Diffusion-Weighted MRI for Quantification of Liver Fibrosis: Preliminary Experience
Abstract
OBJECTIVE. The purpose of this study was to evaluate our preliminary experience using diffusion-weighted MRI for quantification of liver fibrosis.
SUBJECTS AND METHODS. Diffusion-weighted MRI with single-shot echo-planar technique at b values of 50, 300, 500, 700, and 1,000 s/mm2 was prospectively performed on 23 patients with chronic hepatitis and on seven healthy volunteers. The apparent diffusion coefficient (ADC) was measured in four locations in the liver. Liver biopsy results (n = 19) were retrospectively reviewed by two hepatopathologists in consensus to determine stage of fibrosis and grade of inflammation. A Mann-Whitney test was used to compare the ADCs between patients classified with respect to having stage 2 or greater versus stage 1 or less fibrosis and stage 3 or greater versus stage or less 2 fibrosis. Receiver operating characteristics analysis was used to assess the performance of ADC in prediction of the presence of stage 2 or greater and stage 3 or greater fibrosis.
RESULTS. Using a b value of 500 s/mm2 and all combined b values, we found significantly lower hepatic ADCs in stage 2 or greater versus stage 1 or less fibrosis and stage 3 or greater versus stage 2 or less fibrosis. The mean ADCs (× 10–3 mm2/s) with all b values were 1.47 ± 0.11 (SD) versus 1.65 ± 0.10 for stage 2 or greater versus stage 1 or less fibrosis (p < 0.001) and 1.44 ± 0.07 versus 1.66 ± 0.10 for stage 3 or greater versus stage 2 or less fibrosis (p <0.001). Hepatic ADC was a significant predictor of stage 2 or greater and stage 3 or greater fibrosis, with areas under the curve of 0.896 and 0.896, sensitivity of 83.3% and 88.9%, and specificity of 83.3% and 80.0% (ADC with all b values, 1.54–1.53 × 10–3 mm2/s or less).
CONCLUSION. Diffusion-weighted MRI can be used for prediction of the presence of moderate and advanced liver fibrosis.
Introduction
Patients with chronic hepatitis B and hepatitis C virus infections are at high risk of development of hepatic fibrosis and cirrhosis, which can lead to end-stage liver disease, portal hypertension, and hepatocellular carcinoma. In chronic viral hepatitis, evaluation of disease severity and the indications for antiviral therapy usually rely on histologic findings obtained at liver biopsy performed to assess degree of fibrosis (stage) and necroinflammatory changes (grade) [1–3]. Liver biopsy, however, has inherent risks [4–6] and is prone to interobserver variability and sampling error [7–9]. The sensitivity of conventional MRI in the detection of liver fibrosis and early cirrhosis is limited [10], and noninvasive imaging techniques have not yet been definitely established for the detection of liver fibrosis.
With diffusion-weighted MRI (DWI) water diffusion is quantified by calculation of the apparent diffusion coefficient (ADC), which can be used for in vivo quantification of the combined effects of capillary perfusion and diffusion [11]. Several studies have shown a decrease in hepatic ADC in liver cirrhosis [12–16]. There are limited data, however, on the correlation between hepatic ADC and degree of hepatic fibrosis [17, 18]. The objective of our study was to determine with histopathologic findings as the reference standard whether hepatic ADC calculated with DWI can be used to quantify liver fibrosis in patients with chronic liver disease.
Subjects and Methods
Patients
This study, conducted in compliance with the Health Insurance Portability and Accountability Act, had a population of 30 subjects: 23 patients with chronic hepatitis (15 men, eight women; mean age, 54 years; range, 39–77 years) and seven healthy volunteers (five men, two women; mean age, 32 years; range, 28–39 years). Patients with chronic hepatitis were referred from the hepatology clinic of a tertiary care center. Liver disease was diagnosed on the basis of pertinent clinical history, results of liver function tests, and results of percutaneous liver biopsy that was clinically indicated. The causes of liver disease were chronic hepatitis C virus infection (n = 16), chronic hepatitis B virus infection (n = 2), nonalcoholic steatohepatitis (n = 2), autoimmune hepatitis (n = 2), and alcohol abuse (n = 1). None of the healthy volunteers had a history of liver disease or alcohol abuse. MRI was performed on all subjects as part of a prospective research study conducted during the 8-month period December 2004–July 2005. The protocol was approved by our local institutional review board, and informed signed consent was obtained from all participants.
