THE RELATIONSHIP BETWEEN LIVER VOLUME,

Transcription

1 Br. J. clin. Pharmac. (1976), 3, THE RELATIOHIP BETWEEN LIVER VOLUME, ANTIPYRINE CLEARANCE AND INDOCYANINE GREEN CLEARANCE BEFORE AND AFTER PHENOBARBITONE ADMINISTRATION IN MAN C.J.C. ROBERTS, L. JACKSON, M. HALLIWELL & R.A. BRANCH* Departments of Medicine and Pharmacology, University of Bristol and Department of Medical Physics, Bristol General Hospital, Bristol I Liver volume and the clearances of antipyrine and indocyanine green have been measured before and after administration of phenobarbitone (18 mg/day) for 3 weeks to ten healthy subjects. 2 The measurement of liver volume by an ultrasound scanning technique yielded reproducible results which were consistent with predictions of liver size by allometric methods. 3 Before phenobarbitone, antipyrine clearance correlated with liver volume, but there was no correlation between indocyanine green clearance and liver volume. 4 Phenobarbitone administration increased the clearance of antipyrine significantly by 9 ± 14% but there was no significant change in indocyanine green clearance or liver volume. 5 After phenobarbitone the correlation between antipyrine clearance and liver volume persisted. There was no correlation between indocyanine green clearance and liver volume. 6 These results suggest that in non-medicated subjects some of the difference in antipyrine clearance is due to difference in functional hepatic parenchymal mass and that phenobarbitone increases the drug metabolising capacity per unit of hepatic mass but not total liver size. Introduction The elimination of antipyrine and indocyanine green is determined by different rate limiting factors. Antipyrine is a soluble drug which is rapidly and completely absorbed when administered orally. It does not bind to plasma or tissue proteins and is distributed similarly to the body water. As it is lipid soluble it is not excreted by the kidney but undergoes metabolism by the microsomal enzymes in the liver prior to elimination (Brodie & Axelrod, 195). The rate of metabolism is slow so that hepatic extraction is low and hepatic clearance is less than 2% of liver blood flow (LBF) and thus clearance is independent of changes in LBF (Branch, Shand, Wilkinson & Nies, 1974). In contrast indocyanine green is a dye which is almost completely protein bound and is rapidly eliminated by the liver being actively transported into bile without undergoing metabolic transformation (Cherrick, Stein, Leevy & Davidson, 196). The hepatic extraction ratio is high. Consequently the hepatic clearance is high * Present address: Department of Clinical Pharmacology, Vanderbilt University, Nashville, Tennessee, U.S.A. 55 and is limited by LBF (Branch, Shand & Nies, 1973) and has been used to measure LBF (Caesar, Shaldon, Chiandussi, Guevara & Sherlock, 1961). In spite of the differences in the rate limiting factors in the elimination of these two drugs a strong positive correlation has been observed between the clearances of these two agents both in normal subjects and in patients with chronic liver disease when it was postulated that the more fundamental common rate-limiting factor in their elimination was the 'functional hepatic parenchymal mass' (Branch, James & Read, 1975). This hypothesis would suggest that variation in the clearances of either drug between subjects is dependent on the 'functional hepatic parenchymal mass' which in normal subjects would approximate to liver volume. The objective of this study was to develop a method for measuring liver volume using an ultrasound technique and to investigate whether differences between subjects in the clearance of these two drugs could be explained by differences in liver volume. Furthermore microsomal enzyme induction by phenobarbitone is known to be accompanied by

