Abdominal Imaging
ª Springer Science+Business Media New York 2014 Published online: 31 July 2014
Abdom Imaging (2015) 40:258–264 DOI: 10.1007/s00261-014-0204-z
99m
Tc-phytate trans-splenic portal scintigraphy in the diagnosis of cirrhotic portal hypertension and compensatory circulation Aakif Ullah Khan,1 Javaid Irfanullah,2 Syed Qaiser Shah,3 Shazia Fatima2 1
Institute of Radiotherapy and Nuclear Medicine (IRNUM), Peshawar, Pakistan Nuclear Medicine, Oncology and Radiotherapy Institute (NORI), Islamabad, Pakistan 3 Chemistry Department, University of Peshawar, Peshawar, Pakistan 2
Abstract Background and aim: The trans-splenic portal scintigraphy (TPS) was evaluated as a diagnostic tool in the post viral hepatitis cirrhotic patients of various classes of Child Pugh’s (CP) classification. The main aim was to determine the portosystemic shunt index (PSSI) and to compare the results with various clinical grades of disease severity in liver cirrhosis. Methods: TPS was performed on 72 patients and 10 controls and PSSI was derived. All the 72 patients were categorized according to CP classification into three classes. The cirrhotic patients were categorized as CP A (n = 24),CP B (n = 22), and CP C (n = 26)according to CP criteria. PSSI was compared with different CP classes. Results: In the controlled population, the splenic vein was normal in shape and the mean PSSI was calculated to be 0.178 ± 0.031 (n = 10). For CP classes A, B, and C, the mean PSSI was 0.36 ± 0.04, 0.45 ± 0.05, and 0.54 ± 0.04 (n = 26), respectively. There was statistical significance among groups (p £ 0.01).The collateral vessels were mostly uphill or complex. The PSSI index increased as the CP score worsened from A to C. Conclusion: PSSI is a useful and minimally invasive tool to detect and quantify the shunt severity, which correlates well with different clinical grades of disease severity. Key words: Cirrhosis—Portal hypertension— Portosystemic shunt—Portal scintigraphy
According to the World Health Organization estimates, there are 350 million people with chronic hepatitis B
Correspondence to: Aakif Ullah Khan; email: [email protected]
virus infection and 170 million people with chronic hepatitis C virus infection worldwide and Pakistan is one of the world’s most affected countries [1, 2]. Chronic liver disease (CLD) secondary to viral infections is a major concern in the developing world with challenging long-term morbidity primarily due to post cirrhosis portal hypertension (PH) [3, 4]. PH causes shunting of blood from portal system to the systemic circulation which is the main cause of morbidity, mainly due to the upper gastrointestinal bleeds [5, 6]. Evaluation and quantifications of portosystemic shunts are clinically important because they not only predict the risk of bleed but can also foresee the most probable site of bleed. Furthermore, if an intervention is performed, its outcomes can also be assessed quantifiably [7–9]. The measurement of portal pressure by the hepatic venous pressure gradient (HVPG) has been a useful predictor of outcomes in both compensated and decompensated stages of cirrhosis. A high HVPG has been associated with risk of variceal bleeding. The correlation between the outcome of treatment and survival after variceal bleeding has also been established and it is a recommended test for the pharmacological trials. The HVPG estimation has also been found to be useful in the prediction of variceal bleeds but its invasiveness, rate of complications, lack of expertise, consumption of time, and high cost limit its wide spread and routine use [10– 14]. The non-invasive methods like Doppler’s ultrasound have found to be useful in the diagnosis of PH but its utility in the follow-up is limited due to its inability to detect small and subtle changes that may occur after medical treatments or interventions. Furthermore, the operator dependency and inter-observer variability also limit its use in the follow-up of the patients [15–18].
259
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
There is limited data available on the utility of transient elastography (TE) in this clinical scenario as well. TE is an indirect and operator-dependent technique and has not gained a wide acceptance in spite of the claims [19, 20]. There are no reliable, direct, and convenient methods available that might provide the desired information, least invasively in routine clinical practice. The scintigraphic studies offer some promise in quantitation of the portal venous flow impedance by three techniques. A conventional method initially used the liver colloid scintigraphy, in which an estimate regarding PH is made by calculating the capture ratio between liver and spleen. Colloid shift to the bone marrow is considered an advanced stage of PH and cirrhosis of the liver [21]. In its dynamic phase called the liver angioscintigraphy (LAS), the hepatic perfusion index (HPI) is derived which is the ratio between the hepatic artery inflow and total liver perfusion. In CLDs when the HPI is increased (>40%), it depicts the decrease of portal inflow with reactive increase of the flow through the hepatic artery while a HPI of >100% depicts the reversion of portal flow [22, 23]. The second scintigraphic technique involves the rectal mucosal absorption and subsequent flow into the portal channels, and this technique is called per-rectal portal shunt index and was introduced by the classic works of Shiomi et al. [24, 25]. This technique may have the limitations of variable rate of absorption from the rectal mucosa. The third and the most direct method involves injection of the colloidal suspension into the splenic pulp. This is the technique we employed in this study and direct visualization of the splenic vein can be achieved instantaneously and the transit through portal venous system follows. The amount of the tracer escaping the portal venous flow through the alternate pathways is transiently seen in the heart. Then, quantitative indices derived from this dynamic acquisition can be used for the estimation of the hepatic resistivity. The technique is labeled as trans-splenic portal scintigraphy (TPS) which has been described in detail by Gao et al. [26]. In this study, we evaluated the TPS as a diagnostic tool in the post viral hepatitis cirrhotic patients of different Child Pugh’s (CP) classes.
