A retrospective review of the archives of our interventional radiology department from October 2010 to January 2021 yielded the cases of 21 patients with malignant IVC obstructions (11 women and 10 men, median age 61.0; range 18–92 years), who underwent IVC stent implantation. In each case, we used the closed-cell designed, self-expanding nitinol Sinus-XL stent. Patients with benign IVC compression were excluded from the study. The requirement for consent from patients to be included in this retrospective study was waived by our institutional review board (No. 20211125 01). All patients were examined and treated as part of routine care and informed consent for the procedure was obtained before the procedure. Patients’ characteristics and underlying conditions are presented in Table 1.

IVC obstruction syndrome was based on primary tumor compression in five cases, whereas in 16 cases, significant IVC compression was caused by metastatic disease, including metastatic lymph nodes in seven, hepatic metastases in six and a combination of lymph node and hepatic metastases in three patients.

In each case, the indication for endovascular IVC stent implantation was approved by an interdisciplinary team consisting of interventional radiologists, oncologists as well as visceral and vascular surgeons. In 20 patients, stent placement was triggered by worsening of IVC obstruction symptoms. In one case, stent placement was performed as a preventive measure.

For verification of IVC obstruction and procedure planning, 19 patients (90.5%) underwent abdominal contrast-enhanced computed tomography (CT), one patient received abdominal contrast-enhanced magnetic resonance imaging (MRI), and one patient was examined by ultrasound. Diagnostic imaging was performed at a median of 8 days (range 0–64 days) prior to the procedure.

All procedures were performed by the same operator in our local angiography suite (Siemens, Axiom Artis Zee, Forchheim, Germany) via a transfemoral antegrade venous approach. In 20 cases, we used the right common femoral vein as access site. In one case, the left common femoral vein was accessed. IVC stenosis was passed with a 0.035-inch guide wire (Radifocus, Terumo, Tokyo) and a 5F selective angiography catheter (Berenstein configuration, Angiodynamics, Queensbury, NY). IVC stenosis was confirmed by digital subtraction cavography via a 10F sheath (Check-Flo, Cook, Bjaeverskov, Denmark). Prior to cavography, we exchanged the crossing guide wire and the selective catheter for a 0.035-inch support guide wire (Back-Up Meier, Boston Scientific, Natick, Mass) and a calibrated 5F pigtail catheter. We chose oversized self-expanding stents exceeding the normal IVC diameter by approximately 25–30%. The Sinus-XL stent is a laser-cutted nitinol stent compatible with a 10F sheath and a 0.035-inch guide wire available in eleven diameters (16–34 mm) and six lengths (30–100 mm).

After cannulation of the obstructed IVC segment, the stent system was introduced and deployed under fluoroscopic control. All patients had a bolus of 5000 units of unfractionated heparin intraprocedurally. Balloon predilatation was performed in four cases using different types of undersized balloons (Atlas®, Bard, Tempe, AZ; Zelos®, Optimed, Ettlingen, Germany; XXL® Boston Scientific, Maple Grove, MN). In nine cases more than one stent was required and implanted in an overlapping technique. In six cases (27.3%), cavography demonstrated relevant residual stenosis of the treated vessel segment, necessitating additional balloon dilatation. For post-dilatation we used undersized balloons (Atlas®, Bard, Tempe, AZ; Zelos®, Optimed, Ettlingen, Germany; Armada®, Abbott Vascular, Santa Clara, CA). A final cavogram was performed to proof sufficient stent expansion and recanalization of the stenotic IVC segment. Sheaths were removed and access site was closed by manual compression (Figs. 1, 2).

25-year-old female patient with hepatic metastasis of a retroperitoneal leiomyosarcoma. a Preinterventionally performed coronal contrast-enhanced computed tomography demonstrates high-grade stenosis of the IVC in the intrahepatic segment due to the intrahepatic tumor masses. b Corresponding cavography depicts the extent of the central venous obstruction and collateralization. c Fluoroscopic approval of the successful deployment of the Sinus-XL stent within in the malignant stenosis. d Cavography of the IVC after stent deployment demonstrates restored patency with reduced blood flow via collateral vessels

92-year-old male patient with hepatic metastasis of gastric cancer. a Inferior cavography depicts high-grade compression of the intrahepatic IVC with development of collateral vessels. b Balloon post-dilatation after deployment of the nitinol stent within the malignant stenosis. c Final cavography demonstrates significant improvement of the venous backflow and absence of venous collateral blood flow

In nine patients, venous blood pressure measurements were performed during the intervention. At the discretion of the interventional radiologist, pressure gradients were measured across the stenotic segment before and after stent placement. Postinterventionally, the patients received a therapeutic PTT-adapted dose of low-molecular-weight heparin for 7 days, followed by conversion to long-term anticoagulation with enoxaparin or aspirin.

