nature publishing group ORIGINAL CONTRIBUTIONS 531

see related editorial on page x

Randomized Controlled Trial of Transoral Incisionless

ESOPHAGUS
Fundoplication Vs. Proton Pump Inhibitors for
Treatment of Gastroesophageal Reflux Disease
Bart P.L. Witteman, MD, PhD1,2, Jose M. Conchillo, MD, PhD3, Nicolaas F. Rinsma, MD3, Bark Betzel, MD2, Andrea Peeters, MD, PhD4,
Ger H. Koek, MD, PhD3, Laurents P.S. Stassen, MD, PhD1 and Nicole D. Bouvy, MD, PhD1

OBJECTIVES: Transoral incisionless fundoplication (TIF) was developed in an attempt to create a minimally
invasive endoscopic procedure that mimics antireflux surgery. The objective of this trial was to
evaluate effectiveness of TIF compared with proton pump inhibition in a population consisting of
gastroesophageal reflux disease (GERD) patients controlled with proton pump inhibitors (PPIs) who
opted for an endoscopic intervention over lifelong drug dependence.

METHODS: Patients with chronic GERD were randomized (2:1) for TIF or continuation of PPI therapy. American
Society of Anesthesiologists >2, body mass index >35 kg/m2, hiatal hernia >2 cm, and esophageal
motility disorders were exclusion criteria. Primary outcome measure was GERD-related quality of life.
Secondary outcome measures were esophageal acid exposure, number of reflux episodes, PPI usage,
appearance of the gastroesophageal valve, and healing of reflux esophagitis. Crossover for the PPI
group was allowed after 6 months.

RESULTS: A total of 60 patients (TIF n=40, PPI n=20, mean body mass index 26 kg/m2, 37 male) were
included. At 6 months, GERD symptoms were more improved in the TIF group compared with the
PPI group (P<0.001), with a similar improvement of distal esophageal acid exposure (P=0.228)
compared with baseline. The pH normalization for TIF group and PPI group was 50% and 63%,
respectively. All patients allocated for PPI treatment opted for crossover. At 12 months, quality
of life remained improved after TIF compared with baseline (P<0.05), but no improvement in
esophageal acid exposure compared with baseline was found (P=0.171) and normalization of pH
was accomplished in only 29% in conjunction with deteriorated valve appearances at endoscopy and
resumption of PPIs in 61%.

CONCLUSION: Although TIF resulted in an improved GERD-related quality of life and produced a short-term
improvement of the antireflux barrier in a selected group of GERD patients, no long-term objective
reflux control was achieved.
Am J Gastroenterol 2015; 110:531–542; doi:10.1038/ajg.2015.28; published online 31 March 2015

INTRODUCTION include esophagitis, stricture formation, Barrett’s esophagus,

Gastroesophageal reflux disease (GERD) was defined in the and esophageal cancer. The so-called atypical symptoms are
Montreal consensus as a condition that develops when the reflux related to extraesophageal complications of the disease, such as
of stomach contents causes troublesome symptoms and/or com- laryngitis, cough, asthma, and dental erosions. The need for a
plications (1). Typical symptoms include heartburn, regurgi- long-term treatment of GERD has become increasingly apparent
tation, belching, and dysphagia. Esophageal complications of in the past two decades as a result of a growing incidence and
GERD are associated with mucosal injury of the esophagus and prevalence of this chronic disease. The likelihood of developing

1
Department of Surgery, Maastricht University Medical Centre (MUMC), Maastricht, The Netherlands; 2Department of Surgery, Rijnstate Hospital, Arnhem,
The Netherlands; 3Department of Gastroenterology and Hepatology, Maastricht University Medical Centre (MUMC), Maastricht, The Netherlands; 4Department of
Clinical Epidemiolgy and Medical Technology Assessment, Maastricht University Medical Centre (MUMC), Maastricht, The Netherlands. Correspondence:
Bart P.L. Witteman, MD, PhD, Department of Surgery, Maastricht University Medical Centre (MUMC), PO Box 5800, Maastricht 6202 AZ, The Netherlands.
E-mail: [email protected]
Received 22 March 2014; accepted 23 December 2014

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 532 Witteman et al.

GERD increases with the severity of the anatomical disorder and not achievable with TIF, the procedure would only be suitable for
dysfunction of the gastroesophageal junction (GEJ) that repre- few selected surgical candidates. We hypothesized that the proce-
sents the primary defense against reflux of gastric content into dure would be more beneficial to patients who are well controlled
the esophagus (2–7). Although proton pump inhibitors (PPIs) with medication, but opt for a more definitive GERD therapy. Par-
ESOPHAGUS

are effective in controlling symptoms, the disease is not cured ticularly patients who have elements of volume reflux and regur-
and patients will often have a lifelong commitment to drug gitation in their symptom complex may benefit from endoscopic
therapy (8–13). Studies have shown that up to 50% of patients fundoplication. If proven effective, endoscopic fundoplication
treated with PPIs continue to have abnormal acid reflux and up would be an alternative to PPI treatment and might also be cost
to 35% experience a relapse of symptoms after a 3-year follow- effective if PPI cessation can be maintained.
up, requiring titration with increasing doses of PPIs and later The primary objective of this trial was to evaluate the effective-
referral for surgery (14,15). ness of TIF procedure in treating GERD patients who were well
The merits of early surgery in GERD patients have been dem- controlled on daily PPI therapy. The study tested the hypoth-
onstrated in several randomized controlled trials (16–19). Antire- esis that TIF would be as effective as PPI treatment in controlling
flux surgery was shown to be more effective than pharmacological GERD symptoms, esophageal acid exposure, and esophagitis, but
antisecretory therapies in terms of patient satisfaction, clinical out- would be more effective in controlling nonacidic reflux and thereby
come, and healthcare cost reduction in the long-term management improve the quality of life. Secondary objectives were the evalua-
of GERD (20–23). Hence, operative restoration of GEJ competence tion of effectiveness of the TIF procedure in restoring the anatomy,
at the anatomic, mechanic, and physiologic levels seems critical in biomechanics, geometry, and physiology of the antireflux barrier.
GERD treatment, but this has not yet been established without
side effects and/or complications. Disadvantages of surgery are the
invasiveness and fundoplication-related side effects that frequently METHODS
compromise the otherwise excellent postsurgical results (22–24). Study design
Persistent dysphagia, inability to belch and vomit, as well as The trial was designed as a two-center study, performed at two
gas bloating and flatulence are common and difficult to treat University Hospitals in Europe and the United States. Patients were
(20,25–28). To minimize the risk of chronic complications, a 2:1 randomized to either the TIF (intervention group) or continu-
variety of factors need to be taken into account, including techni- ation on PPI therapy (control group). A prospective power analy-
cal aspects of the operation (27,29). Restoration of the anatomy, sis, conducted with GERD– health-related quality-of-life (HRQL)
mechanics, and physiology of the GEJ, with “tailoring” of the valve scores from the phase 2 study with TIF (33), calculated a sample
to the individual patient needs, seems crucial for an adequate size of n=120, including an enrollment loss of 5 and 80% power
repair that should allow air to be vented from the stomach, without to identify a significant difference between groups (Appendix 1).
jeopardizing the efficacy of the antireflux barrier (30). The objective was to demonstrate that TIF is not significantly worse
The field of endoscopic fundoplication has gained popularity than control (equivalence). “Significantly worse” was defined by the
over the past few years, and in an attempt to create a less invasive “equivalence Δ.” With the equivalence Δ of 2, n=120 was adequate
procedure that mimics antireflux surgery with fewer side effects, for demonstrating equivalence if the true TIF mean is better than
transoral incisionless fundoplication (TIF) was developed (31). the control by 4 points. The randomization was conducted by an
The technique was designed to construct a partial fundoplication independent statistician using randomization software (SAS Insti-
at the GEJ, restoring the angle of His, and reducing small hiatal tute, Cary, NC). Allocation assignments were deposited in consec-
hernia, by means of natural orifice surgery, using the EsophyX utively numbered, opaque, sealed envelopes. The study coordinator
device. The device was cleared by the US Food and Drug Adminis- opened envelopes at the time of inclusion after signed informed
tration for treatment of symptomatic chronic GERD in 2007, after consent was obtained. No effort to blind patients or study staff to
it was demonstrated to be safe and effective in a company-spon- treatment assignment was made after randomization. Date of TIF
sored multicenter study (32,33). procedure was considered the enrollment day for the TIF group and
Several initial reports on TIF were encouraging and showed randomization day was the enrollment day for the PPI group. After
effectiveness in reducing typical and atypical GERD symptoms, the 6-month follow-up assessment was completed, patients in the
eliminating daily PPI dependence, as well as in normalizing distal PPI group were allowed to undergo TIF procedure. The crossover
esophageal pH. On the contrary, other early experiences reported possibility was stated upfront in the informed consent and was exe-
substantial failures rates (34–37). The available data on the TIF cuted on patients’ demand, unrelated to outcomes at 6 months of
technique are of limited quality and extracted from uncontrolled follow-up. All patients who had undergone TIF were then followed
studies with generally a short follow-up duration. as a cohort. The research protocol was approved by the Maastricht
All previous studies included patients who were referred for Medical Ethics Committee (METC 07-2-051, NL17303.068.07)
surgical management as they were refractory to PPI treatment and was registered NCT0085759 (clinicaltrials.gov).
and would otherwise probably have undergone conventional
laparoscopic antireflux surgery. Because important components of Study population
antireflux surgery, such as extragastric mobilization of the stomach The study population consisted of GERD patients, controlled
and esophagus, allowing hernia reduction and crural closure, are with PPI therapy, but who opted for an intervention over lifelong

