Dr Yoshiyuki Narahara, Department of Internal Medicine, Division of Gastroenterology, Nippon Medical School, 1-1-5, Sendagi, Bunkyo-ku, Tokyo 113-8603, Japan. Email: firstname.lastname@example.org
Background and Aim: Terlipressin has been shown to be effective in the management of hepatorenal syndrome. However, how terlipressin exerts its effect on the renal artery is unknown. The aim of the present study was to assess the effects of terlipressin on systemic, hepatic and renal hemodynamics in cirrhosis.
Methods: Twenty-eight patients with cirrhosis and portal hypertension were studied. Systemic and hepatic hemodynamics, hepatic and renal arterial resistive indices and neurohumoral factors were measured prior to and 30 min after intravenous administration of 1 mg terlipressin (n = 19) or placebo (n = 9).
Results: After terlipressin, there were significant increases in both mean arterial pressure (P < 0.001) and systemic vascular resistance (P < 0.001), whereas heart rate (P < 0.001) and cardiac output (P < 0.001) decreased significantly. There was a significant decrease in the hepatic venous pressure gradient (P < 0.001). Portal venous blood flow also decreased significantly (P < 0.001). The mean hepatic arterial velocity increased significantly (P < 0.001). Although there was a significant decrease in the hepatic arterial resistive index (0.72 ± 0.08 to 0.69 ± 0.08, P < 0.001) and renal arterial resistive index (0.74 ± 0.07 to 0.68 ± 0.07, P < 0.001), portal vascular resistance was unchanged (P = 0.231). Plasma renin activity decreased significantly (P < 0.005), and there was a significant correlation between this decline and the decrease in renal arterial resistive index (r = 0.764, P < 0.005). The effects of terlipressin on systemic, hepatic and renal hemodynamics were observed similarly in patients with and without ascites. Placebo caused no significant effects.
Conclusion: Terlipressin decreases hepatic and renal arterial resistance in patients with cirrhosis.
Terlipressin (triglycyl-lysine-vasopressin) is a vasopressin analogue that has been shown to be effective in the management of acute variceal hemorrhage.1 In addition, recent trials indicate that terlipressin improves renal function in patients with hepatorenal syndrome.2–7 The effect probably is mediated by an arteriolar vasoconstriction in the splanchnic area with decreased portal pressure as well as by improving renal blood flow following redistribution of effective arterial blood volume. Some attempts have been made to investigate the effects of terlipressin on systemic, portal and renal hemodynamics in patients with cirrhosis.8–12 However, how terlipressin exerts its effects on the hepatic and renal arteries is as yet unknown.
Doppler ultrasonography is useful for non-invasive assessment of the hepatic and renal arterial resistance indices,13,14 because these can be calculated from the systolic and diastolic velocities and are independent of the angle between the vessel and the axis of the ultrasound beam. In several studies, it has been reported that hepatic arterial resistance evaluated by Doppler ultrasonography was significantly higher for cirrhotic than for healthy subjects13,15 and that infusion of vasopressin decreased hepatic arterial resistance.15 In addition, it has been demonstrated that renal arterial resistance evaluated by Doppler ultrasonography was already increased in cirrhotic patients without ascites.14 Doppler hemodynamic evaluation of hepatic and renal arterial resistance may be useful for both pathophysiological and clinical studies in cirrhosis.
The aim of the present study was to investigate using Doppler ultrasonography the effects of terlipressin on the hepatic and renal arteries in patients with cirrhosis.
Between January 2006 and June 2007, 34 patients with cirrhosis and portal hypertension were considered for inclusion in this study. Inclusion criteria for this study were: (i) presence of a portal vein with a straight section for at least 3 cm; (ii) presence of angles between the Doppler beam and the longitudinal axis of the portal vein and the right hepatic artery less than 60°; (iii) presence of high-quality Doppler signals of the right hepatic artery and the interlobar artery of the kidney; and (iv) cooperation of the patients. Exclusion criteria were (i) hepatocellular carcinoma; (ii) cardiopulmonary disease; (iii) hypertension; (iv) coronary and/or peripheral arterial disease; and (v) renal disease (proteinuria and ultrasonographic evidence of obstructive uropathy or parenchymal renal disease). Of the 34 patients, six were excluded because of hepatocellular carcinoma (n = 3), coronary disease (n = 1), or renal disease (n = 2). Therefore, 28 patients were included in the study. A computer made the randomization code with 30 envelopes, with one-third for the placebo and two-thirds for the terlipressin, and 28 were used. Among the 28 patients, 19 were assigned to receive terlipressin and nine were assigned to receive placebo.
