Perendoscopic variceal pressure measurement: a reliable estimation of portal pressure in patients with cirrhosis?

(1)

Article

Reference

Perendoscopic variceal pressure measurement: a reliable estimation of portal pressure in patients with cirrhosis?

SPAHR, Laurent François Joséph, et al.

Abstract

In patients with cirrhosis, the hepatic venous pressure gradient (HVPG) is the reference method for the assessment of portal hypertension (PHT). Variceal pressure (VP) may be measured at endoscopy, but its relationship to the HVPG remains controversial. The aim of the study was to retrospectively compare HVPG and VP values obtained in a cohort of patients with cirrhosis and PHT.

SPAHR, Laurent François Joséph, et al . Perendoscopic variceal pressure measurement: a reliable estimation of portal pressure in patients with cirrhosis? Gastroentérologie clinique et biologique , 2006, vol. 30, no. 8-9, p. 1012-8

PMID : 17075452

Available at:

http://archive-ouverte.unige.ch/unige:36994

Disclaimer: layout of this document may differ from the published version.

(2)

ORIGINAL ARTICLE

Perendoscopic variceal pressure measurement

A reliable estimation of portal pressure in patients with cirrhosis?

Laurent SPAHR (1), Emile GIOSTRA (1), Isabelle MORARD (1), Gilles MENTHA (2), Antoine HADENGUE (1)

(1) Gastroenterogy and Hepatology, (2) Transplantation Unit, Hôpitaux Universitaire de Genève.

SUMMARY

Objectives — In patients with cirrhosis, the hepatic venous pressure gradient (HVPG) is the reference method for the assessment of portal hypertension (PHT). Variceal pressure (VP) may be measured at endoscopy, but its relationship to the HVPG remains controversial.

The aim of the study was to retrospectively compare HVPG and VP values obtained in a cohort of patients with cirrhosis and PHT.

Methods — Within 8 days (range: 6-10 days), 64 patients in a stable condition with biopsy-proven cirrhosis [alcoholic: 47; other 17;

mean age: 56.5 yrs (35-70); mean Child-Pugh’s score: 9.4±1.9;

ascites: 37/64; previous variceal bleeding (=“bleeders”): 24/64) and oesophageal varices (grade 2: 49; grade 3: 15)] underwent both measurement of the HVPG during transjugular liver biopsy and VP at endoscopy using a “home made” pressure sensitive gauge in the absence of needle puncture of the varix. Alcoholic hepatitis was present in 28 patients with alcoholic cirrhosis.

Results — The pressure sensitive gauge was well tolerated. The mean HVPG and VP values were 18.5±3.4 mmHg and 19±3.7 mmHg, respectively. A significant difference was observed between “bleeders” (n = 24) and non “bleeders” (n = 40) in terms of VP values (21.4±3.3 vs 17.2±3.2 mmHg, P < 0.001), but not for HVPG values (19.4±4.1 vs 17.9±2.8 mmHg, P = 0.075).

A positive correlation was observed between VP and HVPG values (r = 0.62, P < 0.0001).

Conclusions — In this group of patients with cirrhosis and oesophageal varices, a “home-made” pressure sensitive gauge allowed a non invasive perendoscopic measurement of VP. The positive correlation between VP and HVPG values suggests that measurement of VP may be a reliable estimate of portal pressure in these patients

RÉSUMÉ

Laurent SPAHR, Emile GIOSTRA, Isabelle MORARD, Gilles MENTHA, Antoine HADENGUE

Objectifs — Chez le malade atteint de cirrhose, le degré d’hyper- tension portale (HTP) est déterminé par la mesure du gradient de pression hépatique (GPH). La pression des varices (PV), influencée par la pression portale, peut être mesurée lors d’endoscopie, mais sa relation avec le GPH est controversée. Le but de ce travail était de comparer rétrospectivement les mesures de GPH et de PV obte- nues dans un groupe de malades atteints de cirrhose compliquée d’HTP.

Méthodes — Dans un intervalle de 8 jours (6-10 jours), 64 malades avec cirrhose histologiquement prouvée [alcoolique : 47 ; autre 17 ; âge moyen : 56,5 ans (35-70) ; score de Child-Pugh moyen : 9,4r1,9 ; ascite : 37/64 ; antécédent d’hémorragie : 24/64) et varices oesophagiennes (grade 2 : 49 ; grade 3 : 15)] ont bénéficié d’une mesure du GPH et de la PV lors d’endoscopie. La mesure de la PV se faisait à l’aide d’un capteur endoscopique confectionné de façon artisanale, sans ponction de la varice. Une hépatite alcoolique était présente chez 28 malades.

