Comparing single- and multi-post labeling delays for the measurements of resting cerebral and hippocampal blood flow for cerebrovascular testing in midlife adults

Kevin P Decker; Faria Sanjana; Nick Rizzi; Mary K Kramer; Alexander M Cerjanic; Curtis L Johnson; Christopher R Martens

doi:10.3389/fphys.2024.1437973

Comparing single- and multi-post labeling delays for the measurements of resting cerebral and hippocampal blood flow for cerebrovascular testing in midlife adults

Front Physiol. 2024 Oct 2:15:1437973. doi: 10.3389/fphys.2024.1437973. eCollection 2024.

Authors

Kevin P Decker¹, Faria Sanjana¹, Nick Rizzi¹, Mary K Kramer², Alexander M Cerjanic^{2

3}, Curtis L Johnson², Christopher R Martens¹

Affiliations

¹ Department of Kinesiology and Applied Physiology, University of Delaware, Newark, DE, United States.
² Department of Biomedical Engineering, University of Delaware, Newark, DE, United States.
³ Department of Neurology, Massachusetts General Hospital, Boston, MA, United States.

Abstract

Objectives: To assess the reliability and validity of measuring resting cerebral blood flow (CBF) and hippocampal CBF using a single-post-labeling delay (PLD) and a multi-PLD pseudo-continuous arterial spin labeling (pCASL) protocol for cerebrovascular reactivity (CVR) testing.

Methods: 25 healthy, midlife adults (57 ± 4 years old) were imaged in a Siemens Prisma 3T magnetic resonance imaging (MRI) scanner. Resting CBF and hippocampal CBF were assessed using two pCASL protocols, our modified single-PLD protocol (pCASL-MOD) to accommodate the needs for CVR testing and the multi-PLD Human Connectome Project (HCP) Lifespan protocol to serve as the reference control (pCASL-HCP). During pCASL-MOD, CVR was calculated as the change in CBF from rest to hypercapnia (+9 mmHg increase in end-tidal partial pressure of carbon dioxide [P_ETCO₂]) and then normalized for P_ETCO₂. The reliability and validity in resting gray matter (GM) CBF, white matter (WM) CBF, and hippocampal CBF between pCASL-MOD and pCASL-HCP protocols were examined using correlation analyses, paired t-tests, and Bland Altman plots.

Results: The pCASL-MOD and pCASL-HCP protocols were significantly correlated for resting GM CBF [r = 0.72; F (1, 23) = 25.24, p < 0.0001], WM CBF [r = 0.57; F (1, 23) = 10.83, p = 0.003], and hippocampal CBF [r = 0.77; F (1, 23) = 32.65, p < 0.0001]. However, pCASL-MOD underestimated resting GM CBF (pCASL-MOD: 53.7 ± 11.1 v. pCASL-HCP: 69.1 ± 13.1 mL/100 g/min; p < 0.0001), WM CBF (pCASL-MOD: 32.4 ± 4.8 v. pCASL-HCP: 35.5 ± 6.9 mL/100 g/min; p = 0.01), and hippocampal CBF (pCASL-MOD: 50.5 ± 9.0 v. pCASL-HCP: 68.1 ± 12.5 mL/100 g^/min; p < 0.0001). P_ETCO₂ increased by 8.0 ± 0.7 mmHg to induce CVR (GM CBF: 4.8% ± 2.6%; WM CBF 2.9% ± 2.5%; and hippocampal CBF: 3.4% ± 3.8%).

Conclusion: Our single-PLD pCASL-MOD protocol reliably measured CBF and hippocampal CBF at rest given the significant correlation with the multi-PLD pCASL-HCP protocol. Despite the lower magnitude relative to pCASL-HCP, we recommend using our pCASL-MOD protocol for CVR testing in which an exact estimate of CBF is not required such as the assessment of relative change in CBF to hypercapnia.

Keywords: and arterial spin labeling (ASL) MRI; brain health; cerebral blood flow (CBF); cerebrovascular reactivity (CVR); magnetic resonance imaging (MRI).

Grants and funding

The author(s) declare that financial support was received for the research, authorship, and/or publication of this article. NIH/NIGMS - P20 GM113125 R01-AG058853.