- Email: [email protected]
Influence of esophageal varices and spontaneous portal-systemic shunts on postprandial splanchnic hemodynamics
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2001 by Am. Coll. of Gastroenterology Published by Elsevier Science Inc.
Vol. 96, No. 2, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(00)02351-0
Influence of Esophageal Varices and Spontaneous Portal-Systemic Shunts on Postprandial Splanchnic Hemodynamics Sebastiano Siringo, M.D., Fabio Piscaglia, M.D., Gianni Zironi, M.D., Soccorsa Sofia, M.D., Stefano Gaiani, M.D., Marcello Zammataro, M.D., and Luigi Bolondi, M.D. Dipartimento di Medicina Interna e Gastroenterologia, Universita´ di Bologna, Bologna; and Divisione di Medicina Generale 2, Ospedale Garibaldi, Catania, Italy
OBJECTIVE: The aim of the study was to assess postprandial splanchnic hemodynamic changes in cirrhosis in relation to variceal status. METHODS: In 9 healthy controls and 56 patients with liver cirrhosis, stratified according to variceal status and presence of spontaneous portal-systemic shunts, the portal vein diameter and flow velocity, the congestion index of the portal vein, and the resistive index of the superior mesenteric artery (SMA-RI) were studied by Doppler ultrasound before and 30, 60, and 120 min after the intake of a standard meal. Comparison of postprandial parameters with basal ones was done within each group by paired t test and among groups by ANOVA and Duncan test. RESULTS: Healthy controls and cirrhotic patients without varices showed similar significant splanchnic hemodynamic changes, namely a reduction of SMA-RI (⫺13% at 30 min) and a consequent increase in portal vein diameter (respectively, ⫹ 32% and ⫹ 17% in the two groups) and velocity (⫹66% and ⫹ 51%). A significant reduction of SMA-RI was also found in patients with varices, irrespective of the variceal size (range, ⫺7 to ⫺11%), but the expected portal vein dilation and velocity increase were progressively blunted with the increase of variceal size (range, 0 –5% for diameter and 5–19% for velocity). Patients with spontaneous portal-systemic shunts showed a response similar to that of patients with large varices. Significant modification of the congestion index of the portal vein did not occur in any group. CONCLUSIONS: Our results show that the hemodynamic response to meal in patients with liver cirrhosis is influenced by the presence and size of esophageal varices and the presence of spontaneous portal-systemic shunts. (Am J Gastroenterol 2001;96:550 –556. © 2001 by Am. Coll. of Gastroenterology)
Current address for Sebastiano Siringo: Divisione di Medicina Generale 2, P.O. Garibaldi, P.zza S. Maria di Gesu` 7, 95124 Catania, Italy.
INTRODUCTION Patients with liver cirrhosis and esophageal varices of different size have different patterns of basal splanchnic hemodynamics, as evaluated by Doppler ultrasound (1). Moreover, it has been shown that the patterns of daily changes in splanchnic hemodynamic parameters are related to variceal size and spontaneous portal-systemic shunts (2). It has also been found that the development of portal-systemic shunts can modify portal hemodynamics in patients with liver cirrhosis (3–7) Many studies investigated the portal hemodynamic response to meal intake in patients with cirrhosis (8 –16). However, the results concerning modification of portal blood flow velocity after a meal are conflicting. A possible reason for this might be that patients were not stratified according to the presence and size of esophageal varices. Because variceal size is associated with different baseline splanchnic hemodynamic patterns (1, 2), a different response to the same stimulation may also be hypothesized. A recent study found that patients with varices have a different portal hemodynamic response to meals in comparison to those without varices, although no difference was found among patients with different variceal size (16). However, it remains to be investigated whether also considering the presence of large spontaneous portal-systemic shunts (SPSS), which influence splanchnic hemodynamics, could produce more specific results for the various variceal classes. The aim of the present study was therefore to assess the splanchnic hemodynamic changes occurring after meal intake in cirrhotic patients with different risk of variceal bleeding according to the variceal size and to the presence of SPSS.
MATERIALS AND METHODS We studied 9 healthy controls and 56 patients with liver cirrhosis in whom a complete abdominal ultrasound and duplex Doppler evaluation of the splanchnic hemodynamics was feasible. The severity of liver disease was assessed
AJG – February, 2001
Postprandial Splanchnic Hemodynamics
Table 1. Characteristics of the Study Population Characteristic
Data
Normal subjects M/F Mean age, yr (⫾SD) Cirrhotic patients M/F Mean age, yr (⫾SD) Etiology of cirrhosis Viral Alcoholic Viral & alcoholic PBC Cryptogenic Child-Pugh class (A/B/C) Esophageal varices (F0/F1/F2/F3) Spontaneous portal-systemic shunt
9 6/3 57.2 (⫾ 13.4) 56 34/22 57.9 (⫾ 9.5) 39 10 4 2 1 28/24/4 16/16/13/11 13
according to the Pugh modification of the Child criteria (17). All patients had upper-GI endoscopy at entry into the study, and esophageal varices were classified according to the endoscopic rules of the Japanese Research Society for Portal Hypertension (18). In all cirrhotic patients without varices, the presence of liver cirrhosis was previously proven by needle liver biopsy, whereas in all the remaining patients, it was demonstrated either by a previous liver biopsy or by unequivocal biochemical, endoscopic, and ultrasonographic findings. The presence of spontaneous portal-systemic shunts (SPSS) was searched for by abdominal US in each patient. Table 1 shows patient characteristics. Twelve patients had no esophageal varices (F0), 10 had F1, 12 had F2, and 9 had F3 varices. Thirteen patients had large SPSS visible at duplex Doppler (umbilical and/or splenorenal); 6 (46.2%) of them had no varices, 5 (38.5%) had F1, and 2 (15.4%) had F3 varices. The following parameters were assessed by an EsaoteHitachi AU590 asynchronous echo color Doppler machine with a 3.5-MHz convex transducer. Portal-vein caliber (PV) was measured at the crossing point with the hepatic artery or just distally to it. Portal-vein blood flow velocity (PFV) was considered the time-averaged maximum velocity over a period of 4 – 6 s. The sample volume was positioned at the crossing point of the portal vein with the hepatic artery and set as wide as ⱖ50% of the vein caliber. The Doppler insonation angle was kept between 30° and 60°. The Con-
551
gestion Index of the portal vein was calculated according to Moriyasu’s formula (19): (cross sectional area/mean PFV)/10 Cross-sectional area was derived from the diameter, assuming a circular shape of the vessel, and mean PFV was the time-averaged maximum velocity ⫻ 0.57 (19). The superior mesenteric artery Resistive Index (SMA-RI) was derived on the proximal tract of the vessel according to the following formula: peak systolic frequency ⫺ end-diastolic frequency/ peak systolic frequency Arterial blood pressure and heart rate (HR) were also measured at each time point. All measurements were begun at 01:00 PM, which is the usual lunch time in Italy, after an overnight fast. After the basal measurements (T0), the subjects were given a standard 1000-kcal isocaloric meal (50% carbohydrates, 20% proteins, 30% lipids), and the measurements were repeated 30, 60, and 120 min, after the end of a meal. All measurements were taken by three skilled operators (S.S., F.P., G.Z.) who had a long joint training, which is important in reducing interobserver variability (20). Each patient was examined by only a single operator, which kept the technical approach consistent at every time point of the procedure. Examinations were performed after the subject had been lying down for ⱖ15 min. Each Doppler measurement was taken during suspended normal respiration. For every parameter, three to six consistent measurements were taken, and the results expressed as their mean. Results were expressed as mean absolute values (⫾ SD). The comparison of the T0 values with those obtained at each following time point was done within each group by the paired t test. The differences with respect to T0 were also expressed as percentage changes to overcome the interindividual variability in absolute basal values; the comparison among groups was assessed by the one-way ANOVA. The source of significance among groups within the ANOVA test was identified by the Duncan test.
