Quantification of macrovesicular and microvesicular hepatic steatosis in rats using 3.0-T ¹H-magnetic resonance spectroscopy.

BACKGROUND AND OBJECTIVE: Hepatic steatosis (HE), which is common among the general population, is present in donor organs, potentially affecting their graft survival as well as the recovery of the donor. Our goal was to develop an experimentally and clinically reliable, noninvasive method to quantify macrovesicular and microvesicular hepatic steatosis using 3-T (1)H-magnetic resonance spectroscopy (MRS). MATERIALS AND METHODS: Macrovesicular and microvescular steatosis were induced in rats using methylcholine deficiency and choline deficiency diets. A MayoBC10 coil was used for radiofrequency transmission and signal recept. Measurements of hepatic fat content were performed using (1)H spectroscopy on a 3.0-T whole-body GE Signa system. The ratio of the areas under the curve of fat (0.8-1.3 ppm) and water (4.7 ppm) was used to determine hepatic fat content, which was compared with the degree of histopathologic and biochemical steatosis. RESULTS: Twenty rats were divided into two groups based on the percentage of microvesicular liver steatosis. Group A (n = 13) was the lower percentage group (microvesicular < 10%) while group B (n = 7), the higher group (microvesicular ≥ 10%). The mean total fatty change in the liver was 58.4% ± 47.2% and 67.6% ± 39.1% in groups A and B, respectively. A highly significant linear correlation between (1)H-MRS and total fatty change was observed in group A (r = .986, P < .001) while there was a relatively poor correlation in group B (r = .764, P = .05). The power to predict fatty change in the liver in groups A and B was significantly different (P = .004). CONCLUSIONS: The degree of hepatic steatosis with a small amount of microvesicular steatosis (<10%) can be precisely predicted using 3-T (1)H-MRS.

Brain alcohol detectability increase with repeated administration in humans: a proton spectroscopy study.

Proton MRS was used to detect brain alcohol after repeated alcohol exposure in human subjects. MRS detectability measurements were made after administration of an alcoholic drink (0.6 g/kg alcohol) and after an identical drink administrated 6 h later. Between-drink differences in the methyl proton triplet resonance of ethyl alcohol were assessed at statistically equivalent and near-peak blood alcohol concentrations (reflecting brain alcohol concentrations) and statistically equivalent internal standard N-acetyl resonance areas after Drinks 1 and 2, respectively. Brain alcohol detectability was not altered in TE 30-ms spectra but was increased in all five subjects after Drink 2 by an average of 70% in TE 270-ms spectra (p < 0.01). This was accompanied by significant between-drink differences in subjective ratings of alcohol's effects, suggestive of induction of acute alcohol tolerance. These findings suggest increased brain alcohol detectability in TE 270-ms spectra after repeated alcohol exposure that may reflect acute alcohol tolerance.

In vivo proton magnetic resonance spectroscopy detection of human alcohol tolerance.

Alcohol tolerance was ascertained with in vivo proton magnetic resonance spectroscopy (MRS) in men who regularly consumed either large (10-20 drinks/week) or small (2-4 drinks/weeks) amounts of beverage alcohol. Brain ethanol concentrations were determined by MRS, and blood ethanol levels were measured by gas chromatography after controlled ethanol administration (0.8 g/kg). Brain-blood ethanol concentration ratios for heavy drinkers were significantly greater than ratios for occasional drinkers (P < 0.002). Inasmuch as ethanol tolerance covaries with the severity of dependence, MRS procedures may facilitate our understanding of alcohol tolerance and treatment of alcoholism.

In vivo proton magnetic resonance spectroscopy of alcohol in rhesus monkey brain.

Brain alcohol was measured in rhesus monkeys (Macaca mulatta) by proton magnetic resonance spectroscopy (MRS) following acute nasogastric alcohol administration (0.8 g/kg). Monkeys were anesthetized with ketamine and xylazine. A 1.5 T whole body imager and a 3-inch surface coil were used to acquire TE 30 and 270 ms spectra from a 7.5 cc voxel localized with a stimulated echo (STEAM) sequence. Venous blood samples were collected during spectral acquisitions for gas chromatographic determination of temporally concordant blood alcohol levels (BALs). Acute alcohol administration did not alter the resonance areas of N-acetylaspartate/N-acetyl containing compounds (NAA), choline containing compounds, or total creatine. The NAA resonance was used as an internal standard to calculate approximate brain alcohol concentrations, which averaged 27 +/- 3% and 27 +/- 8% of temporally concordant BALs (T2-corrected TE 30 and TE 270 ms spectra, respectively). In addition to reconfirming results from prior studies finding incomplete detection of brain alcohol with MRS, these results demonstrate the feasibility of measuring brain alcohol in anesthetized nonhuman primates to examine relationships between alcohol exposure history and MRS-visibility of brain alcohol.

