Radiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online before print April 26, 2007, 10.1148/radiol.2433060433
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Figure E1
Right arrow All Versions of this Article:
2433060433v1
243/3/727    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barash, H.
Right arrow Articles by Abramovitch, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Barash, H.
Right arrow Articles by Abramovitch, R.
(Radiology 2007;243:727-735.)
© RSNA, 2007

Experimental Studies

Functional MR Imaging during Hypercapnia and Hyperoxia: Noninvasive Tool for Monitoring Changes in Liver Perfusion and Hemodynamics in a Rat Model1

Hila Barash, MSc, Eitan Gross, MD, Idit Matot, MD, Yifat Edrei, BSc, Galia Tsarfaty, MD, Gadi Spira, PhD, Israel Vlodavsky, PhD, Eithan Galun, MD, and Rinat Abramovitch, PhD

1 From the Goldyne Savad Institute for Gene Therapy (H.B., Y.E., E. Galun, R.A.), MRI/MRS Lab HBRC (H.B., Y.E., R.A.), Department of Pediatric Surgery (E. Gross), and Department of Anesthesiology and Critical Care Medicine (I.M.), Hadassah Hebrew University Medical Center, POB 12000, Jerusalem 91120, Israel; Department of Radiology, Sheba Medical Center, Tel Hashomer, Israel (G.T.); and Department of Anatomy and Cell Biology (G.S.) and Cancer and Vascular Biology Research Center (I.V.), Bruce Rappaport Faculty of Medicine, Technion, Haifa, Israel. Received March 9, 2006; revision requested May 8; revision received June 27; accepted July 21; final version accepted October 2. Supported in part by Philip Morris USA and Philip Morris International (R.A.), by the Yeshya Horowitz Association through the Center for Complexity Science (R.A., H.B.), by the Belfer Foundation (R.A.), by a grant from the Israeli Ministry of Science (E. Galun), and by the Yeshya Horowitz Foundation. Address correspondence to R.A. (e-mail: rinat{at}hadassah.org.il).


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Purpose: To prospectively assess functional magnetic resonance (MR) imaging during hypercapnia and hyperoxia for monitoring changes in liver perfusion and hemodynamics in rats.

Materials and Methods: All experiments were performed with approval of an animal care and use committee. Functional T2*-weighted gradient-echo MR images of the rat liver were acquired during hyperoxia and graded hypercapnia (n = 24). Additional images were acquired during portal vein ligation (n = 4), induced hypovolemia (n = 5), and 70% hepatectomy (n = 5). Hypercapnic effects were confirmed with Doppler ultrasonography and with gadopentetate dimeglumine. Differences between groups were analyzed by using Wilcoxon rank sum test, except for the graded hypercapnia, for which one-way analysis of variance was used.

Results: Liver signal intensity (SI) increased due to hyperoxia; the percentage change in SI was seven times greater than that in muscle tissue; this reflects higher vascularity of the liver. Liver SI decreased due to hypercapnia; the percentage change in SI was negative in the liver but positive in the muscle (P < .001). Induced hypovolemia resulted in considerable decreases in functional MR imaging response; this reflects lower liver perfusion. Clinical applicability of the functional MR imaging method was proved by monitoring changes in liver perfusion that resulted from liver resection.

Conclusion: In the liver, the magnitude of the percentage change in SI induced by hypercapnia and hyperoxia reflects changes in total blood volume; whereas percentage change in SI values induced by hypercapnia from a negative to a positive value reflects relative changes in portal-to-arterial blood flow ratio.

Supplemental material: http://radiology.rsnajnls.org/cgi/content/full/2433060433/DC1

© RSNA, 2007


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Advanced therapeutic options for liver diseases emphasize the demand for improved diagnostic methods. Existing diagnostic imaging techniques provide limited evaluation of tissue characteristics beyond morphology. Perfusion imaging of the liver has the potential to improve this shortcoming (1). The utility of hepatic perfusion characterization relies on the resolution of each component of the liver's dual blood supply—the portal vein and the hepatic artery—because contributions from each are altered in many liver diseases (1). Changes in relative arterial and portal venous blood flow are known to be associated with the progression of cirrhosis (2,3). In primary and metastatic liver malignancies, there is a relative increase in arterial blood supply to the tumor (4). In addition, physiologic changes, such as regeneration, are also associated with hemodynamic changes (5–7). Because most pathologic conditions of the liver affect blood flow regionally and/or globally, perfusion imaging of the liver has been suggested to improve the sensitivity and specificity of liver diagnostics (1).

Most noninvasive imaging methods for evaluation of hepatic perfusion require administration of an intravenous contrast agent. To date, liver flow scintigraphy and flow quantification performed by using Doppler ultrasonography (US) have been used for characterization of global abnormalities, whereas computed tomography (CT) and magnetic resonance (MR) imaging can provide regional and global parameters (1). While current nuclear medicine techniques, including H2O15 positron emission tomography and single photon emission computed tomography with several agents, can provide measures of blood flow, the techniques result in limited spatial and temporal resolution, and uptake of some agents may be altered in tumors. Dynamic CT has been proposed for quantification of hepatic perfusion at first-pass analysis with extravascular contrast agents (8); however, disadvantages of CT include the required use of iodinated contrast material and the radiation doses associated with dynamic CT imaging (9). Perfusion MR imaging permits sequential whole-liver imaging without radiation. However, techniques for improved spatial and temporal resolution and accurate contrast material quantification still need to be developed, and the separation of blood flow from permeability effects remains challenging. Doppler US and cine phase-contrast velocity mapping with MR imaging are noninvasive methods for acquisition of hepatic perfusion measurements that do not involve administration of an intravenous contrast agent. However, the former is influenced by observer variation (10,11) and angle of insonation and has anatomic limitations (9,12), while the latter is limited to a specific section and blood vessel (13).

