Scand. J. din. Lab. Invest. 37, 635-642, 1977.

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/26/14 For personal use only.

Renal blood flow in cirrhosis: relation to systemic and portal haemodynamics and liver function H. RING-LARSEN Division of Hepatology, Medical Department A, Rigshospitalet, University Hospital, Copenhagen, Denmark

Ring-Larsen, H. Renal blood flow in cirrhosis: relation to systemic and portal haemodynamics and liver function. Scand. J . din. Lab. Invest. 37,635-642, 1977. The RBF was measured by means of the lssXe washout method in seventy patients with cirrhosis. The average RBF in controls was 3.72 ml/g.min compared with 2.34 in the patients without ascites, 1.82 in the decompensated patients, 1.47 in the patients with azotaemia and 1.13 in the patients with additional oliguria. The RBF was not significantly correlated to changes in the systemic or portal haemodynamics. Likewise it was not correlated to any biochemical test of liver function except the serum albumin concentration (P 0.1 3 mmol/l). and those with additional oliguria (diuresis < 400 m1/24 h).

PATIENTS Seventy patients with cirrhosis, histologically verified either by liver biopsy or at autopsy, were studied. Forty-four had alcoholic cirrhosis, twenty-one idiopatic cirrhosis, and five primary biliary cirrhosis. Twenty-four of the patients were women at the age of 22-64 years, mean 49 years, and forty-six were men 22-13

years old, mean 49 years. In two patients, thrombotic occlusion of the portal vein and in one, occlusion of the splenic vein was present. In two patients primary carcinoma of the liver was found at autopsy. Patients with stage 111 hepatic coma were not included in the study. Thirty-two patients were without signs of sodium and water retention, twenty had ascites of various degree without sign of impairment of

TABLE I. Haemodynamic variables in twenty patients with cirrhosis

No.

Init.

1 A0 2 MC 3 HN 4 NB 5 EA 6 KE 7 PL 8 KK 9 WH 10 PJ 11 SE 12 KR 13 CS 14 GB 15 JD 16 FM 17 JP 18 JJ 19 FA 20 LH Mean f SEM

Arterial blood pressure Plasma volume Blood volume Cardiac index (mmHg) (ml/kg) (rnl/kg) (ml/rnin/m2) 80 73 84 75 75 80 94 94 95 78 80 108 16 87 82 75 75 99 74 106 84.5 2.5

54.60 50.44 44.81 85.04 43.53 57.52 40.41 46.40 73.40 79.30 72.60 43.35 74.58 50.90 50.52 54.98 68.1 8 58.95 78.68 82.43 60.53 3.30

72.94 71.98 67.01 105.34 60.00 82.31 61.24 65.00 107.90 124.00 86.30 58.89 103.40 72.12 78.31 77.61 96.19 74.55 104.87 124.13 84.71 4.63

Hepatic Renal blood blood flow Row (I/min) (ml/g.min)

3.7 3.6

1.72 0.59

5.4 3.8 3.2 2.6 4.0 3.0 3.1 2.7 2.0 2.4 3.8

0.85

-

-

3.6 3.5 2.6 3.3 2.1 3.2 0.18

-

1.23

-

1.63 0.57 1.01 1.10 0.89 1.43

-

1.16 2.21

-

1.41 1.22 0.13

3.84 3.80 2.98 2.86 2.61 2.47 2.29 2.29 1.70 1.66 1.65 1.64 1.64 1.51 1.50 1.49 1.36 1.32 1.18 0.74 2.03 0.19

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/26/14 For personal use only.

Renal bloodflow in cirrhosis

renal function, thirteen had ascites and azotaemia (serum creatinine z 0.13 mmol/l) and five had ascites, azotaemia and additional oliguria (diuresis < 400 m1/24 h). Of the twenty patients with ascites, all but six received a sodium restricted diet of 10 mEq/24 h and diuretic therapy of maximum 320 mg furosemide and 150 mg spironolactone daily. Of the thirtytwo non-ascitic patients, eight had previously had ascites and received a maximum of 80 mg furosemide daily to prevent recurrence of sodium and water retention. During hospitalization this treatment could be discontinued in six of these patients. Previous renal or cardiovascular disease was excluded by history, physical examination and laboratory findings. Minor changes at post-mortem microscopical examination of the kidneys, mainly arteriosclerotic changes, and in a few cases thickening of the glomerular basal membrane (in one case with IgM deposit), were present in some patients, but in no case were glomerular changes or tubular necrosis a dominant finding.