Table 2. Hemodyanmic Responses to Meal in Healthy Controls (n ⫽ 9) Parameter
T0
30 min
60 min
120 min
p
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
10 (⫾3)a 26 (⫾5)ab 84 (⫾5)abc 0.3 (⫾2) 121 (⫾14) 80 (⫾11)abc 70 (⫾11)
12 (⫾2)a 41 (⫾8)a 73 (⫾7)a 0.3 (⫾1) 121 (⫾17) 71 (⫾11)a 77 (⫾14)
11 (⫾2) 34 (⫾6)b 76 (⫾3)b 0.3 (⫾1) 117 (⫾14) 73 (⫾9)b 74 (⫾15)
10 (⫾3) 34 (⫾10) 76 (⫾5)c 0.3 (⫾2) 116 (⫾16) 71 (⫾11)c 74 (⫾14)
⬍ 0.01 ab ⬍ 0.006 abc ⬍ 0.003 NS NS abc ⬍ 0.05 NS
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, HR ⫽ heart rate, NS ⫽ nonsignificant.
a
552
Siringo et al.
AJG – Vol. 96, No. 2, 2001
Table 3. Hemodynamic Responses to Meal in Patients With liver cirrhosis and without varices (group; n ⫽ 12) Parameter PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0
30 min abc
10 (⫾2) 22 (⫾7)abc 85 (⫾5)abc 0.5 (⫾3) 132 (⫾19)a 80 (⫾10)a 68 (⫾10)abc
a
12 (⫾1) 32 (⫾9)a 75 (⫾7)a 0.4 (⫾2) 126 (⫾16) 76 (⫾10) 73 (⫾11)a
60 min b
12 (⫾1) 30 (⫾9)b 78 (⫾7)b 0.4 (⫾2) 125 (⫾17)a 73 (⫾10)a 73 (⫾10)b
120 min c
11 (⫾2) 26 (⫾7)c 82 (⫾6)c 0.4 (⫾2) 129 (⫾16) 76 (⫾9) 71 (⫾10)c
p ⬍ 0.0005 ⬍ 0.002 abc ⬍ 0.003 NS a ⬍ 0.02 a 0.03 abc ⬍ 0.009 abc abc
SBP ⫽ systolic blood pressure, DBP ⫽ diastoic blood pressure, NS ⫽ nonsignificant.
RESULTS Healthy controls showed a mild portal vein dilation with a marked increase of PFV and a sustained decrease of SMA-RI after meal ingestion (Table 2). A persistent decrease of the diastolic blood pressure was also observed (Table 2). Cirrhotic patients without varices had a similar hemodynamic response to the meal, with a persistent and significant increase of portal vein diameter, PFV, and HR and a decrease in SMA-RI and systemic blood pressure (Table 3). Patients with F1 and F2 varices had a significant decrease of SMA-RI and increased PFV without any significant PV dilation (Tables 4 and 5). In both groups, the HR increased after a meal (Tables 4 and 5). Patients with F3 varices showed only a persistent decrease of SMA-RI after a meal, without any modification of portal vein diameter and PFV (Table 6). Patients with SPSS had a slight and persistent PV dilation, no increase in the PFV, and a shorter duration of the SMA-RI reduction (Table 7). Moreover, control subjects and patients without varices had the maximum increase in PFV at 30 min, whereas cirrhotics with F1 and F2 had a rather steady PFV increase, and patients with F3 and SPSS had no modification of PFV at all. Because the CI did not show any change in the abovementioned analysis, it was not included in the ANOVA test. Table 8 shows the percentage change of each parameter after a meal with respect to T0 in control subjects and in patients with cirrhosis, stratified according to the variceal size and presence of SPSS. The percentage increase of PV was progressively smaller in patients with increasing variceal size and with SPSS. The subgroup comparison showed that the percentage increase of PV in controls was significantly higher within the first 30 min after the meal than in patients without varices, with varices, and with SPSS. After 60 min, the extent of PV change was similar in controls and in patients without varices, and both groups showed a persistently greater PV percentage increase than did patients with F2 and F3 varices and with SPSS (Table 8). Indeed, patients with F2 and with F3 varices had not had any PV dilation through the study period. After a meal, PFV showed a significant, persistent percentage increase that was inversely related to variceal size. The subgroup comparison showed that only control subjects
and cirrhotics without varices had a marked sustained percentage increase of PFV up to the end of the study. The differences in PV and PFV found in subgroups of different variceal size (Table 8) disappeared if patients with SPSS were included in the various groups of esophageal varices. The negative SMA-RI percentage change found in each group after a meal was significantly greater and more persistent in controls than in cirrhotics. The subgroup analysis showed that during the first 30 min, control subjects and patients without varices had a similar percentage change in SMA-RI that was higher than in patients with F3 and with SPSS. However, each group of cirrhotic patients showed a more transient, negative SMA-RI percentage change with respect the control group. No differences were found in the percentage change of HR, systolic blood pressure, and diastolic blood pressure.
DISCUSSION It has been shown that different patterns of splanchnic hemodynamics exist as baseline values (1) and during the 24-h period (2) in patients without varices, with varices of different size, and with SPSS (2). Patients with cirrhosis have a different splanchnic hemodynamic response to meal ingestion in comparison to normal subjects (8 –15). A follow-up study has also shown that a blunted splanchnic hemodynamic response to a meal was independently associated with death and variceal bleeding (21). A recent study found that the increase of postprandial splanchnic hyperemia is inversely correlated with the severity of liver function impairment (16). In the latter study, patients with varices had a smaller increase of PFV after meal ingestion with respect to those without varices, but no difference was found according to the degree of variceal size. The same occurs in our study if patients with SPSS are not considered separately. It has also been reported that in cirrhotic patients, not selected with regard to the risk of variceal hemorrhage, the Doppler ultrasound hemodynamic response to meal ingestion, expressed either as SMA vasodilatation and/or PFV increase, is smaller and shorter in duration (8, 11, 13, 14) than that in normal subjects (8, 10, 11, 13, 14). In particular, the PV diameter significantly increases in normal subjects
AJG – February, 2001
Postprandial Splanchnic Hemodynamics
553
Table 4. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F1 Varices (n ⫽ 10) Variable
T0
30 min
60 min
120 min
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
12 (⫾2) 21 (⫾5)ab 84 (⫾6)abc 0.6 (⫾3) 126 (⫾20) 75 (⫾12) 73 (⫾9)ab
13 (⫾2) 24 (⫾5)a 75 (⫾6)a 0.5 (⫾2) 124 (⫾18) 75 (⫾9) 79 (⫾13)a
13 (⫾2) 25 (⫾5)b 77 (⫾5)b 0.5 (⫾2) 127 (⫾21) 73 (⫾12) 78 (⫾10)b
12 (⫾2) 23 (⫾5) 80 (⫾4)c 0.6 (⫾2) 124 (⫾24) 73 (⫾12) 75 (⫾11)
p NS ⬍ 0.009 abc ⬍ 0.01 NS NS NS ab ⬍ 0.05 ab
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 5. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F2 Varices (n ⫽ 12) Variable PV (mm) PFV (cm/s) SMA CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0 13 (⫾2) 19 (⫾8)abc 86 (⫾3)ab 0.7 (⫾3) 133 (⫾24) 80 (⫾12)abc 72 (⫾4)a
30 min
60 min
120 min
p
13 (⫾2) 24 (⫾5)a 79 (⫾4)a 0.6 (⫾3) 128 (⫾20) 74 (⫾13)a 75 (⫾4)a
13 (⫾2) 24 (⫾6)b 80 (⫾4)b 0.6 (⫾2) 126 (⫾19) 74 (⫾11)b 776 (⫾9)
13 (⫾2) 22 (⫾7)c 83 (⫾3) 0.6 (⫾2) 127 (⫾14) 74 (⫾12)c 74 (⫾5)
NS abc ⬍ 0.04 ab ⬍ 0.003 NS NS abc ⬍ 0.03 a ⬍ 0.02
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 6. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F3 Varices (n ⫽ 9) Variable
T0
30 min
60 min
120 min
p
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
13 (⫾1) 20 (⫾6) 86 (⫾4)abc 0.8 (⫾3) 129 (⫾23) 76 (⫾11) 65 (⫾6)ab
13 (⫾1) 21 (⫾5) 81 (⫾5)a 0.7 (⫾2) 130 (⫾24) 77 (⫾10) 69 (⫾6)a
13 (⫾1) 21 (⫾5) 82 (⫾4)b 0.7 (⫾2) 132 (⫾19) 77 (⫾10) 69 (⫾6)b
13 (⫾1) 20 (⫾6) 84 (⫾3)c 0.7 (⫾2) 133 (⫾20) 78 (⫾10) 69 (⫾7)
NS NS abc ⬍ 0.02 NS NS NS ab 0.02
SBP ⫽ systolic blood pressire, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 7. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and SPSS (Spontaneous Porta-Systemic Shunts (n ⫽ 13) Variable PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0 12 (⫾2)abc 19 (⫾5) 81 (⫾5)ab 0.7 (⫾3) 134 (⫾16) 80 (⫾10)ab 73 (⫾9)
30 min
60 min
120 min
p
13 (⫾2)a 20 (⫾6) 75 (⫾5)a 0.7 (⫾4) 129 (⫾20) 75 (⫾11)a 77 (⫾13)
13 (⫾2)b 20 (⫾5) 77 (⫾7)b 0.7 (⫾3) 132 (⫾21) 74 (⫾11)b 76 (⫾11)
13 (⫾2)c 19 (⫾5) 80 (⫾3) 0.7 (⫾3) 133 (⫾23) 76 (⫾13) 76 (⫾13)
⬍ 0.01 NS ab ⬍ 0.04 NS NS abc ⬍ 0.03 NS
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
abc
554
Siringo et al.
