Transjugular Intrahepatic Portosystemic Shunt in Hepatorenal
Syndrome: Effects on Renal Function and Vasoactive Systems
MO ´NICA GUEVARA,1PERE GINE `S,1JUAN CARLOS BANDI,1ROSA GILABERT,2PAU SORT,1WLADIMIRO JIME ´NEZ,3
JUAN CARLOS GARCIA-PAGA ´N,1JAUME BOSCH,1VICENTE ARROYO,1AND JUAN RODE ´S1
SEE EDITORIAL ON PAGE 590
Little information exists on the effects of transjugular
intrahepatic portosystemic shunts (TIPS) in the manage-
ment of cirrhotic patients with hepatorenal syndrome
(HRS). The current study was aimed to prospectively
evaluatetheeffectsofTIPS onrenal functionandvasoactive
systems in patients with type I HRS. Glomerular filtration
rate (GFR) (inulin clearance), renal plasma flow (RPF)
(para-aminohippurate clearance), plasma renin activity
(PRA), aldosterone (ALDO), norepinephrine (NE), and
endothelin (ET) were determined in baseline conditions
and at different time intervals after TIPS in 7 patients with
type I HRS. TIPS induced a marked reduction of portal
pressure gradient (PPG) (20 ? 1 to 10 ? 1 mm Hg; P F
.05). Renal function improved in 6 of the7 patients. Serum
creatinine and blood urea nitrogen (BUN) decreased from
5? 0.8and109? 7to1.8? 0.4mg/dL and56? 11mg/dL,
respectively (PF .05 for both), andGFR andRPF increased
from9 ? 4 and 103 ? 33 to 27 ? 7 mL/min and 233 ? 40
mL/min, respectively (PF .05 for both), 30 daysafter TIPS.
These beneficial effects on renal function were associated
withasignificant (PF .05) reductionofPRA (18? 5to3?
1 ng/mL · h), ALDO (279 ? 58 to 99 ? 56 ng/dL), and NE
(1,257 ? 187 to 612 ? 197 pg/mL). ET did not change
significantly (28 ? 8 to27 ? 11 pg/mL). Mean survival was
4.7 ? 2 months (0.3-17 months). Three patients remained
alive more than 3 months after TIPS insertion. In conclu-
sion, TIPS improves renal function and reduces theactivity
of the renin-angiotensin and sympathetic nervous systems
in cirrhotic patients with type I HRS. Nevertheless, the
efficacy of TIPS inthemanagement of thesepatients should
be confirmed in controlled investigations. (HEPATOLOGY
Hepatorenal syndrome (HRS) is a common and severe
complication of patients with advanced cirrhosis and is
characterized by renal failure, marked portal hypertension,
abnormalities in the arterial circulation, and overactivity of
endogenous vasoactive systems.1-5Up to now, the only
with HRS is liver transplantation.3-7However, becauseof the
short survival of patients with HRS and thelimited availabil-
ity of organs, only a small percentage of patients with HRS
can actually reach transplantation. Therefore, it would be of
great valueto haveatherapeutic method that could improve
renal function and increase survival, especially in patients
with type I HRS, which is characterized by a rapid progres-
sion ofrenal failureandavery short survival expectancy.3
The transjugular intrahepatic portosystemic shunt (TIPS)
has been introduced recently in clinical practice for the
with surgical portosystemic shunts, the use of TIPS is
associated with a marked reduction in portal pressure but
with the advantage of a very low operative morbidity and
mortality. It has been shown that TIPS is useful in the
management of acute variceal hemorrhage that cannot be
succesfully controlled with medical therapy and in the
prevention of recurrent variceal hemorrhage.10-14A number
of studies have also shown that the reduction in portal
pressureinduced by TIPS isassociated with beneficial effects
on renal function and neurohumoral factors in patients with
cirrhosis and ascites, including an increase in sodium and
tor and antinatriuretic systems.15-22By contrast, theeffectsof
TIPS in patientswith HRS arenot well characterized. Several
case reports and preliminary data suggest that TIPS may
improve renal function in cirrhotic patients with HRS.23-27
However, no studies have been published that carefully
investigate the changes in renal function and vasoactive
systems after TIPS in patients with HRS. Therefore, the
current study was designed to prospectively investigate the
effects of TIPS on renal function and vasoactivesystems in a
PATIENTS AND METHODS
This study was designed as a prospective protocol to investigate
the effects of the TIPS on renal function and vasoactive factors in
patientswithHRS. Thestudy wasapprovedby theInvestigationand
Ethics Committee of the Hospital Clı ´ nic of Barcelona. Informed
consent wasobtainedfromall patientsbeforeinclusion in thestudy.
In patients with hepatic encephalopathy, informed consent was
Patients. Seven cirrhotic patients with HRS admitted to the Liver
Unit of theHospital Clı ´ nic of Barcelonawereincluded in thisstudy.
Abbreviations: HRS, hepatorenal syndrome; TIPS, transjugular intrahepatic portosy-
syemic shunt; GFR, glomerular filtration rate; PRA, plasma renin activity; ALDO,
aldosterone; NE, norepinephrine; ET, endothelin; RPF, renal plasma flow; PPG, portal
pressuregradient; BUN, bloodureanitrogen.
