Plasma Concentration of Endothelin-1 Does. Not Re¯ ect Renal Vasoconstriction as. Estimated by Duplex Ultrasonography in Cirrhosis. HIROSHI KITAMURA ...
Digestive Diseases and Sciences, Vol. 42, No. 3 (M arch 1997), pp. 542± 545
Plasma Concentration of Endothelin-1 Does Not Re¯ ect Renal Vasoconstriction as Estimated by Duplex Ultrasonography in Cirrhosis HIROSHI KITAMURA, MD, RYO SHIMADA, MD, AKIRA KOBAYASHI, MD, KAZUHIKO NOMURA, MD, TERUMASA NOIKE, MD, HARUHISA HARADA, MD, SHINICHI MIYAGAWA, MD, and SEIJI KAWASAKI, MD
Endothelin, a potent vasoconstrictor, is thought to play a role in liver cirrhosis-related functiona l kidney failure. Our aim was to investigate the correlation between renal vasoconstriction, as detected by a Doppler ultrasound techniqu e, and plasma concentrations of endothelin in cirrhotic patients. Fifty cirrhotic patients underwent Doppler examinatio ns to detect renal vasoconstriction. The plasma concentration of endothelin was measured in 10 patients who exhibited vasoconstriction of the renal microvessels diagnosed by Doppler waveform analysis and was compared to that of patients in whom there was no sign of such vasoconstriction. No difference was observed in the plasma concentration of endothelin between patients in whom renal vasoconstri ction was diagnosed and those in whom it was not. Our results suggested that the circulatin g endothelin does not re¯ ect renal vasoconstri ction, at least in the early phase of the functiona l renal failure associated with cirrhosis of the liver. KEY WORDS: endothelin; Doppler waveform analysis; renal vasoconstriction; resistive index; liver cirrhosis; hepatorenal syndrome.
Endothelin, found in blood vessel endothelium, is the most potent vasoconstri ctor known. The kidney is considered more sensitive than other vascular territories to the vasoconstrictor effect of endothelin (1, 2). Recent studies of liver disease-related kidney failure (hepatore nal syndrome) suggested that renal vasoconstriction plays a part in the etiology of functiona l renal fail ure. Endothelin also has been considered to have a role in vasoconstriction in the kidney (3). However, a question arising from several previous studies is whether circulating endothelin is Manuscript received M ay 16, 1996; accepted October 23, 1996. From the Shinshu University School of Medicine, First Department of Surgery, Matsumoto, Japan. Address for reprint requests: Dr. Hiroshi Kitamura, First Department of Surgery, Shinshu University School of Medicine, 3-1-1 Asahi, M atsumoto, Nagano 390, Japan.
542
relevant to the renal dysfunctio n associated with cirrhosis because of its paracrine nature (4 ± 7). Therefore, in the present study, we set out to investigate the relationshi p between vasoconstriction of the renal vessels and plasma concentrations of endothelin. Recent advances in ultrasonogr aphic studies have demonstrated their ability to assess renal vasoconstriction using Doppler waveform analysis. These ultrasound parameters are regarded as successful predictors of the renal failure that is associated with chronic liver diseases (8, 9). Plasma concentrations of endothelin were measured in cirrhotic patients in whom renal vasoconstri ction had been detected by the Doppler waveform analysis and were compared to that of cirrhotic patients in whom there was no sign of such vasoconstri ction. Digestive Diseases and Sciences, Vol. 42, No. 3 (March 1997)
0163-2116/97/0300-0542$12.50/0 q 1997 Plenum Publishing Corporation
ENDOTHELIN AND RENAL VASOCONSTRICTION
MATERIALS AND METHODS Hepatic Reserve and Resistive Index. Fifty patients with histologically diagnosed liver cirrhosis underwent Doppler examination. None of the subjects had any history of, or current, renal or cardiovascular diseases, nor did they present with overt hepatorenal syndrome at the time of examination. Vessel resistance was measured using a parameter termed the resistive index (RI), which was derived from the Doppler wavefo rm analysis. An RI value of 0.7 was indicative of elevated renal vascular resistance and renal vasoconstriction (according to a study by Platt et al) (10). Patients were also classi® ed according to the Child-Pugh classi® cation (11). Of 50 patients, 19 (38%) were classi® ed as Child-Pugh class A, 21 (42%) as class B, and 10 (20%) as class