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

 TIF Vs. PPIs for GERD 533

drug dependence. Patients were referred to gastroenterologists for practice (16). The usage of all GERD medications was recorded in
GERD analysis and were selected to participate in the trial when a medication diary.
the study criteria were met. The inclusion criteria were 18–75

ESOPHAGUS
years of age, hiatal hernia ≤2 cm, proven reflux while off PPIs, on TIF (intervention group). Under general anesthesia, patients were
daily PPIs for ≥1 year, recurrence of GERD symptoms after ces- positioned in left lateral decubitus position and underwent endo-
sation of PPIs (GERD-HRQL score difference of >10 between on scopic examination of the esophagus and stomach. Subsequently,
and off PPIs), normal or reduced lower esophageal sphincter rest- TIF2 procedure was performed by a team of two experienced
ing pressure (5–40 mm Hg) at manometry, and signed informed physicians (surgeon and gastroenterologist) under continuous
consent. Patients with body mass index ≥35 kg/m2, hiatal hernia endoscopic visualization using the EsophyX-2 device with Sero-
>2 cm, esophagitis grade D, Barrett’s esophagus, esophageal stric- saFuse fasteners (EndoGastric Solutions, Redmond, WA). Using
ture or ulcer, motility disorders, previously splenectomy, gastro- a bite block, the lubricated EsophyX device was inserted over the
paresis, pregnancy (in females), immunosuppression, American endoscope, used as a guide wire. With the use of vacuum suction
Society of Anesthesiologists >2, portal hypertension, or a previous holes in the device, small hernias were reduced by advancing the
antireflux procedure were excluded from participation. Demo- device caudally, with the device secured to the esophageal wall.
graphics, the duration of GERD, frequency, and dose of PPI usage According to the standard TIF2 protocol, with the use of the
at the time of inclusion were recorded. in-built tissue mold and helical retractor, gastroesophageal valves
Baseline measurements included GERD symptoms, assessed were constructed by sequential full-thickness esophagogastric
with the use of a validated questionnaire (GERD-HRQL) after plications, secured by placement of multiple H-shaped polypro-
a cessation period of 14 days (Off-PPI) (Appendix 2) (38). The pylene fasteners, circumferentially around the gastroesophageal
questionnaire was developed to assess typical and atypical GERD junction (Figures 1 and 2) (49,50). After TIF2 procedure, PPI
symptoms and general satisfaction scores. The presence of path- usage was continued for 14 days to allow healing of the mucosa
ologic esophageal acid exposure was confirmed by 24-h pH- around the fasteners and then stopped. In the event of the re-
impedance monitoring while patients were “Off-PPIs” (Ohmega, currence of symptoms, the PPI management scheme was also
Medical Measurement Systems, Enschede, The Netherlands). employed for the TIF group and PPI usage was recorded in a med-
Normal pH was considered pH <4 for <4.2% of a 24-h monitor- ication diary. Revisional standard laparoscopic fundoplication was
ing period. Other baseline evaluations included high-resolution offered in case of treatment failure based on relapse of symptoms,
manometry (Unisensor AG, Attikon, Switzerland) and upper unresponsive to PPI treatment, in conjunction with anatomical
gastrointestinal endoscopy (assessment “On-PPIs”) (39–46). By failure at endoscopy and/or pathologic pH measurements.
endoscopy, hiatal hernia size, esophagitis (Los Angeles classifica-
tion scale), and geometrical aspects (valve length, circumference, Follow-up assessment
adherence, and Hill grade) were assessed (47). The Hill grading The follow-up assessment was conducted after 1 week, and 1, 3,
system of the gastroesophageal valve is predictive of the presence 6, and 12 months following inclusion to record PPI usage and
of reflux (grade I valve: prominent fold of tissue along the lesser adverse events. At 6-month follow-up, all patients underwent
curvature closely apposed to endoscope; grade II valve: promi- identical assessment tests as performed at baseline. Patients in the
nent fold, apposition to endoscope less consistent; grade III PPI group underwent all tests while using PPIs and the TIF group
valve: nonprominent fold without tight apposition to the endo- (in the event PPIs were resumed) discontinued any PPI usage for
scope; grade IV: absent fold and gaping lumen with visualization 14 days before the 6-month follow-up assessment. At 12 months
of the squamous epithelium) (48). after TIF2, the follow-up assessment was repeated for the entire
group, including eventual crossovers from the PPI group.
End points
Primary outcome measures were GERD symptoms. Treatment Safety
success was defined by a significantly (P<0.05) higher number The incidence of serious and nonserious adverse events was
of patients with ≥50% improvement in scores in the TIF group recorded. Serious adverse events were defined as complications
vs. PPI group at 6 months. Secondary outcome measures were necessitating prolonged hospitalization and additional medical
adverse events, esophageal acid exposure, number of reflux epi- or surgical intervention. Nonserious adverse events represented
sodes, PPI usage, appearance of the gastroesophageal valve, and expected side effects and symptoms. Gastrointestinal side effects,
healing of reflux esophagitis. such as epigastric pain, flatulence, bloating, diarrhea, dysphagia,
and the ability to vomit and belch, were assessed using Gastro-
Trial interventions intestinal Symptom Rating Scale questionnaire (51).
PPI (control group). Patients randomized to the PPI group were
treated according to a PPI management scheme (Appendix 3). If Data handling and record keeping
GERD symptoms were well controlled on their previous dose of Case report forms were completed for each subject enrolled into
PPIs for at least a month, the regimen was reduced a step and if the trial and stored in a binder in a secured room. Case report
poorly controlled, increased by one step according to the PPI algo- forms also contained the signed informed consent and source
rithm that provides a treatment plan consistent with good clinical documents of clinical data and patients were identified by their