Résultats — Le capteur de pression endoscopique était bien toléré.

Les valeurs moyennes du GPH et de la PV étaient de 18,5r3,4 mmHg et de 19r3,7 mmHg, respectivement. On observait une différence statistiquement significative entre les patients ayant présenté une hémorragie (n = 24) et ceux qui n’avaient jamais saigné (n = 40) en terme de PV (21,4r3,3 vs 17,2r3,2 mmHg, P < 0,001), mais pas en terme de GPH (19,4r4,1 vs 17,9r2,8 mmHg, P = 0,075). Il existait une corré- lation linéaire entre la PV et le GPH (r = 0,62, P < 0,0001).

Conclusion — dans ce groupe de malades atteints de cirrhose et de varices oesophagiennes, l’application endoscopique sur les varices d’un capteur de pression artisanal permettait d’estimer de façon fia- ble la pression.

Introduction

Clinically significant portal hypertension is defined by an increase in the hepatic venous pressure gradient (HVPG) to a threshold level above 10 mmHg [1]. Portal hypertension (PHT) may be complicated by the presence of oesophageal varices (OV), ascites, and portal-hypertensive related bleeding [2]. The HVPG is the most commonly used method to assess portal pressure in clinical practice [3, 4]. It is usually performed during

transjugular liver biopsy, and may provide information on prog- nosis [5-8], treatment monitoring [9], as well as disease pro- gression [10].

In spite of this, HVPG measurement is not routinely performed [3] for several reasons. First, the HVPG is a reliable indi- cator of portal pressure only in diffuse liver injury such as alcohol [11] and hepatitis C-related cirrhosis [12]. Second, the technique is invasive with potential neck vascular injury [13], and reported as disagreeable by some patients. Third, it requires adapted equipment and application using a standard technical, as recently stated [14].

La mesure de la pression des varices oesophagiennes par voie endoscopique : une estimation fiable

de la pression portale chez des malades atteints de cirrhose ?

(Gastroenterol Clin Biol 2006;30:1012-1018)

Reprints: L. SPAHR, Gastroenterology and Hepatology, University Hospital, 24, Rue Micheli-du-Crest, 1211 Geneva 14, Switzerland.

E-mail : Laurent.Spahr@hcuge.ch

Meeting presentation: Journées Francophones de Pathologie Digestive, Paris, 2003.

(3)

Variceal pressure and portal hypertension

Oesophageal varices are strongly influenced by portal pressure, but there is a non linear relationship between the HVPG level and the risk of variceal bleeding [15, 16]. In patients with cirrhosis, the presence of OV at endoscopy is indicative of clinically significant PHT [17]. The risk of haemorrhage is higher in large as compared to small OV, in relation to an elevated intravariceal pressure and wall tension according to Laplace’s equa- tion [18]. Indeed, the size of OV, the presence of cherry red spots on the variceal wall and the severity of liver dysfunction, all influence the risk of variceal bleed [19]. In addition, the measurement of variceal pressure (VP) may provide information on the bleeding risk [20], and could be of value in predicting the response to pharmacological agents [9].

The methods developed to measure VP include the direct puncture of the varix [21] and the non invasive application of a pressure sensitive device upon the OV. The latter technique may be performed non selectively on the distal end of the oesophagus [22, 23], or selectively on an individual varix [20, 24-26] under endoscopic view. Intravariceal pressure measurements correlates to portal pressure in patients with cirrhosis [21], but is potentially associated with an elevated bleeding risk unless immediately fol- lowed by sclerotherapy [24]. When a non selective manometry of OV is performed, VP correlated linearly to the HVPG values (r = 0.64) [23], and to portal pressure (r = 0.68) measured invasively during temporary TIPS occlusion [25]. Using an endoscopic device, which requires continuous gas perfusion through the system, investigators from the Barcelona group performed individual measurements of VP under direct visual control. They reported high VP values in patients with large OV [26, 27], but no linear correlation between VP and the HVPG (r = 0.48), which they hypothesized may be due to variations in resistance and blood flow in the collaterals.

In order to clarify this issue, we aimed to evaluate the value of a “home-made” pressure sensitive capsule to measure non invasively VP at endoscopy, and to assess its relationship to the

HVPG measured during transjugular liver biopsy in a large group of patients with cirrhosis and OV.

