Perfusion MRI using endogenous deoxyhemoglobin as a contrast agent: preliminary data

Abstract

BACKGROUND: The paramagnetic properties of deoxyhemoglobin shorten T2* as do gadolinium based contrast agents. Induction of abrupt changes in arterial deoxyhemoglobin concentration ([dOHb]) can mimic the action of intra-vascular boluses of gadolinium based contrast agents (GBCAs) used for perfusion imaging. AIM: To demonstrate the feasibility of rapidly changing pulmonary venous hemoglobin saturation for generating boluses of blood with altered T2* properties for measuring flow metrics in the systemic circulation. METHODS: A gas blender with a sequential gas delivery breathing circuit and software enabling prospective arterial blood gas targeting was used to implement rapid lung changes in the partial pressure of blood oxygen (PaO2) while maintaining isocapnea. Lung PaO2 was initially lowered to induce a low baseline deoxyhemoglobin concentration [dOHb]. PaO2 was then rapidly raised to normal for 10 seconds and then rapidly lowered to the initial low baseline creating a oxyhemoglobin (OHb) bolus. R2* changes were measured using blood oxygenation dependent (BOLD) MRI signal changes in large arteries and veins as well as in the microcirculation. This enabled generation of the following maps: bolus arrival time delay (TD) cerebral blood volume (CBV), mean transit time (MTT) and cerebral blood flow (CBF). RESULTS: BOLD signal in the middle cerebral artery showed a sharp increase during the OHb bolus transit indicating minimal dispersion confirming effective rapid modulation of pulmonary venous PO2 with reasonable cortical contrast-to-noise ratio of 3. Signals sorted by amplitude of signal changes and arrival times enabled the visualization of major arteries and veins. Contrast to noise ratio was adequate for a single gas challenge to provide most of the contrast, little improved by averaging over the remaining set of challenges. Values of the flow metrics derived from the perfusion maps were within normal ranges from published literature values. CONCLUSION: Non-invasive induction of abrupt changes in OHb saturation can function as a novel non-invasive vascular contrast agent for use in perfusion imaging.

Cerebral perfusion imaging: Hypoxia-induced deoxyhemoglobin or gadolinium?

“…We have developed considerable expertise in inducing rapid, controlled changes in the P ET O 2 using techniques described as prospective end tidal targeting 14,15 and sequential gas delivery (SGD) 16,17 . Finally, at 3 Tesla, ~25% dOHb in arterial blood provides a T2* signal change sufficient to generate calculations of MTT, CBV and CBF whose magnitudes and distributions are consistent with those previously published for Gd-based perfusion imaging 10 .…”

“…Thus, like Gd, dOHb changes the homogeneity of a magnetic field inside and outside the blood vessels in proportion to its concentration ([dOHb]) 12 . Second, dOHb can be rapidly formed in the blood of pulmonary capillaries by breathing hypoxic gas as previously shown by Vu et al 13 and this process can be rapidly reversed by restoring normoxia by a single breath containing normal levels of oxygen 10 .…”

“…In designing this protocol, we estimated that a physiologically tolerable arterial PO 2 of about 40 mmHg would produce a [dOHb] change of about 25% 20 which would generate a sufficient change in the BOLD signal at 3T to test the value of dOHb as a susceptibility contrast agent 10 .…”

“…Consequently, there remains an urgent world-wide need for an MRI contrast agent that is non-invasive, economical, remains in the blood pool, does not accumulate in the body, is abundant, safe, and is environmentally innocuous. These challenging requirements may be addressed by hypoxia-induced deoxyhemoglobin (dOHb) 9,10 .…”

Quantifying cerebral blood arrival times using hypoxia-mediated arterial BOLD contrast

“…An additional caveat is that vasodilation can lead to changes in blood velocity that may modulate bolus arrival times 14 , or vascular steal and/or redistribution of blood flow that may confound true blood arrival and transit times in both healthy and disease affected tissues. In contrast to BOLD response to hypercapnia and in common with DSC, deoxy-hemoglobin can act as an arterial contrast agent 15,16 . By manipulating blood oxygen saturation at the lungs, hypoxia-BOLD (dOHb-BOLD) signal changes are reflective of both arteries and veins but without confounding effects related to altered hemodynamics.…”