Open, closed or a bit of both: a systematic review and meta-analysis of staged thoraco-abdominal aortic aneurysm repair
Systematic Review

Open, closed or a bit of both: a systematic review and meta-analysis of staged thoraco-abdominal aortic aneurysm repair

Benjamin T. Muston1,2, James Bilbrough2, Ymer Bushati2, Ashley R. Wilson-Smith1,3, Martin Misfeld1,3,4, Tristan Yan1,3

1The Collaborative Research Group (CORE), Sydney, Australia; 2Faculty of Medicine and Health, The University of New South Wales, Sydney, Australia; 3Department of Cardiothoracic Surgery, Royal Prince Alfred Hospital, Sydney, Australia; 4Department of Cardiac Surgery, Heart Center, University of Leipzig, Leipzig, Germany

Correspondence to: Benjamin T. Musto. The Collaborative Research Group (CORE), Sydney, NSW, Australia; Faculty of Medicine and Health, The University of New South Wales, Missenden Road, Camperdown, Sydney, NSW 2050, Australia. Email:

Background: Staged procedures are one strategy found to be beneficial for medium- to high-risk Crawford extent I–III thoraco-abdominal aortic aneurysm (TAAA) repair patients and may be performed through a variety of techniques. This review sought to compare the primary outcomes of spinal cord ischemia (SCI) and long-term mortality between three cohorts grouped by approach: open, endovascular, and hybrid.

Methods: In accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, a total of 919 references were extracted from a search of three online databases (Embase, PubMed, Scopus). Following application of inclusion/exclusion criteria and data extraction, quantitative meta-analysis was undertaken utilizing a random effects model. Kaplan-Meier (KM) curves were digitized and aggregated to graph estimated survival.

Results: A total of 20 studies representing 924 patients were included. SCI was highest in the endovascular group, at 9.8% of weighted means, followed by hybrid, and open groups at 3.2% and 1.4%, respectively. However, 30-day mortality was highest in the open group at 6.0%, followed by the hybrid group at 3.8%, and endovascular at 3.6%. Aggregated long-term survival estimations are shown graphically, extending to 5 years for open and endovascular cohorts, and 3 years for the smaller hybrid cohort.

Conclusions: While all cases incorporated spinal drainage, monitoring and staging for spinal protection, there is innate difference in approach when examining for cord ischemia. This systematic review and meta-analysis of staged TAAA repair describes the first comparison between cohorts of open and endovascular approach, revealing the increased risk of SCI and long-term mortality in endovascular repair.

Keywords: Thoraco-abdominal aortic aneurysm repair (TAAA repair); staged; spinal cord ischemia (SCI)

Submitted May 03, 2023. Accepted for publication Aug 22, 2023. Published online Sep 19, 2023.

doi: 10.21037/acs-2023-scp-20


Whether an emergent or planned operation, extensive thoraco-abdominal aortic aneurysm (TAAA) repair remains a highly complex procedure, with known risk to patient life as well as spinal cord function (1). With post-operative permanent paraplegia of up to 20%, neuroprotection strategies have been placed at the forefront of operative management in recent decades (2). Staging of operations into two or more individual procedures, usually separated by a timeframe of 1 week up to several months, is one technique which is now commonplace in expert aortic centers to reduce the significant rate of spinal cord ischemia (SCI). The aim of this concept is to provide sufficient time for the partially impaired spinal vasculature to repair itself and develop collateral networks (3). The effective preconditioning of the spinal arteries reduces the overall insult to the spinal cord, lowering the rate of major adverse outcomes compared to single-stage procedures (4). Other techniques have also been described to contribute to spinal protection, notably, cerebrospinal fluid drainage (CSFD), intraoperative monitoring, temporary aneurysm sac perfusion (TASP), and segmental artery embolization. A novel form of spinal protection, called minimally invasive segmental spinal artery coil embolization (MISSACE), first reported by Etz et al. in 2015, and currently undergoing multi-center trials in European centers, is a promising technique for future reduction of SCI in TAAA surgery (5).

Any TAAA characterized as an extent I, II, or III aneurysm, according to the Crawford classification (6), is commonly perceived as ‘extensive’, due to the propensity for these to cross the diaphragm and increase the technical complexity of the subsequent repair. While the extent V aneurysm, added by Safi et al. in 1999, has anatomical dilatation on both supra- and infra-diaphragmatic aortic segments, it does not span far enough to commonly be considered ‘extensive’ (7). It follows that the larger the extent of the aortic injury and subsequent dilatation, the larger the required repair or replacement of the native aorta. This is therefore more likely to inhibit the spinal collateral network and lead to SCI, including permanent paraplegia. This report aims to compare the spinal cord and mortality outcomes following the second stage intervention between three grouped combinations of surgical approaches. These are defined as: (I) the “open” group, consisting of two consecutive open surgical procedures, such as supra-diaphragmatic repair followed by infra-diaphragmatic repair, or vice versa; (II) the “endovascular” or “endo” group, consisting of consecutive endovascular procedures such as thoracic endovascular aortic repair (TEVAR) or fenestrated/branched endovascular aortic repair (f/bEVAR); and (III) the “hybrid” group, which for simplicity amalgamates patients who have received one open and one endovascular procedure in either order, i.e., open/endo or endo/open. Time between procedures was not controlled but rather accounted for and presented through demographic data in Table 1.