AJG – Vol. 96, No. 2, 2001
Table 8. Comparisons of Hemodynamic Responses to the Meal Among Controls and Patients With Cirrhosis in Relation to the Presence of SPSS and Variceal Size % Change
Variable PV (mm)
PFV (cm/s)
SMA-RI HR (beats/min) SBP (mm Hg) DBP (mm Hg)
Minutes 30
Controls (n ⫽ 7), % (⫾ SD) ⫹32 (⫾28)
Shunts F0 (n ⫽ 12), F1 (n ⫽ 11), F2 (n ⫽ 12), F3 (n ⫽ 9), (n ⫽ 13), % (⫾ SD) % (⫾ SD) % (⫾ SD) % (⫾ SD) % (⫾ SD) ⫹17 (⫾12)ab
⫹5 (⫾12)
0 (⫾6)a
0 (⫾5)b
⫹8 (⫾9)
60 120 30
⫹17 (⫾23)abc ⫹14 (⫾10)de ⫹6 (⫾12) ⫺2 (⫾8)ad ⫺1 (⫾5)be ⫹6 (⫾8)c ⫹8 (⫾12) ⫹11 (⫾9)ab ⫹4 (⫾17) 0 (⫾9)a ⫺3 (⫾4)b ⫹7 (⫾8) ⫹66 (⫾47)abcd ⫹51 (⫾33)efgh ⫹19 (⫾21)ae ⫹29 (⫾22)bf ⫹5 (⫾16)cg ⫹10 (⫾13)dh
60 120 30 60 120 30 60 120 30 60 120 30 60 120
⫹38 (⫾32)ab ⫹38 (⫾33)cd ⫹35 (⫾40)abc ⫹23 (⫾23)de ⫺13 (⫾7)ab ⫺13 (⫾5)cd ⫺9 (⫾2) ⫺9 (⫾6) ⫺10 (⫾5)abcde ⫺3 (⫾5)a ⫹12 (⫾19) ⫹7 (⫾7) ⫹6 (⫾19) ⫹8 (⫾11) ⫹7 (⫾22) ⫹5 (⫾7) 0 (⫾6) ⫺5 (⫾10) ⫺3 (⫾5) ⫺5 (⫾7) ⫺4 (⫾12) ⫺1 (⫾12) ⫺10 (⫾12) ⫺5 (⫾12) ⫺8 (⫾5) ⫺8 (⫾11) ⫺11 (⫾9) ⫺4 (⫾14)
⫹23 (⫾26) ⫹26 (⫾23) ⫹11 (⫾18)a ⫹17 (⫾20) ⫺11 (⫾4) ⫺8 (⫾6) ⫺8 (⫾4) ⫺6 (⫾6) ⫺4 (⫾4)b ⫺3 (⫾5)c ⫹8 (⫾10) ⫹5 (⫾7) ⫹7 (⫾10) ⫹6 (⫾11) ⫹3 (⫾12) ⫹3 (⫾6) ⫺1 (⫾5) ⫺3 (⫾8) ⫹1 (⫾4) ⫺5 (⫾8) ⫺2 (⫾9) ⫺3 (⫾11) ⫹1 (⫾9) ⫺7 (⫾9) ⫺3 (⫾8) ⫺7 (⫾8) ⫺3 (⫾12) ⫺6 (⫾8)
⫹6 (⫾16)ac ⫹1 (⫾8)bd ⫺6 (⫾6)ac ⫺5 (⫾2) ⫺3 (⫾3)d ⫹7 (⫾8) ⫹6 (⫾9) ⫹7 (⫾10) 0 (⫾7) ⫹3 (⫾10) ⫹4 (⫾8) ⫹1 (⫾10) ⫹2 (⫾8) ⫹2 (⫾6)
⫹6 (⫾15)bd ⫹2 (⫾12)ce ⫺7 (⫾7)bd ⫺5 (⫾9) ⫺2 (⫾5)e ⫹4 (⫾9) ⫹3 (⫾9) ⫹3 (⫾10) ⫺4 (⫾7) ⫹2 (⫾8) 0 (⫾10) ⫹5 (⫾9) ⫺6 (⫾8) ⫺5 (⫾14)
ANOVA p
Duncan Test P ⱕ 0.05
⬍0.00005 Control vs each group. a; b 0.002 a, b, c, d, e 0.05 a, b ⬍0.00005 a, b, c, d, e, f, g, h 0.008 a, b, c, d 0.009 a, b, c, d, e 0.03 a, b, c, d NS 0.05 a, b, c, d, e NS NS NS NS NS NS NS NS NS
NS ⫽ nonsignificant. * All are in relation to T0.
after a meal (11, 13, 15), whereas it has been reported to transiently increase (11) or to not change at all (13, 15) in cirrhotic patients after a meal. The changes of PFV and of SMA impedance indexes were reported to peak at 30 min after food intake both in control subjects and in unselected patients with cirrhosis of the liver (8, 11, 13), but the extent of the changes were smaller and the duration shorter in cirrhotic patients than in controls (8, 11, 13, 14). In summary, the studies published so far have given evidence that cirrhotic patients generally have a more transient postprandial hyperemia than healthy subjects. The present study confirms these previous results and reports new additional findings, namely that the hemodynamic response of the splanchnic circulation to a meal in cirrhotic patients is influenced by the presence and size of esophageal varices and by the presence of SPSS. These previous studies hypothesized that cirrhotic patients have a congestive fasting splanchnic vascular bed that reduces the response to a meal (11, 13) or, as an alternative, that an escape of blood occurs through portal-systemic spontaneous shunting (8, 13, 16). We also support these hypotheses, and particularly the role of portal and splanchnic arterial congestion, to explain the progressive blunting of the postprandial hemodynamic variations in patients with increasing degree of esophageal varices. Indeed, the lack of PV dilation after meal intake in patients with varices may be due to a higher fasting PV congestion (1), reducing the PV compliance to the postprandial blood overinflow. Because of the less congested PV
fasting state (1), patients without varices had a significant and longer postprandial PV dilation with respect to all groups of patients with varices. As a whole, all groups of patients with cirrhosis had a smaller and shorter relative SMA vasodilatation; that implies a smaller percentage of postprandial blood overinflow to the splanchnic bed than in normal subjects. This also may be due, at least partially, to the vasodilation already present in basal conditions, as recently demonstrated by our group (22), which relatively reduces the potential extent of further vasodilation after a meal. The SMA dilation, whatever its extent, determined an increase of PFV in all groups, except in patients with F3 and in those with SPSS. In fact, the increase of the postprandial PFV was maximal in controls subjects and then became progressively smaller as variceal status worsened or as SPSS were detectable at ultrasound examination (Table 8). The various factors causing portal congestion, among which should be considered not only high portal resistance but also SMA and splenic overinflow, may contribute to the disappearance of the PFV increase in patients with large varices or SSPS: they likely act by blocking from downstream (through high intrahepatic portal resistance) further velocity increases and by allowing only limited further decreases of the splanchnic arterial resistance after a meal (because this resistance is already low in the basal state), hence leading to relatively smaller upstream postprandial overinflow. From a hemodynamic point of view, large varices were
AJG – February, 2001
equivalent to SPSS in our study. Indeed, the measurement of azygos venous blood flow (23) showed that large varices divert more blood to systemic venous beds than small ones (24), acting as larger shunts, and that patients with portal hypertension tend to have increased azygos blood flow after a meal (25–28) that may prevent or reduce the postprandial increase of portal pressure. The different behavior of PV after food intake between patients with large varices and those with SPSS may be due to the fact that SPSS decompresses to some extent the splanchnic bed in the fasting state (3–7), whereas large varices do not (29). Our study shows that although 85% of patients with SPSS had F1 or had no varices at all, their splanchnic hemodynamic postprandial response was significantly different from that found in patients with small or no varices and without SPSS and was similar to that of patients with F3 varices: namely, SMA dilation with absent or slight increase of PFV. In conclusion, our data indicate that any investigation of the effects of physiological and pharmacological stimuli on the splanchnic circulation of cirrhotic patients should be done by taking into account not only the risk of bleeding, as expressed by variceal size, but also the presence of SPSS.
