OBES SURG (2018) 28:1342–1350
https://doi.org/10.1007/s11695-017-3012-z
ORIGINAL CONTRIBUTIONS
Bariatric Surgery as an Efficient Treatment for Non-Alcoholic
Fatty Liver Disease in a Prospective Study with 1-Year Follow-up
BariScan Study
Felix Nickel 1 & Christian Tapking 1 & Laura Benner 3 & Janina Sollors 2 &
Adrian T. Billeter 1 & Hannes G. Kenngott 1 & Loay Bokhary 1 & Mathias Schmid 1 &
Moritz von Frankenberg 1 & Lars Fischer 1 & Sebastian Mueller 2 & Beat P. Müller-Stich 1
Published online: 9 November 2017
# Springer Science+Business Media, LLC, part of Springer Nature 2017
Abstract
Background Bariatric surgery gains attention as a potential
treatment for non-alcoholic fatty liver disease (NAFLD).
The present study aimed to evaluate improvement of
NAFLD after the two most common bariatric procedures with
validated non-invasive instruments.
Material and Methods N = 100 patients scheduled for laparo-
scopic sleeve gastrectomy (LSG) or Roux-en-Y gastric bypass
(RYGB) were included. NAFLD was evaluated preoperative-
ly and postoperatively with liver stiffness measurement by
transient elastography and laboratory-based fibrosis scores.
Clinical data included body mass index (BMI), total weight
loss (%TWL), excess weight loss (%EWL), age, gender, co-
morbidities, and the Edmonton obesity staging system
(EOSS).
Felix Nickel and Christian Tapking contributed equally to this work.
* Beat P. Müller-Stich
BeatPeter.Mueller@med.uni-heidelberg.de
Felix Nickel
Felix.Nickel@med.uni-heidelberg.de
Christian Tapking
Christian.Tapking@med.uni-heidelberg.de
Laura Benner
benner@imbi.uni-heidelberg.de
Janina Sollors
Janina.Sollors@stadtmission-hd.de
Adrian T. Billeter
Adrian.Billeter@med.uni-heidelberg.de
Hannes G. Kenngott
Hannes.Kenngott@med.uni-heidelberg.de
Loay Bokhary
Loay.Bokhary@stadtmission-hd.de
Results There were significant improvements of BMI,
%TWL, %EWL, and EOSS after bariatric surgery. Liver stiff-
ness was significantly improved from pre- to postoperative
(12.9 ± 10.4 vs. 7.1 ± 3.7 kPa, p < 0.001) at median follow-
up of 12.5 months. Additionally, there were significant im-
provements of liver fibrosis scores (aspartate aminotransferase
(AST)/alanine aminotransferase (ALT) ratio 0.8 ± 0.3 vs.
1.1 ± 0.4, p < 0.001; NAFLD fibrosis score − 1.0 ± 1.8 vs.
− 1.7 ± 1.3, p < 0.001; APRI score 0.3 ± 0.2 vs. 0.3 ± 0.1,
p = 0.009; BARD score 2.3 ± 1.2 vs. 2.8 ± 1.1, p = 0.008) and
laboratory parameters (ALT, AST, and GGT). After adjust-
ment for baseline liver stiffness, RYGB showed higher im-
provements than LSG, and there was no gender difference.
Improvement of liver stiffness was not correlated to improve-
ment of BMI, %TWL, %EWL, or EOSS.
Mathias Schmid
Mathias.Schmid@stadtmission-hd.de
Moritz von Frankenberg
Moritz.Von.Frankenberg@stadtmission-hd.de
Lars Fischer
l.fischer@klinikum-mittelbaden.de
Sebastian Mueller
Sebastian.Mueller@stadtmission-hd.de
1
Department of General, Visceral and Transplant Surgery, University
of Heidelberg, Im Neuenheimer Feld 110,
69120 Heidelberg, Germany
2
Department of Internal Medicine, Salem Hospital, Zeppelinstrasse
11, 69120 Heidelberg, Germany
3
Institute of Medical Biometry and Informatics, University of
Heidelberg, Im Neuenheimer Feld 130.3,
69120 Heidelberg, Germany
OBES SURG (2018) 28:1342–1350
Conclusions NAFLD seems to be improved by bariatric sur-
gery as measured by validated non-invasive instruments.
Furthermore, it appears that RYGB is more effective than
LSG. No correlation could be detected between NAFLD and
weight loss. The present study highlights the potential of bar-
iatric surgery for successful treatment of NAFLD. Further
research is required to understand the exact mechanisms.
Keywords Transient elastography . Bariatric surgery .
Metabolic surgery . Liver disease . Liver fibrosis . Gastric
bypass . Sleeve gastrectomy
BariScan Study
Introduction
Increasing prevalence of morbid obesity is a growing problem
for industrial and developing countries. The World Health
Organization in 2014 published that over 600 million adults
were obese with obesity defined as a body mass index (BMI)
≥ 30 kg/m2. It is a chronic and multifactorial disease with
immense medical, economic, and social impact [1, 2].
Morbidly obese patients often suffer from metabolic syn-
drome with related diseases such as arterial hypertension, di-
abetes, dyslipidemia, and liver diseases [3].
Non-alcoholic fatty liver disease (NAFLD) is regarded as
the hepatic manifestation of metabolic syndrome and can
progress to non-alcoholic steatohepatitis (NASH) and liver
