In this study, functional magnetic resonance imaging (fMRI) was used to evaluate in vivo hepatic oxygenation changes in chronically ethanol (CE)-treated and pair-fed (PF) control rats. Male Wistar rats were pair-fed an all-liquid diet containing 36% of total calories as either ethanol or dextrin-maltose for 8 weeks. The rats were initially examined under normoxic conditions, and then subjected to 100% oxygen (hyperoxia), 10% oxygen (hypoxia), 5% carbon dioxide (hypercapnia), or an acute dose of ethanol (1.4 g/kg bw intraperitoneally). A T(2)-weighted spin-echo sequence, which may be more selective for sinusoidal (capillary bed) changes, was performed before, during, and after the four challenges. During hyperoxia, both the CE and PF rats showed a statistically significant increase in signal intensity (22% +/- 5% and 48% +/- 6%, respectively, P < 0.05) relative to normoxia, while hypoxic challenge decreased the signal intensity (9% +/- 4%, p>0.05 and 15% +/- 3%, P < 0.05, respectively). The hypercapnic challenge, which causes vasodilation, resulted in a small increase in signal intensity in CE-fed rats (5% +/- 3%, P > 0.05) and a significant increase in the PF rats (15% +/- 4%, P < 0.05), again consistent with expected changes in deoxyhemoglobin. With all three physiological challenges, the degree of change was less in CE rats compared to PF controls. An acute dose of ethanol that causes vasodilation also increased signal intensity, with no significant difference between the two groups. The signal intensity changes seen with fMRI were highly correlated with pulse oximeter readings (r(2) = 0.95; P < 0.05). In conclusion, fMRI was shown to be a good noninvasive indicator of tissue deoxyhemoglobin changes in the liver. In addition, fMRI was able to detect subtle, early effects of CE administration (manifested as an impaired ability of the liver to respond adequately to oxygenation challenges), consistent with microvascular dysfunction.
Copyright 2003 Wiley-Liss, Inc.