Material and methods This was a cross sectional study of patients having CLD with or without significant PH. After approval from the Institutional Ethical Committee and Regional Training and Monitoring Cell, a total of 72 consecutive patients (non-probability) underwent TPS. All these patients were referred to us by tertiary care liver disease clinics. In addition, 10 normal individuals without any hepatosplenic abnormality were also imaged as control subjects
on similar imaging protocols. All patients and control subjects signed an informed consent form for the procedure. The study was conducted at the Nuclear Medicine Department Nuclear Medicine, Oncology and Radiotherapy Institute (NORI). The inclusion criteria were independent of gender and age. Any patient diagnosed as a case of cirrhosis on the basis of clinical, biochemical profile (total bilirubin, total serum proteins, INR values), ultrasound findings (degree of Ascites, splenomegaly, portal vein and splenic vein dilatation), and endoscopic findings (grades of esophageal varices) were included in the study. Patients with co-existing other gastrointestinal abnormalities and malignancy (e.g., hepatocellular carcinoma, metastatic liver disease) were excluded from the study. The pregnant and lactating women were also not included. The diagnosis of cirrhosis secondary to post viral hepatitis was established on the grounds of liver biopsy and/or typical clinical and biochemical features along with relevant imaging studies. At the time of TPS, the following demographic and laboratory data were collected: age, gender, etiology of liver disease, liver biochemistry (liver function tests and serum albumin), hematology especially platelet count and Child-Turcotte-Pugh score or CP score were ascertained. On the basis of CP score patients were classified into three CP classes i.e., A, B, and C. The age and gender distribution of the patients and control population in different CP classes is given in Table 1. Imaging was performed on a SPECT camera (Infinia, GE Healthcare) with a high-resolution, low-energy, parallel hole collimator. The camera detector was positioned in the anterior view over the patient’s upper abdomen so that the field of view (54 9 40 cm) would include the heart, liver, and spleen. The already fasted patients were injected about 740 MBq (20 mCi) of 99m Tc-phytatein 0.5 mL of injectate percutaneously into the splenic parenchyma. The procedure was carried out under ultrasound guidance (Toshiba Nemio) using a 3.5 MHz probe and an intercostal approach. Continuous dynamic images were acquired at 1 frame/s for 60 frames with a matrix size of 64 9 64. From sequential dynamic images, we identified the heartpool image and defined the region of interest (ROI) around heart-pool activity using alight pen, and then copied this ROI over the right lobe of the liver. Time– Table 1. Age and gender distribution of control population and patients with CLD of different CP scores Groups (number)
Mean age
Std. deviation
Male
Female
Control (n = 10) CP A (n = 24) CP B (n = 22) CP C (n = 26)
39.22 37.41 39.38 37.43
±13.23 ±10.19 ±13.86 ±10.42
6 14 8 9
4 10 14 17
260
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
radioactivity curves were generated for the heart and liver ROIs. The portosystemic shunt index (PSSI) was calculated according to H/(H + L), where H is the value for the peak count of the heart time–radioactivity curve and L is the numeric value of the peak count of the liver time–radioactivity curve as described by Gao et al [26]. The pattern of collateral circulation was judged visually by seeing each 1 s sequential image after splenic injection. To keep the study design simple, additional parameters of ultrasound findings were not taken as additional variables.
Results The age distribution in different groups of patients and control population was matching and the effect of age related changes on liver parenchyma and portal circulation could have been eliminated by this study cluster (Table 1). The number of female population in the disease group was higher than control group. Male subjects agreed to volunteer more compared to female subjects which resulted in a high number of male subjects in the control group. While the study design did not allow us to manipulate the gender distribution as the patient’s inclusion was non-probable, random, and consecutive. In the control cluster, the splenic vein appeared immediately like an ‘‘S’’ shaped curve draining into the hepatic vein, and within 2–4 s, the splenoportal venous system was outlined (Fig. 1A, B). Liver parenchyma appeared rapidly and vividly with homogenous radioactivity/tracer distribution. The fading radioactivity of the splenic injection site appeared smaller and well circumscribed with minimal cardiac activity visualization. The mean PSSI was calculated to be 0.178 ± 0.031. A total of 24 patients were imaged under CP class ‘‘A’’. In which there were 14 male patients and 10 female patients with a male to female ration of 1.4:1. The mean PSSI index was calculated to be 0.36 ± 0.04. The patients classified as having a CP class ‘‘B’’ were 22 in total including 8 male and 14 female patients. The male to female ratio in CP class ‘‘B’’ was calculated to be 0.57:1 and the mean PSSI of CP class B was 0.45 ± 0.04. A total of 26 patients were investigated for PSSI index in class ‘‘C’’ CP class including 17 male and 9 female patients and their male to female ratio was 1.8:1. The mean PSSI index calculated for class ‘‘C’’ CP class was 0.54 ± 0.041. With the worsening of clinical grade of cirrhosis, the PSSI increased (Table 2) with significant differences among the groups (p £ 0.01). Identification and delineation of collateral channels in CP class A class was difficult as little blood was being shunted via collateral, but in CP class B and C collateral veins were readily identifiable. In a total of 72 patients, collateral circulation was identifiable in 55 patients (76%
of the cases). In almost half of the patients of class A (11/ 24), no significant collaterals were identifiable, while in all of the class ‘‘B’’ and ‘‘C’’ patients some pattern of blood shunting was identified. The collaterals identified had 3 types of portosystemic shunt patterns. The first was intrahepatic (Fig. 2), characterized as expansion of the splenoportal and intrahepatic portal veins, weak liver-imaging and mild to moderate heart-pool imaging. This type was found in 30 patients imaged in this study. The second type as referred by Gao et al [26] was compensatory, showing a scanty liver uptake and predominantly uphill (left gastric vein) collaterals. A total of 10 patients (14%) showed compensatory collaterals (Fig. 3). The downhill collaterals into inferior mesenteric vein were identified in four cases (5.6%), or complex in 7 cases (10%). The third type was mostly seen in the advance cases of liver cirrhosis and was termed as ‘‘completely extra-hepatic’’ and these showed a highly expanded splenic vein, inconspicuous and thick collateral branches, and a good heart activity. This third type of collaterals was found only in 4 patients (5.6%). In general, the procedure was very well tolerated. All patients complained of a brief pricking pain at the injection site which subsided after the removal of the needle. On immediate post study ultrasound, one patient (1.2%) had a small splenic sub-capsular hematoma, this patient complained of pain for a longer period of time after the study. The condition was also associated with nausea and a slight dip in post procedure blood pressure (100/70–90/60 mm of Hg). A total of three patients (3.6%) complained of weakness after the procedure. The blood pressure in all these three patients was lower than the pre-procedure blood pressure. Except for one patient who had a small hematoma in the spleen, (already mentioned) the other two patients had no collection of blood on post procedure ultrasound. All patients who complained of weakness and fall of blood pressure were in CP class ‘‘C’’. Their condition improved spontaneously overtime with simple measures like making them recline and by raising the foot end of the bed.