Diffusion-Weighted MRI
DWI of the liver was performed on a 1.5-T 32-channel system (Magnetom Avanto, Siemens Medical Solutions) with high-performance gradients (maximum gradient, 45 mTm–1; maximum slew rate, 200 Tm–1s–1). A transverse breath-hold single-shot echo-planar imaging sequence was performed with an eight-element phased-array superficial coil and finger pulse triggering [19]. The following parameters were used: TR/TE range, 1,300/51–71; matrix size, 192 × 256; field of view, 320–400 mm; slice thickness, 7 mm; gap, 1.4 mm; number of signals averaged, 4; number of slices in middle portion of liver, 4 (with a localizer in the coronal plane for slice placement); frequency-selective fat suppression to reduce chemical shift artifacts; parallel imaging with generalized autocalibrating partially parallel acquisitions factor 2 (to decrease acquisition time and improve image quality) [20]. Five breath-hold acquisitions were obtained in the same liver location at b values of 0–50, 0–300, 0–500, 0–700, and 0–1,000 s/mm2 with tridirectional diffusion gradients. The acquisition time was less than 25 seconds.
Automatic voxel-by-voxel analysis on a commercial workstation (Syngo, Siemens Medical Solutions) was used to obtain ADC maps for each b value (50, 300, 500, 700, and 1,000 s/mm2) and for all b values combined. Hepatic ADCs were calculated in four locations within the liver for each of the five b values and for all b values combined (same slice location). An experienced observer placed regions of interest (ROIs) in the same locations for all b values and the combination of all b values. ADCs were measured in the lateral and medial segments of the left lobe and the anterior and posterior segments of the right lobe (Fig. 1A) with round ROIs approximately 1–2 cm in diameter in locations away from normal intrahepatic vasculature and focal liver lesions (one ROI per segment, four ROIs per patient). The final ADC was the average of the four ROIs. A routine MRI examination of the liver was performed after the DWI sequence only if clinically indicated.
Histopathology
Before MRI (mean delay, 42 days; range, 9–70 days), 19 of 23 patients with chronic hepatitis underwent blinded percutaneous liver biopsy by an experienced hepatologist using a 20-gauge Menghini needle without sonographic guidance. Liver biopsy was not performed on four patients with cirrhosis diagnosed on the basis of clinical and imaging criteria [21, 22] or on the healthy volunteers. The liver biopsy findings were retrospectively evaluated by two experienced hepatopathologists in consensus. These reviewers used the Batts-Ludwig classification to record stage of liver fibrosis and grade of necroinflammatory changes [23]. This scoring system has a five-point scale for both staging and grading. Staging is used to evaluate the degree of fibrosis: stage 0, no fibrosis; stage 1, portal fibrosis; stage 2, periportal fibrosis; stage 3, septal fibrosis; stage 4, cirrhosis. Grading of activity refers to the degree of hepatocellular necroinflammatory activity: grade 0, no activity; grade 1, minimal; grade 2, mild; grade 3, moderate; and grade 4, severe activity.
Statistical Analysis
SAS version 9.0 (SAS Institute) was used for analysis. A nonparametric Mann-Whitney test was used to compare hepatic ADCs between patients stratified according to individual fibrosis stage and patients grouped as stage 1 or less versus stage 2 or greater and stage 2 or less versus stage 3 or greater and between patients stratified by inflammation grade (grade 0 vs grade 1 or greater). The Spearman's rank correlation test was used to assess the correlation between hepatic ADC and stage of fibrosis and grade of inflammation. Binary logistic regression and receiver operating characteristic (ROC) curve analyses were conducted to evaluate the utility of the ADC measures for prediction of stage 2 or greater and stage 3 or greater fibrosis and for prediction of grade 1 or greater inflammation. Segmental variation of hepatic ADC was expressed in terms of the coefficient of variation [100% × (SD/mean)] derived from ROI measurements in four liver locations (right posterior, right anterior, left medial, and left lateral lobes) within one subject. By means of a Mann-Whitney test, coefficients of variation for each patient and each b value were compared with different b values and with values for patients stratified according to fibrosis stage. A value of p <0.05 was considered significant.
Results
Histopathologic Findings
The distribution of stages of fibrosis and grades of inflammation is shown in Table 1. The four patients with chronic liver disease who did not undergo liver biopsy were considered to have stage 4 fibrosis on the basis of MRI findings clearly showing liver cirrhosis according to established criteria [21, 22]. Inflammation grade was not available for these patients. No patient had severe inflammation (grade 4). The seven healthy volunteers were assumed to have stage 0 fibrosis and grade 0 inflammation.
a
Includes seven healthy volunteers.
b
Includes four patients with cirrhosis diagnosed on the basis of clinical and MRI criteria.
c
Grade not obtained for four patients with cirrhosis who did not undergo biopsy.
