ARTICLE IN BRIEF
Researchers have developed an ultrathin thermal skin sensor that detects shunt dysfunction in hydrocephalus patients. Neurologists who specialize in treating adult or pediatric hydrocephalus welcomed the development but said it remains to be proved whether the noninvasive device will work in wider clinical practice.
A soft adhesive strip — no bigger than a Band-Aid, yet containing 100 miniaturized temperature sensors, a rechargeable battery, and a Bluetooth transmitter — accurately detected shunt malfunction in a preliminary study involving five adult patients with hydrocephalus.
The common treatment for hydrocephalus, a debilitating neurological disorder that results from the overproduction and/or impaired reabsorption of cerebrospinal fluid (CSF), are shunts to drain excessive CSF in the ventricular system of the brain. But the shunts can fail, and existing diagnostic tests to detect shunt function have limitations: They can be expensive and unreliable, and they are susceptible to interference and patient discomfort, with a potential for harm, according to the authors of a new paper published October 31 in Science Translational Medicine.
Building on more than a decade of prior research in materials science and electronic miniaturization, the researchers at Northwestern University and its Feinberg School of Medicine have developed a device that is soft, ultrathin (with a height of 4 mm at the battery and about 100 microns elsewhere), lightweight (680 milligrams), and skinlike in its physical properties.
The principle of the device is not unlike that of ShuntCheck, which is already commercially available: Apply something to the skin to alter the temperature of CSF flowing toward a shunt and then see if there is a resulting difference in temperature before and after the shunt. Instead of attempting to cool the CSF with ice packs, as ShuntCheck does, however, the new device slightly warms the CSF with its battery.
In the five patients studied, the sensor was highly accurate in detecting the difference between cases flowing or not (p=0.012), as validated by surgery or imaging.
“There is a huge incidence of shunt malfunction; more than half of people with a shunt end up undergoing a revision,” said neurologist Taeun Chang, MD, director of the neonatal neurology and neonatal neurocritical care program at Children's National Health System in Washington, DC, who was not involved with the study.
“Visually examining the ventricular size or shape by computed tomography scans and magnetic resonance imaging does not always accurately determine whether or not a shunt is failing,” Dr. Chang said. “And neurosurgeons don't want to access the shunt itself if they don't have to. It would be a major advance if this group can prove in a larger study that their portable, noninvasive technology works.”
After years of benchtop work, the research team experienced a steep learning curve with their first patient. Just handling the sticky device proved troublesome, as it stuck to surgeons' hands, and manipulating it sometimes destroyed the embedded electronic sensors. Figuring out how to place it precisely over the shunt tubing was also tricky. But modifications based on what they learned “resulted in rapid improvements,” the study authors noted.
Once the kinks were ironed out, the group's assessment of patient 1 using the epidermal sensing array (ESA) indicated a shunt malfunction, with flow of 0.01 ± 0.01 ml/min. After surgical revision, flow increased to 0.06 ± 0.02 ml/min. Patient 5 likewise had evidence of a highly occluded flow (0.027 ± 0.005 ml/min), which was surgically confirmed and corrected.
The second and third patients had no malfunctions, exhibiting a flow rate of 0.36 ± 0.04 ml/min and 0.13 ± 0.02 ml/min, respectively.
Patient 4 proved especially interesting, with initial measurements indicating an occluded flow (0.013 ± 0.002 ml/min). But the patient “had experienced severe and prolonged constipation for the week before the measurement,” the paper noted, “and clinically deteriorated because of a likely pseudo-obstruction.” Instead of going in for a shunt revision, the clinicians administered a bowel regimen, after which the patient's mental status and flow rate both returned to normal.
Overall, the paper concluded, “the skin-like precision sensor systems introduced here have the potential to represent a paradigm shift in clinical diagnostics of shunt malfunction. Compared with radiographic imaging, invasive sampling, and ice-pack cooling, these platforms are unique in their integration of precision soft thermal sensors with wireless transmission capability.”
The co–senior author of the paper described it as a proof of concept that needs to be replicated in a larger clinical trial designed to satisfy the requirements of the US Food and Drug Administration (FDA).
“That's what we're currently planning,” said a co–senior study author Matthew Potts, MD, assistant professor of neurological surgery at Northwestern University's Feinberg School of Medicine.
For now, he said, the device is best able to distinguish between normal flow and a near-total lack of flow.
“What we're excited about is measuring different degrees of flow,” Dr. Potts said. “That comes into play in situations where a shunt is not obstructed but clogged and can't flow to the degree a patient needs. Flow might differ depending on time of day and activities they're doing. We can monitor and see what the shunt does over 24 hours.”