Methods

Between January 2001 and June 2002, 79 patients with cirrhosis were admitted to our hospital for decompensated cirrhosis, variceal haemorrhage, suspicion of alcoholic hepatitis, or as part of an evaluation for liver transplantation. Hepatic hemodynamic and endoscopic measurements, considered as a standard diagnostic work-up in patients with cirrhosis, included an upper gastrointestinal endoscopy, a transjugular liver biopsy with measurement of the HVPG. In addition VP measurement was performed at the time of upper gastrointestinal endoscopy.

Table I summarized the characteristics of the 64 patients in whom both hemodynamic and endoscopic data could be analyzed (see Results).

All patients had clinically significant PHT with OV, and no portal vein thrombosis. The etiology of cirrhosis was alcoholic in the majority of the cases (47 out of 62 patients). Ascites was clinically present in 37 patients.

Both transjugular liver biopsy and VP measurement were performed within a median time interval of 8 days (range: 6-10 days). Twenty four patients reported an episode of previous bleeding from OV. For 12 out of these 24 patients, time from bleeding to haemodynamic measurements (VP and HVPG) was 11 days (range: 5 to 21 days). These patients were all treated as follows: after initial fluid resuscitation, endoscopic treatment, IV octreotide infusion for 72 hours, and blood transfusions (if needed, to reach haemoglobin level of 80-100 gr/dL), patients received a fixed dose of propranolol 80 mg per day, starting 24-48 hours after the index bleed.

During this time period, patients remained in a stable haemodynamic condition.

Endoscopic assessment

Endoscopy was performed using an Olympus (Olympus Optical Co., Tokyo, Japan) videoendoscope. Patients in a fasting state were examined under conscious sedation using intravenous meperidine (25-50 mg). The size of the varices was graded from 1 to 3 [19].

Variceal pressure measurement

We used a “home-made” pressure capsule to measure VP. The device was tested in an artificial varix system, as illustrated and detailed in figure 1, and a linear correlation (r=0.92) was observed (data not shown).

The VP was measured at endoscopy using the pressure sensitive gauge attached to the tip of the endoscope, as schematically represented and illustrated in figure 2. Briefly, a hollow plastic capsule measuring 8 mm in dia- Table I. – Characteristics of 64 patients for whom hemodynamic and endoscopic measurements were ana-

lyzable.

Caractéristiques des 64 malades pour lesquels les mesures hémodynamiques et endoscopiques étaient analysables.

All patients

(n = 64) Bleeders

(n = 24) Non bleeders

(n = 40) P value

Age (yrs, range) 56.5 (35-73) 56.4 (35-68) 56.6 (40-73) 0.68

Gender (M/F) 45/19 18/6 27/13 0.11

Alcoholic cirrhosis 47/64 19/24 28/40 0.27

ASH on liver biopsy 28/47 10/24 18/23 0.18

Pugh’s score 9.4±1.9 8.9±2.1 9.6±1.8 0.33

Child class (A/B/C) 3/32/29 3/11/10 0/21/19 0.37

Ascites 37/64 10/24 27/40 0.08

Beta-blockers 19/64 8/24 11/40 0.11

Size of OV (II/III) 49/15 19/5 30/10 0.15

OV: oesophageal varices; ASH: alcoholic steatohepatitis.

ABBREVIATIONS :

HVPG : hepatic venous pressure gradient PHT : portal hypertension

OV : esophageal varices VP : variceal pressure

(4)

meter was connected to a Teflon catheter and introduced into the operating channel of the endoscope. The whole system was carefully filled with saline, and sealed with a rubber membrane. Particular attention was paid to the absence of air bubble in the system. This system was then connected to a

pressure transducer (Millar Instr, Houston, TX, USA) and the pressure recorded simultaneously on a PC screen using a data acquisition software (Perimed, Stockholm, Sweden). After a careful intubation of the oesophagus, the free pressure (i.e., the pressure in the oesophageal lumen in the absence of contact with the oesophageal wall) was recorded. Then, the endoscopic capsule was gently applied selectively on the varix under direct visual control. Care was taken not to elicit oesophageal contractions. The same physician (L.S) performed all measurements. A second physician (E.G or I.M) was in charge of setting the zero pressure prior to the application of the gauge on the varix, and assessing the quality and stability of the tracing. He was unaware of the endoscopic aspects of the OV, nor of the HVPG value. In addition to meperidine, all patients received IV 40 mg N- butyl scopolamine at the time of measurement of the free pressure in the oesophagus, to reduce as much as possible large oesophageal contractions that influence VP [28]. The criteria required for an acceptable measure included a zero pressure tracing and a VP recording with stable venous fluctuations for at least 10 seconds, with the gauge maintained on the varix under endoscopic view, in the absence of major oesophageal contractions.

