- Email: [email protected]
Renal dysfunction in liver cirrhosis: Renal duplex doppler US vs. Scintigraphy for early identification
Clinical Radiology (1998) 53, 44-48
Renal Dysfunction in Liver Cirrhosis: Renal Duplex Doppler US vs. Scintigraphy for Early Identification E. A. R. AL-KAREEMY, M. A. SOBH, A. M. MUHAMMAD*, M. M. MOSTAFA* and R. A. SABERt
Departments of Internal Medicine, *Radiology and tNuclear Medicine, Faculty of Medicine, Assiut University, Egypt A diagnostic tool to detect early renal dysfunction before it becomes irreversible would be useful in cirrhosis. This study was carried out to evaluate the role of Doppler sonography and Tc-99m DTPA renography in the detection of early renal dysfunction in patients with different grades of liver cirrhosis. Renal arteries of 43 patients with cirrhosis and normal renal function tests were compared with 15 age and gender matched normal subjects as a control group using colour Doppler sonography and Tc-99m DTPA scintigraphy. The patients were categorized into three groups, A (14), B (14) and C (15), according to a modified Child's classification that assesses the severity of liver cirrhosis. Doppler results revealed a highly significant increase in both the pulsatility and resistive indices in groups B and C compared with either group A patients or control subjects and in group C compared with group B (P < 0.001) in the main renal arteries as well as in the interlobar and arcuate arteries. Insignificant differences were observed between group A and controls (PI: control 0.96 _+ 0.08, group A 0.95 _+ 0.07, group B 1.26 _+ 0.06, group C 1.48 - 0.06; RI: control 0.57 _+ 0.02, group A 0.58 _+ 0.02, group B 0.66 _+ 0.01, group C 0.72 _+ 0.02). Abnormal renograms in the form of delayed appearance (34 _+ 14.6 s), diminished blood flow bilaterally with prolonged secretory ( 1 2 _ 4.5 min) and excretory phases (>30 min) and poor response to intravenous frusemide were only observed in group C patients. Radionuclide computed glomerular filtration rate was within the normal range in patients of group A (81 _+ 9.5 ml/min) and group B (78 _+ 8.4 ml/min) and reduced only in patients of group C (34 _+ 14.5 ml/min). Thus Doppler sonography can detect an increase in renal vascular resistance in patients with moderately severe cirrhosis (Child grade B) when renography was normal. We conclude that Doppler sonography can be used for earlier identification of cirrhotic patients with a higher risk of impending renal failure earlier than renography and may also be used to guide therapeutic approaches. A1-Kareemy, E.A.R., Sobh, M.A., Muhammad, A.M., Mostafa, M.M. & Saber, R.A. (1998). Clinical Radiology 53, 44-48. Renal Dysfunction in Liver Cirrhosis: Renal Duplex Doppler US vs. Scintigraphy for Early Identification
Accepted for Publication 6 March 1997
Renal dysfunction may occur in patients with liver cirrhosis without other causative factors [1-3]. Early clinical detection of renal dysfunction with conventional tests is difficult because the serum creatinine level does not increase until late in the course of the disease [4]. An early change that does occur in these patients is intense renal cortical vasoconstriction with resultant elevation of renal arterial resistance [5]. A significant reduction in both mean renal and cortical blood flow in patients with advanced liver cirrhosis has been demonstrated using the 133-xenon washout technique and with selective renal arteriography [6]. Recently Platt et al. studied renal blood flow using Doppler sonography in cirrhotic patients with and without ascites and reported elevation of renal arterial resistance [2]. Halkar et al. used Tc-99m DTPA renography to study renal blood flow in advanced liver cirrhosis where uptake was symmetrically and bilaterally delayed and diminished [7]. An accurate diagnostic test to detect renal dysfunction before it becomes irreversible could be useful in cirrhotic patients. This study was conducted to evaluate the role of Doppler sonography and Tc-99m DTPA renography in the detection of early renal dysfunction in patients with Correspondence to: Dr E. A. R. A1-Kareemy, Assistant Professor of Internal Medicine, Assiut University Hospital, Assiut, Egypt. 9 1998 The Royal College of Radiologists.
different grades of liver cirrhosis who have normal laboratory tests for kidney function.
PATIENTS AND METHODS Patients Fifty-eight subjects were included in this prospective study: 43 with cirrhosis (27 males and 16 females, mean age 42.9---12.7 years) and 15 healthy volunteers (eight males and seven females, mean age 43.2-+ 6.8 years) from the medical personnel in Assiut University Hospital consented as a control group. Patients were selected from the in-patient section of the Internal Medicine Department in our hospital during the period from April 1994 to July 1995. Full clinical evaluation included a thorough history, with particular emphasis on jaundice, bilharziasis, blood transfusion, drug intake and renal diseases, and complete physical examination. Urine analysis, liver and kidney function tests and abdominal ultrasonographic examination were performed. Patients with a past history of renal disease and those with abnormal renal function tests were excluded. All patients had normal blood urea and serum creatinine levels. All of them stopped medications, especially diuretics, 72 h
45
RENAL ULTRASOUND AND SCINTIGRAPHY IN CIRRHOSIS
Table 1 - Modified Child's classification
Grade
1 2 3 4 5
Encephalopathy Ascites Bilirubin (/xmol/1) A l b u m i n (g/l) P r o t h r o m b i n ratio
1
2
3
None None < 35 > 35 < 1.4
Minimal Slight 35-50 28-35 1.4-2
Marked Moderate > 50 < 28 >2
Score A = 5 - 6 points; B = 7 - 9 points; C = 1 0 - 1 5 points.
