Patient recruitment and data recording

This single‑center, observational, retrospective study included patients with active malignancy who had EVT with stent implantation due to SVCS conducted in a cardiology unit between July 2014 and November 2024. The hospital database was examined for all consecutive patients discharged following the implementation of EVT for SVCS. The active malignancy was specified as 1) current requirement for anticancer therapy or 2) at least qualification for such a therapy within the previous 30 days, as well as 3) disqualification from anticancer treatment due to a lack of further therapeutic options. The exclusion criteria were: 1) the absence of a significant stenosis in the SVC on venography and / or 2) a non‑neoplastic etiology of SVCS. The flowchart for patient enrolment is presented in Figure 1. Baseline characteristics were recorded for each confirmed case, including patient demographics, comorbidities, the type of cancer, anticancer treatment that had been received, and the type of SVC occlusion in accordance with the Stanford and Doty Classification.⁶ Clinical symptoms were recorded before the procedure in accordance with the Kishi scoring system, and indications for stent implantation were defined as the Kishi score above 4.⁷ Symptom assessment, including the Kishi scoring system, was conducted 48 hours after the procedure to evaluate the clinical improvement. Improvement in clinical symptoms was defined as follows: 1) complete relief, no symptoms of SVCS; 2) moderate relief, still symptomatic with improvement in the Kishi score by more than 2 points; 3) mild relief, still symptomatic and improvement in the Kishi score of 2 points or less; and 4) no clinical change. Outcome measures specific to stent implantation, including complication rate during the procedure, were noted. After the procedure, the patients were followed‑up for 30 days for recurrent reocclusion and / or death. All the information was derived from medical records with all their limitations.

Procedural details

The stents were implanted following a routine protocol as detailed below. The procedure was performed under local anesthesia. In all cases, a combination of the right transjugular and right transfemoral access was successfully obtained under ultrasound guidance. First, venography was performed to accurately assess the stenosis and to select the optimal stent length and diameter. Vessel diameter and length, as well as significance of the stenosis were assessed on intravascular ultrasound examination (Philips Volcano Eagle Eye or Philips Volcano P18, San Diego, California, United States). In the case of tight stenosis, predilatation with a balloon catheter was performed. The following self‑expanding stents were used: Zilver 635 (Cook Medical, Bloomington, Indiana, United States), S.M.A.R.T Control (Cordis Corp., Miami, Florida, United States), and Venovo (Becton Dickinson, Franklin Lakes, New York, United States). Technical success was defined according to the Society of Interventional Radiology guidelines,⁸ as the full coverage of the stenosis (the stent extending 10 mm beyond the edges of the stenosis on both sides). A routine postdilatation procedure with a balloon catheter was performed following placement of the stent in order to guarantee proper stent apposition. In all patients, pressure measurements in the SVC were performed before and after the stent implantation. All patients were administered 3000 IU of unfractionated heparin during the procedure. After the procedure, all patients received low‑molecular‑weight heparin (LMWH) at 1 mg/kg on a continuous basis.

Statistical analysis

IBM SPSS Statistics package, version 27.0 (IBM, Sheffield, United Kingdom) was applied for statistical analysis. The results for categorical variables were expressed as numbers and percentages, while for continuous variables they were reported as mean and SD or median with interquartile range (IQR), depending on the type of distribution. The t test was used for continuous variables with a normal distribution, while the Mann–Whitney test was applied for those without a normal distribution, to determine the differences between the groups. The Fisher exact test was applied for all contingency Tables (both 2 × 2 and those with more than 2 categories), to establish group differences. In the cases involving paired data, the Wilcoxon matched‑pairs test was conducted to evaluate the significance of the observed differences (eg, blood pressure measurements before and after the procedure). A P value below 0.05 was considered significant.

Patient characteristics

A total of 48 consecutive patients with SVCS underwent EVT with stent implantation between July 2014 and November 2024 at the Department of Pulmonary Circulation, Thromboembolic Diseases and Cardiology at the European Health Center in Otwock. All patients were referred to our center for interventional treatment of SVCS. After exclusions, 40 patients were included in the analysis. A total of 56 stents were implanted in 47 procedures performed in patients with malignancy.

The most prevalent causes of SVCS in the whole study population were NSCLC (67.5%) and SCLC (20%). The majority of patients (60.6%) were at stage 3B or 4 of the disease, whereas all patients with SCLC had an extensive stage of the disease. There were 33 patients (82.5%) who had previously received anticancer treatment (previously treated patients [PTP]) and 7 patients (17.5%) who had not received any anticancer treatment prior to the procedure (not previously treated patients [NPTP]). The PTP group received the following anticancer treatment: 60.6% were treated with both chemotherapy and radiotherapy, 30.3% with chemotherapy alone, and 9.1% with radiotherapy alone. More than 80% of the patients were treated pharmacologically for SVCS, of those, 54.5% received both glucocorticosteroids and diuretics, 12.1% received only diuretics, and 33.3% received only glucocorticosteroids. Thirty‑seven participants (92.5%) received a prophylactic dose of LMWH. There were no significant discrepancies in cancer stage between the PTP and NPTP groups.

