Valve sparing aortic root reconstruction versus composite replacement -- perioperative course and early complications

In patients with aneurysm of the ascending aorta, dilatation of the sinotubular junction is the major cause of aortic valve regurgitation. Valve sparing aortic root replacement in patients without valvular structural defects offers a new form of treatment. The aim of this study was the assessment of the perioperative course and early complications of this method compared to composite replacement in a large single center cohort.

From 1992 to 1999, valve sparing replacement of the ascending aorta (recon) has been performed in 78 patients, while 269 patients underwent replacement by a composite graft (comp). A comparison of matched pairs (n=52) with respect to age, gender, presence of Marfan's syndrome, aortic dissection as well as date of surgery, was chosen. Aortic insufficiency was 2.8+/-0.7 for recon vs. 2.2+/-1.1 for comp preoperatively. Course and length of hospitalization, echocardiographic follow-up, complications, and mortality were compared at 1-year follow-up.

There were no operative deaths. During follow-up, one patient (2%) died 5 months postoperatively (recon) vs. two patients (3.9%) in the comp group. Bypass-time (123+/-31 vs. 153+/-31 min, P<0.0001) and cross-clamp-time (82+/-22 vs. 120+/-23 min, P<0.0001) were significantly shorter in comp. Stay in ICU (1.9+/-1.6 for recon vs. 2.3+/-2.1 days for comp) and post-op hospitalization (18.3+/-5.7 vs. 21.2+/-11.1 days) were comparable. Improvement of NYHA-class was significant after both operations (recon 2.6+/-0.8 vs. 1.3+/-0.5 and for comp 2.4+/-0.6 vs. 1.5+/-0.7, both P<0.0001). One patient (1.9%) in the recon group had to be reoperated for valve failure. Thrombembolic or bleeding complications were observed in 6 patients (12%) in comp, zero in recon (P=0.027).

Valve sparing aortic root reconstruction is feasible with low perioperative morbidity and mortality and good early results. Major advantages of recon are significant reduction of thrombembolic and anticoagulation related complications as opposed to longer cross-clamp and bypass times as well as a valve failure in one patient. Further follow-up is needed to confirm our data in a long-term perspective.