A nonferrous instrumental joystick device for recording behavioral responses during magnetic resonance imaging and spectroscopy.

A nonferrous joystick device was developed to permit subjects to continuously report ethanol-induced alterations in subjective mood states while undergoing a magnetic resonance imaging (MRI) and magnetic resonance spectroscopy (MRS) procedure. The device utilizes air pressure (supplied by a small compressor) that is directed to a series of tubes that terminate in a hand-held unit. The hand-held unit easily fits inside the magnet and resembles a standard computer game joystick except that the ends of the air hoses replace the buttons. The control unit contains three pressure transducers, which are triggered when the tubes are occluded by the subject, activating different pens on an event marker located 6 m from the whole body imager. The unit is safe to use inside a 1.5-Tesla magnetic field and does not disrupt the MRI and MRS recording procedures. Subjective reports of ethanol-induced euphoria and intoxication paralleled the MRS detection of ethanol in the brain. This device could prove to be useful in numerous behavioral studies involving whole-body MRI and MRS.

Induced and spontaneous seizures in man produce increases in regional brain lipid detected by in vivo proton magnetic resonance spectroscopy.

Elevations of brain concentrations of arachidonic acid and other free fatty acids (FFAs) by seizures induced in animals were demonstrated some years ago. Similarly, large shifts of potassium (K+) from intra- to extracellular space during seizure activity have been documented in numerous studies. More recent studies of cell membrane function demonstrated a direct effect of FFAs on membrane K+ conductance, suggesting that FFAs may play a primary role in seizure evolution in brain tissue. Using electroconvulsive therapy (ECT), in which generalized seizures are induced in patients by passage of electrical current, as a controlled human model of seizures, we studied the in vivo biochemical effects of single generalized seizures with localized proton magnetic resonance spectroscopy (1H MRS). We found that ECT reliably induces an elevation in the lipid signal that resonates at approximately 1.2 ppm. We observed a similar increase in brain lipids in a patient with temporal lobe epilepsy temporarily off medication; the signal disappeared after re-medication. Similar observations were noted for a subject with focal gliosis bordering a resected brain tumor. Finally, acute alcohol effects seem also to induce observable lipid changes. The 1H MRS technique does not yet permit direct identification of the specific lipids involved but analysis of cerebrospinal fluid obtained by lumbar puncture before and immediately after ECT may permit more precise characterization of the observed lipid increases. Theoretical and clinical implications of these results for the study of brain FFAs and epilepsy will be discussed.

In vivo 1H spectroscopy of the human brain following electroconvulsive therapy.

Recent advances in the methodology of magnetic resonance spectroscopy now permit localized proton (1H) spectroscopy of the human brain in clinical magnetic resonance systems. In this study, localized 1H spectroscopy was used to observe directly the stimulation of brain metabolic activity in patients undergoing electroconvulsive therapy (ECT) and to compare results obtained before and after treatments. Persistent increases in lactate were expected on the basis of animal data but these increases were small and equivocal 1 hour after ECT. In contrast, a large increase in a lipid signal from before to after ECT was observed in 5 patients when short echo times were used. We postulate that a significant portion of this lipid signal is related to maximal activation of the phosphatidylinositol system (increased levels of diacylglycerol and free fatty acids) have generalized inhibitory effects potentially relevant to both the clinical pathophysiology of seizures and the efficacy of ECT in major affective disorders.

In vivo proton magnetic resonance spectroscopy of alcohol in human brain.

The covariance between blood and brain alcohol levels and subjective reports of mood were examined in 6 healthy adult men after consumption of 0.7 g/kg of beverage alcohol. There was significant (p less than 0.01) temporal concordance between ascending and peak blood alcohol levels and regional brain alcohol levels as measured by in vivo proton Magnetic Resonance Spectroscopy (MRS) when N-acetyl aspartate (N-AA) concentration was used as an internal standard. The frequency of reports of both euphoria and dysphoria also paralleled the ascending limb of the blood and brain alcohol curve. However, peak blood alcohol levels were higher (125.67 +/- 10.91 mg/dl) and earlier (35 min postdrinking) than peak brain alcohol levels (26.25 +/- 6.38 mg/dl) detected 50 min after alcohol intake. This difference in brain and blood alcohol levels appears to be associated with the echo time (TE) parameters of the MRS. A decrease in TE from 270 msec to 50 msec resulted in a marked increase in brain alcohol detectability. MRS measures will permit analysis of regional differences in brain alcohol concentrations and covariance with behavioral, neurophysiologic and neuroendocrine concomitants of acute alcohol intoxication in man.