Although the system's rechargeable battery has only about two hours of operating time, a measurement can be obtained in just five minutes. As a result the device could be used intermittently, on and off, for up to six days.
“The thought is if patients have concerning clinical symptoms,” Dr. Potts said, “we can use this device to investigate it further before we jump to CT scans with radiation or nuclear medicine or even going to surgery.”
Amit B. Ayer, MD, a sixth-year neurosurgery resident at the Feinberg School who has been closely involved in developing the device, said he hopes it will clear up the uncertainty over whether or not a shunt requires a surgical revision.
“As neurosurgeons, we're familiar with the propensity for shunts to fail,” he said. “But the clinical manifestation of failure varies across individuals. It can range from sleepiness and nausea to headache. It can be nothing at all, or a life-and-death emergency.”
Neurologists who specialize in treating adult or pediatric hydrocephalus said it remains to be proved whether the noninvasive device will work in wider clinical practice.
“People have been playing with this concept for a long time, trying to either warm or cool the skin overlying the shunt,” said James B. Golomb, MD, director of the New York University Langone Neurosurgery Adult Hydrocephalus Program. “The question is how practical this particular device will be, how reliable and reproducible the measurements will be in the hands of multiple clinicians, especially if the numbers you get out of this thing are so critically dependent on exactly how you place it.”
As to whether the device will be able to quantify gradations of flow rather than just detect a total obstruction, Dr. Golomb said, “It's less clear whether this system is going to be sufficiently accurate. They claim to be working on it, but it doesn't look like a straightforward calculation.”
Even so, there is an urgent need for such a device, he said.
“My experience is primarily with older people whose hydrocephalus initially caused gait problems that look like Parkinson's disease,” he said. “Some of these patients improve dramatically after a shunt operation but decline again at some point later. The reason for the decline is often hard to figure out. The gold-standard tool we've had for years is a radionuclide shunt patency study or shuntogram, which is invasive, not universally available, and carries risks of infection. But the sooner you can intervene, the better it's going to be. Once a shunt has occluded for months, it may result in irreversible injury. So a noninvasive, safe, and easy way to tell us if a shunt is working is extremely needed.”
Two other neurologists specializing in the treatment of hydrocephalus remarked upon the rapid pace of technological innovation that has made the new device possible.
“There was a conference at the FDA in January of 1999, where I made a presentation about the idea of a smart shunt,” said Michael A. Williams, MD, FAAN, professor of neurology and neurological surgery at the University of Washington, where he is director of Adult and Transitional Hydrocephalus and CSF Disorders. “I couldn't have imagined then anything like the technology described in this paper. It's one thing to have an idea; it's another to get it into the engineering lab and move it forward as this group at Northwestern did. I think they deserve a lot of credit.”
Dr. Chang of Children's National Health System agreed: “It's quite amazing how well they can miniaturize things that can open doors to solving medical problems previously restricted by size or technology.”
The device would be particularly helpful for detecting shunt dysfunction in infants who can't indicate they have a headache, nausea, or fatigue, Dr. Chang noted. “They can decline very subtly when the shunt malfunctions, and then they come into the emergency room in an acute critical state.”
Dr. Chang suggested that parents might want to have one of these devices applied more or less continuously to their infant's shunt when they are having headaches or vomiting in the settling of [meaning clear?] unremarkable neuroimaging.
“If the kid started acting irritable, they could turn on the device with a phone app and have the information sent wirelessly to their neurosurgeon,” she said. “Or it could be set on, automatically, two or three times per day. To me that would be clinically useful.”
Asked about that idea, Dr. Ayer said, “We hope the extended wearability of this device will allow an application where the patient is not in extremis. For someone who's having headaches or other problems, we could envision it as being similar to a Holter monitor.”
Beyond the use of temperature sensors, Dr. Chang said she would like to see the development of noninvasive intracranial pressure sensors. “Right now we have no way of measuring intracranial pressure in a newborn without a surgical intervention,” she said. “If technology has reached the point where temperature sensors can be miniaturized to this degree, why not a pressure probe?”
Dr. Ayer and other co-authors of the paper noted that they are inventors on patents and patent applications related to thermal sensing, including applications in hydrocephalus diagnostics. They are co-founders of Rhaeos, Inc., a company that seeks to commercialize the shunt sensor. Dr. Williams, Dr. Chang, and Dr. Golomb reported no commercial interests or conflicts of interest.