METHODS The following laboratory tests of liver function were performed: serum bilirubin, serum albumin, serum alanine aminotransferase, serum alkaline phosphatase and prothrombin time. Renal function and electrolyte status was estimated by serum creatinine, diuresis and serum and urine sodium and potassium. After fasting for 8 h, which included omission of diuretics, the patients were premedicated with 10 mg diazepam 1 h before the procedure; a catheterization of a renal artery (Seldinger technique) was then performed and the RBF measured by the 133Xe wash-out method according to Ladefoged [22] as previously described [38]. A semilogarithmic plot of the washout curves was used in calculating the mean RBF, as an average of two measurements. In the same session catheterization of the hepatic veins was carried out in forty-eight of the patients (twenty-five, sixteen, four and three in the respective categories). Under local anaesthesia a catheter, Cournand No. 5-8, was inserted through a cubital vein. The following pressures were measured : wedged hepatic venous pressure, hepatic venous pressure, inferior caval venous pressure and right arterial F

637

pressure. The pressures were measured with an Elema-Schonander EMT-35 pressure transducer, and registered at an Elema-Schonander Mingograph-81. In nineteen patients (thirteen, five, one, and nil in the respective categories), while the venous catheter was in the pulmonary artery, pressure was recorded and the cardiac output measured according to the principle of Fick. In twenty patients (fourteen, five, one, and nil in the respective categories) the plasma and blood volume were measured with the EvansBlue technique and 51-Cr tagged erythrocytes according to the method A of 1971 [15]. The total hepatic blood flow was measured in thirteen patients (eleven, two, nil, and nil in the respective categories) by the indocyanine-green extraction method [48], while the catheter was in the hepatic veins. The galactose elimination capacity was estimated in forty-six patients (twenty-three, fourteen, six and three in the respective categories) after intravenous injection of 0.5 g of galactose per kg body weight in the course of 5 min [44]. After 15 min for equilibration the arterial concentration of galactose was measured every 5 min for 40 min. In fortyseven patients (twenty-five, fourteen, five, and two in the respective categories) a splenoportogram with measurement of the intrasplenic pressure was obtained. Unless otherwise stated, differences between RBF in the groups of patients were estimated by means of a rank sum test (Mann-Whitney). Correlations between RBF and other variables were estimated according to Spearman RHO test. The k indicates in all cases 1 SEM. RBF measurements for comparison was carried out in ten patients, five males and five females, 23-44 years of age, mean 35 years, undergoing cardiac catheterization because of suspected but unconfirmed cardiac disease. The controls did not receive sodium restriction or diuretics. The difference in mean age between the cirrhotic patients and the controls was taken into consideration in comparison of RBF [14]. In most cases the procedure was an integral part of the haemodynamic evaluation of the patients, done in the same session as the liver vein catheterization or at the time of coeliac-axis angiography. The right sided cardiac catheterization, donein nineteen patients, was part of the haemodynamic evaluation done immediately

638

H . Ring-Larsen

before portal-systemic shunt operation. In every instance details of the procedure as well as lack of therapeutical benefit from the investigation was explained before consent was obtained. Patients with impaired sensorium did not enter the study.

l/minm*), which is within normal limits for this age group. No significant correlation between cardiac output and RBF was observed. The inferior caval vein pressure was above normal, on the average 7.9 f 0.6 mmHg, but not significantly correlated to the RBF. Portal circulation

Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/26/14 For personal use only.

RESULTS Renal circulation

The mean RBF is shown in Fig. 1 . The average renal perfusion in thecontrols was 3.72 k 0.25 ml/g.min, and the values for the cirrhotic patients were 2.34 k 0.13,1.82*0.14,1.47 k0.18, and 1.13k0.23 ml/gmin in the respective categories. While the RBF was subnormal even in patients with compensated cirrhosis, the age difference between the patients and the controls taken into consideration, (P0.05). Of the twenty patients with ascites fourteen received diuretics and a sodium restricted diet. The average RBF in these patients was 1.83k0.17 ml/g.min compared with 1.81 f0.24 ml/g.min in the patients not receiving this treatment ( P r 0.05).

Intra-splenic pressure was on the average 26.5 k 1 mmHg, wedged hepatic venous pressure 23 f 1 mmHg and the pressure gradient across the liver 13 k0.75 mmHg. Significant correlation between any of these pressure variables and RBF was not demonstrated. This was also the case when only patients with good agreement between intra-splenic pressure and wedged hepatic venous pressure were considered, or when only cases with pathologically elevated pressures were included. In seven patients normal pressures (i.e. intrasplenic pressure 5 15 mmHg, wedged hepatic venous pressure 5 14 mmHg and pressure gradient 1 5 mmHg) were recorded. The RBF in these cases was 2.18 k 0.32 ml/gmin k 0.19 which was significantly less than in the controls (P i0.01). As evident from Table I the total hepatic blood flow was on the average 1.22 I/min (18.3 ml/kg.min). The correlation between this parameter and RBF was not significant. Hepatic function

The RBF was correlated to the serum albumin

Systemic circulation

The mean arterial blood pressure was 81 f2.1 mmHg in the cirrhotic patients, which was less than the predicted 98 mmHg for this age group [9]. No significant correlation between arterial blood pressure and RBF was observed. Some haemodynamic variables of twenty patients (fourteen, five, one, and nil of the respective categories) are shown in Table I. Mean plasma volume and blood volume were 60.5 and 84.7 ml/kg which was significantly larger than those predicted from sex, age and weight [9] (41.6 and 70.6 ml/kg, respectively, P

Renal blood flow in cirrhosis: relation to systemic and portal haemodynamics and liver function.

Scand. J. din. Lab. Invest. 37, 635-642, 1977. Scand J Clin Lab Invest Downloaded from informahealthcare.com by Chulalongkorn University on 12/26/14...
594KB Sizes 0 Downloads 0 Views