Valve-Sparing versus Valve Replacement Techniques for Aortic Root Operations in Marfan Patients: Interim Analysis of Early Outcome
Joseph S. Coselli1, Thoralf M. Sundt2, D. Craig Miller3, Joseph E. Bavaria4, Scott A. LeMaire1, Heidi M. Connolly2, Harry C. Dietz5, Dianna M. Milewicz6, Laura C. Palmero1, Xing Li Wang1, Irina V. Volguina1; 1Baylor College of Medicine and The Texas Heart Institute, Houston, TX; 2Mayo Clinic, Rochester, MN; 3Stanford University, Stanford, CA; 4University of Pennsylvania, Philadelphia, PA; 5Johns Hopkins Hospital, Baltimore, MD; 6University of Texas Health Science Center, Houston, TX
Comment on this Abstract
Objective: It is unclear which aortic root replacement technique works best in patients with Marfan syndrome; therefore, as part of a prospective, international registry study conducted at 21 institutions, we compared early outcomes in Marfan syndrome patients who underwent aortic root replacement with either valve replacement (AVR) or valve-sparing (AVS) methods.
Methods: An interim analysis was performed on the first 99 patients enrolled. All patients met strict Ghent diagnostic criteria for Marfan syndrome and underwent aortic root replacement with either AVR (n = 33) or AVS (n = 66) techniques; the choice of operation was based on clinical factors and surgeon and patient preference. In the AVR group, valve replacement was done with a mechanical composite valve graft in 28 patients (85%) and a bioprosthetic valve in 5 (15%). In the AVS group, David V procedures were performed in 42 patients (64%), David I in 19 (29%), David IV in 4 (6%), and Florida sleeve in 1 (2%). We compared preoperative factors, intraoperative variables, and early postoperative outcomes in the AVR and AVS groups.
Results: Except for age, preoperative factors (see Table), including NYHA class, aortic root size, left ventricular ejection fraction, comorbidities, medications, and smoking, did not differ significantly between the 2 groups. Concomitant procedures were similar in both groups. Despite longer cross-clamp and pump times in the AVS group, there were no significant differences in postoperative complications. No in-hospital or 30-day deaths occurred. One patient suffered a transient neurologic deficit 7 days after an AVS procedure. Valve-related complications included bleeding (n = 13), embolism (n = 2), nonstructural dysfunction (n = 1), and structural deterioration (n = 1). Ten patients required reoperations for bleeding, and 1 patient required early reoperation for revision of an AVS root replacement.
Conclusion: This interim analysis revealed that AVS was the most common operation in Marfan syndrome patients undergoing root replacement. Although AVS procedures, which tended to be used in younger patients, required longer aortic clamp and cardiopulmonary bypass times, the complexity of AVS aortic root replacement did not translate into adverse early outcomes. Subsequent long-term analysis is underway to compare the durability of these 2 approaches.
Perioperative Variables and Outcomes
| Variable | AVR | AVS | p Value |
| Age (yrs) | 40 ± 14 | 31 ± 11 | 0.01 |
| Preoperative aortic root diameter (mm) | 51 ± 8 | 52 ± 6 | 0.6 |
| Emergent or urgent operation (n) | 3/33 (9%) | 6/66 (9%) | 1.0 |
| Aortic dissection (n) | 5/33 (15%) | 8/66 (12%) | 0.8 |
| Aortic clamp time (min) | 111 ± 48 | 185 ± 76 | < 0.001 |
| Cardiopulmonary bypass time (min) | 149 ± 78 | 231 ± 93 | < 0.001 |
| 30-day valve-related complications (n) | 7/33 (21%) | 10/66 (15%) | 0.6 |
| Early reoperation (n) | 5/33 (15%) | 6/66 (9%) | 0.5 |
Back to 88th Annual Meeting
Back to Program Outline
