Background and aim: Self-expandable metal stents (SEMS) are used for non-resectable malignant gastric outlet obstruction (GOO). Studies of covered versus uncovered SEMS have yielded inconsistent results as a result of heterogeneity in design and patient population. We carried out a meta-analysis to compare covered and uncovered gastroduodenal SEMS.
Methods: Using MEDLINE, Embase, and the Cochrane database, we identified 1624 patients from 13 prospective and retrospective studies that evaluated covered and uncovered SEMS for malignant GOO and were published until October 2016. We pooled data on SEMS dysfunction, technical and clinical success, and adverse events using the fixed-effect or random-effects model.
Results: Compared with uncovered SEMS, covered SEMS did not show any significant difference in stent dysfunction risk (risk ratio [RR], 1.02; 95% confidence interval [CI], 0.79-1.32). A subgroup analysis of five randomized trials suggested a trend toward a lower dysfunction risk in covered SEMS (RR, 0.63; 95% CI, 0.45-0.88). Covered SEMS were associated with a lower occlusion risk (RR, 0.44; 95% CI, 0.28-0.68), but with a higher migration risk (RR, 4.28; 95% CI, 2.89-6.34). Technical and clinical success rates were comparable between the groups. Overall adverse events tended to be more frequent in covered SEMS (RR, 1.75; 95% CI, 1.09-2.83).
Conclusions: Outcomes of covered and uncovered gastroduodenal SEMS were comparable, although the lower dysfunction rate of covered SEMS observed in the analysis of randomized trials needs further investigation. Antimigration mechanisms for covered SEMS and identification of patients who can achieve longer patency from uncovered SEMS would help improve the outcomes of gastroduodenal SEMS.
Keywords: gastrointestinal endoscopy; pancreatic neoplasm; pyloric stenosis; self-expandable metallic stent; stomach neoplasm.
© 2016 Japan Gastroenterological Endoscopy Society.