Getting to the root of the matter: management of the aortic root in Type A aortic dissection

In the setting of acute type A aortic dissection, the optimal management of the aortic root remains controversial. The primary goal of operative intervention is to have a surviving patient with a repaired ascending aorta and a resected proximal tear. Decision-making regarding the aortic root is less clear, however, involving both patient-specific factors [i.e. location and extent of the primary tear, competency of the aortic valve (AV), presence or absence of connective tissue disease, and age and overall clinical status of the patient], and surgeon-specific factors (i.e. operator experience with aortic dissection surgery and various aortic root/aortic arch repair/replacement techniques). In a recent round-table of experts held at the 2015 Aortic Valve Repair Summit [ 1 ], some experts favoured AV reimplantation (i.e. David) procedures for type A aortic dissection, others favoured a partial remodelling (i.e. Yacoub) technique with a tongue of graft extending into the non-coronary sinus, and still others recommended a more conservative approach with limited aortic root intervention, citing the overarching goal of leaving the operating room with a live patient. Despite the variety of recommendations, achieving consensus regarding aortic root management in aortic dissection patients would be helpful for the cardiac surgery community. In this issue, Castrovinci et al . [ 2 ] shed more light on this important topic by analysing 296 patients who underwent repair of acute type A dissections from 1999 to 2014. Patients were followed for a mean of 3.5 years for the primary end-points of survival and freedom from aortic root intervention. The majority of patients (60%) underwent conservative management of the aortic root (CR), whereas 40% underwent a root replacement (RR) procedure including 108 modified Bentall operations and 11 AV reimplantation procedures. Indications for RR in the study included Marfan syndrome, primary AV pathology, dilated aortic root (>45 mm) and dissection flap involvement of more than one coronary sinus or extension to the aortic annulus. The majority of patients underwent hemi-arch repair, whereas total arch or elephant trunk procedures were performed in only 22%. The extent of aortic replacement was not significantly different between CR and RR groups. RR patients were significantly younger with more proximal tears and had longer cardiopulmonary bypass (CPB) and myocardial ischaemia times. Six patients in the CR group had unresected primary tears, whereas no RR patients had unresected primary tears. There were no differences between the groups in terms of perioperative complications with in-hospital mortality occurring in 21% of patients and a 5-year survival rate of 68%. Proximal aortic reoperations were required in 16 patients during follow-up: 2 in the RR group versus 14 in the CR group. The freedom from proximal aortic reintervention at 5 years was 96% in the RR group and 91% in the CR group ( P = 0.02). The authors also performed a 1:1 propensity score-matched analysis of 82 patients based on 19 preoperative and 7 intraoperative variables. In this matched cohort, the freedom from proximal reintervention was 98 vs 86% at 5 years for RR and CR patients, respectively ( P = 0.06). As expected, the predominant risk associated with a conservative approach to the aortic root is the need for future reintervention and associated risks related to that procedure. In addition, some patients may experience rupture of the aortic root during follow-up, although this risk appears to be small. The likelihood of future aortic root reintervention is likely related to initial pathology and patient age, with younger patients having a longer time to develop aortic root aneurysms. This probability must be weighed against the risks of a more complicated initial operation, particularly for surgeons who do not perform a large volume of aortic root surgery. Consistent with previous, non-propensity-matched studies [ 3–5 ], the current study revealed no differences in perioperative or in-hospital survival between the CR and RR strategies. In addition, there were no differences in mid-term survival, including mortality that occurred during subsequent proximal aorta reinterventions. A small proportion of patients underwent AV reimplantation RR procedures in the current study, a technique that has been increasingly used in the setting of aortic dissection in large volume centres. While such procedures are associated with longer myocardial ischaemic and CPB times, large registry data have revealed that they are not associated with an increased risk of perioperative mortality [ 6 ]. If a successful AV reimplantation procedure is performed, the patient should have a very low risk of subsequent aortic root reintervention and near zero risk of aortic root rupture [ 8 ]. While the paper from Castrovinci et al . provides insights into aortic root management in acute type A dissection, many questions still remain. Although it seems prudent to recommend a conservative approach for surgeons with limited aortic root experience, indications for RR surgery (and in particular, AV reimplantation) by experienced surgeons remain unclear. In the IRAD database, RR was performed more commonly in younger patients and in those with larger aortic root diameters, Marfan syndrome and aortic regurgitation [ 7 ]. Rylski et al. have shown that RR surgery yields improved long-term survival in Marfan patients [ 8 ]. Preoperative root diameter has been found to be a predictor of subsequent aortic regurgitation and dilatation after CR [ 9 ] and preoperative severe aortic regurgitation has been found to be a predictor of proximal reintervention [ 10 ]. Further questions also remain. For instance, it remains unclear for a given aortic pathology—in patients without any of the risk factors listed above—whether a CR or RR procedure results in superior long-term results. In addition, the optimal type of CR repair (e.g. commissural resuspension, glue apposition of dissected layers, partial Yacoub) needs to be defined. Furthermore, the risk of future aortic root reintervention in CR patients needs to be better defined. And lastly, the role of surgeon experience on perioperative and long-term outcomes needs to be further investigated. While the authors of the current study are to be congratulated for their approach to this challenging topic, further study is definitely warranted.