The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines were followed [12]. Randomized controlled trials (RCTs) and prospective or retrospective observational studies, published between 2000 and 2022, of the English medical literature, reporting on technical success, 30-day mortality, and mortality during follow-up in patients with acTBAD managed with the PETTICOAT or STABILISE technique were eligible and incorporated in the current systematic review. Studies reporting on type A aortic dissection or subacute or chronic TBAD were not considered eligible. In case that a study reported mixed population findings, it was considered eligible, only if outcomes on acute cases could be safely extracted. Furthermore, studies reporting only on TEVAR outcomes, open or hybrid repair were excluded. Case reports and case series with less than 5 patients were also omitted.

A systematic search via Ovid, of MEDLINE and EMBASE, and CENTRAL databases, was conducted with an endpoint set for August 31st, 2022. The PICO model [Patient; Intervention; Comparison; Outcome (Supplementary Table 1)] was applied [13]. The following search items, including Expanding Medical Subject Heading (MeSH terms), were used in various combinations (Table 1): (acute), (complicated), (dissection), (PETTICOAT), (STABILISE), (bare metal stent), (endovascular repair), (technical success), (mortality). Scrutiny was accomplished independently after full-text assessment by two investigators (P.N., K.S.) and discrepancies were resolved after discussion with a third investigator (T.K.).

A Microsoft Excel (Office 365, Microsoft, Redmond, WA, USA) file was generated. Extracted data included study characteristics (authors, journal, date of publication or acceptance, study design, study period, country/center/database, aim) in addition to general information [demographics (age, sex), indication to treat (malperfusion, rupture/impending rupture), technique (PETTICOAT, STABILISE) and technical details (type of endograft, type of bare metal stent, distal extension, balloon, stenting of aortic branches, duration of operation)]. Technical success, mortality at 30-days and mortality during the available follow-up were recorded. Morbidity rupture, stent induced entry tear (SINE), retrograde dissection, endoleak type 1 (EL 1), renal insufficiency, malperfusion, cerebrovascular events (stroke and transient ischemic attack), spinal cord ischemia (SCI; paresis or paraplegia) at 30-days was recorded and analyzed. The available follow-up of each study was extracted when reported. The imaging method of surveillance, false lumen (FL) thrombosis rate of the thoracic and abdominal aorta, any remodeling data, including aortic diameter and volume, were assessed when available. Regarding follow-up outcomes, mortality, rupture, retrograde dissection, EL 1, re-intervention and open conversion were recorded and analyzed. Missing data assessment and funding information were also extracted when available. Regarding potential overlapping studies, the latest available data were included in the analysis.

The quality of the included studies was assessed with the Newcastle-Ottawa Scale (NOS, Supplementary Table 2a) while for the RCT the JADAD tool was used (Supplementary Table 2b) [14, 15]. NOS appraises three main methodological domains: selection methods, comparability on design or analysis, and assessment of outcomes. Individual studies were attributed a higher risk of bias in cases of inadequate confounder control and retrospective nature. Furthermore, any potential loss to follow-up or missing data that was not clearly stated in text were considered an additional confounder. The scale consists of a star system, with a maximum of nine stars. Studies achieving at least seven stars were characterized of higher quality [14]. JADAD is a multidisciplinary panel of six judges which are used to determine the effect of rater blinding on the assessments of quality. The final version of the instrument includes three items. These items were scored consistently by all the raters, as blind assessments produced significantly more consistent scores [15].

The primary outcomes were technical success and 30-day mortality in patients that underwent acTBAD management using the PETTICOAT or STABILISE technique. The mortality during the available follow-up was considered a secondary outcome.

As there was a significant heterogeneity among studies, especially for anatomic modifications during follow-up, the definitions reported by each study are displayed in Table 2 (Ref. [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]).

Continuous data were reported as a mean $\pm$ standard deviation. Categorical data were expressed as absolute numbers with the associated range. The effect of measures for technical success, early and follow-up mortality, as well as the remaining outcomes were presented as percentages or proportions of the included studies for each outcome. For missing data, there was no imputation and the effect of measure of each outcome was estimated on the cohort of the studies reporting on each specific outcome. No comparison between the techniques was executed. Statistical analyses used SPSS 20.0 software (IBM Corp, Armonk, NY, USA).

The initial search yielded 3128 articles. Deduplication was performed automatically using Ovid (474 studies excluded). After exclusion of studies according to the previously reported criteria, thirteen studies were included in this systematic review (Fig. 1) [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]. Three studies were prospective observational studies while one was a randomized controlled trial [10, 18, 24, 26]. The remaining studies were retrospective. Regarding the study of Sobocinski et al. [21], only the anatomic modification data during follow-up were extracted and presented in this analysis in order to overcome any potential overlap with previously reported patients’ outcomes. In studies, reporting on acute, subacute, and chronic cases, only data regarding acTBAD were included [18, 21, 24, 27].

Fig. 1.