Postprandial Splanchnic Hemodynamics
10.
11.
12. 13.
14.
15. 16.
Reprint requests and correspondence: Luigi Bolondi, M.D., Divisione di Medicina Interna, University of Bologna, Via Albertoni 15, 40138 Bologna, Italy. Received Jan. 31, 2000; accepted Aug. 28, 2000.
17.
REFERENCES
19.
1. Siringo S, Bolondi L, Sofia S, et al. The relationship of endoscopy, portal Doppler ultrasound flowmetry, and clinical and biochemical tests in cirrhosis. J Hepatol 1994;20: 11–18. 2. Siringo S, Piscaglia F, Zironi G, et al. Daily modifications of splanchnic hemodynamics in cirrhotics with different degree of variceal bleeding risk. Hepatology 1995;22:492A (abstract). 3. Sacerdoti D, Bolognesi M, Bombonato G, et al. Paraumbilical vein patency in cirrhosis: effects on hepatic hemodynamics evaluated by Doppler sonography. Hepatology 1995;22:1689 – 94. 4. Wexler MJ, MacLean LD. Massive spontaneous portal-systemic shunting without varices. Arch Surg 1975;110:995– 1003. 5. Braillon A, Lee SS, Girod C, et al. Effet du de´veloppement des shunt porto-syste´miques dans le maintien de l’hypertension portale chez le rat. Gastroenterol Clin Biol 1988;12:641–5. 6. Ohnishi K, Nakayama T, Koen H, et al. Interrelationship between type of spontaneous portal systemic shunt and portal vein pressure in patients with liver disease. Am J Gastroenterol 1985;80:561– 4. 7. Watanabe K, Kimura K, Matsutani S, et al. Portal hemodynamic in patients with gastric varices—a study in 230 patients with esophageal and/or gastric varices using portal vein catheterization. Gastroenterology 1988;95:434 – 40. 8. Pugliese D, Ohnishi K, Tsunoda T, et al. Portal hemodynamics after meal in normal subjects and in patients with chronic liver disease studied by echo-Doppler flowmetry. Am J Gastroenterol 1987;82:1052– 6. 9. O’Brien, Keoghan M, Afdhal N, et al. Postprandial changes in
18.
20.
21.
22.
23.
24.
25.
26. 27.
555
portal hemodynamics in patients with cirrhosis. Gastroenterology 1989;96:A640 (abstract). Moneta LG, Taylor DC, Helton WS, et al. Duplex ultrasound measurement of postprandial intestinal blood flow: effect of meal composition. Gastroenterology 1988;95: 1294 –301. Sabba` C, Ferraioli G, Genecin P, et al. Evaluation of postprandial hyperemia in superior mesenteric artery and portal vein in healthy and cirrhotic humans: an operator-blind echoDoppler study. Hepatology 1991;13:714 – 8. Lee SS, Hadengue A, Moreau R, et al. Postprandial hemodynamic response in patients with cirrhosis. Hepatology 1988; 8:647–51. Okazaki K, Miyazaki M, Onishi S, et al. Effects of food intake and various extrinsic hormones on portal blood flow in patients with liver cirrhosis demonstrated by pulsed Doppler with the Octoson. Scand J Gastroenterol 1986;21: 1029 –38. Darnault P, Bretagne JF, Raoul JL, et al. Assessment of splanchnic vascular resistance in cirrhotic patients by measuring the pulsatility index of superior mesenteric artery blood velocity waveforms. Gastroenterology 1988;95:A53 (abstract). Gaiani S, Bolondi L, Li Bassi S, et al. Effect of meal on portal hemodynamics in healthy humans and in patients with chronic liver disease. Hepatology 1989;9:815–9. Ludwig D, Schwarting K, Korbel CM, et al. The postprandial portal flow is related to the severity of portal hypertension and liver cirrhosis. J Hepatol 1998;28:631– 8. Pugh RNH, Murray-Lyon IM, Dawson JL, et al. Transection of the esophagus for bleeding esophageal varices. Br J Surg 1973;60:646 –9. Japanese Research Society for Portal Hypertension. The general rules for recording endoscopic findings on esophageal varices. Jpn J Surg 1980;10:84 –7. Moriyasu F, Nishida O, Ban N, et al. “Congestion Index” of the portal vein. AJR Am J Roentgenol 1986;143:735–9. Sabba` C, Merkel C, Zoli M, et al. Interobserver, and interequipment variability of echo-Doppler examination of the portal vein. Effect of a cooperative training program. Hepatology 1995;21:428 –33. De Vries PJ, de Hooge P, Hoekstra JBL, et al. Blunted postprandial reaction of portal venous flow in chronic liver disease, assessed with duplex Doppler: significance for prognosis. J Hepatol 1994;21:966 –73. Piscaglia F, Gaiani S, Gramantieri L, et al. Superior mesenteric artery impedance in chronic liver diseases: relationship with disease severity and portal circulation. Am J Gastroenterol 1998;93:1925–30. Bosch J, Groszmann RJ. Measurement of azygos venous blood flow by a continuous thermal dilution technique: an index of blood flow through gastroesophageal collaterals in cirrhosis. Hepatology 1984;4:424 –9. Bosch J, Mastai R, Kravetz D, et al. Measurement of azygos venous blood flow in the evaluation of portal hypertension in patients with cirrhosis. Clinical and haemodynamic correlations in 100 patients. J Hepatol 1985;1:125–39. Navasa M, Chesta J, Bosch J, Different effects of protein and charbohydrate meals on splanchnic haemodynamics and portal pressure in patients with cirrhosis. J Hepatol 1988;7:S156 (abstract). Lee SS, Hadengue A, Moreau R, et al. Postprandial hemodynamic responses in patients with cirrhosis. Hepatology 1988; 8:647–51. McCormick PA, Biagini MR, Dick R, et al. Octreotide inhibits the meal-induced increases in the portal venous pressure of virrhotic patients with portal hypertension: a double-blind, placebo-controlled study. Hepatology 1992;16:1180 – 6.
556
Siringo et al.
28. Bendtsen F, Simonsen L, Henriksen H. Effect on hemodynamics of a liquid meal alone and in combination with propranolol in cirrhosis. Gastroenterology 1992;102:1017–23.
AJG – Vol. 96, No. 2, 2001
29. Chao Y, Lin HC, Lee F-Y, et al. Hepatic hemodynamic in patients with esophageal or gastric varices. J Hepatol 1993; 19:85–9.