The diagnosis of cirrhosis was made on the basis of laboratory and ultrasonographic findings or transjugular liver biopsy. Clinical characteristics of the patients in this study are summarized in Table 1. There were no significant differences between patients receiving terlipressin and placebo in relation to background factors, baseline hemodynamic parameters and neurohumoral factors (Tables 1–4). Ten patients (six terlipressin and four placebo) had undergone treatment of esophageal varices previously. None of the patients received vasoactive drugs, which were beta-blockers, nitrates, calcium-channel blockers, angiotensin II type 1-receptor blockers, or angiotensin-converting enzyme inhibitors. In the terlipressin group, 12 patients had moderate to severe ascites and 13 were receiving diuretics (13 patients furosemide, mean 45 mg/day, and 12 patients spironolactone, mean 58 mg/day), whereas in placebo group, six patients had moderate to severe ascites and seven were receiving diuretics (seven patients furosemide, mean 51 mg/day, and six patients spironolactone, mean 64 mg/day). Although the Child–Pugh score in the terlipressin group was significantly higher in patients with ascites than in those without ascites (P < 0.01), the other background factors in terlipressin and placebo groups were not significantly different between patients with and without ascites. All patients with ascites responded to treatment. There were no patients with hepatorenal syndrome. None of the patients had undergone a paracentesis within the last week prior to the study.
Table 1. Clinical characteristics of the patients
Data are given as mean ± SD.
61.4 ± 11.5
63.7 ± 11.1
8.1 ± 1.9
8.2 ± 1.9
Serum total bilirubin (mg/dL)
1.5 ± 0.6
1.3 ± 0.9
Serum albumin (g/dL)
3.0 ± 0.6
3.0 ± 0.6
0.92 ± 0.40
1.01 ± 0.42
Urinary sodium excretion (mEq/day)
64.0 ± 48.7
67.8 ± 38.7
Table 2. Systemic hemodynamics at baseline and 30 min after giving terlipressin and placebo
This study was approved by the Nippon Medical School Ethics Committee and all patients provided written informed consent.
The baseline data of mean arterial pressure (MAP), heart rate (HR), cardiac output (CO), systemic vascular resistance (SVR), pulmonary artery pressure (PAP), pulmonary capillary wedged pressure (PCWP), right atrial pressure (RAP) and hepatic venous pressure gradient (HVPG) were measured by means of a catheterization of the right heart and hepatic vein.
The Doppler hemodynamic studies were carried out at the same time. Doppler indices included the blood flow of the portal vein, the mean velocity of the hepatic artery and the resistive index of the hepatic and renal arteries. One investigator carried out all ultrasound measurements. Systemic hemodynamic indices, HVPG and Doppler indices were estimated from the average of three consecutive measurements.
Measurements of plasma renin activity (PRA), plasma aldosterone concentration (PAC), plasma norepinephrine (NE) concentration, plasma atrial natriuretic peptide (ANP) concentration, and plasma arginine vasopressin (AVP) concentration were carried out at baseline. Diuretics were discontinued only on the day of the investigations. Neurohumoral factors were assessed in 12 patients (four had no ascites and eight had ascites), but were not measured in seven patients in the terlipressin group, whereas all placebo patients assessed neurohumoral factors.
After measurements of baseline data, 1 mg intravenous injection of terlipressin (Glypressin; Ferring AB, Limhamn, Sweden) or placebo (5 mL saline) was given and the measurements were repeated 30 min later.
After an overnight fast, the right jugular vein was cannulated and a Swan-Ganz catheter (7 F Thermodilution Catheter; Biosensors International Co. Ltd, Tokyo, Japan) was placed in the pulmonary artery for measurement of CO, PAP, PCWP and RAP. An automatic blood pressure instrument was used to non-invasively monitor MAP and HR. SVR was calculated as (MAP − RAP/CO) × 80. Subsequent to this, a 5 F balloon-tipped catheter (Clinical Supply Co. Ltd, Gifu, Japan) was inserted into the right hepatic vein for measurement of wedged and free hepatic venous pressure. The HVPG was calculated as wedged hepatic venous pressure—free hepatic venous pressure.