cirrhosis [4, 5]. In the USA, NAFLD was found as the reason
for liver transplantation in 1.2% in 2001, 9.7% in 2009, and
19% in 2011, and it is now the third most frequent reason for
liver transplantation after liver cirrhosis caused by hepatitis C
and alcohol [6, 7]. However, the prevalence varies within
specialized centers. It has been shown recently that since the
introduction of a new generation of antiviral drugs, most ge-
notypes of hepatitis C can be treated successfully and there-
fore has reduced the need of liver transplantation in hepatitis C
patients significantly [8]. Some authors predict NAFLD to be
the leading cause for end-stage liver disease and to need for
liver transplantation in the future [9].Several studies have
shown that weight loss is the most effective treatment of
NAFLD [10–12]. Bariatric surgery is currently regarded as
the only treatment option that provides sustained and success-
ful long-term weight loss and improvement of related comor-
bidities for patients with morbid obesity [13]. Furthermore, it
leads to relevant improvement of health-related quality of life
[14, 15]. Several studies suggested bariatric surgery to be an
effective treatment option for NAFLD, which could potential-
ly avoid progression from NAFLD to end-stage liver disease
and necessity of liver transplantation.
1343
The aim of the present study was to evaluate the potential
of the two most common procedures in bariatric surgery to
treat NAFLD in obese patients measured by validated non-
invasive instruments for the assessment of NAFLD such as
transient elastography (TE) and laboratory-based fibrosis
scores.
Methods
Patients scheduled for laparoscopic sleeve gastrectomy (LSG)
or Roux-en-Y gastric bypass (RYGB) for morbid obesity be-
tween August 2012 and August 2015 who underwent success-
ful liver stiffness measurement by TE before surgery and
consented to participate were included in the present study.
Alcoholic, autoimmune, or viral liver diseases were the exclu-
sion criteria. Indications for bariatric surgery were in accor-
dance with the German S3 guidelines on bariatric surgery.
Clinical data included total weight loss (%TWL), excess
weight loss (%EWL), BMI, sex, age, comorbidities, and the
Edmonton obesity staging system (EOSS). Liver stiffness and
laboratory parameters to calculate fibrosis scores were
assessed before surgery and during follow-ups 1 (only labora-
tory parameters) and 12 months after surgery.
Transient Elastography
Liver stiffness (LS) was measured by TE (Fibroscan ©,
Echosens SA, Paris, France) according to the manufacturer’s
instructions prior to bariatric surgery and within 12 months
[16, 17].TE was performed on the right lobe of the liver in
intercostal position according to established protocols using
the XL probe [18, 19]. Each patient received TE by trained
and experienced physicians until ten valid measurements were
achieved. In total, ten valid measurements, a success rate
(number of measurements divided by number of total tests)
> 60%, and the interquartile range (IQR) of the last ten valid
measurements < 30% of the median of the last ten valid mea-
surement values were needed to be considered as valid mea-
surement according to the manufacturer’s instructions.
Patients with non-valid measurements were excluded from
the present study.
Liver fibrosis stages F0–F4 according to LS measurement
(kPa) were set as follows: no/little fibrosis (F0/F1) < 7.1 kPa,
significant fibrosis (F2) ≥ 7.1 and < 9.5 kPa, advanced fibrosis
(F3) ≥ 9.5 and <12.5 kPa, and cirrhosis (F4) ≥ 12.5 kPa [20].
Fibrosis Scores
Detection of NAFLD can be performed with validated non-
invasive techniques to avoid the need of invasive biopsies.
These include laboratory-based scores and imaging tech-
niques to measure and control NAFLD easily and frequently.
1344
Several serum marker scores were assessed that either reflect
NAFLD activity or fibrosis. The scores are based on parame-
ters assessed from serum samples and/or clinical parameters:
validated predictors for patients with a high risk of advanced
fibrosis are NAFLD fibrosis score [21], the De Ritis ratio
(aspartate aminotransferase (AST)/alanine aminotransferase
(ALT) ratio) [22], BARD score (includes the De Ritis ratio,
diabetes, and BMI) [21], and AST to platelet ratio index
(APRI) [23].
Standard laboratory assessments of all patients from serum
samples included platelet count, high-density lipoprotein
(HDL, mg/dl), low-density lipoprotein (LDL, mg/dl), triglyc-
erides (TG, mg/dl), albumin (g/dl), alkaline phosphatase (AP,