Fromthe1LiverUnit andHepatic Hemodynamic Laboratory,2Radiology Department,
August Pi-Sunyer, University ofBarcelona, Barcelona, Catalunya, Spain.
ReceivedFebruary 10, 1998; acceptedMay 11, 1998.
Supportedby agrantfromtheFondodeInvestigacionesSanitarias(FIS/97/2073) and
Direccio ´nGeneral deInvestigacio ´nCientı ´ ficay Te ´ cnica(SAF96-0131). Dr. Guevarawas
supported by a grant from the Fundacio ´Clı ´ nic per la Recerca Biome ` dica. Dr. Sort was
supportedby afellowshipgrant fromtheHospital Clı ´ nic ofBarcelona.
Address reprint requests to: PereGine ` s, M.D., Liver Unit, Hospital Clı ´ nic, Villarroel,
170, 08036 Barcelona, Catalunya, Spain. Fax: 34-93-451-52-72.
Copyright?1998 by theAmerican Association for theStudy of Liver Diseases.
3Hormonal Laboratory, Hospital Clı ´ nic, Institut d’Investigacions Biome ` diques
five patients and on clinical, biochemical, and ultrasonographical
criteriaproposedrecently by theInternational AscitesClub3: 1) low
glomerular filtration rate (GFR), as indicated by serum creatinine
greater than 1.5 mg/dL; 2) absence of shock, ongoing bacterial
infection, fluid losses, and treatment with nephrotoxic drugs; 3) no
improvement ofrenal function after diuretic withdrawal andplasma
volume expansion; 4) proteinuria lower than 500 mg/d; and 5) no
ultrasonographic evidence of parenchymal renal disease or urinary
tract obstruction. Moreover, all patients had low urineoutput, very
low urine sodium (?10 mEq/L), and urine osmolality greater than
plasma osmolality. All patients had type I HRS, characterized by
rapidly progressive reduction of renal function as defined by a
doubling of the initial serum creatinine to a level greater than 2.5
mg/dL or a50%reductionoftheinitial 24-hour creatinineclearance
to a level lower than 20 mL/min in less than 2 weeks.3The
hepatic function in the four patients in whom liver function tests
before deteriorarion of renal function were available. Two of the
peritonitis before the development of HRS, which was resolved at
least1week beforetheirinclusionintothestudy. Theimpairmentof
renal function in thesetwo patients was rapidly progressivedespite
the resolution of the infection. This type of renal impairment after
is associated with a 100% in-hospital mortality.28No precipitating
factors were found in the remaining five patients. No patient had
received nonsteroidal anti-inflammatory drugs before the develop-
mentoftypeI HRS. Exclusioncriteriaincluded: Child-Pughscore?
12, activeinfection, hepatocellularcarcinomaorothermalignancies,
symptomatic cardiac or respiratory diseases, or recent (within 1
week) gastrointestinal bleeding. Of the seven patients included in
the current study, six patients were male, and one was female. The
median age was 50 years (range, 27-68 years). The etiology of
cirrhosiswasalcoholic in four patientsand hepatitisC virus-related
and hepatitis B virus-related in two and one patient, respectively.
One of the 4 patients with alcoholic cirrhosis had been drinking
actively before admission to the hospital. The remaining three
before admission. Four patients had refractory ascites before the
development of type I HRS. Clinical characteristics of patients and
liver andrenal function testsareshown in Table1.
Study Protocol. Patientswerestudiedafteraminimumof7dayson
observation, renal function continued toworsen despitetheadmin-
istration of plasmaexpanders.3After an overnight fast and bed rest,
blood samples weredrawn at 8:00 AM for thedetermination of liver
tests, renal tests, serum electrolytes, plasma renin activity (PRA),
and the plasma concentration of aldosterone (ALDO), norepineph-
rine (NE), and endothelin (ET). Samples were collected in tubes
under ice containing ethylenediaminetetraacetic acid (PRA and
plus reduced glutation (NE) and ethylenediaminetetraacetic acid
plus aprotinin (ET). Samples were centrifuged at 4°C and stored at
?80°C until analysis. Immediately afterward, GFR, renal plasma
flow(RPF), andwaterexcretionweremeasured. A primingdoseof8
mg/kg body weight of aminohippurate (aminohippurate sodium,
Merk Sharp and Dohme, West Point, PA) and of 50 mg/kg body
weight ofinulin(Inutest, LaevosanGesellschaft, Linz, Austria) were
administered intravenously, followed by a constant infusion of a
saline solution containing aminohippurate and inulin at a rate
60 minutes, urine was collected with a bladder catheter in three
20-minute periods. Urine volume was recorded and aliquots were
osmolality. In themiddleof each period, blood samples weretaken
to measureinulin, aminohippurate, and osmolality. During thefirst
45 minutes of theequilibration period, 5 mL/kg body weight of 5%
dextrose were also given intravenously. The administration of 5%
dextrose was kept constant throughout the study by infusing a
volumeof thesolution equal to theurinevolume. Inulin clearance,
asameasureof GFR, aminohippurateclearance, asan estimation of
RPF, and free water clearance were calculated using standard renal
clearanceformulas. Thefinal valuesoftheseparametersgiveninthe
results are the arithmetic mean of the values obtained in the three
The following day, patients were taken to the Hepatic Hemody-
namic Laboratory for the insertion of TIPS, which was performed
remained in theIntensiveLiver CareUnit for at least 24 hours after
tests were determined every 2 to 3 days during the hospitalization
of renal function, including the measurement of GFR, RPF, and
water and electrolyte excretion and vasoactive factors, was per-
formed. Seven days after TIPS, a hepatic vein catheterization was
also performed to assess shunt patency and measure portocaval
During hospitalization, patients were given a diet containing 40
mEq/d of sodium. Protein content was restricted only in patients
withencephalopathy. Patientswithencephalopathy gradeII ormore
were given intravenous fluids. All patients received lactulose or
lactitol for treatment or prevention of encephalopathy. All patients
received furosemide (40-250 mg/6 to 12 hours intravenously) at
somepoint after TIPSplacement tomaintain aurinevolumegreater
than 400 mL/d. However, furosemide was not given for a period of
24 hours before the assessment of renal function at days 7 and 30.