PRISMA flow. The initial search yielded 3128 articles. After exclusion of studies according to the reported criteria, thirteen studies were included in this systematic review.

In total, 418 patients with acTBAD managed with the PETTICOAT (83%; 346/418) or STABILISE (17%; 72/413) techniques were included; 254 were males (81.4%, 254/312) [17, 18, 19, 22, 23, 24, 25, 26, 27]. The mean age was 56.0 $\pm$ 10.1 years [18, 19, 20, 23, 24, 25, 26, 27]. The indication for TEVAR was described in all studies. The specific indications, including true lumen collapse, FL expansion and persisting high flow to the FL per study are depicted in Table 3 (Ref. [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]), along with the studies’ main characteristics. Eleven studies reported outcomes regarding the PETTICOAT technique while two provided data on the STABILISE approach [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]. In one study, the extended PETTICOAT technique was applied [25].

Four studies reported specific anatomic preoperative characteristics, including aortic diameter and volume, as displayed in Supplementary Table 3 (Ref. [19, 21, 25, 27]). In nine studies, left subclavian artery (LSA) management was reported in detail [10, 17, 19, 20, 21, 22, 23, 25, 27]. In one study, LSA occlusion related to upper limb ischemia was managed conservatively using alprostadil [17]. In the remaining studies, LSA revascularization was performed using conventional bypass from the left common carotid artery (44 cases), or the periscope technique (13 cases) [10, 19, 20, 21, 22, 23, 25, 27].

Regarding the type of bare metal stent, in seven studies the Zenith Dissection Endovascular System (Cook Medical, Bloomington, IN, USA) was used [10, 16, 18, 22, 24, 27]. A combination of the Zenith Dissection Endovascular System and Zenith TX2 endograft for proximal coverage was reported in six of them [10, 16, 18, 24, 27]. In the remaining studies a variety of devices has been used, as depicted in Table 4 (Ref. [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]), along with technical details reported in each study. The reported oversizing of the endograft ranged between 0–15% while the oversizing of the bare metal stent was 0–25% [18, 20, 21, 23, 24, 26]. Three studies provided data on the STABILISE technique and reported the use of a molding balloon with diameter 26 mm–42 mm or 46 mm, dilated up to 2–4 atmospheres , to achieve lamina disruption and stabilization [22, 23, 25].

Three studies reported further management of aortic branches with the application of additional stenting when indicated, as in case of persistent malperfusion or dissection extension [22, 23, 25]. In total, 30 stents were deployed into provisionally selected target vessels [22, 23, 25]. One analysis reported the use of extended PETTICOAT with common iliac artery endograft deployment (Endurant, Medtronic, Santa Anna, CA, USA) [25]. In this study, the endograft limbs were extended into the aortic bare metal stent up to renal arteries, as kissing limbs [25]. Two studies reported the use of self-expanding stents for external iliac artery coverage, in one of them, covered stents were selected [23, 25].

Regarding the intra-operative details, only four studies provided data on operation time while two of them also reported the fluoroscopy duration and three, the contrast volume [10, 17, 19, 20]. The estimated duration of operation was 112 minutes (range 54–519 minutes) [10, 17, 19, 20]. The fluoroscopy time was 13.5 minutes (median; range 11–17 minutes) in one study and 27 $\pm$ 7 minutes in the second one [17, 19]. Contrast volume was estimated at 238 mL (range 89–400 mL) [17, 19, 20].

Technical success was stated in six studies, three of them reported on the STABILISE technique [17, 19, 22, 23, 25, 27]. The estimated technical success rate was 99.5% (range 97.1–100%), 99% for the PETTICOAT and 100% for the STABILISE cohort [22, 23, 25].

Mortality at 30-days was reported in ten studies [10, 16, 17, 18, 19, 20, 22, 23, 25, 27]. In total, twelve deaths were recorded with an estimated 30-day mortality of 3.7% (12/321). Three deaths were related to aortic rupture, leading to an estimated early aorta-related death rate of 1.2% (3/247) [10, 17, 18, 19, 23, 25, 27]. When the cohort was separated into subgroups, the 30-day mortality was 1.4% (1/72) for the STABILISE technique [22, 23, 25], and 4.4% for the PETTICOAT approach (11/249) [10, 16, 17, 18, 19, 20, 27]. All aortic ruptures were reported in the PETTICOAT subgroup, leading to an early aorta-related death rate of 1.4% (3/216) [10, 17, 18, 19, 27].

SINE was recorded in four studies, two of them reporting on the STABILISE technique [17, 23, 25, 27]. Two events were reported, both patients were managed with the PETTICOAT technique [total cohort rate: 2.3% (2/84), PETTICOAT rate: 3.8% (2/57)] [17, 27]. Three cases of retrograde type A aortic dissection were extracted from studies, all belonged to the PETTICOAT group [total cohort rate: 1.6% (3/183), PETTICOAT rate: 2.3% (3/128)].

