Iliocaval reconstruction of chronically thrombosed cylindrical inferior vena cava filters with balloon expandable covered stent-grafts.

The use of a Viabahn VBX endoprosthesis (W. L. Gore & Associates, Flagstaff, Ariz) to exclude chronically thrombosed inferior vena cava (IVC) filters refractory to exclusion with self-expanding stents was evaluated. The mean duration of TrapEase IVC (Cordis, Milpitas, Calif) implantation was 7.6 years (range, 2-11 years). Symptoms included leg pain, edema, color changes, and back pain. The mean Villalta score and venous clinical severity score were 17 (range, 13-23) and 13 (range, 11-15), respectively. Indirect ultrasound evidence of stent patency was demonstrated at a mean of 8 months after intervention. The mean Villalta score and venous clinical severity score had decreased by 13 and 10, respectively, at a mean of 9.5 months after intervention. Iliocaval reconstruction with Viabahn VBX balloon expandable stent-graft exclusion of chronically thrombosed TrapEase IVC filters is safe, with favorable short-term results.

A total of 2.8% to 3.9% of patients with an inferior vena cava (IVC) filter will develop filter-associated IVC thrombosis, which can cause pain, edema, phlegmasia, and ulceration., IVC filter retrieval success rates can be as high as 99.2%. However, cylindrical filters, such as the TrapEase (Cordis, Milpitas, Calif), have had decreased success rates. When filter retrieval is not possible, stent exclusion can be performed using self-expanding stents (SES) and balloon-expandable stents (BES). However, limited data are available on the use of BES grafts for this indication. In the present study, we evaluated the clinical and technical outcomes of iliocaval reconstruction using Viabahn VBX BES grafts (VBES; W. L. Gore & Associates, Flagstaff, Ariz) for IVC thrombosis secondary to chronically thrombosed and embedded cylindrical filters. All three patients provided written informed consent for the report of their case details and imaging studies. The institutional review board approved the present study, and the data were managed in accordance with the Health Insurance Portability and Accountability Act.

Case report

Patient 1

A 63-year-old man had presented with chronic bilateral lower extremity and back pain and limb discoloration. The Villalta score and venous clinical severity score (VCSS) were both 15. Computed tomography (CT) and venography demonstrated a thrombosed and calcified TrapEase IVC filter (Fig, A) with complete occlusion to the bilateral common femoral veins (CFVs; Fig, B). The filter had been placed 11 years previously. Recanalization was required for symptomatic iliocaval occlusion and was performed from the bilateral CFVs and right internal jugular vein (IJV). Pharmacomechanical thrombectomy was followed by filter exclusion and iliocaval reconstruction with parallel, bilateral, overlapping SES dilated using venoplasty to 12 to 16 mm. An acute clot had formed within the SES (Fig, C) secondary to severe narrowing by the filter (Fig, D). The SES were reinforced with parallel 11-mm × 79-mm VBES dilated to 16 mm (Fig, E). In-line flow was established from the lower extremities to the right atrium (Fig, F). The patient was hospitalized for 1 day. Ultrasound (US) at last imaging follow-up at 6 months after the intervention demonstrated patent bilateral CFVs, femoral veins (FVs), and popliteal veins (PVs). Both CFVs demonstrated increased waveforms during forced expiration, suggesting indirect iliac and IVC patency. The Villalta score and VCSS had decreased to 2 and 5, respectively, at the last clinical follow-up at 7 months. He continued taking rivaroxaban.

Fig

A, Contrast-enhanced abdominal and pelvic computed tomography (CT) scan demonstrating a completely thrombosed TrapEase (Cordis) inferior vena cava (IVC) filter with chronically calcified intraluminal thrombus (white arrow). B, Bilateral common femoral veins (CFVs) venography demonstrating complete occlusion of the iliocaval system with robust collateralization. C, Bilateral CFVs venography demonstrating iliocaval reconstruction with placement of overlapping S.M.A.R.T. stents (Cordis) and Wallstents (Boston Scientific, Waltham, Mass) at the level of the IVC filter. Sluggish flow and the development of acute thrombus was present within the stents (gray arrowheads). D, Non–contrast-enhanced CT scan of the abdomen during recanalization demonstrating parallel overlapping S.M.A.R.T. stents and Wallstents at the level of the calcified and thrombosed IVC filter with incomplete expansion of the right-sided stents (black arrow). E, Spot fluoroscopy demonstrating simultaneous deployment of parallel Viabahn VBX stents across the area of persistent narrowing. F, Venography from the bilateral CFVs demonstrating brisk in-line flow throughout the stent construct.

Patient 2

A 66-year-old man had presented with a 2-week history of bilateral lower extremity edema and pain. The Villalta score and VCSS were 13 and 11, respectively. CT demonstrated a calcified and thrombosed TrapEase IVC filter with partial occlusion of the left common and external iliac veins. The filter had been placed 10 years previously. Recanalization was required for symptomatic iliocaval occlusion and was performed from the right CFV, left PV, and right IJV. Pharmacomechanical thrombectomy was followed by attempted IVC filter retrieval using the excimer laser (Spectranetics, Colorado Springs, Colo) and endobronchial forceps (No. 4162; LYMOL Medical, Woburn, Mass) but was unsuccessful due to the large volume of calcified thrombus, which prevented laser sheath advancement. IVC filter exclusion and iliocaval reconstruction were performed with parallel bilateral overlapping SES dilated using venoplasty to 12 to 16 mm. The SES at the level of the IVC filter were narrowed, with repeated high-pressure balloon rupture. They were reinforced with parallel 11-mm × 79-mm VBES dilated to 16 mm. In-line flow was established from the lower extremities to the right atrium. The patient was hospitalized for 2 days. US at the last imaging follow-up at 9 months after the intervention demonstrated patent bilateral CFVs, FVs, and PVs. Both CFVs demonstrated increased waveforms during forced expiration, suggesting indirect iliac and IVC patency. The Villalta score and VCSS had decreased to 3 and 4, respectively, at his last clinical follow-up at 11 months after the intervention. He continued taking rivaroxaban.

