Optimal diagnostics of residual stenosis after lysis

Pre-intervention assessment

The initial assessment of patients with deep vein thrombosis (DVT) has traditionally been based on accurate history, and examination and remains the cornerstone of diagnosis. Symptoms lasting for more than three weeks are considered to be associated with older, collagen-rich thrombus which is less likely to respond to fibrinolytic therapy. Patient history is, however, often subjective and aging thrombus using history alone is not always accurate, explaining why not all thrombi lyse, even if treated within a ‘two-week’ window. Imaging modalities are therefore being developed to overcome this problem, though understanding the functional effect of a stenosis remains a challenge in the venous system.

Transcutaneous duplex ultrasound (DUS) is frequently used as a first-line investigation to assess deep veins. DUS has a low associated cost, it is portable, repeatable, readily accessible and has high sensitivity and specificity for diagnosing both thrombus and, in experienced hands, the presence of a stenosis. It is also routinely used for the surveillance of patients following treatment. Haemodynamically significant lesions are considered to be those that are characterised by: post stenotic turbulence; abnormal Doppler signal at the area of stenosis; continuous flow with the Valsalva maneuver and sluggish with no spontaneous flow and poor augmentation with absent respiratory variation. ¹¹ DUS can be limited, however, by large patient size, overlying bowel gas and sonographer experience. These factors can lead to poor assessment of the iliac veins with visualisation of the external iliac vein and common iliac vein estimated to be 79% and 47%, respectively. ¹ DUS is often used as a first-line investigation for DVT, which if iliofemoral is supplemented by computed tomographic venography (CTV) or a magnetic resonance venogram (MRV). These modalities can reveal complex anatomical pathology such as stenosis, occlusion or atresia of a vein segment, associated soft tissue abnormalities, thrombus within the femoral or profunda veins and/or distant malignancy.

Both CTV and MRV have high sensitivity and specificity for the detection of stenotic lesions in the vein and acute thrombosis.^12–17 Although CTV is fast and readily available in most centers, the benefit of MRV is that it does not involve the use of ionizing radiation, and this may be of significance in a young cohort of patients who are frequently treated. MRV can also be performed without a contrast agent, which will benefit those with contrast allergy or renal impairment, while it is also better able to identify intraluminal anatomical changes associated with previous thrombosis such as partial filling defects, trabeculations or webs, vessel narrowing, thickened vessel walls and pronounced collateralisation.^16,17

More recently, MRV has been used to help identify thrombus susceptible to CDT. MRV fast T₁ mapping can be used as a surrogate measure of the organisation of experimental venous thrombi and can identify thrombi amenable to lysis. ¹⁸ The clinical impact of this technique requires further investigation in man but by identification of highly collagenous lesions in the venous system, it may help identify patients in whom a stent may be beneficial. Access to MRV is still, however, limited in many centers, and it cannot be used in patients with metallic implants, claustrophobia or extreme size.

At our institution, patients are almost universally referred with a DUS. We perform MRV routinely unless this is precluded by time to gain some cross-sectional imaging. If patients have already had a CTV and the thrombus is acute clinically then an MRV is not performed. We have moved to a non-contrast MRV to reduce both radiation exposure in a young patient cohort and the risk of contrast allergy. MRV is used to evaluate thrombus age for suitability for lysis and anatomy, for any pathology that may have contributed to the clot and to plan the procedure i.e. if the clot extends to the IVC we may require bilateral access.

Intra-procedural assessment

In our practice when a haemodynamically significant stenosis is identified on DUS and cross-sectional imaging, we proceed to a combination of venography and intravascular ultrasound (IVUS). After successful CDT, contrast venography remains the gold-standard for the assessment of venous thrombosis and residual stenosis. ¹⁹ Digital subtraction angiography is used with multiplanar (ideally orthagonal) views of the vessels. The presence of a significant stenosis is highlighted by slowed drainage of contrast, the presence of collateral veins and reduced opacification. The presence of large collateral veins is suggestive of a chronic lesion as collaterals take time to develop and these often collapse following accurate placement of a stent. Certain lesions are, however, not identified on venography as many iliac vein lesions are visible only in one plane and membranous lesions may not be visible at all, while diffuse long stenosis common in post-thrombotic disease are also easily missed. Indeed, when assessing chronic venous occlusive disease, it has been shown that venography can miss up to one-third of intraluminal lesions. ²⁰ IVUS has therefore been proposed an adjunctive tool to assess the patency of the venous system during treatment.