U/l), bilirubin (ng/dl), cholesterol (mg/dl), Quick’s method
(%), international normalized ratio (INR), aspartate amino-
transferase (AST, U/l), alanine aminotransferase (ALT, U/l)
and gamma-glutamyltransferase (GGT, U/l).The fibrosis
scores were calculated as shown in Table 1.
Statistical Analysis
Comparisons in the change of liver stiffness in different
groups were performed using the Wilcoxon tests; other char-
acteristics were compared between different time points using
the Wilcoxon signed rank tests. In addition, a linear regression
model was used to predict stiffness reduction per month.
Based on the full model including the predictor of preopera-
tive liver stiffness, time interval between surgery and follow-
up, type of surgery, sex, preoperative BMI, age, EOSS, and
diabetes, a stepwise variable selection was performed. To as-
sess the correlation between changes of liver stiffness and
change of liver fibrosis scores, the method of Spearman was
used. Due to the exploratory nature of this analysis, no adjust-
ment for multiplicity was conducted. Statistical analyses were
performed using R, version 3.2.3 [24].
Results
In total, 100 patients (62 females and 38 males) were included
in the present study with a mean age of 43.6 ± 10.8 years.
Patients general characteristics show that the two groups were
well balanced in terms of age and comorbidities, but the pa-
tients who received LSG had a higher BMI on average
(Table 2).LSG was performed in 59 patients and RYGB in
41 patients. The preoperative BMI of 48.6 ± 7.4 kg/m2 went
down to 43.9 ± 6.9 kg/m2 after 1 month (preop vs. 1 month,
p < 0.001) and to 34.1 ± 6.9 kg/m2 after 12 months (1 month
vs. 12 months, p < 0.001). Mean %TWL was 9.9 ± 4.0% after
1 month and 28.9 ± 7.7% after 12 months (1 month vs.
12 months, p < 0.001). Mean %EWL was 21.3 ± 10.5% after
1 month and 59.8 ± 19.6% after 12 months (1 month vs.
OBES SURG (2018) 28:1342–1350
12 months, p < 0.001). EOSS stage was significantly im-
proved after 12 months but not after 1 month (Fig. 1).
Transient Elastography
Mean liver stiffness measured via TE preoperatively was
12.9 ± 10.4 kPa and after a median follow-up of 12.5 months,
it significantly decreased to 7.1 ± 3.7 kPa (p < 0.001) (Fig. 2).
Forty-eight percent of the patients had liver fibrosis stage F3
or F4 before the operation and this percentage decreased to
16.5% after 12 months. The median fibrosis stage decreased
from F3 toF1.The postoperative liver stiffness improved in
94% of the patients.
Laboratory Parameters and Fibrosis Scores Before and
After Bariatric Surgery
The laboratory parameters changed as follows (Table 3):
HDL significantly decreased 1 month after surgery and
then significantly increased 12 months after surgery. TG,
AST, ALT, and GGT did not change significantly after
1 month, but decreased significantly 12 months after sur-
gery. Bilirubin (higher) and cholesterol (lower) presented
significant changes 1 month after surgery, but not after
12 months. There was no significant change of albumin
after 1 month, but after 12 months (lower).
Liver fibrosis scores changes were as follows (Table 4):
absolute values of AST/ALT ratio, NAFLD fibrosis score,
APRI score, and BARD score significantly improved after
12 months and APRI score already improved significantly
after 1 month.
After 12 months, there was a significantly higher
change in AST/ALT ratio in patients with RYGB than in
patients with LSG, but not between males and females.
The changes in the other scores were not significantly
different after 12 months concerning the type of surgery
or gender (Table 4). There were no significant correlations
between pre- and postoperative changes of liver stiffness
and changes of any liver fibrosis score (AST/ALT ratio
R = − 0.029, p = 0.825, NAFLD fibrosis score
R = − 0.003, p = 0.985, APRI score R = 0.078,
p = 0.556, BARD score R = 0.036, p = 0.783).
Furthermore, there was no significant correlation between
pre- and postoperative changes of liver stiffness and
changes of BMI (R = 0.08, p = 0.479), %TWL
(R = − 0.02, p = 0.882), %EWL (R = − 0.111,
p = 0.329), or EOSS (R = − 0.167, p = 0.14).
A univariate group analysis showed that there was no in-
fluence of a high-preoperative EOSS (≥ 2, p = 0.445), BMI
(≥ 50 kg/m2, p = 0.651), age (≥ 45 years, p = 0.445), gender
(p = 0.291), or presence of diabetes (p = 0.721) on the post-
operative changes of liver stiffness. The linear regression
model after variable selection is shown in Table 5. There
OBES SURG (2018) 28:1342–1350 1345