before the Doppler and scintigraphy studies. The degree of cirrhosis was evaluated according to a modified Child's classification [8]. Features such as the presence or absence of ascites or encephalopathy, serum albumin level, serum bilirubin level and prothrombin ratio categorize the patients into three grades: grade A (5-6 points), grade B (7-9 points) and grade C (10-15 points) (Table 1).
supine or decubitus position depending on which gives the best view of the course of the renal artery. Doppler signals were obtained from the main renal artery at the hilum of the kidney, from the interlobar arteries along the borders of the medullary pyramids and from the arcuate arteries at the corticomedullary junction. To decrease measurement errors, we maximized the size of the Doppler spectrum by using the lowest frequency range possible without causing aliasing and the lowest wall filter (100 Hz). Doppler spectral analysis included measurement of peak systolic velocity, resistive index (RI) and pulsatility index (PI) for at least three vessels in each kidney. We measured the RI and PI for every vessel from at least three wave forms and calculated the average value. A mean RI and PI as the average of both kidneys was calculated for each patient. RI was defined as (peak systolic velocity-minimum diastolic velocity)/ peak systolic velocity and PI was defined as (peak systolic velocity-minimum diastolic vetocity)/mean velocity. Mean RI and PI (• SD) were calculated for each group. A comparison among groups was performed with analysis of variance and the Student t-test with a P-value <0.05 considered significant.
Group A This group was composed of 14 patients (nine males and five females, mean age 42.6 _+ 12.7 years). All had compensated liver cirrhosis, eight post-hepatitic, four bilharzial and two mixed bilharzial and post-hepatitic aetiology. None had hepatic encephalopathy or ascites. Their biochemical results were: serum bilirubin 14.46 _+ 8.78 #mol/1, serum albumin 37 _+ 4.98 g/1 and prothrombin ratio < 1.4 (Child grade A).
Isotope Scanning Using Tc-99m DTPA
Fourteen patients (nine males and five females, mean age 42.2 _+ 13.5 years) of whom three had post-hepatitic cirrhosis, 10 bilharziasis and one cirrhosis of mixed aetiology. Twelve patients had splenomegaly and all had mild ascites. None had hepatic encephalopathy. Their biochemical results were: bilirubin 20.75 + 12.56/~mol/1, albumin 36.57 -4- 6.64 g/1 and prothrombin ratio range 1.4-2 (Child grade B).
Renal scintigraphy was performed using Tc-99m DTPA for 30 patients, 10 patients from each group (A, B and C). After an overnight fast, the patient was allowed to drink and positioned lying supine over a gamma camera that was centred from the xiphisternum to the symphysis pubis. 5 mCi (185 MBq) of Tc-99 m DTPA was injected intravenously as a bolus followed by single-window acquisition of renal blood flow and sequential 4 s dynamic images in the posterior projection using a parallel hole medium energy collimator. Acquisition of the data was performed in frame mode for 30 min. At 20 min, 40 mg Furosemide was given I.V. and sequential dynamic images were obtained until the end of the study. Time-activity curves were recorded over the kidneys and the individual GFR was measured from backgroundsubtracted images of the kidneys between 80 and 180 s after injection (method of Gates) [9].
Group C
Statistical Methods
Fifteen patients (nine males and six females, mean age 43.1 _+ 15.4 years) were classified as Child grade C. All had decompensated liver cirrhosis, six post-hepatitic, two bilharzial and seven mixed bilharzial and post-hepatitic; all had manifestations of hepatocellular failure: 12 had jaundice, nine had encephalopathy and 15 had ascites. Thirteen patients had splenomegaly and 10 patients had portal hypertension and haematemesis. Their biochemical results were: bilirubin 58.72• albumin 29.53 • 4.14 g/1 and prothrombin ratio > 2.
Comparison among groups was performed with analysis of variance (mean + SD) and Student t-test with P-value < 0.05 considered significant.