Regarding post-operative morbidity, 17 cases of renal insufficiency were reported, all in the PETTICOAT cohort [total cohort rate: 6.3% (17/268), PETTICOAT rate: 8.7% (17/196), STABILISE rate: 0% (0/72)] [10, 17, 18, 19, 22, 23, 25]. Regarding cerebrovascular events, 14 adverse events were recorded, two of them in the STABILISE group with a rate of 4.9% (14/288) for the total cohort and, 5.6% (12/216) and 2.8% (2/72) for the PETTICOAT and STABILISE techniques, respectively [10, 17, 18, 19, 22, 23, 25, 27]. SCI was reported in eleven cases (4.3%, 11/253), five of them were characterized as paresis (2.8%, 5/180) and two as paraplegias (0.9%, 2/215) [10, 17, 18, 19, 22, 23, 25, 27]. Among the SCI events, four were recorded to the STABILISE cohort (5.6%, 4/72) and the remaining to the PETTICOAT group (3.9%, 7/181) [10, 17, 18, 19, 22, 23, 25, 27].

All studies reported on the available follow-up which was estimated at 20 months (range 3–168 months) [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]. For the studies reporting on the PETTICOAT technique the estimated follow-up was 22 months (3–40 months) and for the STABILISE technique, it was 17 months (1–168 months). Regarding mortality, 23 patients died during the available follow-up: 7 cases were reported as aorta-related and 4 ruptures were reported [10, 17, 18, 19, 20, 22, 23, 24, 25, 26, 27]. The estimated mortality rate was 5.7% (23/402), with an aorta-related mortality rate at 33% (7/21) for the total cohort [10, 18, 19, 22, 23, 24, 26]. When the cohort was separated into subgroups, the mortality during follow-up was 0% (0/72) for the STABILISE technique [22, 23, 25], and 7.0% for the PETTICOAT approach (23/330) [10, 18, 19, 22, 23, 24, 26]. All aorta-related deaths and ruptures were recorded among patients that were managed with the PETTICOAT technique.

Regarding post-operative adverse events, eleven cases of retrograde type A aortic dissection were reported [10, 16, 18, 19, 22, 23, 24]. The rate for the total cohort was estimated at 3.8% (11/289), for the PETTICOAT group it was 4.3% (10/234) and for the STABILISE, 1.8% (1/55). endoleak type 1A (EL 1A) was reported in eight studies [10, 17, 22, 23, 24, 25, 26, 27]. Twelve endoleaks 1A (ELs 1A) were recorded leading to 3.7% rate (11/295) for the total cohort. Re-interventions were reported in seven studies while the estimated reinterventione rate during follow-up was 10.1% (32/315) [10, 18, 19, 22, 23, 24, 26]. Twenty-three events were detected in the PETTICOAT group (8.8%, 23/260) and nine in the STABILISE group (16.3%, 9/55). Among them one open conversion has been recorded in the study published by Lombardi et al. [24].

As a significant heterogeneity in definitions and assessment of anatomic aortic modifications after TEVAR was detected in the included studies. The anatomic modifications are displayed in Table 5 (Ref. [10, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]), as presented by the included studies. Studies reported a favourable remodelling (expansion of true lumen and total aortic diameter stabilization compared to pre-operative aortic diameters) ranging from 17.6 to 100% for the thoracic aorta [17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27]. In two studies, a true lumen expansion was detected during follow-up compared to the pre-operative estimation [20, 26].

Five out of thirteen studies were considered of high quality ($>$7 stars). The remaining were characterized as low quality (61.5%), due to small number of cases, surgeon, and patients’ selection according to surgeons’ experience and patients’ anatomic characteristics, and limited follow-up and missing data. Regarding the RCT, the application of JADAD demonstrated a moderate quality.

The retrospective nature of most of the included studies introduced certain bias and residual confounders. Studies reporting only on acute cases of TBAD and the use of PETTICOAT and STABILISE techniques were included a priori in this analysis. Thus, further details on both techniques and in other pathologic conditions are missing. The risk of bias varied considerably among studies. Furthermore, technical success, specific patient selection criteria, materials used, and follow-up data were not available in all studies. Variable procedures were performed, including additional stenting of the aortic branches and iliac arteries, that may have affected the potential outcomes, including clinical and anatomic findings. Especially for the PETTICOAT technique, the length and type of the deployed bare metal stents was under-reported and varied among studies. This fact potentially affected the outcomes of the included studies, and further, the findings of the current review. Regarding specific definitions, the heterogeneity was significant among studies, especially when considering true lumen collapse as an indication for repair and further, the methodologic assessment of aortic remodelling from the pre-operative to follow-up setting. Different approaches, including diameter and volumetric analyses, as well as estimation of them in variate anatomic positions did not permit a further estimation of the impact of PETTICOAT and STABILISE in aortic remodelling. Along these lines, a mathematical analysis could not be executed. As case reports and small case series were excluded, the findings of this analysis might have been affected. A meta-analysis could not be excecuted due to lacking comparative data between the techniques. The available follow-up was restricted to 20 months and long-term data are lacking from the literature.