Placement of TIPS and Measurement of Hemodynamic Parameters. A
therapeutic paracentesis followed by intravenous administration of
albumin (8 g/L of ascites removed) was performed before TIPS
insertion in all patients. TIPS placement (Wallstent endoprothesis,
Schneider Europe AG, Bu ¨lach, Switzerland) was performed under
ultrasonographic and radiological guidance as described in detail
TABLE 1. BaselineClinical Characteristics and Renal Function Tests
61 ? 6
— Mean ? SEM6.2 ? 2.8 27 ? 25 ? 0.8109 ? 7
HEPATOLOGY Vol. 28, No. 2, 1998GUEVARA ET AL.
elsewhere.29Measurements of portal venous pressure and inferior
venacavapressureweremadein baselineconditions beforeobtain-
ing a portal venography. A small volume of contrast medium was
used to minimize the risk of nephrotoxicity. After stent placement,
the stent was dilated to an 8-mm diameter and pressure measure-
ments were repeated. If the gradient between inferior vena cava
pressure and portal pressure (PPG) [portal pressure gradient] was
greater than 12 mmHg, thestent was dilated to a10-mmdiameter.
All pressures were measured at least in duplicate using highly
sensitive pressure transducers calibrated before each measurement.
The external zero level was set at the mid-axillary line. Permanent
tracings of pressure measurements were obtained using a Hewlett-
Packard 7754B multichannel recorder (Hewlett-Packard Co.,
Waltham, MA). Antibiotic prophylaxis (ceftriaxone, 2 g intrave-
nously) was given immediately before TIPS insertion. As discussed
previously, 1 week after TIPS insertion, a hepatic vein catheteriza-
tion was performed to determine the PPG. In patients with a PPG
greater than 12 mmHg, thestent was expanded to 10 mmwith the
angioplasty balloon. If PPG did not decreasebelow 12 mmHg after
thismaneuver, acoaxial stent wasinserted.
Analytical Methods. Liver tests, blood ureanitrogen (BUN), creati-
nine, electrolytes, and osmolality were determined using standard
laboratory methods. PRA, ALDO, and ET were measured by
radioimmunoassay as described in detail elsewhere.30,31Theplasma
concentration of NE was measured by radioimmunoassay (IBL
Laboratories, Hamburg, Germany). Normal values of PRA, ALDO,
NE, and ET in our laboratory for healthy subjects on alow-sodium
diet were: 1.2 ? 0.1 ng/mL · h, 24 ? 2 ng/dL, 233 ? 17 pg/mL, and
5 ? 0.3 pg/mL, respectively. Serum and urine inulin and para-
aminohippurate concentrations were measured using colorimetric
Statistical Analysis. Thestatistical analysisof theresultswasmade
Bonferroni comparison. Resultsarepresentedasmeans? SEM. P?
.05 wasconsideredstatistically significant.
fully placed in all cases. A marked decrease in the PPG was
obtainedinall patientsafterTIPSinsertion(pre-TIPS: 20? 1
mm Hg; post-TIPS: 10 ? 1 mm Hg; P ? .001; mean
reduction: 50% ? 4%) (Table 2). In the three patients in
whom the portosystemic gradient 1 week after TIPS was
equal orgreaterthan12mmHg, areductioninthisparameter
was achieved by 10-mm balloon dilation (two patients) or
insertion of a coaxial stent (one patient) (Table 2). One
patient showed a marked increase in serum aminotransfer-
ases and serum bilirubin after TIPS, changes consistent with
the development of ischemic hepatitis. A liver ultrasound
examination disclosed the existence of a fistula between the
hepatic artery and the portal venous system. A progressive
reduction of liver enzymes was observed, and the arteriove-
nous fistula closed spontaneously. No complications related
Effects of TIPS on Renal Function, Arterial Pressure, andVasoac-
tiveSystems. Table3 showsrenal function tests, mean arterial
pressure, and vasoactive substances in baseline conditions
and 7 and 30 days after TIPS insertion. Individual values of
serumcreatininebeforeand after TIPS areshown in Table2.