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 534 Witteman et al.

with normal distribution (tested by Shapiro–Wilk statistic) were

univariately presented by their means and s.d. values and if not
normally distributed the mean and range were used. Categorical
variables were presented by frequencies and percentages. The pri-
ESOPHAGUS

mary outcome parameter was calculated as a ≥50% difference in

the GERD questionnaire scores between baseline and at 6 months
postoperatively. Differences between randomized TIF and PPI
groups were first analyzed in a 2 by 2 cross-tabulation table and
odds -ratio plus 95% confidence intervals were calculated. Sub-
sequently, logistic regression analysis was performed controlling
for baseline values and eight other relevant, possibly confound-
ing variables (gender, age, body mass index, baseline hernia, Hill
grade, esophagitis grade, LES resting pressure, and duration of
GERD symptoms). Secondary outcome parameters were ana-
lyzed by repeated measures analysis of variance and analysis of
covariance. Additional linear regression analysis was performed
to facilitate the statistical control of confounding variables. Mixed
models were used to analyze repeated measurements for the entire
cohort of patients who underwent TIF2 (after eventual crossover
for the PPI group). SPSS-pc, version 20.0 (IBM, Armonk, NY)
was used for data analysis. A P value of <0.05 was considered
statistically significant.

RESULTS
Randomized controlled trial
Recruitment was initiated in 2008, and in 2011, 60 consecutive
patients were randomized and completed the follow-up in one
center. At the time of study completion in one center, the recruit-
ment had not been started in the second center, mainly because
of financial issues. We performed analysis of the single-center
Figure 1. Transoral incisionless fundoplication (TIF) procedure. (a) The data to function as an unplanned interim analysis of the trial and
endoscope is retracted to allow the tissue mold to close. Vacuum suction report the results. A consort flow diagram shows the inclusion
holes enable small hiatal hernia reduction. (b) The device is advanced and allocation at interim analysis (Figure 3).
and hiatal hernia reduced. Visualization is provided with the endoscope in
The majority of patients were Caucasian and 62% were male.
retroflexed view. (c) The helical retractor is engaged to retract tissue into
the mold. (d) Polypropylene H-shaped fasteners are advanced by stylets. Patient characteristics were comparable for the two randomized
(e) After sequential deployment of fastener sets, a full-thickness fundo- groups, with similar distribution of demographics and anatomical
plication is created at the level of the gastroesophageal junction. deterioration of the gastroesophageal junction (Table 1).
(f) Three-dimensional image of the reconstructed gastroesophageal valve. At 6-month follow-up, quality-of-life evaluation showed a signif-
icantly higher improvement of ≥50% after TIF2 (55%) compared
with PPI treatment (5%; P<0.001, Table 2). Satisfaction scores in
the TIF group showed that 50% were satisfied with their general
enrollment number only. Next to the case report forms, an elec- health condition, 26% were neutral, and 24% were dissatisfied; for
tronic database capture system (Microsoft Excel 97–2004) was the PPI group, 0% were satisfied, 15% were neutral, and 85% were
used to collect, organize, and report the study data. Data in the dissatisfied. Esophageal acid exposure time decreased more after
case report forms were entered by the study coordinators. Only PPI treatment, but no significant difference in decrease was found
investigators and monitors had the privilege to read data. Trial between randomized groups (total: P=0.228). Normalization of pH
monitors verified the accuracy of this transcription against the (total proportion of time with a pH of <4 in 24-h period ≤4.2%)
source documents. for TIF group and PPI group was 50% and 63%, respectively.
TIF2 induced a significant increase in lower esophageal sphinc-
Statistical analysis ter resting pressure compared with the PPI group (P=0.004), but
According to the intention-to-treat principle, all enrolled subjects the total number of reflux episodes at impedance measurements
were included in the primary effectiveness analysis. Missing data did not improve significantly (P=0.058). Both groups had similar
imputation was done by last-value-carried-forward techniques outcomes in healing of esophagitis after 6 months compared with
and for baseline by overall mean imputation. Metric variables baseline endoscopy (TIF2: 33 to 14% and PPI: 30 to 10%). In the

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

 TIF Vs. PPIs for GERD 535

ESOPHAGUS
à

Figure 2. Polypropylene H-shaped serosal fasteners are used to secure the full-thickness fundoplication at the level of the gastroesophageal junction.

Randomized n =60
junction, requiring additional insufflations. The patient appeared
to have developed a pneumoperitoneum that was released with the
insertion of a Verres needle. Postoperative upper gastrointestinal
Allocated to TIF goup Allocated to PPI group
n =40 n =20
radiography did not show any signs of perforation and the event
had no further impact on the patient’s recovery.
Lost to follow-up n =3 Lost to follow-up n =0 Postoperative complications at day 1 were mild and transient in
(incomplete data)
all cases. Among these complications, sore throat, cough, belch-
ing, pain, nausea, vomiting, bloating, left shoulder pain, melaena,
Analyzed n =37 Analyzed n =20
at 6 months at 6 months hematemesis, constipation, asthma exacerbation, and tinnitus
were recorded.
Allocated to TIF cohort Crossover to TIF cohort Serious adverse events included pneumonia in three patients, of
n =57 n =20
whom one needed readmission and intravenous antibiotics. One
Lost to follow-up n =7 patient experienced severe epigastric pain 3 weeks after the TIF
(incomplete data, all from procedure. The patient was readmitted and analyzed with an upper
the original TIF group)
gastrointestinal radiography and endoscopy that showed an intact
Analyzed n =50 valve without further abnormalities and the pain resolved sponta-
at 6 months after TIF
neously. We encountered 1 death at 20 months after inclusion. The
Lost to follow-up n =5 patient was male, aged 52 years, and randomized for PPI group. He
(revisional surgery n =3 (2 had undergone TIF procedure after crossover and died 11 months
from PPI/1 from TIF group),
incomplete data n =2) after TIF procedure. Death was ascertained by his general practi-
tioner. The cause of death was assumed cardiac or neurological,
Analyzed n =45
at 12 months after TIF
but the exact cause remains uncertain. Autopsy was not performed
upon family request. The event was reported to the ethics commit-
Figure 3. Consort flow diagram. PPI, proton pump inhibitor; TIF, transoral tee and within the time frame it was considered probably unrelated
incisionless fundoplication. to the trail interventions.
Treatment failure based on relapse of symptoms, unrespon-
sive to PPI treatment, in conjunction with anatomical failure at
endoscopy and/or pathologic pH measurements occurred in three
PPI group, all patients continued the use of PPIs, mostly at a single patients. Revisional laparoscopic fundoplication was performed in
dose per day. Cessation of PPIs occurred in 74% of patients after two patients, randomized to the PPI group (after crossover to TIF)
TIF2, 17% used PPIs occasionally, and 9% daily at 6-month follow- at 6 and 14 months, and in one patient, randomized to the TIF
up (Table 3). group at 10 months after procedure.
Perioperative complications of TIF2 in this trial were limited to Fundoplication-related gastrointestinal side effects, such as
one adverse event. During one TIF2 procedure it was difficult to epigastric pain, flatulence, bloating, diarrhea, dysphagia, and
maintain an adequate endoscopic view on the gastroesophageal the ability to vomit and belch, assessed with Gastrointestinal

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 536 Witteman et al.

