CHICAGO – , including an inability to signal a change in disease activity without laboratory testing or before symptoms arise.
A new device developed at Massachusetts Institute of Technology could change all that.
Using data collected via a passive at-home monitoring device (Emerald sensor, Emerald Innovations Inc.), researchers found that increases in breathing rate, more awakenings at night, and slower walking speed accurately predicted that a person’s Crohn’s disease activity was about to flare, according to a study presented May 7 at Digestive Disease Week® (DDW) 2023.
In some cases, the prediction of a flare came up to 25 days sooner than via traditional measures.
“In order to provide optimal care, providers need to monitor patients closely with regard to accurate active disease.” said Joshua Korzenik, MD, a gastroenterologist at Brigham and Women’s Hospital and assistant professor of medicine at Harvard Medical School in Boston. “The problem is the clinical symptoms are not accurate.”
Tracking flares with technology
Traditionally, measuring flares in Crohn’s disease activity depends on imaging, colonoscopy, and/or laboratory measures of calprotectin or other biomarkers. These approaches can be costly, can involve delays, and can carry risks, Dr. Korzenik said.
“They are also a single snapshot in time,” he added.
To determine how well a noninvasive device could perform, investigators enrolled 120 people with 105 continuing in the study long enough to be evaluable; 44 people whose Crohn’s disease was in remission, 35 with active Crohn’s disease, as well as 26 healthy controls. Among those with Crohn’s disease, 83% were on biologic therapy.
The groups were matched for age and gender, with a mean age of 47 years and mean disease duration of 13 years.
People with certain medical conditions were excluded, as was anyone who owned a large dog that sometimes slept in bed with them because that might throw off the readings.
The participants put the device – which resembles a closed laptop or a large Wi-Fi router – in their homes and were monitored for a mean 306 days. Participants wore an ankle bracelet the first 2 weeks of the study so the device could learn to distinguish them from others in the home.
The device sent out radio waves with frequencies like Wi-Fi for researchers to measure the factors that may be associated with flares, such as sleep quality and cycles, breathing rate, and gait speed.
Traditional clinical measures based on blood and stool samples, along with patient-reported outcome surveys, were taken to compare with the accuracy of the device.
Data from the device were collected and transmitted securely to a cloud database without any interaction from the participant. Data included information on more than 25,000 nights of sleep, 200,000 hours of breathing signals, and 400,000 measurements of walking speed.
Sleep quality and cycles were straightforward, as was breathing rate. But gait speed was a little more complicated to measure. To illustrate, Dr. Korzenik showed the layout of an example apartment with data on how someone moved around. To distill the data, the researchers focused on one path in the home, relatively straight and not obstructed by furniture, and limited the measurements to a certain amount of time. People who spent more time at home during the COVID-19 pandemic did not skew the results, according to Dr. Korzenik, who added that it wasn’t total time walking around but a snippet in time.
A variety of sleep, breathing, and mobility metrics extracted by the device were integrated to assess disease activity. Investigators noted that during flares, sleep quality decreased, and more nocturnal awakenings occurred. They also found that gait speed slowed, and respiratory rate increased with flares.
When the investigators looked at sleep as a function of disease activity in the patient-reported surveys, they found a significant difference between people in remission and those with active disease. For example, people with active disease had a greater number of awakenings at night (P = .0016), less REM sleep at night (P = .0000), and less time in deep sleep (P = .000) compared with those in remission.
The technology “can identify flares with a predictive value that approaches fecal calprotectin,” Dr. Korzenik said.
Machine learning was used to look at severity of disease vs. fecal calprotectin values and “showed the data could be used as a marker of disease,” he added.
Use of a remote monitor, the comparison of validated vs. conventional data, and the large dataset were among the strengths of the study. The single-center design and exclusion of people with some comorbidities are potential limitations.
Further studies are warranted to confirm these findings and guide optimal care of people with Crohn’s disease, the investigators noted.