Subscribe to eTOC

NeuroTech-REM Sleep Behavior Disorder
Smartphone Distinguishes Idiopathic REM Sleep Behavior Disorder from Parkinson's Disease

ARTICLE IN BRIEF

Figure

AN EARLIER STUDY had found that the smartphone could detect and monitor Parkinsons symptoms reliably by assessing voice, balance, gait, finger tapping, and reaction time.

Using only smartphone sensor recordings, investigators were able determine whether a participant had idiopathic REM sleep behavior disorder or Parkinson's disease, and to identify which of seven tasks were most pronounced in distinguishing the groups.

A customized smartphone application can identify motor features that differentiate individuals with sleep study-confirmed idiopathic rapid eye movement sleep behavior disorder (iRBD) from controls and Parkinson's disease, according to a study published online September 19 in Neurology.

In an earlier pilot study, researchers had investigated whether a smartphone could detect and monitor Parkinson's symptoms reliably by assessing voice, balance, gait, finger tapping, and reaction time. This time, the team at the University of Oxford in the United Kingdom explored the use of a larger selection of smartphone tasks, further evaluating rest and postural tremor in the clinic and at home to objectively measure motor symptoms.

“There is growing interest in iRBD as a condition that can precede the onset of Parkinson's,” Christine Lo, MBChB, a co-author of the paper and clinical research fellow at the Oxford Parkinson's Disease Centre, told Neurology Today.

In iRBD, the normal loss of tone (paralysis) during REM sleep does not occur. As a result, Dr. Lo said “people can act out their dreams,” potentially inflicting injuries upon themselves, their partner or both.

Dr. Lo added that it is common for individuals with iRBD to be unaware of their dream enactment. Among the people less likely to seek medical attention are those without a partner to notice abnormal behavior or those experiencing infrequent or milder episodes, she said.

While polysomnography remains the current gold standard for the diagnosis of iRBD, Dr. Lo noted that it is costly, time-consuming, and subject to possibly long waiting times that may discourage some patients.

In the latest study, the researchers demonstrated that it is possible to use smartphone tests alone to distinguish those with iRBD from those without the condition (controls) and those with Parkinson's.

“Though early work, this is exciting, as there is the potential for the smartphone test — itself scalable considering the ubiquity of smartphone use — to be used to screen for and potentially help in the diagnosis of individuals with iRBD,” Dr. Lo told Neurology Today.

The authors acknowledged the study's limitations — among them, fewer women participated in the study, participants may have had more optimally controlled Parkinson's, which could have reduced the magnitude of observed effect sizes, and in the absence of independent data (video reports, self-reported diaries), it may not have been possible to gauge the level of adherence to the test protocols. They recommended further research involving other groups of participants, exhibiting various parkinsonian syndromes, to fully understand the smartphone's ability to assist in differential diagnosis.

STUDY DETAILS

The study enrolled 334 participants with Parkinson's disease, 104 with iRBD, and 84 control subjects. They performed seven tasks to evaluate voice, balance, gait, finger tapping, reaction time, rest tremor, and postural tremor.

Over the course of several days, smartphone recordings captured their performance of these motor tasks both in the clinic and at home under non-controlled conditions. All participants also underwent thorough parallel in-clinic evaluations.

Using only the smartphone sensor recordings, the investigators sought to determine whether a participant had iRBD or Parkinson's and to identify which of the seven tasks were most pronounced in distinguishing the groups.

The observations yielded statistically significant differences between the three groups. For the three pairwise discriminatory comparisons: controls versus iRBD, controls versus Parkinson's, and iRBD versus Parkinson's, mean sensitivity and specificity values ranged from 84.6 percent to 91.9 percent. Postural tremor, rest tremor, and voice were the most discriminatory tasks overall, while reaction time was the least discriminatory.

EXPERT COMMENTARY

As prodromal stages and early identification of Parkinson's have received heightened attention in recent years, the study underscores the importance of initiating potentially neuroprotective treatment before too much damage has occurred, several neurologists told Neurology Today.

