A propensity-matched comparison for endovascular and open repair of thoracoabdominal aortic aneurysms

The aim of this study was to investigate outcomes of patients treated with endovascular repair (ER) with the use of fenestrated and branched stent grafts or open surgery (OS) for thoracoabdominal aortic aneurysm (TAAA) in a current series of patients.

All TAAA patients undergoing repair at three centers between January 2007 and December 2014 were included in a prospective database. Patients were stratified according to treatment by ER or OS, and outcomes were compared using propensity score matching (1:1). Covariates included age, sex, aneurysm extent, hypertension, coronary disease, chronic pulmonary disease, diabetes, and renal function. The primary end points were mortality and paraplegia. Secondary end points included any spinal cord ischemia (SCI), renal and respiratory insufficiency, and a composite of these complications or death at 30 days. All-cause survival and freedom from reintervention were compared in the two groups.

Of 341 patients, 84 (25%) underwent ER and 257 underwent OS (75%). After propensity score matching (65 patients per group), no significant differences were observed in rates of 30-day mortality (7.7% in ER and 6.2% in OS; P = 1) and paraplegia (9.2% and 10.8%; P = 1). Any SCI, renal insufficiency, and respiratory insufficiency were 12.3% and 20% (P = .34), 9.2% and 12.3% (P = .78), and 0% and 12.3% (P = .006) in ER and OS, respectively. The incidence of the composite end point was significantly lower in ER patients (18.5% in ER vs 36.0% in OS; P =.03). According to Kaplan-Meier estimates, all-cause survival at 24 months was 82.8% in ER and 84.9% in OS, with rates unchanged at 42 months (P = .9). Rates of freedom from reintervention were 91.0% vs 89.7% at 24 months and 80.0% vs 79.9% at 42 months in ER vs OS, respectively (P = .3).

A propensity score analysis in patients with TAAA undergoing repair suggests an early benefit from ER compared with OS with regard to the composite end point because of reduced 30-day respiratory complications. No significant differences were found in SCI and renal insufficiency at 30 days and in survival and reintervention rates at midterm.