Functional MR imaging is based on changes in proton signals from tissue that is adjacent to blood vessels that contain paramagnetic deoxyhemoglobin, hence the term blood oxygen level–dependent (BOLD) contrast (14). Changes in deoxyhemoglobin tissue levels cause local magnetic field susceptibility gradients, which can be depicted on T2*-weighted MR images as signal intensity (SI) changes. These changes are related to changes in oxygen saturation, blood flow, and blood volume (15). Although the majority of functional MR imaging studies have focused on brain activation measurements, BOLD contrast can also be applied to hemodynamic processes in other organs such as the liver (16–18).

It has been previously shown that hyperoxia, induced by elevated fraction of inspired O2 concentration (FIO2), and hypercapnia, induced by elevated fraction of CO2 concentration (FICO2), can be used for the analysis of vascular architecture and functionality and for the assessment of vessel maturation in brain and subcutaneous tumors (19–22). Thus, the purpose of our study was to prospectively assess functional MR imaging during hypercapnia and hyperoxia for monitoring changes in liver perfusion and hemodynamics in rats.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Animals
Adult male Sprague-Dawley (Harlan; Jerusalem, Israel) rats (n = 44; weight range, 250–300 g) were used for all animal experiments. All experiments were performed in accordance with the guidelines and approval of the animal care and use committee of the Hebrew University. (The animal care unit holds the U.S. National Institutes of Health approval number OPRR-A01-5011.)

MR Imaging Analysis Technique
MR imaging examinations were performed (H.B.) in anesthetized rats (n = 39) (30 mg pentobarbital [Sigma, Rehovot, Israel]) per kilogram of body weight, intraperitoneally) with a horizontal 4.7-T spectrometer (Biospec; Bruker Medical, Ettlingen, Germany) by using a birdcage coil. Coronal and transverse T1-weighted spin-echo images of the whole liver were acquired for alignment (repetition time msec/echo time msec, 370/18; field of view, 5.8 cm; section thickness, 1.5 mm). Hepatic perfusion and hemodynamics were evaluated on T2*-weighted gradient-echo images (147/10; flip angle, 30°; field of view, 5.8 cm; pixels, 256 x 128; in-plane resolution, 230 µm; five sections; section thickness, 1.5 mm; spectral width, 25 000 Hz; number of signals acquired, four; 65 seconds per image) acquired during breathing of air, an air-CO2 mixture (95% air and 5% CO2), and carbogen (95% oxygen and 5% CO2) as previously described (20,23,24). Five images were acquired at each gas mixture. Zero filling of k-space data was applied, resulting in a matrix of 256 x 256.

US Analysis Technique
Color velocity and spectral Doppler US measurements were acquired (G.T., with 10 years of experience in color and spectral Doppler US) with a 14-MHz linear transducer (15L8s) (Sequoia 512; Acuson, Mountain View, Calif). Rats (n = 5) were anesthetized (H.B., R.A.) (30 mg/kg pentobarbital, intraperitoneally). The upper abdomen was shaved, and imaging gel was applied. Changes in hepatic perfusion were assessed from US measurements of portal vein velocity and hepatic artery maximal velocity. The transducer was fixed on the hepatic artery or on the portal vein during the examination. The threshold level for color display was fixed for all the measurements at a level slightly above that at which the color noise disappeared. Calculation of spectral blood flow parameters (maximum, minimum, and mean velocity) was done automatically. Three measurements (during a period of 4 minutes) were acquired during breathing of air as a baseline, followed by 10 measurements (one measurement every 30 seconds during a period of 5 minutes) that were acquired during breathing of an air-CO2 mixture (95% air and 5% CO2), equivalent to the 5% FICO2 time period in the MR imaging experiment.

Experimental Design
Control group: experiment 1.—After anesthesia was induced, animals (n = 6) were placed inside the MR imager while spontaneously breathing room air for 30 minutes. Animals were then exposed to gas mixtures administered through a mask as follows: 5 minutes of air, followed by 5 minutes of air-CO2 (95% air and 5% CO2), and then 5 minutes of carbogen (95% oxygen and 5% CO2). MR images were acquired during the whole process (Fig 1).


Figure 1
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1: Temporal description of functional MR imaging experiments. Number of animals in each experiment (Exp.) is designated. Composition of gas breathed is: air (A), 95% air and 5% CO2 (5C), 90% air and 10% CO2 (10C), 85% air and 15% CO2 (15C), or 95% oxygen and 5% CO2 (O). bullet = blood samples withdrawn, {circ} = hypovolemia, @ = injection of gadopentetate dimeglumine, Cn = centered phase encoding, Ln = linear phase encoding, X = portal vein ligation.

 
Hypercapnic effects on MR imaging SI: experiments 2–3.—To evaluate the effect of hypercapnia on SI in functional MR images, animals received 10% FICO2 (90% air and 10% CO2) and 15% FICO2 (85% air and 15% CO2) (n = 4 per group). All other conditions were identical to that in experiment 1 (Fig 1).