Patient 3

A 72-year-old man had presented with a 7-day history of bilateral lower extremity swelling and pain. The Villalta score and VCSS were 23 and 14, respectively. CT demonstrated a thrombosed TrapEase IVC filter with completely thrombosed, expanded, and edematous bilateral iliofemoral and PVs. The IVC filter had been placed 2 years previously. Recanalization was required for symptomatic iliocaval occlusion and was performed from the bilateral PVs and right IJV. Pharmacomechanical thrombectomy was followed by IVC filter exclusion and iliocaval reconstruction with parallel bilateral overlapping SES dilated using venoplasty to 12 to 16 mm. Because of the recurrent thrombus formation within the stents that developed overnight, thrombolysis was initiated. Subsequently, the thrombus burden improved; however, the SES were severely narrowed by the IVC filter. Thus, they were reinforced with parallel 11-mm × 79-mm VBES dilated to 16 mm. In-line flow was established from the lower extremities to the right atrium. This patient developed transient acute kidney injury from contrast administration. He was hospitalized for 7 days; his creatinine had increased from 0.89 to 1.53 mg/dL but had normalized to 0.75 mg/dL at discharge. US at the last imaging follow-up at 11 months after the intervention demonstrated patent bilateral CFVs, FVs, and PVs. Both CFVs demonstrated increased waveforms during forced expiration, suggesting indirect iliac and IVC patency. The Villalta score and VCSS had decreased to 6 and 4, respectively, at his last follow-up at 11 months after the intervention. He continued taking enoxaparin.

For all three patients, the intraprocedural activated clotting time was maintained at >250 seconds with heparin, and stent selection was determined by operator preference and availability (Table). Finally, additional stenting was not needed after VBES placement. None of the patients experienced pulmonary embolism or renal vein thrombosis or required reintervention. The contrast agent used was iohexol (Omnipaque 300; mean, 265 mL; range, 230-325 mL), with a mean fluoroscopy time of 187 minutes (range, 90-249 minutes). The mean serum creatinine was 0.78 mg/dL (range, 0.70-0.89 mg/dL). Antiplatelet agents were not used. All patients wore above the knee compression stockings with 30 to 40 mm Hg of compression, and none had had venous ulcers.

Table

Stent type, number, and size used per patient

Stent type	Stent size (No. used)
	Patient 1	Patient 2	Patient 3
S.M.A.R.T stent	12 mm × 80 mm (2); 14 mm × 80 mm (2)	12 mm × 80 mm (6); 14 mm × 80 mm (2)	NA
Wallstent (Boston Scientific, Waltham, Mass)	12 mm × 60 mm (2); 14 mm × 90 mm (2)	14 mm × 90 mm (2); 16 mm × 90 mm (1)	NA
Protege stent (Medtronic, Dublin, Ireland)	NA	NA	10 mm × 80 mm (2)
Vici Venous stent (Boston Scientific)	NA	NA	14 mm × 120 mm (2); 16 mm × 90 mm (2)
Viabahn VBX stent (W. L. Gore)	11 mm × 79 mm (2)a	11 mm × 79 mm (2)a	11 mm × 79 mm (2)a

NA, Not applicable.

Gore Viabahn VBX stents were simultaneously ballooned to 16 mm.

Discussion

Cylindrical filters have been shown to have decreased retrieval success., When retrieval is not possible, iliocaval reconstruction with stent exclusion of chronically embedded IVC filters has had technical and clinical success of 83% to 100%,8, 9, 10 and 60% to 96%,, respectively. Most studies have used SES,,8, 9, 10, 11, 12 although some have demonstrated success with BES.,, IVC filters will typically be excluded with SES; however, when compression is refractory, the options become limited to placing additional overlapping SES, a bare metal BES, or a BES graft such as a VBES. Compared with bare metal BES, the VBES has longer diameters to span the entire IVC filter and is covered to prevent balloon rupture. In the present study, parallel reinforcement of chronic and calcified intrafilter thrombus with VBES restored in-line flow in all three patients. US demonstrated patent PVs to CFVs and indirect iliocaval patency during follow-up (mean, 8 months; range, 6-11 months). The average Villalta score and VCSS had decreased by 13 and 10, respectively, at the last follow-up (mean, 9.5 months; range, 7-11 months). No technical complications developed, and no reinterventions were required. One patient had developed transient acute kidney injury.

VBES have been successfully used in occluded pediatric mesoportal shunts, non–filter-associated iliocaval occlusion, and axillary vein injury. Although stent-grafts have had high success rates to treat hemodialysis outflow stenosis, the iliocaval system has substantially lower flow and the available data are limited. A literature review demonstrated the successful use of iliocaval stent-grafts for aortocaval fistula, thrombosis, IVC injury, and IVC agenesis, with patent stents at the last follow-up (range, 1 week to 7 years).18, 19, 20, 21, 22, 23, 24, 25, 26, 27 The largest retrospective review demonstrated 90% primary patency in 10 patients who had received stent-grafts for chronic iliocaval disease. Overall, the reported data have demonstrated favorable short-term results, similar to the findings from the present study. VBES can theoretically cause filter fracture, embolization, and displacement into nearby structures from the pressure exerted by stent placement. Furthermore, the long-term patency of iliocaval stent-grafts is unknown.

Conclusions

Iliocaval reconstruction with VBES exclusion of chronically thrombosed TrapEase IVC filters is safe, with favorable short-term results.