2 98 C.J.C. ROBERTS, L. JACKSON, M. HALLIWELL & R.A. BRANCH an increase in hepatic mass and LBF in the rat (Ohnhaus, Thorgeirsson, Davies & Breckenridge, 1971) and rhesus monkey (Branch et al., 1974) but its influence on hepatic size has not been previously determined in man. This study has investigated the influence of phenobarbitone on liver volume and the disposition of antipyrine and indocyanine green in man and has examined the interrelationships between these parameters after enzyme induction. Method Three female and seven male healthy subjects aged 2-3 years volunteered to take part in the study which had been approved by an ethical committee. No drugs or oral contraceptives were being taken and alcohol was limited to one pint of beer per day. Each subject received phenobarbitone 12 mg orally at night plus 3 mg twice daily for 3 weeks. Liver volume was measured three times before and once after phenobarbitone. Indocyanine green clearance, antipyrine clearance, serum bilirubin, alkaline phosphatase, aspartate transaminase, albumin and globulin were measured before and after phenobarbitone administration. Plasma phenobarbitone levels were measured at the end of 3 weeks of drug administration. Liver volumes were determined by an ultrasonic scanning technique similar to those of Rasmussen (1972) and Tolbert, Zagzebski, Banjavic & Wiley (1974). Serial transverse scans were recorded at 1 cm intervals from the inferior limit of the liver to the upper limit at which liver outline could be clearly visualised. Liver edge above this level was delineated by serial longitudinal scans in the sagittal plane. Scans were performed during maximum inspiration with the subject supine. Scans were photographed and liver outline transcribed to paper. The whole areas of the liver on the transverse scans and the appropriate areas on the longitudinal scans were measured using a computer fitted with a graphic tracing device. Liver volume was derived as the sum of all the areas measured multiplied by the magnification factor used. After an overnight fast and removal of blood for haematocrit and blank and standard curve measurements, indocyanine green (.5 mg/kg body weight) was administered intravenously. Venous blood samples were taken from the opposite arm through an indwelling catheter at 3 min intervals for 21 minutes. Plasma concentrations were measured within 2 h of administration of the dye by the method of Caesar et al. (1961). Antipyrine (12 mg) was administered orally as a freshly prepared solution by dissolving crystalline antipyrine in water (5 ml). Venous sampling was performed through an indwelling catheter. Samples were taken at 5, 1, 15, 2, 3, 45 min and 1, 2, 3, 5,7,9, 15 and 24 h after drug administration. Antipyrine was assayed by the spectrophotometric method of Brodie, Axelrod, Soberman & Levy (1949). Calculations The clearances of antipyrine and indocyanine green were estimated from the area under the plasma concentration curve (AUC) extrapolated to infinity as clearance = Dose/AUC. The plasma half-life (TL.) was estimated by least squares regression analysis of the log concentration - time profile. Absorption of antipyrine was shown to be rapid, mean time to peak plasma level being 39 ± 8 min (mean ± s.e. mean). Assuming complete absorption of antipyrine and a monoexponential elimination of indocyanine green, the clearance of each drug was calculated from Dose.693 Clearance = x Cpo T2 where Cp = plasma concentration back extrapolated to the time of drug administration. Plasma indocyanine green clearance was corrected to whole blood clearance using the haematocrit. The blood to plasma ratio for antipyrine was assumed to approximate to unity (Branch et al., 1974) and thus blood clearance equated with plasma clearance. Results Liver volume measurement The technique for measuring liver volume was found to yield reproducible results. The mean coefficient of variation for the results of the three base line scans was 5.7%. The values obtained ranged between 1.8% and 2.4% of total body weight with a mean of % (Table 1). Liver volume correlated with body weight (r = +.81, P <.1). Liver volume, antipyrine clearance and indocyanine green clearance before phenobarbitone administration One subject had a very high initial antipyrine clearance (8 ml/min), being 5 ml/min greater than two standard deviations above the mean for the whole group. This increased only minimally after phenobarbitone suggesting that this subject had maximal enzyme induction at the start of the study but no obvious cause was evident. The antipyrine clearance of the other nine subjects