Discussion There are no overall nationwide statistics available for Pakistan regarding the prevalence of hepatitis C. Ali et al. [27] reviewed a total of 84 publications encompassing 139 studies from January 1994 to September 2007 and found moderate to high level of incidence in different parts of Pakistan. The grave consequences of viral and non-viral CLDs are primarily the sequels of increased portal venous pressure secondary to hepatic cirrhosis [28–30]. The general notion is, ‘‘the amount of shunting into the collateral circulation is the outcome of the rise in the portal venous pressure above a certain threshold value.’’
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
261
262
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
A Dynamic (1 s/frame) sequences in a control case, a beautifully outline lined ‘‘S’’ shaped splenic vein can be appreciated in the early part of the study followed by appearance of the liver. B A composite image generated from 1 min dynamic image shows little activity in the heart blood pool. After drawing regions of interest (ROIs) over the composite image separate time activity curves were generated and peak counts over the liver and heart were derived from the curves.
b Fig. 1.
Table 2. Mean PSSI for different Child Pugh’s classes along with total number of patients in each group (n) Groups (number)
Mean PSSI ± SD
Control (n = 10) CP class A (n = 24) CP class B (n = 21) CP class C (n = 26)
0.18 0.36 0.45 0.55
± 0.03 ± 0.04 ± 0.04 + 0.04
Fig. 2. Child Pugh’s class B composite image with region of interest drawn over it, this patient had intrahepatic collateral and a PSSI was calculated to be 0.49.
Fig. 3. Images of the patient with CP class C, A composite image with ROI drawn over it. B Multiple dynamic sequences (1 s/frame) showing upward esophageal collaterals.
To accurately and least invasively quantitate the diversion of portal venous blood to the systemic circulation have been a subject of research in the recent years. If a reliable, reproducible, and readily available diagnostic procedure is clinically validated that can yield the desired information least invasively and cost effectively, it will help millions of CLD patients. The modality should be able to prognosticate the future events and should also be useful in serial studies in order to see the effects of surgical and medical treatments. Nuclear medicine procedures are ideal in clinical scenarios where the dynamics of functional states are being investigated. The subtle changes in the changes in
blood flow cannot be assessed by imaging techniques utilizing measurements of the structural changes like diameters of blood vessels and sizes of organs. Because the structural changes will give us an indirect indication about the ongoing hemodynamic derangement but after intervention the structural changes may persist for quite some time. On the other hand, dynamic scintigraphic techniques give more specific information and after medical or surgical interventions even the slightest changes will be quantifiable. The liver macrophages (also known as Kupffer’s cells) are known to trap and engulf any particulate material when it flows into its parenchyma. A well-known fact was exploited by nuclear medicine to image liver and
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
Table 3. A comparison of PSSI values derived from our study and Gao et al study Groups (number)
Mean PSSI ± SD our study (n = 72)
Mean PSSI ± SD Gao et al’s study (n = 50)
Control CP class A CP class B CP class C
0.18 0.35 0.45 0.55
0.19 0.42 0.56 0.77
± 0.031 (n = 10) ± 0.0426 (n = 24) ± 0.042 (n = 21) + 0.041 (n = 26)
± ± ± ±
0.07 (n = 10) 0.06 (n = 15) 0.07 (n = 19) 0.09(n = 16)
spleen in the past. Colloid labeled with technetium-99m (99mTc) like 99mTc—sulfur colloid, 99mTc—tin colloid, and 99mTc—phytate have been used for this purpose. In the past, this intravenously injected liver scintigraphy agents have also been tried to quantitate the hepatic arterial and total liver perfusion by deriving a HPI, the technique was called as LAS. In LAS, the HPI from normal 20% varied to >100% with increase in hepatic resistivity to portal venous flow [22, 23]. But it cannot account for extra-hepatic circulation and it has limited role in serial follow-up studies after intervention as it is not able to discriminate potential responders from nonresponders [31]. We used 99mTc-phytate in our intra-splenic injection techniques because of its easy availability and no known toxicity. In countries where 99mTc-phytate is not available, 99mTc-tin colloid or 99mTc-sulfur colloid can be used alternatively for this purpose on similar protocol. The rationale of the procedure was simple and outcomes were quantifiable, and any resistance offered by hepatic parenchyma (rise in the portal pressure) will be reflected in the shunting of the tracer back into the systemic blood via the collaterals. Depending upon the severity of portal pressure gradient rise, there is expected to be corresponding rise in the activity reaching the right heart or systemic blood pool. The peak heart count divided by the sum of the peak liver counts and peak heart count will give us important information about the degree of the shunt. The index derived which is labeled as PSSI shows a good correlation with the different CP classes. The collaterals were readily identifiable in 77% of the cases, and were always identifiable in patients with CP class B and C. All the three varieties of intrahepatic, compensated, and extra-hepatic were seen. Most of the collaterals were uphill or complex in nature. In strict sense, this PSSI should have been compared with the portal pressure measured by direct methods like HVPG. But due to the invasiveness of the procedure, the results of the patients of TPS were compared with the clinical data and CP score. The PSSI values derived in our study are comparatively lower than those derived by Gao et al, but the findings correlate well within the groups (Table 3). The future research potentials of this modality can be the evaluation of PSSI in the prediction for the development of varices, because the modality can detect subtle hemodynamic changes and the route of shunting as well.
263
The prognostic implication can pave the ways toward better treatment option application. The role of PSSI estimation over serial studies during and after medical or surgical intervention may also be a matter of interest to many investigators, as PSSI is not an index of size but it is about the dynamics of the blood flow in the portal venous channels. The procedure was well tolerated with no major complications. Only one patient had a small splenic hematoma which needed no further treatment. Only 3 patients complained of weakness and their post procedure blood pressure was found to be mildly lower than the pre-procedure blood pressure. All these patients were re-assured and were asked to take oral fluids. After half an hour of lying in supine positions with foot end raised. These patients regained their pre-procedure blood pressure and were sent home with some instructions.
Conclusion 99m
Tc-phytate TPS is a minimally invasive technique for the accurate quantification of portosystemic shunt and the delineation of collateral flow that correlates well with the CP grades PH. The technique has a very low complications rate that is easily manageable.
Acknowledgments. The authors are thankful to all the technical and non-technical staff of NORI Islamabad for their cooperation in the smooth execution of the research work. And I am also thankful to Mr. Moiz Khan for the English proof reading.