Table 1

Baseline cohort characteristics

Variables Overall Open Endovascular Hybrid P value
Studies 20 6 7 7 N/A
Patients 924 (100.0) 323 (35.0) 402 (43.5) 199 (21.5) N/A
Males 581 (62.9) 173 (53.6) 277 (68.9) 131 (65.8) 0.67
Age (years) 64.6±12.1 59.5±13.8 70.8±10.8 61.9±10.7 0.014
Crawford type N/A
   Extent I 151 112 25 14
   Extent II 326 89 87 150
   Extent III 83 5 66 12
   Other 87 55 29 3
Reporting frequency (%) 68.2 80.8 51.5 89.9
Time between stages (months) 16.2±16.8 25.0±23.0 16.1±18.0 5.2±4.1 0.62
Urgent cases 83 (9.0) 57 (17.6) 16 (4.0) 10 (5.0) 0.27
Aneurysm diameter (cm) 6.6±1.1 6.4±0.87 6.7±1.1 6.6±1.4 0.44
Comorbidities§ N/A
   Cardiac disease 130 (58.7) 58 (79.9) 37 (24.9) 35 (92.5)
   DM 97 (72.1) 21 (56.0) 55 (71.1) 21 (100.0)
   PVD 109 (34.4) 12 (20.1) 14 (23.4) 83 (79.9)
   Hypertension 729 (87.6) 176 (64.4) 373 (100.0) 180 (100.0)
   Smoking history 375 (65.8) 103 (56.0) 168 (68.4) 104 (76.4)
   CVA 56 (39.0) 24 (56.0) 3 (7.5) 29 (74.9)
Previous aortic surgery 267 (28.9) 18 (5.6) 109 (27.1) 140 (70.4) 0.39
Connective tissue disorder 74 (8.0) 38 (11.8) 7 (1.7) 29 (14.6) 0.10

Values are n (%) or mean ± SD (weighted average) unless otherwise specified. , P values are determined by comparison between open and endovascular groups (not given for variables with low reporting frequency); , other aneurysm types making up a minority of the cohort, including: extent IV, extent V and descending thoracic aortic aneurysm; §, n = reporting frequency. N/A, not applicable; DM, diabetes mellitus (any type); PVD, peripheral vascular disease; CVA, cerebrovascular accident; SD, standard deviation.


Literature search

Three electronic databases were selected to complete the initial literature search, specifically PubMed, Embase, and Scopus, from inception of records until 3rd January 2023. The search strategy employed Medical Subject Headings (MeSH) and focused keywords including: (“TAAA” OR “Thoracoabdominal aortic aneurysm” OR “Thoraco-abdominal aortic aneurysm” OR “Thoracic endovascular aortic repair” OR “TEVAR”) AND (“staged” OR “hybrid”).

After removal of duplicate records and those published before the year 2000, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines were followed in accordance with pre-written inclusion and exclusion criteria to screen the remaining records (8). Papers selected for full-text review were then exposed to stricter inclusion criteria which aimed for more homogeneity between the cohorts of included papers. Screening was conducted by two authors independently (Bilbrough J and Bushati Y) with any discrepancies being finalized through team discussion, with ultimate ruling by the leading author (Muston BT). A PRISMA diagram of the search strategy and list of records at each stage is depicted in Figure 1. Once full-text review was completed, the reference lists of all included papers were searched to assess for previously missed publications fitting the inclusion criteria.

Figure 1 PRISMA flowchart of included studies. PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Inclusion and exclusion criteria

Eligibility criteria was established prior to initial screening for this meta-analysis, focusing on maintaining a cohort of patients who had extensive repair of the aorta, spanning both thoracic and abdominal segments. For this reason, Crawford IV repairs were excluded so as to not artificially alter presented data, due to a recorded reduction of SCI in this patient group (9). In order to directly test the differences in SCI and mortality rates between the surgical interventions, similar extent of aortic repair, intervention type, and demographic data were controlled. Strict reporting of baseline patient data and procedural technique was sought to reduce confounding factors when measuring outcomes between studies. Only English language studies were included.