Both the procedure and a typical VP tracing are illustrated in figure 3. The mean value of two acceptable measurements determined the VP.

The patients with recent haemorrhage and bands visible on strangu- lated varices had VP measurement performed on a native residual varix.

Hepatic venous pressure measurement

Transjugular liver biopsy was performed in a fasting state and under sedation (meperidine 25-50 mg IV). A 8F curved catheter (Cordis Europa, Amsterdam, The Netherlands) was introduced into the right hepatic vein under fluoroscopic guidance. The wedged hepatic venous pressure (WHVP) was recorded when the tip of the catheter was wedged in a small hepatic vein, while the free hepatic venous pressure (FHVP) was measured when the tip of the catheter floated in the hepatic vein at the junction with the inferior vena cava. To ensure that the catheter was in a proper wedged position, radiological contrast was injected to visualize small portal branches. The HVPG resulted from the difference between WHVP and FHVP. We used the mid-chest as the external zero reference.

Pressure tracings had to remain stable during at least 30 seconds to be considered interpretable. We performed 2 HVPG measurements in two different vascular territories, and the mean value was kept for analysis.

For technical reasons, we could not keep paper records of the tracings and we used instant pressure reading on the monitor screen.

Histological analysis

Liver biopsy was interpreted by a senior histopathologist expert in liver diseases. The presence of five histologic lesions (liver cell necrosis, Mallory bodies, neutrophilic infiltrate, fibrosis, and fatty infiltration) was required for the diagnosis of alcoholic hepatitis [29].

Statistical analysis

Data were expressed as mean ± SD, and median and range (for time intervals). The non-parametric Wilcoxon signed rank test was used to compare hemodynamic data. Correlation analysis was performed using the non-parametric Spearman rank test. A P < 0.05 was considered statistically significant. All calculations were made using the Statview 5.0 Program (SAS Institute Inc., USA).

Ethical considerations

The Institutional Review Board of the Hôpitaux Universitaires de Genève allowed us to retrospectively review the endoscopic, hemodynamic and clinical data of the patients. In addition, all patients gave a written informed consent for endoscopic procedures.

Results

Haemodynamic data

In 15 patients, VP measurements were rejected due to artefacts related to residual air bubbles in the system (n = 4), large

Artificial varix filled with saline

compressor manometer endoscope

pressure capsule

Rubber membrane

to pressure transducer and PC recording

Connector filled with saline

Fig. 1– Schematic representation of the artificial varix. A plastic tube filled with water and connected to a compressor was equipped with a hole (approximately 2 cm²) recovered and sealed with a thin rubber membrane. Intravariceal pressure was delivered by a compressor in a gradual manner from 10 to 30 mmHg under the direct control of a manometer. At each pressure point, the “zero pressure” was set while the capsule was close to, but not in contact with, the rubber membrane. Then, in a blinded fashion, the examiner applied the endoscopic capsule on the rubber membrane so as to flatten it. An assistant examiner read the VP on the PC screen. Each measurement was performed 3 times.

Représentation schématique de la varice artificielle. Une fenêtre d’environ 2 cm² a été découpée dans un tube de plastique rempli d’eau, lequel était connecté à un compresseur. Une fine membrane de caoutchouc recouvrait de façon étanche la fenêtre. Le compres- seur délivrait une pression croissante (de 10 à 30 mmHg) dans le tube. A chaque mesure, la « pression zéro » était ajustée avant que le capteur ne soit appliqué. Puis, en aveugle, l’examinateur appli- quait le capteur endoscopique sur la membrane, de façon à l’aplat- ir. Le co-examinateur lisait la pression de la varice artificielle sur l’écran de l’ordinateur. Chaque mesure était répétée 3 fois.

Fig. 2– Photographic illustration of the different components of the variceal pressure device. The rubber membrane is applied over the capsule and is then manually tied over the shaft.

Illustration photographique des différents composants du capteur endoscopique. La membrane en caoutchouc est appliquée sur la capsule. Une ligature manuelle assure l’étanchéité.