References 1. Previsani N, Lavanchy D (2002) WHO/CDS/CSR/LYO/ 2002.2:Hepatitis B. Geneva: World Health Organization. Hepatitis B 2. Hepatitis C, Geneva: World Health Organization 2000. World Health Organization fact sheets. http://www.who.int/mediacentre/ factsheets/fs164/en/. Accessed Aug 2012 3. Almani SA, Memon AS, Memon AI, et al. (2008) Cirrhosis of liver: etiological factors, complications and prognosis. J LUMHS 1:61– 66 4. Khan P, Ahmad A, Muhammad N, Khan TM, Ahmad B (2009) Screening of 110 cirrhotic patients for hepatitis B and C at Saidu Teaching Hospital Saidu Sharif Swat. J Ayub Med Coll Abbottabad 21(1):119–121 5. Tsouka A, McLin VA (2012) Complications of chronic liver disease. Clin Res Hepatol Gastroenterol 36(3):262–267 6. Bosch J (2007) Vascular deterioration in cirrhosis: the big picture. J Clin Gastroenterol 41(suppl 3):S247–S253 7. Bosch J, Garcia-Pagan JC, Berzigotti A, Abraldes JG (2006) Measurement of portal pressure and its role in the management of chronic liver disease. Semin Liver Dis 26:348–362 8. Akhtar N, Zuberi BF, Hasan SR, Kumar R, Afsar S (2009) Determination of correlation of adjusted blood requirement index with outcome in patients presenting with acute variceal bleeding. World J Gastroenterol 15:2372–2375 9. Kudo M, Zheng RQ, Kim SR, et al. (2008) Diagnostic accuracy of imaging for liver cirrhosis compared to histologically proven liver cirrhosis: a multicenter collaborative study. Intervirology 51(suppl 1):17–26 10. Ban˜ares R, Catalina MV, Ripoll C, Rinco´n D (2011) Prognostic markers in patients who have recovered from an acute variceal bleeding: role of HVPG measurement. Dis Markers 31(3):165–169 11. Albilllos A, Garcia-Tsao G (2011) Classification of cirrhosis: the clinical use of HVPG measurements. Dis Markers 31(3):121–128
264
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
12. Stanley AJ, Robinson I, Forrest EH, Jones AL, Hayes PC (1998) Hemodynamic parameters predicting variceal hemorrhage and survival in alcoholic cirrhosis. QJM 91:19–25 13. Wadhawan M, Dubey S, Sharma BC, Sarin SK (2006) Hepatic venous pressure gradient in cirrhosis: correlation with the size of varices, bleeding ascites and child’s status. Dig Dis Sci 51:2264– 2269 14. Moitinho E, Escorsell A, Bandi JC, et al. (1999) Prognostic value of early measurement of portal pressure in acute variceal bleeding. Gastroenterology 117:626–631 15. de Franchis R (2006) Noninvasive diagnosis of esophageal varices: is it feasible? Am J Gastroenterol 101:2520–2522 16. Berzigotti A, Gilabert R, Abraldes JG, et al. (2008) Noninvasive prediction of clinically significant portal hypertension and esophageal varices in patients with compensated liver cirrhosis. Am J Gastroenterol 103:1159–1167 17. Go¨rg C, Riera-Knorrenschild J, Dietrich J (2002) Flow patterns in the portal venous system. Br J Radiol 75:919–929 18. Gludd C, Hensiksen JH, Nielsen G (1988) Copenhagen study group for liver diseases—prognostic indicators in alcoholic cirrhotic men. Hepatology 8:222–227 19. Vizzutti F, Arena U, Romanelli RG, et al. (2007) Liver stiffness measurement predicts severe portal hypertension in patients with HCV-related cirrhosis. Hepatology 45:1290–1297 20. Castera L, Pinzani M, Bosch J (2012) Non invasive evaluation of portal hypertension using transient elastography. J Hepatol 56(3):696–703 21. Groshar D, Slobodin G, Zuckerman E (2002) Quantitation of liver and spleen uptake of 99mTc-phytate colloid using SPECT: detection of liver cirrhosis. J Nucl Med 43:312–317 22. Dragoteanu M, Balea IA, Dina LA, et al. (2008) Staging of portal hypertension and portosystemic shunts using dynamic nuclear medicine investigations. World J Gastroenterol 14(24):3841–3848
23. Osada H, Honda N, Takahashi T, et al. (2007) Relationship between 99mTc-GSA scintigraphic indices of liver function reserve and portal circulation in patients with chronic liver disease. Ann Nucl Med 21(5):245–249 24. Shiomi S, Sasaki N, Habu D, et al. (1998) Natural course of portal hemodynamics in patients with chronic liver diseases, evaluated by per-rectal portal scintigraphy with Tc-99m pertechnetate. J Gastroenterol 33(4):517–522 25. Shiomi S, Kuroki T, Ueda T, et al. (1990) Prediction of esophageal varices in cirrhosis by per-rectal portal scintigraphy. Hepatogastroenterology 37(6):544–545 26. Gao L, Yang F, Ren C, et al. (2010) Diagnosis of cirrhotic portal hypertension and compensatory circulation using transsplenic portal scintigraphy with 99mTc-phytate. J Nucl Med 51:52–56 27. Ali SA, Donahue RM, Qureshi H, Vermund SH (2009) Hepatitis B and hepatitis C in Pakistan: prevalence and risk factors. Int J Infect Dis 13(1):9–19 28. Zoli M, Merkel C, Magalotti D, et al. (2000) Natural history of cirrhotic patients with small esophageal varices: a prospective study. Am J Gastroenterol 95:503–508 29. de Franchis R (2005) Evolving consensus in portal hypertension report of the Baveno IV consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 43:167–176 30. Berzigotti A, Gilabert R, Abraldes JG, et al. (2008) Noninvasive prediction of clinically significant portal hypertension and esophageal varices in patients with compensated liver cirrhosis. Am J Gastroenterol 103:1159–1167 31. Lehner B, Kretzschmar U, Bubeck B, Ho¨lting T, Schlag P (1988) Results of liver angiography and perfusion scintigraphy do not correlate with response to hepatic artery infusion chemotherapy. J Surg Oncol 39(2):73–78
ª Springer Science+Business Media New York 2014 Published online: 31 July 2014
Abdom Imaging (2015) 40:258–264 DOI: 10.1007/s00261-014-0204-z
99m
Tc-phytate trans-splenic portal scintigraphy in the diagnosis of cirrhotic portal hypertension and compensatory circulation Aakif Ullah Khan,1 Javaid Irfanullah,2 Syed Qaiser Shah,3 Shazia Fatima2 1
Institute of Radiotherapy and Nuclear Medicine (IRNUM), Peshawar, Pakistan Nuclear Medicine, Oncology and Radiotherapy Institute (NORI), Islamabad, Pakistan 3 Chemistry Department, University of Peshawar, Peshawar, Pakistan 2
