You are on page 1of 1

HEPATOLOGYVol, 34, No. 4, Pt.

2, 2001 AASLD ABSTRACTS 537A

1459 1460
MORTALITY OF GASTROINTESTINAL HEMORRHAGE IN CIRRHOTIC THE INFLUENCE OF SPLEEN ON PORTAL HEMODYNAMICS: A NON
PATIENTS HAS BEEN GREATLY REDUCED IN T%VO DECADES. Arnaud INVASIVE STUDY W I T H DOPPLER ULTRASOUND IN CHRONIC LIVER
Pauwels, Nicolas Carbonell, Olivier Fourdan, Victor Georges L~vy, Raoul Pou- DISEASE AND HEMATOLOGICAL DISORDERS. Fabio Piscaglia, Gabriele
pon, Liver Intensive Care Unit, Paris France Donati, Natascia Celli, Laura Cecilioni, Patrizia Pini, Barbara Stagni, Stefano
]During the past two decades, management of gastrointestinal hemorrhage in Gaiani, Filippo Gherlinzoni, Luigi Bolondi, Univ of Bologna, Bologna Italy
cirrhotic patients has been improved with new vasoactive drugs, endoscopic BACKGROUND AND PURPOSE. Portal hemodynamic changes related to
sclerotherapy and variceal ligation, short-term antibiotic prophylaxis and splenomegalies of different etiology are largely unknown. In particular the
more suitable resuscitatives measures. Aim : To assess improvement of prog- contribute of spleen enlargement to portal hypertension is still controversial.
nosis in cirrhotic patients with gastrointestinal hemorrhage during the 1980- Aim of the present study was therefore to investigate the influence of the spleen
2.000 period. Patients and methods : All cirrhotic patients admitted because of on portal hemodyanamics by assessing splenic and portal Doppler ultrasound
gastrointestinal hemorrhage in our Liver Intensive Care Unit during the years parameters in chronic liver diseases and hematological disorders. SUBJECTS
1L980, 1985, 1990, 1995 and 2000 were retrospectively studied. Results : In- AND METHODS. Study groups were made by: 1) patients with chronic liver
hospital mortality regularly decreased beetween 1980 and 2000 (Table). It was disease (89 with cirrhosis, 35 with chronic hepatitis), 2) patients with spleno-
even more evident when analyzed according to the Child-Pugh's classification megaly without relevant portal hypertension (14 patients with hematological
: in class A patients, mortality decreased from 14 % in 1980 to 0% in 2000; in splenomegaly and 25 liver transplant recipients for cirrhosis without compli-
class B patients, from 44% to 0%; in class C patients, from 76% to 30%. It was cations at 6-24 months after surgery), c) 15 patients with arterial hypertension,
due to a reduction of both bleeding- and sepsis-related mortality (from 26% to d) 22 healthy controls. In all subjects spleen size, portal flow parameters and
7.5% and from 10% to 0%, respectively). In patients with variceal hemorrhage, splenic artery resistance index (RI) were measured by duplex-Doppler ultra-
reduction of bleeding-related mortality was associated with a more and more sound. RESULTS. Splenic artery RI resulted significantly and selectively in-
systematic and combined use of vasoactive drugs (from 17% in 1980 to 100% creased in patients with cirrhosis (0.63, whereas all other groups ranged be-
in 2000) and therapeutic endoscopy in first 24 h (from 0% to 73.5%), and with tween 0.53 and 0.56, p<0.01). Portal flow velocity was significantly decreased
a decrease in early rebleeding incidence. Decrease in bacterial infections was in cirrhosis (p<0.01). The combination of these two parameters provided an
mainly due to introduction of short-term antibiotic prophylaxis in 1990-92. accuracy of 87.5% in distinguishing portal hypertensive from hematological
Conclusion : Mortality of gastrointestinal hemorrhage in cirrhotic patients has splenomegaly. In patients with cirrhosis the degree of spleen enlargement was
been divided by three in two decades. This valuable result has been achieved by associated with increasing portal flow volumes, portal vein diameter and
a reduction of both bleeding- and sepsis-related mortality, concurrently a more variceal size, whereas splenic RI and portal velocity did not differ in connection
and more systematic and combined use of vasoactive drugs, early endoscopic to spleen size. CONCLUSION. Splenomegaly of portal hypertensive or hema-
sclerotherapy or variceal ligation, and short-term antibiotic prophylaxis. tological origin can be differentiated with high accuracy by spleno-portal
Doppler parameters. In patients with cirrhosis, the presence of splenomegaly
1980 1985 1990 1995 2000 suggests a more severe condition, probably due to a more pronounced "for-
% n=80 n=116 n=107 n=I06 n=t18 ward" component of portal hypertension.
Mortality 4Z5 32.8 29 20.8 14,4
Rebleeding 36.2 19,8 17.8 17 11.9
infection 33.7 47,4 23.4 13.2 14.4

