Effect of T1 relaxation on ventilation mapping using hyperpolarized 129Xe multiple breath wash-out imaging

Purpose: To investigate the effect of incorporating T-1 as a function of wash-out breath number (T-1(n)) on estimation of fractional ventilation (r) using hyperpolarized Xe-129 multiple breath wash-out (MBWO) imaging in rats. Methods: MBWO imaging was performed in 8 healthy mechanically ventilated rats at several inter-image delay times (tau) and tidal volumes (TV). r maps were calculated from the imaging data using a model of T-1(n) (assuming that the longitudinal relaxation rate of Xe-129 in the lung is directly proportional to p(A)O(2)) and compared to r maps obtained by assuming a fixed T-1 measured before wash-out breaths (r(1)). Results: Fractional ventilation was overestimated by up to 19.3% when T-1 was fixed. An inverse relationship between bias (Delta r) and ventilation was observed at all tau and TV. Additionally, Delta r significantly increased when TV was decreased (F statistic F (2,7) = 48.97, P < 10(-4)). Histograms from r(1) maps were significantly more skewed toward lower values as compared to r histograms at all tau and TV (P < 0.05) except TV = V-dose - 1 mL. Conclusion: Analysis of hyperpolarized Xe-129 MBWO imaging using a model incorporating T-1(n) corrects for an overestimating bias in the mapping of fractional ventilation in mechanically ventilated rats introduced by assuming a fixed T-1.