Introduction
Gastroparesis is characterised by an association with upper gastrointestinal symptoms
and objective delayed gastric emptying in the absence of mechanical obstruction [1]. It occurs in 1.8 % of the population [2], with a higher prevalence in women. Symptoms include early satiety, postprandial
fullness, nausea, vomiting, and upper abdominal pain. These symptoms are essentially
induced by eating and often lead to an altered quality of life [3] or nutritional deficiencies. The gold standard for the diagnosis of gastroparesis
is gastric emptying scintigraphy, performed after the consumption of a standardised
radiolabelled meal [4].
Gastroparesis is most often idiopathic, but it can also develop in the setting of
diabetes or postoperatively due to vagal damage during surgery [5]. Most treatments of gastroparesis, such as prokinetics, target the symptoms and
thus are frequently ineffective. Surgical therapies include gastric electric stimulation,
but evidence for the efficacy of this approach is lacking. Recent manometric studies
have shown that pyloric pressure, including phasic and tonic contractions, is elevated
in patients with gastroparesis [6]
[7]. This phenomenon, termed pylorospasm, has resulted in new therapeutic approaches
in patients who develop gastroparesis after surgical pyloroplasty [8]. Inspired by the success of peroral endoscopic myotomy (POEM) [9] in patients with achalasia, the use of peroral pyloromyotomy (POP) in a patient
with refractory gastroparesis was first described in 2013 [10]. However, POP is difficult to perform, especially by physicians without experience
in POEM. Here, we describe a porcine model suitable for training physicians in POP.
We also explore the effect of POP on gastric emptying in healthy pigs.
Material and methods
Animals
Four Landrace mini-pigs (three castrated males and one female) with an average weight
of 25.5 kg (22 – 29 kg) were used in this study. The study was approved by the Institutional
Animal Care and Use Committee. All pigs underwent a first gastric emptying scintigraphy
under light sedation. Two days later, POP was performed on each pig ([Fig. 1]). For organisational reasons, a second gastric emptying was carried out 5 days after
the endoscopic procedure to evaluate its impact on gastric emptying. Finally, the
pigs were euthanised after 10 days of follow-up to allow pathological assessment of
the pylorus. Duration of the follow-up was determined according to economic purposes.
Fig. 1 Peroral pyloromyotomy. a, pyloric ring; b, submucosa of the bulb; c, mucosa of the
bulb.
Gastric emptying scintigraphy
The pigs were fasted for 24 hours and then lightly sedated using intramuscular ketamine
(6 g). A laryngoscope was used to establish a nasogastric feeding tube to administer
a meal consisting of one scrambled egg mixed with 68 MBq Tc99-m, 125 g of pellet food
and 80 mL of water. Before the scintigraphy was started, the pigs were lightly anaesthetised
with intravenous propofol at a low dose (5 mg/kg/h) to avoid an impact of the drug
on gastric emptying. The pigs were then placed in a natural position in front of the
camera (Axis Model, Philips Medical). Data were acquired for 90 min. The half emptying
time (T½) is expressed in minutes.
Endoscopic procedure
Two days after scintigraphy, POP was performed as described previously [11]. All procedures were carried out with the animals under general anaesthesia using
intravenous propofol (200 mg/h) after premedication with ketamine (6 g) and endotracheal
intubation. A gastroscope (Karl Stortz 13821PKS, Karl Stortz Endoscopy, Tuttlingen,
Germany) with a distal cap attached (conical cap DH-28GR, Fujifilm, Japan) was used
to define a mucosotomy site 4 cm above the pylorus, on the anterior wall of the great
curvature. After the submucosal injection of a saline solution containing indigo carmine,
a longitudinal mucosal incision was made using a T-type HybridKnife (ERBE Medical,
Tübingen, Germany), assisted by air insufflation and using the Endocut-mode (cut duration:
2; effect: 2; cut interval: 2) via an electrosurgical generator (VIO 300D; ERBE Medical,
Germany). Submucosal tunnel dissection using the SWIFT COAG setting (50 W, effect:
4) revealed the pylorus. After identification of the pyloric muscle, its circumferential
muscular fibres were progressively cut using the HybridKnife or a HookKnife (Olympus,
Tokyo, Japan), until cutting was considered to be complete. Myotomy was considered
endoscopically complete when the full white pyloric ring was cut with visualisation
of the muscle circular layer of the duodenum.