3 PHENOBARBITONE, LIVER VOLUME AND DRUG DISPOSITION 99 Ea c 6 I- u 2 a2) : 15- C.C C, I c 4._ C < 2L * -1 C>Z ( O 2 c 5CD Liver volume (ml) Figure 1 The relationships ( between liver volume and antipyrine clearance (nine healthy subjects) and (b) between liver volume and indocyanine green clearance (ten healthy subjects) before enzyme induction. Open circles identify subjects with an apparently low indocyanine green clearance for their liver volume. For ( r = +.69, P < correlated with liver volume (r = +.69, P <.5) (Figure 1). There was no significant correlation between the clearance of indocyanine green and liver volume. However, the initial indocyanine green clearances in seven subjects approximated to 1 ml/min per ml of liver, whilst in the other three subjects the indocyanine green clearance per unit volume of liver was lower (Figure 1). Influence of phenobarbitone administration The plasma phenobarbitone levels in the ten subjects at the end of the 3 weeks' administration were between mg/ 1 ml. There was no significant change in liver volume (Table 2). Antipyrine clearance increased by 9. ± 14.1% (mean ± s.e. mean) (P<.1) while the volume of distribution of antipyrine remained unchanged. Indocyanine green clearance increased by 15.7 ± 5.9% (P =.54). The volume of distribution of indocyanine green increased by 18.6 ± 9.5% (not statistically significant). The serum bilirubin fell by 4 ± 9.1% (P <.2) and the plasma alkaline phosphatase increased by 17.5 ± 7.3% (P<.5). These changes did not correlate with changes in other parameters measured. There were no changes in serum aspartate transaminase or serum albumin (Table 3). Table 1 The relationship between body weight and liver volume, determined as the mean of three measurements made by ultrasonic scanning technique. Subject number Liver volume (ml) Coefficient of Body weight Liver volume as variation (%) (kg) % of body weight Mean Range

4 91 C.J.C. ROBERTS, L. JACKSON, M. HALLIWELL & R.A. BRANCH E E a Ea ' b 1 5-cci co o c ci) CD 75- C 15- c 1- l. *CZ <~~I c 5 Liver volume (ml). * * Figure 2 The relationships between liver volume and the clearances of ( antipyrine and (b) indocyanine green in ten healthy subjects after 3 weeks' administration of phenobarbitone (18 mg/day). Open circles identify subjects with an apparently low indocyanine green clearance for their liver volume. For ( r = +.86, P <.1. Relationship between liver volume, antipyrine clearance and indocyanine green clearance after phenobarbitone After phenobarbitone administration the antipyrine clearance of all ten subjects correlated with liver volume (r = +.86, P <.1 ) (Figure 2). There was no correlation between the indocyanine green clearance and liver volume. However, the apparent relationship of indocyanine green clear ance to liver volume previously observed was maintained in the same seven subjects, while the clearances in the other three subjects remained low in relation to liver volume (Figure 2). Discussion The concept that 'functional hepatic parenchymal mass' is a major determinant of inter-individual Table 2 Mean values ± s.e. mean before and after 3 weeks administration of phenobarbitone (18 mg/day) to ten subjects. Uver volume (ml) Antipyrine clearance (ml/min) Volume of distribution of antipyrine (litres) Antipyrine plasma T± (h) Indocyanine green clearance (ml/min) Volume of distribution of indocyanine green (litres) Indocyanine green plasma T± (min) Before After phenobarbitone phenobarbitone 1339 ± ± ± 3.1 Mean % change P value (paired t-test) ± ± 14.1 <.1 4 ± ± ± ± ± ± ± ± ± 5.9 (P=.54) 3.4 ± ± ± ± ± ± 7.9 <.1