Table 1 Non-invasive fibrosis scores based on laboratory parameters

Score Formula Ranges

AST/ALT ratio [22] AST (U/l)/ALT (U/l) F3/4 > 1, F0–2 < 1
NAFLD fibrosis score [21] − 1.675 + 0.037 × age (years) + 0.094 × BMI F3/4 > 0.675, F0–2 < 1.455
(kg/m2) + 1.13 × IFG/diabetes (yes = 1, no = 0) Indifferent
+ 0.99 × AST/ALT ratio − 0.013 × platelet (109/l) − 1.455–0.675
− 0.66 × albumin (g/dl)
BARD score [21] AST/ALT ratio ≥ 0.8–2 points; a BMI ≥ 28–1 F3/4 ≥ 2
point; and the presence of diabetes − 1 point F0–2 ≤ 1
APRI [23] (AST (U/L)/upper normal level of AST (U/L))/ F3/4 > 1.5, F0–2 < 0.5
platelet count (109/L) × 100 Indifferent 0.5–1.5

ALT aspartate aminotransferase, ALT alanine aminotransferase, BMI body mass index, IFG impaired fasting glucose

was higher improvement of liver stiffness in patients with
higher preoperative liver stiffness and patients after RYGB
compared to LSG. Patients with a later-measured liver stiff-
ness showed a smaller liver stiffness reduction per month,
meaning that liver stiffness reduction decreases during fol-
low-up. After 12 months and with an adjusted identical base-
line liver stiffness of 9.5 kPa (median), this resulted in an
improvement of 0.591 kPa per month in patients with LSG
and 0.293 kPa per month in patients with RYGB. That meant a
0.298-kPa higher improvement with RYGB per month com-
pared to LSG.
Discussion
In the present study, there was a significant improvement of
NAFLD12.5 months after bariatric surgery as measured by
validated non-invasive instruments. The median fibrosis stage

Table 2 Patients’ general characteristics before surgery

Variable baseline Laparoscopic sleeve gastrectomy Roux-en-Y gastric bypass p value

n 61 39
Age
Mean ± SD 42.9 ± 11.6 45.1 ± 9.5 p = 0.324
Gender n (%)
Male 23 (41.0) 15 (33.3) p = 0.808
Female 36 (59.0) 26 (66.7)
BMI (kg/m2)
Mean ± SD 49.9 ± 7.6 46.6 ± 6.6 p = 0.033
Arterial hypertension
n (%) 37 (60.7) 21 (53.8) p = 0.501
T2DM
n (%) 23 (37.7) 21 (53.8) p = 0.113
Elevated HbA1c*
n (%) 22 (36.0) 17 (43.6) p = 0.452
Hypertriglyceridemia**
n (%) 17 (27.9) 6 (15.4) p = 0.148
Cholesterolemia***
n (%) 19 (31.1) 11 (28.2) p = 0.784
Low HDL****
n (%) 24 (39.3) 13 (33.3) p = 0.544

SD standard deviation, n number, % percent, BMI body mass index, T2DM type 2 diabetes mellitus, HDL high-density lipoprotein
*HbA1c > 6.0%, **triglycerides > 180 mg/gl, ***cholesterol > 200 mg/dl, ****HDL < 40 mg/dl
1346
Fig. 1 The Edmonton obesity
staging system (EOSS stage)
before and after surgery
in TE decreased from F3 to F1. In stepwise backward selec-
tion analysis, reduction of liver stiffness was higher in patients
after RYGB than after LSG. Significant improvement of liver
fibrosis scores and laboratory parameters after bariatric sur-
gery were also found, but did not correlate with improvement
of liver stiffness in TE. The patients in the present study had
significant %TWL, %EWL, and BMI reduction in the follow-
up evaluations in the usual range, which were not related to
the changes of liver stiffness or fibrosis scores (except BARD
score).
In the present study, liver stiffness in TE improved from a
preoperative mean value of 12.9 ± 10.4 kPa to a mean liver
stiffness of 7.1 ± 3.7 kPa. The median fibrosis stage decreased
from F3 toF1.The high preoperative fibrosis stages are in line
with the findings from Luger et al. who performed a liver
biopsy in bariatric surgery patients and showed a prevalence
of NASH of 72% (30%were found with advanced fibrosis or
cirrhosis), a simple steatosis in 11% and normal liver in 17%
[25].A meta-analysis of Hafeez et al. also reported a positive
effect of bariatric surgery on NAFLD. Significant histological
Fig. 2 Liver stiffness (kPa)
before and after surgery
OBES SURG (2018) 28:1342–1350
improvement or resolution of NAFLD was found in 50–80%
of patients 12–24 months after RYGB [26]. This is in line with
the present study that found 94% of patients with improve-
ment of liver stiffness after bariatric surgery in non-invasive
measurement. In a meta-analysis of Mummadi et al., postop-
erative improvement or resolution of liver steatosis, liver fi-
brosis, and steatohepatitis was shown in 91.6, 65.5, and 81.3%
of patients up to 5 years after bariatric surgery, respectively
[27]. The differences between RYGB and LSG concerning
weight loss and improvement of NAFLD are currently under
investigation in a multicenter randomized, blinded trial [28].
Algooneh et al. found a higher resolution rate of NAFLD in
female patients in a study with an average follow-up time of
3.3 years [29], which was not found in the present study. It
was also shown that 56% of patients had a complete resolution
of NAFLD and patients with %EWL > 70% were found to
have the highest improvement of NAFLD in the study by
Algoneeh et al. [29]. In contrast, improvement of liver stiff-
ness was not related with improvements of preoperative BMI,
%TWL, %EWL, or EOSS in the present study. Naveau et al.
OBES SURG (2018) 28:1342–1350 1347

Table 3 Laboratory parameter changes

Parameter Preop mean ± SD 1 month postop Mean ± SD 12 months postop p value p value p value
Mean ± SD preop vs. 1 month vs. preop vs.
1 month 12 months 12 months