(5)

oesophageal contractions (n = 5) or difficulties in positioning the pressure gauge on the varix (n = 6). Thus, we present in table II the results of hemodynamic and endoscopic measures obtained in 64 patients. The mean HVPG value was 18.5±3.4 mmHg, consistent with clinically significant PHT [17]. There was a trend (P = 0.075) towards higher HVPG values in the 24 patients who previously bled from OV (“bleeders”) as compared to the 40 patients who never bled (“non bleeders”). The mean VP value was 19±3.7 mmHg. The intra- observer difference in VP measurement was 5.8%. Variceal pressure values were higher in “bleeders” as compared to “non bleeders” (P < 0.001). Figure 4 illustrates both HVPG and VP values in the 64 patients. The portal-variceal pressure gradient, calculated as the difference between VP and the HVPG and thought to reflect vascular resistance along the collaterals [27], was 3.7±2.9 mmHg. Values were similar in “bleeders” and

“non bleeders”.

Correlations

There was a linear correlation between VP and HVPG values (r = 0.62, P < 0.0001, see figure 5), but not with WHVP (r = 0.2).

Discussion

In this group of patients with cirrhosis and clinically significant PHT, we demonstrate that, i) the non invasive measurement of VP using a “home-made” pressure sensitive gauge is feasible and reproducible, ii) values of VP are elevated, iii) VP values are higher in patients who experienced a previous variceal bleeding episode as compared to those who never bled, and iiii) there is a positive correlation (r = 0.62) between VP and HVPG.

Our findings are in line with results obtained with the continuous gas perfusion endoscopic capsule [26, 27], and we report addi- tional information on the relationship between VP and HVPG values.

Variceal pressure plays a central role in the pathogenesis of variceal haemorrhage [2]. VP is modified following pharmacological intervention [9] and may estimate portal pressure [23, 25]. The non invasive devices (balloon manometry and endoscopic capsule) developed to measure VP rely on the princi- ple that the pressure necessary to compress the varix equals the pressure inside the lumen. The value is expressed as VP. Accord- ing to Laplace’s law, intravariceal pressure is more elevated in large as compared to small OV, and this is associated with an

V

C

C V

Poor quality tracing

artifact mmHg

Sustained satisfactory tracing (18 mmHg)

Fig. 3– Endoscopic view of a variceal pressure measurement. Panel A: the free pressure is recorded while the capsule (C) is not in contact with any structure of the oesophagus. This free pressure is carefully adjusted as the “zero pressure” by the assistant examiner in charge of the recording on the PC screen.

Panel B: the capsule is gently applied to the varix under direct visual control, to obtain a typical tracing with fine venous fluctuations. Panel C: Schematic representation of the capsule adjusted to the varix. Gentle and sustained pressure is applied selectively to the varix until collapse, which corresponds to the pressure inside the varix. There is no puncture of the varix. The variceal pressure is calculated as the difference between the free pressure in the oesophageal lumen (define as the “zero pressure”) and the pressure measured on the varix.

Aspect endoscopique d’une mesure de pression de varice oesophagienne. Panneau A : la pression libre est enregistrée alors que la capsule (C) n’est en contact avec aucune structure de l’œsophage. La pression libre est minutieusement ajustée et enregistrée comme « pression zéro » par le co-examinateur sur l’ordinateur. Panneau B : la capsule est délicatement appliquée sur la varice sous contrôle endoscopique, dans le but d’obtenir un tracé de type veineux avec de fines fluctuations. Panneau C : Représentation schématique de la capsule endoscopique appliquée sur la varice. Une pression délicate mais soutenue aplatit la varice, équivalent à la pression régnant à l’intérieur de la varice. Il n’y a pas de ponction de la varice. La pression de varice est calculée par la différence entre la pression libre (mesurée au niveau de la lumière oesophagienne, et définie comme la pression « zéro ») et la pres- sion mesurée lors de l’application du capteur sur la varice.

a b c d

(6)

increased risk of rupture. Indeed, in our patients, VP values ranged from 12 to 26 mmHg, with highest values in patients with large OV. We also report higher VP in bleeders as compared to non-bleeders, consistent with previous results [27].

In our study, the “home-made” endoscopic pressure capsule was filled with saline in a close circuit and did not require a continuous gas perfusion [26]. The 19% unsatisfactory VP measurements were both related to technical limitations (such as air bubbles in the system), and to patient’s characteristics. Accord- ingly, persistent large oesophageal contractions in spite of anti- spasmodic agent, and small grade 2 OV proved to be the main limitations of VP in our patients. These problems have also been reported by others [9, 26].