Studies were included if they met the following criteria: (I) the entire cohort consisted of TAAA repair patients with isolated outcome data; (II) the cohort specified Crawford types or extent of aneurysm repair; (III) data was presented following the second stage procedure; (IV) the type of repair was described (i.e., open/open, endo/endo, or a combination of both); (V) SCI was a recorded outcome. Studies were excluded if they: (I) included a pediatric population; (II) had a sample size smaller than ten patients; (III) had overlapping cohorts with larger included studies. All conference abstracts, reviews, editorials, and animal studies were also excluded.

Outcome measures

The primary outcome for this study was a pooled analysis of in-hospital SCI, reported as both a ratio with permanent paraplegia and overall temporary or permanent ischemia. SCI was defined as temporary or permanent motor or sensory dysfunction following surgery and was extracted either from direct SCI statistics or through provided modified Tarlov scale scores. Due to differences in use of modified Tarlov scales, SCI was defined as any score less than the highest available, or ‘normal’ ambulation. Paraplegia was defined as permanent motor SCI that did not resolve in the assessed post-operative period.

The secondary outcome was long-term mortality, assessed in two separate methods. The first involved all cohorts reporting mortality, which received weighted mortality analysis to be described quantitatively. The second method incorporated all manuscripts which included Kaplan-Meier (KM) curves, using a graphical approach to portray the data. Mortality was separated and compared by operative technique.

Quality assessment

The quality of each study was assessed using the modified Canadian National Institute of Health Economics (CNIHE) assessment tool for case series (10). Of a possible total of 20 criteria to be met from the CNIHE tool, a study was considered high quality if it scored 17 or higher, moderate quality if it scored between 13 and 16, and low quality if it scored 12 or below. Study quality was independently assessed by two investigators (Muston BT and Bilbrough J) with review and consensus completed by the senior author (Muston BT).

Publication bias

The evaluation of inter-study bias was performed using Stata (version 17.0, StatCorp, College Station, TX, USA) in the form of a funnel plot. The data used to conduct this graphical analysis was extracted from the reporting of the primary outcome of SCI following any of the three intervention types, as this was reported by all studies. As opposed to an odds ratio commonly used when comparing randomized controlled trials, effect sizes should be used on a funnel plot comparing single-arm studies.

Statistical analysis

Baseline characteristics and operative details were extracted from the text, tables, and figures of included papers by three independent authors (Muston BT, Bilbrough J, and Bushati Y). Discrepancies were discussed then finally reviewed by the senior author (Muston BT). Statistical analysis was carried out using Stata (version 17.0, StataCorp) and R (version 4.1.1. R Core Team, Vienna, Austria) utilizing meta-analysis of proportions and means with a random effects model where necessary. Values were considered statistically significant if the reported P value was <0.05. For continuous data with central tendency described using median values and interquartile range, the mean and standard deviation were estimated using calculations described by Wan and colleagues (11). Survival data was calculated using aggregated KM curves collected from included studies, where reported, using the methods described by Guyot and colleagues (12). Digitization of source KM curves was performed using DigitizeIt (version 2.5.9, Braunschweig, Germany) and in the case where multiple cohorts were represented on the same curve, individual KM curves were first generated then subsequently merged with the rest of the data, to be analyzed together.


After independent screening by three authors, 20 studies were included for analysis. Notably, a study by Safi et al., published as an update of a previously written report, was included, but extraction of demographic data from the original paper was required to supplement the outcome data in the latter study (13,14). Furthermore, the study by Tsilimparis et al. was represented twice, with two eligible cohorts (15). These caveats are responsible for the final inclusion of 21 cohorts from 20 papers (14-33). Nine studies were excluded for patient overlap with other included studies after previously adhering to all inclusion criteria.

Quality analysis using the CNIHE tool resulted in nine studies being classed as of a high standard of quality, 11 studies being of medium quality and zero studies being of low quality (Table S1). Therefore, no further sub-group analysis for outcome data or heterogeneity was required as low-quality evidence was not a confounding factor in this meta-analysis.

Baseline study characteristics

Baseline cohort characteristics are reported in Table 1, along with reporting frequencies for each of the three operative methods. A total of 924 patients were followed in this systematic review, of whom 581 (62.9%) were male. The studies ranged in cohort size from ten to 122. The mean age of the overall cohort was 64.6±12.1 years with 83 (9.0%) being urgent or emergency cases. Patient comorbidities were infrequently reported. The mean aneurysm size before second-stage intervention was 6.6±1.1 cm overall, with no significant difference between the operative approaches. Study details are shown in Table S1, with most originating from centers in the USA (n=11), followed by Italian (n=5) and other European centers (n=4). Notably, the time between first and second stages of TAAA repair was significantly different between the comparator groups, attributed to the fact that planned staging reduced time between stages when compared to unintentional staging groups. This is clearly seen amid the open and hybrid cohorts, which spent 25.0±23.0 and 5.2±4.1 months between operations, respectively. The endovascular sub-group was the largest, including 402 patients, followed by the open and hybrid sub-groups at 323 and 199, respectively. Crawford extent II aneurysms were the commonest overall, involving the longest extent of repaired aorta and hence increased risk of SCI. The hybrid cohort had the highest proportion of data points involving this TAAA type, at 83.8% of reported cases, while the open cohort had the lowest ratio, at just 34.1%. Furthermore, 267 of 924 patients had received previous aortic surgery (prior to staged intervention) and 74 cases were reported with an associated connective tissue disorder.