Table 1. Patient characteristics at baseline TIF2, complete reduction of hiatal hernia (if present) was accom-
plished in all cases. Hill grade I was achieved in 90%, Hill grade II
Variable TIF2 (n=40) PPI (n=20) Total P value
in 8%, Hill grade III in 2%, and no patients had a Hill grade IV at
the end of the procedure. Adherence to the endoscope was tight,
ESOPHAGUS

Age (mean) 42.4 49.3 44.7 P=0.05a

(s.d.) (13.3) (11.3) (12.9) moderate, or loose in 83%, 14%, and 3%, respectively (Table 4).
Gender (no.) (%) At 6 (n=53) and 12 (n=45) months after TIF, quality-of-life
Male 24 (60) 14 (70) 38 (63) scores showed a significant improvement compared with baseline
measurements (P<0.05, Table 5). At 6 and 12 months, the number
Female 16 (40) 6 (30) 22 (37) P=0.45b
of reflux episodes and the increase of lower esophageal sphincter
BMI (mean) 26 (3.7) 26 (3.4) 26 (3.6) P=0.95a
pressure were significantly improved compared with baseline.
(s.d.)
Distal esophageal acid exposure, however, did not significantly
Hernia presence (no.) (%)
improve compared with baseline at 6 months (P=0.06) and at 12
No 9 (23) 9 (45) 18 (30) months after TIF (P=0.171). Normalization of pH was accom-
Yes 31 (77) 11 (55) 42 (70) P=0.08b plished after TIF in 44% at 6 months and only in 29% at 12 months.
Hill grade (no.) (%) The use of antisecretory medication was discontinued by 39%, 17%
needed PPIs occasionally and 44% on a daily basis at 12-month
I 2 (6) 1 (6) 3 (6)
follow-up (Table 6).
II 18 (53) 11 (69) 29 (58)
III 11 (32) 4 (25) 15 (30)
IV 3 (9) 0 (0) 3 (6) P=0.40b DISCUSSION
Esophagitis (no.) (%) History of endoscopic antireflux treatment modalities has taught
us that early studies with encouraging results have been proven
None 27 (68) 14 (70) 41 (68)
wrong by later randomized controlled trials, showing the true
A 7 (17) 3 (15) 10 (17) efficacy and durability (52). Accordingly, a randomized controlled
B 6 (15) 3 (15) 9 (15) P=0.97b trial was in need to assess the effectiveness and durability of TIF2
LES resting 15.2 (8–25) 15.5 (6–25) 15.3 P=0.893c compared with standard treatment modalities. In this trial, 60
pressure (6–25) selected GERD patients, responsive to medical treatment, were
(median)
(range)
randomized for either TIF or PPI treatment.
After a 6-month follow-up, TIF resulted in a significantly
Duration of 4.5 5.4 4.8 P=0.882c
GERD (0.05–18.95) (0.33–20.30) (0.05–20.3)
improved GERD symptom score compared with PPI treatment
(median) with a similar decrease in esophageal acid exposure. In addition,
(range) this improvement in GERD-related quality of life was not com-
BMI, body mass index; GERD, gastroesophageal reflux disease; LES, lower promised by fundoplication-related side effects, such as gas bloat-
esophageal sphincter; PPI, proton pump inhibitor; TIF, transoral incisionless
fundoplication.
ing and dysphagia. After crossover for the PPI group, the cohort
Values represent means (s.d.), median (range), or counts (%). of 60 patients who had undergone TIF2 was evaluated at 6 and
a
Analysis of covariance (ANCOVA). 12 months. Symptoms scores, lower esophageal sphincter resting
b
The χ 2L test. pressure, and total number of reflux episodes were still signifi-
c
Mann–Whitney U-test. cantly improved but distal esophageal acid exposure increased in
conjunction with deteriorated endoscopy findings. At 12 months
after TIF2, 5% of patients had undergone revisional surgery to
Symptom Rating Scale questionnaire did not occur. The TIF group control their symptoms. Normalization of distal esophageal acid
showed more improvement in Gastrointestinal Symptom Rating exposure was only achieved in 29%, and 61% needed PPIs occa-
Scale scores compared with PPI group (P=0.001) at 6-month fol- sional or on daily basis.
low-up. The available literature on the TIF procedure, published between
2008 and 2012, showed a variation of outcomes derived from 18
TIF2 cohort results reports on mostly initial experiences. These include retrospective
Further objectives of the study were to evaluate effectiveness and prospective studies with a nonrandomized design and gener-
of the TIF2 procedure in restoring the anatomy, biomechan- ally a short follow-up duration (mean 12.8 months, range 2–36).
ics, geometry, and physiology of the antireflux barrier. The “TIF Only few studies included objective parameters, such as pH meas-
cohort” was composed of all patients who had undergone TIF2 urements, and many reported only subjective outcome measures,
in the study, including crossovers from the PPI group. TIF2 was such as quality of life and PPI usage. Although the results did not
feasible in all patients (n=60), mean procedure time was 33.4 min compare with those of laparoscopic antireflux surgery, overall
(range 17–75), and mean 18 fasteners (range 7–26) were used for outcomes in 15 studies were in favor of TIF. Three studies have
the construction of valves of mean 3.8 cm (range 3–5) in length reported negative outcomes and, additionally, two papers have
and mean 229° (range 160–300°) in circumference. Directly after been published on revisional surgery after previous unsuccessful

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

 TIF Vs. PPIs for GERD 537

Table 2. TIF vs. PPI results at 6-month follow-up

Variable Baseline 6 Months Baseline vs. 6-month

follow-up

ESOPHAGUS
TIF2 (n=40) PPI (n=20) TIF2 (n=37) PPI (n=20) P value

GERD-HRQL (mean) (s.d.) 26.5 (8.0) 28.2 (9.5) 12.4 (10.0) 25.1 (11.2) P<0.001a
% GERD-HRQL >50% improvement 55% (20/37) 5% (1/20) P<0.001b
pH measurements: Total % time pH <4 (median) 10.8 (4–40) 11.3 (5–31) 7.7 (1–40) 6.0 (0–23) P=0.228a
(range)
Impedance measurements: Total number of reflux 111 (42–306) 109 (33–205) 78 (12–248) 101 (29–264) P=0.058a
episodes (median) (range)
LES resting pressure mm Hg (median) (range) 15.2 (8–25) 15.5 (6–25) 18.2 (10–40) 13.6 (4–20) P=0.004a
Esophagitis (%) 33% (13/40) 30% (6/20) 14% (5/37) 10% (2/20) P>0.05a
GERD-HRQL, gastroesophageal reflux disease–health-related quality-of-life; LES, lower esophageal sphincter; PPI, proton pump inhibitor; TIF, transoral incisionless
fundoplication.
a
Analysis of covariance (ANCOVA).
b
The χ 2L test.