DWI Findings
Prediction of stage of fibrosis with ADC— The distribution of hepatic ADCs in patients stratified according to stage of fibrosis is shown in Table 2. There was a trend toward a decrease in hepatic ADC with increasing degree of fibrosis. There was moderate but significant negative correlation between ADC and fibrosis stage, the r values ranging from –0.448 to –0.654 (p = 0.0003–0.01680). The best correlation was observed for the combination of all b values (r = –0.654, p = 0.001).
| Fibrosis Stage | b Value (s/mm2) | |||||
|---|---|---|---|---|---|---|
| 50 | 300 | 500 | 700 | 1,000 | 0-1,000a | |
| 0 | 3.21 ± 0.98 | 1.95 ± 0.25 | 1.60 ± 0.18 | 1.42 ± 0.14 | 1.19 ± 0.12 | 1.66 ± 0.12 |
| 1 | 2.41 ± 1.07 | 1.81 ± 0.25 | 1.61 ± 0.19 | 1.40 ± 0.13 | 1.23 ± 0.12 | 1.68 ± 0.07 |
| 2 | 2.60 ± 1.30 | 1.68 ± 0.25 | 1.46 ± 0.20 | 1.26 ± 0.16 | 1.04 ± 0.12 | 1.56 ± 0.04 |
| 3 | 2.48 ± 0.85 | 1.84 ± 0.26 | 1.38 ± 0.17 | 1.11 ± 0.14 | 1.02 ± 0.11 | 1.50 ± 0.04 |
| 4 | 1.83 ± 0.84 | 1.38 ± 0.20 | 1.22 ± 0.15 | 1.14 ± 0.11 | 1.01 ± 0.10 | 1.41 ± 0.07 |
a
Combination of b values of 0, 50, 300, 500, 700, and 1,000 s/mm2.
There were significant differences between stage 0 and stage 4 fibrosis for all b values and the combination of all b values (p = 0.007–0.02), between stage 0 and stage 2 fibrosis only for a b value of 1,000 s/mm2 (p < 0.05), between stage 0 and stage 3 fibrosis for b values of 700 (p < 0.01) and 1,000 (p < 0.05) s/mm2, between stage 1 and stage 2 fibrosis only for a b value of 1,000 s/mm2 (p < 0.02) and the combination of all b values (p < 0.05), between stage 1 and stage 3 fibrosis for a b value of 700 s/mm2 (p <0.03) and the combination of all b values (p < 0.05), between stage 1 and stage 4 fibrosis for all b values except 50 s/mm2 (p = 0.01–0.03), and between stage 2 and stage 3 fibrosis only for b values of 700 (p < 0.05) and 1,000 (p < 0.02) s/mm2. There were no significant differences between stage 0 and stage 1 fibrosis, stage 2 and stage 4 fibrosis, and stage 3 and stage 4 fibrosis for all b values. Except for b values of 50 s/mm2 (diagnosis of stage 2 or greater and 3 or greater fibrosis) and 300 s/mm2 (diagnosis of stage 3 or greater fibrosis), at all b values there was a significant decrease in hepatic ADC in patients with stage 2 or greater versus stage 1 or less fibrosis and in patients with stage 3 or greater versus stage 2 or less fibrosis (Figs. 1B, 1C, 1D, 1E, 1F, 1G, 1H, Table 3).
| Fibrosis Stage | b Value (s/mm2) | |||||
|---|---|---|---|---|---|---|
| 50 | 300 | 500 | 700 | 1,000 | 0-1,000a | |
| ≤ 1 | 2.88 ± 1.07 | 1.90 ± 0.25 | 1.60 ± 0.19 | 1.41 ± 0.15 | 1.19 ± 0.12 | 1.65 ± 0.10 |
| ≥ 2 | 2.19 ± 0.86 | 1.60 ± 0.23 | 1.33 ± 0.17 | 1.17 ± 0.13 | 1.02 ± 0.11 | 1.47 ± 0.11 |
| p | 0.15 | < 0.03 | < 0.003 | < 0.001 | < 0.001 | < 0.001 |
| ≤ 2 | 2.89 ± 1.06 | 1.85 ± 0.25 | 1.57 ± 0.19 | 1.38 ± 0.14 | 1.16 ± 0.12 | 1.66 ± 0.10 |
| ≥ 3 | 2.09 ± 0.84 | 1.56 ± 0.23 | 1.28 ± 0.16 | 1.13 ± 0.12 | 1.02 ± 0.10 | 1.44 ± 0.07 |
| p | 0.06 | 0.06 | < 0.001 | < 0.001 | < 0.003 | < 0.001 |
Note—Liver apparent diffusion coefficient is significantly decreased in patients with moderate and advanced fibrosis at b values of 500 s/mm2 or greater or a combination of all b values (0, 50, 300, 500, 700, and 1,000 s/mm2). Statistically significant values are displayed in boldface.
a
Combination of b values of 0, 50, 300, 500, 700, and 1,000 s/mm2.
