Summary
Background
The Ross procedure is used in the treatment of selected patients with aortic valve disease. Pulmonary graft stenosis can appear in the long-term follow-up after the Ross intervention, but the factors involved and its clinical implications are not fully known.
Aim
To describe the incidence, clinical impact and predictors of homograft stenosis and reintervention after the Ross procedure in a prospective series in a tertiary referral hospital.
Methods
From 1997 to 2009, 107 patients underwent the Ross procedure (mean age: 30 ± 11 years; 69% men; 21 aged < 18 years), and were followed for echocardiographic homograft stenosis (peak gradient > 36 mmHg) and surgical or percutaneous homograft reintervention.
Results
After 15 years of follow-up (median: 11 years), echocardiographic and clinical data were available in 91 (85%) and 104 (98%) patients, respectively: 26/91 (29%) patients developed homograft stenosis; 10/104 (10%) patients underwent 13 homograft reintervention procedures (three patients underwent surgical replacement, three received a percutaneous pulmonary valve and one needed stent implantation). The other three patients underwent two consecutive procedures in follow-up; one died because of a procedure-related myocardial infarction. Rates of survival free from homograft stenosis and reintervention at 1, 5 and 10 years were 96%, 82% and 75% and 99%, 94% and 91%, respectively. Paediatric patients had worse survival free from homograft stenosis (hazard ratio [HR] 3.50, 95% confidence interval [CI]: 1.56–7.90; P = 0.002), although there were no significant differences regarding reintervention (HR: 2.01, 95% CI: 0.52–7.78; P = 0.31). Younger age of homograft donor was also a stenosis predictor (HR: 0.97, 95% CI: 0.94–0.99; P = 0.046).
Conclusions
The probabilities of homograft stenosis and reintervention 10 years after the Ross procedure were 29% and 10%, respectively; only one patient had a reintervention-related death. Younger donor and recipient age were associated with a higher rate of stenosis.
Résumé
Justification
La procédure de Ross est utilisée dans le traitement des valvulopathies aortiques de façon sélective. La sténose de la greffe pulmonaire peut apparaître lors du suivi au long cours ou au décours d’une intervention de Ross, les facteurs de risque et les implications cliniques ne sont pas parfaitement connus à ce jour.
Objectifs
Décrire l’incidence, l’impact clinique et les prédicteurs de la sténose d’une homogreffe pulmonaire ainsi que le taux de réintervention au décours d’une procédure de Ross dans une série prospective d’un centre hospitalier tertiaire.
Méthode
Entre 1997 et 2009, 107 patients ont bénéficié d’une procédure de Ross (âge moyen : 30 ± 11 ans ; homme : 69 % ; 21 patients âgés de moins de 18 ans). Ces patients ont été suivis en échocardiographie et le diagnostic de sténose de l’homogreffe a été porté sur un gradient pic à pic > 36 mmHg ainsi que sur les résultats au décours de la réintervention.
Résultats
Après un suivi de 15 ans (médian : 11 ans), les données cliniques et échographiques étaient disponibles chez 91 (85 %), 104 (98 %) des patients respectivement. Vingt-six des 91 patients (29 %) ont développé une sténose de l’homogreffe et 10 patients (10 %) ont bénéficié d’une réintervention au cours de 13 procédures, puisque 3 patients ont dû avoir secondairement un remplacement chirurgical, 3 patients ayant bénéficié d’un remplacement pulmonaire percutané et un patient ayant nécessité l’implantation d’un stent. Les trois autres patients ont bénéficié de deux autres procédures lors du suivi ; un patient est décédé au décours d’un infarctus du myocarde lié à la procédure. Le taux de survie sans sténose de l’homogreffe ou réintervention à un an, 5 ans et 10 ans était de 96, 82 et 75 % et de 99, 94 et 91 %, respectivement. Les patients pédiatriques avaient une survie sans sténose pulmonaire moindre (hazard ratio : 3,50, IC 95 % : 1,56–7,90 ; p = 0,002) bien qu’il n’y ait pas de différence significative en ce qui concerne le taux de réintervention (hazard ratio : 2,01, IC 95 % : 0,52–7,78 ; p = 0,31). Un âge plus jeune du donneur de l’homogreffe était également un prédicteur de la sténose (hazard ratio : 0,97, IC 95 % : 0,94–0,99 ; p = 0,046).
