OBES SURG (2017) 27:3092–3101

DOI 10.1007/s11695-017-2748-9

ORIGINAL CONTRIBUTIONS

Reflux, Sleeve Dilation, and Barrett’s Esophagus

after Laparoscopic Sleeve Gastrectomy: Long-Term Follow-Up
Daniel Moritz Felsenreich 1 & Ronald Kefurt 1 & Martin Schermann 2 & Philipp Beckerhinn 3 &
Ivan Kristo 1 & Michael Krebs 4 & Gerhard Prager 1 & Felix B. Langer 1

Published online: 8 June 2017

# Springer Science+Business Media New York 2017

Abstract Methods We performed 24-h pH metries, manometries, gas-

Background Laparoscopic sleeve gastrectomy (SG) has be- troscopies, and questionnaires focusing on reflux (GIQLI,
come the most frequently performed bariatric procedure RSI) in SG patients with a follow-up of more than 10 years
worldwide. De novo reflux might impact patients’ quality of who did not suffer from symptomatic reflux or hiatal hernia
life, requiring lifelong proton pump inhibitor medication. It preoperatively.
also increases the risk of esophagitis and formation of Results From a total of 53 patients, ten patients after adjust-
Barrett’s metaplasia. Besides weight regain, gastroesophageal able gastric banding were excluded. From the remaining 43,
reflux disease (GERD) is the most common reason for con- six patients (14.0%) were converted to RYGB due to intrac-
version to Roux-en-Y gastric bypass. table reflux over a period of 130 months. Ten out of the re-
maining non-converted patients (n = 26) also suffered from
symptomatic reflux. Gastroscopies revealed de novo hiatal
* Gerhard Prager
gerhard.prager@meduniwien.ac.at hernias in 45% of the patients and Barrett’s metaplasia in
15%. SG patients suffering from symptomatic reflux scored
Daniel Moritz Felsenreich significantly higher in the RSI (p = 0.04) and significantly
moritz.felsenreich@meduniwien.ac.at lower in the GIQLI (p = 0.02) questionnaire.
Ronald Kefurt Conclusions This study shows a high incidence of Barrett’s
ronald.kefurt@meduniwien.ac.at esophagus and hiatal hernias at more than 10 years after SG.
Its results therefore suggest maintaining pre-existing large hi-
Martin Schermann
martin.schermann@wienkav.at atal hernia, GERD, and Barrett’s esophagus as relative contra-
indications to SG. The limitations of this study—its small
Philipp Beckerhinn
sample size as well as the fact that it was based on early
philipp.beckerhinn@hollabrunn.lknoe.at
experience with SG—make drawing any general conclusions
Ivan Kristo about this procedure difficult.
ivan.kristo@meduniwien.ac.at
Michael Krebs
michael.krebs@meduniwien.ac.at Keywords Sleevegastrectomy . Reflux . GERD . Conversion
to RYGB . Long-term data
Felix B. Langer
felix.langer@meduniwien.ac.at