HDL 44.1 ± 11.7 38.9 ± 8. 54.6 ± 14.8 < 0.001 < 0.001 < 0.001
LDL 106.8 ± 34.9 106.4 ± 46.9 104.0 ± 30.8 0.456 0.782 0.691
TG 154.5 ± 61.2 139.6 ± 48.7 107.5 ± 54.0 0.083 < 0.001 < 0.001
AST 31.5 ± 17.3 35.3 ± 21.0 21.4 ± 7.2 0.159 < 0.001 < 0.001
ALT 42.3 ± 28.9 50.3 ± 38.4 22.1 ± 11.6 0.127 < 0.001 < 0.001
GGT 56.1 ± 66.5 48.1 ± 50.1 24.3 ± 27.7 0.643 < 0.001 < 0.001
AP 86.4 ± 27.9 79.9 ± 24.4 82.2 ± 24.9 0.065 0.541 0.302
Bilirubin 0.5 ± 0.2 0.6 ± 0.3 0.6 ± 0.3 0.027 0.476 0.221
Albumin 4.4 ± 0.3 4.4 ± 0.3 4.3 ± 0.2 0.237 0.173 0.017
Cholesterol 181.6 ± 37.5 169.4 ± 33.1 175.4 ± 43.6 0.023 0.218 0.514

HDL high-density lipoprotein, LDL low-density lipoprotein, TG triglycerides, AST aspartate aminotransferase, ALT alanine aminotransferase, GGT
gamma-glutamyl aminotransferase, AP alkaline phosphatase

also found that changes in liver stiffness were not related to
changes in BMI and %EWL [30]. Similar results were shown
by Müller-Stich et al. concerning the improvement of neurop-
athy after metabolic surgery, which did not correlate with
weight loss, but improved earlier. This indicates that the
amount of weight loss might not be the most important factor
for improvement of NAFLD, but rather the metabolic changes
that result from the operation, as was already shown for im-
provement of diabetes mellitus type 2 after metabolic surgery
[31, 32].Weight reduction is currently the most important rec-
ommendation for improvement of NAFLD and NASH by
major societies such as the American Gastroenterological
Association. Since bariatric surgery is the most effective treat-
ment for weight loss, NAFLD needs to be considered an in-
dication for bariatric surgery or metabolic surgery [33]. In
addition, there seem to be effects that are independent from
pure weight loss and their mechanisms are subject to further
research including effects of different types of operation on
changes of NAFLD.
The stepwise backward selection with adjusted liver stiff-
ness and time between surgery and measurement revealed that
patients with a high-preoperative liver stiffness had a higher

Table 4 Comparison of pre-and postoperative liver fibrosis scores; differences between points of time are compared between subgroups

Liver fibrosis score Preop mean ± SD 1 month postop mean ± SD 12 months postop p value p value p value
mean ± SD preop 1 month prop vs.
vs. vs. 12 months
1 month 12 months

AST/ALT ratio 0.83 ± 0.28 0.81 ± 0.34 1.09 ± 0.39 0.306 < 0.001 < 0.001
Male vs. female 0.81 ± 0.32 vs. 0.74 ± 0.30 0.87 ± 0.40 vs. 0.80 ± 0.38 1.04 ± 0.32 vs. 0.95 ± 0.25 0.028 0.676 0.724
RYGB vs. LSG 0.79 ± 0.25 vs. 0.83 ± 0.23 0.81 ± 0.29 vs. 0.75 ± 0.23 0.97 ± 0.33 vs. 0.90 ± 0.27 0.031 0.027 0.019
NAFLD fibrosis − 1.03 ± 1.51 − 1.20 ± 1.41 − 1.71 ± 1.30 0.401 0.006 < 0.001
score
Male vs. female − 0.81 ± 1.33 vs. − 1.13 ± 1.31 vs. − 1.86 ± 1.13 vs. 0.166 0.868 0.596
− 0.87 ± 1.50 − 1.22 ± 1.50 − 2.12 ± 1.26
RYGB vs. LSG − 1.2 ± 1.81 vs. −1.21 ± 1.72 vs. −1.98 ± 1.35 vs. 0.017 0.620 0.124
− 1.44 ± 2.29 − 1.59 ± 2.03 − 2.09 ± 1.49
APRI 0.25 ± 0.19 0.31 ± 0.21 0.19 ± 0.08 0.010 < 0.001 0.009
Male vs. female 0.29 ± 0.17 vs. 0.3 ± 0.20 0.29 ± 0.14 vs. 0.27 ± 0.14 0.19 ± 0.08 vs. 0.16 ± 0.05 0.018 0.459 0.233
RYGB vs. LSG 0.27 ± 0.27 vs. 0.25 ± 0.21 0.32 ± 0.28 vs. 0.29 ± 0.20 0.21 ± 0.11 vs. 0.21 ± 0.12 0.991 0.673 0.894
BARD score 2.31 ± 1.19 2.29 ± 1.21 2.83 ± 1.12 0.849 0.015 0.008
Male vs. female 2.11 ± 1.03 vs.1.92 ± 1.02 2.31 ± 1.23 vs. 2.24 ± 1.3 2.57 ± 1.08 vs. 2.50 ± 0.85 0.094 0.784 0.630
RYGB vs. LSG 2.17 ± 1.07 vs. 2.43 ± 1.12 2.51 ± 1.24 vs. 2.21 ± 1.13 2.5 ± 1.39 vs. 2.50 ± 1.51 0.005 0.444 0.091