Two authors reviewed the patients’ medical and radiologic records to gather information on the technical and clinical success of the interventions. As far as possible, missing data were added by telephone interviews with patients and referring physicians. Technical success, clinical success and safety were defined as primary endpoints.

Technical success was designated as successful deployment of the stent within the obstructive lesion accompanied by significant improvement of blood flow in the stenotic IVC segment and by reduced blood flow via collateral vessels.

Clinical success was determined as the improvement of the patient’s symptoms. The introduction of a scoring system allowed the quantification of clinical success assessing the main manifestations, i.e. edema, anasarca and ascites before and after the procedure. To this end, we modified previously published grading systems [6, 10, 11].

Lower body edema was graded as follows: 0 for absence of edema, 1 point for mild edema, 2 points for moderate edematous swelling, 3 points for severe swelling impeding the palpation of foot pulses without pain and immobility and 4 points for massive edematous swelling causing pain and immobility. The edema distribution was scored, ranging from 0.5 points for swelling of the feet to 2 points for the genitals being involved. The presence of complicating factors, for example ulcerations, were graded with an additional point resulting in a maximum score of 7 points. Progress of the edema severity during the follow-up period was rated with another 0.5 points, whereas improvement was assessed with minus 0.5 points. The rating of lower body edema was based on the patients’ clinical reports at three time points during the follow-up period.

Ascites was scored by means of a four-point-scale: 0 for absence of ascites up to 3 indicating the necessity of paracentesis. Cross-sectional imaging and the patients’ data including paracentesis reports allowed the assessment of ascites. After the procedure, reassessment was performed at two time points during the follow-up period. During the follow-up period, a slight progression or reduction was rated by addition or subtraction of 0.5 points, a severe progression or reduction of ascites by addition or subtraction of 1 point.

Scoring of anasarca was based on cross-sectional studies exclusively. Rating was as follows: 0 point for the absence of anasarca, 1 point for slight anasarca and 2 points for severe anasarca. Anasarca scoring during the follow-up was executed analogously to the scoring of ascites.

Considering the procedural safety, complications were classified according to the reporting standards of the Society of Interventional Radiology [12].

Secondary endpoints were defined as procedural parameters, including the number and size of the implanted stents, procedure duration, fluoroscopy time, and radiation exposure in terms of the dose area product. In nine patients, pressure gradient measurements were performed before and after the stenting and the results were also added to the secondary endpoints. Furthermore, lengths of the VCI obstruction were measured based on the intraprocedural cavography as well as cross-sectional imaging.

As another secondary endpoint, we analyzed luminal expansion and stent patency. A residual stenosis of less than 50% on cross-sectional imaging is considered as luminal expansion. Patency was defined as preserved stent perfusion documented by contrast-enhanced imaging or duplex ultrasound. Luminal expansion measurements were available in 15 cases, while contrast enhanced images enabling the assessment of patency were only available in 14 cases. Assisted primary functional patency was defined as patency subsequent to another interventional treatment.

Descriptive data were presented as means ± standard deviation (SD) for normally distributed variables or medians with ranges for non-normalized variables, if appropriate; categorical data were expressed as counts and percentages with n (%). With regard to assessment of normality, the Anderson–Darling test was used rejecting the hypothesis of normality when the p value is less or equal to 0.05. The Wilcoxon test and the Mann–Whitney U test were used for comparison of the pre- and postinterventional data or described subgroups. Kaplan–Meier analysis including the log rank test was used to analyse patients´ survival and stent patency rates.

Correlation analysis of ordinal and metrical data were performed with the test according to Spearman for non-normalized variables. For all evaluations, a p value less than 0.05 was considered to indicate significant differences. Statistical analysis and the evaluation of the data were performed with a specialized computer algorithm (Microsoft Excel V1908 and RStudio 1.2.5033).

Clinical examination was performed in those patients who returned for routine follow-up control in our outpatient clinic. All patients were advised to immediately contact the outpatient clinic at onset of new or worsening symptoms. Imaging follow-up was conducted in the context of re-evaluation of the underlying malignant disease. Median clinical follow-up was 65 days (range 1–790 days) and 10 days (range 0–185) prior to the patients’ decease. The average time between procedure and the last cross-sectional imaging follow-up amounted to 66 days (range 1–775 days) or 35 days (range 1–179 days) between the last imaging follow-up and the patient’s decease.