Before TIPS, patients had markedly increased serum creati-
nine and BUN levels, extremely low RPF and GFR, oliguria,
in renal function were associated with an overactivity of the
renin-angiotensin-aldosterone system and sympathetic ner-
voussystemandincreasedplasmaET levels. Theinsertion of
TIPS induced an improvement of renal function abnormali-
tiesandasuppression oftheactivity ofmajor vasoconstrictor
systems (Table 3). The improvement in renal function was
very slow. Oneweek after treatment, serumcreatinine, BUN,
GFR, and RPF were not significantly different from baseline
values. However, 1 month after TIPS insertion, BUN and
serumcreatininehad decreased markedly, and GFR and RPF
increased twotothreefold over baselinevalues, but werestill
well below normal values(Tables2 and 3). Thesechangesin
renal hemodynamics were associated with a significant in-
crease in urine volume and a nonsignificant increase in free
water clearance and urine sodium. Because of the positive
sodium balance, ascites increased in all patients during the
30-day period. However, only one patient developed tense
ascites during this period, which was treated by paracentesis
and intravenous albumin. Mean arterial pressure did not
change significantly after TIPS. The activity of the renin-
angiotensin-aldosterone system and sympathetic nervous
systemwasmarkedly suppressed after TIPS (Fig. 1, Table3).
Nevertheless, valuesofPRA, ALDO, andNE obtained30days
after TIPS were still higher compared with normal values.
Finally, plasmaET levelsdidnotchangethroughoutthestudy
period. Although no correlations were found between the
reduction in PPG and changes in renal function parameters,
probably because of the low number of patients included,
TABLE 2. Individual Values of Portosystemic PressureGradient and SerumCreatinineBeforeand 7 Days and 30 Days After TIPS
Gradient 1 Week
7 Days After
30 Days After
TIPS (mg/dL)BaselineAfter TIPS
20 ? 1
10 ? 1*
9 ? 2*
Mean ? SEM5 ? 0.8 3.7 ? 1.01.8 ? 0.8*
Abbreviation: NA, not assessed.
*P? .05 comparedwith baseline.
†Valuesofportosystemic gradient after balloon angioplasty.
‡Valueofportosystemic gradient after theinsertion ofacoaxial stent.
§Measuredat 22 daysafter TIPS.
GUEVARA ET AL.HEPATOLOGY August 1998
renal function did not improve in the two patients in whom
PPG didnot decreasebelow 12 mmHg(Table2).
In two of the seven patients, a transient impairment of
Serumcreatinineincreased from4 to 5.9 mg/dL and 5 to 7.7
mg/dL, but decreased rapidly thereafter. None of them
required hemodialysis. This early impairment of renal func-
tionwasattributedtonephrotoxicity by thecontrast medium
used during theinsertion of TIPS and/or progression of HRS.
In two other patients in whom renal function improved
initially, alateimpairment of renal function was observed 32
and 185 days after TIPS. In one of these patients, serum
of apatent shunt, asassessedby Doppler ultrasonography. In
the other patient, serum creatinine increased from 1.4 to 8
mg/dL, and the catheterization of the shunt disclosed the
existence of shunt dysfunction with a PPG of 14 mm Hg.
theremaining patients, serumcreatinineat thetimeof death
or end of follow-up did not differ significantly from values
obtained 30 days after TIPS (1.8 ? 0.8 vs. 2.5 ? 0.6 mg/dL,
Complications and Survival. In the three patients with he-
patic encephalopathy before TIPS insertion, encephalopathy
resolved with standard therapy shortly after TIPS placement
(3 ? 1 days). All of themshowedrecurrenceof encephalopa-
thy during follow-up related to the development of severe
complications (gastrointestinal hemorrhage, renal impair-
ment, and pneumonia in one case each). New encephalopa-
thy developed in two of the four patients without hepatic
encephalopathy before TIPS and was succesfully managed
with medical therapy. The remaining two patients did not
Serum albumin and prothrombin time7 and 30 days after
TIPS werenot significantly different compared with baseline
values. By contrast, serum bilirubin increased significantly
(P ? .05) 7 daysafter TIPS but returned tobaselinevaluesat
30 days (6.2 ? 2.8, 14.1 ? 5.6, and 6.4 ? 3.7 mg/dL,
respectively). When thepatient who developed theischemic
hepatitis after TIPS was excluded fromthis analysis, changes
in serum bilirubin in the remaining patients were still
statistically significant (6.1 ? 2.8, 13.6 ? 6.6, and 6.4 ? 3.2
During follow-up, fivepatients died 9, 22, 35, 45, and 102
days after the insertion of TIPS. Causes of death were liver
failure in two patients, gastrointestinal bleeding caused by
esophageal varices and septicemia in one patient each, and
unknown in one patient. The remaining two patients were
alive 240 and 570 days after the procedure. One of these
patientswassubmittedtoliver transplantation 510 daysafter
TABLE 3. Effects of TIPS on Renal Function, Mean Arterial Pressure
Baseline7 Days 30 DaysP
Mean arterial pres-
PRA (ng/mL · h)
5.0 ? 0.8
109 ? 7
103 ? 33
9 ? 4
3.7 ? 1.0
120 ? 12
138 ? 54
11 ? 5
1.8 ? 0.4*‡
56 ? 11*†‡§ .007
233 ? 40*‡
27 ? 7*‡
126 ? 3 129 ? 2 134 ? 2NS
?0.1 ? 0.06
253 ? 43
0.3 ? 0.2
898 ? 155*‡ 1,080 ? 205*‡
0.1 ? 0.2 NS
2.4 ? 0.46.7 ? 3.09.4 ? 4.2NS
72 ? 4
18 ? 5
279 ? 58
1,257 ? 187
28 ? 7
75 ? 5
6 ? 2
214 ? 92
853 ? 102
33 ? 9
73 ? 3
3 ? 1*‡
99 ? 56*‡
612 ? 197
27 ? 11
NOTE. Normal valuesofPRA, ALDO, NE, andET are1.2? 0.1ng/mL · h,
24 ? 2 ng/dL, 233 ? 17 pg/mL, and5? 0.3 pg/mL, respectively.
