Authors: Conchubhair Winters^1,2, Bu’ Hussain Hayee³, Noor Mohammed^1,2

¹Leeds Teaching Hospitals NHS Trust, Leeds

²University of Leeds

³King’s College Hospital NHS Trust, London

Keywords

1. Gastroparesis

2. G-POEM

3. Myotomy

4. Dilatation

5. Botox injection

Abbreviations

ACG – American College of Gastroenterologists

FD – Functional dyspepsia

GEBT – Gastric emptying breath test

GES – Gastric emptying studies

G-POEM – Gastric per Oral Endoscopic Myotomy

ICC – Interstitial cells of Cajal

NICE – National Institute for Health and Care Excellence

PPI – Proton Pump Inhibitor

WMC – Wireless Motility Capsule

Learning points

1. G-POEM can improve symptoms in appropriately selected patients with medically refractory Gastroparesis.

2. Patient selection is difficult but extremely important in predicting success. 

3. Gastroparesis can overlap with functional dyspepsia.

Introduction to Gastroparesis

Gastroparesis is derived from the Greek words gastro and paresis meaning paralysed stomach. The clinical condition of Gastroparesis is a result of delayed gastric emptying in the absence of a mechanical obstruction. Gastric function is complex and regulated by the interaction of the neurohormonal, myoelectric, and contractile properties of the stomach. The interstitial cells of Cajal (ICC) regulate gastric slow waves to produce gastric contractility. Post-prandial stretching of the gastric smooth muscle cells activate the ICCs, which leads to contraction. Relaxation of the pylorus in a co-ordinated fashion is necessary for gastric emptying. However, the pylorus contains less ICCs and is under the control of a complicated interaction of intrinsic and extrinsic innervation. Emptying of food is a complex process of antro-pyloro-duodenal motor activity. The vagal nerve and intestinal peptides also interact in modulating this process. Disruption to any of these elements can lead to impaired gastric emptying and gastroparesis.

Nausea, vomiting, early satiety, bloating, and abdominal pain are the most common symptoms. The symptoms can be severe and disabling, impacting quality of life, and causing malnutrition. While nausea and vomiting symptoms correlate weakly with the severity of delayed gastric emptying, abdominal pain, which is present in 90% of patients, does not.^{[1, 2]} There is a significant overlap with functional upper gastrointestinal disorders. Recent evidence suggests that gastroparesis and functional dyspepsia are spectrums of the same disorder, with gastroparesis the diagnosis given to those with a greater element of motor dysfunction.^[3] The diagnosis of gastroparesis requires objective evidence of delayed gastric emptying using scintigraphy or a breath test.

The condition is more common in females. UK epidemiological data reports several aetiologies for gastroparesis: idiopathic (39%), diabetic (38%), and drug-induced (20%), with the remaining cases a combination of post-surgical, neurological and connective tissue diseases.^[4] A systematic review by Dilmaghani et al. found that prevalence is increasing, healthcare utilisation is increasing, and all-cause mortality is increasing in gastroparesis.^[5]

Investigation of Gastroparesis

The work up for anyone with nausea, vomiting and abdominal pain should begin with a detailed history and clinical examinations. Gastroscopy or imaging should be used to exclude a gastric outlet obstruction. Retained gastric food at endoscopy can point towards, but should not be used to diagnose gastroparesis due to a poor positive predictive value of only 55%.^[6] Explore the patient’s drug history to ensure they are not on any medication which could be impairing gastric emptying. Drugs which could impair gastric emptying include opioids, tricyclic antidepressants, calcium-channel blockers, clonidine, dopamine agonists, lithium, nicotine, or progesterone.

Objective evidence of impaired gastric emptying is needed using gastric emptying studies or stable-isotope gastric emptying breath test (GEBT) which correlate well with each other ^[7]. Wireless motility capsule (WMC) has previously been used, but is no longer available. Gastric emptying studies (GES) is the most widely available test in the UK. A standard protocol for GES involves the consumption of a 250kcal radionucleotide labelled meal, then scintigraphy images every 60 minutes for 4 hours. Retention of >10% at 4 hours or >60% at 2 hours suggests delayed gastric emptying. Where this standardised approach is not used, the centre should know their normal values. It is important that any medications which could alter gastric emptying are stopped for at least 48 hours. These drugs include those which impair gastric emptying as listed above, but also drugs that aid gastric emptying such as prokinetic antiemetics, erythromycin or prucalopride. Cannabinoids and elevated blood sugars can also impair results. GEBT should be avoided if there is severe lung, liver, pancreatic or small bowel disease or in the cases of allergies to egg, milk, wheat or Spirulina.