Doppler hemodynamic measurements
Ultrasonographic examination was carried out at rest in the supine position. A Doppler ultrasonography scanner (Power Vision 6000; Toshiba, Tokyo, Japan) equipped with a single convex multifrequency electronic probe was used in B mode with a pulsed system (3.75 MHz).
The portal vein was scanned longitudinally and the Doppler sample volume was then set at the center of the lumen, 1–2 cm before the bifurcation of the portal vein. Cross-sectional area and mean velocity measurements were obtained. The blood flow was calculated as cross-sectional area × mean velocity × 60. Cross-sectional area was calculated as π × r2, where r is the radius. Portal vascular resistance (PVR) was calculated as HVPG/portal venous blood flow (PVBF).
Under the right intercostal scanning of the liver, the right hepatic artery, where it crossed the portal vein around the porta hepatis was identified using color Doppler ultrasonography. After the Doppler sample volume was located in the right hepatic artery, the time velocity waveform of the Doppler signal was recorded. Peak systolic, end diastolic and mean velocity were measured, and the hepatic arterial resistive index ([peak systolic velocity—end diastolic velocity]/peak systolic velocity) was determined.13 In addition, the resistive index of the renal artery was determined at the interlobar artery of the kidney.14
Intraobserver variability for measuring the hepatic and renal arterial resistive indices was tested in 10 preliminary patients. The variation was less than 5%.
Assays of neurohumoral factors
Plasma renin activity and PAC were measured by radioimmunoassay (TFB, Inc., Tokyo, Japan). Normal values ranged from 0.3 to 2.9 ng/mL per h for PRA and from 29.9 to 159.0 pg/mL for PAC. Plasma NE concentrations were determined by fluorescent detection after separation by high performance liquid chromatography (Tosoh Co., Tokyo, Japan). Normal values of NE concentrations were from 100 to 450 pg/mL. Plasma ANP concentrations and plasma AVP concentrations were measured by radioimmunoassay (Shionogi & Co. Ltd, Osaka, Japan and Mitsubishi Kagaku Iatron, Inc., Tokyo, Japan). Normal values of ANP concentrations were less than 40 pg/mL and normal values of AVP concentrations were from 0.3 to 3.5 pg/mL.
All results are expressed as mean ± standard deviation. Comparisons between terlipressin and placebo groups were carried out using the χ2-test with Yates' correction or Fisher's exact test and the Mann-Whitney U-test. The Wilcoxon test or paired t-test was used to compare differences within the patient groups. Comparisons between patients with and without ascites were carried out using the χ2-test with Yates' correction or Fisher's exact test and the Mann-Whitney U-test or unpaired t-test. Correlation was analyzed by the Spearman rank correlation test. A two-tailed P value of 0.05 was considered statistically significant.
Changes in systemic hemodynamics with terlipressin
Table 2 shows the results of systemic hemodynamic changes. Following terlipressin administration, MAP increased by 17% (P < 0.001) and SVR by 36% (P < 0.001). HR and CO decreased 11% (P < 0.001) and 12% (P < 0.001), respectively. PAP, PCWP and RAP increased after terlipressin infusion (P < 0.001). Systemic hemodynamic indices at baseline were not significantly different between patients with and without ascites. There were no significant differences between patients with and without ascites in the changes in MAP, HR, SVR, PAP, PCWP and RAP, but the changes in CO were significantly lower in patients with ascites (P < 0.005).
Changes in hepatic hemodynamics with terlipressin
Table 3 shows the results of hepatic hemodynamic changes. After terlipressin, HVPG and PVBF decreased 15% (P < 0.001) and 30% (P < 0.001), respectively. PVR was unchanged (P = 0.231). The mean hepatic arterial velocity (HAV) increased 31% (P < 0.001). HVPG, PVBF, PVR and HAV at baseline were not significantly different between patients with and without ascites. There were no significant differences between patients with and without ascites in the changes in HVPG, PVBF, PVR and HAV.
