Scandinavian Journal of Gastroenterology

ISSN: 0036-5521 (Print) 1502-7708 (Online) Journal homepage: http://www.tandfonline.com/loi/igas20

Transjugular intrahepatic portosystemic shunt treatment of variceal bleeding in an unselected patient population K. Rosenqvist, R. Sheikhi, R. Nyman, F. Rorsman, P. Sangfelt & C. Ebeling Barbier To cite this article: K. Rosenqvist, R. Sheikhi, R. Nyman, F. Rorsman, P. Sangfelt & C. Ebeling Barbier (2017): Transjugular intrahepatic portosystemic shunt treatment of variceal bleeding in an unselected patient population, Scandinavian Journal of Gastroenterology, DOI: 10.1080/00365521.2017.1386795 To link to this article: http://dx.doi.org/10.1080/00365521.2017.1386795

Published online: 09 Oct 2017.

Submit your article to this journal

View related articles

View Crossmark data

Full Terms & Conditions of access and use can be found at http://www.tandfonline.com/action/journalInformation?journalCode=igas20 Download by: [UNIVERSITY OF ADELAIDE LIBRARIES]

Date: 11 October 2017, At: 20:02

SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY, 2017 https://doi.org/10.1080/00365521.2017.1386795

ORIGINAL ARTICLE

Transjugular intrahepatic portosystemic shunt treatment of variceal bleeding in an unselected patient population K. Rosenqvista, R. Sheikhib, R. Nymana, F. Rorsmanb, P. Sangfeltb and C. Ebeling Barbiera

Downloaded by [UNIVERSITY OF ADELAIDE LIBRARIES] at 20:02 11 October 2017

a Department of Radiology, Institution of Surgical Science, Uppsala University Hospital, Uppsala, Sweden; bDepartment of Hepatology, Institution of Medical Science, Uppsala University Hospital, Uppsala, Sweden

ABSTRACT

ARTICLE HISTORY

Objective: To evaluate transjugular intrahepatic portosystemic shunt (TIPS) in variceal bleeding in a clinical setting. Materials and methods: Retrospective review of 131 patients (116 with liver cirrhosis) treated with TIPS with covered stent grafts in a single centre from 2002 to 2016. Results: Survival at 1 and 2 years was 70% and 57% in patents with, and 100% at 2 years in patients without liver cirrhosis, respectively. A high Child–Pugh score and severe hepatic encephalopathy (HE) within 12 months post-TIPS were related to increased mortality. Re-bleeding occurred in 8% within 12 months and was related to TIPS dysfunction and a post-TIPS portosystemic gradient (PSG) of
5 mmHg. The main cause of TIPS dysfunction was that the stent did not fully reach the inferior vena cava. There was no correlation between the PSG and the occurrence of HE. Conclusions: TIPS was safe and prevented re-bleeding in patients with variceal bleeding, with or without liver cirrhosis, regardless of Child–Pugh class and of how soon after bleeding onset, the TIPS procedure was performed. A post-TIPS PSG of
5 mmHg was associated with an increased risk for re-bleeding and there was no correlation between the post-TIPS PSG and the occurrence of HE.

Received 9 July 2017 Accepted 22 September 2017 KEYWORDS

Variceal bleeding; TIPS; transjugular intrahepatic portosystemic shunt; portal hypertension; liver cirrhosis; hepatic encephalopathy; portosystemic gradient; TIPS dysfunction; re-bleeding

Introduction

Definitions

Specific primary treatments for acute variceal bleeding include vasoactive drugs, antibiotic prophylaxis and endoscopic haemostasis with ligation or tissue adhesive compounds, depending on the bleeding site [1]. The creation of a transjugular intrahepatic portosystemic shunt (TIPS) using covered stents has proved to be a safe and effective secondary prevention in variceal bleeding [2–5]. The aim of this study was to evaluate the safety and efficacy of TIPS in patients with portal hypertension and variceal bleeding in a clinical setting. Endpoints were survival, re-bleeding, TIPS dysfunction and adverse events.

