Background: The aetiology of fatigue after stroke is unknown. We explored the relationship between fatigue and C-reactive protein (CRP) as a marker of inflammation.
Methods: This cross-sectional study recruited inpatients with a stroke (onset within the previous three months) over a five-week period. Those with dysphasia or confusion severe enough to prevent informed consent and those with current infection were excluded. A semi-structured interview determined a) fulfilment of a case definition for fatigue and b) severity of fatigue (fatigue assessment scale, FAS). Venous blood was taken for CRP. A hospital anxiety and depression score (HADS) was used to screen for emotional distress.
Results: Of the 28 patients recruited (mean age 72.7 years, proportion men 47%), 15 (53%) fulfilled the case definition for fatigue. C-reactive protein data were logarithmically transformed for analysis. C-reactive protein levels did not differ significantly between those with and without fatigue, according to the case definition (n=28, p=0.35). After exclusion of those with pre-stroke fatigue and those with high scores on the HADS (suggestive of emotional distress), the geometric mean CRP of the fatigued group was 16.04 mg/l (95% CI: 7.12–36.14) compared with 5.16 mg/l (95% CI: 2.7–9.85) in the non-fatigued group (n=21, p=0.025, unpaired t test), but the relationship between FAS and CRP was not statistically significant (r=0.37, p= 0.098).
Conclusion: This pilot study is the first to demonstrate an association between fatigue after stroke and higher CRP, after excluding patients with pre-stroke fatigue and those with probable mood disorders. If this finding is confirmed in a larger number of patients, it might provide a target for treating fatigue after stroke.
Keywords C-reactive protein, fatigue, inflammation, stroke
Declaration of Interests Dr Gillian Mead is Assistant Editor of The Journal of the Royal College of Physicians of Edinburgh. This article was submitted to the JRCPE’s Clinical Editor and has been peer-reviewed in line with usual JRCPE procedures.