C. The RIs of these three classes were compared to verify correlation between RI and a conventional estimation for hepatic reserve. The indocyanine green clearance test [ICG(159 )] (12, 13) as a measure of hepatic reserve was also carried out in 35 subjects to determine the correlation with RI. Plasma Endothelin Concentration and RI. Of those 50 cirrhotic patients, 10 patients had an RI of 0.7 or greater. These patients underwent blood sampling immediately after the Doppler examination for measurement of the plasma concentration of endothelin. To compare with the cirrhotic group, 10 patients with an RI of less than 0.7 were randomly selected for the blood sampling. Another 10 patients with normal hepatic reserve (not included in the cirrhotic group) also served as controls for plasma endothelin concentration. Ultrasound Technique. A real-time ultrasound examination of each kidney was performed with a 5.0-MHz transducer (model SSD-2000; Aloka, Tokyo, Japan), and a pulsed Doppler evaluation of the intrarenal arteries was carried out. Doppler signals were obtained from either the accurate arteries at the corticomedullary junction, from the interlobar arteries along the border of the medullary pyramids, or from both. The identi® cation of vascular segments was based upon the imaged anatomy and the audible and spectral characteristics of ¯ ow velocity patterns. The Doppler sample volume was set at 2 mm. The ultrasound examination was performed with the patient in the supine position. The RI, de® ned as [(peak systolic frequency shift 2 minimum diastolic frequency shift)/peak systolic frequency shift], was determined by positioning markers on the monitor screen and allowing the system to automatically calculate the indices. The RI for each patient was calculated as an average value of both kidneys. In each kidney, three to ® ve waveforms were recorded in two different regions and averaged. Enzyme Immunoassay for Endothelin. A sandwich-type enzyme immunoassay (EIA) kit for endothelin-1 (Wako, Osaka, Japan) was used to measure plasma concentrations of endothelin (14). A monoclonal antibody that recognizes the N-terminal portion of endothelin was used as an immobilizing antibody. A polyclonal antibody against the Cterminal peptide was used as the enzyme-labeled antibody, after being coupled to horseradish peroxidase. A 1-ml sample of plasma was acidi® ed with 4% acetic acid and applied to Sep-Pak C18 cartridges (W aters Associates, Mildford, Digestive Diseases and Sciences, Vol. 42, No. 3 (M arch 1997)
Fig 1. Relationship between Child-Pugh classi® cation and RI. There were signi® cant differences in RI between class C and the other two groups.
Massachusetts). Cartridges were washed with distilled water, and immunoreactive endothelin-1 was eluted with 10 ml of a solvent consisting of 4% acetic acid and 86% ethanol. The eluted endothelin-1 was then dried and reconstituted for EIA. The sensitivity of the EIA was 0.5 pg/ml. The normal level of endothelin-1 and the sensitivity in this assay were less than 2.3 pg/ml and 0.5 pg/ml, respectively. Informed consent was obtained from each patient concerning the ultrasound examination and blood sampling. Statistical analysis of the results was carried out using an unpaired t test and two-variable regression analysis. Results are expressed as mean 6 SD . P , 0.05 were considered to be statistically signi® cant.
RESULTS Hepatic Reserve and RI. Figure 1 shows a comparison of RI with three groups categorized by ChildPugh classi® cation (A: 0.647 6 0.047, B: 0.642 6 0.038, C: 0.717 6 0.055). The RIs of patients in class C were signi® cantly higher than in other two groups. Figure 2 show s the regression analysi s between ICG(15 9 ) values and the RI. The RI correlated sig2 ni® cantly with the ICG(15 9 ) values (r 5 0.252, P , 0.005). The mean RI, mean ICG(15 9 ), Child-Pug h classi® cation, and ascites formation for both groups are summarized in Table 1. It shows the patients with higher RI values seem to have more severe hepatic reserve. Plasma Endothelin Concentration and RI. Figure 3 show s plasma endothelin concentration in three groups; normal liver (1.420 6 0.463 pg/ml), cirrhosis
543
KITAMURA ET AL
Fig 2. A statistically signi® cant correlation was found between RI 2 and ICG(15 9 ) values (r 5 0.252, P , 0.005).