In the present study, we observed a positive correlation between HVPG and VP, with mean values of 18.5 and 19 mmHg, respectively. These results of VP in our patients differ from those published by Rigau et al. [27], who reported less elevated VP values, in the range of 15 mmHg. These findings deserve comments and thus we put forward a number of hypotheses to explain our results. First, could liver failure affect VP? Azygos blood flow [30], which directly influences VP [26], increases in parallel with liver failure. Since the mean Pugh’s score was higher in our patients as compared to those in the study by Rigau (9.4 vs 7.4), a role for this parameter may be considered. Second, what is the role of ascites? By increasing intra-abdominal pressure [18], ascites raises VP[31]. In our study, 58% of patients had clinically evident ascites. In these patients, there was a trend towards higher VP values as compared to those without ascites (18.7±3 vs 17.2±4.1, P = 0.23). Third, may variceal bleeding affect VP? High VP is associated with elevated variceal wall tension which may even- tually lead to rupture, according to Laplace’s law. This concept is supported by our results and those by Rigau et al. [27]. In our Table II. – Hemodynamic and endoscopic values in a group of 64 patients with cirrhosis and clinically

significant portal hypertension.

Mesures hémodynamiques chez les 64 malades avec cirrhose selon l’état hémorragique des varices œsophagiennes.

All patients

n = 64 Bleeders

n = 24 Non bleeders

n = 40 P value

HVPG (mmHg) 18.5±3.4 19.4±4.1 17.9±2.8 0.075

FHVP (mmHg) 11±5 11.5±5.7 11.4±4.6 0.320

WHVP (mmHg) 28.8±6.5 29.8±7.7 28.2±5.5 0.574

VP (mmHg) 19±3.7 21.4±3.3 17.2±3.3 0.001

PVPG (mmHg) 3.7±2.9 3.4±2.5 3.9±3.2 0.808

HVPG: hepatic venous pressure gradient; WHVP: wedge hepatic venous pressure; FHVP: free hepatic venous pressure; VP: variceal pressure; PVPG: portal-variceal pressure gradient.

5 10 15 20 25 30

HVPG (mmHg) VP (mmHg)

Bleeders Non bleeders Bleeders Non bleeders

P = NS

P < 0.001

Fig. 4– Individual values of HVPG and VP in patients with a previous variceal hemorrhage (“bleeders”, n = 24) and patients who never bled (“non bleeders”, n = 40).

Valeurs individuelles du gradient de pression hépatique (HVPG) et de la pression variqueuse (VP) chez les malades ayant déjà présenté un épisode hémorragique (« bleeders », n = 24) et chez ceux qui n’ont jamais saigné (« non bleeders », n = 40).

1 2

5 10 15 20 25 30

VP (mmHg)

HVPG(mmHg)

r = 0.62 P < 0.0001

Fig. 5– Correlation between the HVPG and the VP value in 64 patients with cirrhosis and clinically significant PHT.

Corrélation entre les valeurs de gradient de pression hépatique (HVPG) et de pression variqueuse (VP) mesurées chez 64 malades avec cirrhose et hypertension portale cliniquement significative.

(7)

study, 37% of the patients had a previous variceal bleed, which in 12 out of 24 of the cases was a recent event. It is at present unclear whether endoscopic therapy influences portal haemodynamics [32, 33], or if VP may vary early after bleeding from OV. Finally, in our group of patients in whom alcoholic cirrhosis predominates, could alcoholic steatohepatitis influence VP?

Acute administration of alcohol to patients with cirrhosis increases portal pressure and collateral blood flow [34], but this effect is relatively short-lived and presumed to be absent several days after hospital admission. Alcoholic steatohepatitis, how- ever, may increase portal pressure [35], and could have partici- pated in the splanchnic and collateral vessels hemodynamic alterations.

In conclusion, in this group of patients with PHT and OV, measurement of VP using a “home-made” endoscopic pressure sensitive capsule proved feasible and may provide a reliable estimate of portal pressure. Variceal pressure measurement, but not the HVPG, allowed to discriminate patients who had a previous variceal bleed from those who never bled. The possible role of factors such as ascites, degree of liver failure, recent hemorrhage and concomitant alcoholic steatohepatitis remains to be established.

Given the fact that VP measurement is technically demand- ing and potentially affected by artefacts in relation to either the device itself or the patient’s characteristics, its use cannot be recommended in routine clinical practice. If determination of portal pressure is clinically indicated, it should still be performed invasively by performing a hepatic venous catheteriza- tion using the technical standard [14] until further studies are available.