Abstract Background and aim: The trans-splenic portal scintigraphy (TPS) was evaluated as a diagnostic tool in the post viral hepatitis cirrhotic patients of various classes of Child Pugh’s (CP) classification. The main aim was to determine the portosystemic shunt index (PSSI) and to compare the results with various clinical grades of disease severity in liver cirrhosis. Methods: TPS was performed on 72 patients and 10 controls and PSSI was derived. All the 72 patients were categorized according to CP classification into three classes. The cirrhotic patients were categorized as CP A (n = 24),CP B (n = 22), and CP C (n = 26)according to CP criteria. PSSI was compared with different CP classes. Results: In the controlled population, the splenic vein was normal in shape and the mean PSSI was calculated to be 0.178 ± 0.031 (n = 10). For CP classes A, B, and C, the mean PSSI was 0.36 ± 0.04, 0.45 ± 0.05, and 0.54 ± 0.04 (n = 26), respectively. There was statistical significance among groups (p £ 0.01).The collateral vessels were mostly uphill or complex. The PSSI index increased as the CP score worsened from A to C. Conclusion: PSSI is a useful and minimally invasive tool to detect and quantify the shunt severity, which correlates well with different clinical grades of disease severity. Key words: Cirrhosis—Portal hypertension— Portosystemic shunt—Portal scintigraphy
According to the World Health Organization estimates, there are 350 million people with chronic hepatitis B
Correspondence to: Aakif Ullah Khan; email: [email protected]
virus infection and 170 million people with chronic hepatitis C virus infection worldwide and Pakistan is one of the world’s most affected countries [1, 2]. Chronic liver disease (CLD) secondary to viral infections is a major concern in the developing world with challenging long-term morbidity primarily due to post cirrhosis portal hypertension (PH) [3, 4]. PH causes shunting of blood from portal system to the systemic circulation which is the main cause of morbidity, mainly due to the upper gastrointestinal bleeds [5, 6]. Evaluation and quantifications of portosystemic shunts are clinically important because they not only predict the risk of bleed but can also foresee the most probable site of bleed. Furthermore, if an intervention is performed, its outcomes can also be assessed quantifiably [7–9]. The measurement of portal pressure by the hepatic venous pressure gradient (HVPG) has been a useful predictor of outcomes in both compensated and decompensated stages of cirrhosis. A high HVPG has been associated with risk of variceal bleeding. The correlation between the outcome of treatment and survival after variceal bleeding has also been established and it is a recommended test for the pharmacological trials. The HVPG estimation has also been found to be useful in the prediction of variceal bleeds but its invasiveness, rate of complications, lack of expertise, consumption of time, and high cost limit its wide spread and routine use [10– 14]. The non-invasive methods like Doppler’s ultrasound have found to be useful in the diagnosis of PH but its utility in the follow-up is limited due to its inability to detect small and subtle changes that may occur after medical treatments or interventions. Furthermore, the operator dependency and inter-observer variability also limit its use in the follow-up of the patients [15–18].
259
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
There is limited data available on the utility of transient elastography (TE) in this clinical scenario as well. TE is an indirect and operator-dependent technique and has not gained a wide acceptance in spite of the claims [19, 20]. There are no reliable, direct, and convenient methods available that might provide the desired information, least invasively in routine clinical practice. The scintigraphic studies offer some promise in quantitation of the portal venous flow impedance by three techniques. A conventional method initially used the liver colloid scintigraphy, in which an estimate regarding PH is made by calculating the capture ratio between liver and spleen. Colloid shift to the bone marrow is considered an advanced stage of PH and cirrhosis of the liver [21]. In its dynamic phase called the liver angioscintigraphy (LAS), the hepatic perfusion index (HPI) is derived which is the ratio between the hepatic artery inflow and total liver perfusion. In CLDs when the HPI is increased (>40%), it depicts the decrease of portal inflow with reactive increase of the flow through the hepatic artery while a HPI of >100% depicts the reversion of portal flow [22, 23]. The second scintigraphic technique involves the rectal mucosal absorption and subsequent flow into the portal channels, and this technique is called per-rectal portal shunt index and was introduced by the classic works of Shiomi et al. [24, 25]. This technique may have the limitations of variable rate of absorption from the rectal mucosa. The third and the most direct method involves injection of the colloidal suspension into the splenic pulp. This is the technique we employed in this study and direct visualization of the splenic vein can be achieved instantaneously and the transit through portal venous system follows. The amount of the tracer escaping the portal venous flow through the alternate pathways is transiently seen in the heart. Then, quantitative indices derived from this dynamic acquisition can be used for the estimation of the hepatic resistivity. The technique is labeled as trans-splenic portal scintigraphy (TPS) which has been described in detail by Gao et al. [26]. In this study, we evaluated the TPS as a diagnostic tool in the post viral hepatitis cirrhotic patients of different Child Pugh’s (CP) classes.