Hypercapnic effects on partial pressure of CO2: experiments 1–3.—Blood samples (0.3 mL) were withdrawn (R.A.) from the animals for measurement of partial pressure of CO2 (PCO2) (I.M.) at two time points: (a) immediately after 5 minutes of air inhalation (approximately 35 minutes after anesthesia was induced) and (b) after an additional 5 minutes of either inhalation of 5% FICO2, 10% FICO2, or 15% FICO2 (n = 6, 4, and 4, respectively, from the same animals that were imaged) (Fig 1).

Effect of portal vein ligation on functional MR imaging SI: experiment 4.—A midline laparotomy was performed (E. Gross, H.B.) with the use of anesthesia (n = 4). With the bowel retracted, the portal vein was identified and totally ligated by using a 3-0 silk thread. Immediately after abdominal closure, the rats were imaged while breathing air, air-CO2 (95% air and 5% CO2), and carbogen (95% oxygen and 5% CO2) (Fig 1).

Source of Hypercapnic Effects on Functional MR Imaging SI
To further clarify the source of signal changes in the liver during hypercapnia, we isolated each of these effectors in the following experiments.

Gadopentetate dimeglumine: experiment 5.—We (R.A., H.B.) used gadopentetate dimeglumine to eliminate the effects of flow and deoxyhemoglobin-oxyhemoglobin ratio on SI. For a limited period of time, this contrast agent maintains steady-state levels that enable the assessment of changes in blood volume (25). A 25-gauge catheter was positioned in the tail vein (n = 5). Gadopentetate dimeglumine (0.5 mmol/kg, Magnetol; Soreq Radiopharmaceuticals, Ness Ziona, Israel) was injected slowly before imaging with inhalation of gas mixtures. T1-weighted gradient-echo images were acquired during breathing of air, air-CO2, and air (95% air and 5% CO2) alternately. Ten images (discarding the first) were acquired at each gas mixture (23 seconds per image; section thickness, 1.5 mm; 93.5/4.6; field of view, 5.8 cm; pixels, 256 x 128; in-plane resolution, 230 µm; number of signals acquired, two) for a total of 11.5 minutes (Fig 1).

Linear and centered phase encoding: experiment 6.—At linear phase encoding, the center of k-space is acquired during steady-state saturation and SI is sensitive to changes in flow. At centered phase encoding, the center of k-space is acquired when the magnetization is fully relaxed, resulting in a lower sensitivity to flow (24). Gradient-echo images were acquired (R.A., H.B.) during enrichment with air, air-CO2 (95% air and 5% CO2), and carbogen (95% oxygen and 5% CO2) by using two orders of k-space acquisitions (linear and centered phase encoding, n = 6). Images were acquired with a 5-second inter-image delay to allow full recovery of the magnetization between frames (Fig 1).

Effects of hypovolemia on functional MR imaging SI: experiment 7.—A 25-gauge catheter was washed with heparin and positioned (R.A., H.B.) in the tail vein. After imaging with inhalation of gas mixtures was performed, 3 mL of blood (15% of total blood volume) was withdrawn from the rats (n = 5) while they were inside the MR imager, and imaging was performed again. Blood hematocrit levels were measured (I.M.) before and after induction of hypovolemia.

Effect of partial hepatectomy.—Resection of 70% of the total liver mass was performed (G.S.), according to the method of Higgins and Anderson (26) with the use of light anesthesia (diethyl ether) by removing the median and left lateral lobes (n = 5). Rat livers were imaged (H.B.) while the rats were breathing air, air-CO2 (95% air and 5% CO2), and carbogen (95% oxygen and 5% CO2) before and 2 days after hepatectomy.

Statistical Analysis
MR imaging data were analyzed (H.B.) on a personal computer with software written in-house (Y.E.) by using Interactive Data Language programming language (Research Systems, Boulder, Colo). Maps of mean SI (S) values for each pixel during inhalation of gases (Sair, SCO2, and SO2) were calculated from four images for each gas (discarding the repeats obtained during gas changes [65 seconds]) (Figure E1, http://radiology.rsnajnls.org/cgi/content/full/2433060433/DC1). It has been previously shown that there is a rapid response to changes in the inhaled gas and a delay of 1 minute is sufficient for the functional MR imaging signal to reach stability (23). The percentage change in SI induced by hypercapnia ({Delta}SCO2) and that induced by hyperoxia ({Delta}SO2) were calculated (for each pixel greater than or equal to the noise threshold) by using the following equations:

Formula
and

Formula

Results are expressed as means ± standard deviations. A region of interest was defined by using anatomic images and included the whole liver (approximately 8000 pixels) or back muscle (approximately 600 pixels) and excluded large blood vessels and flow artifacts. Mean values were calculated from these regions of interest in n animals as indicated, and three sections per rat were used. Reproducibility of {Delta}S values was verified with repeated sampling (five regions of interest, three times each). Differences between groups of continuous data were compared with a one-sided exact Wilcoxon rank sum test, except for graded hypercapnia, for which we used one-way analysis of variance. Statistical analyses were performed with statistical software (SAS, version 8; SAS Institute, Cary, NC). P values less than .05 were considered to indicate a significant difference.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 
Hyperoxia and Hypercapnia in Liver and Muscle
Inhalation of carbogen resulted in enhanced SI on functional MR images. Mean {Delta}SO2 values in the liver were seven times greater than those in adjacent muscle (22.3% ± 8.9 vs 3.2% ± 6.2, n = 16, P < .001) (Fig 2). In contrast, air-CO2 inhalation caused a decrease of SI on functional MR images, which generated negative {Delta}SCO2 values in the liver (–19.5% ± 9.7), as opposed to positive values recorded in adjacent muscle tissue (0.07% ± 3.7, P < .001) (Fig 2).