RYGB Roux-en-Y gastric bypass, LSG laparoscopic sleeve gastrectomy

1348
Table 5 Parameters influencing postoperative liver stiffness reduction
Estimate Std. p value
Error
(Intercept) 0.184 0.137 0.182
Preoperative liver stiffness (kPa) 0.082 0.005 < 0.001
Time interval between surgery and − 0.031 0.006 < 0.001
follow-up (months)
Type of surgery (sleeve gastrectomy) − 0.298 0.112 0.009
Estimate shows the difference in reduction of liver stiffness per month
between sleeve gastrectomy and gastric bypass; with a 1-kPa higher pre-
operative liver stiffness, patients showed a 0.082-kPa higher reduction per
month; the reduction was highest in the beginning after surgery and be-
came 0.031 kPa less with every extra month after surgery
liver stiffness reduction after surgery. Furthermore, reduction
of liver stiffness was higher for RYGB than LSG, whereas
age, BMI, EOSS, present diabetes, and sex showed no influ-
ence. Similar to the present study, Aguilar-Olivos et al. found
that effects of bariatric surgery on NAFLD were higher after
RYGB than after other interventions (LSG, gastric banding,
biliopancreatic diversion) [34].Taitano and colleagues report-
ed a resolution of steatosis in 75% of the patients with %EWL
of 62% after RYGB [35]. However, Algooneh et al. also found
positive effects on NAFLD for LSG, with a higher number of
NAFLD resolutions in female patients [29]. As shown by
Yoneda et al., NAFLD leads to a stepwise increase of liver
fibrosis, which can be measured reliably with TE [36].In a
controlled matched pair study by Billeter et al., LSG was
reported to improve NAFLD better than RYGB in a highly
selected group with metabolic syndrome. However, RYGB
significantly improved the observed parameters as well
[37].Differences in study designs and patient selection criteria
can explain the partly differing results, as the previous study
by Billeter et al. was limited to patients with preoperative
diabetes mellitus type 2, and the present study included pa-
tients with and without diabetes mellitus type 2. In the present
study, the comparison between LSG and RYGB was adjusted
for baseline liver stiffness. However, as a limitation, it must be
stated that in the present study, the patients who received LSG
had a higher preoperative BMI on average. Thus, prospective
randomized trials are needed to evaluate differences between
LSG and RYGB for improvement of NAFLD. Although TE
has been validated, possible discrepancies between TE and
biopsy results have to be taken into consideration and should
be investigated in randomized controlled trials.
High levels of serum transaminases can be associated with
obesity and with progression of chronic liver disease [38–40].
In the present study, significant increase of AST and ALT
levels after 1 month was found, and a significant decrease after
12 months was found in comparison to preoperative values.
Burza et al. also showed a decrease of liver transaminases in 2-
and 10-year follow-up evaluations. An increase of AST and
OBES SURG (2018) 28:1342–1350
ALT levels shortly after bariatric surgery was shown in the
present study, but has been rarely reported in the literature
before [41]. The reason for that may be fast metabolic changes
due to surgery and slower adaptation of the metabolism of the
liver.
Liver fibrosis scores improved after surgery in the present
study. Nascimento et al. also found improved NAFLD fibrosis
score 12 months after bariatric surgery [42]. Several studies
have shown a good validity to detect liver fibrosis for APRI
score, NAFLD fibrosis score, and FIB-4 score [21, 23, 43,
44].However, in the present study, improvement of fibrosis
scores was not statistically related to improvement of liver
stiffness in TE. Improvement of NAFLD fibrosis score was
related to improvement of BMI and improvement of BARD
score was correlated to improvement of EOSS, %TWL, and
%EWL. Later findings are not surprising since BMI forms
part of the calculation of these scores. Improvement of AST/
ALT ratio was significantly higher in patients with RYGB.
Changes of the other scores were independent from the type
of surgery and gender.
The present study did not use liver biopsy as the gold stan-
dard to evaluate patients with chronic liver disease. TE can
produce a relatively high failure rate of 4.5–6% according to
the literature and can barely be performed in patients with
ascites [45, 46]. Compared to liver biopsy, TE is totally non-
invasive, fast, and can easily be reproduced [47]. Additionally,
TE samples a much larger part of the liver than a single biopsy
(approximately by factor 100) [48]. There are already studies
comparing liver biopsy and non-invasive instruments in
NAFLD patients [30, 49–51]. Further studies including liver
biopsy samples before surgery and in follow-up evaluations
are required to show correlations between liver stiffness and
the histological classification after bariatric surgery. However,
the non-invasive methods used in the present study were val-
idated and in combination can be suggested as valid tools.
Furthermore, NAFLD can be diagnosed by either histological
or imaging findings according to the American Association
