2.1 All-cause mortality
Multiple observational studies have been published in recent years to ascertain the impact of TEVAR in acute or subacute un-TBAD. Most studies mentioned have an identical time-based classification of un-TBAD (<14 days for acute, 15-90 days for subacute, and >90 days for chronic since symptom onset) unless otherwise stated. Among these studies, some have investigated the mid- and long-term all-cause mortality rates in patients with different subtypes of TBAD receiving TEVAR. One such example is a retrospective study conducted by Tjaden et al. [6]. Two hundred and sixty four patients who had undergone TEVAR for TBAD (170 acute and 94 chronic) were selected from the Global Registry of Endovascular Aortic Treatment for analysis. Throughout a mean follow-up period of 26 months, the study reported that the total all-cause mortality is significantly higher in patients with chronic TBAD than in those with acute TBAD (19.2% versus 8.8%; P = 0.02). However, on further analysis, it was found that the specific overall survival (calculated as 1 minus all-cause mortality) for acute un-TBAD (n= 69, 26% of study group) was significantly higher (P<0.05) than any other subtypes of TBAD (93% vs 83% at 2 years, respectively).
In another retrospective study by Torrent et al. [7] which involved the evaluation of the data from the Vascular Quality Initiative TEVAR and complex endovascular aneurysm repair. Over 600 un-TBAD patients were included (446 acute and 242 subacute). The analysis of the information from the registry established that the 1-year mortality rates for acute and subacute un-TBAD patients treated by TEVAR were 13.3% and 8.2% respectively. Furthermore, the difference in mortality rates for the 2 groups was not significant (P=0.129). This finding is supported by Xie. et al. [8] who reported no difference in all-cause mortality between acute and sub-acute (4.2% and 8.3% respectively) un-TBAD patients treated with TEVAR. The study (follow-up duration up to 106 months) consisted of 267 patients divided into 2 groups (130 acute and 137 subacute) depending on when TEVAR was performed. Additionally, Xiang et al. [9] also reported findings for 238 acute un-TBAD patients, again divided into 2 groups depending on the timing of TEVAR (142 acute and 96 subacute). By 5 years, there was no difference seen between acute and subacute patients in terms of all-cause mortality (7.3% and 12.4% respectively). These results suggest that there are relatively low mid- to long-term (1-5 years) all-cause mortality rates for acute and subacute un-TBAD patients post-TEVAR. Furthermore, the difference between the 2 groups was also insignificant (P=0.39).
Other retrospective studies have also reported on the survival rates for un-TBAD patients post-TEVAR. Spinelli et al. [10] retrospectively analysed patient data from the Global Registry for Endovascular Aortic Treatment. In this registry, 172 patients with acute TBAD underwent TEVAR (102 complicated TBAD and 70 un-TBAD). The overall survival rates for acute un-TBAD at 1 year and 3 years were reported (96.8%±3.1% and 90.4%±9.5%, respectively). Bi et al. [11] also published overall survival rates at 3-year post-TEVAR and found 95.5% for acute (<15 days) and 100% for subacute (15-92 days) TBAD. The study included a total of 53 acute and subacute TBAD patients randomized into 3 groups (22 acute TBAD + TEVAR; 18 subacute TBAD + TEVAR; 13 TBAD + non-operative treatment). Also, number of mortalities due to complications (i.e. aortic rupture, low cardiac output, bleeding, multiple organ failure) across the follow-up period for both groups of patients that received TEVAR were reported to be zero, whilst the non-operative group only showed 4 deaths. The results of this study, however, is not completely compatible with this review as only 47 acute/subacute un-TBAD patients were included in this study.
Further studies have sought to compare the difference in mortality outcomes in set patient groups that received different forms of treatments for acute un-TBAD. In a study by Qin and colleagues [12], patients with acute un-TBAD across 3 medical centres were retrospectively identified. Three hundred and thirty eight fulfilled the study’s inclusion criteria, of which, 184 received TEVAR while the remaining 154 received BMT. The all-cause mortality was significantly higher (P=0.01) in those receiving BMT across a follow-up period of up to 11 years. Correspondingly, Iannuzzi et al. [13], retrospectively reviewed 9165 acute un-TBAD patients (95% best medical therapy; 2.9% TEVAR; 2% open repair). Analysis revealed acute un-TBAD patients treated with TEVAR have a significantly lower (P<0.01) all-cause mortality of 19% when compared to other treatment modalities (37% best medical therapy and 34% open surgical repair). Median follow-up periods for the 3 treatment groups were 1.5 years or longer.
A retrospective study by Xiang et al. [14] included 357 acute un-TBAD patients (191 TEVAR and 166 BMT). The median follow-up duration for both groups was at least 3 years. The freedom from all-cause mortality at 1-, 3- and 5-years in the TEVAR group was significantly higher (P=0.028) than the BMT group. The conclusion was acute un-TBAD patients treated by TEVAR showed a lower all-cause mortality in the mid- to long-term than BMT treatment and/or open surgery.
Prospective studies and randomized controlled trials, also provide valuable supporting evidence for all-cause mortality in TEVAR-treated un-TBAD patients. Firstly, the INvestigation of STEnt Grafts in Aortic Dissection trial (INSTEAD) [15] aimed to distinguish the mid-term results between 140 un-TBAD patients, randomized into 2 different groups (72 TEVAR and OMT versus 68 OMT only). Over a follow-up period of 2 years, data collected from the study showed that there was no significant difference between the 2 groups in terms of all-cause mortality (survival probability 95.6%±2.5% OMT versus 88.9%±3.7% TEVAR; P=0.15). The follow-up INSTEAD-XL trial, over 5 years, showed the risk of all-cause mortality was significantly lower (P=0.13) for TEVAR and OMT than OMT alone (11.1% versus 19.3% respectively) [16]. However, selection criteria for the INSTEAD trial also included chronic un-TBAD patients (2 to 52 weeks after onset), which is irrelevant to the scope of this present review. Secondly, the prospective randomized ADSORB trial [17] focused on acute un-TBAD patients (n=61). Randomization in the allocation of patients was 31 patients in the TEVAR + BMT group and 30 in the BMT only group. It is reported that at 1-year there was only one death in the TEVAR group. Again, data from the studies indicate a lower all-cause mortality for TEVAR, further proving its superiority.
