RSNA 2009 

Abstract Archives of the RSNA, 2009


SSK19-09

Noninvasive Evaluation of Chronic Liver Disease Using 1H and 23Na MR with Subsequent Examination of Underlying Physiological Phenomena Using Shift Reagent-aided 23Na and 31P MRS

Scientific Papers

Presented on December 2, 2009
Presented as part of SSK19: Physics (MR Spectroscopy)

Participants

Paige Nicole Hopewell MS, Presenter: Nothing to Disclose
Navin Bansal PhD, Abstract Co-Author: Nothing to Disclose

PURPOSE

The inability to noninvasively detect hepatitis in chronic liver diseases prompted the evaluation of nonalcoholic fatty liver disease (NAFLD) progression using 1H MRI/S and single quantum (SQ) and multiple quantum-filtered (MQF) 23Na MRI. The physiological mechanisms underlying these findings were examined using shift reagent (SR)-aided 23Na and 31P MRS.

METHOD AND MATERIALS

MR data were collected before and 2, 5, 10 and 15 weeks after placing rats on a methionine- and choline-deficient diet (MCDD). Fat-to-water ratios (F/W) were calculated from 1H MRI with chemical selective suppression applied at 4.7 ppm (H2O) or 1.3 ppm (CH2)n. 1H MRS were collected with localized adiabatic selective refocusing (LASER) and fit to combined Lorentzian/Gaussian functions. SQ and MQF 23Na MRI were collected to measure total tissue and intracellular Na+. SR-aided 23Na and 31P MRS experiments were performed to measure relative intra- and extracellular spaces and [Na+] before histological sample collection. A linear mixed model was used for statistical analysis (p<0.02).

RESULTS

F/W from 1H MRI and methylene-to-water ratios ((CH2)n/H2O) from 1H MRS increased dramatically (747% and 1520% of baseline values, respectively) and negatively correlated with 23Na MRI signal intensity (SI). Undetectable at baseline, unsaturated fatty acids (UFA)/(CH2)n peaked at week 10 (0.23±0.01) and correlated with 23Na MRI SI. SQ and MQF 23Na SI decreased by week 5 (66.7% and 73.3%) and increased at week 10 (121.9% and 163.6% of week 5). Histological evaluation showed increased lipid at week 2, mild fibrosis by week 10 and advanced fibrosis at week 15. SR data showed increased rECS, yet no change in ATP/Pi.

CONCLUSION

The difference between F/W and (CH2)n/H2O is explained by the ~2 ppm bandwidth of the suppression pulse that affects surrounding peaks (e.g., UFA signal at 5.3 ppm during H2O suppression). Many lipid resonances observed with 1H MRS, not only (CH2)n and H2O, are important in NAFLD. The decrease in SQ and MQF 23Na MRI SI at week 2 and 5 and increase at weeks 10 and 15 is hypothesized to be mostly from Na+’s insolubility in lipid. However, hepatitis and fibrosis-associated macromolecule deposition can also increase MQF SI and was observed with histological evaluation at weeks 10 and 15.

CLINICAL RELEVANCE/APPLICATION

23Na MRI with localized 1H MRI can noninvasively assess chronic liver disease and correlates with  biopsy, prompting further clinical development.

Cite This Abstract

Hopewell, P, Bansal, N, Noninvasive Evaluation of Chronic Liver Disease Using 1H and 23Na MR with Subsequent Examination of Underlying Physiological Phenomena Using Shift Reagent-aided 23Na and 31P MRS.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8010194.html