磁共振兼容CO₂负荷系统实现与脑血管反应性测量方法对比

Implementation of a magnetic resonance-compatible CO ₂ challenge system and comparison of cerebrovascular reactivity assessment methods

摘要

目的：自主搭建磁共振兼容CO₂负荷脑血管反应性（CVR）测量系统，并以CO₂吸入BOLD-CVR（CO₂-CVR）为标准，评估屏气BOLD-CVR（BH-CVR）和静息态BOLD-CVR（RS-CVR）的替代可行性。方法：前瞻性纳入30例健康志愿者，参考双管路开放式呼吸回路方案搭建磁共振兼容CO₂负荷系统。所有受试者在联影uMR 770 3.0T扫描仪上依次完成CO₂吸入、BH和RS这3种BOLD-fMRI任务的扫描。CO₂-CVR和BH-CVR基于一般线性模型计算，RS-CVR采用低频BOLD信号回归方法计算。3种CVR图经灰质均值归一化后，在体素水平和感兴趣区水平分别评估方法间一致性。结果：CO₂负荷系统运行稳定，全组扫描均顺利完成。体素水平分析显示，CO₂-CVR与BH-CVR的空间相关系数显著高于CO₂-CVR与RS-CVR（0.531±0.111vs 0.438±0.130,P=0.012）。感兴趣区水平各方法间ICC值普遍偏低，RS-CVR在椎基底动脉供血区呈现系统性低估。Bland-Altman分析显示CO₂-CVR与RS-CVR存在显著负向比例偏差（β=−0.547,P<0.001），CO₂-CVR与BH-CVR的比例偏差较轻。结论：成功搭建稳定的磁共振兼容CO₂负荷系统。BH-CVR在空间一致性和个体稳定性上均优于RS-CVR，是CO₂-CVR更可靠的替代方案。RS-CVR可反映CVR空间分布趋势，但在后循环区域存在系统性偏差，定量解读需谨慎。

Abstract

Objective To develop a magnetic resonance (MR)-compatible CO₂ challenge system for cerebrovascular reactivity (CVR) assessment, and further evaluate the feasibility of breath-hold blood oxygenation level-dependent (BOLD)-CVR (BH-CVR) and resting-state BOLD-CVR (RS-CVR) as alternatives to the reference standard CO₂-inhalation BOLD-CVR (CO₂-CVR). Methods Thirty healthy volunteers were prospectively enrolled. An MR-compatible CO₂ challenge system was built based on a dual-limb open-circuit breathing apparatus. All subjects sequentially underwent CO₂ inhalation, breath-hold, and resting-state BOLD-fMRI scans with a United Imaging uMR 770 3.0T scanner. CO₂-CVR and BH-CVR were computed using a general linear model, whereas RS-CVR was computed using a low-frequency BOLD signal regression method. After gray-matter mean normalization, inter-method agreement was evaluated at both the voxel level and region-of-interest level. Results The CO₂ challenge system operated reliably, and all subjects completed the entire scanning protocol successfully. Voxel-level analysis showed that the spatial correlation coefficient between CO₂-CVR and BH-CVR was significantly higher than that between CO₂-CVR and RS-CVR (0.531±0.111vs 0.438±0.130,P=0.012). ICC values at the region-of-interest level were generally low across methods, and RS-CVR systematically underestimated CVR in the vertebrobasilar territory. Bland-Altman analysis revealed a significant negative proportional bias for CO₂-CVR and RS-CVR (β=−0.547,P<0.001), whereas the proportional bias for CO₂-CVR and BH-CVR was mild. Conclusion A stable MR-compatible CO₂ challenge system is successfully developed. BH-CVR demonstrates superior spatial agreement and inter-individual stability compared with RS-CVR, representing a more reliable alternative to CO₂-CVR. RS-CVR can reflect the overall spatial distribution pattern of CVR but exhibits systematic bias in the posterior circulation; and therefore, its quantitative interpretation should be performed with caution.

关键词

脑血管反应性 / 功能磁共振成像 / 磁共振兼容CO₂负荷系统 / 屏气 / 静息态

Key words

cerebrovascular reactivity / functional magnetic resonance imaging / magnetic resonance-compatible CO₂ challenge system / breath-hold / resting-state

参考文献

[1] Fisher JA, Venkatraghavan L, Mikulis DJ . Magnetic resonance imaging—based cerebrovascular reactivity and hemodynamic reserve[J]. Stroke, 2018, 49(8): 2011-2018.

[2] Liu PY, Lin ZX, Hazel K, et al. Cerebrovascular reactivity MRI as a biomarker for cerebral small vessel disease—related cognitive decline: multi—site validation in the MarkVCID consortium[J]. Alzheimers Dement, 2024, 20(8): 5281-5289.

[3] van Niftrik CH, Sebök M, Germans MR, et al. Increased risk of recurrent stroke in symptomatic large vessel disease with impaired BOLD cerebrovascular reactivity[J]. Stroke, 2024, 55(3): 613-621.

[4] Sayin ES, Duffin J, Poublanc J, et al. MRI—based classification of cerebral hemodynamic failure with resting perfusion metrics and cerebrovascular reactivity[J]. Stroke, 2025, 56(10): 3034-3046.