Primary outcome evaluation: long-term SCI and paraplegia

All studies included data on overall SCI rates as well as permanent paraplegia, as per previously outlined inclusion criteria. Reported long-term follow-up for the primary outcome exceeded 2 years, with a mean of 27 months between studies. Pooled rate of SCI (temporary or permanent) was 5.4% [95% confidence interval (CI), 5.1–5.8%; I2=31.7%; Table 2]. When subgroups were compared, the open approach resulted in the lowest proportion of SCI, at 1.4% (95% CI, 1.3–1.5%; I2=0.0%), followed by the hybrid cohort with 3.2% (95% CI, 2.8–3.6%; I2=0.0%) and the endovascular approach, which showed the highest presence of SCI postoperatively, at 9.8% (95% CI, 9.2–10.4%; P<0.01; I2=23.8%) of cases.

Table 2

Operative outcomes and early morbidity

Operative outcomes Overall Open Endovascular Hybrid P value I2 (%)
Patients 924 (100.0) 323 (35.0) 402 (43.5) 199 (21.5) N/A
SCI (%) 5.4 1.4 9.8 3.2 <0.01 31.7
Paraplegia (%) 2.0 0.69 2.6 2.9 0.17
30-day mortality (%) 4.6 6.0 3.6 3.8 0.069 55.4
Hospital LOS (days) 19.5±9.0 26.6±7.4 15.2±7.6 14.8±5.9 0.34
ICU LOS (days) 11.2±7.4 16.1±7.7 5.3±1.7 7.4±2.0 0.42
Adverse events N/A
   MI 6 (24.2) 1 (22.6) 3 (22.1) 2 (31.2)
   Stroke/CVA 15 (41.3) 8 (64.4) 1 (15.4) 6 (56.3)
   AKI 83 (57.9) 25 (64.4) 8 (38.1) 50 (87.4)
   Respiratory failure 16 (15.6) 7 (10.8) 5 (15.4) 4 (23.6)
Endoleak 2 (8.0) 11 (36.2) 0.29§

Values are n (%) or mean ± SD (weighted average) unless otherwise specified. , P values are determined by comparison between open and endovascular groups; , n = reporting frequency; §, endoleak assessed for endovascular and hybrid groups. N/A, not applicable; SCI, spinal cord ischemia; LOS, length of stay; ICU, intensive care unit; MI, myocardial infarction; CVA, cerebrovascular accident; AKI, acute kidney injury; SD, standard deviation.

Following sensitivity analysis, these results remained robust, with minimal deviation from the pooled effect size. Permanent paraplegia was also quantified using weighted means, showing the open approach subgroup to have the lowest rate at 0.7% (95% CI, 0.6–0.8%). When this metric is divided by the weighted means for SCI, the propensity for SCI to become permanent paralysis/paraplegia was highest in the hybrid subgroup, at 90.5%. The endovascular approach proved to be more favorable in this analysis than the open approach, with paraplegia occurring from associated SCI 26.2% vs. 49.9% of the time, respectively.

Secondary outcome evaluation: 30-day mortality following second-stage intervention

Data for the analysis of 30-day mortality were present in 19 studies, totaling 808 of 924 patients. The overall pooled 30-day mortality rate was 4.6% (95% CI, 4.3–4.8%; I2=55.4%; Table 2). Pooled mortality was greater in the open cohort than the endovascular cohort, at 6.0% (95% CI, 5.5–6.4%) vs. 3.6% (95% CI, 3.2–3.9%), however, this was not statistically significant (P=0.069). Hybrid operations achieved a similar 30-day mortality rate to the endovascular approach (3.8%; 95% CI, 3.3–4.2%).

KM curves were used to assess the less frequently reported long-term mortality rate and were aggregated and digitized to present this extended mortality rate (Figures 2,3). Actuarial survival at 1, 3, and 5 years for the open cohort was 76.5%, 61.7%, and 55.0%, respectively. This compared to the 1- and 3-year survival of the endovascular cohort (76.1% and 45.8%, respectively) shows a similar initial level of mortality but superior lasting survival for the open cohort. Five-year data was not presented for the endovascular cohort. Interestingly, the hybrid cohort showed the most impressive actuarial survival data, with 92.7%, 88.8%, and 85.8% at 1, 3, and 5 years, respectively. However, this was the least frequently reported and is likely impacted by bias and confounders.

