The case for catheter-directed thrombolysis in selected patients with acute proximal deep vein thrombosis

In the most extreme form of acute DVT, phlegmasia cerulea dolens, acute thrombosis in the proximal deep vein(s) causes total or near-total occlusion of venous drainage from the lower extremity. The resultant venous congestion leads to compromised circulation, jeopardizing viability of the affected limb. In such cases, it is well-accepted that immediate clot degradation, either by thrombectomy or thrombolysis, is necessary to restore venous patency and avoid limb gangrene and amputation.⁶  Thrombolysis is also indicated in less extreme instances in which DVT is not limb-threatening, but results in severe symptoms that are persistent or progressive despite initial treatment with anticoagulation.

The role of routine thrombolytic therapy in unselected cases of DVT, in contrast, is controversial. Rather than limb salvage, the desired goal in this setting is prevention of PTS. PTS results from venous hypertension, which is a consequence of incomplete clot lysis, residual venous obstruction, and venous valvular reflux.^7-9  Anticoagulation prevents thrombus formation and propagation, effectively reducing the risk of pulmonary embolism and recurrent venous thromboembolism. The clearance of existing thrombus, however, relies on the activity of the endogenous fibrinolytic system. The rate of thrombus resolution varies between and within individuals. In certain cases, impaired thrombus resolution leads to incomplete venous recanalization and residual vein obstruction.

The open vein hypothesis postulates that early effective thrombus removal may prevent residual venous obstruction, valvular reflux, and PTS. This concept laid the foundation for the adjunctive use of thrombolytic therapy in acute DVT to prevent PTS. In early studies exploring rates of venous patency and valvular reflux after acute DVT, addition of CDT provided superior results over anticoagulation alone,^10-12  suggesting its potential role in prevention or mitigation of PTS.

Methods for dissolving existing thrombus are categorized into pharmacologic (infusion of fibrinolytic drug to augment thrombus degradation), mechanical (use of mechanical force to break down and remove thrombus), and combined approaches.

In pharmacologic thrombolysis, fibrinolytic drugs are administered either systemically via a peripheral venous catheter (systemic thrombolysis) or locally at the site of thrombus through a percutaneous infusion catheter. Studies evaluating systemic thrombolysis in DVT demonstrated improved clot lysis and reduced risk of PTS, but these benefits were overshadowed by significant risk of major bleeding.¹³  In pharmacologic CDT, direct imaging-guided intrathrombus infusion achieves higher local concentration of fibrinolytics while minimizing drug levels in the systemic circulation, potentially enhancing efficacy while reducing bleeding complications. In a retrospective study comparing systemic and catheter-directed thrombolysis, pharmacologic CDT was associated with lower rates of valvular reflux and PTS.¹⁴  In addition, balloon angioplasty with or without stent placement may be considered in cases of anatomical abnormalities or residual flow obstruction after CDT.

Mechanical thrombectomy involves open surgical procedures to remove the clot or percutaneous insertion of a catheter-based device that aspirates or macerates the thrombus.¹⁵  The invasive nature of surgical thrombectomy limits its utilization to patients with limb-threatening DVT and high bleeding risk. Catheter-directed mechanical thrombectomy uses endovascular devices such as rotational wires, saline jet, or ultrasound wave to disperse the clot, as well as aspiration for immediate clot removal. Because their stand-alone use is not sufficiently effective, mechanical approaches are often used in conjunction with CDT.⁵  Pharmacomechanical CDT, the combination of pharmacologic CDT and catheter-directed mechanical thrombectomy, enables a reduction in dose and infusion time of the fibrinolytic agent and a shorter period of intensive care monitoring with similar success rates to pharmacologic CDT alone.^16-18 

Several randomized controlled trials have evaluated pharmacologic or pharmacomechanical CDT as a means of preventing PTS. Early trials were limited in quality by small sample size and use of nonvalidated scales for measuring PTS.^11,19,20  Two higher quality trials, Catheter-Directed Thrombolysis for Deep Vein Thrombosis (CaVenT) and Acute Venous Thrombosis: Thrombus Removal With Adjunctive Catheter-Directed Thrombolysis (ATTRACT), were completed more recently and used the Villalta scale to grade PTS.^21-23 