Finally, the tunnel entry was closed using haemoclips (Boston Resolution, Boston Scientific,
Natick, Massachusetts, United States). In the only case of perforation, antibiotics
were injected intraperitoneally during surgery and then again 4 hours later.
Statistical analysis
A descriptive statistical analysis of the data using means ± SD was employed. Comparison
of linked samples was performed using the Student test for paired data. Fisher’s exact
test was used to compare qualitative data. A P value less than 0.05 was considered to indicate statistical significance.
Feasibility
Endoscopic pyloromyotomy was technically possible in all four pigs (feasibility: 100 %).
The pylorus muscle and its thick white fibres were readily identified in each pig
([Table 1]).
Table 1
Results for peroral endoscopic pylorotomy and gastric emptying.
|
Pig 1
|
Pig 2
|
Pig 3
|
Pig 4
|
Mean
|
|
Procedure duration, min
|
55
|
50
|
55
|
60
|
55 (± 4.08 SD)
|
|
Complications, n
|
|
|
|
|
|
|
Perforation
|
1
|
0
|
0
|
0
|
25 %
|
|
Bleeding
|
0
|
0
|
0
|
0
|
0 %
|
|
Anticipated sacrifice
|
0
|
0
|
0
|
0
|
0 %
|
|
T½ gastric emptying, min
|
|
|
|
|
|
|
Before POP
|
301.89
|
89
|
180
|
183
|
188.47 (± 87.3 SD)
|
|
After POP
|
125
|
38
|
85
|
90
|
84.5 (± 35.74 SD) P = 0.03 (CI 95 %: 20.15 – 187.8)
|
|
Technical difficulty
|
Loop + + +
|
Loop +
|
Loop + +
|
Loop + +
|
|
|
Feasibility, n
|
1
|
1
|
1
|
1
|
100 %
|
|
Pathological analysis
|
|
|
|
|
|
|
% section of pyloric muscle, %
|
> 90
|
> 90
|
> 90
|
> 90
|
> 90
|
POP, peroral endoscopic pylorotomy.
The average duration of the pyloromyotomy was 55 (± 4 SD) minutes. No bleeding occurred
during or after the procedure. A perforation occurred in the first pig during the
myotomy phase, but the endoscopic procedure was completed successfully. The perforation
was treated with intraperitoneal injection of delayed action amoxicillin (25 mg/kg)
administered immediately, as defined in the study protocol. There were no fatalities
during the procedure or during the 10 days of follow-up.
The mean weight of the animals was 25.5 kg before and 24 kg after the POP. During
follow-up, no sign of suffering or a change in behaviour was noted.
Gastric emptying
The mean duration of gastric emptying (T½) in the four pigs before POP was 188.47 minutes (± 87.03 SD). After POP, the T½ was 2.2-fold shorter, with an average duration of 84.5 minutes (± 35.74 SD). This
difference was statistically significant (P = 0.029). The duration of gastric emptying in each of the four pigs before and after
POP is shown in [Fig. 2].
Fig. 2 Individual results for gastric emptying scintigraphy showing T½ values before and after peroral endoscopic pylorotomy (POP).
Histology
Necropsy confirmed the only perforation that developed during endoscopy as a pneumoperitoneum
without peritonitis.
In agreement with the endoscopic observations, sectioning of the pyloric muscle in
each pig was histologically complete (> 90 % of the length and thickness of the muscle).
An example of pathologic assessment of the pylorus is shown in [Fig. 3].
Fig. 3 Pathologic assessment of a complete section of the pylorus (horizontal axis and haematoxylin
and eosin (H&E) staining): 1, normal gastric muscularis propria; 2, granuloma replacing
the submucosa (tunnel dissection); 3, normal gastric mucosa; 4, fibrosis replacing
the pyloric muscle (area of the pyloromyotomy); 5, serosa and subserosa.
