ORIGINAL 

Annals of Nuclear Medicine Vol. 7, No. 3, 141-145, 1993 

Difference in regional hepatic blood flow in liver segments -Non-invasive measurement of regional hepatic arterial and portal blood flow in human by positron emission tomography with H215O-

Hiroki TANIGUCHI, Atsushi OGURO, Kazumi TAKEUCHI, Keigo MIYATA, Toshio TAKAHASHI, Tadashi INABA*, and Hisamitsu NAKAHASHI* 

First Department of Surgery, Kyoto Prefectural University of Medicine and Nishtjin Hospital* 

Organ blood flow can be quantitatively measured by positron emission tomography (PET). As the liver has dual blood supplies, arterial and portal, regional hepatic blood flow had not been measured quantitatively. However, we succeeded in simultaneously measuring both regional hepatic arterial and portal blood flow by PET in non-stressed patients. Mean regional portal hepatic blood flow in patients with normal liver and cirrhotic liver was 57.5 and 36.7 ml/minutes/100 g, respectively. Mean regional arterial blood flow was 42.5 and 30.7 ml/minutes/100 g, respectively. A significant difference between regional portal hepatic blood flows in normal and cirrhotic patients was noted. Mean regional portal hepatic blood flow in the lateral, medial, anterior, and posterior segments of the liver was 29.8, 43.4, 50.0, and 40.9 ml/minutes/100 g, respectively. Mean regional arterial blood flow in each liver segment was 37.6, 30.0, 28.2, and 31 .6 ml/minutes/100 g, respectively. A significant difference between regional portal hepatic blood flows in lateral and anterior segment was noted. The p value was less than 0.025 and the 95 % confidence interval of the difference between means was from -20.2 to -2.7 ml/minutes/100 g by ANOVA. These results showed that regional hepatic blood flow is not the same in all the liver segments. 

Key words: regional arterial hepatic blood flow, regional portal hepatic blood flow, seg-mental hepatic blood flow, PET 

INTRODUCTION 

Organ blood flow can be quantitatively measured with position emission tomography (PET)1, for instance in the brain2-3 and heart.4 Although regional hepatic blood flow (rHBF) had not been measured by PET, because the liver has dual blood supplies, arterial and portal, we were able to measure regional hepatic arterial (rHBFa) and portal (rHBFp) blood flows by PET simultaneously, separately, non-invasively, and without stress. 

MATERIALS AND METHODS 

Theory 
If regional hepatic blood flow (rHBF) is expressed as shown schematically in Fig. 1 and a one compartment model6 is applied, the following simultaneous dif-ferential equations hold. The rate of change of radioactivity in the liver can be expressed as equation (A), when organs which drain into the portal vein are considered as one organ called the portal organ. In this equation, Fa and Fp are the tracer flows into the liver by way of the hepatic artery and from the portal organ, respectively, Ca(t) and Cp(t) are the tracer concentrations in the arterial blood and portal organ, respectively, Ch(t) is the radioactivity in the liver which is washed out by hepatic venous flow (Fh), Vh is the volume of the region of interest (ROI) in the liver, Kp is the con-stant for the portal organ, and Kh is the constant for the liver. If ROI for the portal organ is set to the same volume as for the liver, then the rate of change of radioactivity is represented by equation (B). From this condition, the tracer flow into the portal organ is Fp at a concentration equal to Ca(t), and the radioactivity washes out of the portal organ at a rate defined by Fp. Consequently, hepatic venous flow is the sum of arterial and portal blood flow (equation (C)). Because Fa/Vh. Fp/Vh, and Fh/Vh are rHBFa(=fa), rHBFp(=fp), and rHBF(=fh), respectively, equation (B) can now be expressed as follows : 
When this equation is substituted into equation (A), the solution of equation (A) becomes 
Since the PET radioactivity of the ROIs determined in the liver from tn to tn+1 is 
the combination of serial measurements and non-linear multiple regression analysis permits the calculation of rHBFa, rHBFp, Kh, and Kp. 

Materials 
For eighteen (12 males and 6 females) patients, rHBF was measured simultaneously in all 4 liver segments (lateral, medial, anterior, and posterior segments) by PET. They ranged in age from 32 to 74 with a mean age of 57.9 years. Five patients had normal liver function, 2 had chronic hepatitis, and 11 were cirrhotic, and had been diagnosed histo-logically. 