Figure 2A
View larger version (97K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2a: Effect of hypercapnia and hyperoxia on SI at liver functional MR imaging. Rats were exposed to gas mixtures: air, air–5% CO2, and carbogen. Hepatic hemodynamics were determined on T2*-weighted gradient-echo images (147/10). Representative functional MR imaging maps of rat liver show percentage SI change due to (a) hyperoxia and (b) hypercapnia (values as indicated in color bar; bar = 1 cm). (c, d) Graphs show (c) mean {Delta}SO2 and (d) {Delta}SCO2 values ± standard deviations (error bars) in liver versus those in adjacent back muscle (n = 16, including baseline values of animals in partial hepatectomy and phase-encoding experiments; * = P < .001).

 

Figure 2B
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2b: Effect of hypercapnia and hyperoxia on SI at liver functional MR imaging. Rats were exposed to gas mixtures: air, air–5% CO2, and carbogen. Hepatic hemodynamics were determined on T2*-weighted gradient-echo images (147/10). Representative functional MR imaging maps of rat liver show percentage SI change due to (a) hyperoxia and (b) hypercapnia (values as indicated in color bar; bar = 1 cm). (c, d) Graphs show (c) mean {Delta}SO2 and (d) {Delta}SCO2 values ± standard deviations (error bars) in liver versus those in adjacent back muscle (n = 16, including baseline values of animals in partial hepatectomy and phase-encoding experiments; * = P < .001).

 

Figure 2C
View larger version (15K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2c: Effect of hypercapnia and hyperoxia on SI at liver functional MR imaging. Rats were exposed to gas mixtures: air, air–5% CO2, and carbogen. Hepatic hemodynamics were determined on T2*-weighted gradient-echo images (147/10). Representative functional MR imaging maps of rat liver show percentage SI change due to (a) hyperoxia and (b) hypercapnia (values as indicated in color bar; bar = 1 cm). (c, d) Graphs show (c) mean {Delta}SO2 and (d) {Delta}SCO2 values ± standard deviations (error bars) in liver versus those in adjacent back muscle (n = 16, including baseline values of animals in partial hepatectomy and phase-encoding experiments; * = P < .001).

 

Figure 2D
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2d: Effect of hypercapnia and hyperoxia on SI at liver functional MR imaging. Rats were exposed to gas mixtures: air, air–5% CO2, and carbogen. Hepatic hemodynamics were determined on T2*-weighted gradient-echo images (147/10). Representative functional MR imaging maps of rat liver show percentage SI change due to (a) hyperoxia and (b) hypercapnia (values as indicated in color bar; bar = 1 cm). (c, d) Graphs show (c) mean {Delta}SO2 and (d) {Delta}SCO2 values ± standard deviations (error bars) in liver versus those in adjacent back muscle (n = 16, including baseline values of animals in partial hepatectomy and phase-encoding experiments; * = P < .001).

 
Hypercapnia Effects
Inhalation for 5 minutes of increasing concentrations of FICO2 (5%, 10%, or 15%) increased blood PCO2 levels by 10.4%, 29%, and 85%, respectively, compared with the levels during air inhalation (Table) (P < .01). The PCO2 level during 15% FICO2 inhalation was also significantly greater than the level during 10% FICO2 and 5% FICO2 (P < .01). Five-minute inhalation of increasing concentrations of FICO2 (10% and 15%) increased {Delta}SCO2 values by 75% and 160%, respectively (Table), compared with the levels during 5% FICO2 inhalation (P < .01). Relative changes in PCO2 and in {Delta}SCO2 highly correlated (R2 = 0.979) (Fig 3); that is, with an increase in blood PCO2 levels, {Delta}SCO2 values decreased.


View this table:
[in this window]
[in a new window]

  		Mean PCO2 and {Delta}SCO2 Values Induced with FICO2 Concentrations

 

Figure 3
View larger version (6K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3: Graph shows effect of hypercapnia on SI at MR imaging. Correlation (trend line) between change in blood PCO2 level and {Delta}SCO2 was deduced from animals that received either 5%, 10%, or 15% FICO2. Mean blood PCO2 levels and {Delta}SCO2 values ± standard deviations were measured before and during exposure to FICO2 concentrations. (R2 = 0.984, n ≥ 4 for all measurements).

 
Effect of Portal Vein Ligation
Total portal vein ligation caused enhancement of SI during hypercapnia, which resulted in positive {Delta}SCO2 values (1.1% ± 8.7, P < .01) that were indistinguishable from {Delta}SCO2 values of normal muscle (Fig 4). In addition, mean {Delta}SO2 values of the liver in rats that underwent portal vein ligation decreased dramatically (2.9% ± 5.4, P < .01) compared with those in animals that did not undergo portal vein ligation (22.4% ± 5.7), because of reduction in blood content of the liver (Fig 4).


Figure 4A
View larger version (11K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4a: Graphs illustrate the effects of portal ligation and hypovolemia on functional MR imaging SI during hyperoxia and hypercapnia. Graphs show comparison of mean (a) {Delta}SCO2 and (b) {Delta}SO2 values ± standard deviations (error bars) (n liver) of liver and muscle (n muscle) between normal rats (n = 15), rats that underwent portal vein ligation (pl liver) (n = 4), and rats during hypovolemic shock (hypo liver) (n = 5). Hypovolemic shock was induced by drawing 3 mL of blood (15% of blood volume) (* = P < .01).

