A recent systematic review and meta-analysis showed that, in patients with iliofemoral deep vein thrombosis (DVT), percutaneous mechanical thrombectomy was associated with a higher cumulative six-month primary patency and a lower incidence of major bleeding compared to thrombolysis alone. These findings were recently published online in the European Journal of Vascular Medicine (VASA).
The authors, Michael KW Lichtenberg (Venous Center Klinikum Arnsberg, Arnsberg, Germany) and colleagues, conclude: “Our meta-analysis of thrombolysis alone versus mechanical thrombectomy with or without thrombolysis in patients with iliofemoral DVT revealed similar incidences of acute successful thrombus removal. However, mechanical thrombectomy increased six-month cumulative primary patency and, subject to publication bias, decreased frequency of major bleeding.”
They add that no differences across groups became obvious with regard to valvular reflux or pulmonary embolism and that the level of risk and clinical relevance of haemolysis resulting from mechanical thrombectomy “needs to be evaluated in future studies”.
Detailing the rationale behind the study, Lichtenberg et al outline that iliofemoral DVT accounts for about 25% of lower limb DVT. Patients with acute iliofemoral DVT carry a high risk of developing potentially chronic debilitating post-thrombotic syndrome (PTS). Residual thrombus may increase the risk of PTS as well as recurrent DVT, and pulmonary embolism, they note. Whether early thrombus removal in addition to standard care of anticoagulation and compressive therapy decreases the risk of PTS has not been clarified yet, the authors write.
Lichtenberg and colleagues identified observational and randomised trial published between January 2001 and February 2019 using MEDLINE. They included studies on DVT treated with either thrombolysis alone or percutaneous mechanical thrombectomy adjunctive to conventional anticoagulation and compressive intervention.
The investigators then conducted a meta-analysis of proportions to assess effectiveness outcomes of successful lysis and primary patency, post-thrombotic symdrome, valvular reflux, recurrent DVT, as well as safety outcomes of major bleeding, haematuria, and pulmonary embolism.
Lichtenberg et al identified 77 records, of which 17 studies comprising 1,417 patients were eligible for inclusion.
They report that pooled proportion of successful lysis was similar between groups (thrombolysis alone: 95% [I2=68.4%]; percutaneous mechanical thrombectomy: 96% [I2=0%]; Qbet [Cochran’s Q between groups] 0.3, p=0.61). However, pooled proportion of six-month primary patency was lower after thrombolysis alone than after percutaneous mechanical thrombectomy (68% [I2=15.6%] versus 94%; Qbet 26.4, p<0.001).
In addition, they found that considerable heterogeneity between groups did not allow for between-group comparison of thrombolysis alone and recurrent DVT.
Major bleeding was more frequent after thrombolysis alone than after percutaneous mechanical thrombectomy (6% [I2=0%] versus 1% [I2=0%]; Qbet 12.3, p<0.001).
Incidence of haematuria was lower after thrombolysis alone as compared to percutaneous mechanical thrombectomy (2% [I2=56%] versus 91.3% [I2=91.7%]; Qbet 714, p<0.001).
Finally, the incidences of valvular reflux and pulmonary embolism were similar across groups (thrombolysis alone: 61% versus percutaneous mechanical thrombectomy: 53%; Qbet 0.7, p=0.39 and thrombolysis alone: 2% versus percutaneous mechanical thrombectomy: 1%; Qbet 1.1, p=0.3, respectively).
In the discussion of their findings, Lichtenberg and colleagues consider how these results might affect practice. “On one hand, thrombolysis supports standard treatment of anticoagulation by means of improved thrombus clearance, but on the other, long lasting infusion of up to 72 hours and the need of intensive care unit monitoring due to an increased risk of bleeding is necessary,” they begin.
The authors continue: “Percutaneous mechanical thrombectomy is shown to accelerate thrombus maceration and removal and to effectively decrease residual thrombus burden by means of aspiration. Thus, with rapid reperfusion by percutaneous mechanical thrombectomy, thrombolytic medication can be reduced or completely avoided. It is therefore to be expected that the incidence of bleeding decreases without compromising effectiveness.”
Lichtenberg et al acknowledge some limitations to their study. Firstly, they note that comparison is based on study-level results from small to medium scale studies with different patients and procedure characteristics such as additional stenting in varying frequencies. For example, the percutaneous mechanical thrombectomy group included studies that applied mechanical thrombectomy alone as well as studies that combined mechanical thrombectomy. “This approach might have increased heterogeneity,” Lichtenberg et al comment.
Furthermore, they note that outcome definitions, including primary patency, recurrent DVT, and severity of post-thrombotic syndrome as well as follow-up time were not consistent across studies and that evidence is limited by moderate to considerable confounding, selection, misclassification, or information bias of included studies.
In addition, due to the small number of studies regarding lysis, primary patency, recurrent DVT, and pulmonary embolism, “publication bias must be interpreted with caution,” they remark.
Finally, due to the differences in the mechanisms of action for the analysed percutaneous mechanical thrombectomy devices, the investigators caution that “it may not be correct to group the findings and assume the results can be consistently assumed for each device i.e. class effect”.