Abbreviation: NS, not significant.
†Comparedwith 7 days.
tionsin baselineconditionsand7 and30 daysafter TIPS.
Individual values of PRA and plasma ALDO and NE concentra-
HEPATOLOGY Vol. 28, No. 2, 1998GUEVARA ET AL.
the procedure. Mean survival time in the whole group of
patientswas140 ? 68 days.
The results of the current study indicate that TIPS im-
proves renal hemodynamics and reduces the activity of
vasoconstrictor and antinatriuretic systems in cirrhotic pa-
tients with type I HRS. The improvement of renal function
was characterized by increased renal perfusion, as indicated
by arisein RPF andGFR andmarkedreductionsin BUN and
serum creatinine levels. It is interesting to note that in some
patients, particularly thosewithmoresevererenal failure, the
improvement of renal function did not occur immediately
after TIPS insertion. Rather, the increase in renal perfusion
occurred very slowly and took placeover a period of several
weeks. Similar findings were observed in the few reports
assessing the effects of surgical portacaval shunt in patients
with HRS.34,35Theseresultsarealsoin keeping with previous
studies investigating the effects of TIPS on renal function in
patients with refractory ascites, showing that the improve-
ment in sodium and water excretion is delayed with respect
to the reduction in portal pressure achieved by TIPS inser-
tion.16,18In most patients included in the current study, the
up. Although this is not a controlled investigation and a
control group of patients was not studied, it is very unlikely
that thebeneficial effectson renal function wereunrelated to
the use of TIPS, because spontaneous improvement of renal
function is very uncommon in patients with type I HRS, a
condition characterized by arapidly progressiverenal failure
and an extremely short survival.36,37It should bepointed out
that despite the improvement of renal hemodynamics, RPF
and GFR were below normal values 1 month after TIPS.
Moreover, therenal capacity toexcretesodiumandwater did
not increase significantly, and marked sodium and water
retentionwerestill present 1monthafter TIPS. Thismay bea
result of an incompletesuppression of theactivity of antina-
triuretic systems and/or a lack of normalization of portal
Theincidenceof complicationsafter TIPSwasnot particu-
larly high despitethecoexistenceof advanced liver and renal
failure. The impairment in renal function following TIPS,
in cirrhotic patients with renal failure,17,19was transient,
occurred in only two patients, and none of them required
hemodialysis. The low volume of contrast medium used
probably accounted for this low incidence of renal impair-
ment after TIPS. Thiscomplicationcanbereducedfurther by
using CO2instead of iodine-containing contrast agents. The
that reported in series of patients treated with TIPS for
indications other than HRS.38-41In all cases, encephalopathy
was successfully managed with standard medical therapy.
Finally, similar to previous reports in patients with esopha-
geal varices or refractory ascites, a moderate impairment of
liver function was observed after TIPS.16,19,41All these find-
ings suggest that the presence of HRS is not associated with
an increased risk of complicationsafter TIPS in patientswith
Portal hypertensionisamajorfactorinthedevelopment of
ascites. Although a threshold of portal pressure required for
the development of ascites has not been defined precisely,
ascites rarely develops when portal pressure, as estimated by
portosystemic gradient, is lower than 12 mm Hg.42,43In a
only 4 patients (4%) had a portosystemic gradient below 12
mmHg (mean valuein thewholeseriesof patients, 20.4 ? 5
mmHg; range, 9-40.5 mmHg) (Gine ` sP , unpublishedresults,
January 1997). Consistent with thesefindings, it hasrecently
been established that recurrence of ascites after TIPS is
succesfully prevented in patients in whom portosystemic
gradient is maintained below 12 mm Hg.29Data from the
current study are insufficient to establish a correlation
between portosystemic gradient after TIPS andan increasein
renal perfusion, but the observation that renal function did
not improve in the two patients in whom portosystemic
gradient did not decreasebelow 12 mmHg suggeststhat this
level isalsopredictiveofresponsein patientswith HRS.
