Ilse Kant1,2, Jeroen de Bresser1,3, Simone van Montfort2, Ellen Aarts1,2, Ilona Bader2, Yarit Wiggerts2, Georg Winterer4, Claudia Spies4, Arjen Slooter2, and Jeroen Hendrikse1
1Radiology, UMC Utrecht, Utrecht, Netherlands, 2Intensive Care, UMC Utrecht, Utrecht, Netherlands, 3Radiology, Leiden University Medical Center, Leiden, Netherlands, 4Anesthesiology and Intensive Care, Charité Universitätsmedizin, Berlin, Germany
Synopsis
Physical frailty develops with
increasing age and is a chronic state of vulnerability that is associated with
disability and cognitive decline. Few studies have assessed the underlying structural
brain abnormalities of physical frailty. We therefore examined the association
between brain volumes on MRI and physical
frailty in a group of 214 non-demented elderly participants. Frail
participants showed a lower total brain volume and a lower grey matter volume
compared to pre-frail and non-frail participants. Furthermore, pre-frail
participants showed more cortical infarcts compared to non-frail participants.
These brain abnormalities could be the underlying substrate of the physical
frailty phenotype.
Introduction
Physical frailty develops with
increasing age and is a chronic state of vulnerability that is associated with
disability and cognitive decline. Frailty is most often described using the
physical frailty phenotype. This phenotype is assessed with five frailty
components: slowness, weakness, exhaustion, weight loss and a low level of
activity. Few studies have assessed the underlying structural brain abnormalities
of physical frailty. We therefore examined the association between brain
volumes and physical
frailty.Methods
A total of
214 non-demented participants (mean age 72.4 ± 4.9 years) were included on 2
sites (UMC Utrecht and Charité Universitätsmedizin Berlin) as part of the
Biomarker Development for Postoperative Cognitive Decline study (BioCog).
Participants were classified as non-frail (N=75), pre-frail (N=107) or frail
(N=32) by the Fried frailty phenotype. Brain volumes (total brain, white
matter, grey matter, white matter
hyperintensity and intracranial volume) were
automatically segmented on 3D T1 and 3D FLAIR images by the lesion prediction
algorithm, a lesion filling algorithm and CAT12. All scans were checked by a
radiologist (JB) for cortical infarcts and major artifacts that might hinder
accurate segmentations. The associations between brain volumes and frailty were
examined by linear regression analyses, corrected for age, gender, intracranial
volume and study center. Results
Participants
in the frail group showed a lower total brain volume (β (95% CI)= -22.40
(-40.38 – -4.41), p=0.02) and a lower grey matter volume (β (95% CI)=-12.26 (-23.09 –
-1.43), p=0.03) compared to the pre-frail group (see table 1). Participants in
the frail group also showed a lower total brain volume (β (95% CI)=-19.67 (-37.84 –
-1.50), p=0.03) and lower grey matter volume (β (95% CI)=-12.19 (-23.84 – -0.54), p=0.04)
compared to participants in the non-frail group (see table 1). Pre-frail
participants showed more cortical infarcts > 1.5 cm compared to non-frail participants
(OR (95% CI)= 4.66 (1.00 –
21.73), p=0.05, see table 2). No other significant differences were
found between the non-frail and pre-frail group.
Conclusion
Individuals with physical
frailty showed more brain abnormalities compared to pre-frail and non-frail individuals.
These brain abnormalities could be the underlying structural correlates of the
physical frailty phenotype.
Acknowledgements
The authors present this abstract on behalf of the BioCog Consortium,
clinicaltrials.gov identifier: NCT02265263, ethical approval number EA2/092/14 (Berlin), 14-469 (Utrecht).References
No reference found.