The Gastroparesis Cardinal Symptom Index (GCSI) is a validated symptom score which attempts to score symptoms to decide who has severe symptoms and would benefit from invasive management. GCSI can also help in tracking response to treatment.^[8] For each symptom patients are asked to score how severe their symptoms have been in the last two weeks. The score is usually presented as a mean of the symptom scores. An average score of >2.3 is classed as severe symptoms.

Table 1. Gastroparesis Cardinal Symptom Index.

 Gastric electrophysiological studies, Body Surface Gastric Mapping (BSGM) may help in differentiating gastroparesis phenotypes which will respond to pyloric interventions but is still predominantly used in research.

EndoFLIP uses a solid state endoscopically inserted catheter to measure the pressure across the pylorus, providing a topographical read out of the distensibility and diameter of the lumen. EndoFLIP is showing promise in predicting who will respond to  pyloric interventions, with pyloric distensibility index >8–10 mm²/mm Hg being unlikely to respond^[9].

Management of Gastroparesis

Gastroparesis is a chronic disorder with no cure. Patient education is important. Management is symptomatic. Patient discussions should outline the risk and benefits of each management option, while managing expectations. Dietary and medical management should be explored before more invasive options. Anxiety as a confounder should be explored and appropriately managed. Medical management fails to control symptoms adequately in up to 40%^[10]. More invasive management options can be considered in those with chronic (>6 months) medically refractory severe symptoms (GCSI >2.3). The most appropriate invasive management option is unclear, with the choice often coming down to local availability and expertise.

Diet

Dietary management involves small frequent meals (4-6/day) of a small particle low fat diet^[11]. Diabetic gastroparesis adds an additional level of complexity due to the risk of hypoglycaemia or diabetic ketoacidosis. Dieticians should be involved if there are nutritional concerns, you are struggling with blood sugar control or if patients are following increasingly restrictive diets. Murray et al found many gastroparesis patients on highly restrictive diets had symptoms of an eating disorder. These patients showed greater gastrointestinal symptoms scores but no association with severity of delayed emptying.^[12] Care should be taken when giving dietary advise. Jejunal feeding, venting gastrostomies and parental feeding may be required in refractory cases.

Medical Management

Prokinetic dopamine receptor antagonists, such as metoclopramide, have been shown to improve gastric emptying^[13]. With metoclopramide, there are concerns about tardive dyskinesia and extrapyramidal side effects in long-term use. Domperidone is an alternative dopamine receptor antagonist which does not cross the blood-brain barrier but can prolong the QTc and cause potentially life-threatening arrhythmias. A network meta-analysis recently found that, other than cisapride which is no longer on the market, domperidone was the only other drug that was superior to placebo in reducing global symptoms.^[14] Neurokinin-1-receptor antagonists, such as aprepitant, and oral metoclopramide showed benefit in pooled and subgroup analysis. Other antiemetics and neuromodulators are often used but the evidence base for their effectiveness is limited. In subgroup analysis, none of the medications included were better than placebo at reducing symptoms in diabetic gastroparesis.^[14] Erythromycin, a macrolide antibiotic, is a motilin receptor agonist which promotes gastric emptying and was NICE approved for off-license use in diabetic gastroparesis.^[15] Tachyphylaxis occurs with Erythromycin, so courses should be limited to 4 weeks. 5-HT₄ agonists, such as prucalopride, have shown some promising results in small studies, however again there is tachyphylaxis but this seems to reset after a few weeks drug break^[16].  

Pain management in gastroparesis is problematic. There are no clinical trials addressing this question, and opiate analgesia can further impair gastric emptying. Given the significant overlap with functional dyspepsia, we would recommend following the BSG guidelines for the management of functional dyspepsia.^[3]

Optimising glycaemic control in diabetic gastroparesis is recommended, but the evidence base for an improvement in gastroparesis symptoms is weak.^[6]

Gastric Electrical Stimulators

Gastric electric stimulators have been controversial. Randomised controlled trials to date suggest improvement in nausea and vomiting but no improvement in abdominal pain, gastric emptying or quality of life.^[6] When compared to G-POEM in a propensity-matched analysis, G-POEM showed better long-term efficacy.^[17] Studies looking at the combination of a surgical pyloroplasty and gastric electrical stimulator have had some success but further studies looking at G-POEM and electrical stimulators are needed. If available, gastric electrical stimulator as a single modality should be reserved for those with nausea and vomiting predominant symptomologies.

