Background and aims: Gastric peroral endoscopic myotomy (G-POEM) has been regarded as a novel and minimally invasive therapy for refractory gastroparesis. This study reports the long-term outcomes and possible predictive factors for successful outcomes after G-POEM in an Asian population.
Methods: This is a retrospective single-centre study of 16 patients who underwent G-POEM for refractory gastroparesis from August 2016 to October 2017. This study included 11 males and 5 females; in addition, 13 patients had postsurgical gastroparesis, and 3 patients had diabetes. The patients included had severe and refractory gastroparesis, as indicated by a Gastroparesis Cardinal Symptom Index (GCSI) score ≥20, and evidence of a delay on gastric emptying scintigraphy (GES). The primary outcome parameter was an assessment of the long-term clinical efficacy of the procedure. The secondary outcome parameter was the detection of possible predictive factors for success and the determination of cut-off values for such predictors.
Results: Technical success was achieved in 100% of the patients, with a mean procedure time of 45.25±12.96 min. The long-term clinical response was assessed in all patients during a median follow-up of 14.5 months. Clinical success was achieved in 13 (81.25%) patients. There was a significant reduction in the GCSI scores and GES values after the procedure compared to the baseline values, with P values of <0.0001 and 0.012, respectively. Univariate regression analysis showed that the GCSI and GES had significant associations with the future clinical outcomes of the patients, but this finding was not confirmed in multivariate analysis. A GCSI cut-off score of ≤30 had a high sensitivity and a negative predictive value (NPV) of 100% for predicting a successful procedure. GES (half emptying time ≤221.6 min and 2-hour retention ≤78.6%) had a high specificity and a positive predictive value (PPV) of 100%.
Conclusions: G-POEM is a safe and effective treatment option with a long-term efficacy of 81.6%. GCSI and GES could serve as good predictive measures.