Using ROC analysis, we found hepatic ADC to be a significant predictor of stage 2 or greater and of stage 3 or greater fibrosis (Table 4). For example, for prediction of stage 2 or greater fibrosis with all b values, we found an area under the curve (AUC) of 0.896 with a sensitivity of 83.3%, specificity of 83.3%, positive predictive value of 84.0%, negative predictive value of 83.0%, and accuracy of 83.3% for a hepatic ADC of 1.54 mm2/s or less. For prediction of stage 3 or greater fibrosis at all b values, we found an AUC of 0.896 with a sensitivity of 88.9%, specificity of 80.0%, positive predictive value of 92.8%, negative predictive value of 92.8%, and accuracy of 87.5% for a hepatic ADC of 1.53 mm2/s or less. Corresponding ROC curves are shown in Figures 2 and 3.
| Prediction of Stage 2 or Greater Fibrosis | Prediction of Stage 3 or Greater Fibrosis | |||||||
|---|---|---|---|---|---|---|---|---|
| b Value (s/mm2) | AUC | ADC Criterion (value × 10-3 mm2/s) | Sensitivity (%) | Specificity (%) | AUC | Criterion (value × 10-3 mm2/s) | Sensitivity (%) | Specificity (%) |
| 50 | 0.677 | ≤ 2.97 | 92.3 | 40.0 | 0.717 | ≤ 1.41 | 40.0 | 100.0 |
| 300 | 0.714 | ≤ 1.48 | 50.0 | 93.3 | 0.716 | ≤ 1.39 | 50.0 | 94.7 |
| 500 | 0.786 | ≤ 1.45 | 71.4 | 75.0 | 0.835 | ≤ 1.34 | 70.0 | 85.0 |
| 700 | 0.882 | ≤ 1.31 | 92.3 | 73.3 | 0.901 | ≤ 1.17 | 66.7 | 100.0 |
| 1,000 | 0.868 | ≤ 1.08 | 85.7 | 81.2 | 0.832 | ≤ 1.03 | 80.0 | 90.0 |
| 0-1,000a | 0.896 | ≤ 1.54 | 83.3 | 83.3 | 0.896 | ≤ 1.53 | 88.9 | 80.0 |
Note—Sensitivity and specificity are calculated when apparent diffusion coefficient is used to diagnose stage 2 or greater or stage 3 or greater fibrosis.
a
Combination of b values of 0, 50, 300, 500, 700, and 1,000 s/mm2.
Prediction of grade of inflammation with ADC—There was a trend toward a decrease in hepatic ADC with increasing degree of inflammation. There was weak to moderate correlation between ADC and inflammation grade, the r values ranging from –0.292 to –0.516 (p = 0.01–0.139). The best correlation was observed for a b value of 700 s/mm2 (r = –0.516, p = 0.01). Hepatic ADC was significantly decreased in patients with mild to moderate inflammation (grades 1–3) versus no inflammation (grade 0) (Table 5). There was a significant difference at b values of 300, 500, and 700 s/mm2 and at the combination of all b values. Using ROC analysis, we found ADC to be a significant predictor of grade 1 or greater inflammation with an AUC of 0.875, sensitivity of 85.7%, specificity of 75.0%, positive predictive value of 87.5%, negative predictive value of 75.0%, and accuracy of 83.3% for an ADC 1.35 mm2/s or less (b = 700 s/mm2). AUC values corresponding to all b values are shown in Table 6.
| Inflammation Grade | b Value (s/mm2) | |||||
|---|---|---|---|---|---|---|
| 50 | 300 | 500 | 700 | 1,000 | 0-1,000a | |
| 0 | 3.34 ± 0.99 | 2.09 ± 0.25 | 1.68 ± 0.19 | 1.47 ± 0.15 | 1.20 ± 0.12 | 1.71 ± 0.07 |
| ≥ 1 | 2.58 ± 1.02 | 1.71 ± 0.25 | 1.45 ± 0.18 | 1.26 ± 0.14 | 1.10 ± 0.11 | 1.56 ± 0.10 |
| p | 0.07 | < 0.04 | < 0.03 | < 0.004 | 0.09 | < 0.04 |
Note—Statistically significant values are displayed in boldface.
