Acute iliofemoral DVT – What evidence is required to justify catheter-directed thrombolysis?

Deep vein thrombosis (DVT) is complicated by the post-thrombotic syndrome (PTS) in up to 50% of cases, often resulting in disabling symptoms and reduced quality of life, ¹ particularly in the context of iliofemoral thrombosis. Substantial research has been undertaken to determine the best management options to prevent PTS, with percutaneous catheter-directed thrombolysis (CDT) being increasingly utilised in the past two decades.

There have been three multicentre randomised-controlled trials (RCTs) in the last ten years evaluating the use of CDT in acute DVTs. The first – post-thrombotic syndrome after Catheter-directed thrombolysis for deep Vein Thrombosis (CaVenT) trial – published in 2012, reported an absolute PTS risk reduction of 28% over a 5-year period when patients were treated with percutaneous CDT compared to standard treatment (compression stockings and anticoagulation), but with no additional benefit in quality of life outcomes. ²

The Acute venous Thrombosis: Thrombus Removal with Adjunctive Catheter-directed Thrombolysis (ATTRACT) trial, published in 2017, cast doubt over the benefit of CDT. ³ It randomised 692 patients to pharmacomechanical thrombolysis or standard treatment (anticoagulation). There was no significant difference in the rate of PTS or quality of life outcomes between both arms after 24 months.

Most recently, the ultrasound-accelerated Catheter-directed thrombolysis Versus Anticoagulation for the prevention of post-thrombotic syndrome (CAVA) trial, published in 2020, found that in patients with acute iliofemoral DVT there was no significant difference between the rate of PTS or quality of life at 12 months. ⁴

Therefore, at a glance, it would be easy to dismiss CDT as a potentially beneficial treatment option; however, the methods and outcomes of these trials should be considered further. In the CaVenT trial, fewer than 50% of patients in both trial arms suffered from iliofemoral DVT. Moreover, adjunctive endovascular procedures, which are considered necessary to achieve lasting venous patency, including deep venous stenting, were only performed in a minority of patients.

Similarly, although the ATTRACT trial represents the largest RCT, nearly half of the patients had a femoropopliteal DVT. Therefore, the study was not powered to detect significant differences in patients with iliofemoral DVT. However, in a sub-analysis of this patient population, there was a notable reduction in symptoms, PTS severity and improved quality of life scores in the thrombolysis arm. ⁵ Furthermore, almost 94% of the patients who met the inclusion criteria for the ATTRACT trial were excluded, with notable reasons being age (younger than 16 or older than 75 years), significant comorbidities (such as active cancer) and pregnancy. Although this reduces the effect of confounding, the cohort of patients included may not be generalisable to the patients commonly seen in clinical practice.

Interestingly, further analysis of the CAVA trial in the patients who achieved successful recanalisation post ultrasound-accelerated CDT (defined as regained patency ≥90%) demonstrated that there was a significant reduction in symptom severity and improved quality of life over both the standard treatment arm and those with unsuccessful recanalization. ⁶ However, only 53.2% of the patients in the intervention arm achieved successful recanalisation. This is reflected by the relatively low rates of adjunctive procedures – only 55% of patients underwent adjunctive endovascular procedures that include deep venous stenting.

Furthermore, anticoagulation protocol was not consistent across studies – the CaVenT trial used low-molecular-weight heparin and warfarin, while the ATTRACT and CAVA trials introduced direct oral anticoagulants (DOACs) as they became more available. A recent study of 309 patients with acute proximal DVT found DOAC-treated patients had a 54% lower risk of PTS than patients treated with vitamin-K antagonists, ⁷ suggesting the use of DOACs may independently reduce PTS rates and may represent a confounding factor in these trials.

This raises an important point regarding the multicentre approach of these trials. Multiple centres were used partly in an effort to overcome strict inclusion criteria, low recruitment numbers and prolonged trial duration. An inherent consequence of this is variation in treatment protocol delivered within and between the trials. This leads to difficulty in interpretation of results and variable technical success, as mirrored by the relatively low rates of recanalisation across the studies, when reported.

As an example to the contrary, a single centre prospective cohort study on 42 patients with acute iliofemoral DVT treated with ultrasound-accelerated CDT reported 63.8% primary patency rates at 38.5 month mean follow-up. ⁸ PTS development, assessed via the Villalta score, did not occur in 70% of participants; these patients had improved quality of life outcomes compared to those that developed PTS.

One of the main subjects of controversy is the definition of PTS using the Villalta score. Although widely used, ⁹ its lack of specificity and exclusion of certain factors (such as venous claudication) has been criticised. This is compounded by its use as a binary outcome in trials (i.e. PTS defined as an absolute score ≥5), instead of measuring a change in score.

Interpreting the trial results without considering the aforementioned limitations may reduce access to treatment to patients who may still benefit from CDT. In addition, trial data has important medico-legal and insurance implications. Based solely on the trial outcomes, healthcare professionals may struggle to justify offering CDT to patients with iliofemoral DVTs, particularly should complications from using this treatment arise. However, the trials have provided valuable information regarding the management of femoropopliteal DVTs. In an additional sub-analysis of the ATTRACT trial, CDT did not improve the rate of PTS, severity of PTS or quality of life in patients with femoropopliteal DVTs, confirming that CDT is not an appropriate initial management option for this cohort of patients. ¹⁰

Guidelines currently recommend that CDT is only considered for patients who present with acute symptomatic iliofemoral DVT, have a good functional baseline and have a low risk of bleeding. ¹¹ Ideally, further studies should include recruitment of patients with iliofemoral DVT, increased involvement of specialist centres with high caseloads of CDT, a focus on adjunctive procedures to maximise vein patency and consistent anticoagulation protocols. The issue of centre volume to outcome is well established in vascular surgery, ¹² however it does raise questions regarding service delivery and potential generalisability of findings, especially in certain locations, if the case is proven for CDT. Steps should be taken to define a more appropriate measure as a primary outcome in view of the limitations of the Villalta score. Funding organisations should be aware of the limitations of the current state of evidence and the need to continue to support future trials.

However, if further randomised-controlled trials struggle to receive funding in light of the previous attempts, specialist centre registries should be set up to continue evaluating the evidence for CDT. Indeed, one could argue that based on the sub-group analyses in the ATTRACT and CAVA trials, as well as other single centre observational studies, there is already enough evidence to justify carrying out CDT for this defined cohort of patients.