Early mortality
The observed pooled early mortality was 5.5%. This is in accordance with an older review on aortic root replacement (4.5 to 5.3%) and with the recent report of the Society of Thoracic Surgeons database from the U.S. that estimates early mortality after bioprosthetic aortic root replacement to be 6.2% (45, 46). However, these studies include also acute and emergent operative indication like endocarditis and aortic dissection. Early mortality was mainly due to low cardiac output (22.7%) and multi-organ failure (18.1%). Surgical indication was endocarditis in 9.2% of patients and type A aortic dissection in 7.2%. The high mortality could (partly) be explained by operation in emergent setting and partly by the additional procedures (e.g. arch replacement, CABG). Nevertheless, overall early mortality seems not changed significantly last 2 decades.
Late mortality and reintervention outcome There was a high mortality rate (4.8%/pt-year) for a pooled mean age of 65.9 years, which is higher than the general population mortality. Translated to our microsimulation-based life-expectancy, there is a life-expectancy of 14.3 year for a 60 year old patient receiving a bioprosthetic root replacement, while there is a life-expectancy of 22.5 years for the 60 year old U.S. “healthy” population (12). From previous research there is evidence of significant “excess mortality” in (elective) isolated aortic valve replacement, compared to the age-matched general population (47). Additionally, patients in this study were diagnosed with a dilated aortic root as well, with about 13% suffering from a dissection of the root and/or connective tissue disease, which are conditions that may influence patient survival due to complication other than valve-related events.
This microsimulation model shows a life-time reintervention risk of 9% for patients older than 60 years, which is comparable to previous predictions on biological aortic valve prostheses (3). It is known that younger patients, especially younger than 60 years, are more likely to have a reintervention after biological aortic valve replacement, mainly due to progressive SVD (3, 48). Of the 8 studies that explicitly tested association between age and reintervention, 3 found indeed that older age is associated with lower reintervention hazard.
Thromboembolic eventsWe found a high incidence of thromboembolic events, with a life-time risk of more than 20% after bioprosthetic aortic root replacement. Data on TE events are not comprehensive, thus discriminating between TIA and disabling ischemic CVA is not possible. However, a previous systematic review and microsimulation study on aortic valve replacement with isolated biological stented valve, published by Puvimanasinghe et al. (3), reports similar TE event rates (1.4%/patient-year). Additionally, the incidence of thromboembolic events is known to increase with age (49, 50) and might partly explain this high incidence of thromboembolic events.
Subsequently the question arises whether there is a difference with patients receiving a classical Bentall prosthesis. Although this comparison is hampered by the differences in patient characteristics, mainly due to the younger age in patient receiving mechanical valves; a recently published meta-analysis on the Bentall procedure (mean age 50 years), shows lower thromboembolic event rates (0.77%/patient-year) (51). Another study on mechanical valve replacement in non-elderly showed suboptimal survival and considerable lifetime risk of anticoagulation-related complications (52). However, the anticoagulation therapy after mechanical valve implantation in these patients plays a protective role in prevention of thromboembolic events, as it also occurs irrespective of the aortic valve replacement due to the aging process (50).
Additionally, TE hazard is less likely to occur during long-term follow-up, suggesting a larger hazard in the early postoperative period, which may be related to anticoagulation therapy. There were also some studies that included (a small portion of) patients diagnosed with endocarditis and aortic dissection, which may have led to additional TE rate. However, excluding these studies from the analysis did not changed the results. According to the current US and European guidelines on the management of valvular heart disease, antiplatelet therapy is reasonable and may be considered for the first 3 months after biological valve replacement (2, 53). Additionally, the European guidelines state that the need for a 3 months postoperative period of anticoagulation therapy has been challenged in patients with bioprostheses, with low-dose aspirin being favored as an alternative.
Hence, it is questionable whether the proposed anticoagulation therapy is appropriate in patient receiving a bioprosthetic aortic root replacement. Nevertheless, due to a lack of data on the exact anticoagulation therapy and patient compliance, it is not possible to make broad inference about this possible association. Further studies are needed to determine the most optimal anticoagulation therapy after biological aortic valve replacement.Endocarditis and type of prosthesesAlthough the rate of endocarditis after bioprosthetic aortic root replacement varies widely in the literature (54), our findings are comparable to an older study on biological aortic valve replacement [3]. Nine studies included patients with acute endocarditis, of which three with an endocarditis rate of more than 2.8%/patient-year (22, 26, 55), all including stentless valve prostheses. However, these studies included a relatively high proportion of patients with active endocarditis which may explain the higher re-endocarditis rate, although the severity of endocarditis was not provided. Hence, the trend toward more endocarditis in stentless valves could possibly be explained by the latter. Based on these data it is reasonable to assume that stented bioprosthetic grafts are at least not inferior to stentless bioprosthetic grafts to be used in case of endocarditis, although the extent of endocarditis may allow for different inference. Moreover, we found no difference in other valve-related events between stentless and stented prosthesis. We believe that both prostheses are safe to use in the average patient undergoing aortic root replacement.