Table 3. PPI usage for TIF vs. PPI group at baseline and at Other adverse events and side effects were probably related to
6-month follow-up the insertion of the 18 mm diameter device, with subsequently
repetitive axial rotations and the “wrapping” of the greater gas-
PPI usage Baseline 6 Months
tric curvature around the esophagus. These side effects were of
TIF group (n=40) TIF group (n=37) a transient nature. Most importantly, the feared postsurgical
None 0% (0/40) 74% (28/37)
fundoplication side effects, such as dysphagia and gas bloating,
resolved within a short period of time. No long-term fundo-
Occasional 0% (0/40) 17% (6/37)
plication-related complications occurred. These outcomes
Daily single dose 100% (40/40) 9% (3/37) correspond with the results of systematic reviews in which
Daily double dose 0% (0/40) 0% (0/37) partial fundoplication techniques were associated with signifi-
cantly less fundoplication-related side effects compared with
PPI group (n=20) PPI group (n=20)
total 360° fundoplication (66).
None 0% (0/20) 0% (0/20) Earlier, we reported a previous experience with TIF1 proce-
Occasional 0% (0/20) 0% (0/20) dure after a cohort study in 38 patients who were referred for
Daily single dose 100% (20/20) 90% (18/20) antireflux surgery. In contrast with the improved quality of life
and decrease in PPI usage, postoperative pH measurements
Daily double dose 0% (0/20) 10% (2/20)
did not significantly improve compared with baseline, and after
PPI, proton pump inhibitor; TIF, transoral incisionless fundoplication.
prolonged follow-up duration, many patients needed additional
PPIs and 37% of patients had undergone revisional laparoscopic
Nissen fundoplication after mean 17 months (range 7–34) (36).
TIF procedures. In the 18 studies, with a total of a composite 728 Disruption of the polypropylene H-fasteners, leading to failure
patients, 50 cases (7%) of revisional laparoscopic fundoplication of the TIF valve as a barrier to reflux, was seen in the majority of
have been reported (31–37,53–65). these patients.
Although TIF has been considered relatively safe, serious adverse In the present study, TIF2 enabled to create valves, scored as a
events have been reported in the literature and include esophageal Hill grade I in 90% at endoscopy directly after the procedure, but
perforations on device insertion, hemorrhage requiring transfu- after 12 months a Hill grade I was only seen in 35% of patients.
sion, permanent numbness of the tongue, pneumonia, and pneu- The lack of durability has been an important cause of treatment
mothorax. Serious adverse events during the TIF2 procedure in failure of previous endoscopic procedures for GERD and also
present trial were a pneumoperitoneum. The air leakage had prob- appears to be of concern for the TIF- valve (67,68). Literature data
ably occurred because of the required high insufflation pressures have shown feasibility of the TIF procedure in the creation of a
via the small perforations made with fastener deployment. Seri- Hill grade I gastroesophageal valve, and after the procedure most
ous postoperative adverse events included three pneumonias and patients experience a relatively symptom-free period, allowing
a readmission for unexplained severe epigastric pain. One patient PPI cessation. This symptom-free period may be induced by ini-
died 11 months after the TIF2 procedure, most likely because of an tial placebo effects, but considering objective outcomes in GERD
unrelated cause. control, it is more likely explained by an initially well-functioning

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 538 Witteman et al.

antireflux barrier. Resumption of symptoms in previous stud- Because extragastric mobilization of the fundus to release the
ies has been dedicated to a “suture loss.” The “pull-through” of tension of the fundoplication cannot be achieved with TIF, the
fasteners, verified by endoscopy or laparoscopy (in revisional continuous force on the repair eventually causes the anatomy
laparoscopic procedures), has been reported by others (34–37). of the GEJ to return to its original shape, causing the valve to
ESOPHAGUS

unravel (34). This finding is supported by a study of Testoni et al.

(54,65) who found that a higher number of fasteners, used in the
Table 4. Endoscopy at baseline and 6 and 12 months after TIF2 procedure, was predictive for positive outcomes.
procedure There are several limitations to this study. First, the trial has not
been completed and at interim analysis half of the required partici-
Endoscopy Baseline After TIF2 6 Months 12 Months
after TIF2 after TIF2 pants were included. The study as reported is therefore underpow-
ered and results should be interpreted with caution.
Number of 60 60 53 45
patients
It is also important to stress that we included highly selected
GERD patients, only with a small (≤2 cm) hiatal hernia in the trial.
Hiatal hernia 70% (42/60) 0% 37% (20/53) 55% (25/45)
(HHD)
A large hiatal hernia, while entirely acceptable for laparoscopic
GERD treatment, is considered a contraindication for TIF.
HHD 1 cm 38% (16/42) — 50% (10/20) 43% (11/25)
Another limitation is the short follow-up duration of only 6
HHD 2 cm 62% (26/42) — 30% (6/20) 38% (10/25) months for the comparative part of the trial. The crossover design
HHD 3 cm 0% — 20% (4/20) 19% (4/25) was chosen to facilitate recruitment and seemed to be valid because
Esophagitis 32% (19/60) — 25% (13/53) 27% (12/45) the outcomes of PPI treatment with an extensive follow-up dura-
Grade A 53% (10/19) — 38% (5/13) 40% (5/12)
tion are widely known (69).
GERD is a symptom-driven disease and in this trial we used
Grade B 47% (9/19) — 46% (6/13) 60% (7/12)
a score for cardinal symptoms as our primary outcome meas-
Grade C 0% — 16% (2/13) 0% ure that is in correspondence with the paper written by Dent et
Grade D 0% — 0% 0% al. (70) in which consensus statements for clinical trial design
Barrett 0% — 0% 0% were enumerated in order to improve quality and comparability
of clinical trials evaluating therapies for GERD. In this trial, a
Z-line 85% (51/60) — 89% (47/53) 90% (40/45)
normal ≥50% improvement of symptom score was considered clinically
Z-line 15% (9/60) — 11% (6/53) 10% (5/45)
significant. The ≥50% was chosen in accordance with previous
irregular
Hill grade I 6% (4/60) 90% (54/60) 39% (21/53) 35% (16/45)
Hill grade II 58% (34/60) 8% ((5/60) 42% (22/53) 35% (16/45) Table 6. PPI usage at baseline and 6 and 12 months after TIF2
procedure
Hill grade III 30% (18/60) 2% (1/60) 19% (20/53) 25% (11/45)
Hill grade IV 6% (4/60) 0% 0% 5% (2/45) PPI usage Baseline 6 Months after 12 Months after
n=60 TIF2 n=53 TIF2 n=45
Tight 12% ((7/60) 83% (50/60) 59% (31/53) 42% (19/45)
adherence None 0% 66% (35/53) 39% (17/45)

Moderate 50% (30/60) 15% (9/60) 27% (14/53) 29% (13/45) Occasional 0% 19% (10/53) 17% (8/45)
adherence
Daily single dose 100% (60/60) 11% (6/53) 36% (16/45)
Loose 38% (23/60) 2% (1/60) 14% (8/53) 29% (13/45)
Daily double dose 0% 4% (2/53) 8% (4/45)
adherence
PPI, proton pump inhibitor; TIF, transoral incisionless fundoplication.
TIF, transoral incisionless fundoplication.

Table 5. Outcomes parameters at baseline and 6 and 12 months after TIF2 procedure

Variable Baseline 6 Months 12 Months

TIF2 (n=60) TIF2 (n=53) P valuea TIF2 (n=45) P valuea

GERD-HRQL (mean) (s.d.) 27.1 (8.4) 11.1 (9.7) P<0.05 10.3 (7.8) P<0.05
pH measurements: Total %time pH<4 (mean) (s.d.) 11.0 (7.1) 7.9 (8.0) P=0.06 9.1 (8.7) P=0.171
Impedance, no. of reflux episodes (mean) (s.d.) 110.0 (56.4) 71.1 (43.6) P<0.05 73.8 (40.9) P<0.05
LES resting pressure mm Hg (mean) (s.d.) 15.3 (4.8) 17.8 (5.9) P<0.05 17.6 (5.9) P<0.05
GERD-HRQL, gastroesophageal reflux disease–health-related quality-of-life; LES, lower esophageal sphincter; TIF, transoral incisionless fundoplication.
a
Mixed model analysis.