Index
Control (n = 15)
Group A (n = 14)
Group B (n = 14)
Group C (n = 15)
Colour Coded Doppler Ultrasonographic Examination
PI
0.96 + 0.08
0.95 -4- 0.07 NS*
1.26 -+ 0.06 P<0.001* P<0.001**
RI
0.57 + 0.02
0.58 -+ 0.02 NS*
0.66 -+ 0.01 P<0.001* P<0.001**
1.48 -+ 0.06 P<0.001* P<0.001* P<0.001*** 0.72 -+ 0.02 P<0.001* P<0.001** P<0.00I***
Group B
Colour coded Doppler ultrasonographic examinations were performed with an Acuson computed sonography 128 XP/10 (Mountain View, USA) with a 3.5 MHz vector array transducer. The main renal, interlobar and arcuate arteries of both kidneys were assessed, including at least three renal vessels in each kidney, recorded according to the method described by Platt et al. [2]. The examinations were performed with the patient in the 9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
Table 2 - Pulsatility index (PI) and resistive index (RI) of both kidneys (mean _+ SD) in 15 controls and 43 cirrhotic patients
* Controls vs. patients group; ** g r o u p A vs. groups B a n d C; *** g r o u p B vs. g r o u p C.
46
CLINICAL RADIOLOGY
1.6 1.4 1.2
-- 1.0 o.8 a. 0.6 0.4 0.2 0
Control
Group A
Group B
Group C
(n = 15)
(n = 14)
(n = 14)
(n = 15)
Fig. 1 - Pulsatility index (PI) and resistive index (RI) of both kidneys (mean) in 15 controls and 43 cirrhotic patients. [B, PI; II, RI.
RESULTS The result of duplex Doppler examination, PI and RI indices in the three arteries of both kidneys in various groups revealed a significant increase in the mean of both indices in groups B (PI: 1.26 + 0.06, RI: 0.66 + 0.01) and C (PI: 1.48 + 0.06, RI: 0.72 _+ 0.02) compared with group A (PI: 0.95+0.07, RI: 0.58_+0.02) or controls (PI: 0.96_+0.08, RI: 0.57_+0.02) (P<0.001). The values of both indices were higher in group C than in group B (P<0.001) while no significant difference was observed between group A and the control group (Table 2, Figs 1, 2, 3 & 4). Tc-99m DTPA renograms in the studied groups showed normal curves in groups A and B, with normal perfusion (i.e. rapid appearance of the radioactive bolus in both kidneys (during the first 10 _+ 4.5 s for Group A and 11 -+ 3.4s for group B)), and secretory (the DTPA time vs. activity curve was steep and reached a maximal activity at 4 _+ 1.5 min for group A and 4 - 2.5 min for group B) and excretory phases (all radioactive material was excreted completely within 30 rain with good response to injected furosemide). In group C there was a delay in appearance of the activity (34 _+ 14.6s), a prolonged secretory (12_+ 4.5rain), and excretory phase (more than 30rain) with delayed tracer concentration, prolonged cortical retention and poor response to intravenous furosemide. The radionuclide-computed GFRs for cirrhotic patients group A and group B were within the normal range (group A: 81 _+ 9.5 ml/min, group B: 78 _+ 8.4ml/min) while in group C the GFR was significantly lower (34 + - 14.5ml/min) (P<0.001) (Table 3, Figs 5 & 6).
Fig. 3 - Duplex Doppler scan in a patient of Child's group B, showing high peak systolic velocity with reduction in diastolic flow (PI = 1.2, RI = 0.66).
Fig. 2 - Duplex Doppler scan in a patient of Child's group A, showing normal renal systolic velocity with minimal or no diastolic reduction (plateau flow) (PI = 0.9, RI = 0.58).
Fig. 4 - Duplex Doppler scan in a patient of Child's group C, showing high peak systolic velocity with prolonged and marked reduction in diastolic filling (PI = 1.5, RI = 0.72).
Table 3 - Tc-99m DTPA renogram in various groups of patients with liver cirrhosis
Phase
Group A (n = 10)
Group B (n = 10)
Group C (n = 10)
Perfusion Secretory Excretory Response to furosemide Radionuclide computed GFR (ml/min)
N N N N 81 • 9.5
N N N N 78 • 8.4 NS *
Diminished Prolonged Prolonged Poor 34 + 14.5 P < 0.001 * *
N : normal; GFR = glomerular filtration rate. *Group A vs. group B; **group C vs. group A or B.
DISCUSSION Recent advances in Doppler sonography and Tc-99m DTPA scintigraphy provide accurate information on the vascular status of highly vascular organs such as the kidney, liver and the spleen [10]. Such techniques are now used to evaluate the renal function in patients with liver cirrhosis as early renal dysfunction is often clinically masked and a normal serum creatinine may be maintained in the presence of a markedly decreased GFR as a result of decreased muscle mass secondary to poor nutrition and hepatic failure [2,11].
9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
47
RENAL ULTRASOUND AND SCINTIGRAPHY IN CIRRHOSIS
10 0 0 0
Count/Time RT K - -
-
(min)
Count/Time
5000
(min)
4000
0
,
,
,
,
0
I , 592
,
L
,
I 1185
10 0 0 0
0
,
,
Count/Time
(rain)
R T K - -
LTK .....
BG
,
,
0
1 , 592
,
,
,
r 1185
..........
5000
/s
0
. . . . 0
I
. . . .