Re-bleeding was defined as haematemesis or melena with transfusion of two or more units of blood. TIPS dysfunction was defined as stenosis or occlusion of the TIPS. Hepatic encephalopathy (HE) was considered to be mild when it subdued with treatment of the precipitation factor and/or with laxatives, and it was considered to be severe when recurrent or permanent [8]. Complications were defined as major if they required in-hospital treatment.

Materials and methods Study group and patient data Informed consent was collected from the patients with portal hypertension and variceal bleeding who were treated with TIPS at our hospital between January 2002 and May 2016. After approval from the ethical board, the medical records and radiology images were retrospectively reviewed. The Child–Pugh score [6] and the model for end-stage liver disease (MELD) score [7] were used to assess liver disease in patients with cirrhosis. There was no standard follow-up, but whenever TIPS dysfunction was suspected, patients were referred for re-intervention.

Technical procedure TIPS was performed under general anaesthesia according to standard methods as previously described [9], with or without the guidance of a wire in a portal venous branch. The pressures in the right atrium and in the portal vein (PV) were measured to determine the portosystemic gradient (PSG). PTFE-covered TIPS stents with a diameter of 10 mm (GOREV VIATORRV Tips Endoprosthesis; Gore Medical, W.L. Gore & Associates, Inc., Flagstaff, AZ) were used and dilated to 8 mm (n ¼ 102) or 10 mm (n ¼ 25). In four patients, the diameter after dilatation was not recorded. Embolisation of collaterals was performed in one patient. R

R

Statistical analysis StatView 5.0.1 (SAS Institute, Cary, NC) was used for statistical analyses. Differences between groups were estimated with

CONTACT Kerstin Rosenqvist [email protected], [email protected] Interventional Radiology, Uppsala University Hospital, 75185 Uppsala, Sweden ß 2017 Informa UK Limited, trading as Taylor & Francis Group

Department of Radiology, Section of

2

K. ROSENQVIST ET AL.

Table 1. Baseline characteristics.

Downloaded by [UNIVERSITY OF ADELAIDE LIBRARIES] at 20:02 11 October 2017

Characteristics No. of patients Age (year) mean (range) Cause of portal hypertension, no. of patients (%) Alcohol cirrhosis HCV/HBV cirrhosis Autoimmune liver/biliary disease NAFLD Othera Site of variceal bleeding, no. of patients (%) Oesophageal Gastric Stoma Rectal Caput medusa Child–Pugh classification  no. of patients (%) Class A Class B Class C Unknown MELD score Mean (range) 18 or higher no. of patients (%) Unknown Ascites, no. of patients (%) Bilirubin (mg/dl) mean Albumin (g/litre) mean Creatinine (mg/dl) mean Hepatic encephalopathy, no. of patients (%) Portal vein thrombosis, no. of patients (%) TIPS at primary episode of variceal bleeding, no. of patients (%) Endoscopic treatment at the time of bleeding (more than one option possible), no. of patients (%) Band ligation Injection sclerotherapy Sengstaken sond or Danis stent Vasoactive drug therapy at the time of bleeding, no. of patients (%) Terlipressin Somatostatin

Cirrhotic patients

Non-cirrhotic patients

116 57 (20–78)

15 43 (20–71)

62 15 16 11 12

(53) (13) (14) (10) (10)

0 0 5 0 10

(0) (0) (33) (0) (67)

89 20 6 0 1

(77) (17) (5) (0) (1)

13 0 1 1 0

(86) (0) (7) (7) (0)

20 60 30 6

(17) (52) (26) (5)

13 (5–30) 23 (20) 7 (6) 67 (58) 2.2 18.6 0.88 9 (8) 10 (9) 50 (43)

– – – – – 3 (20) 1.1 23.4 0.83 0 (0) 7 (47) 8 (53)

75 (65) 30 (26) 28 (24)

13 (87) 1 (7) 1 (7)

65 (56) 6 (5)

3 (20) 0 (0)

No.: number; a: portal vein thrombosis (n ¼ 7), cryptogenic cirrhosis (n ¼ 5), cirrhosis without specified cause (n ¼ 3), idiopathic liver trauma (n ¼ 2), inflammatory bowel disease (n ¼ 2), Felty’s syndrome (n ¼ 1), a-1-antitrypsin deficiency (n ¼ 1), cystic fibrosis (n ¼ 1) and common variable immune deficiency (n ¼ 1).