Figure 2 KM curve showing non-parametric survival analysis of: (A) endovascular cohort; (B) hybrid cohort; (C) open cohort. KM, Kaplan-Meier.
Figure 3 KM curve overlay for all included cohorts. KM, Kaplan-Meier; endo, endovascular.

Spinal cord protection techniques and comorbidities

All but two studies reported using some form of spinal cord protection outside of standard practice of induced hypothermia and rewarming. In total, 597 patients from a possible 805 patients received CSFD, whether prophylactic or interventional, following surgery. An additional 46 patients from the endovascular cohort were reported to have received TASP as an alternative SCI prevention modality. Limited data were presented on use of motor evoked potential (MEP) and somatosensory evoked potential (SSEP) monitoring from collected studies. Major adverse events following second-stage procedures are listed in Table 2. As expected, hospital and intensive care unit (ICU) length of stay (LOS) were longest in the open group, at 26.6±19.1 and 16.1±12.6 days, respectively. Notably, acute kidney injury (AKI) was more common in the open subgroup than those undergoing endovascular intervention, at 4.5% (95% CI, 4.0–5.0%) vs. 1.6% (95% CI, 1.5–1.7%), respectively.

Study bias evaluation

Publication bias was assessed using a funnel plot, with data provided from the pooled primary outcome of all included studies, in this case being SCI after intervention. The resulting funnel plot (Figure 4) shows some skew, suggesting the possibility of some bias in the publication of results between studies. The reasons for this skew are likely due to publication bias, heterogeneity between studies, differing methodology or random chance. Notably, this meta-analysis showed some moderate heterogeneity, requiring the use of a random-effects model, as despite I2<50%, the Q test was not statistically significant (P=0.2).

Figure 4 Funnel plot for SCI overall. CI, confidence interval; SCI, spinal cord ischemia.


While the benefit of staging as a method of spinal cord protection for extensive TAAA repair has long been recognized, with recent meta-analyses being performed for both open and endovascular cohorts (34,35), comparisons between staged cohorts have been lacking. Moreover, the current literature has a paucity of well-defined staged repair cohorts, whereby the morbidity and mortality data are published for a homogeneous patient group which can then be compared between centers. Nevertheless, this systematic review reported a pooled SCI rate of 5.4% and a pooled paraplegia rate of 2.0% from 924 patients distributed between 21 studies. The discrepancy between surgical approaches was noteworthy, showing open repair to be most favorable when solely analyzing this metric at 1.4%, and endovascular repair to be least favorable, at 9.8% of patients suffering SCI. These figures are verified in the literature yet have not previously been directly compared. Pini et al., during a very comprehensive meta-analysis of 18 studies reporting endovascular TAAA repair for Crawford extents I, II, III, and V, found a similar rate of 9% when reporting overall SCI percentage (34). The rate of permanent paraplegia in their review were found to be slightly higher, at 6%, where this present review reports a rate of 2.6%. However, this may be due to variable reporting of permanent paraplegia between studies. This study was also valuable in portraying the importance of excluding the Crawford IV cohort when completing SCI analysis. A significantly lower percentage of patients suffered SCI in their Crawford IV cohort when compared to all other Crawford extents, at 6% vs. 13%, respectively. This is evidence for separation of these patients in future analyses in order to preserve homogeneity when comparing staging approaches. The reduced aortic coverage during repair of the extent IV aneurysm cohort can artificially reduce the risk of SCI when grouped into an overall TAAA repair population, which is why Crawford IV patients were excluded in this analysis.

Mortality differences also appeared between approaches for staged TAAA, however, this was most elucidated in the short-term. We found pooled 30-day mortality to be highest in the open cohort and lowest in the endovascular cohort, at 4.6% and 3.6%, respectively. Interestingly, this is a reversal of the order found in SCI outcomes. Differences in complexity between cases may be a reason for this discrepancy, as it is well established that surgical approach will be dictated based on the morphology of the aneurysm and complexity of the case. Usually, increased complexity leads to an open approach, which may be a confounding factor affecting the rate of mortality. An analysis of long-term mortality was also conducted, using digitized and aggregated KM curves for each surgical approach (Figure 2). While this is a useful visual representation of the survival data over 3 or more years, a relatively low sample size limits the comparability of the data. Four of the six open repair cohorts supplied a KM curve for aggregation, while 3/7 hybrid and only 2/8 endovascular cohorts had a KM curve able to be included in the final digitized graph. Regardless, at the 36-month interval, the hybrid group had the highest survival, at 88.7%, followed by the open group at 61.7%. The endovascular cohort had too few numbers-at-risk to make a meaningful interpretation of the aggregated survival estimate (Figure 3). It should also be noted that mortality was only assessed for the period during and following second stage intervention in this review, and that this only represents one of the four periods where the patient is at risk of death (Figure 5). Post-operative mortality was used in this review due to reporting frequency in included studies, however this is not to shroud or minimize the apparent risk of death for staged procedures. When using this review to report mortality in comparison to any non-staged procedures, a point of clarification must be made to also involve inter-operative mortality to fully express the incidence of death.

