Catheter-directed thrombolysis did not benefit, may harm patients with DVT

Roy Silverstein, MD

The vast majority of clinical research in VTE over the past 3 decades has focused on identification of risk factors and development of best practices to prevent VTE in high-risk settings, and to prevent recurrent VTE in patients with an episode of VTE. Although these studies have led to remarkable progress, surprisingly little attention has been paid to developing strategies to lower the risk for post-acute local complications of lower-extremity DVT. These complications, known collectively as post-thrombotic syndrome (PTS), have significant negative impact on quality of life and cause considerable long term morbidity, including a more than 2.5-fold increased risk for recurrent DVT.
More aggressive acute treatment of proximal DVT with catheter-delivered localized thrombolytic therapy (CDT), with or without mechanical thrombus removal, has been shown in small clinical trials to lower risk of PTS and has been promoted by some clinicians as a modality that should be used much more frequently.
Bashir and colleagues used a large claims-based database of more than 90,000 US hospital discharges for proximal DVT without PE over a 6-year period ending in 2010 to assess acute outcomes in patients treated with CDT.
Perhaps the most important finding of this study is that CDT was used rarely; only 4% of hospitalized patients were treated. An obvious selection bias was demonstrated in that young, white males admitted to academic medical centers in urban centers were most likely to be treated. The authors used a state-of-the-art statistical methodology called propensity scores to try to overcome the selection biases by defining matched control populations and found that, by all measures, patients treated with CDT plus anticoagulation did worse than those treated with anticoagulation alone.
Not surprisingly, CDT led to longer length of stay, increased bleeding complications (including intracranial), increased transfusion utilization and increased costs. Perhaps more surprising, CDT-treated patients also had more inferior vena cava filters placed and higher likelihood of pulmonary embolism. These short-term complications were associated with a 1.4-fold increase in mortality (P=.015).
Patients treated in centers that performed more than five CDT procedures per year did better than those treated in centers that performed few procedures, and patients treated in the last 2 years of the observation period did better, suggesting perhaps that procedure morbidity can be lowered by increasing operator experience. However, these data may have been influenced by changes in the characteristics of patients hospitalized for DVT; in many cases, DVT is now managed without hospitalization.
This well-done observational study should cause clinicians who treat DVT to think carefully about the place of CDT in acute management. The authors suggest that this therapy should be limited to those patients at highest risk of PTS (eg, iliofemoral DVT) and lowest risk of bleeding. I would take this further — in light of the data showing increased risk for PE and strongly suggesting excess mortality — and say that use of CDT should be limited to patients enrolled in randomized clinical trials to assess both short-term and long-term outcomes.
Given the expense and difficulty of performing adequately powered studies of this type, an alternative approach could be mandatory enrollment of all CDT patients in a well-designed registry that requires follow-up for at least a year, and that assesses PTS and recurrent VTE as well as short-term complications. If coupled with robust use of statistical tools, such as those used by the authors of this study, such a registry could define a role for CDT in management of patients with acute proximal DVT. In the meantime, an approach that recognizes that less treatment can often be better than more seems prudent.