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

 TIF Vs. PPIs for GERD 539

studies from Cadiere’s group evaluating TIF but may have set Study Highlights
the bar too low because based on symptoms alone, this trail has
outcomes in favor of TIF because symptom scores remained sig- WHAT IS CURRENT KNOWLEDGE
✓ Gastroesophageal reflux disease (GERD) has a growing

ESOPHAGUS
nificantly improved, even after 12-month follow-up (33). How-
ever, looking at pH measurements, reflux control decreased in incidence and prevalence.
correspondence with deteriorated valve appearance at endos- ✓ Merits of early GERD surgery have been demonstrated.
copy and the high resumption of PPIs at 12-month follow-up. ✓ Transoral incisionless fundoplication (TIF) was developed
This was the main reason for our research group not to complete to endoscopically mimic antireflux surgery.
the trial and include another 60 patients, before the durability WHAT IS NEW HERE
of TIF has been improved. The reason why symptoms scores ✓ TIF produced short-term improvement of the antireflux barrier.
remained improved remains unclear, but it suggests that placebo ✓ TIF produced no long-term reflux control.
effect may play a role in this study evaluating a novel treatment ✓ TIF is no equivalent alternative to proton pump inhibitors
modality. (PPIs).
In conclusion, this randomized controlled trial tested the
hypothesis that TIF would be as effective as PPI treatment in
controlling GERD symptoms, esophageal acid exposure, and REFERENCES
esophagitis, but would be more effective in controlling reflux 1. Vakil N, van Zanten SV, Kahrilas P et al. The Montreal definition and
classification of gastroesophageal reflux disease: a global evidence-based
and thereby improve the quality of life. Although a short-term consensus. Am J Gastroenterol 2006;101:1900–20.quiz 1943.
improvement of the antireflux barrier was found, supporting the 2. Stylopoulos N, Rattner DW. The history of hiatal hernia surgery: from
validity of the concept of TIF2, no long-term reflux control was Bowditch to laparoscopy. Ann Surg 2005;241:185–93.
3. Jones MP, Sloan SS, Rabine JC et al. Hiatal hernia size is the dominant
achieved and, therefore, the study was terminated after interim determinant of esophagitis presence and severity in gastroesophageal reflux
analysis. The endoscopically created antireflux barrier deterio- disease. Am J Gastroenterol 2001;96:1711–7.
rated with increased follow-up duration and, therefore, in this 4. Mittal RK. Hiatal hernia: myth or reality? Am J Med 1997;103:33S–39S.
5. Mittal RK. Hiatal hernia and gastroesophageal reflux: another attempt to
form, TIF2 is no equivalent alternative for PPIs in GERD treat- resolve the controversy. Gastroenterol 1993;105:941–3.
ment, even in this highly selected population. A next-generation 6. Moss SF, Kidd M, Modlin IM. The status of the hiatus: the role of hernia in
endoscopic fundoplication technique may improve outcomes, gastroesophageal reflux disease. J Clin Gastroenterol 2007;41:S144–S153.
7. Rydberg L, Ruth M, Lundell L. Mechanism of action of antireflux proce-
if equipped with a more solid suturing or stapling mechanism dures. Br J Surg 1999;86:405–10.
to better withstand intraabdominal traction forces and warrant 8. El-Serag HB. Time trends of gastroesophageal reflux disease: a systematic
durability. review. Clin Gastroenterol Hepatol 2007;5:17–26.
9. Lundell L. Surgery of gastroesophageal reflux disease: a competitive or
complementary procedure? Dig Dis 2004;22:161–70.
ACKNOWLEDGMENTS 10. Metz DC. Managing gastroesophageal reflux disease for the lifetime of
We thank the study coordinators Liza Toemen, Gitte Snijders, and the patient: evaluating the long-term options. Am J Med 2004;117 Suppl
5A:49S–55S.
Ralf Vijgen for their assistance with follow-up of patients and Dana 11. Moayyedi P, Talley NJ. Gastro-oesophageal reflux disease. Lancet
Hamers for the TIF technique illustrations. This paper was a poster 2006;367:2086–100.
presentation at the Digestive Disease Week 2013, Orlando, Florida, 12. Nandurkar S, Talley NJ. Epidemiology and natural history of reflux disease.
Baillieres Best Pract Res Clin Gastroenterol 2000;14:743–57.
USA. 13. Smout AJ. The patient with GORD and chronically recurrent problems.
Best Pract Res Clin Gastroenterol 2007;21:365–78.
CONFLICT OF INTEREST 14. Fass R. Proton-pump inhibitor therapy in patients with gastro-oesophageal
reflux disease: putative mechanisms of failure. Drugs 2007;67:1521–30.
Guarantor of the article: Bart P.L. Witteman, PhD. 15. Gerson LB, Boparai V, Ullah N et al. Oesophageal and gastric pH profiles
Specific author contributions: Bart P.L.Witteman: literature search, in patients with gastro-oesophageal reflux disease and Barrett's oe-
conception and design, analysis and interpretation, writing the sophagus treated with proton pump inhibitors. Aliment Pharmacol Ther
2004;20:637–43.
article, final approval of the article; José M. Conchillo: conception 16. Lundell L, Miettinen P, Myrvold HE et al. Continued (5-year) followup of a
and design, analysis and interpretation, critical revision of the arti- randomized clinical study comparing antireflux surgery and omeprazole in
cle, final approval of the article; Nicolaas F. Rinsma: literature search, gastroesophageal reflux disease. J Am Coll Surg 2001;192:172–9.discussion
179-181.
analysis and interpretation, critical revision of the article, final 17. Anvari M, Allen C, Marshall J et al. A randomized controlled trial of
approval of the article; Bark Betzel and Andrea Peeters: statistical laparoscopic Nissen fundoplication versus proton pump inhibitors for
expertise, final approval of the article; Ger H. Koek: conception and treatment of patients with chronic gastroesophageal reflux disease: one-year
follow-up. Surg Innov 2006;13:238–49.
design, critical revision of the article, final approval of the article; 18. Spechler SJ. Comparison of medical and surgical therapy for complicated
Laurents P.S. Stassen: critical revision of the article, final approval gastroesophageal reflux disease in veterans. The Department of Veterans
of the article; Nicole D. Bouvy: literature search, conception and Affairs Gastroesophageal Reflux Disease Study Group. N Engl J Med
1992;326:786–92.
design, analysis and interpretation, critical revision of the article, 19. Spechler SJ, Lee E, Ahnen D et al. Long-term outcome of medical and
writing the article, final approval of the article. surgical therapies for gastroesophageal reflux disease: follow-up of a rand-
Financial support: This study was supported by Maastricht omized controlled trial. JAMA 2001;285:2331–8.
20. Allgood PC, Bachmann M. Medical or surgical treatment for chronic
University funding. gastrooesophageal reflux? A systematic review of published evidence of
Potential competing interests: None. effectiveness. Eur J Surg 2000;166:713–21.

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 540 Witteman et al.

