To supplement the data for the pooled analysis, we retrieved our data on all consecutive patients treated using the MFM between May 2013 and August 2020. Because five of our patients were also included in the study by Ibrahim et al. [8], we excluded these patients from this data set. The study was approved by the local ethics committee (University Witten/Herdecke, S-143/2022), patient´s informed consent was waived.

In our hospital, MFM was used in patients with a type B aortic dissection and thoracoabdominal or thoracic aneurysms who were unfit for open surgery.

A computed tomography (CT) angiography was performed preoperatively. According to the instructions for use, a minimal overlap of at least 6 cm for straight aortas and 8 cm for angulated aortic segments should be used (Supplementary Table 1). In addition, there has to be a proximal and distal landing zone of normal arterial wall of at least 2 cm; stenotic side branches have to be treated with stent implantation prior to implantation of the MFM, and stents have to be oversized according to the sizing table in the instructions for use. The instructions for use were not followed in some patients in our cohort. In these cases, there was no other viable alternative due to anatomic restrictions or their comorbidities. In a review by Sultan et al. [10], it was also suggested that the MFM was not recommended for an aneurysm diameter of $>$6.5 cm. In aneurysms with this large diameter, adventitial elastolysis develops, the structural integrity of the aortic wall is lost, and the MFM cannot remodel the aneurysm [10].

Femoral access was obtained using a groin incision or percutaneously. After angiography and under systematic heparinization, a stiff guidewire (Lunderquist, Cook Medical, Bloomington, IN, USA) was positioned in the ascending aorta. Then, the MFM stent was placed through a 20 F introducer sheath. When multiple stents were planned, the stent with the smallest diameter was placed first, followed by the stent with the larger diameter. A remodeling balloon (Reliant; Medtronic, Minneapolis, MN, USA) was only used in patients with an aneurysm or penetrating aortic ulcer. Within 30 days of implantation, a CT angiography (1 mm axial slices) was performed.

Follow-up visits were scheduled after 6 months, and then, yearly thereafter. All patients received postoperative clopidogrel 75 mg daily for at least 12 weeks.

Technical success was defined according to the reporting standards for endovascular aortic repair (EVAR) as the successful introduction and deployment of the device without surgical conversion, mortality, type I or III endoleak, or limb obstruction within the first 24 hours [11]. Clinical success was defined as successful deployment of the endovascular device without aneurysm-related mortality, type I or III endoleak, graft infection, aneurysm expansion ($\ge$5 mm during follow-up), aneurysm rupture or conversion to open repair [11]. MFM has an open cell design and type IA/B and III endoleaks have been described as failure mode I and endoleak type III as failure mode II [10]. Therefore, endoleak type I or III were disregarded in the definition of technical and clinical success.

Acute renal failure was defined as an increase in serum creatinine $\ge$0.3 mg/dL within 48 hours or an increase in baseline serum creatinine $\ge$1.5 times or urine volume 0.5 mg/kg/h for 6 hours [12]. We also assessed patient demographics (age, sex, comorbidities, and risk factors), aortic pathology (dissection, penetrating aortic ulcer, aneurysm, and classification), maximal diameter of the aneurysm, previous aortic procedures, number of covered aortic branches, adjuvant procedures, non-adherence to the instructions for use, 30-day outcomes and complications, aneurysm-related outcomes (maximal aneurysm diameter, occluded covered aortic branches, reinterventions because of aneurysm progression, and aneurysm-related death), and all-cause mortality.

Statistical analysis of this retrospective cohort was performed using SPSS (version 27; IBM Corporation, Armonk, NY, USA). Continuous variables were reported as mean and standard deviation and categorical data as absolute numbers and percentages. Statistical significance was defined as p 0.05. Non-normally distributed continuous data were compared using the Mann–Whitney U test; Students’ test was used for normally distributed continuous data. Categorical data were compared using the Pearson $\chi$${}^2$ test (and Fisher’s exact test when n 5). Kaplan–Meier analysis was used for all-cause and aneurysm-related mortality.

The systematic review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [13]. The systematic review was registered at PROPERO (CDR42023454147). The MEDLINE database was systematically searched from January 1st, 2010, to June 1st, 2022 (search terms included [“multilayer” OR “Cardiatis”] AND [“aneurysm” OR aorta*]) and updated on January 1st, 2023. Two authors (DÖ and RZ) confirmed the eligibility of the studies independently. Randomized controlled trials (RCTs), cohort studies, and single-arm studies (including registry studies) were accepted. We only accepted studies in which the MFM (Cardiatis, Isnes, Belgium) was used. Patients with an aneurysm, dissection, penetrating aortic ulcer, false aneurysm, or intramural hematoma of the aorta were included. No limits were applied regarding publication language or status. The references of each identified trial were used to identify any further relevant studies. When multiple studies describing the same population were published, the most complete report was used. Studies that were only available as abstracts were excluded since a quality assessment could not be performed. Studies with 5 patients were also excluded.

The following characteristics were extracted: author and year of publication, country, total number of patients, total number of stents, indication, location of the aortic pathology, number of over-stented side branches, and mean follow-up.