Material and methods This was a cross sectional study of patients having CLD with or without significant PH. After approval from the Institutional Ethical Committee and Regional Training and Monitoring Cell, a total of 72 consecutive patients (non-probability) underwent TPS. All these patients were referred to us by tertiary care liver disease clinics. In addition, 10 normal individuals without any hepatosplenic abnormality were also imaged as control subjects
on similar imaging protocols. All patients and control subjects signed an informed consent form for the procedure. The study was conducted at the Nuclear Medicine Department Nuclear Medicine, Oncology and Radiotherapy Institute (NORI). The inclusion criteria were independent of gender and age. Any patient diagnosed as a case of cirrhosis on the basis of clinical, biochemical profile (total bilirubin, total serum proteins, INR values), ultrasound findings (degree of Ascites, splenomegaly, portal vein and splenic vein dilatation), and endoscopic findings (grades of esophageal varices) were included in the study. Patients with co-existing other gastrointestinal abnormalities and malignancy (e.g., hepatocellular carcinoma, metastatic liver disease) were excluded from the study. The pregnant and lactating women were also not included. The diagnosis of cirrhosis secondary to post viral hepatitis was established on the grounds of liver biopsy and/or typical clinical and biochemical features along with relevant imaging studies. At the time of TPS, the following demographic and laboratory data were collected: age, gender, etiology of liver disease, liver biochemistry (liver function tests and serum albumin), hematology especially platelet count and Child-Turcotte-Pugh score or CP score were ascertained. On the basis of CP score patients were classified into three CP classes i.e., A, B, and C. The age and gender distribution of the patients and control population in different CP classes is given in Table 1. Imaging was performed on a SPECT camera (Infinia, GE Healthcare) with a high-resolution, low-energy, parallel hole collimator. The camera detector was positioned in the anterior view over the patient’s upper abdomen so that the field of view (54 9 40 cm) would include the heart, liver, and spleen. The already fasted patients were injected about 740 MBq (20 mCi) of 99m Tc-phytatein 0.5 mL of injectate percutaneously into the splenic parenchyma. The procedure was carried out under ultrasound guidance (Toshiba Nemio) using a 3.5 MHz probe and an intercostal approach. Continuous dynamic images were acquired at 1 frame/s for 60 frames with a matrix size of 64 9 64. From sequential dynamic images, we identified the heartpool image and defined the region of interest (ROI) around heart-pool activity using alight pen, and then copied this ROI over the right lobe of the liver. Time– Table 1. Age and gender distribution of control population and patients with CLD of different CP scores Groups (number)
Mean age
Std. deviation
Male
Female
Control (n = 10) CP A (n = 24) CP B (n = 22) CP C (n = 26)
39.22 37.41 39.38 37.43
±13.23 ±10.19 ±13.86 ±10.42
6 14 8 9
4 10 14 17
260
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
radioactivity curves were generated for the heart and liver ROIs. The portosystemic shunt index (PSSI) was calculated according to H/(H + L), where H is the value for the peak count of the heart time–radioactivity curve and L is the numeric value of the peak count of the liver time–radioactivity curve as described by Gao et al [26]. The pattern of collateral circulation was judged visually by seeing each 1 s sequential image after splenic injection. To keep the study design simple, additional parameters of ultrasound findings were not taken as additional variables.
Results The age distribution in different groups of patients and control population was matching and the effect of age related changes on liver parenchyma and portal circulation could have been eliminated by this study cluster (Table 1). The number of female population in the disease group was higher than control group. Male subjects agreed to volunteer more compared to female subjects which resulted in a high number of male subjects in the control group. While the study design did not allow us to manipulate the gender distribution as the patient’s inclusion was non-probable, random, and consecutive. In the control cluster, the splenic vein appeared immediately like an ‘‘S’’ shaped curve draining into the hepatic vein, and within 2–4 s, the splenoportal venous system was outlined (Fig. 1A, B). Liver parenchyma appeared rapidly and vividly with homogenous radioactivity/tracer distribution. The fading radioactivity of the splenic injection site appeared smaller and well circumscribed with minimal cardiac activity visualization. The mean PSSI was calculated to be 0.178 ± 0.031. A total of 24 patients were imaged under CP class ‘‘A’’. In which there were 14 male patients and 10 female patients with a male to female ration of 1.4:1. The mean PSSI index was calculated to be 0.36 ± 0.04. The patients classified as having a CP class ‘‘B’’ were 22 in total including 8 male and 14 female patients. The male to female ratio in CP class ‘‘B’’ was calculated to be 0.57:1 and the mean PSSI of CP class B was 0.45 ± 0.04. A total of 26 patients were investigated for PSSI index in class ‘‘C’’ CP class including 17 male and 9 female patients and their male to female ratio was 1.8:1. The mean PSSI index calculated for class ‘‘C’’ CP class was 0.54 ± 0.041. With the worsening of clinical grade of cirrhosis, the PSSI increased (Table 2) with significant differences among the groups (p £ 0.01). Identification and delineation of collateral channels in CP class A class was difficult as little blood was being shunted via collateral, but in CP class B and C collateral veins were readily identifiable. In a total of 72 patients, collateral circulation was identifiable in 55 patients (76%
of the cases). In almost half of the patients of class A (11/ 24), no significant collaterals were identifiable, while in all of the class ‘‘B’’ and ‘‘C’’ patients some pattern of blood shunting was identified. The collaterals identified had 3 types of portosystemic shunt patterns. The first was intrahepatic (Fig. 2), characterized as expansion of the splenoportal and intrahepatic portal veins, weak liver-imaging and mild to moderate heart-pool imaging. This type was found in 30 patients imaged in this study. The second type as referred by Gao et al [26] was compensatory, showing a scanty liver uptake and predominantly uphill (left gastric vein) collaterals. A total of 10 patients (14%) showed compensatory collaterals (Fig. 3). The downhill collaterals into inferior mesenteric vein were identified in four cases (5.6%), or complex in 7 cases (10%). The third type was mostly seen in the advance cases of liver cirrhosis and was termed as ‘‘completely extra-hepatic’’ and these showed a highly expanded splenic vein, inconspicuous and thick collateral branches, and a good heart activity. This third type of collaterals was found only in 4 patients (5.6%). In general, the procedure was very well tolerated. All patients complained of a brief pricking pain at the injection site which subsided after the removal of the needle. On immediate post study ultrasound, one patient (1.2%) had a small splenic sub-capsular hematoma, this patient complained of pain for a longer period of time after the study. The condition was also associated with nausea and a slight dip in post procedure blood pressure (100/70–90/60 mm of Hg). A total of three patients (3.6%) complained of weakness after the procedure. The blood pressure in all these three patients was lower than the pre-procedure blood pressure. Except for one patient who had a small hematoma in the spleen, (already mentioned) the other two patients had no collection of blood on post procedure ultrasound. All patients who complained of weakness and fall of blood pressure were in CP class ‘‘C’’. Their condition improved spontaneously overtime with simple measures like making them recline and by raising the foot end of the bed.
