Affecting more than 1 million Americans each year, sepsis is the leading cause of in-hospital death and costs the healthcare system more than $20 billion annually. Because sepsis treatment is extremely challenging, researchers at the University of Michigan are working to develop a sensor that would reduce recovery time for patients, saving hospitals millions of dollars.

The current standard-of-care for sepsis patients is early, aggressive fluid therapy, in parallel with antibiotics, to stabilize blood pressure and maintain proper cardiac function. However, the amount of fluid administered is not “one size fits all” and providing optimal fluid therapy is a challenge. Some patients fall victim to fluid overload, which can lead to complications such as trouble breathing. Having a better gauge on patients’ fluid volumes would substantially reduce sepsis mortality and ICU recovery days.

A prototype of their ultrasound system. 

According to Ross Kessler, MD, one of the principal investigators in the study, it is very difficult to evaluate where a patient is on the “fluid spectrum.” The current solutions for fluid management are either too invasive, time consuming, or unreliable. One of the more advanced methods uses ultrasound to measure the diameter of the inferior vena cava (IVC), the large vein that carries blood from the lower body to the heart. Dr. Kessler and his colleague, Nik Theyyunni, MD, who is also part of the team, are co-directors of the U-M Emergency Ultrasound and are highly experienced using this method. However, both attest that this practice is time consuming, requires special training, and is only feasible to provide a limited number of “spot checks” on a patient’s fluid volume.  

In an effort to improve this ultrasound method, Kessler and Theyyunni teamed up with Grant Kruger, PhD, whose health informatics research includes investigating technologies for the automation of tissue measurements utilizing ultrasound imaging techniques. Together, they are working on a specialized, disposable sensor that non-invasively and continuously tracks the fluid levels of patients.

Instead of measuring the diameter of the IVC, the sensor is placed on the side of the patient’s neck and measures the internal jugular (IJ) vein. The device saves physician time by enabling a nurse to apply the wearable patch-like sensor. Once placed, the sensor interfaces with a bedside monitor. Using a proprietary signal processing algorithm, the monitor localizes the position of the vein and tracks geometric changes of the IJ as it expands and collapses.

If the IJ is large in size with little variability, the monitor indicates the patient has a high fluid level. The resulting measurement can be automatically repeated periodically and graphed over time to reveal trends in a patient's fluid volume. Kessler says the monitor will be similar to a pulse oximeter in that nurses and physicians will have continuous access to fluid volume at-a-glance.


Recent studies have shown that early identification of fluid overload can reduce mortality, recovery days, and days spent on a ventilator. Improved sepsis care protocols have also shown evidence of increasing the availability of ICU beds, while reducing patient care costs for third party payers. After going through a customer discovery process with MCIRCC, the team estimates that their device will reduce ventilator days/hospital stay by two days on average, saving hospitals an estimated $4,000 per sepsis patient.

“Working with MCIRCC has enabled us to take our idea and create a pathway where we can actually develop, validate, and eventually commercialize this technology. We’re excited to bring this product to the bedside so patients can actually benefit from it,” said Kessler.

While initially designed for early sepsis treatment, Kessler, Kruger, and Theyyunni envision other clinical uses for their technology such as monitoring fluid loss during dialysis to reduce potential treatment related side effects. The device may also be beneficial for monitoring blood loss due to internal hemorrhaging.

The team won the 2018 Coulter Translational Research Partnership Award which is funding the system development and pilot study. The initial study will use a cohort of 20-30 dialysis patients, where the drastic fluid shifts will be helpful in testing the accuracy of the sensor and algorithm.