1
Division of General Surgery, Department of Surgery, Medical Introduction
University of Vienna, Vienna, Austria
2
Department for Surgery, Hospital Rudolfsstiftung, Vienna, Austria Laparoscopic sleeve gastrectomy (SG) has become the most
3
Department for Surgery, Hospital Hollabrunn, Hollabrunn, Austria commonly performed bariatric procedure in the world. It rep-
4
Division of Endocrinology, Department of Internal Medicine, resented 37% of all procedures worldwide in 2013. In many
Medical University of Vienna, Vienna, Austria countries, such as the USA, Canada, or Australia, SG has
OBES SURG (2017) 27:3092–3101
already become the most frequently performed bariatric pro-
cedure [1].
GERD can be seen as one of SG’s most important side
effects and is still controversially discussed in the litera-
ture. Severe reflux does not only impact a patient’s quality
of life and forces them to permanently depend upon pro-
ton pump inhibitors, it can actually lead to esophagitis,
which can in turn cause Barrett’s esophagus—a potential
factor for the development of esophageal carcinoma [2].
This process, or cascade, does undoubtedly take years and
can therefore only be identified by long-term studies.
Short-term studies often do not detect de novo GERD
or even show an improvement of preoperative reflux
[3–5]. This effect is likely caused by the initially impres-
sive weight loss leading to a decreased intra-abdominal
pressure [6]. Studies presenting a longer follow-up have
shown an increase in symptomatic reflux after SG [7, 8],
possibly due to weight regain or the development of de
novo hiatal hernias caused by the pressure within the gas-
tric sleeve for years [9].
In our long-term series, severe GERD following SG repre-
sented one of the most common indications for conversion to
RYGB, [10–12], there are alternative options to the conver-
sion to RYGB, such as the LINX ring [13], re-sleeve gastrec-
tomy [14, 15], cardiopexy [16], and permanent conservative
reflux management [17].
The aim of this study was to determine the 10-year out-
come of SG in terms of GERD and long-term complications,
such as Barrett’s esophagus or de novo hiatal hernia develop-
ment—which were contraindications to the procedure—with-
in a series of the first patients in Austria that had not been
converted to RYGB within the follow-up period.
Methods
All consecutive patients who underwent SG in three Austrian
bariatric centers from January 2003 to December 2005 were
included in this retrospective multicenter study. Informed con-
sent was obtained from all individual participants included in
the study. The Ethics Committee of the Medical University of
Vienna, serving as the local Institutional Review Board
(Reference number 1434/2015), approved of this study.
Surgical Technique and Indications
The method of how SG is performed at our center has
been described by Langer et al. [10]. A 42–48-French
bougie was used for calibration in this series. The bou-
gie size and extent of resection, resulting in the preser-
vation of the antrum, were common in terms of surgical
technique in the early years of SG. Symptomatic reflux,
Barrett’s esophagus, and hiatal hernias, diagnosed
3093
through preoperative gastroscopies, were contraindica-
tions to SG. An intraoperational visualization of the left
hiatal crus was done routinely to rule out undiagnosed
hiatal hernias. Despite preoperative gastroscopies and
intraoperative visualization of the hiatal crura, the pos-
sibility of undetected hiatal hernias may not be entirely
eliminated.
Also, only patients who were not infected with
Helicobacter pylori or in whom H. pylori had been eradicated
underwent SG.
Follow-up
All patients were contacted by the Medical University Vienna
via mail and telephone. Great efforts were made to achieve a
complete 10-year follow-up in all patients. As it is absolutely
mandatory for Austrian inhabitants to inform the registration
office about their current home address, we were able to reach
all patients based on the information from the central state
registry. Ten patients were excluded from the present study
due to the fact that they had been converted to SG from
AGB. Sixteen out of the remaining 43 patients (37%) were
converted to RYGB in the postoperative follow-up period, and
one patients died unrelated to surgery. Twenty of a total of 26
patients who had not been converted to RYGB so far followed
our invitation to participate in a complete functional workup
consisting of gastroscopy, pH-metry, 24-h manometry,
Gastrointestinal Quality of Life Index (GIQLI), and Reflux
Symptom Index (RSI) questionnaires.
Quality of Life Questionnaires
The GIQLI, developed by Eypasch et al., is a validated ques-
tionnaire. It contains 36 questions with five response catego-
ries each. Patients’ responses are summed in order to receive a
numerical score reaching from 0 to 144 points [18]. The RSI is
a validated score developed by Belafsky et al. to detect
laryngopharyngeal reflux and consists of nine items with a
total of 45 points to be scored, 5 per question. A score higher
than 10 is interpreted as abnormal, while higher than 13 indi-
cates pathologic [19].
Statistical Analysis
Data in this study is presented either within the median and
range, by mean and standard deviation or as percentages,
where appropriate. When comparing groups of data, chi-
square tests and the non-parametric Mann-Whitney U test
were used. All univariate analyses were two-tailed, with sig-