Methods 
Serial measurements were performed with the patients in the fasting state in a recumbent position on the bed of a whole body PET scanner (HEAD-TOME III SET-120W, Shimadzu Co., Kyoto, Japan). Following the intravenous injection of 20 to 30 mCi (370 to 740 MBq) of H215O produced by a medical cyclotron (BC-1710. Japan Steel Works, Muroran, Japan), twelve PET measurements were taken in the first minute and another 8 measurements were taken in the subsequent 4 minutes. Three PET scans 10 mm in width and spaced at 15 mm intervals were per-formed. ROIs were set on the liver segments of the PET images, always referring to the X-ray computed tomographic imagings on the same slices. From the PET images, the radioactivity in 4 liver segments was calculated. The rHBF of the segments was determined by means of 3 scans, and the rHBF in the whole liver was defined by means of 4 segments. Blood samples were taken from the left brachial artery for 5 minutes concurrent with PET measure-ments, at a rate of I every 5 seconds for the first 30 seconds and at intervals of 30 seconds thereafter, to give a total of 12 samples in 4 minutes. Radioactivity concentrations in the blood samp]es were measured immediately in a precalibrated well counter. They were approximated by non-linear multiple regression analysis with the following two equations. 
where tmax was the time when the radioactivity con-centration in the blood sample was maximal. 
In the current study, the specific gravity of the liver, Kp, and Kh was assumed to be 1. Physical decay of 150 was corrected every 2.5 seconds. The Simplex method6 was applied for non-linear multiple regression analysis and data processing was per-formed by means of a software newly produced by us with a personal computer (PC9801 RA2, NEC Co. Ltd., Tokyo, Japan). Statistical evaluation was performed with one way analysis of variance (ANOVA). 

RESULTS 

The result of non-linear multiple regression analysis is shown in Fig. 2. The dots and a line indicate the radioactivity concentrations in a hepatic ROI ob-tained by PET, and y= Ch(t) obtained by the Simplex method, respectively. 
The rHBFp in the whole liver ranged from 17.9 to 110.5 with a mean value of 41.0 ml/minutes/100 g. The mean rHBFa in the whole liver was 31.9 ml/ minutes/100 g, ranging from 16.1 to 70.3. The rHBF in the whole liver ranged from 40.6 to 135.6 ml/ minutes/100 g, and mean rHBF in the whole liver (Table). The mean rHBFp/rHBF ratio in the whole liver was 0.56. There was a significant correlation between rHBF in the whole liver and the 1 5 minute indocyanine green clearance test (p<0.03, R=0.53, y=1.12x+93.7) (Fig. 3). 
The mean rHBFps in the lateral, medial, anterior, and posterior segments of the liver were 29.8, 43.4, 50.0 and 40.9 ml/minutes/100 g, and 95 % confidence intervals for the segments were 17.5-42.1, 31.2-55.7, 37.7-62.3, 28.6-53.2, respectively. The mean rHBFas for the liver segments were 37.6, 30.0, 28.2, and 31.6 ml/minutes/100 g with 95% confidence intervals being 31.2-44.1, 23.5-36.4, 21.7-34.6, and 25.2-38.1, respectively. The mean rHBFS for the segments were 67.4, 73.4, 78.2, and 72.5 ml/minutes/100 g, and 95% confidence intervals were 54.5-80.4, 60.4-86.4, 65.2-91.2, and 59.5-85.5, respectively (Fig. 4). 
There was no significant difference among the 4 segment of the liver in the rHBFp, rHBFa, or rHBF. However, a significant difference between lateral and anterior segment rHBFps Was noted. The p value was less than 0.025 and the 95 % confidence interval of difference of means was from -20.2 to -2.7 ml/ minutes/100 g as determined by ANOVA. 

DISCUSSION 

There are several ways of measuring of hepatic blood flow, such as the dye clearance method with indocyanine green,7 the ultrasonic doppler method,8 electromagnetic flowmetry,5 laser doppler method,10 and reflectance spectrophotometry.n However, the first three methods cannot measure rHBF, and the last three methods require laparotomy. 
Scintiangiography with radioactive colloid12-13 can measure rHBFa and rHBFp separately and simultaneously, but semi-quantitatively. Neverthe-less, because ROI is restricted by the other organs such as the lungs, the concentration of radioactivity in the whole liver cannot be measured. If single photon emission computed tomography (SPECT) is used, this problem is solved, but it is still semi-quantitative measurement because of the dispersion of single photons. Another problem with this method is that it requires curve analysis of the arterial phase and the portal phase. There are several methods of curve analysis, and the best method has not been determined. Scintiphotography with rare gases such as 133Xe14-15 and 85Kr16 takes only a short time, and if the arterial catheter technique is used, rHBFa and rHBFp can be measured separately.15 However, 133Xe is soluble in fatty tissue and some invasions are carried out in a patient with a catheter. Moreover, if the hepatic artery is occluded with a balloon catheter to measure portal blood flow, this flow is thought to increase due to hepatic arterial buffer response.17 
Because the position tracer is used, accurate quantification by means of PET is much better than by SPECT. And PET in which a biological radio-tracer is used can perform physiological measure ments are possible and radiotoxicity is limited, as the tracer has a short half life. In spite of these many merits,18 quantitative measurement of rHBF by PET has not been performed, primarily because of the complexity of the hepatic blood supply. However, we succeeded in measuring segmental rHBF by PET ; both rHBFa and rHBFp simultaneously, separately, and easily. Besides, we found that rHBF is not the same to all liver segments. 
Nevertheless, a few problems are associated with this method. The first relates to choosing values for the constants Kp and Kh. These constants are con-sidered to relate to the organ blood equilibrium parti-tion coefficients for the tracer. The liver, especially chronically damaged liver, had portosystemic shunts. And there is the time delay between the radioactivity of a radiotracer in the portal organ and that in the liver. The latter must be considered when the tracer has a very short half life. These factors should be included in Kp, too. However, in the present study. Kp and Kh were assigned values of 1. As a result, rHBFp, and the rHBFp/rHBF ratio were estimated to be lower than the values for them that have been commonly known. Moreover, it is thought that Kh may change with the progress of chronic liver disease_ An analysis of these constants will be needed. 
The second problem, which is the most important one, stems from the assumption that water dis-tribution is defined by a one compartment model. Although this assumption can be tested for an organ that has a single blood supply, such as the brain and the heart,4,19 it nevertheless remains extremely diffi-cult. If a two or more compartment model is used, the analysis is of no practical value since it becomes very complicated. A final problem arises from the cost of the PET systems,20 although they undoubted-ly will become cheaper as they become more widely used. 
With the demonstration of a new non-invasive technique for the quantitative measurement of both rHBFa and rHBFp simultaneously, a new tool has become available for studying the pathophysiology of liver disorders. 