ACKNOWLEDGMENT - This work was supported by Funding HUG PRD-0006.

REFERENCES

1.Consensus conference: complications of portal hypertension in adults (Paris, December 405, 2003). Short text]. Gastroenterol Clin Biol 2004;28(Suppl 2):B318-23.

2.Lebrec D, Vinel JP, Dupas JL. Complications of portal hypertension in adults: a French consensus. Eur J Gastroenterol Hepatol 2005;17:403-10.

3.Huet PM, Pomier-Layrargues G. The hepatic venous pressure gradient: “remixed and revisited”. Hepatology 2004;39:295-8.

4.Vanbiervliet G, Pomier-Layrargues G, Huet PM. [Invasive diagnosis of portal hypertension in cirrhosis: a critical evaluation of the hepatic venous pressure gradient measurement]. Gastroenterol Clin Biol 2005;29:988-96.

5.Merkel C, Bolognesi M, Sacerdoti D, Bombonato G, Bellini B, Bighin R, et al. The hemodynamic response to medical treatment of portal hypertension as a predictor of clinical effectiveness in the primary prophylaxis of variceal bleeding in cirrhosis. Hepatology 2000;32:930-4.

6.McCormick PA, Patch D, Greenslade L, Chin J, McIntyre N, Burroughs AK. Clinical vs haemodynamic response to drugs in portal hypertension. J Hepatol 1998;28:1015-9.

7.Moitinho E, Escorsell A, Bandi JC, Salmeron JM, Garcia-Pagan JC, Rodes J, Bosch J. Prognostic value of early measurements of portal pressure in acute variceal bleeding. Gastroenterology 1999;117:626-31.

8.Vorobioff J, Groszmann RJ, Picabea E, Gamen M, Villavicencio R, Bordato J, et al. Prognostic value of hepatic venous pressure gradient measurements in alcoholic cirrhosis: a 10-year prospective study.

Gastroenterology 1996;111:701-9.

9.Escorsell A, Bordas JM, Castaneda B, Llach J, Garcia-Pagan JC, Rodes J, et al. Predictive value of the variceal pressure response to continued pharmacological therapy in patients with cirrhosis and portal hypertension. Hepatology 2000;31:1061-7.

10.Burroughs AK, Groszmann R, Bosch J, Grace N, Garcia-Tsao G, Patch D, et al. Assessment of therapeutic benefit of antiviral therapy in chronic hepatitis C: is hepatic venous pressure gradient a better end point? Gut 2002;50:425-7.

11.Boyer TD, Triger DR, Horisawa M, Redeker AG, Reynolds TB.

Direct transhepatic measurement of portal vein pressure using a thin needle. Comparison with wedged hepatic vein pressure. Gastroente- rology 1977;72:584-9.

12.Perello A, Escorsell A, Bru C, Gilabert R, Moitinho E, Garcia- Pagan JC, et al. Wedged hepatic venous pressure adequately reflects portal pressure in hepatitis C virus-related cirrhosis. Hepatology 1999;30:1393-7.

13.Shah PM, Babu SC, Goyal A, Mateo RB, Madden RE. Arterial mis- placement of large-caliber cannulas during jugular vein catheteriza- tion: case for surgical management. J Am Coll Surg 2004;198:939-44.

14.Groszmann RJ, Wongcharatrawee S. The hepatic venous pressure gradient: anything worth doing should be done right. Hepatology 2004;39:280-2.

15.Lebrec D, De Fleury P, Rueff B, Nahum H, Benhamou JP. Portal hypertension, size of esophageal varices, and risk of gastrointestinal bleeding in alcoholic cirrhosis. Gastroenterology 1980;79:1139-44.

16.Viallet A, Marleau D, Huet M, Martin F, Farley A, Villeneuve JP, et al. Hemodynamic evaluation of patients with intrahepatic portal hypertension. Relationship between bleeding varices and the portohepa- tic gradient. Gastroenterology 1975;69:1297-300.

17.R. DF. Updating Consensus in Portal Hypertension: report of the Ba- veno III Consensus Workshop on definitions, methodology and therapeutic strategies in portal hypetension. J Hepatol 2000;33:46-52.

18.Escorsell A, Gines A, Llach J, Garcia-Pagan JC, Bordas JM, Bosch J, et al. Increasing intra-abdominal pressure increases pressure, volume, and wall tension in esophageal varices. Hepatology 2002;36:936-40.