5 PHENOBARBITONE, LIVER VOLUME AND DRUG DISPOSITION 911 Table 3 Mean values ± s.e. mean before and after 3 weeks administration of phenobarbitone 8 mg/ day) to ten subjects. Serum bilirubin Serum alkaline phosphatase Serum albumin Serum aspartate transaminase Mean % change after phenobarbitone P value (Paired t-test) -4 ± 9.1 < ± 7.3 < ± ± 8.6 variation in the disposition of certain drugs eliminated by the liver has been supported by the correlation of liver volume to the clearance of antipyrine. However, the failure to correlate liver volume to the clearance of indocyanine green suggests that this concept is not valid for all drugs. Differences between the mechanisms of elimination may account for this variation between the two drugs. The concept of 'functional hepatic mass' has been further investigated following chronic administration of phenobarbitone. Doses of phenobarbitone adequate to induce hepatic enzymes and a fall in serum bilirubin in healthy subjects over a three-week period did not change liver volume but caused an increase in antipyrine clearance per unit of hepatic mass. The method established for the measurement of liver volume was found to be useful for the purpose of this study and to have certain advantages over previously reported methods. The combination of scans in two planes allowed visualisation of the upper and lower borders of the liver and obviated the need for angulated scans as used by Rasmussen (1972). Furthermore the recording of parallel scans at 1 cm intervals removed the need for complicated geometry at analysis. Straightforward summation of areas multiplied by the magnification factor yielded liver volume. Confidence in the method was derived from the acceptable coefficient of variation from repeated measurements of liver volume in the same individual together with the close correlation between liver volume and body weight. When liver volume was expressed as a percentage of body weight a small range was obtained and this range was in close agreement with allometric predictions of liver size based on autopsy findings (Ludwig, 1972). The absence of significant protein or tissue binding of antipyrine together with its slow and complete hepatic metabolism make antipyrine an ideal agent for study of hepatic drug metabolizing capacity. Whilst there is a wide interindividual variation in in vivo rates of drug clearance (Davies, Thorgeirsson, Breckenridge & Orme, 1973), variation in microsomal drug metabolizing enzyme concentration and in vitro activity has been found to be comparatively small (Schoene, Fleischmann, Remmer & Olderhausen, 1972; Lecamwasam, Franklin & Turner, 1975). Attempts to correlate rates of drug metabolism in vitro with those in vivo in man have been unsuccessful (Davies et al., 1973). The correlation between antipyrine clearance and liver volume in the present study suggests that variation in hepatic parenchymal mass might explain this failure. Thus it is postulated that variance in antipyrine clearance is composed of variance in drug metabolizing enzyme concentration and activity plus variance in mass of functional hepatic parenchyma. This hypothesis would be consistent with the finding that in experimental models antipyrine biotransformation in vivo does correlate with that in vitro in animals selected for similarity in weight (Statland, Astrup, Black and Oscholm, 1973; Vesell, Lee, Passaranti & Shively, 1973). Similarity in liver size might explain the smaller intrapair variance in antipyrine half-life seen in identical twins as compared to fraternal twins (Vesell & Page, 1969). The disposition of indocyanine green differs greatly from that of antipyrine being actively and rapidly excreted into bile without undergoing biotransformation. Two rate-limiting factors for its clearance are liver blood flow and the capacity of the active transport mechanism. The former tends to be rate limiting in lower dosages whilst the latter is rate limiting in high doses (Paumgartner, Probst, Kraines & Leevy, 197). Using a dose of.5 mg/kg in the present study no clear cut relationship between indocyanine green clearance and liver volume was found. However, the apparent proportionality between indocyanine green clearance and liver volume in the seven subjects represented by closed circles in Figure 1 suggests the possibility that the rate-limiting factor for indocyanine green clearance was liver blood flow with each gram of liver being able to clear 1 ml/min thus implying that these seven subjects had a high maximal hepatic removal rate (Rm). In the three subjects represented by open circles it is possible that the rate-limiting factor was a low Rm. This would be consistent with the previously reported wide range (53-9%) of arterio-hepaticvenous extraction ratios in normal individuals (Caesar et al., 1961; Cherrick et al., 196; Wiegand, Ketterer & Rapaport, 196). The alternative possibility is that the liver blood flow in the latter three subjects was low per unit of liver volume. Thus individual variation in the clearance