1
592
1185
Fig. 5 - Time activity curve from normal Tc-99m DTPA renogram of a patient of Child' s group B. The renogram consists of several segments; the initial steep upward rise (perfusion), then the more gradually rising second segment (secretory phase) and the downwards sloping segment (excretory phase).
20 0 0 0
-
C o u n t / T i m e
20 0 0 0
(rain)
Count/Time
RT K - -
10 0 0 0
10 0 0 0
0
, 0
,
,
,
(min)
L K - -
I , 592
20
,
=
000
-
,
I 1185
. ~ ' ~ r
0
,
Count/Time
(min)
R T K - -
L.K.- ....
BG
i
,
~
0
I 592
,
,
,
A
I 1185
..........
10 0 0 0
0 0
"'--'--'--'--"-'--'--'-'--' 592
1185
Fig. 6 - The Tc-99m DTPA renogram curves of a patient of Child's group C, showing an initial peak and then a flat response indicating prolonged secretory and excretory phases. 9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
48
CLINICAL RADIOLOGY
In our study, 43 cirrhotic patients and 15 normal subjects were examined with Doppler ultrasound. The pulsatility and resistive indices in the hilar and intrarenal arteries were significantly higher in Child's grade B and C patients compared with grade A or the control group. Thus abnormal vascular impedance is detectable only in patients with impaired liver function and is related to the severity of liver cirrhosis. This finding is consistent with that reported by Platt et al. and Colli e t al. [2,3]. Abnormal Tc-99m DTPA renograms were observed only in group C patients in the form of diminished blood flow bilaterally, decreased tracer concentration in the parenchyma (as demonstrated by a prolonged secretory phase) and prolonged parenchymal transit (as demonstrated by a prolonged excretory phase). In addition, the cortical activity showed impaired response to diuretic administration. Radionuclide computed GFR was within the normal range in patients of group A and B while it was decreased in group C patients. Although the GFR was normal in Child's grade B patients; RI and PI indices were increased; this suggests that renal impairment in liver cirrhosis was related to elevated renal vascular resistance due to renal vasoconstriction more than to glomerular filtration rate abnormalities [2,3,7,11 ]. The same findings were observed by Halkar e t al. in a patient with hepatorenal syndrome (case report) [7]. We did not find other reports using scintigraphy in evaluating renal function in cirrhosis. Epstein e t al. demonstrated a significant reduction in calculated mean renal blood flow with preferential reduction in cortical perfusion in patients with cirrhosis using 133xenon washout and selective renal arteriography [6]. In comparing Doppler with scintigraphy in this study, we found that the Doppler indices can detect an increase in renal vascular impedance (increased PI and RI) in patients with group B modified Child's classification while renography did not detect any significant changes in this group.
Both methods can detect significant changes indicating impairment in patients with advanced cirrhosis (group C). We conclude that duplex ultrasonographic examination of renal arteries is capable of detecting subtle derangements in renal haemodynamics in patients with cirrhosis at an earlier stage than DTPA renography, and may be useful in identifying patients with higher risk of impending renal failure. REFERENCES
1 Epstein M. The hepatorenal syndrome. Hospital Practice 1989;24:6576. 2 Platt JF, Marn CS, Baliga PK, Rubin JM. Renal dysfunction in hepatic disease: early identification with renal duplex Doppler US in patients who undergo liver transplantation. Radiology 1992; 183:801-806. 3 Colli A, Cocciolo M, Riva R, Martine E. Abnormal renovascular impedance in patients with hepatic cirrhosis: detection with duplex US. Radiology 1993;187:561-563. 4 Gentilini P, Laffi G. Renal functional impairment and sodium retention in liver cirrhosis. Digestion 1989;43:1-32. 5 Levenson DJ, Skorecki KL, Newell GC, Narins RG. Acute renal failure associated with hepatobiliary disease. In: Brenner BM & Laarus JM eds. Acute Renal Failure, 2nd ed. New York: Churchill-Livingstone, 1988:535-580. 6 Epstein M. Renal complications of liver disease. Clinical Symposia 1985;37:30-32. 7 Halkar RK, Iffer JA, Taylor A Jr. Tc-99m DTPA and I 131 hippurate renography findings in hepatorenal syndrome. Clinical Nuclear Medicine 1992;17:469-472. 8 Pugh RH, Murray-Lyon IM, Dawson JL. Transection of oesophagus for bleeding oesophageal varices. British Journal of Surgery 1973;60:646649. 9 Gates GF. Glomerular filtration rate: estimation from fractional renal accumulation of Tc-99m DTPA (stannous). American Journal of Roentgenology 1982;138:565-570. 10 Zwiebel WJ, Fruechte D. Basics of abdominal and pelvic duplex. Instrumentation, anatomy and vascular Doppler signatures. Seminars' in Ultrasound, CT and MRI 1992;13:3-21. 11 Papadakis MA, Arieff AI. Unpredictability of clinical evaluation of renal function in cirrhosis. American Journal of Medicine 1987;82:945-952.