21. Cookson R, Flood C, Koo B et al. Short-term cost effectiveness and long- 47. Jobe BA, Kahrilas PJ, Vernon AH et al. Endoscopic appraisal of the gastro-
term cost analysis comparing laparoscopic Nissen fundoplication with esophageal valve after antireflux surgery. Am J Gastroenterol 2004;99:
proton-pump inhibitor maintenance for gastro-oesophageal reflux disease. 233–43.
Br J Surg 2005;92:700–6. 48. Hill LD, Kozarek RA, Kraemer SJ et al. The gastroesophageal flap valve: in
22. Lundell L, Miettinen P, Myrvold HE et al. Seven-year follow-up of a vitro and in vivo observations. Gastrointest Endosc 1996;44:541–7.
ESOPHAGUS

randomized clinical trial comparing proton-pump inhibition with surgical 49. Jobe BA, O'Rourke RW, McMahon BP et al. Transoral endoscopic fundopli-
therapy for reflux oesophagitis. Br J Surg 2007;94:198–203. cation in the treatment of gastroesophageal reflux disease: the anatomic and
23. Mahon D, Rhodes M, Decadt B et al. Randomized clinical trial of laparo- physiologic basis for reconstruction of the esophagogastric junction using a
scopic Nissen fundoplication compared with proton-pump inhibitors for novel device. Ann Surg 2008;248:69–76.
treatment of chronic gastro-oesophageal reflux. Br J Surg 2005;92:695–9. 50. Bell RC, Cadiere GB. Transoral rotational esophagogastric fundoplica-
24. Dallemagne B, Weerts J, Markiewicz S et al. Clinical results of laparoscopic tion: technical, anatomical, and safety considerations. Surg Endosc
fundoplication at ten years after surgery. Surg Endosc 2006;20:159–65. 2011;25:2387–99.
25. Coelho JC, Wiederkehr JC, Campos AC et al. Conversions and complica- 51. Kulich KR, Madisch A, Pacini F et al. Reliability and validity of the
tions of laparoscopic treatment of gastroesophageal reflux disease. J Am Gastrointestinal Symptom Rating Scale (GSRS) and Quality of Life in Reflux
Coll Surg 1999;189:356–61. and Dyspepsia (QOLRAD) questionnaire in dyspepsia: a six-country study.
26. Kamolz T, Granderath FA, Bammer T et al. Dysphagia and quality of life Health Qual Life Outcomes 2008;6:12.
after laparoscopic Nissen fundoplication in patients with and without 52. Jafri SM, Arora G, Triadafilopoulos G. What is left of the endoscopic antire-
prosthetic reinforcement of the hiatal crura. Surg Endosc 2002;16:572–7. flux devices? Curr Opin Gastroenterol 2009;25:352–7.
27. Lundell L. Complications after anti-reflux surgery. Best Pract Res Clin 53. Repici A, Fumagalli U, Malesci A et al. Endoluminal fundoplication (ELF)
Gastroenterol 2004;18:935–45. for GERD using EsophyX: a 12-month follow-up in a single-center experi-
28. Wills VL, Hunt DR. Dysphagia after antireflux surgery. Br J Surg ence. J Gastrointest Surg 2010;14:1–6.
2001;88:486–99. 54. Testoni PA, Corsetti M, Di Pietro S et al. Effect of transoral incisionless
29. Bessell JR, Finch R, Gotley DC et al. Chronic dysphagia following laparo- fundoplication on symptoms, PPI use, and ph-impedance refluxes of GERD
scopic fundoplication. Br J Surg 2000;87:1341–5. patients. World J Surg 2010;34:750–7.
30. Hayden J, Jamieson G. Optimization of outcome after laparoscopic antire- 55. Velanovich V. Endoscopic, endoluminal fundoplication for gastroesophageal
flux surgery. ANZ J Surg 2006;76:258–63. reflux disease: initial experience and lessons learned. Surgery 2010;148:
31. Cadiere GB, Rajan A, Rqibate M et al. Endoluminal fundoplication (ELF)- 646–51. discussion 651-3.
-evolution of EsophyXTM, a new surgical device for transoral surgery. 56. Demyttenaere SV, Bergman S, Pham T et al. Transoral incisionless fun-
Minim Invasive Ther Allied Technol 2006;15:348–55. doplication for gastroesophageal reflux disease in an unselected patient
32. Cadiere GB, Rajan A, Germay O et al. Endoluminal fundoplication by a population. Surg Endosc 2010;24:854–8.
transoral device for the treatment of GERD: a feasibility study. Surg Endosc 57. Bell RC, Freeman KD. Clinical and pH-metric outcomes of transoral
2008;22:333–42. esophagogastric fundoplication for the treatment of gastroesophageal reflux
33. Cadiere GB, Buset M, Muls V et al. Antireflux transoral incisionless fun- disease. Surg Endosc 2011;25:1975–84.
doplication using EsophyX: 12-month results of a prospective multicenter 58. Barnes WE, Hoddinott KM, Mundy S et al. Transoral incisionless fundo-
study. World J Surg 2008;32:1676–88. plication offers high patient satisfaction and relief of therapy-resistant
34. Hoppo T, Immanuel A, Schuchert M et al. Transoral incisionless fundopli- typical and atypical symptoms of GERD in community practice. Surg Innov
cation 2.0 procedure using EsophyX for gastroesophageal reflux disease. J 2011;18:119–29.
Gastrointest Surg 2010;14:1895–901. 59. Trad KS, Turgeon DG, Deljkich E. Long-term outcomes after transoral
35. Furnee EJ, Broeders JA, Draaisma WA et al. Laparoscopic Nissen fundopli- incisionless fundoplication in patients with GERD and LPR symptoms.
cation after failed EsophyX fundoplication. Br J Surg 2011;97:1051–5. Surg Endosc 2012;26:650–60.
36. Witteman BP, Strijkers R, de Vries E et al. Transoral incisionless fundoplica- 60. Petersen RP, Filippa L, Wassenaar EB et al. Comprehensive evaluation of endo-
tion for treatment of gastroesophageal reflux disease in clinical practice. scopic fundoplication using the EsophyX device. Surg Endosc 2012;26:1021–7.
Surg Endosc 2012;26:3307–15. 61. Narsule CK, Burch MA, Ebright MI et al. Endoscopic fundoplication for
37. Witteman BP, Kessing BF, Snijders G et al. Revisional laparoscopic antire- the treatment of gastroesophageal reflux disease: initial experience. J Thorac
flux surgery after unsuccessful endoscopic fundoplication. Surg Endosc Cardiovasc Surg 2012;143:228–34.
2013;27:2231–6. 62. Bell RC, Mavrelis PG, Barnes WE et al. A prospective multicenter registry
38. Velanovich V, Vallance SR, Gusz JR et al. Quality of life scale for gastroe- of patients with chronic gastroesophageal reflux disease receiving transoral
sophageal reflux disease. J Am Coll Surg 1996;183:217–24. incisionless fundoplication. J Am Coll Surg 2012;215:794–809.
39. Kahrilas PJ, Sifrim D. High-resolution manometry and impedance-pH/ 63. Muls V, Eckardt AJ, Marchese M et al. Three-year results of a multicenter
manometry: valuable tools in clinical and investigational esophagology. prospective study of transoral incisionless fundoplication. Surg Innov
Gastroenterology 2008;135:756–69. 2012;20:321–30.
40. Conchillo JM, Schwartz MP, Selimah M et al. Acid and non-acid reflux patterns 64. Bergman S, Mikami DJ, Hazey JW et al. Endolumenal fundoplication
in patients with erosive esophagitis and non-erosive reflux disease (NERD): with EsophyX: the initial North American experience. Surg Innov 2008;15:
a study using intraluminal impedance monitoring. Dig Dis Sci 2008;53: 166–70.
1506–12. 65. Testoni PA, Vailati C, Testoni S et al. Transoral incisionless fundoplication
41. Bredenoord AJ. Impedance-pH monitoring: new standard for measuring (TIF 2.0) with EsophyX for gastroesophageal reflux disease: long-term
gastro-oesophageal reflux. Neurogastroenterol Motil 2008;20:434–9. results and findings affecting outcome. Surg Endosc 2010;26:1425–35.
42. Zerbib F, Roman S, Ropert A et al. Esophageal pH-impedance monitoring 66. Broeders JA, Mauritz FA, Ahmed Ali U et al. Systematic review and meta-
and symptom analysis in GERD: a study in patients off and on therapy. analysis of laparoscopic Nissen (posterior total) versus Toupet (posterior
Am J Gastroenterol 2006;101:1956–63. partial) fundoplication for gastro-oesophageal reflux disease. Br J Surg
43. Sifrim D, Dupont L, Blondeau K et al. Weakly acidic reflux in patients with 2010;97:1318–30.
chronic unexplained cough during 24 hour pressure, pH, and impedance 67. Schiefke I, Zabel-Langhennig A, Neumann S et al. Long term failure of
monitoring. Gut 2005;54:449–54. endoscopic gastroplication (EndoCinch). Gut 2005;54:752–8.
44. Bernhard A, Pohl D, Fried M et al. Influence of bolus consistency and 68. Pearl JP, Marks JM. Endolumenal therapies for gastroesophageal reflux
position on esophageal high-resolution manometry findings. Dig Dis Sci disease: are they dead? Surg Endosc 2007;21:1–4.
2008;53:1198–205. 69. Miner PBJr.. Review article: physiologic and clinical effects of proton pump
45. Bredenoord AJ, Smout AJ. High-resolution manometry. Dig Liver Dis inhibitors on non-acidic and acidic gastro-oesophageal reflux. Aliment
2008;40:174–81. Pharmacol Ther 2006;23 Suppl 1:25–32
46. Fox MR, Bredenoord AJ. Oesophageal high-resolution manometry: moving 70. Dent J, Kahrilas PJ, Vakilà N et al. Clinical trial design in adult reflux disease:
from research into clinical practice. Gut 2008;57:405–23. a methodological workshop. Aliment Pharmacol Ther 2008;28:107–26.