with RI less than 0.7 (1.610 6 0.484 pg/ml), and cirrhosis with RI of 0.7 or higher (1.570 6 0.386 pg/ml). Regardless of RI values, plasma endothelin levels in cirrhotic groups were slightly higher compared to the normal liver controls, but the differences did not reach statistical difference. Furthermore, there was no signi® cant difference in the plasma endothelin concentrations between the groups with RI values of more (N 5 10) and less (N 5 10) than 0.7 (Figure 3). In addition, all patients exhibited a plasma endothelin concentration of less than 2.3 pg/ml. DISCUSSION In function al kidney failure associated with liver cirrhosis, several circulatin g substances such as catecholamines, throm boxane, or angioten sin II are thought to play roles in the vasoconstri ction of the microvessels in the kidney. Endothelin is a relatively newly discovered peptide that is known to be the most potent of the vasoconstrictors. In relation to liver disease, this potential has received attention as one of T ABLE 1. C OMPARISON OF M EAN RI, M EAN ICG(15 9 ), C HILD -P UGH C LASSIFICATION , AND A SCITES F ORMATION RI , Number of patients Mean RI Mean ICG(15 9 ) % Child-Pugh A B C Ascites formation
544
0.7
RI .
0.7
10 0.638 6 0.028 23 6 8
10 0.740 6 0.03 34 6 9
4 5 1 1
3 1 6 3
Fig 3. Plasma endothelin concentration in three groups. Cirrhotic patients were divided on the basis of an RI value of 0.7 and were compared with controls with normal liver. There were no signi® cant differences in the plasma endothelin concentrations among the three groups. The horizontal bar represents the upper limit of the normal level (2.3 pg/ml).
the most important contributo rs to function al renal failure. However, it has not been establishe d whether an elevated plasma endothelin level is merely a coincidental event, such as decreased renal disposal, or a counter elevation against visceral vasodilatio n. In fact, decreased plasma levels of endothelin in cirrhotic patients have been reported (6). In the protocols of previous studies, patients with liver diseases were ranked according to clinical classi® cation, such as Child-Pug h, or the presence of overt hepatorenal syndrome, to observe correlation s with the plasma concentration of endothelin. For instance, Schrader et al and Moore et al reported high plasma levels of endothelin in cirrhotic patients with ascites (15, 16). Uchihara et al noted that plasma endothelin concentrations were signi® cantly higher in patients with cirrhosis and ascites than in patients with cirrhosis but no ascites or in normal patients (17). In contrast, Lerman et al reported normal endothelin levels in cirrhotic patients who presented with sodium retention and alteration of the renin± angiotensin system (5). Textor et al stressed the importance of the suppression of renal prostacyclin excretion for renal vaso constrictio n, rather than stimulatio n of thromboxa ne or endothelin (7). Veglio et al reported that plasma concentrations of endothelin were signi® cantly lower in nonuremic cirrhotic patients than in normal subjects (6). These results are con¯ icting, and the causative relationsh ip between endothelin and renal dysfunction are still controversial. Conversely, Asbert et al presented data of a Digestive Diseases and Sciences, Vol. 42, No. 3 (March 1997)
ENDOTHELIN AND RENAL VASOCONSTRICTION
protocol in which sodium intake and the use of diuretics were strictly controlled so that the experimental condition s would be more uniform and reliable (4). They suggested that endothelin is not involved in the pathogenesis of renal vasoconstriction in cirrhosis, since the plasma concentration of endothelin was similar in cirrhotic patients with ascites, with or without renal failure (4). These discussions prompted us to study selected patients based on the evidence of renal vasoconstriction as assessed by an ultrasound Doppler techniqu e. At the same time, patients were assessed by clinical stagin g an d serum labora tory tests. Our results seemed to show that the vasoconstriction of renal vessels occurs with progression of liver disease, which is compatible with the results of previous studies (8, 9) (Figures 