for the Study of Liver Diseases [33].
Conclusion
Bariatric surgery is known to be effective to reduce weight in
morbidly obese patients and to improve associated comorbid-
ities. The present study shows that NAFLD can also be im-
proved after bariatric surgery as measured by TE and liver
fibrosis scores. In the present study, RYGB seemed to be more
effective than LSG in reduction of liver stiffness only when
adjusted for baseline values. However, differences between
RYGB and LSG should be confirmed in randomized trials.
As improvement of liver stiffness did not correlate with im-
provement of BMI, %TWL, EWL%, and EOSS, underlying
mechanisms do not seem to be related to weight loss alone and
OBES SURG (2018) 28:1342–1350
are subject to further research. Since NAFLD will be a leading
cause for necessity of liver transplantation in the future, bar-
iatric surgery should be considered an important option of
treatment for NAFLD.
Acknowledgements We would like to thank Carly R. Garrow for
proofreading the manuscript.
Compliance with Ethical Standards
Conflict of Interest The authors declare that they have no conflict of
interest.
Statement of Informed Consent Informed consent was obtained from
all individual participants included in the present study.
Ethical Approval All procedures performed in studies involving hu-
man participants were in accordance with the ethical standards of the
institutional and/or national research committee and with the 1964
Helsinki declaration and its later amendments or comparable ethical stan-
dards. Ethical approval was obtained from the local ethics committee
(S-181/2009, S-618/2011, S-500/2012 and S-629/2013).
References
1. Ng M. Global, regional, and national prevalence of overweight and
obesity in children and adults during 1980–2013: a systematic anal-
ysis for the Global Burden of Disease Study 2013. Lancet.
2014;386(9945):766–81.
2. WHO. Fact Sheet. 2015 [cited 2016 17.08.2016]; Available from:
http://www.who.int/mediacentre/factsheets/fs311/en/.
3. Huang Z et al. Body weight, weight change, and risk for hyperten-
sion in women. Ann Intern Med. 1998;128(2):81–8.
4. Sanyal AJ. A technical review on nonalcoholic fatty liver disease.
Gastroenterology. 2002;123(5):1705–25.
5. Day CP. Non-alcoholic fatty liver disease: current concepts and
management strategies. Clin Med. 2006;6(1):19–25.
6. Charlton MR et al. Frequency of outcomes of liver transplantation
for nonalcoholic steatohepatitis in the United States.
Gastroenterology. 2011;141(4):1249–53.
7. Agopian VG et al. Liver transplantation for nonalcoholic
steatohepatitis: the new epidemic. Ann Surg. 2012;256(4):624–33.
8. Belli LS et al. Impact of direct anti-viral agents on inactivation/
delisting of liver transplant candidates listed for decompensated C
cirrhosis: a European study. In: International Liver Congress (ILC):
abstract PS036; 2016.
9. Angulo P. Nonalcoholic fatty liver disease. N Engl J Med.
2002;346(16):1221–31.
10. Palmer M, Schaffner F. Effect of weight reduction on hepatic ab-
normalities in overweight patients. Gastroenterology. 1990;99(5):
1408–13.
11. Park HS, Kim MW, Shin ES. Effect of weight control on hepatic
abnormalities in obese patients with fatty liver. Kor Med Sci.
1995;10(6):414–21.
12. Ueno T, Sugawara H, Sujaku K. Therapeutic effects of restricted
diet and exercise in obese patients with fatty liver disease. J
Hepatol. 1997;27(1):103–7.
13. Sjöström L et al. Association of bariatric surgery with long-term
remission of type 2 diabetes and with microvascular and
macrovascular complications. JAMA. 2014;311(22):2297–304.
1349
14. Nickel F et al. Gastrointestinal quality of life improves significantly
after sleeve gastrectomy and Roux-en-Y gastric bypass-a prospec-
tive cross-sectional study within a 2-year follow-up. Obes Surg.
2017;27(5):1292–7.
15. Nickel F et al. Influence of bariatric surgery on quality of life, body
image, and general self-efficacy within 6 and 24 months-a prospec-
tive cohort study. Surg Obes Relat Dis. 2017;12(2):313–9.
16. Echosens. FibroScan® user manual. Paris: Frankreich; 2015.
17. Sandrin L et al. Transient elastography: a new noninvasive method
for assessment of hepatic fibrosis. Ultrasound Med Biol.
2003;29(12):1705–13.
18. Mueller S, Seitz HK, Rausch V. Non-invasive diagnosis of alcohol-
ic liver disease. World J Gastroenterol. 2014;20(40):14626–41.
19. Boursier K, Konaté A, Gorea G. Reproducibility of liver stiffness
measurement by ultrasonographic elastometry. Clin Gastroenterol
Hepatol. 2008;6(11):1263–9.
20. Castéra L et al. Prospective comparison of transient elastography,
Fibrotest, APRI, and liver biopsy for the assessment of fibrosis in
chronic hepatitis C. Gastroenterology. 2005;128(2):343–50.
21. Cichoz-Lach H. The BARD score and the NAFLD fibrosis score in
the assessment of advanced liver fibrosis in nonalcoholic fatty liver
disease. Med Sci Monit. 2012;18(12):735–40.
22. Botros M, Sikaris KA. The De Ritis ratio: the test of time. Clin
Biochem Rev. 2013;34(3):117–30.
23. Wai CT. A simple noninvasive index can predict both significant
fibrosis and cirrhosis in patients with chronic hepatitis C.
Hepatology. 2003;38(2):518–26.