nificance set at a p value of <0.05. SPSS v22 for Windows
was used for all statistical calculations.
3094
Results
A total of 53 patients, 79% female, underwent SG in the three
participating Austrian bariatric centers between January 2003
and December 2005. Forty-one patients received SG as a first
bariatric procedure, one patient each after gastric stimulation
(TANTALUS®), endoscopic gastric balloon placement and
10 after AGB, who were excluded from this study. At the time
of surgery, the mean age of this study’s collective was
38.4 ± 12.4 years, and the mean weight and BMI were
139.9 ± 27.8 kg and 49.5 ± 9.6 kg/m2, respectively. The me-
dian postoperative time of this long-term study ended up at
129 months. No GERD patient underwent SG, as reflux and
hiatal hernia were contraindications to SG. Two patients died
unrelated to SG.
Within the 10-year follow-up period, a total of 16 patients
(37%) were converted to RYGB, ten (23%) due to weight
regain, six (14%) due to reflux as the main symptom, and
one patients were converted for acute leakage (2%).
Symptomatic reflux was found in 38% of the non-
converted patients (9 females/1 male), as defined and recog-
nized by clinically interviewing patients about their symp-
toms, such as heartburn, regurgitation, an acidic taste in the
mouth, pain with swallowing or a sore throat after eating,
coughing, increased salivation, or chest pain. Seven (70%)
of the 10 non-converted patients suffering from reflux take
PPI on a daily basis. Three take PPI only if necessary, depend-
ing on their symptoms. Two patients without reflux symptoms
take PPI for other reasons.
Gastroscopy
Twenty of our participating non-converted patients underwent
gastroscopy performed at a minimum of 10-year follow-up
(Table 1, Fig. 1). De novo hiatal hernia developed in nine
patients (45%) during the follow-up period. Also, patients
suffering from symptomatic reflux were significantly more
likely to develop hiatal hernia (n = 7; 70%) than the ones
who did not. A significant correlation could also be found
between symptomatic reflux and esophagitis visible during
gastroscopy (p = 0.05).
Columnar lined esophagus (CLE) was found in a total of 10
patients (symptomatic reflux, 7; no reflux, 3). However, the
CLE is significantly longer in patients who suffer from symp-
tomatic reflux (4.0 mm) than in patients who do not (2.3 mm)
(p = 0.013).
Sleeve enlargement, defined by an easily possible inversion
with the gastroscope, correlating to a minimum sleeve diam-
eter of 5 cm, was found in 60% of the patients. In 15% (n = 3)
of the patients—with a mean age of 47.3 ± 14.7 years—
H. pylori was found, combined with active gastritis in every
case. The most striking finding was Barrett’s esophagus with-
out dysplasia in three patients (15%). However, the histology
OBES SURG (2017) 27:3092–3101
revealed no correlation between an infection with H. pylori
and the incidence of Barrett’s esophagus (p = 0.3).
Manometry
Fourteen of the participating non-converted patients (54%)
underwent manometry (Table 2). There is no preoperative
manometry data available, which is certainly a limitation of
this study. On average, only 30.0% out of 10 swallows could
be performed properly, i.e., patients managed a complete bo-
lus transit. No significant difference could be found between
patients with (28.6%) or without symptomatic reflux (31.7%).
In total, 50.0% of the patients (n = 7) managed no complete
bolus transit (in 10 trials). Also, there was neither any signif-
icant difference concerning lower esophageal sphincter pres-
sure (LESP) (p = 0.58) nor the liquid bolus transit time
(p = 0.90) between symptomatic GERD patients and asymp-
tomatic patients. No correlation could be found between
sleeve enlargement or hiatal hernia and bolus transit
(p = 0.27/p = 0.48) or abnormal LESP (p = 0.42/p = 0.65),
respectively. The same goes for a correlation between bolus
transit and LESP (p = 0.16).
24-h pH-Metry
Only 12 patients (46%) underwent 24-h pH-metry. An abnor-
mally high number of refluxes was found in 41.7% (n = 5) of
the patients (62.7 ± 37.4 on average) within 24 h. The acid
exposure time was found increased in 58.3% (n = 7) of the
patients; in 9.7 ± 8.0% (normal <4.2%) of the time over a
period of 24 h, patients’ pH value measured below 4.0. The
average DeMeester Score (n = 12) was 52.7 ± 45.2 (normal
<14.72) and increased in 75% of the patients (n = 9).
Interestingly, the score was increased not only in patients with
but also in those without symptomatic reflux in this study
(p = 0.81) (Table 3).
Two patients were not even able to discontinue their PPI
10 days prior to the 24-h pH-metry due to intense heartburn.
Thus, acid suppression had to be used in order to perform the
examinations. Both of their gastroscopies showed hiatal her-
nias and CLE areas at the gastroesophageal junction.
Questionnaires
Twenty of our non-converted patients (77%) completed two
different types of questionnaires focusing on gastrointestinal
quality of life using the GIQLI (Fig. 2) and about reflux-
related symptoms using the RSI (Fig. 3).
In the GIQLI, patients with symptomatic reflux (105.3 ± 18.4)
scored significantly lower than the ones without (122.9 ± 10.4)
(p = 0.02). Patients suffering from hiatal hernia scored
112.4 ± 16.0 and those with Barrett’s scored 113.3 ± 27.3; both
did not show any significant difference to those without
OBES SURG (2017) 27:3092–3101 3095