Figure 5 Timeline of possible mortality. (A) Single stage TAAA repair; (B) two-stage TAAA repair. Most reported risks are during each operation (blue boxes), or solely in the postoperative period, however total risk should be marked by mortality in its entirety. Op, operation; TAAA, thoraco-abdominal aortic aneurysm.

The nature of endovascular aortic repair using long landing zones, compared to the precise sutured attachment of open grafting, increases the extent of aortic coverage, making this approach more liable to SCI. Tenorio et al. suggest that this is a potential shortcoming of the endovascular repair in their review of the literature (36), which may be reflected in the present study’s results as both endo and hybrid groups were inferior to the open group. Previous literature has established the importance that aortic coverage holds when analyzing rates of SCI (37,38), whether by multiple stent grafts or through extensive involvement by open repair, first denoted by Greenberg and colleagues (39). Multivariate analysis in a study assessing neurological complications associated with endovascular repair of the aorta by Buth et al. found statistically significant correlation between aortic length coverage and rates of SCI, an issue less prevalent in open repair (40).

The hybrid group described in this review included a mixed population undergoing TEVAR, f/bEVAR, frozen elephant trunk and open procedures as their first or second stage. However, this sub-group did have a high proportion of Crawford extent II patients (83.8%) and intentionally staged procedures, leading to a short, pooled interval between operations of 5.2 months. Good long-term survival was also shown through the composite KM curve (Figure 2) with 3-year survival of 86%. Our staged hybrid cohort proved to have substantially more promising results than a previous meta-analysis conducted by Moulakakis et al., who found 14 single-stage hybrid cohorts comprising of 528 patients (41). This study established an operative mortality of 14.3% and a risk of SCI at 7.0%, both of which were higher than our results, which should be attributed to the lack of staging.

Further research should be conducted not only on prospective patient-matched randomized controlled trials, in order to make comparisons more directly between the surgical cohorts, but also longitudinal studies taking into account mortality rates between and after staged procedures, whether they are done with an open, endovascular or hybrid approach. This would allow incidence statistics to be unveiled regarding the risks of delaying complete aneurysmal repair during the staging process.


Heterogeneity between and within sub-groups formed the predominant basis for impaired validity in this study. Firstly, between groups, inconsistent reporting of aortic coverage and co-morbidities as well as varying prevalence of Crawford extent II reduces the ability for the cohorts to be fairly compared to each other. While a great degree of caution was undertaken during selection of inclusion criteria, population differences still exist between intervention groups. Secondly, within groups, there was a notable difference in the time interval between staged procedures, which can be attributed to the number of intentionally vs. unintentionally staged patients. This could possibly skew survival data in the favor of those with longer intervals as these patients have survived longer after the first procedure and are therefore more likely to be stable and less complicated (13). Some cohorts within each sub-group also received different procedures despite being merged under the same heading. For example, the endovascular cohort had patients undergoing f/bEVAR initially before then receiving TEVAR, while other patients had both operations in the reverse order.

Furthermore, a preponderance for retrospective observational studies with most (15/21) analyzing cohorts of fewer than 40 patients can limit the generalizability of these data. Many of these reports (17/21) pulled data from only one institution, which heightens the impact of surgeon skill and the protocols of the single center on the survival outcomes.


This systematic review and meta-analysis describes the post-operative rates of SCI, permanent paraplegia and mortality following staged repair of TAAAs. We compared the outcomes between cohorts receiving total open, total endovascular and hybrid repair of the aorta and found that the open cohort had the lowest rate of SCI and permanent paraplegia, at just 1.4% and 0.7%, respectively. Hybrid repair showed promising long-term survival at 5-year with preliminary data. Overall, this study describes the first comparison between cohorts of open and endovascular approach, revealing the increased risk of SCI and long-term mortality in endovascular repair.


Funding: None.


Conflicts of Interest: The authors have no conflicts of interest to declare.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See:


  1. Bashir M, Harky A, Adams B, et al. Renal protection in thoracoabdominal aortic aneurysm surgery. Gen Thorac Cardiovasc Surg 2019;67:192-5. [Crossref] [PubMed]
  2. Etz CD, Weigang E, Hartert M, et al. Contemporary spinal cord protection during thoracic and thoracoabdominal aortic surgery and endovascular aortic repair: a position paper of the vascular domain of the European Association for Cardio-Thoracic Surgery†. Eur J Cardiothorac Surg 2015;47:943-57. [Crossref] [PubMed]
  3. Heidemann F, Tsilimparis N, Rohlffs F, et al. Staged procedures for prevention of spinal cord ischemia in endovascular aortic surgery. Gefasschirurgie 2018;23:39-45. [Crossref] [PubMed]
  4. Gombert A, Kirner L, Ketting S, et al. Editor's Choice - Outcomes After One Stage Versus Two Stage Open Repair of Type II Thoraco-abdominal Aortic Aneurysms. Eur J Vasc Endovasc Surg 2019;57:340-8. [Crossref] [PubMed]
  5. Etz CD, Debus ES, Mohr FW, et al. First-in-man endovascular preconditioning of the paraspinal collateral network by segmental artery coil embolization to prevent ischemic spinal cord injury. J Thorac Cardiovasc Surg 2015;149:1074-9. [Crossref] [PubMed]
  6. Crawford ES, Crawford JL, Safi HJ, et al. Thoracoabdominal aortic aneurysms: preoperative and intraoperative factors determining immediate and long-term results of operations in 605 patients. J Vasc Surg 1986;3:389-404. [Crossref] [PubMed]
  7. Safi HJ, Miller CC 3rd. Spinal cord protection in descending thoracic and thoracoabdominal aortic repair. Ann Thorac Surg 1999;67:1937-9; discussion 1953-8. [Crossref] [PubMed]
  8. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA Statement. Open Med 2009;3:e123-30.
  9. Coselli JS, LeMaire SA, Preventza O, et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg 2016;151:1323-37. [Crossref] [PubMed]
  10. Moga C, G.B., Schopflocher D, Harstall C. Development of a quality appraisal tool for case series studies. in 21st Cochrane Colloquium (John Wiley & Sons, Québec City, Canada, 2013).
  11. Wan X, Wang W, Liu J, et al. Estimating the sample mean and standard deviation from the sample size, median, range and/or interquartile range. BMC Med Res Methodol 2014;14:135. [Crossref] [PubMed]
  12. Guyot P, Ades AE, Ouwens MJ, et al. Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol 2012;12:9. [Crossref] [PubMed]
  13. Safi HJ, Miller CC 3rd, Estrera AL, et al. Staged repair of extensive aortic aneurysms: long-term experience with the elephant trunk technique. Ann Surg 2004;240:677-84; discussion 684-5. [Crossref] [PubMed]
  14. Safi HJ, Miller CC 3rd, Estrera AL, et al. Optimization of aortic arch replacement: two-stage approach. Ann Thorac Surg 2007;83:S815-8; discussion S824-31. [Crossref] [PubMed]
  15. Tsilimparis N, Haulon S, Spanos K, et al. Combined fenestrated-branched endovascular repair of the aortic arch and the thoracoabdominal aorta. J Vasc Surg 2020;71:1825-33. [Crossref] [PubMed]
  16. Bertoglio L, Kahlberg A, Gallitto E, et al. Role of historical and procedural staging during elective fenestrated and branched endovascular treatment of extensive thoracoabdominal aortic aneurysms. J Vasc Surg 2022;75:1501-11. [Crossref] [PubMed]
  17. Coselli JS, Krause HM, Green SY, et al. A 23-year experience with the reversed elephant trunk technique for staged repair of extensive thoracic aortic aneurysm. J Thorac Cardiovasc Surg 2022;163:1252-64. [Crossref] [PubMed]
  18. Di Marco L, Murana G, Leone A, et al. Hybrid repair of thoracoabdominal aneurysm: An alternative strategy for preventing major complications in high risk patients. Int J Cardiol 2018;271:31-5. [Crossref] [PubMed]
  19. Etz CD, Zoli S, Mueller CS, et al. Staged repair significantly reduces paraplegia rate after extensive thoracoabdominal aortic aneurysm repair. J Thorac Cardiovasc Surg 2010;139:1464-72. [Crossref] [PubMed]
  20. Gallitto E, Faggioli G, Fenelli C, et al. Multi-Staged Endovascular Repair of Thoracoabdominal Aneurysms by Fenestrated and Branched Endografts. Ann Vasc Surg 2022;81:48-59. [Crossref] [PubMed]
  21. Gombert A, Ketting S, Rückbeil MV, et al. Perioperative and long-term outcome after ascending aortic and arch repair with elephant trunk and open thoracoabdominal aortic aneurysm repair. J Vasc Surg 2022;75:824-32. [Crossref] [PubMed]
  22. Haensig M, Schmidt A, Staab H, et al. Endovascular Repair of the Thoracic or Thoracoabdominal Aorta Following the Frozen Elephant Trunk Procedure. Ann Thorac Surg 2020;109:695-701. [Crossref] [PubMed]