1461 1462
CLINICAL CHARACTERIZATION FOR PORTAL HYPERTENSIVE GAS- PLASMA ENDOTHELIN-I (ET-I) LEVELS IN CHILDREN WITH CIRRHO-
TROPATHY -EVALUATIONS FOR ITS SEVERITY AND LIVER FUNC- SIS AND THEIR RELATIONSHIP TO RENAL FUNCTION AND THE SE-
TIONS IN PROGRESS OF CHRONIC LIVER DISEASES-. Yasuhiro Nish- VERITY OF PORTAL HYPERTENSION. Ahmed F Abdalla, Mohammed Ezz
izaki, Masaru Itakura, Shigeyuki Motegi, Chikashi Muramatsu, Fuminori el-Regal, Ashraf M Bakr, Hala El-Marsafa~w, Talal Amer, lkbal Abu-Hashem,
Kobayashi, Tokai University Tokyo Hospital, Tokyo Japan; Syohei Matsuzaki, Mansoura Univ, Mansoura Egypt; Mohammed K Abdel-Khalik, Cairo Univ,
Tokai University, Isehara Japan Cairo Egypt; Hassan H A-Kader, Upstate Medical Univ, Cairo Egypt
Aim: Portal hypertensive gastropathy (PHG) is a major complication in chronic liver Plasma Endothelin-I (ET-I) is a potent vasoconstrictor peptide, which is in-
diseases which causes upper abdominal complaints and hemorrhage. Several classi- volved in the pathogenesis of several disorders. ET-I levels are increased in
fications have been provided but endoscopic findings are not uniform in parts of the adult patients with cirrhosis. However, little is known about ET-I levels in
gastric mucosa. To determine details in relation between endoscopic severity of the children with cirrhosis. Therefore, we measured ET-I levels in 19 children with
PHG and liver functions including degree of portal hypertension in progress of cirrhosis (8 patients with ascites, and 11 without ascites), and 1l age and
chronic liver diseases, 4 different findings of PHG were evaluated at 3 divided
sex-matched normal children by radioimmune assay. We also correlated the
portions of the stomach and were compared. Materials and Methods: 102 patients
with chronic liver diseases (12 chronic hepatitis (CH), 27 cirrhosis with Child A ET-I levels with mean blood pressure, creatinine clearance, and severity of
(LC-A), 13 Child B (LC-B), 31 Child C (LC-C), 19 portal vein obstruction (PVO)) portal hypertension as measured by portal flow volume and portal flow veloc-
who underwent gastroendoscopy were included. Endoscopic findings of PHG were ity. Patients with cirrhosis and ascites had higher plasma ET-I levels compared
classified according to McComacks classification, i.e. superficial reddening (SR), to patients without ascites (6.8 --- 0.62 pg/ml vs. 4.6 + 0.35 pg/ml; p < 0.01)
snake skin appearance (SSA), cherry red spot (CRS), diffuse hemorrhage (DH), and and controls (3.6 -+ 0.27 pg/ml; p < 0.005). Plasma ET-I levels were higher in
were evaluated separately in three portions of stomach, such as upper (U), middle cirrhotic patients without ascites compared to controls (p < 0.005). Plasma
(M), lower (L). Liver functions were evaluated by Childs classification, prothrombin ET-I levels correlated positively with the mean blood pressure (r = 0.58; p <
time (PT%), cholinesterase (Ch-E), and degree of portal hypertension i.e. indocya- 0.05), and negatively with renal function as measured by creatinine clearance
nin green test (k-ICG) and spleen size measured by ultrasound. Patients with Heli- (r = - 0.7; p <0.005). However, no correlation was detected between ET-I
cobacter pylori infection were eliminated. Logistic regression analysis was per- levels and portal flow volume (r = - 0.02; p = 0.4) or portal flow velocity (r =
formed. ANOVA, chi squre were used for statstical analysis. Results: Positive rate - 0.16; p = 0.4). We conclude that plasma ET-t levels are increased in children
PHG findings in CH, LC-A, LC-B, LC-C and PVO were 50, 85, 84, 93 and 100% with cirrhosis, with or without ascites, compared to controls. Patients with
respectively. SR was frequently obsesrved at L (41-69%) but had no relation between cirrhosis and ascites have higher plasma ET-I levels compared to those without
progress of liver disease. SSA was more frequently observed in M (25-55%) and U ascites. The degree of elevation is not related to the severity of portal hyper-
(29--39%) than in L (0-16%) in CH, LC-A, LC-B, LC-C and PVO, however SSA in L
tended to be increased along to progress of liver diseases. Positive rate of CRS in U tension. This increase tends to maintain systemic blood pressure, but is asso-
was higher in CH (17%), LC-A (44%), LC-C (55%) than in L and M in each disease ciated with decline in renal function.
group. DH in U with PVO (47%) was markedly higher than in all portion with and
without PVO. In comparison analysis between each PHG findings and liver func-
tions excluded patients with PVO showed that, PT% was lower in patients with CRS
and also CH-E and pit were lower in CRS and DH. Spleen index was larger than SSA,
and k-ICG was lower in DH when compared with patients with SSA. In Logistic
regression analysis, progress of liver damage (from CH to LC-C) had an independent
effect on the risk for appearances of SSA in L, CRS in U and DH in U in odds ratios
of 2.8, 1.6,10.9 respectively. Conclusions: The incidence of PHG was increased as
liver disease progressed. SR showed significantly higher in rates at L but had no
relation between progress of liver disease. SSA and CRS were appeared along to
progress of liver damage. DH was frequently observed at U in patients with PVO.
CRS and DH reflected liver dysfunction and portal hypertension in progress of liver
diseases.

You might also like