Conclusion
Les probabilités de sténose d’une homogreffe et d’une réintervention 10 ans après une procédure de Ross sont de respectivement 29 et 10 % ; seul un patient est décédé au décours d’une réintervention. Un âge plus jeune du donneur et du patient sont associés avec un taux accru de sténose.
Background
The Ross procedure is a therapeutic alternative in the treatment of selected groups of patients with aortic valve disease. At present, it is considered a particularly useful option in children and young adults because, besides providing haemodynamic advantages and acceptable durability, it has good adaptation to the growth of paediatric patients, and offers the possibility to avoid anticoagulation . The limited use of this procedure results, in part, from its greater technical complexity, but also because it has some long-term complications that require one or even several subsequent cardiac interventions.
Pulmonary graft stenosis is one of the complications that can appear in the long-term follow-up of patients after the Ross intervention. However, the factors involved in such degeneration and its clinical implications are not fully known. Knowing the possible triggers and clinical characteristics associated with a worse outcome would optimize patient selection, and help to prevent and treat complications arising in follow-up. So, our main objective was to describe the incidence and clinical impact of homograft stenosis, and to identify clinical, surgical and echocardiographic variables associated with this complication.
Methods
Study population
This prospective, observational, longitudinal study included all consecutive adult and paediatric (aged < 18 years) patients who were operated on using the Ross procedure at our institution (a tertiary referral hospital), from the introduction of the technique in November 1997 until July 2009.
Surgical technique
All interventions were performed electively by only four cardiovascular surgeons, using cardiopulmonary bypass with standard bicaval and aortic cannulation. Early in the series, the patients were operated on under moderate hypothermia (five patients, at 28 °C); subsequently, the procedures were performed under mild hypothermia (33 °C). The integrity of the native pulmonary valve was evaluated in all cases by transverse pulmonary arteriotomy. After resection of the pathological aortic valve, the pulmonary autograft was implanted in the aortic position as a full aortic root replacement with interrupted sutures on an annulus of autologous pericardium, anastomosing the coronary buttons to the autograft with a running suture, and the autograft to the ascending aorta. In recent years, we supported the neosinotubular junction with 2 or 3 cm of Dacron graft in all patients. The right ventricular outflow tract was restored with a pulmonary homograft implantation, with distal and proximal running polypropylene sutures. In all cases we used a pulmonary homograft cryopreserved in liquid nitrogen at –196 °C, treated in the Sectorial Tissue Bank Institution, and following the usual standards set by the Spanish Association of Tissue Banks. Human leukocyte antigen (HLA) typing and ABO compatibility testing were not performed.
Echocardiographic baseline and follow-up studies
The baseline echocardiographic study was performed before surgery by transthoracic echocardiography. All cases were scanned at our centre using the Acuson Sequoia 256 ® (Siemens, Mountain View, CA, USA) and iE33 ® (Philips, Amsterdam, Netherlands) ultrasound systems. The measurements, taken after at least 5 minutes of rest, and in the absence of other factors that might lead to overestimation of the flow rate, were interventricular septum, diastolic and systolic diameters and left ventricular posterior wall with M mode; left ventricular ejection fraction with the biplane Simpson’s method; assessment of valvular structure and function with bidimensional and Doppler echocardiography, with calculation of maximum and mean gradients by means of the simplified Bernoulli equation, and evaluation of valvular regurgitation according to the recommendations of scientific societies ; measurement of aortic and pulmonary annuli from the parasternal long axis and short axis, respectively, and evaluation of right ventricular size and function.
Clinical and echocardiographic data were collected at discharge, then at 6 and 12 months, and annually thereafter. Echocardiographic follow-up studies were performed according to an identical protocol to baseline echocardiography, with special attention to the study of the structure and function of the pulmonary autograft and pulmonary homograft. Both annuli were measured, and the maximum and mean transvalvular gradients were determined by the simplified Bernoulli equation. Valvular regurgitation was evaluated according to the recommendations of scientific societies .
