The Ross procedure in children: preoperative haemodynamic manifestation has significant effect on late autograft re-operation.

OBJECTIVES: The Ross procedure is the aortic valve-replacement procedure of choice in children. Nonetheless, late autograft re-operation for dilatation and/or valve regurgitation is of concern. We examined whether preoperative haemodynamic manifestation (e.g., stenosis, regurgitation and mixed aortic valve disease) affected late re-operation risk.
METHODS: Medical records of 227 children who underwent the Ross procedure (1991-2004) were reviewed. Competing-risks methodology determined time-related prevalence and associated factors for two mutually exclusive end-states after the Ross procedure: (1) death prior to subsequent autograft re-operation and (2) autograft re-operation, with the remainder of patients being alive and free from subsequent autograft re-operation.
RESULTS: There were 162 male patients (71%) in this study. Median age at surgery was 12.1 years (range: 1 week-18 years). The haemodynamic aortic valve dysfunction was primarily stenosis (n=40, 18%), primarily regurgitation (n=109, 48%) and mixed disease (n=78, 35%). Underlying pathology was rheumatic fever (n=104, 46%), congenital heart disease (n=113, 50%) and endocarditis (n=8, 3%). Competing-risks analysis showed that, at 10 years following the Ross procedure, ∼5% of patients had died, 16% had undergone autograft re-operation with aortic valve replacement and 79% were alive and free from autograft re-operation. Ten-year freedom from autograft re-operation for patients with preoperative stenosis, regurgitation and mixed disease was 97%, 69% and 93%, respectively, (p<0.001 for regurgitation vs others). Risk factors for increased risk of autograft re-operation were rheumatic fever (parameter estimates (PEs): 2.09 ± 0.75, p=0.006), and earlier year of surgery (PE: 0.20 ± 0.06, p=0.001). Ten-year freedom from homograft replacement was 81% and was not dependent on haemodynamic manifestation (PE: -0.16 ± 0.38, p=0.68). Significant factors for homograft replacement included fresh homografts (PE: 2.2 ± 0.63, p=0.01) and annular enlargement (PE: 1.11 ± 0.3, p=0.01). Ten-year freedom from cardiac re-operation other than auto-/homograft was 85%, higher in patients with preoperative aortic regurgitation (PE: 1.01 ± 0.42, p=0.02). Concomitant cardiac surgery was a significant factor for late cardiac re-operation other than auto-/homograft replacement (PE: 1.79 ± 0.39, p<0.001).
CONCLUSIONS: The Ross procedure in children is associated with excellent survival. Late autograft re-operation may be required; however, it is more common in children with preoperative aortic regurgitation, especially those with rheumatic fever. Better patient selection in later era has mitigated the risk of autograft re-operation. Continued improved candidate selection, along with modifications in autograft implantation and root/sinotubular stabilisation techniques, may further decrease late autograft failure.