Several systematic review and meta-analyses have attempted to summarize the findings of observational and interventional studies in order to identify the best treatment for un-TBAD. One such review [18] which included 6 studies, reported that overall, there was no significant difference in all-cause mortality at 1 year (5.1% versus 5.4%, P=0.96) nor 5 years (15.3% versus 26.3%, P=0.75) between acute un-TBAD patients that received TEVAR or best medical therapy. Contrastingly, another systematic review and meta-analysis by Hossack et al. [19], which also included six studies, reported that the risk of late (>30 days) all-cause mortality was significantly lower (P<0.001) for acute un-TBAD patients who received TEVAR and BMT when compared to those treated with BMT only.
Evaluating the outcome of mid- to long-term all-cause mortality in the findings of the various studies, reviews and analyses demonstrates that the treatment of choice for un-TBAD regardless of time of symptom onset within the first 90 days (acute or subacute) should be TEVAR.
Aorta-related mortality and rupture
Despite gaining consensus for superior long-term outcomes, TEVAR may cause late aortic adverse events which sometimes do lead to mortality. The cumulative freedom from aortic-related adverse events was reported as 71.8% at 5 years post-TEVAR [12]. Numerous studies reported late aortic-specific mortality, which commonly resulted from aortic rupture and to a lesser extent retrograde type A dissection (RTAD).
The most notable clinical trials investigating TEVAR versus OMT for un-TBAD are the INSTEAD, INSTEAD-XL and the ADSORB trials. The ADSORB trial reported no aortic rupture or aortic-related mortality within the first year post-TEVAR [17]. The INSTEAD-XL trial had a 6.9% aortic-specific mortality rate (5/72) in the first year of follow-up in the TEVAR group. Yet, there were no aortic-related deaths over the next four years of follow-up suggesting an optimal aortic-specific survival in the long term [16].
Observational studies investigating patients with acute un-TBAD undergoing TEVAR reported varying aortic-specific mortality rates, resulting mainly from rupture. A study by Xiang and colleagues [14], who proposed performing TEVAR in the acute phase of un-TBAD, showed freedom from aortic-related death at 1- and 5-years post-TEVAR as 97.8% and 94.1%, respectively. Further, Qin et al. [12] revealed an aortic-related mortality rate of 4.3% (8/184) over 29 months of follow-up. Both studies reported significant post-TEVAR aortic-related survival over OMT (P<0.05). Another study by Nakamura et al [20], who offered TEVAR during the subacute phase of dissection, ascertained the aortic-related mortality by reporting a rate of 4% over 37 months of follow-up.
Rupture and RTAD were the most commonly reported causes of aortic-related death in patients who underwent TEVAR for unTBAD. Rupture is associated with high mortality at the indexed event [21]. Other observational studies reported the incidence of aortic rupture to be 2.5%-5% [12, 14, 20]. Xiang et al. [14] showed a cumulative incidence rate of aortic rupture of 2.1% after 1-year and 5.1% after 5-years of follow-up. These variations in the incidence could be attributed to the different follow-up periods and timings of intervention adopted in each respective study.
Some studies reported the influence of different timings of interventions on the rate of post-TEVAR aortic-related mortality. A study by Tjaden and colleagues [6], with a mean follow-up of 26 months, showed that TEVAR in the acute phase of unTBAD was associated with more aortic-related mortality than in the chronic phase (2.7 vs. 2.1; P=1.0). In the same study, four cases of post-TEVAR aortic rupture occurred only in the acute group of intervention.
Xie and colleagues [8], demonstrated that aortic-related death is more likely to occur in the subacute than the acute intervention group (4.5% vs. 2.5; P=0.5). They also showed that risk of aortic rupture was higher in the subacute group (3.8% vs. 1.7%; P=0.45). Similarly, Xiang et al. [9] demonstrated that aortic-related mortality was higher in the subacute intervention group than in the acute group (6.8% vs. 5.2%; P =.86) five years post-TEVAR.
Three meta-analyses presented data on the long-term aortic-related outcomes in TBAD patients who underwent TEVAR versus OMT. Li and colleagues [22] demonstrated superior benefit in long-term survival with TEVAR compared to OMT (HR = 0.71; 95% CI: 0.52-0.95). Furthermore, this meta-analysis demonstrated that TEVAR for TBAD was protective from later rupture compared to OMT (OR = 0.21; 95%CI: 0.10-0.43). Another meta-analysis of four studies showed lower rates of aortic rupture post-TEVAR (TEVAR group 3.5%, OMT group 9.4%), which again demonstrated the OMT group has higher risks of rupture in the long term (OR 2.49 P=0.01) [18].The third meta-analysis by Hossack et al. [19] proved the significant benefit of TEVAR in protecting un-TBAD patients from late aortic adverse events compared to OMT (HR 1.56, 95%CI 1.14 - 2.13).
The presented data gives an understanding into the aortic-related survival in the long-term. However, there remains a paucity of robust evidence on the long-term aortic-specific mortality post-TEVAR in patients with acute un-TBAD. It is worth noting that, however, multiple studies showed both selection bias and significant loss of patients during follow-up. Finally, many studies reported mortalities due to unknown causes, which may affect actual rate of aortic-specific mortality.