6 912 C.J.C. ROBERTS, L. JACKSON, M. HAL-LIWELL & R.A. BRANCH of indocyanine green may depend on variations in the capacity of the active transport mechanism and blood flow per unit of liver volume as well as on total functional hepatic mass. Phenobarbitone was administered for a period of three weeks in order to allow ample time for maximum enzyme induction to occur (Breckenridge, Orme, Davies, Thorgeirsson & Davies, 1973) and the dose used was a large one, causing a degree of sedation in all subjects. Independent evidence of enzyme induction was provided by the fall in serum bilirubin thought to be due to induction of the enzyme UDP-glucuronyl transferase (Black, Perrett & Carter, 1973) and by the rise in plasma alkaline phosphatase. Phenobarbitone was found to have no effect on liver volume. This is in contrast to a 27% increase in liver weight found in rats after 8 days' phenobarbitone administration (Ohnhaus et al., 1971) and a 34% increase in liver weight in rhesus monkeys after 12 days of phenobarbitone administration (Branch et al., 1974). This interspecies difference in response to phenobarbitone may be related to the very much lower concentration of microsomal protein found in human liver tissue (32 to 38 mg/g of liver) in comparison to that in the rat (5 to 6 mg/g of liver) (Schoene et al., 1972). Phenobarbitone caused an increase in the clearance of antipyrine without a change in the volume of distribution and consequently a shortening of the plasma T2; this reduction in T1 is in agreement with the findings of other workers (Davis et al., 1974; Vesell & Page, 1969). It is also consistent with the finding that phenobarbitone administration to rats caused an increase in the maximal velocity of ethyl morphine metabolism per gram of liver in vitro and an increase in the cytochrome P45 concentration as well as an increase in total liver mass (Ohnhaus et al., 1971). The effect of phenobarbitone on indocyanine green clearance was less clear cut. Although the increase in mean indocyanine green clearance in the present study does not quite reach statistical significance it is consistent with previous reports. Increased indocyanine green clearance has been found in patients taking phenobarbitone (Melikian, Eddy & Paton, 1972) and phenobarbitone has been shown to enhance indocyanine green excretion in animals (Klaassen & Plaa, 1968). A possible explanation for these findings is the increase in liver blood flow induced by phenobarbitone (Branch et al, 1974). The relationship between antipyrine clearance and liver volume was even more impressive after phenobarbitone. The absence of a change in liver volume after phenobarbitone together with the consistency of the correlation between antipyrine clearance and liver volume indicate that the increase in antipyrine clearance induced by phenobarbitone in man is due to an increase in drug metabolizing efficiency per unit of functional hepatic mass. Furthermore the increased coefficient of correlation between antipyrine clearance and liver volume implied that variance in drug metabolising capacity per unit mass of liver decreased. This would be consistent with the decreased variance in antipyrine half-life known to occur after enzyme induction with phenobarbitone (Vesell and Page, 1969). The increase in mean indocyanine green clearance did not affect its relationship to liver volume, the same three subjects having a low clearance for their liver volume. In the other seven subjects the apparent relationship between liver volume and indocyanine green clearance was preserved. This would further support the view that different factors are consistently rate limiting between individuals. Liver size has been shown to operate as a fundamental rate-limiting factor for the disposition of antipyrine and can partially account for individual variation in rates of clearance of the drug. The findings