Duncan F, Lewis SJ, Greig CA, Dennis MS, Sharpe M, MacLullich AMJ, and Mead GE. Exploratory Longitudinal Cohort Study of Associations of Fatigue After Stroke. Stroke. 2015
Fatigue is a familiar complaint in any clinical practice. Both experience and literature confirms the invisible symptom of post-stroke fatigue but there is no current treatment. The problem in studying this phenomenon is that fatigue can be a difficult variable to quantify and there are many potential confounders. An individual with an ACA stroke may be abulic, a widow forced to relocate to a nursing home may be depressed or a gentleman with a brainstem stroke might develop sleep apnea leading to poor sleep quality. When considering a patient’s reduced physical activity, is this a contributor to post stroke fatigue or simply a consequence?
Duncan et al study associations between fatigue and physical activity in stroke in hopes of better establishing a causal relationship. The study had three main aims: 1) Investigate if fatigue was significantly associated with measured physical activity at one, six and 12 months post stroke 2) Examine relationships between physical activity and patient characteristics at baseline and reported fatigue 3) Determine if physical activity was a predictor of later fatigue independent of other variables. The team measured levels of physical activity using an accelerometer and step count. 157 patients who had a stroke of any kind in the preceding month were enrolled. Unfortunately, only 136 attended any of the assessments with 58 returning for the final 12 month study visit.
The investigators looked for the presence of many conditions that have been previously noted in the literature to be associated with post-stroke fatigue. Stroke subtype and patient characteristics were extracted from the medical record. At time of enrollment, an NIHSS, mini-mental status exam as well as a physical activity scale for the elderly questionnaire was performed. At the one, six and 12 month assessments, the patients reported level of fatigue by filling out the Fatigue Assessment Scale (FAS) and also had a structured interview probing for clinically significant fatigue. Activity level was assessed using step count with an accelerometer worn at all times on the thigh unaffected by stroke. Additional assessment tools included anxiety and depression questionnaires, a quality of Life assessment (EuroQoL), Epworth Sleepiness Scale probing for sleep disorders and finally blood pressure.
Accelerometer (ActivPaL) data was available for 64%, 66% and 58% respectively of the total enrolled at each of the three assessments. The authors report that a positive fatigue case definition (structured interview mentioned above) was associated with lower daily step counts. At all three time point, more fatigue was associated with lower step count, higher depression, anxiety scores and sleepiness as well as a poorer quality of life. Lower daily step count and greater anxiety at one month independently predicted more fatigue at six and 12 months. The authors conclude that lower step counts at one month independently predicted fatigue for up to 12 months and therefore physical activity might serve as a potential target for intervention to prevent post-stroke fatigue.
How do we interpret these results? It is important to note that the study was powered to enroll 170 patients at baseline and have 120 at 12 month follow up. They had a high dropout and as per above only had 58 participants at 12 months. Another important consideration is that the NIHSS was more severe in the population without activity data limiting the generalizability of their results. The investigators should be applauded for using multiple scales to assess for any variables such as anxiety, depression, sleepiness, gender, stroke characteristics and pre-stroke fatigue that has been shown in the literature to have a relationship to post-stroke fatigue. The numbers once again limit application as they only had 18 patients at the 12 month time point where the case definition of fatigue was fulfilled. They were unable to perform logistic regression due to power so our ability to draw relationships is limited. While the concept that increasing physical activity may drive away fatigue is laudable for the clear cardiovascular and therapy benefits, more data is needed prior to stating that reduced activity leads to post stroke fatigue.