9 1998 The Royal Collegeof Radiologists,ClinicalRadiology, 53, 44-48.
Renal Dysfunction in Liver Cirrhosis: Renal Duplex Doppler US vs. Scintigraphy for Early Identification E. A. R. AL-KAREEMY, M. A. SOBH, A. M. MUHAMMAD*, M. M. MOSTAFA* and R. A. SABERt
Departments of Internal Medicine, *Radiology and tNuclear Medicine, Faculty of Medicine, Assiut University, Egypt A diagnostic tool to detect early renal dysfunction before it becomes irreversible would be useful in cirrhosis. This study was carried out to evaluate the role of Doppler sonography and Tc-99m DTPA renography in the detection of early renal dysfunction in patients with different grades of liver cirrhosis. Renal arteries of 43 patients with cirrhosis and normal renal function tests were compared with 15 age and gender matched normal subjects as a control group using colour Doppler sonography and Tc-99m DTPA scintigraphy. The patients were categorized into three groups, A (14), B (14) and C (15), according to a modified Child's classification that assesses the severity of liver cirrhosis. Doppler results revealed a highly significant increase in both the pulsatility and resistive indices in groups B and C compared with either group A patients or control subjects and in group C compared with group B (P < 0.001) in the main renal arteries as well as in the interlobar and arcuate arteries. Insignificant differences were observed between group A and controls (PI: control 0.96 _+ 0.08, group A 0.95 _+ 0.07, group B 1.26 _+ 0.06, group C 1.48 - 0.06; RI: control 0.57 _+ 0.02, group A 0.58 _+ 0.02, group B 0.66 _+ 0.01, group C 0.72 _+ 0.02). Abnormal renograms in the form of delayed appearance (34 _+ 14.6 s), diminished blood flow bilaterally with prolonged secretory ( 1 2 _ 4.5 min) and excretory phases (>30 min) and poor response to intravenous frusemide were only observed in group C patients. Radionuclide computed glomerular filtration rate was within the normal range in patients of group A (81 _+ 9.5 ml/min) and group B (78 _+ 8.4 ml/min) and reduced only in patients of group C (34 _+ 14.5 ml/min). Thus Doppler sonography can detect an increase in renal vascular resistance in patients with moderately severe cirrhosis (Child grade B) when renography was normal. We conclude that Doppler sonography can be used for earlier identification of cirrhotic patients with a higher risk of impending renal failure earlier than renography and may also be used to guide therapeutic approaches. A1-Kareemy, E.A.R., Sobh, M.A., Muhammad, A.M., Mostafa, M.M. & Saber, R.A. (1998). Clinical Radiology 53, 44-48. Renal Dysfunction in Liver Cirrhosis: Renal Duplex Doppler US vs. Scintigraphy for Early Identification
Accepted for Publication 6 March 1997
Renal dysfunction may occur in patients with liver cirrhosis without other causative factors [1-3]. Early clinical detection of renal dysfunction with conventional tests is difficult because the serum creatinine level does not increase until late in the course of the disease [4]. An early change that does occur in these patients is intense renal cortical vasoconstriction with resultant elevation of renal arterial resistance [5]. A significant reduction in both mean renal and cortical blood flow in patients with advanced liver cirrhosis has been demonstrated using the 133-xenon washout technique and with selective renal arteriography [6]. Recently Platt et al. studied renal blood flow using Doppler sonography in cirrhotic patients with and without ascites and reported elevation of renal arterial resistance [2]. Halkar et al. used Tc-99m DTPA renography to study renal blood flow in advanced liver cirrhosis where uptake was symmetrically and bilaterally delayed and diminished [7]. An accurate diagnostic test to detect renal dysfunction before it becomes irreversible could be useful in cirrhotic patients. This study was conducted to evaluate the role of Doppler sonography and Tc-99m DTPA renography in the detection of early renal dysfunction in patients with Correspondence to: Dr E. A. R. A1-Kareemy, Assistant Professor of Internal Medicine, Assiut University Hospital, Assiut, Egypt. 9 1998 The Royal College of Radiologists.
different grades of liver cirrhosis who have normal laboratory tests for kidney function.
PATIENTS AND METHODS Patients Fifty-eight subjects were included in this prospective study: 43 with cirrhosis (27 males and 16 females, mean age 42.9---12.7 years) and 15 healthy volunteers (eight males and seven females, mean age 43.2-+ 6.8 years) from the medical personnel in Assiut University Hospital consented as a control group. Patients were selected from the in-patient section of the Internal Medicine Department in our hospital during the period from April 1994 to July 1995. Full clinical evaluation included a thorough history, with particular emphasis on jaundice, bilharziasis, blood transfusion, drug intake and renal diseases, and complete physical examination. Urine analysis, liver and kidney function tests and abdominal ultrasonographic examination were performed. Patients with a past history of renal disease and those with abnormal renal function tests were excluded. All patients had normal blood urea and serum creatinine levels. All of them stopped medications, especially diuretics, 72 h
45
RENAL ULTRASOUND AND SCINTIGRAPHY IN CIRRHOSIS
Table 1 - Modified Child's classification
Grade
1 2 3 4 5
Encephalopathy Ascites Bilirubin (/xmol/1) A l b u m i n (g/l) P r o t h r o m b i n ratio
1
2
3
None None < 35 > 35 < 1.4
Minimal Slight 35-50 28-35 1.4-2
Marked Moderate > 50 < 28 >2
Score A = 5 - 6 points; B = 7 - 9 points; C = 1 0 - 1 5 points.