Changes in hepatic and renal arterial resistive indices with terlipressin
The changes in hepatic and renal arterial resistive indices with terlipressin in the individual patients are shown in Fig. 1. There were no significant differences between alcoholic cirrhosis and viral cirrhosis in the changes in hepatic and renal arterial resistive indices. There was a slight but significant decrease in hepatic arterial resistive index (HA-RI) (−4%, P < 0.001) after infusion of terlipressin (Table 3). HA-RI at baseline was not significantly different between patients with and without ascites. The changes in HA-RI were significantly lower in patients with ascites than in those without ascites (P < 0.05). Four of the 12 patients with ascites had no decrease in HA-RI, whereas seven patients without ascites decreased in HA-RI. Renal arterial resistive index (RA-RI) decreased significantly (−8%, P < 0.001) after terlipressin (Table 3). RA-RI at baseline was not significantly different between patients with and without ascites. No difference in the changes in RA-RI was found between patients with and without ascites.
Changes in neurohumoral factors with terlipressin
Plasma renin activity, PAC and NE decreased significantly (P < 0.005, P < 0.05 and P < 0.005, respectively), and AVP increased significantly (P < 0.005) after terlipressin. ANP was unchanged (P = 0.285) (Table 4). There was a significant correlation between the change in PRA and the change in RA-RI (r = 0.764, P < 0.005) (Fig. 2). However, there were no correlations between the change in PAC or NE and the change in RA-RI. PRA, PAC, NE and AVP at baseline were not significantly different between patients with and without ascites. ANP at baseline was significantly lower in patients with ascites than in those without ascites (P < 0.05). There were no significant differences between patients with and without ascites in the changes in neurohumoral factors.
Effects of giving a placebo
As shown in Tables 2–4, giving a placebo had no significant effects on systemic and hepatic hemodynamics, hepatic and renal arterial resistive indices and neurohumoral factors. These parameters at baseline and the changes in those were not significantly different between patients with and without ascites.
Four patients had diarrhea and five patients presented with abdominal pain after terlipressin administration, both symptoms resolved without treatments. No adverse effects related to placebo were seen.
Terlipressin, an agonist of the V1 vasopressin receptors, is inactive in its native form, but is transformed into the biologically active form, lysine-vasopressin, through enzymatic cleavage of glycyl residues by tissue peptidases.16 Because of this modification, terlipressin has a prolonged biological half-life compared with other vasopressin analogues (such as ornipressin).17 In addition, Escorsell et al. demonstrated that the peak reduction in HVPG was achieved at 30 min after giving 1 mg terlipressin.11 Furthermore, Hansen et al. showed that 1 mg terlipressin produced a significant reduction in portal venous blood flow at 30 min in healthy pigs.18 Therefore, the hemodynamic measurements in the present study were carried out at 30 min.
Previous studies demonstrated that the injection of 2 mg terlipressin was effective on systemic and splanchnic hemodynamics in patients with cirrhosis;8,9,12 however, Escorsell et al. reported that giving 1 mg terlipressin was effective in such patients.11 Intravenous injection of 1 mg terlipressin is the recommended dose for the treatment of hepatorenal syndrome and 1 mg terlipressin every 4–6 h improves renal function in patients with hepatorenal syndrome.19 Therefore, 1 mg terlipressin was given in this study.
The patient population in the present study had heterogeneity with respect to the severity of cirrhosis as well as to previous studies.8–12 Ten patients had no ascites and 18 had ascites. Although all had portal hypertension, the different clinical stages reflect a difference in circulatory, hormonal and renal function. This may significantly affect the results. However, no study has investigated that the effects of terlipressin on systemic, hepatic and renal hemodynamics are different between different clinical stages in cirrhotic patients. Therefore, we also evaluated differences between cirrhotic patients with and without ascites in all measurements.