Vol. 96, No. 2, 2001 ISSN 0002-9270/01/$20.00 PII S0002-9270(00)02351-0
Influence of Esophageal Varices and Spontaneous Portal-Systemic Shunts on Postprandial Splanchnic Hemodynamics Sebastiano Siringo, M.D., Fabio Piscaglia, M.D., Gianni Zironi, M.D., Soccorsa Sofia, M.D., Stefano Gaiani, M.D., Marcello Zammataro, M.D., and Luigi Bolondi, M.D. Dipartimento di Medicina Interna e Gastroenterologia, Universita´ di Bologna, Bologna; and Divisione di Medicina Generale 2, Ospedale Garibaldi, Catania, Italy
OBJECTIVE: The aim of the study was to assess postprandial splanchnic hemodynamic changes in cirrhosis in relation to variceal status. METHODS: In 9 healthy controls and 56 patients with liver cirrhosis, stratified according to variceal status and presence of spontaneous portal-systemic shunts, the portal vein diameter and flow velocity, the congestion index of the portal vein, and the resistive index of the superior mesenteric artery (SMA-RI) were studied by Doppler ultrasound before and 30, 60, and 120 min after the intake of a standard meal. Comparison of postprandial parameters with basal ones was done within each group by paired t test and among groups by ANOVA and Duncan test. RESULTS: Healthy controls and cirrhotic patients without varices showed similar significant splanchnic hemodynamic changes, namely a reduction of SMA-RI (⫺13% at 30 min) and a consequent increase in portal vein diameter (respectively, ⫹ 32% and ⫹ 17% in the two groups) and velocity (⫹66% and ⫹ 51%). A significant reduction of SMA-RI was also found in patients with varices, irrespective of the variceal size (range, ⫺7 to ⫺11%), but the expected portal vein dilation and velocity increase were progressively blunted with the increase of variceal size (range, 0 –5% for diameter and 5–19% for velocity). Patients with spontaneous portal-systemic shunts showed a response similar to that of patients with large varices. Significant modification of the congestion index of the portal vein did not occur in any group. CONCLUSIONS: Our results show that the hemodynamic response to meal in patients with liver cirrhosis is influenced by the presence and size of esophageal varices and the presence of spontaneous portal-systemic shunts. (Am J Gastroenterol 2001;96:550 –556. © 2001 by Am. Coll. of Gastroenterology)
Current address for Sebastiano Siringo: Divisione di Medicina Generale 2, P.O. Garibaldi, P.zza S. Maria di Gesu` 7, 95124 Catania, Italy.
INTRODUCTION Patients with liver cirrhosis and esophageal varices of different size have different patterns of basal splanchnic hemodynamics, as evaluated by Doppler ultrasound (1). Moreover, it has been shown that the patterns of daily changes in splanchnic hemodynamic parameters are related to variceal size and spontaneous portal-systemic shunts (2). It has also been found that the development of portal-systemic shunts can modify portal hemodynamics in patients with liver cirrhosis (3–7) Many studies investigated the portal hemodynamic response to meal intake in patients with cirrhosis (8 –16). However, the results concerning modification of portal blood flow velocity after a meal are conflicting. A possible reason for this might be that patients were not stratified according to the presence and size of esophageal varices. Because variceal size is associated with different baseline splanchnic hemodynamic patterns (1, 2), a different response to the same stimulation may also be hypothesized. A recent study found that patients with varices have a different portal hemodynamic response to meals in comparison to those without varices, although no difference was found among patients with different variceal size (16). However, it remains to be investigated whether also considering the presence of large spontaneous portal-systemic shunts (SPSS), which influence splanchnic hemodynamics, could produce more specific results for the various variceal classes. The aim of the present study was therefore to assess the splanchnic hemodynamic changes occurring after meal intake in cirrhotic patients with different risk of variceal bleeding according to the variceal size and to the presence of SPSS.
MATERIALS AND METHODS We studied 9 healthy controls and 56 patients with liver cirrhosis in whom a complete abdominal ultrasound and duplex Doppler evaluation of the splanchnic hemodynamics was feasible. The severity of liver disease was assessed
AJG – February, 2001
Postprandial Splanchnic Hemodynamics
Table 1. Characteristics of the Study Population Characteristic
Data
Normal subjects M/F Mean age, yr (⫾SD) Cirrhotic patients M/F Mean age, yr (⫾SD) Etiology of cirrhosis Viral Alcoholic Viral & alcoholic PBC Cryptogenic Child-Pugh class (A/B/C) Esophageal varices (F0/F1/F2/F3) Spontaneous portal-systemic shunt
9 6/3 57.2 (⫾ 13.4) 56 34/22 57.9 (⫾ 9.5) 39 10 4 2 1 28/24/4 16/16/13/11 13
according to the Pugh modification of the Child criteria (17). All patients had upper-GI endoscopy at entry into the study, and esophageal varices were classified according to the endoscopic rules of the Japanese Research Society for Portal Hypertension (18). In all cirrhotic patients without varices, the presence of liver cirrhosis was previously proven by needle liver biopsy, whereas in all the remaining patients, it was demonstrated either by a previous liver biopsy or by unequivocal biochemical, endoscopic, and ultrasonographic findings. The presence of spontaneous portal-systemic shunts (SPSS) was searched for by abdominal US in each patient. Table 1 shows patient characteristics. Twelve patients had no esophageal varices (F0), 10 had F1, 12 had F2, and 9 had F3 varices. Thirteen patients had large SPSS visible at duplex Doppler (umbilical and/or splenorenal); 6 (46.2%) of them had no varices, 5 (38.5%) had F1, and 2 (15.4%) had F3 varices. The following parameters were assessed by an EsaoteHitachi AU590 asynchronous echo color Doppler machine with a 3.5-MHz convex transducer. Portal-vein caliber (PV) was measured at the crossing point with the hepatic artery or just distally to it. Portal-vein blood flow velocity (PFV) was considered the time-averaged maximum velocity over a period of 4 – 6 s. The sample volume was positioned at the crossing point of the portal vein with the hepatic artery and set as wide as ⱖ50% of the vein caliber. The Doppler insonation angle was kept between 30° and 60°. The Con-
551
gestion Index of the portal vein was calculated according to Moriyasu’s formula (19): (cross sectional area/mean PFV)/10 Cross-sectional area was derived from the diameter, assuming a circular shape of the vessel, and mean PFV was the time-averaged maximum velocity ⫻ 0.57 (19). The superior mesenteric artery Resistive Index (SMA-RI) was derived on the proximal tract of the vessel according to the following formula: peak systolic frequency ⫺ end-diastolic frequency/ peak systolic frequency Arterial blood pressure and heart rate (HR) were also measured at each time point. All measurements were begun at 01:00 PM, which is the usual lunch time in Italy, after an overnight fast. After the basal measurements (T0), the subjects were given a standard 1000-kcal isocaloric meal (50% carbohydrates, 20% proteins, 30% lipids), and the measurements were repeated 30, 60, and 120 min, after the end of a meal. All measurements were taken by three skilled operators (S.S., F.P., G.Z.) who had a long joint training, which is important in reducing interobserver variability (20). Each patient was examined by only a single operator, which kept the technical approach consistent at every time point of the procedure. Examinations were performed after the subject had been lying down for ⱖ15 min. Each Doppler measurement was taken during suspended normal respiration. For every parameter, three to six consistent measurements were taken, and the results expressed as their mean. Results were expressed as mean absolute values (⫾ SD). The comparison of the T0 values with those obtained at each following time point was done within each group by the paired t test. The differences with respect to T0 were also expressed as percentage changes to overcome the interindividual variability in absolute basal values; the comparison among groups was assessed by the one-way ANOVA. The source of significance among groups within the ANOVA test was identified by the Duncan test.