Table 1 Endoscopic and

histologic upper-GI findings at Non-converted Symptomatic reflux Non-symptomatic reflux p value
10 years after SG Gastroscopy macroscopic (n = 20) (n = 10) (n = 10)

Fundus residual (%) 20 10 30 0.29

Hiatal hernia (%) 45 70 20 0.02
Bile in the stomach (%) 25 20 30 0.63
a
Enlarged sleeve (%) 60 70 50 0.39
Gastritis and ulcers (%) 45 50 40 0.67
Esophagitis (%) 30 50 10 0.05
CLE (GE junction) (%) 50 70 30 0.08
Mean size (mm)b 3.5 (R 2–5) 4.0 (R 3–5) 2.3 (R 2–3) 0.01
Chronic gastritis (%) 80 90 70 0.29
Active gastritis (%) 45 40 50 0.67
Dysplasia (%) 0 0 0 N/A
Barrett’s esophagus (%) 15 10 20 0.56
Acanthosis/parakeratosis (%) 20 30 10 0.29
Hyperregeneratory
esophagopathy (%) 60 70 50 0.39
Helicobacter pylori (%) 15 10 20 0.56

SG sleeve gastrectomy, CLE columnar lined esophagus, GE gastroesophageal

a
Enlarged sleeve was defined as inversion with a gastroscope equaling a 5-cm diameter
b
n = 10

(p = 0.69/p = 0.70). Patients with enlarged sleeve or H. pylori Discussion

scored low in the GIQLI (110.7 ± 19.4/107.3 ± 23.6) but did not
show any significant difference to patients without (p = 0.28
/p = 0.47).
In total, 15.0% (n = 3) showed a pathologic score of more
than 13 points in the RSI. Patients suffering from symptomatic
reflux (8.4 ± 7.4) scored significantly higher than the ones
who do not (2.8 ± 3.1) (p = 0.04). Those with hiatal hernias
scored 6.8 ± 5.8 (p = 0.71), and patients with Barrett’s esoph-
agus scored 10.3 ± 11.4 (p = 0.08). The presence of H. pylori
and enlarged sleeve does not affect RSI scores (p = 0.12/
p = 0.86).
This study presents endoscopic and functional long-term re-
sults of SG with a follow-up of at least 10 years. One out of
three patients of this study’s population underwent conversion
for weight regain or GERD. De novo hiatal hernia was found
in 45% and Barrett’s esophagus in 15% of the patients.
Summing up patients who were converted to RYGB and the
non-converted patients who took part in the study, GERD was
found in up to 38% of the patients (15 females/1 male) in the
long-term follow-up. What makes these numbers even more
striking is the fact that hiatal hernia, Barrett’s esophagus, and

