Clinical signiﬁcance of small-bowel villous edema in patients
with liver cirrhosis: A capsule endoscopy study
Ichiro Otani,* Shiro Oka,* Shinji Tanaka,
Akiyoshi Tsuboi,* Sayoko Kunihara,* Yuko Nagaoki,*
Hiroshi Aikata* and Kazuaki Chayama*
Departments of *Gastroenterology and Metabolism, and
Endoscopy, Hiroshima University Hospital, Hiroshima, Japan
capsule endoscopy, liver cirrhosis, portal
hypertensive enteropathy, villous edema.
Accepted for publication 25 September 2017.
Dr Shiro Oka, Department of Gastroenterology
and Metabolism, Hiroshima University Hospital,
1-2-3 Kasumi, Minami-ku, Hiroshima 734-8551,
Declaration of conflict of interest: The authors
have no conﬂict of interest to declare.
Financial support: The study was supported by
departmental resources only.
Background and Aim: The relationship between the presence of villous edema (VE) in
portal hypertensive enteropathy and clinical factors remains unclear. The aim of this study
was to reveal the clinical factors related to VE in patients with liver cirrhosis (LC), and
investigate the clinical signiﬁcance of VE.
Methods: Between February 2009 and September 2016, 363 consecutive patients with LC
underwent capsule endoscopy for diagnosing portal hypertensive enteropathy at Hiroshima
University Hospital. We evaluated the relationship between the presence of VE and
patients’ clinical characteristics, ﬁndings of esophagogastroduodenoscopy and computed
tomography, and survival time.
Results: Villous edema was observed in 131 patients (36%), and severe lesions were found
in 71 (20%). The presence of VE was signiﬁcantly greater in patients with Child–Pugh
classiﬁcation B or C, esophageal varices, portal hypertensive gastropathy (PHG), ascites,
portal vein thrombosis (PVT), and splenomegaly. In multivariate analysis, Child–Pugh
class B or C, esophageal varices, PVT, and splenomegaly were signiﬁcant predictive
factors for the presence of VE. Severe VE was signiﬁcantly greater in patients with
Child–Pugh class B or C, serum albumin level ≤ 3.2 mg/dL, PHG, and PVT. In multivariate
analysis, PHG, Child–Pugh class B or C, PVT, were signiﬁcant predictive factors for
Conclusions: Clinical factors related to portal hypertension were signiﬁcantly correlated
with VE. In particular PVT was correlated with the appearance and exacerbation of VE.
Periodic capsule endoscopy in LC patients may lead to early detection of portal hyperten-
sion and PVT.
For patients with liver cirrhosis (LC), increased intrahepatic vascu-
lar resistance and blood ﬂow of the portal venous system cause
portal hypertension. Jonas et al.
reported that portal hypertension
alters the gastrointestinal tract mucosa and increases susceptibility
to injury. Impaired oxygenation of the mucosa is the probable
mechanism for this increased susceptibility
; decreased hemoglo-
bin oxygen saturation has been documented in the duodenal
mucosa of patients with LC.
The GI tract of patients with LC is
subject to various pathological lesions. Mucosal edema, erythema,
angioectasia, and varices have been detected mainly in the
stomach and colon of these patients.
Recently, capsule endoscopy (CE) has improved the detectabil-
ity of small-bowel lesions
and has become the gold standard
non-invasive technique. With the spread of CE, portal hyperten-
sion in the small bowel
has been reported. Portal hypertensive
enteropathy (PHE) is deﬁned as inﬂammatory-like lesions (edema-
tous, erythematous, granular, and friable lesions) and/or vascular
lesions (cherry red spots, telangiectasias, or angiodysplasia-like
lesions and varices).
Akyuz et al.
reported that PHE was found
in 90.5% of patients with portal hypertension. We also reported that
PHE was detected by CE in 67% of patients with LC.
However, the clinical signiﬁcance of PHE is unclear, and factors
reported to be associated with the appearance of PHE differ
between studies. While one study reported that PHE is signiﬁcantly
associated with Child–Pugh class C cirrhosis, varices, the presence
of portal hypertensive gastropathy (PHG), and portal hypertensive
another reported that there was no correlation between
the presence of PHE and these factors (Table 1).
The difference in clinical factors related to PHE in patients
with LC was caused by the inclusion of different types of PHE
ﬁndings such as erosions, edematous, erythematous, and telangi-
ectasias. Therefore, we focused on one of the typical PHE
ﬁndings, villous edema (VE), which is the edematous change of
the villi found in the small bowel of LC patients,
small-bowel edema is strongly correlated with portal hyperten-
The aim of this study was to reveal the clinical factors
related to VE in patients with LC and investigate the clinical signif-
icance of VE.
Journal of Gastroenterology and Hepatology 33 (2018) 825–830
© 2017 Journal of Gastroenterology and Hepatology Foundation and John Wiley & Sons Australia, Ltd