The Norwegian multicenter CaVenT trial randomized 209 acute IFDVT patients to receive standard anticoagulation alone or with adjunctive pharmacologic CDT using alteplase.²¹  The primary end points were 6-month iliofemoral patency and frequency of PTS (defined as Villalta score ≥5) at 2 years. Addition of pharmacologic CDT led to improvement in both outcomes with greater 6-month patency rates (65.9% vs 47.4%, P = .012) and a reduction in PTS at 2 years (41.1% vs 55.6%, P = .047). Pharmacologic CDT was associated with a higher risk of major bleeding (3% vs 0%). Most bleeding occurred at the venous access puncture site. A follow-up report demonstrated persistent benefits of pharmacologic CDT at preventing PTS at 5 years (42.5% in the CDT group vs 70.8% in the control group, P < .0001).²² 

In the recently published ATTRACT trial, 692 patients with acute proximal DVT were randomized to pharmacomechanical CDT or standard treatment. At 24 months, PTS (defined as Villalta score ≥5) was present in 47% of subjects in the CDT arm compared with 48% in the control group (P = .56).²³  Major bleeding was uncommon, but more frequent in the CDT group (1.7% vs 0.3%, P = .049).

The discrepant findings between CaVenT and ATTRACT may be attributable, in part, to patient selection. Although CaVenT primarily enrolled patients with IFDVT, 43% of patients in ATTRACT had femoropopliteal DVT, a population at inherently lower risk of PTS.²⁴  Indeed, although the overall results of ATTRACT were negative, there was a trend toward reduced moderate/severe PTS with pharmacomechanical CDT in the subgroup of patients with IFDVT (18.4% vs 28.2%).²⁴ 

To assess the efficacy of CDT in preventing moderate/severe PTS in patients with IFDVT, we performed a meta-analysis of CaVenT and the subgroup of patients with IFDVT in ATTRACT (Figure 1).^22,23,25  In the pooled analysis, CDT was associated with a 39% reduction in the risk of long-term moderate/severe PTS (risk ratio, 0.61; 95% CI, 0.43-0.86; P = .005). These results must be interpreted with caution. First, there were differences in design between the 2 trials (CaVenT used pharmacologic CDT as the intervention and measured PTS at 5 years; ATTRACT used pharmacomechanical CDT as the intervention and measured PTS at 2 years). Second, the data from ATTRACT are derived from a secondary end point (moderate/severe PTS) among a subgroup of patients (those with IFDVT). Third, some patients in CaVenT had isolated femoral rather than IFDVT²⁶ ; thus, the results of our meta-analysis do not provide proof of the efficacy of CDT, but they do highlight its potential value in selected patients, particularly those with IFDVT.

Consistent with a potential reduction in moderate/severe PTS, pharmacomechanical CDT was associated with improved disease-specific QOL in ATTRACT at 30 days and 6 months. Patients with IFDVT in the CDT arm experienced greater gains in QOL from baseline to 24 months compared with patients in the control arm.²⁷ 

CDT is indicated in patients with limb-threatening DVT or progressive thrombosis despite anticoagulation. In patients with non–limb-threatening DVT, we consider patient characteristics, values, and preferences; characteristics of the DVT itself; and availability of resources in selecting patients for CDT (Figure 2). In general, CDT may be considered in patients with IFDVT^21,28  and recent symptom onset (<14 days),²⁹  who are determined to be at low bleeding risk and are willing to accept a modest but increased risk of bleeding to reduce their risk of moderate/severe PTS. To achieve the best outcomes, CDT should be performed in centers with experienced interventionalists and resources for intensive care monitoring.

Although CDT is not suitable for all patients with proximal DVT and available evidence is inconclusive, it has potential value as a means of preventing moderate/severe PTS in carefully selected patients with IFDVT (Figure 1). Individualized risk–benefit analysis based on patient characteristics, DVT characteristics, patient values and preferences, and available resources is key to appropriate patient selection (Figure 2).

Contribution: T.C. and A.C. searched the literature and wrote the manuscript.

Conflict-of-interest disclosure: A.C. has served as a consultant for Genzyme, Kedrion, and Synergy, and his institution has received research support on his behalf from Alexion, Bayer, Bioverativ, Novo Nordisk, Pfizer, Shire, Spark, and Syntimmune. T.C. declares no competing financial interests.

Correspondence: Adam Cuker, Hospital of the University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104; e-mail: adam.cuker@uphs.upenn.edu.