Introduction Management of dysphagia due to malignant obstruction of the oesophagus can be challenging. Endoscopically placed self-expanding metal stents (SEMS) remain an important tool and have been performed under direct vision in our institution since 2004. A previous audit suggested a significant morbidity and mortality rate and therefore the aim of this study was to identify risk factors for 30 day mortality and re-intervention.
Methods A retrospective review of all patients undergoing oesophageal SEMS insertion was performed over the period 2004–2014. Data collected included patient demographics, tumour stage/position, stent type inserted, survival following SEMS insertion, bioprofile and other therapies. Univariate analysis was performed and was followed by multivariate logistic regression to identify independent variables associated with 30 day mortality and re-stenting.
Results 396 patients (median age 74 years, 252 males) had 514 stents were inserted over the study period and were included in the study. Regarding 30 day mortality, univariate analysis showed association with distal tumours (OR 1.6), stent length >10 cm (OR 3.0), oncological treatment (OR 0.4), raised CRP (OR 6.1), raised white cell count (OR 6.0), anaemia (OR 2.8) and low albumin (OR 4.9). Multivariate analysis showed white cell count (OR 8.1, 2.0–21.9, p < 0.001, albumin <35 g/l (OR 6.9, 2.6–18.8, p < 0.001) and oncological treatment (OR 0.3, 0.1–0.8, p = 0.013) were independent predictors of 30 day mortality. Regarding re-intervention, univariate analysis showed association with age > 70 (OR 0.6), junctional tumours (OR 0.5), stent length >10 cm (OR 0.6), raised CRP (OR 0.5) and raised white cell count (OR 0.4). Multivariate analysis showed age > 70 (OR 0.5, 0.3–0.9, p = 0.032) and raised white cell count (OR 0.4, 0.2–0.8, p = 0.009) were independent predictors of re-intervention.
Conclusion Raised inflammatory markers were associated with poorer outcomes following oesophageal SEMS insertion and oncological treatment was protective. Longer survival appears to be associated with re-intervention following initial stent insertion. This data may provide a framework for the creation of a prognostic scoring system prior to SEMS insertion.
Disclosure of Interest None Declared