In our study, terlipressin increased MAP and SVR, and decreased HR and CO, which are consistent with the systemic hemodynamic effects described previously.8–12 Patients with cirrhosis and portal hypertension exhibit a hyperdynamic circulation with increased HR and CO and decreased MAP and SVR.20 The changes may be brought about by a peripheral arterial vasodilatation, which appears to be most excessive in the splanchnic area.21 According to the peripheral arterial vasodilation hypothesis,21 splanchnic arterial vasodilatation produced by vasodilators decreased effective arterial blood volume. Indeed, it has been demonstrated that patients with cirrhosis and portal hypertension have increased blood pooling in the splanchnic area and a reduction of the central blood volume.22,23 V1 receptors are involved in smooth muscle contraction and are particularly abundant in the splanchnic area.24 Thus, it was postulated that terlipressin attenuated the hyperdynamic circulation due to increased effective arterial blood volume following arteriolar vasoconstriction in the splanchnic area.
The present study also showed that terlipressin decreased renal arterial resistance. The decrease in renal arterial resistance can be explained largely by the increase in MAP in response to terlipressin. In our study, MAP increased by 17%, which might have a significant impact on the renal circulation. Thus, it was postulated that the systemic vasoconstriction produced by terlipressin increased MAP, which improved renal perfusion pressure and reduced renin release and sympathetic nervous system activity, as indicated by the decrease in PRA, PAC and NE. This reduction in vasoconstrictors could decrease renal arterial resistance and increase renal blood flow (RBF). According to research carried out on decompensated cirrhotic patients, a constant 4-h infusion of ornipressin results in an increase in RBF and a decrease in renal vascular resistance.25 Ornipressin, like terlipressin, appears to reverse splanchnic vasodilatation without increasing renal vascular resistance because of the preferential distribution of V1 receptors in the splanchnic area. Gadano et al. recently reported that terlipressin increased RBF in patients with cirrhosis and refractory ascites.10 Although we did not measure RBF, we measured RA-RI using Doppler ultrasonography. The decreases in renal arterial resistance and PRA in the present study suggest that renal arterial blood flow is increased.
In comparing cirrhotic patients with and without ascites, RA-RI, PRA, PAC and NE at baseline were not significantly different and the changes in these parameters were not different between both groups. These results indicate that terlipressin decreases renal arterial resistance in cirrhotic patients with and without ascites.
In the present study, PVBF and hepatic arterial resistance decreased and HAV increased after giving terlipressin. Decreased hepatic arterial resistance and increased HAV suggest that hepatic arterial blood flow is increased. Thus, the reduction in PVBF induced by terlipressin may be compensated by increased hepatic arterial blood flow. Iwao et al. showed that vasopressin infusion decreased PVBF and hepatic arterial resistance in cirrhotic patients.15 Several studies have shown that, when PVBF decreases, hepatic arterial blood flow increases, and vice versa. This phenomenon has been described as the hepatic arterial buffer response (HABR) and is an intrinsic regulatory mechanism of the liver to maintain total hepatic blood flow.26–28 Terlipressin infusion had a marked systemic hemodynamic effect manifested by a significant increase in MAP and reflex bradycardia in cirrhotic patients. These effects may alter hepatic artery hemodynamic response because of the passive effect of the increased arterial blood pressure and hepatic artery autoregulation in response to increased arterial blood pressure.27
The changes in HA-RI were significantly lower in cirrhotic patients with ascites than in those without ascites. The reason terlipressin infusion dose not decrease HA-RI in some cirrhotic patients with ascites is not clear. One reason may be individual differences of HABR. Some studies indicated that the HABR was blunted in some cirrhotic patients.15,28 It has been suggested that the HABR is mediated by adenosine, a potent vasodilating substance.29 Blunted HABR in cirrhotic patients may be explained by hyposensitivity of adenosine receptors of the hepatic artery.30
Four patients in the present study had diarrhea and five had abdominal pain after giving terlipressin. Other side-effects reported in other studies in patients treated with terlipressin, such as ischemic heart disease, arrhythmia, arterial hypertension, peripheral ischemia or bronchospasm,2–6 were not observed in the current investigation.
In conclusion, terlipressin decreases renal arterial resistance and PRA, PAC and NE. Decreases in renal arterial resistance and PRA suggest that renal arterial blood flow is increased. Terlipressin also decreases hepatic arterial resistance. Decreased hepatic arterial resistance may be a compensatory mechanism to maintain total hepatic blood flow. The benefit of terlipressin for patients with hepatorenal syndrome may be achieved by the changes observed. In addition, further study should be done to clarify the effects of terlipressin on systemic, hepatic and renal hemodynamics in patients with hepatorenal syndrome.