 

Figure 4B
View larger version (8K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4b: Graphs illustrate the effects of portal ligation and hypovolemia on functional MR imaging SI during hyperoxia and hypercapnia. Graphs show comparison of mean (a) {Delta}SCO2 and (b) {Delta}SO2 values ± standard deviations (error bars) (n liver) of liver and muscle (n muscle) between normal rats (n = 15), rats that underwent portal vein ligation (pl liver) (n = 4), and rats during hypovolemic shock (hypo liver) (n = 5). Hypovolemic shock was induced by drawing 3 mL of blood (15% of blood volume) (* = P < .01).

 
Gadopentetate Dimeglumine and Phase-encoding Experiments
During steady-state levels of gadopentetate dimeglumine, 5% FICO2 resulted in induced SI elevation in blood vessels, whereas SI of liver parenchyma remained stable (Fig 5). With linear phase encoding, SI was sensitive to changes in flow, while with centered phase encoding, there was a lower sensitivity to flow (24). The liver response to hyperoxia was similar with both linear and centered phase encoding, which indicated that hyperoxia did not affect liver blood flow. The response to hypercapnia was twofold lower at centered phase encoding than at linear phase encoding (–40% ± 11 and –20% ± 13, respectively).


Figure 5A
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5a: Eliminating effects of flow and blood oxygenation during hypercapnia with gadopentetate dimeglumine. Gadopentetate dimeglumine (0.5 mmol/kg) was injected slowly before inhalation of gas mixtures (n = 5). T1-weighted gradient-echo MR images (93.5/4.6) were acquired during breathing of air and air-CO2 (FICO2) alternately (as indicated). Ten images were acquired at each gas mixture. MR images show SI (a) in liver parenchyma (region marked by white triangle) and (b) in the rim of a large vessel in the liver (arrow). (c) Graph shows SI/108 in the vessel ({square}) was significantly higher (P < .01) during CO2 inhalation than in the liver parenchyma ([{diamondsuit}]). Bar = 1 cm.

 

Figure 5B
View larger version (125K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5b: Eliminating effects of flow and blood oxygenation during hypercapnia with gadopentetate dimeglumine. Gadopentetate dimeglumine (0.5 mmol/kg) was injected slowly before inhalation of gas mixtures (n = 5). T1-weighted gradient-echo MR images (93.5/4.6) were acquired during breathing of air and air-CO2 (FICO2) alternately (as indicated). Ten images were acquired at each gas mixture. MR images show SI (a) in liver parenchyma (region marked by white triangle) and (b) in the rim of a large vessel in the liver (arrow). (c) Graph shows SI/108 in the vessel ({square}) was significantly higher (P < .01) during CO2 inhalation than in the liver parenchyma ([{diamondsuit}]). Bar = 1 cm.

 

Figure 5C
View larger version (17K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5c: Eliminating effects of flow and blood oxygenation during hypercapnia with gadopentetate dimeglumine. Gadopentetate dimeglumine (0.5 mmol/kg) was injected slowly before inhalation of gas mixtures (n = 5). T1-weighted gradient-echo MR images (93.5/4.6) were acquired during breathing of air and air-CO2 (FICO2) alternately (as indicated). Ten images were acquired at each gas mixture. MR images show SI (a) in liver parenchyma (region marked by white triangle) and (b) in the rim of a large vessel in the liver (arrow). (c) Graph shows SI/108 in the vessel ({square}) was significantly higher (P < .01) during CO2 inhalation than in the liver parenchyma ([{diamondsuit}]). Bar = 1 cm.

 
Hypercapnic Effect on Portal Vein and Hepatic Artery Velocities
Hepatic artery or portal vein velocity values were measured by using Doppler US. During 4 minutes of air inhalation, these values were stable; however, inhalation of 5% FICO2 increased portal vein velocity by 13% ± 8 and decreased hepatic artery maximal velocity by 12% ± 6 compared with baseline values (n = 5).

Hypovolemic Effects on Functional MR Imaging Signal
Withdrawing 15% of blood volume (3 mL of blood) caused a 25% decrease in hematocrit levels. At that point, mean values of {Delta}SCO2 were significantly less negative (–3.6% ± 7) than during normovolemia (–19.5% ± 9.7, P < .01) (Fig 4). Subsequently, mean {Delta}SO2 values declined (5.4% ± 4.5) compared with values during normovolemia (P < .01) (Fig 4).

Partial Hepatectomy Effects on Liver Perfusion
Two days following 70% partial hepatectomy, mean {Delta}SO2 values declined significantly (before hepatectomy, 29.4% ± 6; following hepatectomy, 0.4% ± 6.5; P < .001; n = 5) (Fig 6); this mirrors a decrease in liver vascularity and blood content. Mean {Delta}SCO2 values increased and became positive (before hepatectomy, –21.3% ± 6; following hepatectomy, 11.2% ± 10; P < .01); this demonstrates a change in liver blood supply ratio.


Figure 6A
View larger version (71K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6a: Liver hemodynamic changes during regeneration at functional MR imaging. Liver resection (70%) was performed by removing median and left lateral lobes. MR images were acquired before and 2 days after hepatectomy (n = 5). (a) Top row: Before hepatectomy (Pre). Bottom row: Two days after hepatectomy. Left column: T1-weighted spin-echo (SE) anatomic MR images (370/18). Middle column: {Delta}SO2 MR maps. Right column: {Delta}SCO2 MR maps. Values as indicated in color bar; bar = 1 cm. (b) Graph shows mean {Delta}SO2 and {Delta}SCO2 values ± standard deviations (error bars) during regeneration. (* = P < .01 compared with values before hepatectomy).

