Early and mid-term mortality and morbidity of contemporary international endovascular treatment for type B aortic dissection - A systematic review and meta-analysis

Abstract Background Effectiveness and optimal timing of endovascular treatment for type B aortic dissection (AD) remain controversial. Method An extensive search of literature (January 1999-December 2017) was conducted using PubMed, Cochrane Library and Science-Direct databases for studies on endovascular repair for acute/chronic type B AD; ≥10 patients; not reviews; and reporting predefined baseline data and ≥50% of predefined study outcomes, which were extracted and analysed according to Preferred Reporting Items for Systematic reviews and Meta-Analyses. Primary outcomes were in-hospital all-cause mortality and aorta-related mortality. Result Based on 92 publications (5956 patients), pooled estimate for overall in-hospital mortality was 7.0% [95% CI, 6.2%–7.8%]. Major perioperative complications included stroke (4.2% [3.6%–4.9%]), spinal cord ischemia (3.3% [2.8%–3.9%]), retrograde type A AD (3.2% [2.7%–3.9%]), type I endoleak (4.9% [3.8%–6.2%]), visceral ischemia (3.1% [2.5%–3.8%]) and acute renal failure requiring haemodialysis (5.1% [4.3%–5.9%]). Mid-term mortality incidence was 8.9% [7.2%–10.9%], and secondary intervention rate was 12.5% [10.5%–15.0%] with 6.1% [5.3%–7.2%] conversion to open surgery. Institutions with ≥40 endovascular treatment caseload had significantly lower rates of in-hospital and aorta-related mortality, stroke, type I endoleak, renal failure and retrograde type A AD. Patients treated in the acute phase had significantly higher incidence of in-hospital or aorta-related mortality and renal failure. Patients with chronic dissection required significantly more often secondary intervention during follow-up. Conclusion Endovascular stent-graft for type B AD therefore appeared feasible and safe with a low incidence of mortality and perioperative complications, particularly for delayed intervention and centres with ≥40 caseload. Standardized and long-term follow-up data are warranted.