Discussion There are no overall nationwide statistics available for Pakistan regarding the prevalence of hepatitis C. Ali et al. [27] reviewed a total of 84 publications encompassing 139 studies from January 1994 to September 2007 and found moderate to high level of incidence in different parts of Pakistan. The grave consequences of viral and non-viral CLDs are primarily the sequels of increased portal venous pressure secondary to hepatic cirrhosis [28–30]. The general notion is, ‘‘the amount of shunting into the collateral circulation is the outcome of the rise in the portal venous pressure above a certain threshold value.’’
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
261
262
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
A Dynamic (1 s/frame) sequences in a control case, a beautifully outline lined ‘‘S’’ shaped splenic vein can be appreciated in the early part of the study followed by appearance of the liver. B A composite image generated from 1 min dynamic image shows little activity in the heart blood pool. After drawing regions of interest (ROIs) over the composite image separate time activity curves were generated and peak counts over the liver and heart were derived from the curves.
b Fig. 1.
Table 2. Mean PSSI for different Child Pugh’s classes along with total number of patients in each group (n) Groups (number)
Mean PSSI ± SD
Control (n = 10) CP class A (n = 24) CP class B (n = 21) CP class C (n = 26)
0.18 0.36 0.45 0.55
± 0.03 ± 0.04 ± 0.04 + 0.04
Fig. 2. Child Pugh’s class B composite image with region of interest drawn over it, this patient had intrahepatic collateral and a PSSI was calculated to be 0.49.
Fig. 3. Images of the patient with CP class C, A composite image with ROI drawn over it. B Multiple dynamic sequences (1 s/frame) showing upward esophageal collaterals.
To accurately and least invasively quantitate the diversion of portal venous blood to the systemic circulation have been a subject of research in the recent years. If a reliable, reproducible, and readily available diagnostic procedure is clinically validated that can yield the desired information least invasively and cost effectively, it will help millions of CLD patients. The modality should be able to prognosticate the future events and should also be useful in serial studies in order to see the effects of surgical and medical treatments. Nuclear medicine procedures are ideal in clinical scenarios where the dynamics of functional states are being investigated. The subtle changes in the changes in
blood flow cannot be assessed by imaging techniques utilizing measurements of the structural changes like diameters of blood vessels and sizes of organs. Because the structural changes will give us an indirect indication about the ongoing hemodynamic derangement but after intervention the structural changes may persist for quite some time. On the other hand, dynamic scintigraphic techniques give more specific information and after medical or surgical interventions even the slightest changes will be quantifiable. The liver macrophages (also known as Kupffer’s cells) are known to trap and engulf any particulate material when it flows into its parenchyma. A well-known fact was exploited by nuclear medicine to image liver and
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
Table 3. A comparison of PSSI values derived from our study and Gao et al study Groups (number)
Mean PSSI ± SD our study (n = 72)
Mean PSSI ± SD Gao et al’s study (n = 50)
Control CP class A CP class B CP class C
0.18 0.35 0.45 0.55
0.19 0.42 0.56 0.77
± 0.031 (n = 10) ± 0.0426 (n = 24) ± 0.042 (n = 21) + 0.041 (n = 26)
± ± ± ±
0.07 (n = 10) 0.06 (n = 15) 0.07 (n = 19) 0.09(n = 16)
spleen in the past. Colloid labeled with technetium-99m (99mTc) like 99mTc—sulfur colloid, 99mTc—tin colloid, and 99mTc—phytate have been used for this purpose. In the past, this intravenously injected liver scintigraphy agents have also been tried to quantitate the hepatic arterial and total liver perfusion by deriving a HPI, the technique was called as LAS. In LAS, the HPI from normal 20% varied to >100% with increase in hepatic resistivity to portal venous flow [22, 23]. But it cannot account for extra-hepatic circulation and it has limited role in serial follow-up studies after intervention as it is not able to discriminate potential responders from nonresponders [31]. We used 99mTc-phytate in our intra-splenic injection techniques because of its easy availability and no known toxicity. In countries where 99mTc-phytate is not available, 99mTc-tin colloid or 99mTc-sulfur colloid can be used alternatively for this purpose on similar protocol. The rationale of the procedure was simple and outcomes were quantifiable, and any resistance offered by hepatic parenchyma (rise in the portal pressure) will be reflected in the shunting of the tracer back into the systemic blood via the collaterals. Depending upon the severity of portal pressure gradient rise, there is expected to be corresponding rise in the activity reaching the right heart or systemic blood pool. The peak heart count divided by the sum of the peak liver counts and peak heart count will give us important information about the degree of the shunt. The index derived which is labeled as PSSI shows a good correlation with the different CP classes. The collaterals were readily identifiable in 77% of the cases, and were always identifiable in patients with CP class B and C. All the three varieties of intrahepatic, compensated, and extra-hepatic were seen. Most of the collaterals were uphill or complex in nature. In strict sense, this PSSI should have been compared with the portal pressure measured by direct methods like HVPG. But due to the invasiveness of the procedure, the results of the patients of TPS were compared with the clinical data and CP score. The PSSI values derived in our study are comparatively lower than those derived by Gao et al, but the findings correlate well within the groups (Table 3). The future research potentials of this modality can be the evaluation of PSSI in the prediction for the development of varices, because the modality can detect subtle hemodynamic changes and the route of shunting as well.
263
The prognostic implication can pave the ways toward better treatment option application. The role of PSSI estimation over serial studies during and after medical or surgical intervention may also be a matter of interest to many investigators, as PSSI is not an index of size but it is about the dynamics of the blood flow in the portal venous channels. The procedure was well tolerated with no major complications. Only one patient had a small splenic hematoma which needed no further treatment. Only 3 patients complained of weakness and their post procedure blood pressure was found to be mildly lower than the pre-procedure blood pressure. All these patients were re-assured and were asked to take oral fluids. After half an hour of lying in supine positions with foot end raised. These patients regained their pre-procedure blood pressure and were sent home with some instructions.
Conclusion 99m
Tc-phytate TPS is a minimally invasive technique for the accurate quantification of portosystemic shunt and the delineation of collateral flow that correlates well with the CP grades PH. The technique has a very low complications rate that is easily manageable.
Acknowledgments. The authors are thankful to all the technical and non-technical staff of NORI Islamabad for their cooperation in the smooth execution of the research work. And I am also thankful to Mr. Moiz Khan for the English proof reading.
