is the primary cause of CSF shunt malfunction—is not readily diagnosed, and the specific reasons for mechanical failure are not easily discerned. Prior attempts to measure CSF flow noninvasively have lacked the ability to either quantitatively or qualitatively obtain data. To address these needs, this preliminary study evaluates an ultrasonic transit time flow sensor in pediatric and adult patients with external ventricular drains (EVDs). One goal was to confirm the stated accuracy of the sensor in a clinical setting. A second goal was to observe the sensor’s capability to record real-time continuous CSF flow. The final goal was to observe recordings during instances of flow blockage or lack of flow in order to determine the sensor’s ability to identify these changes. METHODS A total of 5 pediatric and 11 adult patients who had received EVDs for the treatment of hydrocephalus were studied in a hospital setting. The primary EVD was connected to a secondary study EVD that contained a fluid-filled pressure transducer and an in-line transit time flow sensor. Comparisons were made between the weight of the drainage bag and the flow measured via the sensor in order to confirm its accuracy. Data from the pressure transducer and the flow sensor were recorded continuously at 100 Hz for a period of 24 hours by a data acquisition system, while the hourly CSF flow into the drip chamber was recorded manually. Changes in the patient’s neurological status and their time points were noted. RESULTS The flow sensor demonstrated a proven accuracy of ± 15% or ± 2 ml/hr. The flow sensor allowed real-time continuous flow waveform data recordings. Dynamic analysis of CSF flow waveforms allowed the calculation of the pressure-volume index. Lastly, the sensor was able to diagnose a blocked catheter and distinguish between the blockage and lack of flow. CONCLUSIONS The Transonic flow sensor accurately measures CSF output within ± 15% or ± 2 ml/hr, diagnoses the blockage or lack of flow, and records real-time continuous flow data in patients with EVDs. Calculations of a wide variety of diagnostic parameters can be made from the waveform recordings, including resistance and compliance of the ventricular catheters and the compliance of the brain. The sensor’s clinical applications may be of particular importance to the noninvasive diagnosis of shunt malfunctions with the development of an implantable device.