IVUS allows a three-dimensional assessment of the vessel lumen, wall characteristics as well as side branches, collateral vessels and external structures such as crossing arteries. IVUS is not affected by patient size or oedema and can provide detailed measurements using direct planimetry for the accurate sizing of venous stents. The tomographic perspective of ultrasound also enables a more accurate assessment of vessels that are difficult to image by angiography, including diffusely diseased segments, hyperechogenic eccentric wall thickening, the presence of multiple vascular channels separated by hyperechogenic structures, and the presence of sequelae of DVT such as synechiae and organised thrombus. IVUS can therefore be used to accurately measure both the diameter, cross-sectional area and the length of the stenosis allowing for accurate sizing and placement of a venous stent.

It is widely accepted in the literature that the determination of a hemodynamically significant venous stenosis is difficult. In most papers that describe venous stenting, venous hemodynamic assessment is lacking. The significance of the stenosis is commonly determined with the presence of stenosis and the corresponding clinical signs and symptoms. There are many tests available (e.g. decreased plethysmographic outflow fraction, increased hand/foot pressure differential, increased hyperemia pressure differential, pull-through gradient of >2 to 3 mm Hg and increased pressure after intraarterial papaverine hydrochloride injection); however, none have been proven to determine outcome. Despite the lack of positive hemodynamic results with available methods, balloon dilation and stenting of stenotic iliac veins guided with morphologic area stenosis of more than 50% on IVUS results appear to have apparent clinical benefits for the patients. ²⁰

The choice of 50% as an indicator of significant stenosis is an arbitrary measure and it needs to be considered in the context of the patient together with features from multiple imaging sources. It has also been suggested that a reduction in vein area is a more significant measure than a pure diameter reduction. This can be easily calculated on IVUS using automated measures. Any area measuring under 100 mm² in the iliac outflow tract is considered to indicate a significant stenosis requiring treatment, though evidence for this arbitrary number is lacking.

Following stent placement, IVUS can also reveal inadequate stent expansion, poor vessel wall apposition or incomplete coverage of the stenotic lesion. This is particularly relevant in the setting of May–Thurner syndrome where the anteroposterior compression by the right common iliac artery onto the left common iliac vein makes the stent appear fully expanded on a two-dimensional view.

The routine use of IVUS significantly increases the cost of the procedure (∼£750 at our institution). In our experience, there is a learning curve of 30 to 40 procedures based on the heterogeneity of this patient cohort, but there are no data for this. We feel that IVUS is worth the time and increased cost, and use it together with venogram routinely on the completion of lysis to accurately evaluate and the vessel and stent residual disease.

Post-procedural imaging

DUS is an ideal modality for surveillance imaging due to its low cost, non-invasive nature and availability. Post stenting DUS does not suffer from significant artifact. B mode can be used to identify stent stenosis or thrombus lining the stent and colour flow can identify velocities, post stenotic turbulence and abnormal Doppler signals at the area of stenosis.

We use a follow-up schedule of DUS at day one, two weeks, 6 weeks, 3 months, 6 months and then annually.

If there are concerns with the stent that are inconclusive on DUS, patients will usually have additional imaging with CTV. Nitinol venous stents do give off artefact on CT. However, this is minor and given the large diameter of the venous stents, this does not hinder the identification of significant lesions. Re-intervention is offered in patients who have an asymptomatic stenosis greater than 60%, demonstrate progression of in stent thrombus formation or develop recurrence of symptoms at any stage. The majority of these interventions occur within the first six weeks, with animal studies suggesting this is how long it takes to line the stent with an endothelial layer.

Conclusion

The treatment of acute DVT has developed significantly over the past few years, and CDT is now widely accepted. Controversy, however, exists over the decision to treat residual or underlying lesions with a venous stent. There is no current method of directly measuring venous pressure in a robust and reproducible fashion following intervention and we rely on imaging both before and during a procedure to make clinical decisions. Magnetic resonance (MR) techniques have evolved considerably, which together with IVUS can provide a detailed assessment of the deep venous system. Nevertheless, there is still no single perfect imaging modality that can identify a functional stenosis in the venous system. Until this has been developed, the decision to proceed to stent placement following lysis is multifactorial, reliant on the use of a number of imaging modalities in combination, but most importantly good clinical acumen and experience, which may only be achieved in high-volume centres performing these treatments regularly.