 

Figure 6B
View larger version (16K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6b: Liver hemodynamic changes during regeneration at functional MR imaging. Liver resection (70%) was performed by removing median and left lateral lobes. MR images were acquired before and 2 days after hepatectomy (n = 5). (a) Top row: Before hepatectomy (Pre). Bottom row: Two days after hepatectomy. Left column: T1-weighted spin-echo (SE) anatomic MR images (370/18). Middle column: {Delta}SO2 MR maps. Right column: {Delta}SCO2 MR maps. Values as indicated in color bar; bar = 1 cm. (b) Graph shows mean {Delta}SO2 and {Delta}SCO2 values ± standard deviations (error bars) during regeneration. (* = P < .01 compared with values before hepatectomy).

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ADVANCES IN KNOWLEDGE
 References
 
While vascularized tissues, such as gray matter in the brain, show relatively little signal enhancement (4%) in response to 5% FICO2 (27), results of our study showed that increased CO2 concentration has an overwhelming effect (20%) on the SI of the whole liver. The liver is a hypervascularized organ and has a dual blood supply. In the normal liver, approximately three-fourths of the blood supply is derived from the portal vein, whereas only about one-fourth is derived from the hepatic artery (28,29). Thus, hepatic blood is more deoxygenated, extending the response range to changes in tissue oxygenation. The changes in SI at functional MR imaging due to changes in FIO2 and FICO2 result from alterations in blood volume, flow, and deoxyhemoglobin concentration (15).

Results of our study demonstrate that elevated PCO2 blood levels alter liver perfusion, with an increase in portal blood flow relative to hepatic artery flow. Because portal blood flow carries deoxygenated blood, {Delta}SCO2 values were negative. Thus, higher deoxyhemoglobin levels produce a decrease in functional MR imaging SI. As expected, increasing concentrations of FICO2 produced elevated PCO2 levels and decreasing {Delta}SCO2 values. Moreover, ligation of the portal vein resulted in positive {Delta}SCO2 values, because the only blood source was oxygenated blood from the hepatic artery. In addition, the arterial buffer response probably has some effect on the hepatic artery flow (30). These results were further confirmed by using Doppler US, which illustrated increase in portal flow and reduction in arterial blood flow during CO2 enrichment.

High liver blood content (15% of total blood volume) and hypervascularity explain the high {Delta}SO2 values induced by hyperoxia, compared with those of muscle. In the brain, the effects of hyperoxia are attenuated (31) because of the high basal oxyhemoglobin level. When we reduced the total blood volume by inducing hypovolemia or the liver blood volume by performing portal vein ligation, the mean {Delta}SO2 values in the liver decreased to levels similar to those observed in muscle. Nevertheless, during hypovolemia, because the flow ratio (portal vs arterial) did not change considerably, mean {Delta}SCO2 values remained negative. The decrease in blood content and the resultant physiologic response of the body to hypovolemia (eg, redistribution of blood to vital organs and reduced responsiveness of blood vessels, which are already dilated or vasoconstricted to their maximum level) explain the limited observed changes in functional MR imaging SI relative to that with normovolemia. These results also confirm our assumption that {Delta}SCO2 values become positive during portal vein ligation because of discontinuation of portal flow rather than resulting from the reduction in liver blood volume. Thus, in the liver, a change in the magnitude of the SI induced by using 5% FICO2 ({Delta}SCO2) reflects changes in liver blood volume, whereas a change in SI from a negative to a positive value reflects relative changes in portal-to-arterial blood flow ratio.

By using gadopentetate dimeglumine, we followed changes associated with increased FICO2 concentration. For a limited period of time, this contrast agent eliminated the effect of blood flow and enabled the assessment of changes in blood volume (25). Increased SI induced by using gadopentetate dimeglumine appeared only in blood vessels throughout CO2 inhalation; this reflects vasodilation. Therefore, the observed changes in SI in the parenchyma during CO2 inhalation did not originate from changes in vascular tone but from changes in flow. By changing the order of phase-encoding acquisition, we can eliminate the effect of flow on functional MR imaging SI. We did not observe changes in {Delta}SO2 during hyperoxia between the two phase-encoding methods (sensitive and insensitive to flow); however, {Delta}SCO2 values were more negative with centered phase encoding. These data suggest that increased FIO2 concentration has no effect on liver blood flow, whereas CO2 enrichment causes increases in portal flow. These results were confirmed by using Doppler US.

We further demonstrated the ability of this method to noninvasively follow physiologic changes in liver perfusion. During the first 3 days following liver resection, there is a decrease in sinusoidal density (5) and thus a decline in liver perfusion and blood content. The lower magnitude of both {Delta}SO2 and {Delta}SCO2 during the early phase of regeneration demonstrated this reduction in liver perfusion. In addition, following partial hepatectomy, there was elevated blood flow, along with an increase in arterial versus portal perfusion due to higher O2 consumption (6,7). The switch of {Delta}SCO2 values from negative to positive enabled us to detect these changes in the portal-to-arterial blood flow ratio.