19.Prediction of the first variceal hemorrhage in patients with cirrhosis of the liver and esophageal varices. A prospective multicenter study. The North Italian Endoscopic Club for the Study and Treatment of Eso- phageal Varices. N Engl J Med 1988;319:983-9.

20.Nevens F, Bustami R, Scheys I, Lesaffre E, Fevery J. Variceal pressure is a factor predicting the risk of a first variceal bleeding: a prospective cohort study in cirrhotic patients. Hepatology 1998;27:15-9.

21.Sarin SK, Sethi KK, Nanda R. Measurement and correlation of wedged hepatic, intrahepatic, intrasplenic and intravariceal pressures in patients with cirrhosis of liver and non-cirrhotic portal fibrosis. Gut 1987;28:

260-6.

22.Gertsch P, Fischer G, Kleber G, Wheatley AM, Geigenberger G, Sauerbruch T. Manometry of esophageal varices: comparison of an endoscopic balloon technique with needle puncture. Gastroenterology 1993;105:1159-66.

23.Pontes JM, Leitao MC, Portela F, Nunes A, Freitas D. Endosonogra- phic Doppler-guided manometry of esophageal varices: experimental validation and clinical feasibility. Endoscopy 2002;34:966-72.

24.Hou MC, Lin HC, Kuo BI, Liao TM, Lee FY, Chang FY, et al. Se- quential variceal pressure measurement by endoscopic needle puncture during maintenance sclerotherapy: the correlation between variceal pressure and variceal rebleeding. J Hepatol 1998;29:772-8.

25.Brensing KA, Neubrand M, Textor J, Raab P, Muller-Miny H, Scheurlen C, et al. Endoscopic manometry of esophageal varices:

evaluation of a balloon technique compared with direct portal pressure measurement. J Hepatol 1998;29:94-102.

26.Bosch J, Bordas JM, Rigau J, Viola C, Mastai R, Kravetz D, et al. No- ninvasive measurement of the pressure of esophageal varices using an endoscopic gauge: comparison with measurements by variceal puncture in patients undergoing endoscopic sclerotherapy. Hepatology 1986;6:667-72.

27.Rigau J, Bosch J, Bordas JM, Navasa M, Mastai R, Kravetz D, et al.

Endoscopic measurement of variceal pressure in cirrhosis: correlation with portal pressure and variceal hemorrhage. Gastroenterology 1989;

96:873-80.

28.Miller LS, Kim JK, Dai Q, Mekapati J, Izanec J, Chung C, et al.

Mechanics and hemodynamics of esophageal varices during peristaltic contraction. Am J Physiol Gastrointest Liver Physiol 2004;

287:G830-5.

29.Spahr L, Rubbia-Brandt L, Pugin J, Giostra E, Frossard JL, Borisch B, et al. Rapid changes in alcoholic hepatitis histology under steroids:

correlation with soluble intercellular adhesion molecule-1 in hepatic venous blood. J Hepatol 2001;35:582-9.

(8)

30.Braillon A, Cales P, Valla D, Gaudy D, Geoffroy P, Lebrec D. In- fluence of the degree of liver failure on systemic and splanchnic haemodynamics and on response to propranolol in patients with cirrhosis. Gut 1986;27:1204-9.

31.Kravetz D, Bildozola M, Argonz J, Romero G, Korula J, Munoz A, et al. Patients with ascites have higher variceal pressure and wall tension than patients without ascites. Am J Gastroenterol 2000;

95:1770-5.

32.Lo GH, Liang HL, Lai KH, Chang CF, Hwu JH, Chen SM, et al.

The impact of endoscopic variceal ligation on the pressure of the portal venous system. J Hepatol 1996;24:74-80.

33.Pereira-Lima JC, Zanette M, Lopes CV, de Mattos AA. The influence of endoscopic variceal ligation on the portal pressure gradient in cirrhotics. Hepatogastroenterology 2003;50:102-6.

34.Luca A, Garcia-Pagan JC, Bosch J, Feu F, Caballeria J, Groszmann RJ, et al. Effects of ethanol consumption on hepatic hemodynamics in patients with alcoholic cirrhosis. Gastroenterology 1997;112:1284-9.

35.Valla D, Flejou JF, Lebrec D, Bernuau J, Rueff B, Salzmann JL, et al. Portal hypertension and ascites in acute hepatitis: clinical, hemodynamic and histological correlations. Hepatology 1989;

10:482-7