ARTICLE IN BRIEF
A non-invasive MRI-based assessment combined with clinical data proved effective in forecasting functional outcomes after stroke in a small study.
The use of diffusion spectrum imaging, combined with clinical assessment measures, may help predict motor improvement after stroke, according to a new study published in the June 20 online edition of Neurology.
Investigators used an early and automated MRI-based assessment to forecast functional outcomes after stroke in this small study (12 subjects, 12 controls). The findings, though preliminary, offer a “proof of principle” that contra-lesional diffusion MRI measures may provide insight for personalized rehabilitation planning after an ischemic motor stroke, the authors wrote.
Lead author Cristina Granziera, MD, PhD, research group leader in the department of neurology at the University of Lausanne in Switzerland, told Neurology Today that the study aimed to explore the potential prognostic value of DSI (diffusion spectrum imaging), a relatively new and advanced technology in this field, in combination with simple demographic and clinical data — i.e. patient age and the National Institutes of Health Stroke Scale (NIHSS) — for predicting motor improvement after stroke.
“Our goal is to create a tool that is easy and quick — this model uses age, a routine clinical assessment, and a new measure of connectivity integrity,” she said. “It's potentially very simple and feasible with the emergency routine. If this method is found to be effective and accurate and we can get scan times reduced, this could have really important clinical implications.”
Bruce H. Dobkin, MD, director of the Neurological Rehabilitation and Research Program, and co-director of the Stroke Program at the Geffen School of Medicine at UCLA, who was not involved in the study, said that the investigators were able to gather some statistically important information from a small sample size. “Underlying all this is the notion that if the physicians and therapists had an idea about the amount of residual descending pathways for motor functioning within, for example, the corticospinal tract — then that information combined with the clinical exam would help us anticipate whether there was enough residual neural substrate to help promote improved movement and improved functional use of the dormant limb.
“They are making attempts to understand how much of the motor pathways are attacked — and how much, in a sense, are spared,” he added.
To date, the investigators wrote, no study has specifically focused on the structural remodeling of the motor network in stroke patients with motor deficits, despite the knowledge that motor deficits are a frequent cause of disability. So, Dr. Granziera and colleagues set out to do just that.
The study included 12 patients with ischemic stroke that affected the motor cortex and/or subcortical structures involved in motor function and 12 gender- and age-matched healthy controls. “We scanned the patients during the first week after stroke — in what we call the connectivity acute phase. Then we took a second time point after one month,” said Dr. Granziera. Then the third time point was after six months because, she said, the literature has shown this to be the peak of neuroplasticity after stroke.
The mean age of the study participants and standard deviation was 58.4 ± 17.0 years, which, the authors noted, was relatively young for stroke patients. They assessed post-stroke plasticity using diffusion spectrum MRI, which allows the mapping of the diffusion process of (mainly water) molecules in biological tissues in vivo and non-invasively.
The investigators found that contra-lesional structural changes in the motor network could be measured using DSI and were different in patients compared with healthy subjects (0.001 ≤ p < 0.05); moreover, the imaging changes in the contra-lesional motor tracts of patients correlated with functional motor improvement. The diffusion data at the first time point (one week post-stroke), combined with patient age and the acute clinical NIHSS score, proved to be important predictors of motor improvement six months after stroke (R2= 0.96, p= 0.0007).
“These highly sophisticated diffusion data are mostly useful in looking at complex intersections of fiber trajectories in a voxel as it happens after a lesion or some other damage to the brain, where restructuring of surrounding networks is occurring,” Dr. Granziera said. After stroke, fibers are going to be making complicated patterns in the brain, and likely in complex directions — so this type of imaging (DSI) allows for more specific information on these diffusion patterns as opposed to a more regional or one-directional imaging technology, she added.
Dr. Granziera said the investigators hope to confirm these data in larger cohorts and to explore other applications of the technology. “Further down the line, we're looking into implications of this data for robotic rehabilitation,” she added, noting that they are also currently testing this assessment model in multiple sclerosis patients.
Enrique C. Leira, MD, an associate professor of neurology and member of the Cerebrovascular Division at the University of Iowa College of Medicine in Iowa City, said that this study could have important implications for future clinical research in stroke recovery, but it's not yet close to primetime for practice. Since they only used twelve patients who were rather young and with single cerebral lesions and deficits (all of which the authors acknowledge), the value of this imaging tool needs to be further tested in a larger population of older stroke patients with co-morbidities that is more representative, he said.
“Even if these findings are proven accurate, they wouldn't necessarily have an immediate practical application for patient care, though I do see a potential use for clinical trials for recovery of stroke patients — perhaps that automated system could be used to not only predict recovery and stratify patients accordingly at the initial assessment, but also to monitor objectively the recovery and the formation of new connections over time,” he said.
“I think that they have shown the proof of principle that this technique may work in monitoring recovery,” Dr. Leira added, “but to gauge the value of this technique we have to really see it work in an actual recovery trial.”
Dr. Dobkin added that the study offers a nice window into the adaptations that are happening in the brain “that compensate for the partial loss of the corticospinal tract in the affected hemisphere — but the findings aren't likely to be adequate for day-to-day decision making after stroke in regard to patients who have some selective movement that is not yet very skillful.” Right now, he said, nothing is better for predicting motor recovery than a clinical examination done by someone with experience in treating and following stroke patients.
Aside from stratification for clinical trials, these data may be useful in studies that involve physical therapy or a drug intervention to try to enhance skills learning and recovery, or in neural repair strategies — in trying, for example, to regenerate some of the corticospinal tract, Dr. Dobkin said. “With these fine-grained imaging techniques, we might be able to better appreciate what's going on in the nervous system that is a result of these neural repair strategies.”