Surgical Management

Roux-en-Y, partial gastrectomy, pyloroplasty and surgical jejunostomy have previously been used in gastroparesis, but they are more invasive than endoscopic options with increased risk of morbidity. A meta-analysis comparing G-POEM to surgical pyloroplasty found no difference in the clinical success rate based on GCSI or GES, or in the adverse event rate.^[18]

Endoscopic Management Options in Gastroparesis

Endoscopic management options in gastroparesis are aimed at attenuating pyloric tone to improve gastric emptying. Endoscopic management is unable to alter other aspects of gastroparesis, such as impaired fundic accommodation, antral contractility, or antro-duodenal reflux. Research is ongoing, using devices such as Endoflip and BSGM, attempting to identify patients with impaired pyloric relaxation who may benefit from pylorus-directed management.

Botox

Intrapyloric injection of Botulinum Toxin (Botox) has shown variable results in clinical trials. Botox failed to show a statistically significant improvement in two small randomised controlled trials and hence has not been suggested by national guidelines such as the ACG^{[19, 20]}. Four other small prospective studies to date have shown improvements in GES and GCSI^[21-24]. Botox has historically been avoided when G-POEM is being considered due to the submucosal fibrosis it can cause adding complexity to the procedure. Desprez et al. found that a distensibility index of <10 mm2/mmHg measured using EndoFLIP could predict those patients who would respond to intrapyloric Botox injection^[25]. No response to Botox is a poor predictor of response to other endoscopic therapies such as G-POEM. Eriksson et al. when comparing surgical myotomy and G-POEM found that Botox increased the likelihood of a clinically successful myotomy, irrespective of the Botox response.^[26] They postulated that the benefit was a result of reduced myelofibroblast production leading to less scar tissue formation post myotomy.

Botox is most commonly used as a bridge or short-term management option. The duration of action of Botox is variable but usually a few months, and it is known to exhibit tachyphylaxis making it a poor long-term management.

Pneumatic dilatation

Pneumatic dilatation involves endoscopic balloon dilatation to diameters or 20-30mmHg. The evidence base is limited with retrospective studies reporting clinical response rates of 50% but diminishing to 1/3 at 2 years.^[10] A more recent EndoFLIP guided pyloric dilatation study reported improved symptom scores in 57% at 4 months following dilatation.^[27]The risk of adverse events appear to be low in the limited evidence available. Pneumatic dilatation could be considered if therapies with increased efficacy (e.g. G-POEM) are not available or appropriate.

Pyloric Stenting

Transpyloric stenting showed promising symptomatic responses of 75% but was limited by the high stent migration rate of 59%. Even when the researchers endoscopically sutured the stent it migrated in half of cases.^[28] Transpyloric stenting should be reserved as a short term management option.

G-POEM

The first gastric per oral endoscopic myotomy (G-POEM) was published by Khashab in 2013.^[29] G-POEM uses a similar submucosal tunnelling and myotomy technique as used in oesophageal per oral endoscopic myotomy. G-POEM should be considered in patients with medically refractory gastroparesis. G-POEM has been shown to be efficacious in improving symptoms and gastric emptying in a range of gastroparesis aetiologies with varying success rates. ^[30] Post-infectious gastroparesis should not routinely be considered for G-POEM.^[31] Patients on regular opiates should have them weaned off and gastric emptying studies repeated before considering G-POEM.

Patient Selection

Patient selection for G-POEM is important and non-selective use of G-POEM should be discouraged. Significant overlap with functional dyspepsia means many FD patients will score high on the GCSI. We recommend G-POEM is offered for severe symptoms which usually correspond to a GCSI score of >2.3, with higher scores predicting response.^[32] Vomiting predominant symptomatology's are more likely to respond to G-POEM than patients with abdominal pain predominant symptoms. Although a retention of >10% at 4 hours on GES is abnormal, the likelihood of a response to G-POEM is higher in those with >20% retention at 4 hours.^[32]

Table 2. Patients who are more likely to respond to pylorus directed therapy.

G-POEM Procedure

Patients should be carefully counselled about the risk of adverse events, which meta-analysis data suggest occurring in 8-12% of cases.^{[33, 34]} The most common adverse event is bleeding, with abdominal pain and gastric ulceration also being common. Tunnel leak, perforation and capnoperitoneum are less common, but the patient should be advised of the possibility of a needle paracentesis for decompression of leaked carbon dioxide. The patient will be asked to consume a liquid only diet for at least 24 hours before the procedure to reduce food retention within the stomach.