[5] Greenberg SM, Albert MS, An H, et al. MarkVCID2 consortium for clinical validation of biomarkers of cerebral small vessel disease: validation framework and baseline characteristics[J]. Ann Neurol, 2026, 99(2): 449-458.

[6] Woods JG, Achten E, Asllani I, et al. Recommendations for quantitative cerebral perfusion MRI using multi—timepoint arterial spin labeling: acquisition, quantification, and clinical applications[J]. Magn Reson Med, 2024, 92(2): 469-495.

[7] Uniken Venema SM, Bhogal A, Dankbaar JW, et al. Benefits and challenges of multi—delay arterial spin labeling in clinical practice: measuring perfusion and cerebrovascular reactivity in intracranial steno—occlusive disease[J]. Insights Imaging, 2025, 16(1): 197.

[8] Sleight E, Stringer MS, Marshall I, et al. Cerebrovascular reactivity measurement using magnetic resonance imaging: a systematic review[J]. Front Physiol, 2021, 12: 643468.

[9] Pinto J, Bright MG, Bulte DP, et al. Cerebrovascular reactivity mapping without gas challenges: a methodological guide[J]. Front Physiol, 2021, 11: 608475.

[10] Liu PY, De Vis JB, Lu HZ . Cerebrovascular reactivity (CVR) MRI with CO₂ challenge: a technical review [J]. Neuroimage, 2019, 187: 104-115.

[11] Fisher JA, Mikulis DJ . Cerebrovascular reactivity: purpose, optimizing methods, and limitations to interpretation — a personal 20—year odyssey of (Re) searching[J]. Front Physiol, 2021, 12: 629651.

[12] Urback AL, MacIntosh BJ, Goldstein BI . Cerebrovascular reactivity measured by functional magnetic resonance imaging during breath—hold challenge: a systematic review[J]. Neurosci Biobehav Rev, 2017, 79: 27-47.

[13] Liu PY, Li Y, Pinho M, et al. Cerebrovascular reactivity mapping without gas challenges[J]. Neuroimage, 2017, 146: 320-326.

[14] Tancredi FB, Lajoie I, Hoge RD . A simple breathing circuit allowing precise control of inspiratory gases for experimental respiratory manipulations[J]. BMC Res Notes, 2014, 7: 235.

[15] Esteban O, Markiewicz CJ, Blair RW, et al. fMRIPrep: a robust preprocessing pipeline for functional MRI[J]. Nat Methods, 2019, 16(1): 111-116.

[16] Tustison NJ, Avants BB, Cook PA, et al. N4ITK: improved N3 bias correction[J]. IEEE Trans Med Imaging, 2010, 29(6): 1310-1320.

[17] Fischl B . FreeSurfer[J]. Neuroimage, 2012, 62(2): 774-781.

[18] Avants BB, Tustison NJ, Song G, et al. A reproducible evaluation of ANTs similarity metric performance in brain image registration[J]. Neuroimage, 2011, 54(3): 2033-2044.

[19] Jenkinson M, Beckmann CF, Behrens TE, et al. FSL[J]. NeuroImage, 2012, 62(2): 782-790.

[20] Cox RW . AFNI: software for analysis and visualization of functional magnetic resonance neuroimages[J]. Comput Biomed Res, 1996, 29(3): 162-173.

[21] Greve DN, Fischl B . Accurate and robust brain image alignment using boundary—based registration[J]. Neuroimage, 2009, 48(1): 63-72.

[22] Jezzard P, Balaban RS . Correction for geometric distortion in echo planar images from B0 field variations[J]. Magn Reson Med, 1995, 34(1): 65-73.

[23] Pruim RH, Mennes M, van Rooij D, et al. ICA—AROMA: a robust ICA—based strategy for removing motion artifacts from fMRI data[J]. Neuroimage, 2015, 112: 267-277.

[24] Rovai A, Lolli V, Trotta N, et al. CVRmap—a complete cerebrovascular reactivity mapping post—processing BIDS toolbox[J]. Sci Rep, 2024, 14(1): 7252.

[25] Liu PY, Liu GK, Pinho MC, et al. Cerebrovascular reactivity mapping using resting—state BOLD functional MRI in healthy adults and patients with moyamoya disease[J]. Radiology, 2021, 299(2): 419-425.

[26] Liu CF, Hsu J, Xu X, et al. Digital 3D brain MRI arterial territories atlas[J]. Sci Data, 2023, 10(1): 74.

[27] Liu PY, Baker Z, Li Y, et al. CVR—MRICloud: an online processing tool for CO₂—inhalation and resting—state cerebrovascular reactivity (CVR) MRI data [J]. PLoS One, 2022, 17(9): e0274220.

[28] Raghavan V, Sobczyk O, Sayin ES, et al. Assessment of cerebrovascular reactivity using CO₂—BOLD MRI: a 15—year, single center experience [J]. J Magn Reson Imaging, 2024, 60(3): 954-961.

[29] Zhao MY, Woodward A, Fan AP, et al. Reproducibility of cerebrovascular reactivity measurements: a systematic review of neuroimaging techniques[J]. J Cereb Blood Flow Metab, 2022, 42(5): 700-717.

[30] Sobczyk O, Sayin ES, Sam K, et al. The reproducibility of cerebrovascular reactivity across MRI scanners[J]. Front Physiol, 2021, 12: 668662.

基金资助

国家自然科学基金(82071280)

国家自然科学基金(82572374)