Relationship Between Lower Limb Movement Detected by Activity Monitors and Functional Mobility After SCI
Abstract
For someone who has sustained a spinal cord injury (SCI), one of the first questions they will ask is, “Will I walk again?” For some people this question is easy for clinicians to answer. If they have good strength and feeling below the level of their injury they will likely walk and if they have little to no movement or feeling, they will likely use a wheelchair. However, for the many people that fall between these two extremes, a clinician (doctor or physical therapist) is forced to rely on what they know from their previous experiences to predict someone’s outcome. After someone sustains a SCI, they generally have only 4-6 weeks of focused therapy in the hospital to try to regain function or learn new techniques that allow them to compensate for their injury. In this short time, focusing on the wrong type of therapy can have costs. For individuals who will ultimately use a wheelchair, it has been shown that training in walking takes away from time that could be spent learning important wheelchair-related skills like how to get in and out of the chair, push it without damaging their shoulders, and navigate obstacles in the environment like curbs and ramps. Also, individuals may have poorer quality of life outcomes like depression and be at risk for damaging the shoulder and wrist if they must bear significant weight through their arms when trying to walk. Alternatively, if someone has the potential to walk but doesn’t receive training, they miss out on benefits like improved heart health, bone health, and decreased risk of pressure sores. They may also miss a window when the body has higher levels of “neuroplasticity,” or an ability to reorganize nerve cells in the spinal cord to compensate for what was injured, during which walking training may be more effective. Research has been conducted to develop models that forecast outcomes, but for the group of people with some remaining strength and sensation, they aren’t very useful. In fact, for some people, the predicted likelihood of walking is equivalent to a coin flip. This study will use accelerometers, like those found in Smart watches, on the wrist and ankles to determine if movement of the limbs is related to someone’s strength and feeling. This builds on previous studies that have shown that small movements after SCI in animals have been linked to recovery, as well as studies that show a relationship between walking ability and movement in other populations with neurological diseases. These monitors may be able to pick up on subtle differences that current clinical tests cannot detect. In our prediction model, we will also include important factors that have been left out of previous models and can impact walking once someone is outside of a hospital environment, such as caregiver support, the need for bracing, environmental barriers like stairs, and one’s ability to cope with his/her injury. Data will be collected through these monitors for 24 hours/day for 5 days, creating a rich data set. We will use machine learning to determine how movement and personal traits like age are related to how they get around. We will collect data from people who have been injured for more than one year, as recovery of strength and feeling are generally stable by this point. We will also collect data from a group of people who are newly injured and follow-up with them at a year after their injury. We will use this data to demonstrate the relationship between the accelerometer data and environmental and personal factors and walking ability. With this data, we will have information need to conduct a larger, multisite trial that focuses on individuals who have new SCI. In the end we will be able to predict not just if someone will walk or use a wheelchair, but how much help they will need, how fast they can walk, and how far they can walk. This data has much more meaning, as it could tell someone if they’ll need bracing, if they could cross a street before the light changes, and
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Mar 10, 2021
- Source ID
- W81XWH2010724
Entities
People
- Michael Boninger
Organizations
- United States Army
- University of Pittsburgh