Study Design

This STROBE compliant prospective, multicentre, observational study has been approved by the Medical Ethical Committee of the promoter Centre (Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico - Milan, IT) with approval number1954 (RadIntv04/2021) and performed according to the Declaration of Helsinki. All participants signed an informed consent for the study.

Data from endovascular procedures occurred in ten interventional radiology centres (ASST dei Sette Laghi - Varese, IT; IRCSS Azienda Ospedaliero-Universitaria di Bologna - Bologna, IT; AOU “Maggiore della Carità” - Vercelli, IT; POC SS Annunziata - Taranto, IT; Cardarelli Hospital - Naples, IT; Policlinico Casilino - Rome, IT; IRCCS “A Gemelli” - Rome, IT; A.O.U. Citta della Salute e della Scienza - Torino, IT; Azienda Ospedaliera per l’emergenza Cannizzaro - Catania, IT; Institut Mutualiste Montsouris - Paris, FR) were prospectively collected from March 2021 to May 2022.

Inclusion Criteria

age > 18 years old;

informed consent signed;

No ultrasound guidance was used for common femoral artery (CFA) access;

used sheaths ≥ 5F;

Use of the VCDs indicated (FemoSeal, AngioSeal, Mynx)

use of VCDs following manufacturer’s instructions for use (IFU);

operators with at least 3 years of experience in the specific chosen VCD’s positioning

bed rest for at least 2–3 h after the use of VCD;

in case of unsuccessful haemostasis with VCDs and manual compression is required, bed rest is suggested for at least 6 h

Exclusion Criteria

CFA occlusion or stenosis ≥ 80%;

history of severe allergy to contrast media or any VCDs’ component;

absence of valid consent to participate to the study.

From a larger common femoral artery closure device positioning dataset, a subgroup of 5–7 Fr vascular introducer calibre was selected.

Outcome Measures

Primary outcomes were considered effectiveness and safety.

Effectiveness is meant as successful haemostasis, i.e. the possibility to obtain complete haemostasis with the use of VCDs, without any complications nor additional manual compression for AngioSeal and FemoSeal and 90 s of compression (as IFU indicate) for Mynx control.

As unsuccessful haemostasis was considered any case requiring immediate manual compression (or after 90 s of the expected compression for Mynx control) or the onset of one or more of the following complications after primary effective of VCD.

Safety is meant as the rate of complications. Complications were classified into minor and major according to CIRSE guidelines [11]. Haematoma was defined as the occurrence of perivascular blood suffusion larger than 4 cm. Pseudoaneurysm (PSA) at the access site after the VCD positioning is considered a complication regardless of the diameter. “Arterio-venous fistula (AVF)” was defined as the appearance of arterio-venous communication at the access site after the VCD positioning. Ipsilateral acute limb ischaemia is defined as the occlusion of the CFA following the use of VCDs.

Secondary outcomes were considered the relationship between the mechanism of action of the VCDs (dependent variables) and some patient’s characteristics (covariates variables) with the haemostasis result.

Dependent variables and covariates variables are listed in the following paragraph.

Variables

The outcome variables considered were: successful haemostasis; haematoma (alone); pseudoaneurysm (with/without haematoma); artero-venous fistula (AVF; with/without haematoma); and failure management.

The dependent variable considered was the VCDs mechanism of action: balloon occluders (passive approximators: Cordis MynxGrip) vs. non-balloon collagen plug occluders (passive approximators: Terumo AngioSeal; Terumo FemoSeal).

Covariates variables considered were: Age; Sex (male/female); French (sheath used for vascular access); Skin/Vessel distance (mm) (evaluated at CT scan if available; otherwise by US); Arterial diameter (mm) (evaluated at CT scan if available; otherwise by US); Access direction (retrograde or anterograde); Access side (right/left); Calcifications (considered present if gross calcifications were visible on unsubtracted DSA images and/or pre-procedural CT); BMI (Kg/m2); INR; PTL (109/l); PT %; PTT (s); Previous surgery; Diabetes mellitus; Hypertension; Smoking; Dialysis; Hyperlipidaemia; COVID-19; Cirrhosis; Haematological disorders; Anti-coagulation therapy; and Intra-procedural heparin.

Follow-up

During the hours after the procedure, clinical observation was performed in all patients included in the study.

Moreover, in all patients an ultrasound examination was performed before discharge. For discharged patients, a new hospital access within 30 days, due to complications related to the VCD, is considered a failure management.

Statistical Analysis

Data were anonymized and collected on an electronic dataset (Excel, Microsoft, Redmond, Washington, USA). Twenty-four different variables were investigated as potential predictors of four different outcomes.

Power Analysis

According to the literature, complication rates of VCDs can be approximated to 10% in clinical settings [1, 3,4,5,6, 12]. Moreover, a significant bias is the absence of randomization. To overcome this issue, a minimum enrolment ratio criterion of 1:6 was introduced. According to this data, the target sample size was fixed to 1264 (158:1106) patients, which provides 80% power at the 5% (2-sided) level of significance [13, 14].

Data Analysis

Descriptive statistics were produced for cases’ demographic, clinical and laboratory characteristics. Number and percentages are presented for categorical variables, and mean and standard deviation (SD) are presented for normally distributed variables. 95% confidence interval (CI) was given when appropriate.

Shapiro–Wilk test was used to assess normality.

Crosstabs and the Fisher’s exact test were used to assess relations among VCDs mechanism of action and the outcome variables. Univariate and multivariate analyses of variance were also performed to weight the effects of all variables on closure device choice upon the four outcomes.

SPSS version 25.0 (IBM, Armonk, New York, USA) was used for all statistical analyses. In all cases, two-tailed tests were used. p values were considered significant when < 0.05.