Fig. 1 Non-converted patients’ 80

p=0.02
current gastroscopy (%).
70
Symptomatic reflux was defined
by interviewing patients about 60 p=0.05
symptoms
50

40

30

20

10

0
Hiatal hernia Remnant Enlarged Gastris / Esophagis CLE Barre´s
fundus Sleeve Ulcers metaplasia

Symptomac reflux No reflux

3096 OBES SURG (2017) 27:3092–3101

Table 2 Manometry results at

10 years after SG Non- Symptomatic Non-symptomatic p value
converted reflux reflux
Manometry (n = 14) (n = 8) (n = 6)

LESP (mmHg) 25.3 ± 11.0 26.8 ± 13.3 23.3 ± 7.4 0.58

(Normal 10–35 mmHg)
Patients increased (%) 14.3 25.0 0
Patients decreased (%) 7.1 12.5 0
Swallow bolus transit: 10 trials 30.0 ± 40.4 28.8 ± 36.0 31.7 ± 49.2 0.90
(%)
(Normal >80%)
0 of 10 trials pos. (%) 50.0 37.5 66.7
Patients decreased (%) 78.6 87.5 66.7
Liquid bolus transit timea (s) 6.1 ± 2.0 5.4 ± 2.0 7.8 ± 0.4 0.19
(n = 7)
(Normal <12 s)
Patients not measurable 50.0 37.5 66.7
Patients increased (%) 0 0 0
Integrated relaxation pressure 17.5 ± 8.4 18.9 ± 10.5 15.7 ± 4.7 0,50
(Normal <15 mmHg)
Patients increased (%) 57.1 62.5 50.0
Distal contractile integral 2135 ± 1681 1814 ± 1095 2562 ± 2298 0.43
(Normal
450–8000 mmHg/s/cm)
Patients increased (%) 0 0 0
Patients decreased (%) 0 0 0

SG sleeve gastrectomy, LESP lower esophageal sphincter pressure

a
Only in normal swallow trials

GERD were all contraindications to SG diagnosed through
gastroscopies preoperatively performed on all patients.
The question whether GERD can be improved, even cured
or—on the contrary—is actually caused by sleeve gastrecto-
my (de novo reflux), has not been clearly answered yet. Thus,
a number of studies, presenting short- to long-term data, can
be found with a wide range of varying results on this topic. A
brief overview of the literature focusing on GERD after SG is
provided in Table 4 for reference [3, 4, 8, 20–22].
Sleeve Gastrectomy and Reflux
Short-term studies do not detect de novo GERD or do
even find an improvement of preoperative reflux [3–5,
22]. This effect is likely caused by the initially impressive
weight loss which in turn leads to a decreased intra-
abdominal pressure [6].
Sharma et al., for example, performed SG on patients suf-
fering from GERD preoperatively (as well as on asymptomatic

Table 3 The 24-h ph-metry

results at 10 years after SG Non-converted Symptomatic reflux Non-symptomatic reflux p value
24 h ph-metry (n = 12) (n = 6) (n = 6)

Acid exposure (%) 9.7 ± 8.0 9.7 ± 7.5 9.7 ± 9.2 0.99
(Normal <4.2%)
Patients increased (%) 58.3 66.7 50
Reflux activity (nr.) 62.7 ± 37.4 63.4 ± 37.8 62.0 ± 41.5 0.96
(Normal <73)
Patients increased (%) 41.7 50.0 20
DeMeester Score 52.7 ± 45.2 56.0 ± 40.9 49.3 ± 52.9 0.81
(Normal <14.72)
Patients increased (%) 75.0 83.3 66.7

SG sleeve gastrectomy
OBES SURG (2017) 27:3092–3101 3097

Fig. 2 Non-converted patients’

current GIQLI. Symptomatic
reflux was defined by
interviewing patients about
symptoms. GIQLI
Gastrointestinal Quality of Life p=0,02
Index

ones) and found GERD improvement 12 months after the pro-
cedure using questionnaires. The Carlsson Dent Score and the
Severity of Symptoms Score, both used in this study, showed a
significant improvement of GERD from 2.88 to 1.63 and 2.28
to 1.06, respectively. The authors thus conclude that GERD
does not necessarily have to be a contraindication to SG even
though scintigraphy showed an increase of reflux in 71.7% of
their patients [23]. Rebecchi et al. performed gastroscopy, ma-
nometry, and pH-metry postoperatively in a study population
of 71 patients, 43% of whom suffered from GERD preopera-
tively. Contrary to our findings, they detected an improved
reflux in 86% of their patients and a decreased DeMeester
Score of 39.5 ± 16.5 down to 10.6 ± 5.8 after 2 years [5].
Ece et al. used a GERD questionnaire and 24 h pH-metries
in a study of 402 patients preoperatively. In 59 patients with a
DeMeester Score of over 14.7, a routine hiatoplasty was per-
formed. The authors found that only 3.3% of patients showed
reflux symptoms after 12 months; one additional patient had