before the Doppler and scintigraphy studies. The degree of cirrhosis was evaluated according to a modified Child's classification [8]. Features such as the presence or absence of ascites or encephalopathy, serum albumin level, serum bilirubin level and prothrombin ratio categorize the patients into three grades: grade A (5-6 points), grade B (7-9 points) and grade C (10-15 points) (Table 1).
supine or decubitus position depending on which gives the best view of the course of the renal artery. Doppler signals were obtained from the main renal artery at the hilum of the kidney, from the interlobar arteries along the borders of the medullary pyramids and from the arcuate arteries at the corticomedullary junction. To decrease measurement errors, we maximized the size of the Doppler spectrum by using the lowest frequency range possible without causing aliasing and the lowest wall filter (100 Hz). Doppler spectral analysis included measurement of peak systolic velocity, resistive index (RI) and pulsatility index (PI) for at least three vessels in each kidney. We measured the RI and PI for every vessel from at least three wave forms and calculated the average value. A mean RI and PI as the average of both kidneys was calculated for each patient. RI was defined as (peak systolic velocity-minimum diastolic velocity)/ peak systolic velocity and PI was defined as (peak systolic velocity-minimum diastolic vetocity)/mean velocity. Mean RI and PI (• SD) were calculated for each group. A comparison among groups was performed with analysis of variance and the Student t-test with a P-value <0.05 considered significant.
Group A This group was composed of 14 patients (nine males and five females, mean age 42.6 _+ 12.7 years). All had compensated liver cirrhosis, eight post-hepatitic, four bilharzial and two mixed bilharzial and post-hepatitic aetiology. None had hepatic encephalopathy or ascites. Their biochemical results were: serum bilirubin 14.46 _+ 8.78 #mol/1, serum albumin 37 _+ 4.98 g/1 and prothrombin ratio < 1.4 (Child grade A).
Isotope Scanning Using Tc-99m DTPA
Fourteen patients (nine males and five females, mean age 42.2 _+ 13.5 years) of whom three had post-hepatitic cirrhosis, 10 bilharziasis and one cirrhosis of mixed aetiology. Twelve patients had splenomegaly and all had mild ascites. None had hepatic encephalopathy. Their biochemical results were: bilirubin 20.75 + 12.56/~mol/1, albumin 36.57 -4- 6.64 g/1 and prothrombin ratio range 1.4-2 (Child grade B).
Renal scintigraphy was performed using Tc-99m DTPA for 30 patients, 10 patients from each group (A, B and C). After an overnight fast, the patient was allowed to drink and positioned lying supine over a gamma camera that was centred from the xiphisternum to the symphysis pubis. 5 mCi (185 MBq) of Tc-99 m DTPA was injected intravenously as a bolus followed by single-window acquisition of renal blood flow and sequential 4 s dynamic images in the posterior projection using a parallel hole medium energy collimator. Acquisition of the data was performed in frame mode for 30 min. At 20 min, 40 mg Furosemide was given I.V. and sequential dynamic images were obtained until the end of the study. Time-activity curves were recorded over the kidneys and the individual GFR was measured from backgroundsubtracted images of the kidneys between 80 and 180 s after injection (method of Gates) [9].
Group C
Statistical Methods
Fifteen patients (nine males and six females, mean age 43.1 _+ 15.4 years) were classified as Child grade C. All had decompensated liver cirrhosis, six post-hepatitic, two bilharzial and seven mixed bilharzial and post-hepatitic; all had manifestations of hepatocellular failure: 12 had jaundice, nine had encephalopathy and 15 had ascites. Thirteen patients had splenomegaly and 10 patients had portal hypertension and haematemesis. Their biochemical results were: bilirubin 58.72• albumin 29.53 • 4.14 g/1 and prothrombin ratio > 2.
Comparison among groups was performed with analysis of variance (mean + SD) and Student t-test with P-value < 0.05 considered significant.
