Venous Blood Oxygenation Measurements Using TRUST and T2‐TRIR MRI During Hypoxic and Hypercapnic Gas Challenges.

Background: Oxygen extraction fraction (OEF) and cerebral metabolic rate of oxygen (CMRO2) may serve as biomarkers in several diseases. OEF and CMRO2 can be estimated from venous blood oxygenation (Yv) levels, which in turn can be calculated from venous blood T2 values (T2b). T2b can be measured using different MRI sequences, including T2‐relaxation‐under‐spin‐tagging (TRUST) and T2‐prepared‐blood‐relaxation‐imaging‐with‐inversion‐recovery (T2‐TRIR). The latter measures both T2b and T1 (T1b) but was found previously to overestimate T2b compared to TRUST. It remained unclear, however, if this bias is constant across higher and lower oxygen saturations. Purpose: To compare TRUST and T2‐TRIR across a range of O2 saturations using hypoxic and hypercapnic gas challenges. Study Type: Prospective. Population: Twelve healthy volunteers (four female, age 36 ± 10 years). Field Strength/Sequence: A 3T; turbo‐field echo‐planar‐imaging (TFEPI), echo‐planar‐imaging (EPI), and fast‐field‐echo (FFE). Assessment: TRUST‐ and T2‐TRIR‐derived T2b, Yv, OEF, and CMRO2 were compared across different respiratory challenges. T1b from T2‐TRIR was used to estimate Hct (HctTRIR) and compared with venipuncture (HctVP). Statistical Tests: Shapiro–Wilk, one‐sample and paired‐sample t‐test, repeated measures ANOVA, Friedman test, Bland–Altman, and correlation analysis. Bonferroni multiple‐comparison correction was performed. Significance level was 0.05. Results: A significant bias was observed between TRUST‐ and T2‐TRIR‐derived T2b, Yv, and OEF values (−13 ± 11 msec, −5.3% ± 3.5% and 5.9 ± 4.1%, respectively). For Yv and OEF, this bias was constant across the range of measured values. T1b was significantly lower during severe hypoxia and hypercapnia compared to baseline (1712 ± 86 msec and 1634 ± 79 msec compared to 1757 ± 90 msec). While no significant bias was found between HctVP and HctTRIR (0.02% ± 0.06%, P = 0.20), the correlation between these Hct values was significant but weak (r = 0.19). Data Conclusion: Given the constant bias, TRUST‐ and T2‐TRIR‐derived venous T2b values can be used interchangeably to estimate Yv, OEF, and CMRO2 across a broad range of oxygen saturations. Hct from T2‐TRIR‐derived T1‐values only weakly correlated with Hct from venipuncture. Evidence Level: 2 Technical Efficacy: Stage 2. [ABSTRACT FROM AUTHOR]