Fig. 3 Non-converted patients’

current RSI. Symptomatic reflux
was defined by interviewing
patients about symptoms. GRSI
Reflux Symptom Index

p=0,04
3098
to receive PPI treatment. They conclude that carefully
selecting patients and surgical technique may positively affect
reflux symptoms [24].
However, Sheppard et al. found 58% of their patients in-
creasing the dose of PPI after SG in order to improve reflux
symptoms. GERD had not been a contraindication to SG ei-
ther of this study [25].
Studies presenting mid-term outcomes of SG have sug-
gested the occurrence of GERD as a more or less important
side effect. The largest series so far (with 4832 patients) has
been presented by DuPree et al., who observed new-onset
GERD in 8.6% of the preoperatively non-GERD patients after
3 years. They therefore determined reflux a contraindication to
SG [26]. Also, a postoperative increase in GERD was found in
21% of the patients of Kular et al. after SG [27]. Reflux was no
contraindication here; they did, however, encourage their pa-
tients suffering from reflux preoperatively to undergo one
anastomosis gastric bypass instead. Boza et al. [28] found
the highest incidence of new-onset GERD with 26.7% at a
5-year follow-up of SG performed with 60 Fr. bougie calibra-
tions. Himpens et al. [8] observed reflux at 6 years postoper-
ative in 23% of the patients, while only 3.6% had suffered
from GERD preoperatively.
Sleeve and Hiatal Hernia Repair
A number of studies have been published in recent years in-
cluding patients treated with a combination of SG and
hiatoplasty in case of hiatal hernia. Sucandi et al. combined
SG with hiatoplasty in 67 patients suffering from reflux pre-
operatively. In total, 29.9% of their patients showed clinical
improvement of their symptoms while 47.7% were even
found to be cured [29]. Samakar et al. published a similarly
designed study, in which 34.6% of their patients showed a
resolution of their symptoms after having had SG with
hiatoplasty. However, 15.6% of their patients were found to
have new onset reflux after the procedure [28]. Lyon et al.
found an improvement of reflux in their 262 patients using
the Visick Score after SG, which was combined with
hiatoplasty only if needed. In their conclusion, they argue that
symptomatic reflux and hiatal hernia do not have to be treated
as contraindications to SG, contrary to our findings [30]. In a
study of 68 patients, however, Sieber et al. found an increase
in GERD after SG of 16.2% despite administering hiatoplasty
in case of hiatal hernia; both reflux and hiatal hernia were not
seen as contraindications to SG [31].
Soricelli et al. published a study with a total of 378 patients,
15.8% of whom initially suffered from preoperative GERD
and 14.5% from hiatal hernia. Eighteen months after SG with
hiatoplasty (if needed), they report GERD remission in 73.3%
of their patients. In total, 22.9% of patients who only had had
SG developed de novo GERD [32].
OBES SURG (2017) 27:3092–3101
In the study of Daes et al. with 382 SG patients with a
follow-up of 6–22 months, 45.5% were diagnosed with
GERD preoperatively and hiatal hernia was discovered in
37.2% intraoperatively. In total, 94% of patients suffering
from GERD preoperatively turned out asymptomatic at fol-
low-up. This study shows that the intraoperative search for
hiatal hernia and its repair, if detected, may prevent postoper-
ative reflux [22].
Boules et al. found in a study of 83 bariatric patients who
had had concomitant hiatal hernia repair that only 39% of
hiatal hernias were diagnosed preoperatively; the rest intraop-
eratively. They conclude that the incidence of hiatal hernia
may well be underestimated in bariatric patients [33]. Thus,
it may be difficult to distinguish between real Bde novo^ hiatal
hernias and pre- and intraoperatively missed small hernias,
which developed to significant hernias over the time.
GERD Treatment
Apart from the occurrence of GERD or de novo GERD, re-
spectively, an important point is its treatment, if it does devel-
op after SG. In our study, 11.3% of the patients suffering from
GERD after SG were converted to RYGB within the follow-
up period. Langer et al. also found, beside weight regain,
intractable GERD to be a common indication for conversion
to RYGB [10]. A recent prospective, randomized study of 86
patients by Rebecchi et al., on the other hand, has shown that
RYGB may lead to an increase of Bweakly acidic reflux^:
5 years after the procedure, 24 h pH-metry revealed a signif-
icant improvement of acid exposure. However, weakly acidic
reflux was found in 74% in a group of previously non-reflux
patients [34].
There are, of course, alternative ways to treat GERD after SG,
such as an implantation of a LINX® ring device. At this point,
only one study has been published on this approach; Desart et al.
treated postoperative reflux (DeMeester: mean score of 56.6)
after SG in their study of seven patients by implanting a magnetic
anti-reflux system (LINX® ring device) around the gastrojejunal
junction after a mean period of 18.1 months. Using a GERD
score, which asked questions about the severity and frequency
of typical reflux symptoms, 2 to 4 weeks after the implantation of
the device, they reported a symptomatic improvement of 12
points in all of their patients [13].
The treatment of GERD by implanting a magnetic ring device
has generally proven successful in more recent years. The first
series of patients suffering from primary GERD who were treated
with a LINX® ring device was published in 2010 by Bonavina
et al. GERD was eliminated successfully in a series of 44 pa-
tients, measured via 24-h pH-metry at 1 to 2 years follow-up
[35]. Reynolds et al. compared the LINX® device to the laparo-
scopic Nissen fundoplication in a matched pair analysis of 100
patients and found a similarly successful control of reflux symp-
toms after both methods [36].
OBES SURG (2017) 27:3092–3101
Another suggested way of treating GERD after SG is
performing a re-sleeve gastrectomy. Nedelcu et al., for example,
did a study on 61 patients who received a re-Sleeve; indications
were weight regain, insufficient weight loss, and GERD (in four
patients). Preoperative examinations were gastrografin swallow
and CT, and all four GERD patients went into remission [14].
Silecchia et al. published a study featuring 19 patients suffering
from GERD after SG. They were divided into groups organized
by hiatal hernia and residual fundus. All patients received a
fundectomy (re-sleeve) and hiatoplasty, if necessary, and experi-
enced a remission after 3 months. However, the authors also