Table 2. Hemodyanmic Responses to Meal in Healthy Controls (n ⫽ 9) Parameter
T0
30 min
60 min
120 min
p
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
10 (⫾3)a 26 (⫾5)ab 84 (⫾5)abc 0.3 (⫾2) 121 (⫾14) 80 (⫾11)abc 70 (⫾11)
12 (⫾2)a 41 (⫾8)a 73 (⫾7)a 0.3 (⫾1) 121 (⫾17) 71 (⫾11)a 77 (⫾14)
11 (⫾2) 34 (⫾6)b 76 (⫾3)b 0.3 (⫾1) 117 (⫾14) 73 (⫾9)b 74 (⫾15)
10 (⫾3) 34 (⫾10) 76 (⫾5)c 0.3 (⫾2) 116 (⫾16) 71 (⫾11)c 74 (⫾14)
⬍ 0.01 ab ⬍ 0.006 abc ⬍ 0.003 NS NS abc ⬍ 0.05 NS
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, HR ⫽ heart rate, NS ⫽ nonsignificant.
a
552
Siringo et al.
AJG – Vol. 96, No. 2, 2001
Table 3. Hemodynamic Responses to Meal in Patients With liver cirrhosis and without varices (group; n ⫽ 12) Parameter PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0
30 min abc
10 (⫾2) 22 (⫾7)abc 85 (⫾5)abc 0.5 (⫾3) 132 (⫾19)a 80 (⫾10)a 68 (⫾10)abc
a
12 (⫾1) 32 (⫾9)a 75 (⫾7)a 0.4 (⫾2) 126 (⫾16) 76 (⫾10) 73 (⫾11)a
60 min b
12 (⫾1) 30 (⫾9)b 78 (⫾7)b 0.4 (⫾2) 125 (⫾17)a 73 (⫾10)a 73 (⫾10)b
120 min c
11 (⫾2) 26 (⫾7)c 82 (⫾6)c 0.4 (⫾2) 129 (⫾16) 76 (⫾9) 71 (⫾10)c
p ⬍ 0.0005 ⬍ 0.002 abc ⬍ 0.003 NS a ⬍ 0.02 a 0.03 abc ⬍ 0.009 abc abc
SBP ⫽ systolic blood pressure, DBP ⫽ diastoic blood pressure, NS ⫽ nonsignificant.
RESULTS Healthy controls showed a mild portal vein dilation with a marked increase of PFV and a sustained decrease of SMA-RI after meal ingestion (Table 2). A persistent decrease of the diastolic blood pressure was also observed (Table 2). Cirrhotic patients without varices had a similar hemodynamic response to the meal, with a persistent and significant increase of portal vein diameter, PFV, and HR and a decrease in SMA-RI and systemic blood pressure (Table 3). Patients with F1 and F2 varices had a significant decrease of SMA-RI and increased PFV without any significant PV dilation (Tables 4 and 5). In both groups, the HR increased after a meal (Tables 4 and 5). Patients with F3 varices showed only a persistent decrease of SMA-RI after a meal, without any modification of portal vein diameter and PFV (Table 6). Patients with SPSS had a slight and persistent PV dilation, no increase in the PFV, and a shorter duration of the SMA-RI reduction (Table 7). Moreover, control subjects and patients without varices had the maximum increase in PFV at 30 min, whereas cirrhotics with F1 and F2 had a rather steady PFV increase, and patients with F3 and SPSS had no modification of PFV at all. Because the CI did not show any change in the abovementioned analysis, it was not included in the ANOVA test. Table 8 shows the percentage change of each parameter after a meal with respect to T0 in control subjects and in patients with cirrhosis, stratified according to the variceal size and presence of SPSS. The percentage increase of PV was progressively smaller in patients with increasing variceal size and with SPSS. The subgroup comparison showed that the percentage increase of PV in controls was significantly higher within the first 30 min after the meal than in patients without varices, with varices, and with SPSS. After 60 min, the extent of PV change was similar in controls and in patients without varices, and both groups showed a persistently greater PV percentage increase than did patients with F2 and F3 varices and with SPSS (Table 8). Indeed, patients with F2 and with F3 varices had not had any PV dilation through the study period. After a meal, PFV showed a significant, persistent percentage increase that was inversely related to variceal size. The subgroup comparison showed that only control subjects
and cirrhotics without varices had a marked sustained percentage increase of PFV up to the end of the study. The differences in PV and PFV found in subgroups of different variceal size (Table 8) disappeared if patients with SPSS were included in the various groups of esophageal varices. The negative SMA-RI percentage change found in each group after a meal was significantly greater and more persistent in controls than in cirrhotics. The subgroup analysis showed that during the first 30 min, control subjects and patients without varices had a similar percentage change in SMA-RI that was higher than in patients with F3 and with SPSS. However, each group of cirrhotic patients showed a more transient, negative SMA-RI percentage change with respect the control group. No differences were found in the percentage change of HR, systolic blood pressure, and diastolic blood pressure.
DISCUSSION It has been shown that different patterns of splanchnic hemodynamics exist as baseline values (1) and during the 24-h period (2) in patients without varices, with varices of different size, and with SPSS (2). Patients with cirrhosis have a different splanchnic hemodynamic response to meal ingestion in comparison to normal subjects (8 –15). A follow-up study has also shown that a blunted splanchnic hemodynamic response to a meal was independently associated with death and variceal bleeding (21). A recent study found that the increase of postprandial splanchnic hyperemia is inversely correlated with the severity of liver function impairment (16). In the latter study, patients with varices had a smaller increase of PFV after meal ingestion with respect to those without varices, but no difference was found according to the degree of variceal size. The same occurs in our study if patients with SPSS are not considered separately. It has also been reported that in cirrhotic patients, not selected with regard to the risk of variceal hemorrhage, the Doppler ultrasound hemodynamic response to meal ingestion, expressed either as SMA vasodilatation and/or PFV increase, is smaller and shorter in duration (8, 11, 13, 14) than that in normal subjects (8, 10, 11, 13, 14). In particular, the PV diameter significantly increases in normal subjects
AJG – February, 2001
Postprandial Splanchnic Hemodynamics
553
Table 4. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F1 Varices (n ⫽ 10) Variable
T0
30 min
60 min
120 min
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
12 (⫾2) 21 (⫾5)ab 84 (⫾6)abc 0.6 (⫾3) 126 (⫾20) 75 (⫾12) 73 (⫾9)ab
13 (⫾2) 24 (⫾5)a 75 (⫾6)a 0.5 (⫾2) 124 (⫾18) 75 (⫾9) 79 (⫾13)a
13 (⫾2) 25 (⫾5)b 77 (⫾5)b 0.5 (⫾2) 127 (⫾21) 73 (⫾12) 78 (⫾10)b
12 (⫾2) 23 (⫾5) 80 (⫾4)c 0.6 (⫾2) 124 (⫾24) 73 (⫾12) 75 (⫾11)
p NS ⬍ 0.009 abc ⬍ 0.01 NS NS NS ab ⬍ 0.05 ab
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 5. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F2 Varices (n ⫽ 12) Variable PV (mm) PFV (cm/s) SMA CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0 13 (⫾2) 19 (⫾8)abc 86 (⫾3)ab 0.7 (⫾3) 133 (⫾24) 80 (⫾12)abc 72 (⫾4)a
30 min
60 min
120 min
p
13 (⫾2) 24 (⫾5)a 79 (⫾4)a 0.6 (⫾3) 128 (⫾20) 74 (⫾13)a 75 (⫾4)a
13 (⫾2) 24 (⫾6)b 80 (⫾4)b 0.6 (⫾2) 126 (⫾19) 74 (⫾11)b 776 (⫾9)
13 (⫾2) 22 (⫾7)c 83 (⫾3) 0.6 (⫾2) 127 (⫾14) 74 (⫾12)c 74 (⫾5)
NS abc ⬍ 0.04 ab ⬍ 0.003 NS NS abc ⬍ 0.03 a ⬍ 0.02
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 6. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and F3 Varices (n ⫽ 9) Variable
T0
30 min
60 min
120 min
p
PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
13 (⫾1) 20 (⫾6) 86 (⫾4)abc 0.8 (⫾3) 129 (⫾23) 76 (⫾11) 65 (⫾6)ab
13 (⫾1) 21 (⫾5) 81 (⫾5)a 0.7 (⫾2) 130 (⫾24) 77 (⫾10) 69 (⫾6)a
13 (⫾1) 21 (⫾5) 82 (⫾4)b 0.7 (⫾2) 132 (⫾19) 77 (⫾10) 69 (⫾6)b
13 (⫾1) 20 (⫾6) 84 (⫾3)c 0.7 (⫾2) 133 (⫾20) 78 (⫾10) 69 (⫾7)
NS NS abc ⬍ 0.02 NS NS NS ab 0.02
SBP ⫽ systolic blood pressire, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
Table 7. Hemodynamic Responses to Meal in Patients With Liver Cirrhosis and SPSS (Spontaneous Porta-Systemic Shunts (n ⫽ 13) Variable PV (mm) PFV (cm/s) SMA-RI CI SBP (mm Hg) DBP (mm Hg) HR (beats/min)
T0 12 (⫾2)abc 19 (⫾5) 81 (⫾5)ab 0.7 (⫾3) 134 (⫾16) 80 (⫾10)ab 73 (⫾9)
30 min
60 min
120 min
p
13 (⫾2)a 20 (⫾6) 75 (⫾5)a 0.7 (⫾4) 129 (⫾20) 75 (⫾11)a 77 (⫾13)
13 (⫾2)b 20 (⫾5) 77 (⫾7)b 0.7 (⫾3) 132 (⫾21) 74 (⫾11)b 76 (⫾11)
13 (⫾2)c 19 (⫾5) 80 (⫾3) 0.7 (⫾3) 133 (⫾23) 76 (⫾13) 76 (⫾13)
⬍ 0.01 NS ab ⬍ 0.04 NS NS abc ⬍ 0.03 NS
SBP ⫽ systolic blood pressure, DBP ⫽ diastolic blood pressure, NS ⫽ nonsignificant.
