Neuroimaging, serum biomarkers, and patient characteristics as predictors of upper limb functioning 12 weeks after acute stroke: an observational, prospective study

ABSTRACT

Objective: To evaluate the potential of neuroimaging, serum biomarkers, stroke etiology, and clinical characteristics as predictors of upper limb functioning 12 weeks after stroke.

Methods: This was a prospective, observational study of patients (18–85 years-old) hospitalized due to acute ischemic stroke in the territory of the middle cerebral artery. Patients were hospitalized at a stroke rehabilitation center, where they underwent a standardized rehabilitation program. Clinical, imagiology, laboratory (biomarkers: C-reactive protein, D-dimer, and fibrinogen, and S100 calcium binding protein β [S100β]), and functionality assessments were conducted four different times: within 24 hours, and at 48 hours, 3 weeks, and 12 weeks after acute stroke.

Results: Upper limb functioning at 12 weeks was significantly associated with Alberta Stroke Program Early CT Score (ASPECTS) score (OR:2.012 [CI:1.349–3.000]; P = 0.001) and S100β protein levels (OR:0.997 [CI:0.994–0.999]; P = 0.007). Advanced age was associated with poor upper limb functioning. S100β protein levels < 140.5 ng/L at 48 hours and ASPECTS scores ≥ 7.5 within 24 hours of admission predicted good hand functioning at 12 weeks. Upper limb functioning and general functioning were significantly correlated (P < 0.001), with strong negative correlations (all correlation coefficients ≤–0.586) for all comparisons.

Conclusion: ASPECTS score ≥ 8 within 24 hours and S100β protein < 140.5 ng/L at 48 hours predict better upper limb functioning, while advanced age predicts worse upper limb functioning 12 weeks after stroke.

KEYWORDS:

• Stroke
• biomarkers
• s100β
• aspects
• rehabilitation
• upper limb
• hand
• functionality

Acknowledgments

The authors thank Graça Fernandes, BSN (Stroke Unit, Centro Hospitalar Universitário de Coimbra) for providing nursing care throughout the study; Marisa Loureiro, MSc (Laboratory of Biostatistics and Medical Informatics, Faculty of Medicine of the University of Coimbra) for providing assistance during statistical analysis; and Tiago Campos, MSc (ARC Publishing) for providing editorial assistance.

Conflicts of interest

The authors have no conflict of interest to declare.

Additional information

Funding

There were no external sources of funding.