Our study had limitations because all of our experiments were performed in rats. Use of this method in humans requires some adjustments. In rodents, we have found consistent {Delta}SCO2 and {Delta}SO2 baseline values that were different between rats and mice (32,33). Thus, it is essential to establish baseline values for healthy humans and to assess the variability between ages and sexes. Additionally, in humans, there will be a need to replace the gradient-echo sequence with echo-planar MR imaging, which has previously been demonstrated to be feasible in the liver (34) and would shorten the additional time of the routine examination. The total blood volume is also higher in humans than in rats, and the percentage change in the liver is substantially higher than in the brain (27). Thus, these changes could be easily observed with 1.5-T imagers that are currently acceptable for humans. Finally, our study was performed in anesthetized rats, and anesthesia, which is not required for adult humans for functional MR imaging, is known to effect regional blood flow. BOLD MR imaging can be easily included at routine liver MR imaging examinations, and inhalation of 5% CO2 has been shown previously to be nontoxic to humans (35,36). The proposed functional MR imaging method does not enable measurement of arterial and portal blood flow values but rather reflects changes in the ratio between them.

In conclusion, we found that with the use of hypercapnia and hyperoxia, functional MR imaging offers a noninvasive tool that reflects changes in intrahepatic hemodynamics in a way that might facilitate evaluation of pathologic conditions of the liver. In the liver, the changes in the magnitude of both {Delta}SCO2 and {Delta}SO2 reflect changes in liver blood volume, whereas changes in {Delta}SCO2 values from negative to positive reflect relative changes in portal-to-arterial blood flow ratio.

Practical applications: Functional MR imaging with hyperoxia and hypercapnia may prove to be an additional noninvasive tool for monitoring perfusion and hemodynamic changes in the liver. By using this method, we showed in rats its applicability for evaluating liver perfusion and hemodynamics in two clinical settings: acute bleeding and liver resection. Further studies are needed to evaluate the feasibility of this method in characterizing pathologic conditions of the liver, such as cirrhosis, hepatic cancer, and liver metastasis.


   ADVANCES IN KNOWLEDGE
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
References
 


   ACKNOWLEDGMENTS
 
We thank Mery Clausen, BA, for assisting in the editing of the manuscript.


   FOOTNOTES
 

Abbreviations: BOLD = blood oxygenation level dependent • FICO2 = fraction of inspired CO2 • FIO2 = fraction of inspired O2 • PCO2 = partial pressure of CO2 • SI = signal intensity

Authors stated no financial relationship to disclose.

Author contributions: Guarantor of integrity of entire study, R.A.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, H.B., I.M., E. Galun, R.A.; experimental studies, H.B., E. Gross, I.M., Y.E., G.T., G.S., E. Galun, R.A.; statistical analysis, H.B., Y.E., R.A.; and manuscript editing, all authors


   References
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ADVANCES IN KNOWLEDGE
 References
 