abc
554
Siringo et al.
AJG – Vol. 96, No. 2, 2001
Table 8. Comparisons of Hemodynamic Responses to the Meal Among Controls and Patients With Cirrhosis in Relation to the Presence of SPSS and Variceal Size % Change
Variable PV (mm)
PFV (cm/s)
SMA-RI HR (beats/min) SBP (mm Hg) DBP (mm Hg)
Minutes 30
Controls (n ⫽ 7), % (⫾ SD) ⫹32 (⫾28)
Shunts F0 (n ⫽ 12), F1 (n ⫽ 11), F2 (n ⫽ 12), F3 (n ⫽ 9), (n ⫽ 13), % (⫾ SD) % (⫾ SD) % (⫾ SD) % (⫾ SD) % (⫾ SD) ⫹17 (⫾12)ab
⫹5 (⫾12)
0 (⫾6)a
0 (⫾5)b
⫹8 (⫾9)
60 120 30
⫹17 (⫾23)abc ⫹14 (⫾10)de ⫹6 (⫾12) ⫺2 (⫾8)ad ⫺1 (⫾5)be ⫹6 (⫾8)c ⫹8 (⫾12) ⫹11 (⫾9)ab ⫹4 (⫾17) 0 (⫾9)a ⫺3 (⫾4)b ⫹7 (⫾8) ⫹66 (⫾47)abcd ⫹51 (⫾33)efgh ⫹19 (⫾21)ae ⫹29 (⫾22)bf ⫹5 (⫾16)cg ⫹10 (⫾13)dh
60 120 30 60 120 30 60 120 30 60 120 30 60 120
⫹38 (⫾32)ab ⫹38 (⫾33)cd ⫹35 (⫾40)abc ⫹23 (⫾23)de ⫺13 (⫾7)ab ⫺13 (⫾5)cd ⫺9 (⫾2) ⫺9 (⫾6) ⫺10 (⫾5)abcde ⫺3 (⫾5)a ⫹12 (⫾19) ⫹7 (⫾7) ⫹6 (⫾19) ⫹8 (⫾11) ⫹7 (⫾22) ⫹5 (⫾7) 0 (⫾6) ⫺5 (⫾10) ⫺3 (⫾5) ⫺5 (⫾7) ⫺4 (⫾12) ⫺1 (⫾12) ⫺10 (⫾12) ⫺5 (⫾12) ⫺8 (⫾5) ⫺8 (⫾11) ⫺11 (⫾9) ⫺4 (⫾14)
⫹23 (⫾26) ⫹26 (⫾23) ⫹11 (⫾18)a ⫹17 (⫾20) ⫺11 (⫾4) ⫺8 (⫾6) ⫺8 (⫾4) ⫺6 (⫾6) ⫺4 (⫾4)b ⫺3 (⫾5)c ⫹8 (⫾10) ⫹5 (⫾7) ⫹7 (⫾10) ⫹6 (⫾11) ⫹3 (⫾12) ⫹3 (⫾6) ⫺1 (⫾5) ⫺3 (⫾8) ⫹1 (⫾4) ⫺5 (⫾8) ⫺2 (⫾9) ⫺3 (⫾11) ⫹1 (⫾9) ⫺7 (⫾9) ⫺3 (⫾8) ⫺7 (⫾8) ⫺3 (⫾12) ⫺6 (⫾8)
⫹6 (⫾16)ac ⫹1 (⫾8)bd ⫺6 (⫾6)ac ⫺5 (⫾2) ⫺3 (⫾3)d ⫹7 (⫾8) ⫹6 (⫾9) ⫹7 (⫾10) 0 (⫾7) ⫹3 (⫾10) ⫹4 (⫾8) ⫹1 (⫾10) ⫹2 (⫾8) ⫹2 (⫾6)
⫹6 (⫾15)bd ⫹2 (⫾12)ce ⫺7 (⫾7)bd ⫺5 (⫾9) ⫺2 (⫾5)e ⫹4 (⫾9) ⫹3 (⫾9) ⫹3 (⫾10) ⫺4 (⫾7) ⫹2 (⫾8) 0 (⫾10) ⫹5 (⫾9) ⫺6 (⫾8) ⫺5 (⫾14)
ANOVA p
Duncan Test P ⱕ 0.05
⬍0.00005 Control vs each group. a; b 0.002 a, b, c, d, e 0.05 a, b ⬍0.00005 a, b, c, d, e, f, g, h 0.008 a, b, c, d 0.009 a, b, c, d, e 0.03 a, b, c, d NS 0.05 a, b, c, d, e NS NS NS NS NS NS NS NS NS
NS ⫽ nonsignificant. * All are in relation to T0.
after a meal (11, 13, 15), whereas it has been reported to transiently increase (11) or to not change at all (13, 15) in cirrhotic patients after a meal. The changes of PFV and of SMA impedance indexes were reported to peak at 30 min after food intake both in control subjects and in unselected patients with cirrhosis of the liver (8, 11, 13), but the extent of the changes were smaller and the duration shorter in cirrhotic patients than in controls (8, 11, 13, 14). In summary, the studies published so far have given evidence that cirrhotic patients generally have a more transient postprandial hyperemia than healthy subjects. The present study confirms these previous results and reports new additional findings, namely that the hemodynamic response of the splanchnic circulation to a meal in cirrhotic patients is influenced by the presence and size of esophageal varices and by the presence of SPSS. These previous studies hypothesized that cirrhotic patients have a congestive fasting splanchnic vascular bed that reduces the response to a meal (11, 13) or, as an alternative, that an escape of blood occurs through portal-systemic spontaneous shunting (8, 13, 16). We also support these hypotheses, and particularly the role of portal and splanchnic arterial congestion, to explain the progressive blunting of the postprandial hemodynamic variations in patients with increasing degree of esophageal varices. Indeed, the lack of PV dilation after meal intake in patients with varices may be due to a higher fasting PV congestion (1), reducing the PV compliance to the postprandial blood overinflow. Because of the less congested PV
fasting state (1), patients without varices had a significant and longer postprandial PV dilation with respect to all groups of patients with varices. As a whole, all groups of patients with cirrhosis had a smaller and shorter relative SMA vasodilatation; that implies a smaller percentage of postprandial blood overinflow to the splanchnic bed than in normal subjects. This also may be due, at least partially, to the vasodilation already present in basal conditions, as recently demonstrated by our group (22), which relatively reduces the potential extent of further vasodilation after a meal. The SMA dilation, whatever its extent, determined an increase of PFV in all groups, except in patients with F3 and in those with SPSS. In fact, the increase of the postprandial PFV was maximal in controls subjects and then became progressively smaller as variceal status worsened or as SPSS were detectable at ultrasound examination (Table 8). The various factors causing portal congestion, among which should be considered not only high portal resistance but also SMA and splenic overinflow, may contribute to the disappearance of the PFV increase in patients with large varices or SSPS: they likely act by blocking from downstream (through high intrahepatic portal resistance) further velocity increases and by allowing only limited further decreases of the splanchnic arterial resistance after a meal (because this resistance is already low in the basal state), hence leading to relatively smaller upstream postprandial overinflow. From a hemodynamic point of view, large varices were
AJG – February, 2001
equivalent to SPSS in our study. Indeed, the measurement of azygos venous blood flow (23) showed that large varices divert more blood to systemic venous beds than small ones (24), acting as larger shunts, and that patients with portal hypertension tend to have increased azygos blood flow after a meal (25–28) that may prevent or reduce the postprandial increase of portal pressure. The different behavior of PV after food intake between patients with large varices and those with SPSS may be due to the fact that SPSS decompresses to some extent the splanchnic bed in the fasting state (3–7), whereas large varices do not (29). Our study shows that although 85% of patients with SPSS had F1 or had no varices at all, their splanchnic hemodynamic postprandial response was significantly different from that found in patients with small or no varices and without SPSS and was similar to that of patients with F3 varices: namely, SMA dilation with absent or slight increase of PFV. In conclusion, our data indicate that any investigation of the effects of physiological and pharmacological stimuli on the splanchnic circulation of cirrhotic patients should be done by taking into account not only the risk of bleeding, as expressed by variceal size, but also the presence of SPSS.
