Risk stratification and outcomes of transluminal endografting for abdominal aortic aneurysm: 7-year experience and long-term follow-up.
PURPOSE: To determine early and late outcomes of transluminal endografting (TE) in patients with abdominal aortic aneurysm (AAA), stratified by predicted risk of procedure-related mortality with conventional operation. MATERIALS AND METHODS: A retrospective study was conducted in consecutive risk-stratified AAA patients undergoing TE at a not-for-profit cardiovascular referral center from March 1994 through November 2000 with follow-up through February 2001. With use of conventional risk strata (0 = low, 1 = minimal, 2 = moderate, and 3 = high), predicted procedure-related mortalities were 0%-1% in stratum 0 (n = 40), 1%-3% in stratum 1 (n = 118), 3%-8% in stratum 2 (n = 116), and 8%-30% in stratum 3 (n = 31). Main outcome measures were: (i) TE procedural success, (ii) procedure-related mortality, (iii) major nonfatal complications, (iv) composite adverse outcome (ii + iii), (v) length of stay (LOS), (vi) freedom from AAA rupture, (vii) late survival, (viii) late complications, and (ix) endoleaks and their classification and management. RESULTS: Women were significantly less likely than men to qualify for and undergo endografting: 24 of 91 (26.4%) women underwent TE, compared to 281 of 684 (41.1%) men. Of 305 attempted TE procedures, 291 (95.4%) were successful, four (1.3%) were urgently converted to open repair, and 10 (3.3%) were aborted. Procedure-related mortalities occurred in eight cases (2.6%) overall and one of 40 (2.5%), one of 118 (0.8%), four of 116 (3.4%), and two of 31 (6.5%) cases for risk strata 0-3, respectively. Perioperative survivors were significantly younger than nonsurvivors (74.3 y +/- 9 vs 81.6 y +/- 5.1; P =.0087). Forty-six patients (15.1%) had major complications. Composite adverse outcome was worse for patients in stratum 3 than those in stratum 1 (P =.0296) and those in strata 0, 1, and 2 combined (P =.026). Procedure-related mortality declined with institutional experience, from 4% among the first 100 patients undergoing TE to 1% among the last 105. For strata 0-3, median LOS were 2, 3, 3, and 4 days, respectively. Seventy patients (22.9%) had 75 endoleaks, of which 30 necessitated additional procedures, 17 self-resolved, and 22 were untreated as of March 1, 2001. Five patients with endoleak died of unrelated causes. One late-onset type IA endoleak (26 mo) resulted in the only AAA rupture and death in the follow-up period among the 291 patients who underwent successful transluminal endograft implantation. Actuarial survival rates at 1 year after TE were 90.3% +/- 1.9% for the overall study group and 97.5% +/- 2.5%, 94% +/- 2.5%, 86.9% +/- 3.3%, and 81.3% +/- 7.7% for risk strata 0-3, respectively. At 5 years, overall actuarial survival was 69.6% +/- 6.1%. Thirty-eight late deaths were attributable to post-TE AAA rupture (n = 1), AAA rupture late after failed TE with no further treatment (n = 1), other cardiovascular disorders (n = 7), cancer (n = 15), other causes (n = 10), and unknown causes (n = 4). Late deaths occurred in risk strata 0-3 at the following rates: two of 40 (5%), 10 of 118 (8.5%), 16 of 116 (13.8%), and 10 of 31 (32.3%), respectively (stratum 0 vs stratum 3, P =.0017; stratum 1 vs stratum 3, P =.003). CONCLUSIONS: TE is safe and confers durable protection against AAA rupture in treated populations. Still, protection is not absolute in patients with endoleaks, because late AAA enlargement and even rupture can occur. Given current knowledge, technology, and practice, careful patient selection and close surveillance of patients after implantation of transluminal endografts is essential.