REFERENCES 

1 . Huang SC, Carson RE, Phelps ME : Measurement of local blood flow and distribution volume with short-lived isotopes : A general input technique. J Cereb Blood Flow Metab 2: 99-108, 1982 
2. Herscovitch P, Markham J, Raichle ME : Brain blood flow measured with intravenous H215O ; Theory and error analysis. J Nucl Med 24 : 782-789, 1983 
3. Huang SC, Carson RE, Hoffman EJ, Mcdonald N, Barrio JR, Phelps ME : Quantitative measurement of local cerebral blood flow in humans by positron computed tomography and 15O-water. J Cereb Blood Flow Metab 3: 141-153, 1983 
4. Bergman SR, Fox KAA, Rand AL, McElvany KD. Welch MJ, Markham JM, Sobel BE : Quantification of regional myocardial blood flow in vivo with H215O. Circulation 70 : 724-733, 1984 
5. Ketty SS: The theory and applications for the ex-change of inert gas at the lungs and tissues. Pharmacol Rev 3: 1-41, 1951 
6. Nelder JA, Mead R : A simplex method for function minimization. Computer J 7 : 308-313, 1965 
7. Caesar J, Shaldon S, Chiandussi L, Guevara L. Sharlock S : The use of indocyanine green in the measurement of hepatic blood flow and as a test of hepatic function. Clin Sci 21 : 43-57, 1961 
8 . Moriyasu F, Nishida O, Nakamura T et al : Measure-ment of portal vascular resistance in patients with portal hypertension. Gastroenterol 90: 710-717, 1986 
9. Moreno AH, Burchell AR, Reddy RV: Portal blood flow in cirrhosis of the liver. J Clin. Invest 46 : 436-445, 1967 
10. Shepherd AP, Riedel GL : Kiel JW, Haumschild DJ, Maxwell LC : Evaluation of infrared Laser-Doppler blood flowmeter. Am JPhysiol 252 : G832-G839, 1987 
11. Sato N, Hagihara. Kamada T, et al : Non destructive measurement of tissue oxidative functions : Mcasure-ment of concentration of respiratory enzymes and the rate of oxygen consumption in living liver tissue by reflectance spectrophotometry. In Frontiers in Biological and Energetics, Scarpa A, et al eds. Academic Press, New York, 1978, pp 1507 
12. Sarper R, Fajman WA, Rypius EB, Henderson JM, Tarcan YA, Galambos JT, Warren WD : A non-invasive method for measuring portal venous/total hepatic blood flow by hepatosplenic radionuclide angiography. Radiology 141 : 179-184, 1981 
13. Fleming JS, Ackery DM, Walmsley BH, Karran SJ : Scintigraphic estimation of arterial and portal blood supplies to the liver. J Nucl Med 24: 1108-1113, 1983 
14. Rees JR, Reeding VJ, Ashfield R : Hepatic blood-flow measurement with xenon 133; Evidence for separate hepatic-arterial and portal-venous pathway. Lancet 2 : 562-563, 1964 
15. Yasuhara Y, Miyauchi S. Hamamoto K : A measure-ment of regional portal blood flow with Xe-133 and balloon catheter in man. Eur J Nucl Med 1 5 : 346-350, 1989 
16. Hollenberg M, Dougherty J : Liver blood flow measured by portal venous and hepatic arterial routes with Kr85. Am JPhysiol 210: 926-932, 1966 
17. Lautt WW, Legare DJ, Ezzat WR : Quantitation of hepatic arterial buffer response to graded change in portal blood flow. Gastroenterology 98: 1024-1028, 1990 
18. Tilyou SM: Yesterday, today, and tomorrow : The evolution of positron emission tomography. J Nucl Med 32: 15N-26N, 1991 
19. Huang SC, Carson RE, Hoffman EJ, Carson J, MacDonald N, Barrio JR, Phelps ME : Quantitative measurement of local cerebral blood flow in humans by position computed tomography and 150-water. J Cereb Blood Flow Metab 3 : 141-153, 1983 
20. Chilton HM, Hawkins RA, Maddahi JM, Phelps ME, Hubner KF, Frick MP : Planning and financing a PET center. J. Nucl Med 32: 35N-51N, 1991