Postprandial Splanchnic Hemodynamics
10.
11.
12. 13.
14.
15. 16.
Reprint requests and correspondence: Luigi Bolondi, M.D., Divisione di Medicina Interna, University of Bologna, Via Albertoni 15, 40138 Bologna, Italy. Received Jan. 31, 2000; accepted Aug. 28, 2000.
17.
REFERENCES
19.
1. Siringo S, Bolondi L, Sofia S, et al. The relationship of endoscopy, portal Doppler ultrasound flowmetry, and clinical and biochemical tests in cirrhosis. J Hepatol 1994;20: 11–18. 2. Siringo S, Piscaglia F, Zironi G, et al. Daily modifications of splanchnic hemodynamics in cirrhotics with different degree of variceal bleeding risk. Hepatology 1995;22:492A (abstract). 3. Sacerdoti D, Bolognesi M, Bombonato G, et al. Paraumbilical vein patency in cirrhosis: effects on hepatic hemodynamics evaluated by Doppler sonography. Hepatology 1995;22:1689 – 94. 4. Wexler MJ, MacLean LD. Massive spontaneous portal-systemic shunting without varices. Arch Surg 1975;110:995– 1003. 5. Braillon A, Lee SS, Girod C, et al. Effet du de´veloppement des shunt porto-syste´miques dans le maintien de l’hypertension portale chez le rat. Gastroenterol Clin Biol 1988;12:641–5. 6. Ohnishi K, Nakayama T, Koen H, et al. Interrelationship between type of spontaneous portal systemic shunt and portal vein pressure in patients with liver disease. Am J Gastroenterol 1985;80:561– 4. 7. Watanabe K, Kimura K, Matsutani S, et al. Portal hemodynamic in patients with gastric varices—a study in 230 patients with esophageal and/or gastric varices using portal vein catheterization. Gastroenterology 1988;95:434 – 40. 8. Pugliese D, Ohnishi K, Tsunoda T, et al. Portal hemodynamics after meal in normal subjects and in patients with chronic liver disease studied by echo-Doppler flowmetry. Am J Gastroenterol 1987;82:1052– 6. 9. O’Brien, Keoghan M, Afdhal N, et al. Postprandial changes in
18.
20.
21.
22.
23.
24.
25.
26. 27.
555
portal hemodynamics in patients with cirrhosis. Gastroenterology 1989;96:A640 (abstract). Moneta LG, Taylor DC, Helton WS, et al. Duplex ultrasound measurement of postprandial intestinal blood flow: effect of meal composition. Gastroenterology 1988;95: 1294 –301. Sabba` C, Ferraioli G, Genecin P, et al. Evaluation of postprandial hyperemia in superior mesenteric artery and portal vein in healthy and cirrhotic humans: an operator-blind echoDoppler study. Hepatology 1991;13:714 – 8. Lee SS, Hadengue A, Moreau R, et al. Postprandial hemodynamic response in patients with cirrhosis. Hepatology 1988; 8:647–51. Okazaki K, Miyazaki M, Onishi S, et al. Effects of food intake and various extrinsic hormones on portal blood flow in patients with liver cirrhosis demonstrated by pulsed Doppler with the Octoson. Scand J Gastroenterol 1986;21: 1029 –38. Darnault P, Bretagne JF, Raoul JL, et al. Assessment of splanchnic vascular resistance in cirrhotic patients by measuring the pulsatility index of superior mesenteric artery blood velocity waveforms. Gastroenterology 1988;95:A53 (abstract). Gaiani S, Bolondi L, Li Bassi S, et al. Effect of meal on portal hemodynamics in healthy humans and in patients with chronic liver disease. Hepatology 1989;9:815–9. Ludwig D, Schwarting K, Korbel CM, et al. The postprandial portal flow is related to the severity of portal hypertension and liver cirrhosis. J Hepatol 1998;28:631– 8. Pugh RNH, Murray-Lyon IM, Dawson JL, et al. Transection of the esophagus for bleeding esophageal varices. Br J Surg 1973;60:646 –9. Japanese Research Society for Portal Hypertension. The general rules for recording endoscopic findings on esophageal varices. Jpn J Surg 1980;10:84 –7. Moriyasu F, Nishida O, Ban N, et al. “Congestion Index” of the portal vein. AJR Am J Roentgenol 1986;143:735–9. Sabba` C, Merkel C, Zoli M, et al. Interobserver, and interequipment variability of echo-Doppler examination of the portal vein. Effect of a cooperative training program. Hepatology 1995;21:428 –33. De Vries PJ, de Hooge P, Hoekstra JBL, et al. Blunted postprandial reaction of portal venous flow in chronic liver disease, assessed with duplex Doppler: significance for prognosis. J Hepatol 1994;21:966 –73. Piscaglia F, Gaiani S, Gramantieri L, et al. Superior mesenteric artery impedance in chronic liver diseases: relationship with disease severity and portal circulation. Am J Gastroenterol 1998;93:1925–30. Bosch J, Groszmann RJ. Measurement of azygos venous blood flow by a continuous thermal dilution technique: an index of blood flow through gastroesophageal collaterals in cirrhosis. Hepatology 1984;4:424 –9. Bosch J, Mastai R, Kravetz D, et al. Measurement of azygos venous blood flow in the evaluation of portal hypertension in patients with cirrhosis. Clinical and haemodynamic correlations in 100 patients. J Hepatol 1985;1:125–39. Navasa M, Chesta J, Bosch J, Different effects of protein and charbohydrate meals on splanchnic haemodynamics and portal pressure in patients with cirrhosis. J Hepatol 1988;7:S156 (abstract). Lee SS, Hadengue A, Moreau R, et al. Postprandial hemodynamic responses in patients with cirrhosis. Hepatology 1988; 8:647–51. McCormick PA, Biagini MR, Dick R, et al. Octreotide inhibits the meal-induced increases in the portal venous pressure of virrhotic patients with portal hypertension: a double-blind, placebo-controlled study. Hepatology 1992;16:1180 – 6.
556
Siringo et al.
28. Bendtsen F, Simonsen L, Henriksen H. Effect on hemodynamics of a liquid meal alone and in combination with propranolol in cirrhosis. Gastroenterology 1992;102:1017–23.
AJG – Vol. 96, No. 2, 2001
29. Chao Y, Lin HC, Lee F-Y, et al. Hepatic hemodynamic in patients with esophageal or gastric varices. J Hepatol 1993; 19:85–9.