G-POEM is performed under general anaesthesia and takes approximately 60 minutes to perform. After injection of a lifting agent, a 1.5-2cm mucosal incision is made in the antrum, usually on the greater curve, 4-5cm from the pylorus. A tunnel is created within the submucosal space until the pyloric ring is exposed. Careful myotomy of the pyloric ring is performed, usually as an incision of 2-3cm length. At a minimum the inner circular layer of the muscularis propria should be cut, but a full thickness myotomy of the circular and oblique muscle layers is recommended, leaving the serosa unbreeched. The mucosal incision tunnel entry site is closed using endoscopic clips or endoscopic suturing.

Figure 1. Endoscopic images of a G-POEM. a: mucosal incision, b: access to submucosal created by widening the mucosal flap, c: submucosal tunnelling, d: pyloric ring exposed, e: myotomy completed, f: endoscopic suturing of mucosal flap. 

Where the tunnel closure is secure, such as with endoscopic suturing, many centres are discharging patients the same day. Post-procedural dietary advice varies, but many advise a liquid diet for 5-7 days before building up to a regular small meal gastroparesis diet. A course of high dose PPI therapy is often started to aid mucosal healing.

Most physician's follow-up patients with a repeat GCSI, and some with GES, in the first few months and then again between 6 and 12 months.

Clinical Efficacy

Technical success is reported in nearly 100% of cases. Clinical success can be defined as an improvement in the total GCSI of ≥1 or a 25% improvement in at least two sections of the GCSI. Several meta-analyses have reported short term clinical success rates of approximately 80% in the first year.^{[18, 33]} The only meta-analysis to look at longer term outcomes suggested a 61% clinical success rate at 12 months.^[34] Up until recently we only had observational data. In 2022, a randomised controlled trial comparing G-POEM to sham was stopped early due to superiority of G-POEM over sham, with clinical success rates (improvement in GCSI by at least 50%) of 71% vs 22% (p=0.001).^[30]

One single centre retrospective case series of 97 patients found a 13% loss of response per year in those who initially responded.^[35] Redo G-POEM is challenging due to significant fibrosis but has been reported. Initial failure to respond despite an adequate myotomy (often EndoFLIP guided) suggests further pyloric intervention is not indicated.

Predictors of clinical success

Several studies and meta-analysis have attempted to identify predictors of clinical success to G-POEM. Nausea and vomiting predominant symptoms scores, diabetic gastroparesis, higher retentions scores at GES, higher GCSI scores and shorter duration since diagnosis of gastroparesis have been highlighted by several as predictors of clinical success.^{[18, 30, 32, 35, 36]}

Summary

Gastroparesis can be a severe and debilitating condition. The aetiology is varied, and the pathophysiology remains poorly understood. Dietary and medical management remain the mainstay of treatment. Pylorus targeting endoscopic therapies have a role to play in selected patients with refractory gastroparesis. G-POEM has the potential to improve symptoms and gastric emptying in gastroparesis. However, more research is required to help identify those patients most likely to benefit from pylorus directed therapies.

Author Biographies

Dr Conchubhair Winters

Dr Conchubhair Winters completed his Gastroenterology training in Yorkshire and Humbre. Dr Winters is completing an MD at the University of Leeds on barriers to colonoscopy and robotic colonoscopy. Following completion of a post-CCT fellowship in submucosal endoscopy he is currently working as a Consultant Gastroenterologist in Northern Care Alliance. He has a special interest in luminal therapeutic endoscopy.

Dr Noor Mohammed

Dr Noor Mohammed is a consultant Gastroenterologist and interventional endoscopist at Leeds Teaching Hospitals NHS trust. He is the training lead for Yorkshire Endoscopy Training Academy and Leeds Teaching Hospitals NHS Trust. 

His research degree (MD) was in use of advanced endoscopic modalities such as Chromoendoscopy and NBI in IBD. During his advanced endoscopy fellowship in Leeds, he developed an interest in submucosal endoscopy. He underwent training in ESD under the proctorship of Prof N Yahagi, Keio University, Tokyo, Japan. He has successfully set up an ESD service in Leeds since January 2016. He also completed training in POEM by Prof H Inoue, Showa University Koto-Toyosu, Tokyo, Japan; and has set up a POEM service in Leeds since 2018. He now seems to spend more time in submucosa than in the lumen!

He has successfully set up the UK’s first ESD academy since 2021 and has successfully trained and proctored several endoscopists in the UK. He is involved in guideline review/update/writings with BSG.

Dr Bu’ Hussain Hayee