Asokkumar, Ravishankar

Abstract Background Gastric variceal bleeding carries significant mortality in the setting of portal hypertension. Among the endoscopic treatment options, endoscopic ultrasound (EUS)-guided glue and/or coil injection is a novel approach, but its role in the treatment of gastric varices is not established due to a lack of robust data. Methods We conducted a comprehensive search of several databases (inception to June 2019) to identify studies evaluating EUS in the treatment of gastric varices. Our primary goals were to estimate the pooled rates of treatment efficacy, obliteration and recurrence of gastric varices, early and late rebleeding, and adverse events with EUS-guided therapy in gastric varices. We also searched for studies that evaluated direct endoscopic glue (END-glue) injection for treatment of gastric varices, and used the pooled rates as comparators. Results 23 studies (851 patients) evaluating EUS-guided therapy were included. The pooled treatment efficacy was 93.7 % (95 % confidence interval [CI] 89.5 – 96.3, I 2 = 53.7), gastric varices obliteration was 84.4 % (95 %CI 74.8 – 90.9, I 2 = 77), gastric varices recurrence was 9.1 % (95 %CI 5.2 – 15.7, I 2 = 32), early rebleeding was 7.0 % (95 %CI 4.6 – 10.7, I 2 = 0), and late rebleeding was 11.6 % (95 %CI 8.8 – 15.1, I 2 = 22). The rates were comparable to END-glue therapy (28 studies, 3467 patients) except for obliteration, which was significantly better with EUS-guided therapy. On subgroup analysis, EUS-coil/glue combination showed superior outcomes. Conclusions EUS-guided therapy demonstrated clinical efficacy for treatment of gastric varices in terms of obliteration, recurrence, and long-term rebleeding, and may be superior to END-glue.

INTRODUCTION: Options to treat medically refractory gastroparesis are limited. Gastric peroral endoscopic myotomy (G-POEM) is a novel minimally invasive technique in endosurgery that has shown promising results. Data is limited as to its efficacy, safety, predictive factors and its comparability to surgical pyloroplasty. The aim of our study was to synthesize and analyze the clinical outcomes of GPOEM by meta-analysis methods. METHODS: We searched multiple databases from inception through March 2019 to identify studies that reported on G-POEM and surgical pyloroplasty in gastroparesis. Our primary outcome was to analyze the pooled rates of gastroparesis cardinal symptom index (GCSI) score and 4-hr gastric emptying study (GES) results, with G-POEM. Similar outcomes with surgical pyloroplasty was used as a comparator group. RESULTS: 332 and 375 patients underwent G-POEM (11studies) and pyloroplasty (7studies), respectively. The pooled rate of clinical success, based on the GCSI score, with G-POEM was 75.8% (95% CI 68.1-82.1, I 2 = 50) and with pyloroplasty was 77.3% (95% CI 66.4-85.4, I 2 = 0), with no significance, P = 0.81. The pooled rate of clinical success, based on the 4-hour GES results, with GPOEM was 85.1% (95% CI 68.9-93.7, I 2 = 74) and with pyloroplasty was 84% (95% CI 64.4-93.8, I 2 = 81), with no significance, P = 0.91. The pooled rate of mean pre-G-POEM GCSI score was 3.4 (95% CI 3.1-3.7, I 2 = 89) and post-G-POEM GCSI score was 1.8 (95% CI 1.4-2.1, I 2 = 89), with statistical significance ( P = 0.001). The pooled pre-G-POEM 4-hr GES was 49.9% (95% CI 41.6-58.3, I 2 = 94) and post-G-POEM 4-hr GES was 20.6% (95% CI 12.3-28.8, I 2 = 77), with statistical significance ( P = 0.001). The pre and post pyloroplasty 4-hr GES results were 36% (95% CI 32.7-39.3, I 2 = 88) and 3.9% (95% CI 2.3-5.5, I 2 = 43), respectively ( P = 0.001). The pre and post pyloroplasty GCSI score was not analyzed due to paucity in data. The overall adverse events were comparable (11%, 5.9-19.8, I 2 = 66 vs 11.4%, 5-24, I 2 = 69, P = 0.95). Based on meta-regression analysis, idiopathic-gastroparesis, prior treatment with botulinum toxin and gastric-stimulator seemed to predict clinical success with G-POEM. CONCLUSION: G-POEM demonstrates clinical success in treating refractory gastroparesis with comparable outcomes to pyloroplasty. Idiopathic gastroparesis, prior treatment with botulinum injection and gastric stimulator appear to have positive predictive effects on the 4-hr GES results after G-POEM.