a
Combination of b values of 0, 50, 300, 500, 700, and 1,000 s/mm2.
| b Value (s/mm2) | AUC | ADC Criterion (value × 10-3 mm2/s) | Sensitivity (%) | Specificity (%) |
|---|---|---|---|---|
| 50 | 0.739 | ≤ 2.49 | 64.7 | 85.7 |
| 300 | 0.786 | ≤ 1.94 | 83.3 | 71.4 |
| 500 | 0.792 | ≤ 1.48 | 66.7 | 87.5 |
| 700 | 0.875 | ≤ 1.35 | 85.7 | 75.0 |
| 1,000 | 0.712 | ≤ 1.08 | 66.7 | 87.5 |
| 0-1,000a | 0.881 | ≤ 1.61 | 71.4 | 83.3 |
Note—Sensitivity and specificity are calculated when hepatic apparent diffusion coefficient is used to diagnose grade 1 or greater inflammation.
a
Combination of b values of 0, 50, 300, 500, 700, and 1,000 s/mm2.
Segmental ADC variation—Regional variation in hepatic ADC was significantly smaller with use of the combination of all b values (coefficient of variation, 6.85% ± 2.70%) than with use of individual b values (p < 0.0001–0.005). The greatest ADC variation (23.32% ± 14.78%) was observed at a b value of 50 s/mm2; this variation was significantly greater than that observed at other b values (b = 300 s/mm2, 12.93% ± 6.85%; b = 500 s/mm2, 13.31% ± 7.02%; b = 700 s/mm2, 10.42% ± 5.13%; b = 1,000 s/mm2, 10.69% ± 4.46%) (p < 0.0007–0.001). We did not observe significant differences in the coefficients of variation of ADC between patients stratified according to fibrosis stage (stage 0, 12.49% ± 8.10%; stage 1, 11.60% ± 5.44%; stage 2, 14.80% ± 8.60%; stage 3, 13.97% ± 9.15%; stage 4, 13.47% ± 13.27%) (p = 0.19–1.0).
Discussion
In patients with chronic viral hepatitis, the histopathologic findings are used to assess prognosis, guide antiviral therapy, and predict treatment efficacy [2, 24]. Although it is a relatively safe procedure when performed by experienced clinicians, liver biopsy has poor patient acceptance and is not risk-free. Previous findings [4–6] have suggested a risk of hospitalization of 1–5%, a 0.57% risk of severe complications, and a mortality rate of 1:1,000–1:10,000. Liver biopsy also is prone to interobserver variability and sampling error [8, 9]. The development of noninvasive biomarkers of liver fibrosis and inflammation would reduce biopsy-related risk and costs and facilitate earlier diagnosis and improved monitoring of progression of chronic viral hepatitis. A number of serologic markers of liver fibrosis have been developed in chronic hepatitis C. The techniques include simple tests such as aspartate transaminase-to-alanine transaminase ratio, platelet count, and prothrombin index and more complex tests, such as the FibroTest (BioPredictive) developed by Imbert-Bismut and colleagues [25], which is based on a combination of basic serum markers. This test panel performed with 75% sensitivity and 85% specificity in the diagnosis of fibrosis in Metavir stage F2 or greater [26]. In a subsequent publication, the same group [27] found poorer performance (AUC, 0.76 ± 0.03) of the FibroTest and ActiTest (BioPredictive) (activity index, which incorporates alanine transaminase) in patients treated for chronic hepatitis C.
MRI has become an increasingly important imaging technique in the evaluation of chronic liver disease. Conventional MRI has a limited role, however, in the evaluation of disease activity in chronic hepatitis. Using gadolinium-enhanced dynamic MRI, Semelka et al. [28] described two parenchymal enhancement patterns of chronic hepatitis: early patchy enhancement correlating with inflammatory changes in the liver and late linear enhancement correlating with the presence of fibrosis. Several morphologic criteria have been described [10, 29, 30]. These criteria, however, are limited in sensitivity and specificity and are subject to interobserver variability. For example, the caudate-to-right lobe ratio measured on contrast-enhanced images had limited value in the diagnosis of cirrhosis (sensitivity, 71.7%; specificity, 77.4%; accuracy, 74.2% at a caudate-to-right lobe ratio > 0.90) [10]. A 2006 study [31] showed the usefulness of MRI double enhanced with superparamagnetic iron oxide and gadolinium in the diagnosis of advanced liver fibrosis. The sensitivity and specificity were greater than 90%. The protocol described, however, involved injection of two contrast media, which may be difficult to implement in the clinical setting. Other recently developed methods, such as sonographic transient elastography (FibroScan, Echosens) [32–34], perfusion-weighted MRI [35, 36], and MR elastography [33, 37], perform well in the prediction of advanced fibrosis and cirrhosis.