Index
Control (n = 15)
Group A (n = 14)
Group B (n = 14)
Group C (n = 15)
Colour Coded Doppler Ultrasonographic Examination
PI
0.96 + 0.08
0.95 -4- 0.07 NS*
1.26 -+ 0.06 P<0.001* P<0.001**
RI
0.57 + 0.02
0.58 -+ 0.02 NS*
0.66 -+ 0.01 P<0.001* P<0.001**
1.48 -+ 0.06 P<0.001* P<0.001* P<0.001*** 0.72 -+ 0.02 P<0.001* P<0.001** P<0.00I***
Group B
Colour coded Doppler ultrasonographic examinations were performed with an Acuson computed sonography 128 XP/10 (Mountain View, USA) with a 3.5 MHz vector array transducer. The main renal, interlobar and arcuate arteries of both kidneys were assessed, including at least three renal vessels in each kidney, recorded according to the method described by Platt et al. [2]. The examinations were performed with the patient in the 9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
Table 2 - Pulsatility index (PI) and resistive index (RI) of both kidneys (mean _+ SD) in 15 controls and 43 cirrhotic patients
* Controls vs. patients group; ** g r o u p A vs. groups B a n d C; *** g r o u p B vs. g r o u p C.
46
CLINICAL RADIOLOGY
1.6 1.4 1.2
-- 1.0 o.8 a. 0.6 0.4 0.2 0
Control
Group A
Group B
Group C
(n = 15)
(n = 14)
(n = 14)
(n = 15)
Fig. 1 - Pulsatility index (PI) and resistive index (RI) of both kidneys (mean) in 15 controls and 43 cirrhotic patients. [B, PI; II, RI.
RESULTS The result of duplex Doppler examination, PI and RI indices in the three arteries of both kidneys in various groups revealed a significant increase in the mean of both indices in groups B (PI: 1.26 + 0.06, RI: 0.66 + 0.01) and C (PI: 1.48 + 0.06, RI: 0.72 _+ 0.02) compared with group A (PI: 0.95+0.07, RI: 0.58_+0.02) or controls (PI: 0.96_+0.08, RI: 0.57_+0.02) (P<0.001). The values of both indices were higher in group C than in group B (P<0.001) while no significant difference was observed between group A and the control group (Table 2, Figs 1, 2, 3 & 4). Tc-99m DTPA renograms in the studied groups showed normal curves in groups A and B, with normal perfusion (i.e. rapid appearance of the radioactive bolus in both kidneys (during the first 10 _+ 4.5 s for Group A and 11 -+ 3.4s for group B)), and secretory (the DTPA time vs. activity curve was steep and reached a maximal activity at 4 _+ 1.5 min for group A and 4 - 2.5 min for group B) and excretory phases (all radioactive material was excreted completely within 30 rain with good response to injected furosemide). In group C there was a delay in appearance of the activity (34 _+ 14.6s), a prolonged secretory (12_+ 4.5rain), and excretory phase (more than 30rain) with delayed tracer concentration, prolonged cortical retention and poor response to intravenous furosemide. The radionuclide-computed GFRs for cirrhotic patients group A and group B were within the normal range (group A: 81 _+ 9.5 ml/min, group B: 78 _+ 8.4ml/min) while in group C the GFR was significantly lower (34 + - 14.5ml/min) (P<0.001) (Table 3, Figs 5 & 6).
Fig. 3 - Duplex Doppler scan in a patient of Child's group B, showing high peak systolic velocity with reduction in diastolic flow (PI = 1.2, RI = 0.66).
Fig. 2 - Duplex Doppler scan in a patient of Child's group A, showing normal renal systolic velocity with minimal or no diastolic reduction (plateau flow) (PI = 0.9, RI = 0.58).
Fig. 4 - Duplex Doppler scan in a patient of Child's group C, showing high peak systolic velocity with prolonged and marked reduction in diastolic filling (PI = 1.5, RI = 0.72).
Table 3 - Tc-99m DTPA renogram in various groups of patients with liver cirrhosis
Phase
Group A (n = 10)
Group B (n = 10)
Group C (n = 10)
Perfusion Secretory Excretory Response to furosemide Radionuclide computed GFR (ml/min)
N N N N 81 • 9.5
N N N N 78 • 8.4 NS *
Diminished Prolonged Prolonged Poor 34 + 14.5 P < 0.001 * *
N : normal; GFR = glomerular filtration rate. *Group A vs. group B; **group C vs. group A or B.
DISCUSSION Recent advances in Doppler sonography and Tc-99m DTPA scintigraphy provide accurate information on the vascular status of highly vascular organs such as the kidney, liver and the spleen [10]. Such techniques are now used to evaluate the renal function in patients with liver cirrhosis as early renal dysfunction is often clinically masked and a normal serum creatinine may be maintained in the presence of a markedly decreased GFR as a result of decreased muscle mass secondary to poor nutrition and hepatic failure [2,11].
9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
47
RENAL ULTRASOUND AND SCINTIGRAPHY IN CIRRHOSIS
10 0 0 0
Count/Time RT K - -
-
(min)
Count/Time
5000
(min)
4000
0
,
,
,
,
0
I , 592
,
L
,
I 1185
10 0 0 0
0
,
,
Count/Time
(rain)
R T K - -
LTK .....
BG
,
,
0
1 , 592
,
,
,
r 1185
..........
5000
/s
0
. . . . 0
I
. . . .
1
592
1185
Fig. 5 - Time activity curve from normal Tc-99m DTPA renogram of a patient of Child' s group B. The renogram consists of several segments; the initial steep upward rise (perfusion), then the more gradually rising second segment (secretory phase) and the downwards sloping segment (excretory phase).