Thepathogenesisof HRSisnot known completely, but it is
generally accepted that renal vasoconstriction is related to
markedabnormalitiesinthesystemic circulation, includinga
marked reduction in total systemic vascular resistance and
arterial pressure and increased activity of systemic vasocon-
strictor factors, mainly therenin-angiotensin system and the
tor factors is most likely caused by a reduction in effective
arterial blood volume (i.e., the volume that is sensed by
arterial baroreceptors) secondary to the vasodilation of the
arterial circulation. This arterial vasodilation, which occurs
mainly in the splanchnic circulation, would result in an
abnormal distribution of blood volume, characterized by a
tral blood volume.44Several lines of evidence indicate that
vasoconstrictor systems play a major role in the disturbed
renal hemodynamics observed in patients with cirrhosis and
ascites. First, patients with impaired renal hemodynamics
have a more marked activation of renin-angiotensin and
sympathetic nervous systems compared with patients with
ascitesandnormal renal hemodynamics.45,46Second, patients
with increased activity of vasoconstrictor systems are more
likely todeveloprenal failurewhen treatedwith nonsteroidal
anti-inflammatory drugs, which inhibit therenal synthesisof
vasodilatorprostaglandins, thanpatientswithnormal activity
of vasoconstrictor systems.45,46Third, cirrhotic patients with
ascites and increased activity of these systems have a very
high risk of developing HRS during the evolution of their
disease.37Moreover, a recent study has shown that the
suppression of the overactivity of renin-angiotensin and
sin and albumin for 15 days is associated with a marked
improvement in RPF and GFR in cirrhotic patients with
HRS.47Therefore, it is likely that the improvement in renal
hemodynamics after TIPS in patients included in thecurrent
study was related, at least in part, to the suppression of
vasoconstrictor systems. As with the improvement of renal
delayedwithrespect totheinsertionofTIPS. Similarfindings
were obtained in investigations in which TIPS was used for
the management of refractory ascites.16,18The reduction in
theactivity of vasoconstrictor systemscouldberelatedtothe
redistribution of blood volume caused by the insertion of
TIPS, with a decrease in noncentral blood volume and an
increase in central blood volume.18The lack of changes in
arterial pressure after TIPS should not be interpreted as an
indication against an improvement of systemic hemodynam-
ics, because after TIPS, patients were able to maintain a
GUEVARA ET AL.HEPATOLOGY August 1998
similar arterial pressure compared with baseline conditions
fusion in HRS may be caused by the existence of a direct
hepatorenal reflex raised by sinusoidal portal hyperten-
sion.48,49Therefore, the possibility exists that the improve-
ment of renal function wasrelated toareduction in therenal
sympatheticnerveactivity secondary tothedecreaseinportal
pressure. It hasbeen shown that thedirect inhibition of renal
sympathetic nerve activity by anesthetic blockade of renal
sympathetic innervation isassociatedwith amarkedincrease
inGFR incirrhotic patientswithHRS.50However, after direct
renal sympathetic inhibition, thereisarapidincreaseinGFR,
whereas the improvement of GFR after TIPS in our patients
occurred very slowly. Further studies areneeded to elucidate
themechanism(s) by which TIPS improves renal function in
In conclusion, theresultsof thecurrent study indicatethat
TIPS improves renal function and reduces the activity of
fore, TIPS could be beneficial in the management of this
severe clinical condition. Nevertheless, because this study
wasperformed with asmall number of patientsand acontrol
group was not included, definitive conclusions about the
efficacy and safety of TIPS in themanagement of typeI HRS
cannot be obtained. Prospective, controlled studies in large
thisprocedurecan beadvocatedfor usein clinical practice.
Cela, R.N., Angels Baringo, R.N., Carme Escofet, R.N.,
Eula ` lia Calvo, R.N., Laura Rocabert, R.N., and the nursing
staff of the intensive care unit for their technical assistance.
The authors also thank Drs. Escorsell, Bru, Mas, and Sal-
mero ´n for their participation in thestudy.
The authors are indebted to Raquel
1. Levy M. Hepatorenal syndrome. Kidney Int 1993;43:737-753.
2. Schrier RW, Niederberger M, Weigert A, Gine ` s P . Peripheral arterial
vasodilation: determinant of functional spectrum of cirrhosis. Semin
3. Arroyo V, Gine ` s P , Gerbes AL, Dudley FJ, Gentilini P , Laffi G, Reynolds
TB, et al. Definition and diagnostic criteria of refractory ascites and
hepatorenal syndromein cirrhosis. HEPATOLOGY 1996;23:164-173.
4. Epstein M. Hepatorenal syndrome. In: Epstein M, ed. The Kidney in
Liver Disease. 4th ed. Philadelphia: Hanley & Belfus, 1996:75-108.
5. Bataller R, Gine ` sP , GuevaraM, ArroyoV. Hepatorenal syndrome. Semin
6. GonwaTA, KlintmalmGB, Levy M, JenningsLS, GoldsteinRM, Husberg
BS. Impact of pretransplant renal function on survival after liver
transplantation. Transplantation 1995;59:361-365.
7. Laffi G, La Villa G, Gentilini P . Pathogenesis and management of the
hepatorenal syndrome. Semin Liver Dis1994;14:71-81.