report five complications [15].
For further reference, Crawford et al. present a comprehen-
sive and very recent overview over possible ways of dealing
with GERD after SG including radiofrequency ablation (RFA;
see also below), reflux management systems, revisional sur-
gery, and preventive procedures [37].
Barrett’s Metaplasia
This study shows an increase in Barrett’s esophagus from 0%
(preoperatively) to 14% (three patients) over the course of the
follow-up period. Interestingly, two of these patients do not
suffer from GERD and only one has symptomatic reflux. In
comparison, Barrett’s metaplasia can be found in 1.6% of the
average population [38].
The availability of data on the incidence of Barrett’s esoph-
agus in obese patients is generally quite limited at this point.
Drahos et al. showed in their study of over 60,000 patients that
a significantly higher number with Barrett’s esophagus were
obese (BMI > 30 kg/m2) as well (20.7%), as opposed to those
without Barrett’s (19.2%) [2]. One could deduct that obese
patients are more likely to develop Barrett’s esophagus as they
do suffer from reflux more often than patients with normal
weight due to a higher intra-abdominal pressure.
Braghetto et al. were the first to present the incidence of
Barrett’s esophagus after SG. They only found it in 1.2% of
their 231 patients (23.5% of which showed reflux symptoms)
at 5 years after SG [7]. A plausible reason for the striking
difference between the numbers presented by Braghetto
et al. and our results (1.2% at 5 vs. 14% at 10 years follow-
up) is certainly that the development of Barrett’s metaplasia
does take several years.
In any case, Michel Gagner has recently argued that no
consensus has been found on the question whether Barrett’s
esophagus should be considered an absolute contraindication
to SG [39]. Nineteen (0.95%) of the patients of Gorodner et al.
had Barrett’s esophagus before bariatric surgery, for example.
All of them received RYGB, and the authors recommend this
procedure due to a regression rate of 36% and no progression
to dysplasia [40]. Similarly, Kindel et al. recommend RYGB
instead of SG due to its superior reflux control in morbidly
obese patients with Barrett’s esophagus [41].
3099
Quality of Life
In our study, patients who suffer from symptomatic reflux
scored significantly lower in the GIQLI in this study. This
proves how high the impact of GERD can be on a person’s
gastrointestinal quality of life. The study of Overs et al. on
gastrointestinal quality of life following bariatric procedures,
on the other hand, found the highest GIQLI in SG (120.5),
compared to RYGB (113.0) and a preoperative control group
(96.0) [42].
Limitations of the Study
First of all, this study is based on a rather limited number of
patients (53), which might be too small to draw any final
conclusions about general success or failure rates of SG.
Also, as a retrospective analysis, it includes patients from
SG’s early days and the surgical technique has been slightly
modified since. These minor alterations include the size of the
bougie, which has changed from 48fr to 36fr—a difference of
a mere 4 mm—and the extent of the resection, which used to
be opposite the crow’s foot, is started closer today. Thus, the
results presented here might not be entirely comparable to the
10-year results of SG the way it is performed today. Surgeons
improve the surgical technique with the numbers of LSG per-
formed. This early series also includes the learning curve,
which might influence the incidence of de novo or intraoper-
atively missed hiatal hernias. Furthermore, there is no preop-
erative manometry and 24-h pH-metry data available for the
patients participating in this study. At 10 years, the sample size
for these examinations is rather small compared to the patient
collective. Finally, the fact that 19% of the patients were con-
verted from gastric banding to SG might have affected the
occurrence and severity of reflux, even though this study did
not prove a higher rate of reflux in these patients.
Conclusions
To sum up, this study presents a high rate of reflux after SG
in patients without clinical GERD symptoms preoperative-
ly, along with high rates of hiatal hernias and Barrett’s
esophagus, which had both been ruled out via gastros-
copies preoperatively. Thus, preoperative large hiatal her-
nia, GERD, and Barrett’s esophagus should be treated as
(relative) contraindications to this procedure. Patients with
small hiatal hernia may be treated with SG and hiatoplasty,
as a number of studies have shown. In case of post-SG
Barrett’s esophagus, treatment modalities such as perma-
nent PPI and RFA might be less effective than conversion
to RYGB. In any case, gastroscopies of the gastroesopha-
geal junction at regular intervals should be integrated in
routine SG follow-up to detect Barrett’s metaplasia in time.
3100
Due to the limitations of this study, i.e., its small sample
size and the fact that it is based on early experience with SG, it
is difficult to draw any general conclusions about this proce-
dure, which could only be done based on evidence gained
from further studies.
Compliance with Ethical Standards All procedures performed in
studies involving human participants were in accordance with the ethical
standards of the research committee of the Vienna Medical University and
with the 1964 Helsinki declaration and its later amendments or compara-
ble ethical standards.
Conflict of Interest The authors DM Felsenreich, R Kefurt, M
Schermann, P Beckerhinn, I Kristo, M Krebs, G Prager, and FB Langer
declare that they have no conflict of interest.
Human and Animal Rights and Informed Consent Informed con-
sent was obtained from all individual participants included in the study.
This article does not contain any studies with animals performed by any
of the authors.
References
1. Angrisani L, Santonicola A, Iovino P et al. Bariatric Surgery
Worldwide 2013. Obes Surg. 2015
2. Drahos J, Li L, Jick SS, et al. Metabolic syndrome in relation to
Barrett’s esophagus and esophageal adenocarcinoma: results from a
large population-based case-control study in the clinical practice
research Datalink. Cancer Epidemiol. 2016;42:9–14.
3. van Rutte PW, Smulders JF, de Zoete JP, et al. Outcome of sleeve
gastrectomy as a primary bariatric procedure. Br J Surg. 2014;101:
661–8.
4. Spivak H, Rubin M, Sadot E, et al. Laparoscopic sleeve gastrecto-
my using 42-French versus 32-French bougie: the first-year out-
come. Obes Surg. 2014;24:1090–3.
5. Rebecchi F, Allaix ME, Giaccone C, et al. Gastroesophageal reflux
disease and laparoscopic sleeve gastrectomy: a physiopathologic