24. R Core Team, A language and environment for statistical comput-
ing, R.C. Team, Editor 2015, R Foundation for Statistical
Computing, Vienna, Austria.
25. Luger M et al. Prevalence of liver fibrosis and its association with
non-invasive fibrosis and metabolic markers in morbidly obese pa-
tients with vitamin D deficiency. Obes Surg. 2016;
26. Hafeez S, Ahmed MH. Bariatric surgery as potential treatment for
nonalcoholic fatty liver disease: a future treatment by choice or by
chance? J Obes. 2013;
27. Mummadi RR et al. Effects of bariatric surgery on nonalcoholic
fatty liver disease: systematic review and meta-analysis. Clin
Gastroenterol Hepatol. 2008;6(12):1396–402.
28. Fischer L et al. BariSurg trial: sleeve gastrectomy versus Roux-en-
Y gastric bypass in obese patients with BMI 35–60 kg/m2—a
multi-centre randomized patient and observer blind non-inferiority
trial. BMC Surg. 2015;15(1):87.
29. Algooneh A et al. Non-alcoholic fatty liver disease resolution fol-
lowing sleeve gastrectomy. Surg Endosc. 2016;30(5):1983–7.
30. Naveau S et al. The diagnostic accuracy of transient elastography
for the diagnosis of liver fibrosis in bariatric surgery candidates with
suspected NAFLD. Obes Surg. 2014;24(10):1693–701.
31. Müller-Stich BP et al. Gastric bypass leads to improvement of dia-
betic neuropathy independent of glucose normalization—results of
a prospective cohort study (DiaSurg 1 study). Ann Surg.
2013;258(5):760–5.
32. Kenngott HG et al. DiaSurg 2 trial—surgical vs. medical treatment
of insulin-dependent type 2 diabetes mellitus in patients with a body
mass index between 26 and 35 kg/m2: study protocol of a random-
ized controlled multicenter trial—DRKS00004550. Trials.
2013;14(183)
33. Chalasani N et al. The diagnosis and management of nonalcoholic
fatty liver disease: practice guideline by the American Association
for the Study of Liver Diseases, American College of
Gastroenterology, and American Gastroenterological Association.
Hepatology. 2012;55(6):2005–23.
34. Aguilar-Olivos NE et al. The role of bariatric surgery in the man-
agement of nonalcoholic fatty liver disease and metabolic syn-
drome. Metabolism. 2016;65(8):1196–207.
1350
35. Taitano A, Markow M, Finan J. Bariatric surgery improves histo-
logical features of nonalcoholic fatty liver disease and liver fibrosis.
J Gastrointestinal Surg. 2015;19(3):429–37.
36. Yoneda M et al. Transient elastography in patients with non-
alcoholic fatty liver disease (NAFLD). Gut. 2007;56(9):1330–1.
37. Billeter AT et al. Combined non-alcoholic fatty liver disease and
type 2 diabetes mellitus: sleeve gastrectomy or gastric bypass?—a
controlled matched pair study of 34 patients. Obes Surg.
2016;26(8):1867–74.
38. Clark JM, Brancati FL, Diehl AM. The prevalence and etiology of
elevated aminotransferase levels in the United States. Am J
Gastroenterol. 2003;98(5):960–7.
39. Torgerson JS et al. Are elevated aminotransferases and decreased
bilirubin additional characteristics of the metabolic syndrome?
Obes Res. 1997;5(2):105–14.
40. Ekstedt M et al. Long-term follow-up of patients with NAFLD and
elevated liver enzymes. Hepatology. 2006;44(865–873)
41. Burza MA et al. Long-term effect of bariatric surgery on liver en-
zymes in the Swedish Obese Subjects (SOS) study. PLoS One.
2013;8(3)
42. Nascimento TM et al. Comparison of hepatic profile in pre and
postoperative of bariatric surgery: private vs public network. Arg
Bras Cir Dig. 2015;28(4):274–7.
43. Giannini EG et al. A simple approach to noninvasively identifying
significant fibrosis in chronic hepatitis C patients in clinical prac-
tice. J Clin Gastroenterol. 2006;40(6):521–7.
OBES SURG (2018) 28:1342–1350
44. Vallet-Pichard A et al. FIB-4: an inexpensive and accurate marker
of fibrosis in HCV infection. Comparison with liver biopsy and
fibrotest. Hepatology. 2007;46(1):32–6.
45. Jeong WK et al. Principles and clinical application of ultrasound
elastography for diffuse liver disease. Ultrasonography. 2014;33(3):
149–60.
46. Foucher J et al. Prevalence of factors associated with failure liver
stiffness measurement using Fibroscan in a prospective study of
2114 examinations. Eur J Gastroenterol Hepatol. 2006;18(4):411–
2.
47. Sandrin L, Fourquet B, Hasquenoph JM. Transient elastography: a
new noninvasive method for assessment of hepatic fibrosis.
Ultrasound Med Biol. 2003;29(12):1705–13.
48. Scott DR, Levy MT. Liver transient elastography (Fibroscan): a
place in the management algorithms of chronic viral hepatitis.
Antivir Ther. 2010;15(1):1–11.
49. Wong GL et al. Assessment of fibrosis by transient elastography
compared with liver biopsy and morphometry in chronic liver dis-
eases. Clin Gastroenterol Hepatol. 2008;6(9):1027–35.
50. Desai NK et al. Comparison of controlled attenuation parameter
and liver biopsy to assess hepatic steatosis in pediatric patients. J
Pediatr. 2016;173:160–3.
51. Pfeifer L et al. Prospective evaluation of acoustic radiation force
impulse (ARFI) elastography and high-frequency B-mode ultra-
sound in compensated patients for the diagnosis of liver fibrosis/
cirrhosis in comparison to mini-laparoscopic biopsy. Ultraschall
Med. 2015;36(6):581–9.