Endoscopic Pyloromyotomy for Refractory Gastroparesis

Gastroparesis is a disorder triggered by numerous causes and it is defined by symptoms and with an objective evidence of delayed gastric emptying in the absence of obstruction (albeit pyloric spasms may play a role in a subset of patients). Gastroparesis may be a consequence of medication, surgery or diabetes but in approximately one third of patients, the cause remains unknown and the patients are diagnosed with idiopathic gastroparesis. Effective treatment for gastroparesis is challenging especially in patients with severe symptoms. The efficacy of prokinetics is dubious since they have not proven real clinical efficacy in placebo controlled trials. In refractory gastroparesis, endoscopic or surgical treatments may therefore be considered. Endoscopic treatments include intrapyloric injection of botulinum toxin and transpyloric insertion of a metallic stent. Surgical options involve implantation of a gastric "pacemaker" (gastric stimulation), pyloroplasty and subtotal gastrectomy. The partial effectiveness of botulinum toxin injection, stents and pyloroplasty suggests that disruption of the pyloric muscle may lead to a decreased intrapyloric tone and consequently to a symptomatic improvement in some patients with refractory gastroparesis.

Recently, a new endoscopic technique, gastric endoscopic per oral pyloromyotomy (G-POEM) has been introduced with promising preliminary results. Uncontrolled studies with so far limited number of patients have demonstrated a significant symptomatic improvement in approximately 70% of patients and improved or normalized of gastric emptying in more than a half of patients after G-POEM. A prospective uncontrolled study suggested that patients with idiopathic or post-surgical gastroparesis experiences higher success rate after G-POEM (70-80%) compared to patients with diabetic gastroparesis (50%).

G-POEM is, in principle, adaptation of POEM (per-oral endoscopic myotomy) in the stomach. POEM is now considered a standard treatment for esophageal achalasia and it has been shown to be safe and effective. In contrast to achalasia, pathophysiology of pyloric function in patients with gastroparesis is less understood and the explanation of how and why G-POEM should work is some-how hypothetical. For example, presumed pylorospasm has not been demonstrated as the predictive factor for treatment success of G-POEM yet. Refractory gastroparesis is often accompanied by psychological or even psychiatric disturbances and hence a placebo" effect of G-POEM cannot be ruled out. Therefore, the real clinical efficacy of G-POEM can only be demonstrated in a clinical randomized sham-controlled trial.

To assess the severity of gastroparesis-related symptoms, the Gastroparesis Cardinal Symptom Index (GCSI) has been developed for this item. The GCSI is part of a larger questionnaire PAGI-SYM (Patient Assessment of Upper Gastrointestinal Symptom severity index) established for assessment of patient-reported symptoms in gastroparesis (dyspepsia and gastroesophageal reflux). PAGI-SYM as well as GCSI subscale scores varied significantly by global disease severity, with higher (worse) scores observed in those subjects who rated their gastroparesis as moderate to severe.

The aim of this prospective, sham-controlled, cross-over study (cross-over for patients randomized to the sham arm) is to compare short and long-term efficacy and safety of G-POEM in patients with refractory gastroparesis. Symptoms and objective parameters of gastric emptying will be the main outcome criteria. The reason of using a sham protocol is to control for the potential confounders (therapeutic effects of touch and belief, which are components of the placebo effect).

Investigators plan to randomize 86 patients (43 in the active arm, ratio 1:1 active vs. sham). Sample size is calculated based on expected therapeutic success of G-POEM in 50% of patients vs. 20% in the sham group; significance level 0,05; study power 0,8; beta error 0,2; adjustment for 15% expected drop out.

Patients will be randomised in blocks of 6, stratified according to the etiologies: (idiopathic, diabetic, and post-surgical; patients after esophagectomy with gastric pull-through will not be included). Control visits will be scheduled at 3, 6, 12, 24, and 36 months. The primary outcome will be the proportion of patients with treatment success in the active group vs. sham group at 6 months after the procedure. Several secondary outcomes will also be assessed, including procedure-related parameters and safety parameters and change in Gastric Emptying Study (GET) after G-POEM vs. sham. After 6 months, patients randomized to the sham group will be offered G-POEM procedure and further followed up (cross-over part of the study) providing that they did not have a therapeutic effect of the sham procedure.