The authors “have used an innovative approach and show some promising results, with significant potential implications for clinical research and clinical care,” said F. Rainer von Coelln, DrMed, assistant professor of neurology at the University of Maryland School of Medicine.

Among the study's strengths are the pairwise comparisons of ample patients with a Parkinson's diagnosis versus healthy controls and individuals with sleep study-confirmed iRBD. The researchers “were able to work with a solid sample size for all three groups, even though one would have hoped for a larger group of control subjects (n=84, compared to 334 PD patients),” Dr. von Coelln said.

By capitalizing on the highly sensitive accelerometers, microphone, and touch screen features incorporated into smartphones, they combined different motor and non-motor aspects, casting a wider net than most previous studies assessing parkinsonian symptoms. Allowing participants to complete tasks under real-life conditions — either during clinic visits or at home rather than in a controlled laboratory — adds to the study's usefulness in guiding the development of future applications for research and patient care, he said.

Considering the possibility — not proven thus far — that patients with the slightest tremor or voice change may develop Parkinson's sooner than those without any detectable abnormality, a neurologist may decide to monitor them more often based on a smartphone's findings, said Carlos Singer, MD, professor of neurology and division chief of Parkinson's disease and movement disorders at the University of Miami's Leonard M. Miller School of Medicine.

It has been well established that individuals with iRBD have a high chance of developing Parkinson's or one of two similar neurodegenerative conditions. “Many of these patients already have a little bit of something that cannot be detected by the clinician but can be detected by the smartphone,” Dr. Singer said, while adding that neurologists still don't know when they could develop Parkinson's — for instance, in a year or 20 years.

“You would be more on guard,” Dr. Singer said of the patients with any aberration in motor or non-motor skills. Meanwhile, “you may be more reassuring to the ones who don't have it that they're in the small percentage who may not develop it. It's the art of trying to be compassionate with the patient.”

Although the smartphone application isn't ready for clinical neurology practice outside the scope of a trial, the study offers a “well-designed approach to the initial steps that need to be taken in order to use this technology effectively,” said Joohi Jimenez-Shahed, MD, associate professor of neurology and associate director of the neurology residency program in the Parkinson's Disease Center and Movement Disorders Clinic at Baylor College of Medicine in Houston.

The researchers “allude to the future use of their algorithms to monitor disease progression” with remote applications that help patients better understand their disease or measure symptom changes over time, she said.

However, Dr. Jimenez-Shahed, who is also the director of the deep brain stimulation program at Baylor, emphasized the need for more than “a single snapshot” following patients over the course of seven days. The test should be repeated at intervals over a long period of time, such as months, to confirm that this type of measurement can accurately monitor a person's condition.

In addition to allowing multiple measurements over time, another advantage of the smartphone application is that it helps track different aspects of motor symptoms that occur in Parkinson's, said Michelle A. Burack, MD, PhD, assistant professor of neurology and a movement disorders specialist at the University of Rochester Medical Center in Rochester, NY.

“One of the challenges with Parkinson's,” Dr. Burack explained, “is it's such a multidimensional disease, so you can't rely on a single metric to really capture” the impact of the condition.

Although the differences between groups were statistically significant, Dr. Burack said further work is needed to leverage these distinctions clinically. “We know that not everyone with iRBD is going to develop Parkinson's disease in the next year or two,” she said. “We need to see if these measurements predict conversion to Parkinson's disease.”

Dr. Burack hailed the study as “a step toward possibly identifying, with a device, a group of people who are at risk of developing Parkinson's in the near future. That would be the group you would want to offer some sort of neuroprotective therapy.”

LINK UP FOR MORE INFORMATION:

• Arora S, Baig F, Lo C, et al Smartphone motor testing to distinguish idiopathic REM sleep behavior disorder. controls, and PD http://n.neurology.org/lookup/doi/10.1212/WNL.0000000000006366?. Neurology 2018; Epub 2018 Sept 19.