In 18 cases (45%) obstruction concerned a single vessel (SVC in 15 cases, right or left brachiocephalic vein [BCV] in 3 cases), in 20 cases (50%) obstruction concerned 1 BCV and SVC, and in 2 cases (5%) both BCVs and SVC were obstructed. All patients in the NPTP group had obstruction involving 2 vessels (SVC and 1 BCV). A summary of all baseline patient parameters is presented in Table 1.

The median (IQR) Kishi score in the studied population was 6 (6–7), and there was no difference in the Kishi score between the PTP and NPTP groups (6 [6–7] vs 7 [6–8], respectively; P = 0.13). The symptoms of SVCS along with the frequency of their occurrence are presented in Table 2.

Procedure efficacy and clinical follow‑up

The procedure was successful in 97.5% of the cases (39 of 40 patients). In 1 patient, who had previously been treated with chemotherapy, the SVC obstruction could not be crossed with the guidewire. Predilatation with a balloon catheter was performed prior to 28 stent placements (56%; Figure 2). In all the procedures with at least 1 stent implanted, the postdilatation with a balloon catheter was performed, thus there were no cases with residual stenosis exceeding 30%. In 21 patients (52.5%), residual stenosis smaller than 30% was noted, despite postdilatation with a balloon catheter. The majority of patients (70% of the total study population) had only 1 stent implanted. In all other cases, more than 1 stent was necessary owing to the complexity of the lesion (Figure 3). No significant differences were found between the size of the stents and the balloon catheters depending on whether or not the patient had previously received anticancer treatment. All the procedural details are presented in Table 3. There was a considerable decrease in the median (IQR) pressure measured in the SVC following stent implantation, as compared with the values before the procedure in the entire study population (23 [15–29] vs 8 [5.75–13] mm Hg; P <⁠0.001; Figure 4).

Within 48 hours of the procedure, clinical improvement of the symptoms was achieved in 38 patients (95%). No clinical changes were noticed in 2 patients who had previously received anticancer treatment (Table 4). During 30 days of follow‑up, 3 cases of stent (S.M.A.R.T Control) reocclusion were observed. All these patients had previously been treated with radio- and chemotherapy. Two patients (a 70‑year‑old woman, NSCLC stage 3C and a 65‑year‑old woman, NSCLC stage 3B) experienced compression of the stents due to tumor progression. In both cases, dilatation with a balloon catheter was performed initially, followed by implantation of 1 additional stent in 1 case and 2 additional stents in the other case, restoring the flow through the vessel. In the third case (a 50‑year‑old man with stage 3B thymus cancer), stent reocclusion was observed within 14 days of the procedure. In this patient, a recanalization attempt was unsuccessful, because the lesion could not be passed through with the guidewire. The patient died 2 weeks after the second procedure.

In the total study population, the primary rate of patency in the follow‑up period was 92.3% (36 of 39 patients), and the rate of secondary patency reached 97.4% (38 of 39 patients).

After the procedure, all patients received anticoagulant treatment with subcutaneous LMWH at 1 mg/kg body weight.

Complications

The study population experienced 4 types of complications (Table 4). In 1 patient, cardiac arrhythmia (sustained ventricular tachycardia) was noted in the catheter laboratory immediately after the procedure of the stent (S.M.A.R.T Control) implantation, and the patient died. The patient (a 65‑year‑old woman) had no significant comorbidities, including no history of cardiac arrhythmia or heart failure. In the second case (a 68‑year‑old man), the stent (S.M.A.R.T Control; 12 mm in diameter, 60 mm in length) slipped into the right atrium shortly after placement. Nevertheless, the stent was partially relocated using a balloon catheter (12 mm in diameter, 60 mm in length) back to the SVC. The new position of the stent was secured with a second stent (80 mm in length, 12 mm in diameter, S.M.A.R.T Control). In the third case (an 81‑year‑old man), the stent (Zilver 635) rupture was noted during balloon catheter postdilatation with small extravasation of contrast most likely into the tumor mass, accompanied by cough without desaturation. Immediate balloon catheter inflation was performed to stop the leakage, with resolution of the symptoms. In the other 2 cases (a 50‑year‑old man and a 75‑year‑old woman), there were complications related to vascular access (hematoma and pseudoaneurysm). However, they did not require further intervention. A comprehensive overview of all outcomes of the stent implantation procedures, including complications, is presented in Table 4.

Limitations

This study is subject to several limitations. Firstly, it was of retrospective, single‑center, and noncontrolled design. This may be the reason for the incomplete observations, which may result from a lack of a follow‑up protocol or a lack of detailed data on cancer diagnosis as well as observational bias. Secondly, due to the heterogeneity of the population, survival analysis was not possible, and only in‑hospital deaths were included in the analysis. Thirdly, the study is limited by a small sample size. Fourthly, the examined population was divergent both in terms of the type of cancer and the treatment that had been administered prior to the study. Hence, it is a challenge to generalize and draw broad conclusions from the results we have obtained.

Conclusions

To sum up, EVT with stent placement is an effective type of treatment in terms of symptom relief in patients with SVCS due to underlying malignancy. The procedure is characterized by an acceptable incidence of complications, whether it is the primary or subsequent treatment in patients who have previously undergone radio- and / or chemotherapy.