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

 TIF Vs. PPIs for GERD 541

APPENDIX 1
Population size determination
A prospective power analysis conducted with gastroesophageal reflux disease health-related quality-of-life (GERD-HRQL) scores from

ESOPHAGUS
the phase 2 study with EsophyX-TIF1 (1) revealed that a study population of n=120, including enrollment loss of 5%, will have 80%
power to identify a significant difference between TIF2 and proton pump inhibitors (PPIs) in the two-arm trial with a 2:1 randomiza-
tion ratio (bold in Table). The objective was to demonstrate that TIF is not significantly worse than control (equivalence). “Significantly
worse” was defined by the “equivalence Δ .” With the equivalence Δ of 2, n=120 was adequate for demonstrating equivalence if the true
transoral incisionless fundoplication (TIF) mean was better than the control by 4 points.

(A) Control (B) Equivalence? (C) Total?b (D) s.d.c (assumed to be the Power Total N (2 arms Enrollment goal Enrollment goal
mean-TIF meana (extra marginb) same for both studies) combined) if 5% loss if 10% loss

4 2 6 12 0.90 156 164 173

4 2 6 12 0.85 132 139 147
4 2 6 12 0.80 114 120 127
3 2 5 12 0.90 222 234 247
3 2 5 12 0.85 189 199 210
3 2 5 12 0.80 162 171 180
2 2 4 12 0.90 348 366 387
2 2 4 12 0.85 294 309 327
2 2 4 12 0.80 252 265 280
4 2 6 13 0.90 183 193 203
4 2 6 13 0.85 153 161 170
4 2 6 13 0.80 132 139 147
3 2 5 13 0.90 261 275 290
3 2 5 13 0.85 219 231 243
3 2 5 13 0.80 189 199 210
2 2 4 13 0.90 408 429 453
2 2 4 13 0.85 342 360 380
2 2 4 13 0.80 294 309 327
a
Mean=mean difference from baseline; first number “4” extrapolated from 9-question HRQL mean difference of 2.4 in TIF1 Phase 2 data.
b
The ratio B/C determines and power (trial’s change of success if all assumptions true) determines the sample size requirement.
c
Standard deviation also extrapolated from in TIF1 Phase 2 data.

1. Cadiere GB, Buset M, Muls V, et al. Antireflux transoral incisionless fundoplication using EsophyX: 12-month results of a prospective
multicenter study. World J Surg 2008;32:1676–88.

APPENDIX 2
Patient name:____________________
Date: ____________________
GERD-HRQL SCALE
0= No symptoms
1= Symptoms noticeable, but not bothersome
2= Symptoms noticeable and bothersome, but not every day
3= Symptoms bothersome everyday
4= Symptoms affect daily activities
5= Symptoms are incapacitating—unable to do daily activities

© 2015 by the American College of Gastroenterology The American Journal of GASTROENTEROLOGY

 542 Witteman et al.

Please circle the correct response to the following questions:

1. How bad is your heartburn? 0 1 2 3 4 5

ESOPHAGUS

2. Heartburn when lying down? 0 1 2 3 4 5

3. Heartburn when standing up? 0 1 2 3 4 5
4. Heartburn after meals? 0 1 2 3 4 5
5. Does heartburn change your diet? 0 1 2 3 4 5
6. Does heartburn wake you from sleep? 0 1 2 3 4 5
7. Do you have difficulty swallowing? 0 1 2 3 4 5
8. Do you have pain with swallowing? 0 1 2 3 4 5
9. Do you have gassy or bloating feelings? 0 1 2 3 4 5
10. If you take medication, does it affect your daily life? 0 1 2 3 4 5
How satisfied are you with your present condition? Satisfied Neutral Dissatisfied

APPENDIX 3
PPI management scheme
The use of proton pump inhibitors (PPIs) was recorded in medication diary using generic names listed in the table below and classified
according to the daily dose and frequency.

Proton pump inhibitors

Generic name Brand names Pill size (mg)

Esomeprazole Nexium 20, 40

Lansomeprazole Prevacid, Dakar, Lanso, Lanzor, Prezal, Lanzol 15, 30
Omeprazole Prilosec, Losec, Logastric 10, 20, 40
Pantoprazole Protonix, Zurcal, Pantozol, Zurcale 20, 40
Rabeprazole Aciphex, Pariet 20

Dosage will be categorized as: “Double dose” = ≥30 or 40 mg per day, “Full dose” = 30 or 40 mg per day, “Half dose” = 15 or 20 mg
per day, “Occasional” or “on demand” = < “full dose” taken for < 50% of days in the follow-up period.

PPI management algorithm

0. None
1. Half-a-single dose
2. PPI (previous effective) at single dose
3. PPI at double dose (b.i.d.) if symptoms are more severe than moderate despite two months’ therapy at single dose
4. Increase every 2 months by single dose to triple dose (2 q.a.m., 1 q.h.s.) and then quadruple dose (2 b.i.d.)

Reference (16): Lundell L, Miettinen P, Myrvold HE, et al. Continued (5-year) followup of a randomized clinical study comparing antire-
flux surgery and omeprazole in gastroesophageal reflux disease. J Am Coll Surg 2001;192:172–179; discussion 179–181.

The American Journal of GASTROENTEROLOGY VOLUME 110 | APRIL 2015 www.amjgastro.com

Reproduced with permission of the copyright owner. Further reproduction prohibited without
permission.