DWI is based on intravoxel incoherent motion and is used for noninvasive quantification of water diffusion and capillary and blood perfusion [11]. Several studies have shown that the ADC of cirrhotic liver is lower than that of normal liver [12–16], possibly because of the presence of a larger amount of connective tissue in the liver, narrowed sinusoids, and decreased blood flow [38]. There are limited data, however, on correlation between hepatic ADC and histologic stage of fibrosis. In only two studies [17, 18], to our knowledge, have investigators correlated hepatic ADC with stage of fibrosis. Boulanger et al. [17] used DWI at b values of 50–250 s/mm2 to examine 18 hepatitis C virus patients and 10 control subjects. Those investigators found no significant difference between the hepatitis C virus patients and the controls (ADC, 2.30 ± 1.28 and 1.79 ± 0.25 × 10–3 mm2/s, respectively). The ADCs of patients with hepatitis were even higher that those of controls. A potential explanation for the findings is that differences between fibrotic and nonfibrotic liver cannot be detected at small b values (< 300 s/mm2), which can increase the amount of perfusion contamination in ADC measurement [18, 39]. Using a low b value (128 s/mm2), Koinuma et al. [18] evaluated a large population of patients (n = 163), 31 of whom underwent liver biopsy. In agreement with our findings, their results showed a significant negative correlation between hepatic ADC and fibrosis score but no correlation between ADC and inflammation grade. We found a negative correlation between ADC and stage of fibrosis and a weaker negative correlation between ADC and grade of inflammation. The multiple b values (low and high) enabled more precise calculation of ADC with less perfusion contamination [18] and less regional ADC variation.
The diagnosis of stage 2 or greater fibrosis is clinically important because, owing to cost, risk of toxicity, and limited efficacy, only patients with stage 2 or greater fibrosis should receive antiviral treatment [2]. We found ADC to be a significant predictor of stage 2 or greater fibrosis, the AUC being 0.896 (for all b values combined). For comparison, an AUC of 0.79 has been reported with use of transient elastography (FibroScan) [32] and an AUC of 0.74–0.87 with use of serum markers (FibroTest) in the diagnosis of fibrosis in stages 2–4 [40, 41]. ADC can be used as an indication for and in surveillance of antiviral treatment. We believe these uses are important potential clinical applications of DWI, even if ADC cannot be used to differentiate individual stages of fibrosis. With an AUC of 0.896 (for all b values combined), ADC also can be used to predict the presence of advanced fibrosis and cirrhosis, which gives it prognostic value. With an AUC of 0.875, sensitivity of 85.7%, and specificity of 75% for an ADC (b = 700 s/mm2) less than 1.35 mm2/s, DWI also can be used to predict the presence of grade 1 or greater inflammation. Prediction of inflammation grade is important because this criterion has been correlated with the rate of progression to cirrhosis [42] and with response to antiviral treatment [43].
Our study had several limitations. First, because we are reporting our initial experience, the results were limited by the sample size, in particular the small number of patients with intermediate levels of hepatic fibrosis. Second, comparison of ADCs between groups stratified according to individual fibrosis stage did not show differences between all groups, and the correlation between fibrosis stage and ADC was only moderate. Third, biopsy was not performed on four patients who had an imaging diagnosis of cirrhosis.
Future work is needed to assess a larger number of patients and to correlate DWI findings with findings obtained with newer methods of perfusion MRI [35, 37] and MR elastography [33, 36] and with serologic markers of fibrosis.
In conclusion, our findings suggest that hepatic ADC measured with DWI can be used to quantify liver fibrosis when the b value is 500 s/mm2 or greater. DWI can be easily incorporated into a routine MRI protocol. It may be possible to use DWI findings to determine the indication for antiviral treatment and for follow-up of patients with chronic hepatitis.
Footnotes
Address correspondence to B. Taouli ([email protected]).
Funded by a Wylie J. Dodds research award from the Society of Gastrointestinal Radiologists (2004).
References
1.