8. Ro ¨ssleM, Haag K, Ochs A, Sellinger M, No ¨ldgeG, Perarnau JM, Berger
E, et al. The transjugular intrahepatic portosystemic stent-shunt proce-
durefor variceal bleeding. N Engl J Med1994;330:165-171.
9. ShiffmanML, JeffersL, HoofnagleJH, TralkaTL. Theroleoftransjugular
intrahepatic portosystemic shunt for treatment of portal hypertension
and itscomplications: aconferencesponsored by theNational Digestive
DiseaseAdvisory Board. HEPATOLOGY 1995;25:1591-1597.
10. Ring EJ, LakeJR, RobertsJP , Gordon RL, LaBergeJM. ReadAE, Sterneck
11. CabreraJ, Maynar M, GranadosR, Gorriz E, ReyesR, Pulido-DuqueJM,
Sanroman JLR, et al. Transjugular intrahepatic portosystemic shunt
(TIPS) vs. esclerotherapy in the elective treatment of variceal haemor-
rhage. Gastroenterology 1996;110:832-839.
12. Jalan R, Forrest EH, Stanley AJ, Redhead DN, Forbes J, Dillon JF,
MacGilchrist AJ, et al. A randomized trial comparing transjugular
intrahepatic portosystemic stent-shunt withvariceal bandligationinthe
prevention of rebleeding fromesophageal varices. HEPATOLOGY 1997;26:
13. Grace ND. TIPS: the search for a proper role. In: Arroyo V, Bosch J,
BrugueraM,Rode ´ sJ,eds.TherapyinLiverDisease.ThePathophysiologi-
cal BasisofTherapy. Barcelona: Masson S.A., 1997:435-440.
14. Grace ND. TIPS: the long and short of it. Gastroenterology 1997;112:
15. Ferral H, Bjarnason H, Wegryn SA, Rengel GJ, Nazarian GK, Rank JM,
Tadavarthy SM, et al. Refractory ascites: early experience in treatment
16. Somberg KA, Lake JR, Tomlanovich SJ, Laberge JM, Feldstein V, Bass
NM. Transjugular intrahepatic portosystemic shunt for refractory asci-
tes: Assessment of clinical and humoral response and renal function.
17. Quiroga J, Sangro B, Nun ˜ez M, Bilbao I, Longo J, Garcı ´ a-Villarreal L, et
al. Transjugular intrahepatic portal-systemic shunt in the management
ofrefractory ascites: Effectonclinical, renal, humoral andhemodynamic
parameters. HEPATOLOGY 1995;21:986-994.
18. Wong F, Sniderman K, Liu P , Allidina Y, Sherman M, Blendis L.
Transjugular intrahepatic portosystemic stent shunt: effectson hemody-
namics and sodiumhomeostasis in cirrhosis and refractory ascites. Ann
19. Ochs A, Ro ¨ssle M, Haag K, Hauentsein KH, Deibert P , Siegerstetter V,
Huonker M, et al. The transjugular intrahepatic portosystemic stent-
shunt procedure for refractory ascites. N Engl J Med 1995;332:1192-
20. Somberg KA. Transjugular intrahepatic portosystemic shunt in the
treatment ofrefractory ascitesandhepatorenal syndrome. In: EpsteinM,
ed. TheKidney in Liver Disease, 4th ed. Philadelphia: Hanley & Belfus,
21. Wong F. Blendis L. Transjugular intrahepatic portosystemic shunt for
refractory ascites: tipping thesodium balace. HEPATOLOGY 1995;22:358-
22. CrenshawWB, GordonFD, McEniffNJ, PerryLJ, Hartnell G, Anastopou-
los H, Jenkins RL. et al. Severe ascites: efficacy of the transjugular
intrahepaticportosystemicshuntintreatment. Radiology 1996;200:185-
23. LakeJR, Ring E, LaBergeJ, Gordon R, RobertsJ, Ascher N. Transjugular
intrahepatic portocaval stent shunt in patients with renal insufficiency.
Transplant Proc 1993;25:1766-1767.
24. Alam I, Bass NM, LaBerge JM, Ring EJ, Somberg KA.: Treatment of
hepatorenal syndromewith thetransjugular intrahepatic portosystemic
shunt (TIPS). Gastroenterology 1995;108:1024A.
25. Spahr L, Fenyeves D, N’Guyen VV, Roy L, Legault L, Dufresne MP ,
Pomier-Layrargues G. Improvement of hepatorenal syndrome by tran-
sjugular intrahepatic portosystemic shunt. AmJ Gastroenterol 1995;90:
26. Lerut J, Goffette P , Laterre PF, Donataccio M, Reynaert MS, Otte JB.
Sequential treatment of hepatorenal syndromeandposthepatic cirrhosis
by intrahepatic portosystemic shunt (TIPSS) and liver transplantation.
27. SturgisTM.Hepatorenal syndrome:resolutionaftertransjugularintrahe-
patic portosystemic shunt. J Clin Gastroenterol 1995;20:241-243.
28. Follo A, Llovet JM, NavasaM, Planas R, Forns X, FrancitorraA, Rimola
A, et al. Renal impairment after spontaneous bacterial peritonitis in
cirrhosis: incidence, clinical course, predictive factors and prognosis.
