Acute type A aortic dissection (AAAD) is associated with high in-hospital mortality, exceeding 20% according to the International Registry of Aortic Dissection (IRAD). Five-year survival stands at approximately 63%. Predictors of long-term morbidity and mortality include advanced age, female sex, atherosclerosis, and renal impairment. A recent meta-analysis by Li et al. demonstrated that the size of the residual patent FL is associated with poorer long-term survival compared to complete FL thrombosis. Therefore, residual flow in the FL is considered to promote dilation and, ultimately, rupture of the aorta. Additionally, partial FL thrombosis has also been independently associated with worse long-term survival in type B aortic dissection.
Many survivors of aortic dissection may have an indication for long-term oral anticoagulation (OAC) therapy. Currently, there are no clear recommendations for OAC use in patients undergoing AAAD repair. The study by Song et al. found an association between early initiation of anticoagulation and a higher incidence of FL patency, yet paradoxically observed better overall long-term clinical outcomes. Other studies have reported conflicting results, failing to demonstrate a clear clinical impact, whether negative or positive of OAC in this context.
The aim of this study was to investigate the long-term clinical impact of OAC and its effects on FL patency in patients who underwent surgery for AAAD. The study analyzed 188 patients (median age, 62 years; 74% male) who underwent AAAD repair, comparing patients who received chronic postoperative OAC (n = 59) with those who received antiplatelet therapy alone (n = 129). The median age was similar between groups, 60 years (range: 18–79 years; anticoagulated group) vs. 64 years (range: 22–86 years; non-anticoagulated group) (p = 0.11); anticoagulated patients were more frequently male (88% vs. 67%, p = 0.003). After a median follow-up of 8.4 years (range: 2 months–30 years), 58 patients died, 18 due to aortic-related causes, and 37 patients underwent aortic reintervention. After multivariable analysis, anticoagulation did not show a significant effect on long-term survival or reintervention risk. Analysis of 127 postoperative CT scans showed a patent FL in 53% of anticoagulated patients vs. 38% of non-anticoagulated patients (p = 0.09). The FL was also partially thrombosed in 8% vs. 28% (p = 0.01) and fully thrombosed in 39% vs. 34% (p = 0.63), respectively. Among patients with follow-up CT scans, there were 6 late aortic-related deaths, 1 among anticoagulated patients and 5 among non-anticoagulated patients.
The authors concluded that chronic anticoagulation following AAAD repair favors long-term FL patency, which is not associated with an increased risk of late mortality or reintervention in this study. Chronic anticoagulation can be safely administered to patients with repaired AAAD regardless of specific indications.
COMMENTARY:
Current scientific evidence indicates that in chronic aortic dissection, FL patency and even partial thrombosis are associated with disease progression and a poor long-term prognosis compared to patients with complete FL thrombosis. This association also applies to surgically treated AAAD cases. However, it is noteworthy that the incidence of long-term FL patency has significantly decreased due to advancements in surgical techniques, such as the increasingly frequent use of frozen elephant trunks. Therefore, it could intuitively be assumed that anticoagulant use in these patients would reduce the incidence of FL thrombosis, which would be linked to a poorer prognosis. However, the study by Vendramin et al. suggests that oral anticoagulation (OAC) does not have as unfavorable an effect as might initially be assumed. In this retrospective single-center analysis, a higher proportion of patients with persistent FL patency was observed in the OAC group, but this was not associated with an increase in adverse events. These findings provide a degree of reassurance when chronic anticoagulation therapy is required in patients who have undergone surgical intervention for AAAD, especially those with comorbidities and a higher risk of thromboembolic events, such as the presence of mechanical valve prostheses or concomitant atrial fibrillation.
The overall mortality of treated type A dissections over time was 24%. Of the 188 patients who survived until discharge, FL evolution could be accurately evaluated in 127, thanks to appropriately timed imaging follow-ups. Among these patients, 39 received oral anticoagulation therapy with warfarin, mainly due to the presence of mechanical prostheses, while the remaining 88 did not receive oral anticoagulation. Therefore, it is important to interpret the results of this study with caution, as it includes a limited number of evaluated patients. Although the baseline characteristics of the two groups were very similar, the type of surgery performed differed significantly. The non-anticoagulated group had a much higher proportion of isolated hemiarch replacements compared to the anticoagulated group (87% vs. 29%). Consequently, this study may present a selection bias, as a relationship between the extent of repair and FL thrombosis seems to exist. In cases where more residual aorta was left, there was a higher likelihood of FL patency (44% vs. 34% in this study) and a lower incidence of complete thrombosis (31% vs. 48%) compared to more aggressive and extensive surgeries.
A notable aspect of this study is that, although a slightly higher but non-significant incidence of FL patency was observed in the OAC group compared to the non-anticoagulated group (54% vs. 38%), and a significantly higher incidence of partial thrombosis was observed in the non-anticoagulated group (28% vs. 8%), the incidence of complete FL thrombosis was comparable between the two groups. From another perspective, one could also argue that if FL patency is considered as the combination of both completely and partially patent FL cases, there was no difference between the groups, with 62% in the OAC cohort and 66% in the non-anticoagulated cohort. Conversely, if thrombosis is considered as the combination of complete and partial thrombosis, then it occurred in 46% of the OAC group and in 62% of those without anticoagulants. Therefore, it could be inferred that OAC does not promote FL patency per se but rather prevents thrombosis.
In addition to flow rate and duration of FL patency, other factors influence thrombus formation and growth of the residual dissected aorta. Recent computational modeling studies suggest that the geometry of the aortic dissection and its evolution over time also play influential roles. Therefore, while OAC may favor the persistence of FL patency, it is insufficient alone to lead to a poorer prognosis compared to patients who are not on anticoagulants. The long-term prognosis of patients undergoing AAAD repair and the mechanisms involved are determined by multiple factors, requiring a multifaceted approach.
The findings of the study by Vendramin et al. add further complexity to the debate on the effect of oral anticoagulation on FL patency, and its interpretation may vary depending on how FL patency or partial thrombosis is defined. However, the true value of this study lies in its demonstration that OAC use in this patient group does not seem to have a significant impact on mortality or the incidence of reoperations. This has substantial intrinsic value and a direct practical implication for the care of our patients.
REFERENCES:
Vendramin I, Piani D, Lechiancole A, de Manna ND, Sponga S, et al. Do oral anticoagulants impact outcomes and false lumen patency after repair of acute type A aortic dissection? J Thorac Cardiovasc Surg. 2023 Jul;166(1):38-48.e4. doi: 10.1016/j.jtcvs.2021.09.009.
Song SW, Yoo KJ, Kim DK, Cho BK, Yi G, et al. Effects of early anticoagulation on the degree of thrombosis after repair of acute DeBakey type I aortic dissection. Ann Thorac Surg. 2011 Oct;92(4):1367-74; discussion 1374-5. doi: 10.1016/j.athoracsur.2011.04.111.