Fisher’s exact test for categorical data and with the Mann–Whitney U-test for continuous data. Simple unadjusted analyses were performed between the parameters displayed in Tables 2 and 3 as independent variables and overall mortality (Table 2) and re-bleeding (Table 3) as dependent variables, respectively. Parameters displaying significant associations in the simple analyses were included in multiple regression analyses. Cox’s proportional hazards model and the Kaplan–Meier method were used to estimate survival in the three Child–Pugh groups. The significance level was set at 0.05 in all analyses.

Results One patient was lost to follow-up. The remaining 131 were included in the study, 116 of them had liver cirrhosis. Ninetysix of the 131 patients (73%) were treated with non-selective beta-blockers. At the time of variceal bleeding, patients were treated with vasoactive drugs and antibiotics, and endoscopic treatment with insufficient result. The patients were then referred for TIPS as rescue treatment. Baseline characteristics are presented in Table 1. In three patients with extensive chronic portal vein thrombosis (PVT), the first attempt of TIPS was abandoned as the

risk of complications was considered too high. However, TIPS was successful at a second attempt in two patients when they presented with re-bleeding and in the third patient five days after the first attempt. Thus, the primary technical success rate of TIPS was 98% (n ¼ 128/131). The mean PSG dropped from 18 ± 7 mmHg before TIPS to 5 ± 2 mmHg after TIPS. There was no association between the site of the variceal bleed and the frequency of re-bleeding, or with mortality. Fourteen patients had a liver transplant (median 267 days post-TIPS). There was no correlation between TIPS dysfunction and liver transplantation. Five of the patients were included in a previously reported study [10]. The median follow-up time was 35 months (4–108 months).

Survival Overall survival was 92%, 70% and 57% at 6 weeks, 1 year and 2 years, respectively, in patients with liver cirrhosis, and 100% at 2 years in patients without cirrhosis. The cause of death was related to liver disease in 23/53 Child–Pugh B and C patients, but was not in Child–Pugh A patients. An increased mortality was associated with having a higher Child–Pugh score, a higher MELD score, liver cirrhosis

SCANDINAVIAN JOURNAL OF GASTROENTEROLOGY

3

Table 2. Simple and multiple regression analysis of factors associated with overall mortality in patients treated with transjugular intrahepatic portosystemic shunt (TIPS). Parameters in the entire cohort (n ¼ 131)

Parameters available in patients with liver cirrhosis (n ¼ 116)

Age, mean ± SD (years) Liver cirrhosis, n (%) Child–Pugh, mean ± SD MELD score, mean ± SD Severe HE within 12 months from TIPS, n (%)

Alive (n ¼ 73)

Dead (n ¼ 58)

p-Value simple analysis

CC simple analysis

p-Value multiple analysis

CC multiple analyses

53 ± 14 60 (82) 8±2 12 ± 5 2 (3)

62 ± 8 56 (97) 9±2 14 ± 5 10 (17)

.0002 .01 .0006 .021 .004

– 0.22 0.32 0.22 0.25

.015 – .025 .74 .02

– – 0.27 – 0.21

MELD: model of end-stage liver disease; HE: hepatic encephalopathy; n: number; CC: correlation coefficient.

Downloaded by [UNIVERSITY OF ADELAIDE LIBRARIES] at 20:02 11 October 2017

Table 3. Simple and multiple regression analysis of factors associated with re-bleeding in patients treated with transjugular intrahepatic portosystemic shunt (TIPS). Parameters in the entire cohort (n ¼ 131)

Parameters available in patients with liver cirrhosis (n ¼ 116)

No re-bleeding (n ¼ 121)

Re-bleeding (n ¼ 10)

p-Value simple analysis

CC simple analysis

p-Value multiple analysis

Child–Pugh, mean ± SD MELD score, mean ± SD

5.2 ± 2.3 9 (7) 8 ± 1.9 13 ± 5.3

7.6 ± 3.5 5 (50) 10 ± 1.4 14 ± 3

.003