Impact of hospital volume on outcomes following treatment of thoracic aortic aneurysms and type-B dissections.

Previous research suggests an association between hospital volume and outcomes in high-risk surgical pathologies. The association between hospital volume and outcomes in patients with isolated descending thoracic aortic aneurysms (DTAAs) and type-B thoracic aortic dissections (TBADs) is conflicting. We aimed to investigate this in a literature review and meta-analysis. A systematic review of the literature was performed to identify studies reporting mortality and morbidity following repair (elective or emergency) of DTAA and/or TBAD using the Medline and Embase Databases (2000-2015). Hospital volume was assessed based on the number of patients treated per institution: low volume (1-5 cases per year), medium volume (6-10) and high volume (>10). The primary outcome of interest was all-cause mortality during inpatient stay and at 30 days. Eighty-four series of non-dissecting DTAA or TBAD were included in data synthesis (4219 patients; mean age: 62 years; males: 73.5%). For all patients (emergency and elective) undergoing DTAA repair, in-hospital mortality was 8% [95% confidence interval (CI): 6-8%]. Results were not superior in high-volume centres (8 vs 6 vs 11% for high-, medium- and low-volume, respectively). Sub-analyses for emergency and elective repairs showed no significant differences. For TBAD repairs, in the combined population (emergency and elective), results reached borderline significance (P = 0.0475), favouring high-volume centres (6 vs 11 vs 14%), but this association disappeared when emergency and elective repairs were analysed separately. Nine series reported outcomes at 1 year and 5 series followed DTAA and 18 TBAD treatment. No meaningful long-term comparisons were possible due to the lack of data. No significant associations were detected between hospital volume and subsequent mortality following DTAA or TBAD treatment. Data were heterogeneous and long-term results were scarcely reported. A well-designed longitudinal study of sufficient size is required to inform future strategies in this area.