29. Casado M, Bosch J, Garcı ´ a-Paga ´ n JC, Bru C, Ban ˜ares R, Bandi JC,
Escorsell A, et al. Clinical events following transjugular intrahepatic
portosystemic shunt placement: correlation with hemodynamic find-
ings. Gastroenterology 1998;114:1-10.
30. Gine ` s P , Jime ´ nez W, Arroyo V, NavasaM, Lo ´pez C, Tito ´L, SerraA, et al.
Atrial natriuretic factor in cirrhosis with ascites: plasma levels, cardiac
releaseandsplanchnic extraction. HEPATOLOGY 1988;8:636-642.
31. Salo ´ J, Ferna ´ ndez-Esparrach G, Gine ` s P , Gine ` s A, Guevara M, Sort P ,
Jime ´ nez W, et al. Urinary endothelin-like immunoreactivity in patients
with cirrhosis. J Hepatol 1997;27:810-816.
32. Heyrovsky A. A new method for the determination of inulin in plasma
andurine. Clin ChimActa1956;1:470-474.
33. Brun C. A rapid method for the determination of para-aminohippuric
acidandkidney function tests. J LabClin 1951;37:955-958.
34. Schroeder ET, Anderson GH, Smulyan H. Effects of a portacaval or
peritoneovenous shunt on renin in the hepatorenal syndrome. Kidney
HEPATOLOGY Vol. 28, No. 2, 1998GUEVARA ET AL.
35. Ariyan S, Sweeny T, Kerstein MD. The hepatorenal syndrome: recovery Download full-text
after portacaval shunt. Ann Surg1975;181:847-849.
36. Papper S. Hepatorenal syndrome. In: EpsteinM, ed. TheKidney inLiver
Disease. 2nded. New York: Elsevier, 1983:87-106.
37. Gine ` sA, Escorsell A, Gine ` sP , Salo ´J, Jime ´ nez W, IngladaL, NavasaM, et
al. Incidence, predictive factors, and prognosis of the hepatorenal
syndromein cirrhosiswith ascites. Gastroenterology 1993;105:229-236.
38. Blei AT. Hepatic encephalopaty in theageof TIPS. HEPATOLOGY 1994;20:
39. Sanyal AJ, Freedman AM, Shiffman ML, Purdum PP , Luketic VA,
CheathamAK. Portosystemicencephalopathy aftertransjugularintrahe-
patic portosystemic shunt: results of a prospective controlled study.
40. RiggioO, Merli M, Pedretti G, Servi R, Meddi P , LionetteR, Rossi P , et al.
Hepatic encephalopathy after transjugular intrahepatic portosystemic
shunt. Incidenceandrisk factors. DigDisSci 1996;41:578-584.
41. Ro ¨ssleM. Thetransjugular intrahepatic portosystemic shunt. J Hepatol
42. Rector WG. Portal hypertension: a permissive factor only in the
development ofascitesandvariceal bleeding. Liver 1986;6:221-226.
43. Morali GA, Sniderman KW, Deitel KM, Tobe S, Witt-Sullivan H, Simon
M, HeathcoteJ, etal. Issinusoidal portal hypertensionanecessary factor
for thedevelopment ofhepatic ascites? J Hepatol 1992;16:249-250.
44. Henriksen J, Møller S. Hemodynamics, distribution of blood volume,
andkineticsofvasoactivesubstancesincirrhosis. In: EpsteinM, ed. The
Kidney in Liver Disease. Philadelphia: Hanley & Belfus, 1996:241-258.
45. Gine ` s P , Arroyo V, Rode ` s J. Disorders of renal function in cirrhosis:
Pathophysiology and clinical aspects. In: Zakim D, Boyer T, eds.
Hepatology: A Textbook of Liver Diseases. Philadelphia, Saunders,
46. Arroyo V, Gine ` s P , Jime ´ nez W, Rode ´ s J. Ascites, renal failure, and
electrolytedisordersin cirrhosis. Pathogenesis, diagnosisandtreatment.
In:McIntyreN, BenhamouJP , BircherJ, RizzettoM, Rode ´ sJ, eds.Oxford:
OxfordUniversity Press, 1991:427-471.
47. Guevara M, Gine ` s P , Ferna ` ndez-Esparrach G, Sort P , Salmero ´n JM,
Jime ´ nez W, Arroyo V, et al. Reversibility of hepatorenal syndrome by
prolonged administration of ornipressin and plasmavolumeexpansion.
48. LangF, TschernkoE, SchulzeE, O¨ttl I, Ritter M, Vo ¨lkl H, Hallbrucker C,
et al. Hepatorenal reflex regulating kidney function. HEPATOLOGY 1991;
49. Safirstein R, Levitt MF. A hepatorenal depressor reflex: apossibleclueto
thepathogenesisofthehepatorenal syndrome.HEPATOLOGY 1991;14:734-
50. Solı ´ s-Herruzo JA, Duran A, Favela V, Castellano G, Madrid JL, Mun ˜oz-
Yagu ¨e MT, Morillas JD, et al. Effects of lumbar sympathetic block on
kidney function in cirrhotic patients with hepatorenal syndrome. J
GUEVARA ET AL.HEPATOLOGY August 1998