References 1. Previsani N, Lavanchy D (2002) WHO/CDS/CSR/LYO/ 2002.2:Hepatitis B. Geneva: World Health Organization. Hepatitis B 2. Hepatitis C, Geneva: World Health Organization 2000. World Health Organization fact sheets. http://www.who.int/mediacentre/ factsheets/fs164/en/. Accessed Aug 2012 3. Almani SA, Memon AS, Memon AI, et al. (2008) Cirrhosis of liver: etiological factors, complications and prognosis. J LUMHS 1:61– 66 4. Khan P, Ahmad A, Muhammad N, Khan TM, Ahmad B (2009) Screening of 110 cirrhotic patients for hepatitis B and C at Saidu Teaching Hospital Saidu Sharif Swat. J Ayub Med Coll Abbottabad 21(1):119–121 5. Tsouka A, McLin VA (2012) Complications of chronic liver disease. Clin Res Hepatol Gastroenterol 36(3):262–267 6. Bosch J (2007) Vascular deterioration in cirrhosis: the big picture. J Clin Gastroenterol 41(suppl 3):S247–S253 7. Bosch J, Garcia-Pagan JC, Berzigotti A, Abraldes JG (2006) Measurement of portal pressure and its role in the management of chronic liver disease. Semin Liver Dis 26:348–362 8. Akhtar N, Zuberi BF, Hasan SR, Kumar R, Afsar S (2009) Determination of correlation of adjusted blood requirement index with outcome in patients presenting with acute variceal bleeding. World J Gastroenterol 15:2372–2375 9. Kudo M, Zheng RQ, Kim SR, et al. (2008) Diagnostic accuracy of imaging for liver cirrhosis compared to histologically proven liver cirrhosis: a multicenter collaborative study. Intervirology 51(suppl 1):17–26 10. Ban˜ares R, Catalina MV, Ripoll C, Rinco´n D (2011) Prognostic markers in patients who have recovered from an acute variceal bleeding: role of HVPG measurement. Dis Markers 31(3):165–169 11. Albilllos A, Garcia-Tsao G (2011) Classification of cirrhosis: the clinical use of HVPG measurements. Dis Markers 31(3):121–128
264
A. U. Khan et al.: Diagnosis of cirrhotic portal hypertension and compensatory circulation
12. Stanley AJ, Robinson I, Forrest EH, Jones AL, Hayes PC (1998) Hemodynamic parameters predicting variceal hemorrhage and survival in alcoholic cirrhosis. QJM 91:19–25 13. Wadhawan M, Dubey S, Sharma BC, Sarin SK (2006) Hepatic venous pressure gradient in cirrhosis: correlation with the size of varices, bleeding ascites and child’s status. Dig Dis Sci 51:2264– 2269 14. Moitinho E, Escorsell A, Bandi JC, et al. (1999) Prognostic value of early measurement of portal pressure in acute variceal bleeding. Gastroenterology 117:626–631 15. de Franchis R (2006) Noninvasive diagnosis of esophageal varices: is it feasible? Am J Gastroenterol 101:2520–2522 16. Berzigotti A, Gilabert R, Abraldes JG, et al. (2008) Noninvasive prediction of clinically significant portal hypertension and esophageal varices in patients with compensated liver cirrhosis. Am J Gastroenterol 103:1159–1167 17. Go¨rg C, Riera-Knorrenschild J, Dietrich J (2002) Flow patterns in the portal venous system. Br J Radiol 75:919–929 18. Gludd C, Hensiksen JH, Nielsen G (1988) Copenhagen study group for liver diseases—prognostic indicators in alcoholic cirrhotic men. Hepatology 8:222–227 19. Vizzutti F, Arena U, Romanelli RG, et al. (2007) Liver stiffness measurement predicts severe portal hypertension in patients with HCV-related cirrhosis. Hepatology 45:1290–1297 20. Castera L, Pinzani M, Bosch J (2012) Non invasive evaluation of portal hypertension using transient elastography. J Hepatol 56(3):696–703 21. Groshar D, Slobodin G, Zuckerman E (2002) Quantitation of liver and spleen uptake of 99mTc-phytate colloid using SPECT: detection of liver cirrhosis. J Nucl Med 43:312–317 22. Dragoteanu M, Balea IA, Dina LA, et al. (2008) Staging of portal hypertension and portosystemic shunts using dynamic nuclear medicine investigations. World J Gastroenterol 14(24):3841–3848
23. Osada H, Honda N, Takahashi T, et al. (2007) Relationship between 99mTc-GSA scintigraphic indices of liver function reserve and portal circulation in patients with chronic liver disease. Ann Nucl Med 21(5):245–249 24. Shiomi S, Sasaki N, Habu D, et al. (1998) Natural course of portal hemodynamics in patients with chronic liver diseases, evaluated by per-rectal portal scintigraphy with Tc-99m pertechnetate. J Gastroenterol 33(4):517–522 25. Shiomi S, Kuroki T, Ueda T, et al. (1990) Prediction of esophageal varices in cirrhosis by per-rectal portal scintigraphy. Hepatogastroenterology 37(6):544–545 26. Gao L, Yang F, Ren C, et al. (2010) Diagnosis of cirrhotic portal hypertension and compensatory circulation using transsplenic portal scintigraphy with 99mTc-phytate. J Nucl Med 51:52–56 27. Ali SA, Donahue RM, Qureshi H, Vermund SH (2009) Hepatitis B and hepatitis C in Pakistan: prevalence and risk factors. Int J Infect Dis 13(1):9–19 28. Zoli M, Merkel C, Magalotti D, et al. (2000) Natural history of cirrhotic patients with small esophageal varices: a prospective study. Am J Gastroenterol 95:503–508 29. de Franchis R (2005) Evolving consensus in portal hypertension report of the Baveno IV consensus workshop on methodology of diagnosis and therapy in portal hypertension. J Hepatol 43:167–176 30. Berzigotti A, Gilabert R, Abraldes JG, et al. (2008) Noninvasive prediction of clinically significant portal hypertension and esophageal varices in patients with compensated liver cirrhosis. Am J Gastroenterol 103:1159–1167 31. Lehner B, Kretzschmar U, Bubeck B, Ho¨lting T, Schlag P (1988) Results of liver angiography and perfusion scintigraphy do not correlate with response to hepatic artery infusion chemotherapy. J Surg Oncol 39(2):73–78