Discussion
Peroral endoscopic pyloromyotomy is a new functional technique that evolved from endoscopic
submucosal dissection (ESD) and POEM. In Europe, the anaesthetised pig is considered
to be a valid training model and recommended for ESD learning. Our study demonstrates
that the same model can be adapted for POP learning and development. The endoscopists
who participated in this study had no previous experience with either POP or POEM,
but each had already conducted 60 ESDs in animal models and 30 ESDs in human patients.
Thus, even among these inexperienced operators, the feasibility of POP was 100 %.
The duration of the procedure was typically 55 minutes. Because the anatomy and histology
of pigs and humans are similar, this porcine model not only allows training in this
technique but also familiarises physicians with the muscular fibrous section of the
pylorus. The success of this learning approach was demonstrated by the ability of
operators with experience only in ESD to identify the muscular fibres easily during
each procedure. This was confirmed by histologic analysis, which verified the completeness
of the myotomy in all four pigs.
These encouraging results on animals recommend this model for surgeons who plan to
operate on humans. In fact, endoscopy is more difficult in pigs than in humans because
of differences in their gastric anatomy. The porcine stomach is U-shaped, resulting
in higher pressure on the gastroscope, and it has larger gastric loops. Seen in retrovision,
the pylorus is positioned next to the cardia, which renders POP more difficult. The
J-shaped stomach of humans should facilitate the procedure.
The safety of POP was indicated by the occurrence of a perforation in only one pig,
which was uncomplicated. The perforation occurred during endoscopy and was inconsequential
because of the establishment of a submucosal dissection tunnel, which allowed the
perforation site to be isolated from the gastric secretions. However, this complication
was due to the many gastric loops in the U-shaped porcine stomach. In humans, the
systematic use of CO2 insufflation and the J-shape of the stomach would minimise the risk of perforation
during the procedure.
There is currently no animal model validated for the evaluation of endoscopic or surgical
procedures aimed to treat gastroparesis. The simplicity of our study protocol, combined
with the availability of a gastric emptying study, highlight the feasibility of using
mini-pigs as an animal model. Our study therefore provides a proof of concept of the
beneficial effect of POP on gastric emptying, even in healthy pigs. Moreover, it is
the first study to use scintigraphy for thorough evaluation of the procedure.
In 2011, in the first animal study on POP, Kawai et al. [12] used manometry to assess its effects on pyloric pressure, but they were unable to
evaluate the impact on gastric emptying. We determined that the T½ of gastric emptying in healthy pigs was accelerated more than twofold after POP,
confirming the efficiency of the procedure.
The variation of gastric emptying duration between each pig before and after POP could
be compared with healthy humans. Indeed a wide inter-individual variation has been
described in pivotal studies. But, as it is the first study using gastric emptying
scintigraphy in pigs, there is no reference concerning these data [13].
To better characterise and understand the mechanism underlying the efficacy of POP,
we are currently working with a porcine model of gastroparesis after surgical vagotomy.
By examining gastric emptying under these conditions, we will be able to assess the
efficiency not only of POP but also of other endoscopic/surgical procedures.
Pylorospasm is a common finding in patients with gastroparesis [7]; thus, POP may be the treatment of choice for refractory gastroparesis.
POP could also represent an alternative procedure in hypertrophic pyloric stenosis
if suppliers are able to miniaturize their devices for use in infants (operative channel
of paediatric gastroscopes: 2 mm).
Conclusion
A twofold decrease in T½ of gastric emptying was achieved after POP, as verified by gastric emptying scintigraphy.
The efficacy of the procedure provides indirect proof of the involvement of the pyloric
ring in delayed gastric emptying and suggests new therapies for patients with gastroparesis.
Our protocol combining gastric emptying scintigraphy and POP validated the use of
anaesthetised mini-pigs as a learning and training model for POP or other endoscopic/surgical
procedures related to gastric emptying. Pathologic confirmation of pyloric sectioning,
the low complication rate, and the absence of mortality demonstrate the feasibility
and safety of POP, even when performed by physicians without experience in POEM.
A prospective study on gastroparetic pigs after surgical vagotomy is ongoing in parallel
with prospective work on patients with refractory gastroparesis and a significantly
impaired quality of life.