  23. Hawkins RB, Mehaffey JH, Narahari AK, et al. Improved outcomes and value in staged hybrid extent II thoracoabdominal aortic aneurysm repair. J Vasc Surg 2017;66:1357-63. [Crossref] [PubMed]
  24. Hughes GC, Andersen ND, Hanna JM, et al. Thoracoabdominal aortic aneurysm: hybrid repair outcomes. Ann Cardiothorac Surg 2012;1:311-9. [Crossref] [PubMed]
  25. Iannacone EM, Robinson NB, Rahouma M, et al. Elephant trunk simplifies thoracoabdominal aortic aneurysm repair without impacting operative risk. J Card Surg 2022;37:4685-91. [Crossref] [PubMed]
  26. Kawajiri H, Tenorio ER, Khasawneh MA, et al. Staged total arch replacement, followed by fenestrated-branched endovascular aortic repair, for patients with mega aortic syndrome. J Vasc Surg 2021;73:1488-1497.e1. [Crossref] [PubMed]
  27. King RW, Dias AP. Staging Endovascular Thoracic and Thoracoabdominal Aortic Aneurysm Repairs and the Risk of Post-operative Spinal Cord Ischemia. Ann Vasc Surg 2022;85:299-304. [Crossref] [PubMed]
  28. LeMaire SA, Carter SA, Coselli JS. The elephant trunk technique for staged repair of complex aneurysms of the entire thoracic aorta. Ann Thorac Surg 2006;81:1561-9; discussion 1569. [Crossref] [PubMed]
  29. O'Callaghan A, Mastracci TM, Eagleton MJ. Staged endovascular repair of thoracoabdominal aortic aneurysms limits incidence and severity of spinal cord ischemia. J Vasc Surg 2015;61:347-354.e1. [Crossref] [PubMed]
  30. Orrico M, Ronchey S, Setacci C, et al. The "bare branch" for safe spinal cord ischemia prevention after total endovascular repair of thoracoabdominal aneurysms. J Vasc Surg 2019;69:1655-63.
  31. Spinella G, Finotello A, Pisa FR, et al. Geometrical Evaluation of Aortic Sac Remodeling During Two-Step Thoracoabdominal Aortic Aneurysm Endovascular Repair. Ann Vasc Surg 2020;67:43-51. [Crossref] [PubMed]
  32. Thompson MA, Lowry AM, Caputo F, et al. Ultra-Hybrid Repair: Open Thoracoabdominal Completion After Descending Stent Grafting. Semin Thorac Cardiovasc Surg 2022; Epub ahead of print. [Crossref]
  33. Vivacqua A, Idrees JJ, Johnston DR, et al. Thoracic endovascular repair first for extensive aortic disease: the staged hybrid approach†. Eur J Cardiothorac Surg 2016;49:764-9. [Crossref] [PubMed]
  34. Pini R, Faggioli G, Paraskevas KI, et al. A systematic review and meta-analysis of the occurrence of spinal cord ischemia after endovascular repair of thoracoabdominal aortic aneurysms. J Vasc Surg 2022;75:1466-1477.e8. [Crossref] [PubMed]
  35. Sharples L, Sastry P, Freeman C, et al. Endovascular stent grafting and open surgical replacement for chronic thoracic aortic aneurysms: a systematic review and prospective cohort study. Health Technol Assess 2022;26:1-166. [Crossref] [PubMed]
  36. Tenorio ER, Dias-Neto MF, Lima GBB, et al. Endovascular repair for thoracoabdominal aortic aneurysms: current status and future challenges. Ann Cardiothorac Surg 2021;10:744-67. [Crossref] [PubMed]
  37. Gravereaux EC, Faries PL, Burks JA, et al. Risk of spinal cord ischemia after endograft repair of thoracic aortic aneurysms. J Vasc Surg 2001;34:997-1003. [Crossref] [PubMed]
  38. Marcheix B, Dambrin C, Bolduc JP, et al. Midterm results of endovascular treatment of atherosclerotic aneurysms of the descending thoracic aorta. J Thorac Cardiovasc Surg 2006;132:1030-6. [Crossref] [PubMed]
  39. Greenberg R, Resch T, Nyman U, et al. Endovascular repair of descending thoracic aortic aneurysms: an early experience with intermediate-term follow-up. J Vasc Surg 2000;31:147-56. [Crossref] [PubMed]
  40. Buth J, Harris PL, Hobo R, et al. Neurologic complications associated with endovascular repair of thoracic aortic pathology: Incidence and risk factors. a study from the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) registry. J Vasc Surg 2007;46:1103-1110; discussion 1110-1.
  41. Moulakakis KG, Mylonas SN, Antonopoulos CN, et al. Combined open and endovascular treatment of thoracoabdominal aortic pathologies: a systematic review and meta-analysis. Ann Cardiothorac Surg 2012;1:267-76. [Crossref] [PubMed]
Cite this article as: Muston BT, Bilbrough J, Bushati Y, Wilson-Smith AR, Misfeld M, Yan T. Open, closed or a bit of both: a systematic review and meta-analysis of staged thoraco-abdominal aortic aneurysm repair. Ann Cardiothorac Surg 2023;12(5):418-428. doi: 10.21037/acs-2023-scp-20

Article Options

Download Citation