Strader DB, Wright T, Thomas DL, Seeff LB. Diagnosis, management, and treatment of hepatitis C. Hepatology 2004; 39:1147 –1171
2.
Kim AI, Saab S. Treatment of hepatitis C. Am J Med 2005; 118:808 –815
3.
Gish RG. Current treatment and future directions in the management of chronic hepatitis B viral infection. Clin Liver Dis 2005; 9:541 –565
4.
Janes CH, Lindor KD. Outcome of patients hospitalized for complications after outpatient liver biopsy. Ann Intern Med 1993; 118:96 –98
5.
Froehlich F, Lamy O, Fried M, Gonvers JJ. Practice and complications of liver biopsy: results of a nationwide survey in Switzerland. Dig Dis Sci 1993; 38:1480 –1484
6.
Cadranel JF, Rufat P, Degos F. Practices of liver biopsy in France: results of a prospective nationwide survey. Hepatology 2000; 32:477 –481
7.
Maharaj B, Maharaj RJ, Leary WP, et al. Sampling variability and its influence on the diagnostic yield of percutaneous needle biopsy of the liver. Lancet 1986; 1:523–525
8.
Regev A, Berho M, Jeffers LJ, et al. Sampling error and intraobserver variation in liver biopsy in patients with chronic HCV infection. Am J Gastroenterol 2002; 97:2614 –2618
9.
Bedossa P, Dargere D, Paradis V. Sampling variability of liver fibrosis in chronic hepatitis C. Hepatology 2003; 38:1449 –1457
10.
Awaya H, Mitchell DG, Kamishima T, Holland G, Ito K, Matsumoto T. Cirrhosis: modified caudateright lobe ratio. Radiology 2002; 224:769 –774
11.
Le Bihan D. Diffusion/perfusion MR imaging of the brain: from structure to function. Radiology 1990; 177:328–329
12.
Namimoto T, Yamashita Y, Sumi S, Tang Y, Takahashi M. Focal liver masses: characterization with diffusion-weighted echo-planar MR imaging. Radiology 1997; 204:739–744
13.
Ichikawa T, Haradome H, Hachiya J, Nitatori T, Araki T. Diffusion-weighted MR imaging with a single-shot echoplanar sequence: detection and characterization of focal hepatic lesions. AJR 1998; 170:397 –402
14.
Amano Y, Kumazaki T, Ishihara M. Single-shot diffusion-weighted echo-planar imaging of normal and cirrhotic livers using a phased-array multicoil. Acta Radiol 1998; 39:440–442
15.
Kim T, Murakami T, Takahashi S, Hori M, Tsuda K, Nakamura H. Diffusion-weighted single-shot echoplanar MR imaging for liver disease. AJR 1999; 173:393 –398
16.
Taouli B, Vilgrain V, Dumont E, Daire JL, Fan B, Menu Y. Evaluation of liver diffusion isotropy and characterization of focal hepatic lesions with two single-shot echo-planar MR imaging sequences: prospective study in 66 patients. Radiology 2003; 226:71–78
17.
Boulanger Y, Amara M, Lepanto L, et al. Diffusion-weighted MR imaging of the liver of hepatitis C patients. NMR Biomed 2003; 16:132 –136
18.
Koinuma M, Ohashi I, Hanafusa K, Shibuya H. Apparent diffusion coefficient measurements with diffusion-weighted magnetic resonance imaging for evaluation of hepatic fibrosis. J Magn Reson Imaging 2005; 22:80 –85
19.
Murtz P, Flacke S, Traber F, van den Brink JS, Gieseke J, Schild HH. Abdomen: diffusion-weighted MR imaging with pulse-triggered single-shot sequences. Radiology 2002; 224:258–264
20.
Taouli B, Martin AJ, Qayyum A, et al. Parallel imaging and diffusion tensor imaging for diffusion-weighted MRI of the liver: preliminary experience in healthy volunteers. AJR 2004; 183:677 –680
21.
Ito K, Mitchell DG. Hepatic morphologic changes in cirrhosis: MR imaging findings. Abdom Imaging 2000; 25:456–461
22.
Ito K, Mitchell DG, Siegelman ES. Cirrhosis: MR imaging features. Magn Reson Imaging Clin N Am 2002; 10:75–92
23.
Batts KP, Ludwig J. Chronic hepatitis: an update on terminology and reporting. Am J Surg Pathol 1995; 19:1409 –1417
24.
Kim WR. The burden of hepatitis C in the United States. Hepatology 2002; 36:S30–S34
25.