  1. Pandharipande PV, Krinsky GA, Rusinek H, Lee VS. Perfusion imaging of the liver: current challenges and future goals. Radiology 2005;234:661–673.[Abstract/Free Full Text]
  2. Zoli M, Magalotti D, Bianchi G, et al. Functional hepatic flow and Doppler-assessed total hepatic flow in control subjects and in patients with cirrhosis. J Hepatol 1995;23:129–134.[CrossRef][Medline]
  3. Blendis L, Wong F. The hyperdynamic circulation in cirrhosis: an overview. Pharmacol Ther 2001;89:221–231.[CrossRef][Medline]
  4. Breedis C, Young G. The blood supply of neoplasms in the liver. Am J Pathol 1954;30:969–977.[Medline]
  5. Michalopoulos GK, DeFrances MC. Liver regeneration. Science 1997;276:60–66.[Abstract/Free Full Text]
  6. Yoshioka S, Miyazaki M, Shimizu H, et al. Hepatic venous hemoglobin oxygen saturation predicts regenerative status of remnant liver after partial hepatectomy in rats. Hepatology 1998;27:1349–1353.[CrossRef][Medline]
  7. Shimizu H, Miyazaki M, Yoshioka S, et al. Changes in hepatic venous oxygen saturation related to the extent of regeneration after partial hepatectomy in rats. Am J Surg 1999;178:428–431.[CrossRef][Medline]
  8. Miles KA, Hayball M, Dixon AK. Colour perfusion imaging: a new application of computed tomography. Lancet 1991;337:643–645.[Medline]
  9. Reddy SI, Grace ND. Liver imaging: a hepatologist's perspective. Clin Liver Dis 2002;6:297–310, ix.[CrossRef][Medline]
  10. Rappaport AM, Wanless IR. Physioanatomic considerations. In: Schiff L, Schiff ER, eds. Diseases of the liver. 7th ed. Philadelphia, Pa: Lippincott, 1993; 1–41.
  11. Cosgrove D, Eckersley R, Blomley M, Harvey C. Quantification of blood flow. Eur Radiol 2001;11:1338–1344.[CrossRef][Medline]
  12. Aube C, Oberti F, Korali N, et al. Ultrasonographic diagnosis of hepatic fibrosis or cirrhosis. J Hepatol 1999;30:472–478.[CrossRef][Medline]
  13. Nayler GL, Firmin DN, Longmore DB. Blood flow imaging by cine magnetic resonance. J Comput Assist Tomogr 1986;10:715–722.[Medline]
  14. Ogawa S, Lee TM, Kay AR, Tank DW. Brain magnetic resonance imaging with contrast dependent on blood oxygenation. Proc Natl Acad Sci U S A 1990;87:9868–9872.[Abstract/Free Full Text]
  15. van Zijl PC, Eleff SM, Ulatowski JA, et al. Quantitative assessment of blood flow, blood volume and blood oxygenation effects in functional magnetic resonance imaging. Nat Med 1998;4:159–167.[CrossRef][Medline]
  16. Foley LM, Picot P, Thompson RT, Yau MJ, Brauer M. In vivo monitoring of hepatic oxygenation changes in chronically ethanol-treated rats by functional magnetic resonance imaging. Magn Reson Med 2003;50:976–983.[CrossRef][Medline]
  17. Semple SI, Wallis F, Haggarty P, et al. The measurement of fetal liver T2* in utero before and after maternal oxygen breathing: progress towards a non-invasive measurement of fetal oxygenation and placental function. Magn Reson Imaging 2001;19:921–928.[CrossRef][Medline]
  18. Dor Y, Djonov V, Abramovitch R, et al. Conditional switching of VEGF provides new insights into adult neovascularization and pro-angiogenic therapy. Embo J 2002;21:1939–1947.[CrossRef][Medline]
  19. Abramovitch R, Itzik A, Harel H, Nagler A, Vlodavsky I, Siegal T. Halofuginone inhibits angiogenesis and growth in implanted metastatic rat brain tumor model: an MRI study. Neoplasia 2004;6:480–489.[CrossRef][Medline]
  20. Abramovitch R, Dafni H, Smouha E, Benjamin LE, Neeman M. In vivo prediction of vascular susceptibility to endothelial growth factor withdrawal: magnetic resonance imaging of C6 rat glioma in nude mice. Cancer Res 1999;59:5012–5016.[Abstract/Free Full Text]
  21. Rauscher A, Sedlacik J, Barth M, Haacke EM, Reichenbach JR. Nonnvasive assessment of vascular architecture and function during modulated blood oxygenation using susceptibility weighted magnetic resonance imaging. Magn Reson Med 2005;54:87–95.[CrossRef][Medline]
  22. Robinson SP, Rijken PF, Howe FA, et al. Tumor vascular architecture and function evaluated by non-invasive susceptibility MRI methods and immunohistochemistry. J Magn Reson Imaging 2003;17:445–454.[CrossRef][Medline]
  23. Abramovitch R, Frenkiel D, Neeman M. Analysis of subcutaneous angiogenesis by gradient echo magnetic resonance imaging. Magn Reson Med 1998;39:813–824.[Medline]
  24. Neeman M, Dafni H, Bukhari O, Braun RD, Dewhirst MW. In vivo BOLD contrast MRI mapping of subcutaneous vascular function and maturation: validation by intravital microscopy. Magn Reson Med 2001;45:887–898.[CrossRef][Medline]
  25. Pastor CM, Planchamp C, Pochon S, et al. Kinetics of gadobenate dimeglumine in isolated perfused rat liver: MR imaging evaluation. Radiology 2003;229:119–125.[Abstract/Free Full Text]
  26. Higgins GM, Anderson RM. Restoration of the liver of the white rat following partial surgical removal. Arch Pathol 1931;12:186–202.
  27. Vesely A, Sasano H, Volgyesi G, et al. MRI mapping of cerebrovascular reactivity using square wave changes in end-tidal PCO2. Magn Reson Med 2001;45:1011–1013.[CrossRef][Medline]
  28. Schenk WG Jr, McDonald JC, McDonald K, Drapanas T. Direct measurement of hepatic blood flow in surgical patients: with related observations on hepatic flow dynamics in experimental animals. Ann Surg 1962;156:463–471.[Medline]
  29. Chiandussi L, Greco F, Sardi G, Vaccarino A, Ferraris CM, Curti B. Estimation of hepatic arterial and portal venous blood flow by direct catheterization of the vena porta through the umbilical cord in man: preliminary results. Acta Hepatosplenol 1968;15:166–171.[Medline]
  30. Lautt WW. Mechanism and role of intrinsic regulation of hepatic arterial blood flow: hepatic arterial buffer response. Am J Physiol Gastrointest Liver Physiol 1985;249(5 pt 1):G549–G556.[Abstract/Free Full Text]
  31. Losert C, Peller M, Schneider P, Reiser M. Oxygen-enhanced MRI of the brain. Magn Reson Med 2002;48:271–277.[CrossRef][Medline]
  32. Edrei Y, Gross E, Pikarsky E, et al. Characterization and early detection of liver matastasis by functional MRI. Hepatology 2006;44(4)(suppl 1):490A-491A.
  33. Ben-Moshe T, Barash H, Kang TB, et al. Role of caspase 8 in hepatocyte response to infection and injury in mice. Hepatology (in press).
  34. Marciani L, Bush D, Wright P, et al. Monitoring of gallbladder and gastric coordination by EPI. J Magn Reson Imaging 2005;21:82–85.[CrossRef][Medline]
  35. Ziyeh S, Rick J, Reinhard M, Hetzel A, Mader I, Speck O. Blood oxygen level-dependent MRI of cerebral CO2 reactivity in severe carotid stenosis and occlusion. Stroke 2005;36:751–756.[Abstract/Free Full Text]
  36. Lythgoe DJ, Williams SC, Cullinane M, Markus HS. Mapping of cerebrovascular reactivity using BOLD magnetic resonance imaging. Magn Reson Imaging 1999;17:495–502.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Figure E1
Right arrow All Versions of this Article:
2433060433v1
243/3/727    most recent
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Barash, H.
Right arrow Articles by Abramovitch, R.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Barash, H.
Right arrow Articles by Abramovitch, R.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
RADIOLOGY RADIOGRAPHICS RSNA JOURNALS ONLINE