vindicate the need for measurement of liver size when comparing rates of drug metabolism between healthy subjects and between patients. Whilst total functional hepatic mass is likely to be an important factor in the disposition of drugs eliminated by the liver it will not explain all the inter-individual variation when multiple factors are involved in the elimination process. We thank Miss J. Ford for her technical assistance. References BLACK, M., PERRETT, R.D. & CARTER, A.E. (1973). Hepatic bilirubin UDP-glucuronyl transferase activity and cytochrome P45 content in a surgical population and the effect of preoperative drug therapy. J. Lab. clin. Med., 81, BRANCH, RA., JAMES, J.A. & READ, A.E. (1975). Major determinants of drug disposition in chronic liver disease: A study with indocyanine green and antipyrine. Br. J. clin. Pharmac., 2, BRANCH, R.A., SHAND, D.G. & NIES, A.S. (1973). Haemodynamic drug interactions: the reduction of oxyphenbutazone clearance by d, I-propranolol in the dog.j. Pharmac. exp. 7her., 187,, BRANCH, R.A., SHAND, D.G., WILKION, G.R. &

7 PHENOBARBITONE, LIVER VOLUME AND DRUG DISPOSITION 913 NIES, A.S. (1974). Increased clearance of antipyrine and d-propranolol after phenobarbital treatment in the monkey. J. clin. Invest., 53, BRECKENRIDGE, A., ORME, M.L.E., DAVIES, L., THORGEIRSSON, S.S. & DAVIES, D.S. (1973). Dose-dependent enzyme induction. Pharmac. Rev., 19, BRODIE, B.B. & AXELROD, J. (195). The fate of antipyrine in man. J. Pharmac. exp. Ther., 98, BRODIE, B.B., AXELROD, J., SOBERMAN, R. & LEVY, B.B. (1949). The estimation of antipyrine in biological materials. J. biol Chem., 179, CAESAR, J., SHALDON, S., CHIANDUSSI, L., GUEVARA, L. & SHERLOCK, S. (1961). The use of indocyanine green in the assessment of hepatic blood flow as a test of hepatic function. Clin. Sci., 21, CHERRICK, G.R., STEIN, S.W., LEEVY, C.M. & DAVIDSON, C.S. (196). Indocyanine green: Observations on its physical properties, plasma decay and hepatic extraction. J. clin. Invest., 39, DAVIES, D.S., THORGEIRSSON, S.S., BRECKEN- RIDGE, A. & ORME, M.L.E. (1973). Inter-individual differences in rates of drug oxidation in man. Drug Met. Dis., 1, KLAASSEN, C.D. & PLAA, G.L. (1968). Studies on the mechanism of phenobarbital enhanced sulfobromophthalein disappearance. J. Pharmac. exp. Ther., 161, LECAMWASAM, D.S., FRANKLIN, C. & TURNER, P. (1975). Effect of phenobarbitone on hepatic drugmetabolising enzymes and urinary D-glucaric acid excretion in man. Br. J. clin. Pharmac., 2, LUDWIG, J. (1972). Normal weights and measurements. In Current Methods of Autopsy Practice. ed. Ludwig, J. pp Philadelphia: W.B. Saunders and Co. MELIKIAN, V., EDDY, J.D. & PATON, A. (1972). The stimulant effect of drugs on indocyanine green clearance by the liver. Gut, 13, OHNHAUS, E.E., THORGEIRSSON, S.S., DAVIES, D.S. & BRECKENRIDGE, A. (1971). Changes in liver blood flow during enzyme induction. Biochem. Pharmac., 2, PAUMGARTNER, G., PROBST, P., KRAINES, R. & LEEVY, C.M. (197). Kinetics of indocyanine green removal from the blood. Ann. N. Y. Acad. Sci., 17, RASMUSSEN, S.N. (1972). Liver volume determination by ultrasound scanning. Br. J. Radiol., 45, SCHOENE, B., FLEISCHMANN, R.A., REMMER, A. & OLDERHAUSEN, H.F.V. (1972). Determination of drug metabolising enzymes in needle biopsies of human liver. Eur. J. clin. Pharmac., 4, STATLAND, B.E., ASTRUP, P., BLACK, C.H. & OSCHOLM, E. (1973). Plasma antipyrine half-life and hepatic microsomal antipyrine hydroxylase activity in rabbits. Pharnacology, 1, TOLBERT, D.D., ZAGZEBSKI, J.A., BANJAVIC, R.A. & WILEY, A.L. (1974). Quantitation of tumour volumes and response to therapy with ultrasound B-scans. Radiology, 113, VESELL, E.S. & PAGE, J.G. (1969). Genetic control of phenobarbital induced shortening of plasma antipyrine half lives in man. J. clin. Invest., 48, VESELL, E.S., LEE, C.J., PASSARANTI, G.R. & SHIVELY, C.A. (1973). Relationship between plasma antipyrine half lives and hepatic microsomal drug metabolism in dogs. Pharmacology, 1, WEIGAND, D.B., KETTERER, S.G. & RAPAPORT, E. (196). The use of indocyanine green for the evaluation of hepatic function and blood flow in man. Am. J. dig. Dis., 5, (Received November 24, 1975)