Imbert-Bismut F, Ratziu V, Pieroni L, Charlotte F, Benhamou Y, Poynard T. Biochemical markers of liver fibrosis in patients with hepatitis C virus infection: a prospective study. Lancet 2001; 357:1069 –1075
26.
Bedossa P, Poynard T. An algorithm for the grading of activity in chronic hepatitis C. Hepatology 1996; 24:289–293
27.
Poynard T, McHutchison J, Manns M, Myers RP, Albrecht J. Biochemical surrogate markers of liver fibrosis and activity in a randomized trial of peginterferon alfa-2b and ribavirin. Hepatology 2003; 38:481–492
28.
Semelka RC, Chung JJ, Hussain SM, Marcos HB, Woosley JT. Chronic hepatitis: correlation of early patchy and late linear enhancement patterns on gadolinium-enhanced MR images with histopathology—initial experience. J Magn Reson Imaging 2001; 13:385–391
29.
Ito K, Mitchell DG, Gabata T. Enlargement of hilar periportal space: a sign of early cirrhosis at MR imaging. J Magn Reson Imaging 2000; 11:136 –140
30.
Ito K, Mitchell DG, Kim MJ, Awaya H, Koike S, Matsunaga N. Right posterior hepatic notch sign: a simple diagnostic MR finding of cirrhosis. J Magn Reson Imaging 2003; 18:561–566
31.
Aguirre DA, Behling CA, Alpert E, Hassanein TI, Sirlin CB. Liver fibrosis: noninvasive diagnosis with double contrast material-enhanced MR imaging. Radiology 2006; 239:425–437
32.
Ziol M, Handra-Luca A, Kettaneh A, et al. Noninvasive assessment of liver fibrosis by measurement of stiffness in patients with chronic hepatitis C. Hepatology 2005; 41:48–54
33.
Rouviere O, Yin M, Dresner MA, et al. MR elastography of the liver: preliminary results. Radiology 2006; 240:440–448
34.
Castera L, Vergniol J, Foucher J, et al. Prospective comparison of transient elastography, FibroTest, APRI, and liver biopsy for the assessment of fibrosis in chronic hepatitis C. Gastroenterology 2005; 128:343 –350
35.
Annet L, Materne R, Danse E, Jamart J, Horsmans Y, Van Beers BE. Hepatic flow parameters measured with MR imaging and Doppler US: correlations with degree of cirrhosis and portal hypertension. Radiology 2003; 229:409–414
36.
Hagiwara M, Rusinek H, Lee VS, et al. Quantification of liver fibrosis in patients with chronic liver disease using 3D perfusion-weighted MR imaging: preliminary experience. Radiology 2007 (in press)
37.
Huwart L, Peeters F, Sinkus R, et al. Liver fibrosis: non-invasive assessment with MR elastography. NMR Biomed 2006; 19:173 –179
38.
Annet L, Peeters F, Abarca-Quinones J, Leclercq I, Moulin P, Van Beers BE. Assessment of diffusion-weighted MR imaging in liver fibrosis. J Magn Reson Imaging 2007; 25:122–128
39.
Yamada I, Aung W, Himeno Y, Nakagawa T, Shibuya H. Diffusion coefficients in abdominal organs and hepatic lesions: evaluation with intravoxel incoherent motion echo-planar MR imaging. Radiology 1999; 210:617–623
40.
Poynard T, Imbert-Bismut F, Munteanu M, et al. Overview of the diagnostic value of biochemical markers of liver fibrosis (FibroTest, HCV FibroSure) and necrosis (ActiTest) in patients with chronic hepatitis C. Comp Hepatol 2004; 3:8
41.
Bourliere M, Penaranda G, Renou C, et al. Validation and comparison of indexes for fibrosis and cirrhosis prediction in chronic hepatitis C patients: proposal for a pragmatic approach classification without liver biopsies. J Viral Hepat 2006; 13:659–670
42.
Guido M, Fagiuoli S, Tessari G, et al. Histology predicts cirrhotic evolution of post transplant hepatitis C. Gut 2002; 50:697 –700
43.
Kumagai N, Kuramochi S, Toda K, et al. Assessment of histological features and outcome of interferon therapy in chronic hepatitis C. J Gastroenterol 1996; 31:69–74
Information & Authors
Information
Published In
Copyright
© American Roentgen Ray Society.
History
Submitted: February 19, 2007
Accepted: May 20, 2007
First published: November 23, 2012
Keywords
Authors
Metrics & Citations
Metrics
Citations
Export Citations
To download the citation to this article, select your reference manager software.