evaluation. Ann Surg. 2014;260:909–14. discussion 914-905
6. Chiu S, Birch DW, Shi X, et al. Effect of sleeve gastrectomy on
gastroesophageal reflux disease: a systematic review. Surg Obes
Relat Dis. 2011;7:510–5.
7. Braghetto I, Csendes A. Prevalence of Barrett’s esophagus in bar-
iatric patients undergoing sleeve gastrectomy. Obes Surg. 2015
8. Himpens J, Dobbeleir J, Peeters G. Long-term results of laparo-
scopic sleeve gastrectomy for obesity. Ann Surg. 2010;252:319–24.
9. Deitel M, Crosby RD, Gagner M. The first international consensus
summit for sleeve Gastrectomy (SG), New York City, October 25-
27, 2007. Obes Surg. 2008;18:487–96.
10. Langer FB, Bohdjalian A, Shakeri-Leidenmuhler S, et al.
Conversion from sleeve gastrectomy to Roux-en-Y gastric by-
pass—indications and outcome. Obes Surg. 2010;20:835–40.
11. Homan J, Betzel B, Aarts EO, et al. Secondary surgery after sleeve
gastrectomy: Roux-en-Y gastric bypass or biliopancreatic diversion
with duodenal switch. Surg Obes Relat Dis. 2015;11:771–7.
12. Weiner RA, Theodoridou S, Weiner S. Failure of laparoscopic
sleeve gastrectomy—further procedure? Obes Facts.
2011;4(Suppl 1):42–6.
13. Desart K, Rossidis G, Michel M, et al. Gastroesophageal reflux
management with the LINX(R) system for gastroesophageal reflux
disease following laparoscopic sleeve Gastrectomy. J Gastrointest
Surg. 2015;19:1782–6.
OBES SURG (2017) 27:3092–3101
14. Nedelcu M, Noel P, Iannelli A, et al. Revised sleeve gastrectomy
(re-sleeve). Surg Obes Relat Dis. 2015;11:1282–8.
15. Silecchia G, De Angelis F, Rizzello M, et al. Residual fundus or
neofundus after laparoscopic sleeve gastrectomy: is fundectomy
safe and effective as revision surgery? Surg Endosc. 2015;29:
2899–903.
16. Galvez-Valdovinos R, Cruz-Vigo JL, Marin-Santillan E, et al.
Cardiopexy with ligamentum teres in patients with hiatal hernia
and previous sleeve gastrectomy: an alternative treatment for gas-
troesophageal reflux disease. Obes Surg. 2015;25:1539–43.
17. Khan A, Kim A, Sanossian C, et al. Impact of obesity treatment on
gastroesophageal reflux disease. World J Gastroenterol. 2016;22:
1627–38.
18. Eypasch E, Wood-Dauphinee S, Williams JI, et al. The gastrointes-
tinal quality of life index. A clinical index for measuring patient
status in gastroenterologic surgery. Chirurg. 1993;64:264–74.
19. Belafsky PC, Postma GN, Koufman JA. Validity and reliability of
the reflux symptom index (RSI). J Voice. 2002;16:274–7.
20. Vage V, Sande VA, Mellgren G, et al. Changes in obesity-related
diseases and biochemical variables after laparoscopic sleeve gas-
trectomy: a two-year follow-up study. BMC Surg. 2014;14:8.
21. Gibson SC, Le Page PA, Taylor CJ. Laparoscopic sleeve gastrecto-
my: review of 500 cases in single surgeon Australian practice. ANZ
J Surg. 2015;85:673–7.
22. Daes J, Jimenez ME, Said N, et al. Improvement of gastroesopha-
geal reflux symptoms after standardized laparoscopic sleeve gas-
trectomy. Obes Surg. 2014;24:536–40.
23. Sharma A, Aggarwal S, Ahuja V, et al. Evaluation of gastroesoph-
ageal reflux before and after sleeve gastrectomy using symptom
scoring, scintigraphy, and endoscopy. Surg Obes Relat Dis.
2014;10:600–5.
24. Ece I, Yilmaz H, Acar F et al. A new algorithm to reduce the
incidence of gastroesophageal reflux symptoms after laparoscopic
sleeve gastrectomy. Obes Surg. 2016.
25. Sheppard CE, Sadowski DC, de Gara CJ, et al. Rates of reflux
before and after laparoscopic sleeve gastrectomy for severe obesity.
Obes Surg. 2015;25:763–8.
26. DuPree CE, Blair K, Steele SR, et al. Laparoscopic sleeve gastrec-
tomy in patients with preexisting gastroesophageal reflux
diseasennn: a national analysis. JAMA Surg. 2014;149:328–34.
27. Kular KS, Manchanda N, Rutledge R. Analysis of the five-year
outcomes of sleeve gastrectomy and mini gastric bypass: a report
from the Indian sub-continent. Obes Surg. 2014;24:1724–8.
28. Samakar K, McKenzie TJ, Tavakkoli A, et al. The effect of laparo-
scopic sleeve gastrectomy with concomitant hiatal hernia repair on
gastroesophageal reflux disease in the morbidly obese. Obes Surg.
2016;26:61–6.
29. Sucandy I, Chrestiana D, Bonanni F, et al. Gastroesophageal reflux
symptoms after laparoscopic sleeve gastrectomy for morbid obesi-
ty. The importance of preoperative evaluation and selection. N Am
J Med Sci. 2015;7:189–93.
30. Lyon A, Gibson SC, De-loyde K, et al. Gastroesophageal reflux in
laparoscopic sleeve gastrectomy: hiatal findings and their manage-
ment influence outcome. Surg Obes Relat Dis. 2015;11:530–7.
31. Sieber P, Gass M, Kern B, et al. Five-year results of laparoscopic
sleeve gastrectomy. Surg Obes Relat Dis. 2014;10:243–9.
32. Soricelli E, Iossa A, Casella G, et al. Sleeve gastrectomy and crural
repair in obese patients with gastroesophageal reflux disease and/or
hiatal hernia. Surg Obes Relat Dis. 2013;9:356–61.
33. Boules M, Corcelles R, Guerron AD, et al. The incidence of hiatal
hernia and technical feasibility of repair during bariatric surgery.
Surgery. 2015;158:911–6. discussion 916-918
34. Rebecchi F, Allaix ME, Ugliono E, et al. Increased esophageal
exposure to weakly acidic reflux 5 years after laparoscopic Roux-
en-Y gastric bypass. Ann Surg. 2016;264:871–7.
OBES SURG (2017) 27:3092–3101
35. Bonavina L, DeMeester T, Fockens P, et al. Laparoscopic sphincter
augmentation device eliminates reflux symptoms and normalizes
esophageal acid exposure: one- and 2-year results of a feasibility
trial. Ann Surg. 2010;252:857–62.
36. Reynolds JL, Zehetner J, Wu P, et al. Laparoscopic magnetic
sphincter augmentation vs laparoscopic Nissen fundoplication: a
matched-pair analysis of 100 patients. J Am Coll Surg. 2015;221:
123–8.
37. Crawford C, Gibbens K, Lomelin D et al. Sleeve gastrectomy and
anti-reflux procedures. Surg Endosc. 2016.
38. Ronkainen J, Aro P, Storskrubb T, et al. Prevalence of Barrett’s
esophagus in the general population: an endoscopic study.
Gastroenterology. 2005;129:1825–31.
3101
39. Gagner M. Is sleeve gastrectomy always an absolute contraindica-
tion in patients with Barrett’s? Obes Surg. 2016;26:715–7.
40. Gorodner V, Buxhoeveden R, Clemente G, et al. Barrett’s esopha-
gus after Roux-en-Y gastric bypass: does regression occur? Surg
Endosc. 2017;31:1849–54.
41. Kindel TL, Oleynikov D. The improvement of gastroesophageal
reflux disease and Barrett’s after bariatric surgery. Obes Surg.
2016;26:718–20.
42. Overs SE, Freeman RA, Zarshenas N, et al. Food tolerance and
gastrointestinal quality of life following three bariatric procedures:
adjustable gastric banding, Roux-en-Y gastric bypass, and sleeve
gastrectomy. Obes Surg. 2012;22:536–43.