Outcome measures included: technical success, mesenteric ischemia at 30 days, neurological complications at 30 days, patent side branches, thrombosis of the aneurysms or false lumen of the dissection, reintervention at 1 year and maximal follow-up, rupture at 1 year and maximal follow-up, and all-cause mortality at 30 days, 1 year, and maximal follow-up.

Quality assessment of the single-arm studies was performed using the methodological index for non-randomized studies (MINORS) [14]. The MINORS quality assessment contains 12 items, of which the first 8 are specifically for non-comparative studies. MINORS has a high test–retest reliability and good internal consistency [14]. In the assessment of follow-up length, two points were assigned when the mean follow-up length was longer than 12 months and one point when the mean follow-up length was longer than 6 months. The quality assessment was performed by RZ and DÖ.

Since no comparative studies were available, single-arm meta-analyses were performed using OpenMeta (Brown University, Providence, RI, USA). To assess heterogeneity, the I${}^2$ statistic was used (I${}^2$ $>$ 75% was used as a threshold indicating significant heterogeneity). All analyses were performed on an intention-to-treat basis. When analyzing complications or reinterventions, the number of complications/reinterventions was used and not the number of patients.

A total of 37 consecutive patients were treated using MFM, the demographics of whom are depicted in Table 1. Most of the patients were male (54%). All patients were American Society of Anesthesiologists (ASA) III or IV. A prior aortic procedure was performed in 10 patients, while MFM (off-label) was used in 2 patients to treat a type 1A endoleak.

3.1.2 Postoperative Outcomes

The mean follow-up was 19.4 months (SD 18.4). In 83.8% of patients with an aneurysm, the aneurysm sack or the false lumen (in patients with a dissection) remained perfused (Table 2). The maximal diameter of the aneurysm in patients increased from 53.0 mm to 58.2 mm. All-cause mortality and aneurysm-related mortality are displayed in Fig. 1. There was a trend towards higher overall survival rates in treated patients when the instructions for use were followed (p = 0.08); aneurysm-related survival was significantly improved in patients treated within the instructions for use (p = 0.03).

Fig. 1.

All-cause and aneurysm-related mortality. (a) All-cause mortality. (b) Aneurysm-related mortality. (c) All-cause mortality: within and outside IFU. (d) Aneurysm-related mortality: within and outside IFU. IFU, instructions for use.

A reintervention was necessary in eight (21.6%) patients due to the progression of the aneurysm sack, table 3. In seven patients, an additional MFM or thoracic endovascular aortic repair ((T)EVAR) was implanted because of mode I or III failure, in one patient a debranching procedure was performed. Other reinterventions included stenting of aortic side branches (n = 1), nephrectomy (n = 1), iliac-renal bypass (n = 2), and thrombectomy (n = 1).

A total of 353 potentially eligible studies were identified (Fig. 2). After screening the titles and abstracts for inclusion, a total of 12 studies met the inclusion criteria. The characteristics of the included studies are depicted in Table 4 (Ref. [4, 8, 9, 15, 16, 17, 18, 19, 20, 21, 22, 23]). There were five registered studies describing the same patient population [15, 17, 23, 24, 25], the study with the most patients and most complete follow-up data was included in the meta-analysis [23].

Fig. 2.

Study flow chart. MFM, multilayer flow modulator.

In most studies, a thoracoabdominal aneurysm was the indication used for MFM implantation. However, only patients with an aortic dissection were included in one study [15]. Since no comparative studies could be included, the quality assessment was based on the first eight questions in the MINORS quality assessment tool (Supplementary Table 2). In none of the studies, the study size was prospectively calculated, meaning the maximal score of 16 was not reached in the quality assessment. In addition, the outcome assessments were not performed by an unbiased researcher in most studies.

Outcomes are shown in Fig. 3 and indicate a technical success of 97.8% (350/358). At 30 days, three cases of mesenteric ischemia (0.8%) and four cases of neurological complications (1.2%) were identified. Neurological complications included one case of a transient ischemic cerebrovascular event, two patients who suffered a stroke, and one case of a hemorrhagic cerebrovascular stroke. Most covered aortic side branches remained patent during the follow-up study (1005 of 1036, 97.0%). Additionally, the aneurysm sack or false lumen remained perfused in 68.5% (102 of 149) of the patients during the follow-up study. Reinterventions were necessary for 14.2% of patients (33 of 232) after 1 year and 24.4% (41 of 168) at the maximal follow-up. The incidence of aneurysm rupture was 2.1% (5 of 539) at 1 year and 4.4% (9 of 206) at the maximal follow-up. All-cause mortality was 3.4% (11 of 322) at 30 days, 15.4% at 1 year (35 of 228), and 20.9% (40 of 191) at the maximal follow-up (mean of 14.3 months).

Fig. 3.

Meta-analyses. (a) Technical success. (b) Mesenteric ischemia at 30 days. (c) Neurological complications at 30 days. (d) Patency of side branches at maximal follow-up. (e) Complete thrombosis of the aneurysm sac at maximal follow-up. (f) Reinterventions at 1 year. (g) Reinterventions at maximal follow-up. (h) Rupture at 1 year. (i) Rupture at maximal follow-up. (j) All-cause mortality at 30 days. (k) All-cause mortality at 1 year. (l) All-cause mortality at maximal follow-up. Ev, event; Trt, treated patient.