20 0 0 0
-
C o u n t / T i m e
20 0 0 0
(rain)
Count/Time
RT K - -
10 0 0 0
10 0 0 0
0
, 0
,
,
,
(min)
L K - -
I , 592
20
,
=
000
-
,
I 1185
. ~ ' ~ r
0
,
Count/Time
(min)
R T K - -
L.K.- ....
BG
i
,
~
0
I 592
,
,
,
A
I 1185
..........
10 0 0 0
0 0
"'--'--'--'--"-'--'--'-'--' 592
1185
Fig. 6 - The Tc-99m DTPA renogram curves of a patient of Child's group C, showing an initial peak and then a flat response indicating prolonged secretory and excretory phases. 9 1998 The Royal College of Radiologists, Clinical Radiology, 53, 44-48.
48
CLINICAL RADIOLOGY
In our study, 43 cirrhotic patients and 15 normal subjects were examined with Doppler ultrasound. The pulsatility and resistive indices in the hilar and intrarenal arteries were significantly higher in Child's grade B and C patients compared with grade A or the control group. Thus abnormal vascular impedance is detectable only in patients with impaired liver function and is related to the severity of liver cirrhosis. This finding is consistent with that reported by Platt et al. and Colli e t al. [2,3]. Abnormal Tc-99m DTPA renograms were observed only in group C patients in the form of diminished blood flow bilaterally, decreased tracer concentration in the parenchyma (as demonstrated by a prolonged secretory phase) and prolonged parenchymal transit (as demonstrated by a prolonged excretory phase). In addition, the cortical activity showed impaired response to diuretic administration. Radionuclide computed GFR was within the normal range in patients of group A and B while it was decreased in group C patients. Although the GFR was normal in Child's grade B patients; RI and PI indices were increased; this suggests that renal impairment in liver cirrhosis was related to elevated renal vascular resistance due to renal vasoconstriction more than to glomerular filtration rate abnormalities [2,3,7,11 ]. The same findings were observed by Halkar e t al. in a patient with hepatorenal syndrome (case report) [7]. We did not find other reports using scintigraphy in evaluating renal function in cirrhosis. Epstein e t al. demonstrated a significant reduction in calculated mean renal blood flow with preferential reduction in cortical perfusion in patients with cirrhosis using 133xenon washout and selective renal arteriography [6]. In comparing Doppler with scintigraphy in this study, we found that the Doppler indices can detect an increase in renal vascular impedance (increased PI and RI) in patients with group B modified Child's classification while renography did not detect any significant changes in this group.
Both methods can detect significant changes indicating impairment in patients with advanced cirrhosis (group C). We conclude that duplex ultrasonographic examination of renal arteries is capable of detecting subtle derangements in renal haemodynamics in patients with cirrhosis at an earlier stage than DTPA renography, and may be useful in identifying patients with higher risk of impending renal failure. REFERENCES
1 Epstein M. The hepatorenal syndrome. Hospital Practice 1989;24:6576. 2 Platt JF, Marn CS, Baliga PK, Rubin JM. Renal dysfunction in hepatic disease: early identification with renal duplex Doppler US in patients who undergo liver transplantation. Radiology 1992; 183:801-806. 3 Colli A, Cocciolo M, Riva R, Martine E. Abnormal renovascular impedance in patients with hepatic cirrhosis: detection with duplex US. Radiology 1993;187:561-563. 4 Gentilini P, Laffi G. Renal functional impairment and sodium retention in liver cirrhosis. Digestion 1989;43:1-32. 5 Levenson DJ, Skorecki KL, Newell GC, Narins RG. Acute renal failure associated with hepatobiliary disease. In: Brenner BM & Laarus JM eds. Acute Renal Failure, 2nd ed. New York: Churchill-Livingstone, 1988:535-580. 6 Epstein M. Renal complications of liver disease. Clinical Symposia 1985;37:30-32. 7 Halkar RK, Iffer JA, Taylor A Jr. Tc-99m DTPA and I 131 hippurate renography findings in hepatorenal syndrome. Clinical Nuclear Medicine 1992;17:469-472. 8 Pugh RH, Murray-Lyon IM, Dawson JL. Transection of oesophagus for bleeding oesophageal varices. British Journal of Surgery 1973;60:646649. 9 Gates GF. Glomerular filtration rate: estimation from fractional renal accumulation of Tc-99m DTPA (stannous). American Journal of Roentgenology 1982;138:565-570. 10 Zwiebel WJ, Fruechte D. Basics of abdominal and pelvic duplex. Instrumentation, anatomy and vascular Doppler signatures. Seminars' in Ultrasound, CT and MRI 1992;13:3-21. 11 Papadakis MA, Arieff AI. Unpredictability of clinical evaluation of renal function in cirrhosis. American Journal of Medicine 1987;82:945-952.
9 1998 The Royal Collegeof Radiologists,ClinicalRadiology, 53, 44-48.