Definition of study endpoints
For this study, three main objectives were considered in the follow-up: the appearance of homograft stenosis; the need for homograft reintervention; and, finally, the study of possible predictors of these complications. We decided to analyse just the homograft stenosis because of the low rate and grade of insufficiency. We observed only 22 cases of homograft regurgitation in our series: none of them severe, five moderate (two with significant stenosis) and 17 mild, with no homograft reintervention for this indication. The outcome “homograft stenosis” was defined as a maximum transvalvular gradient of at least 36 mmHg. This cut-off point was selected because it is the one considered in the guidelines as at least moderate stenosis and, therefore, requires closer clinical and echocardiographic follow-up. The “homograft reintervention” outcome included surgical replacement and percutaneous interventions (valvuloplasty, stenting or percutaneous implantation of pulmonary prosthesis). The criteria used to reoperate on patients were those recommended by clinical practice guidelines , and included: symptomatic worsening; objective worsening of right ventricular function and/or progressive right ventricular dilatation; and medically intractable infective endocarditis or endocarditis with significant functional involvement of the homograft.
The appearance of autograft dysfunction, autograft reintervention and global mortality and its causes were also registered in the follow-up.
Statistical analysis
Statistical analyses were performed using SPSS 15 for Windows (SPSS Inc., Chicago, IL, USA). Continuous variables are expressed as mean ± standard deviation (range). Categorical data are described using frequencies and percentages. Actuarial estimates of freedom from homograft stenosis, autograft dysfunction and reintervention were acquired using the Kaplan-Meier method. Possible determinants of clinical and echocardiographic outcomes were assessed by the Cox proportional hazards test, and for dichotomous qualitative variables we also used the log-rank test. We used multivariable Cox models for the study of possible predictors of homograft stenosis, initially including all variables with significance < 0.10 in the univariate analysis, and removed the non-significant variables in successive backward steps to get variables independently associated with this outcome. The multivariable analysis was not performed for the reintervention endpoint because there were insufficient events. Statistical significance was set at P < 0.05.
Results
Preoperative patient characteristics
From November 1997 to July 2009, a total of 107 Ross interventions were performed at our centre. Preoperative patient characteristics are presented in Table 1 . Our sample consisted mainly of young men with normal left ventricular systolic function and severe aortic stenosis of congenital aetiology; 20% were paediatric patients.
| Total ( n = 107) | Adults ( n = 86) | Paediatric (aged < 18 years) ( n = 21) | |
|---|---|---|---|
| Recipient age (years) | 30 ± 11 [3–54] | 34 ± 9 [18–54] | 12 ± 4 [6–17] |
| Recipient sex | |||
| Male | 74 (69) | 61 (71) | 13 (62) |
| Female | 33 (31) | 25 (29) | 8 (38) |
| Congenital aetiology | 70 (67) | 49 (58) | 21 (100) |
| Previous cardiac surgery | 24 (25) | 14 (18) | 10 (55) |
| Donor age (years) | 42 ± 13 [15–69] | 44 ± 13 [18–69] | 36 ± 16 [15–54] |
| Donor sex | |||
| Male | 57 (62) | 51 (67) | 6 (38) |
| Female | 35 (38) | 25 (33) | 10 (62) |
| Interventricular septum (mm) | 13 ± 7 | 14 ± 7 | 10 ± 3 |
| LV diastolic diameter (mm) | 59 ± 11 | 60 ± 11 | 55 ± 11 |
| LV systolic diameter (mm) | 37 ± 11 | 37 ± 11 | 33 ± 8 |
| LV posterior wall (mm) | 10 ± 3 | 10 ± 3 | 10 ± 4 |
| LV ejection fraction (%) | 67 ± 10 | 66 ± 10 | 71 ± 6 |
| Aortic annulus (mm) | 24 ± 5 | 25 ± 4 | 22 ± 5 |
| Pulmonary annulus (mm) | 22 ± 3 | 22 ± 3 | 22 ± 3 |
| Freezing time (months) | 7 ± 3 [1–57] | 6 ± 5 [1–29] | 8 ± 6 [0.73–57] |
| Homograft size (mm) | 25 ± 3 [19–32] | 26 ± 3 [22–32] | 24 ± 4 [19–32] |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