1 and 2, and Table 1). However, the elevation of plasma concentrations of endothelin did not occur in parallel manner with the events in the kidney (Figure 3). This may support the idea that circulating endothelin does not re¯ ect renal vasoconstriction, at least in the early stage of the functional kidney failure associated with cirrhosis. A possible explanation of such an idea involves that endothelin can be produced in local blood vessel endothelium, and its effect as a paracrine hormone is dif® cult to assess by means of detectors for vasoconstriction such as Doppler ultrasound . In conclusion , further studies regarding the local productio n of endothelin and its effect on renal vasculature are necessary. The studies should include the use of endothelin antagonist s, and Doppler ultrasound may still be useful for detecting the vascular relaxation when those antagonists are administe red to patients who exhibit low plasma concentrations of endothelin. REFERENCES 1. Loutzenhiser R, Epstein M , Hayashi K, Horton C: Direct visualization of effects of endothelin on the renal microvasculature. Am J Physiol 258:F61± F68, 1990 2. Rabelink TJ, Kaasjager KA, Boer P, Stroes EG, Braam B, Koomans HA: Effects of endothelin-1 on renal function in humans: Implications for physiology and pathophysiology. Kidney Int 46:376 ± 381, 1994
Digestive Diseases and Sciences, Vol. 42, No. 3 (M arch 1997)
3. Moore KW, Wendon J, Frezer M, Karani J, W illiams R, Badr K: Plasma endothelin immunoreactivity in liver disease and the hepatorenal syndrome. N Engl J M ed 327:1774 ± 1778, 1992 4. Asbert M, Gines A, Gines P, Jimenez W , Claria J, Salo J, Arroyo V, Rivera F, Rodes J: Circulating levels of endothelin in cirrhosis. Gastroenterology 104:1485± 1491, 1993 5. Lerman A, Click RL, Narr BJ, Wiesner RH, Krom RA: Elevation of plasma endothelin associated with systemic hypertension in humans following orthotopic liver transplantation. Transplantation 51:646 ± 650, 1991 6. Veglio F, Pinna G, Melchio R, Rabbia F, Panarelli M, Gagliardi B, Chiandussi L: Plasma endothelin levels in cirrhotic subjects. J Hepatol 15:85± 87, 1992 7. Textor SC, Wilson DJ, Lerman A, Romero JC, Burnett JC Jr, Wiesner R, Dickson ER, Krom RA: Renal hemodynamics, urinary eicosanoids, and endothelin after liver transplantation. Transplantation 54:74 ± 80, 1992 8. Platt JF, Ellis JH, Rubin JM , M erion RM , Lucey M R: Renal duplex ultrasonography: A noninvasive predictor of kidney dysfunction and hepatorenal failure in liver disease. Hepatology 20:362± 369, 1994 9. Colli A, Cocciolo M, Riva C, M artinez E: Abnormal renovascular impedance in patients with hepatic cirrhosis: Detection with duplex US. Radiology 187:561± 563, 1993 10. Platt JF, Rubin JM, Ellis JH, DiPietro MA: Duplex Doppler US of the kidney: Differentiation of obstructive from nonobstructive dilatation. Radiology 171:515± 517, 1989 11. Pugh RN, Murray-Lyo n IM , Dawson JL, Pietroni M C, Williams R: Transection of the esophagus for bleeding oesophageal varices. Br J Surg 60:646 ± 649, 1973 12. Kaplowitz N, Eberle D, Yamada T: Biochemical tests for liver disease. In HepatologyÐ A Text Book of Liver Disease. D Zakim, TDB Boyer (eds). Philadelphia, WB Saunders, 1982, pp 583± 612 13. Moody FG, Leyton FR, Joagin SA: Estimation of the functional reserve of human liver. Ann Surg 180:592± 598, 1974 14. Suzuki N, M atsumoto H, Kitada C, M asaki T, Fujino M: A sensitive sandwich-enzyme immunoassay for human endothelin. J Immunol Methods 118:245± 250, 1989 15. Schrader J, Tebbe U, Borries M , Ruschitzka F, Schoel G, Kandt M, Warneke G, Zuchner C, W eber M H, Neu U: Plasma endothelin in normal probands and patients with nephrologicrheumatologic and cardiovascu lar diseases. Klin Wochenschr 68:774 ± 779, 1990 16. Moore KP, W endon J, Frazer M , Gimson AES, Karani J, Williams R, Badr KF: Plasma endothelin (ET) levels in acute and chronic renal failure, chronic liver disease and hepatorenal syndrome (HRS). J Am Soc Nephrol 3:652, 1991 (abstract) 17. Uchihara M, Izumi N, Sato C, M arumo F: Clinical signi® cance of elevated plasma endothelin concentration in patients with cirrhosis. Hepatology 16:95± 99, 1992
545
