Six Months vs Extended Oral Anticoagulation After a First Episode of Pulmonary Embolism: The PADIS-PE Randomized Clinical Trial

Authors of the publication: Couturaud F, Sanchez O, Pernod G, Mismetti P, Jego P, Duhamel E, Provost K, dit Sollier CB, Presles E, Castellant P, Parent F, Salaun PY, Bressollette L, Nonent M, Lorillon P, Girard P, Lacut K, Guégan M, Bosson JL, Laporte S, Leroyer C, Décousus H, Meyer G, Mottier D; PADIS-PE Investigators

Journal: JAMA. 2015
Volume: 314
Issue : 1
Page numbers: 31-40

Summary

To respect copyright, please do not copy the abstract of the paper!

Pulmonary embolism (PE) is considered to be unprovoked in the absence of a temporary or reversible risk factor (such as surgery, trauma, immobilization, pregnancy, oral contraceptive use or hormone replacement therapy) within the 6 weeks to 3 months prior to diagnosis [1]. The risk of recurrence is much higher at discontinuation of therapy after a first unprovoked PE as compared to provoked PE (4.5% vs 2.5% per year)[2]. For patients with PE secondary to a transient (reversible) risk factor, oral anticoagulation is recommended for 3 months[3], whereas the optimal duration of anticoagulation beyond 3 to 6 months in patients with unprovoked PE is uncertain [2, 4-6]. The 2014 ESC guidelines on the diagnosis and management of acute PE indicated that extended oral anticoagulation should be considered for patients with a first episode of unprovoked PE and low bleeding risk (recommendation class IIa)[3].

The aim of the PADIS-PE trial, recently published by Couturaud and colleagues in the JAMA [7], was to determine the benefits and harms of an additional 18 months of treatment with vitamin K antagonist (warfarin) vs placebo, after an initial 6 month non-randomized treatment period. This was a multicenter (14 centers), randomized, double-blind study involving patients with a first episode of unprovoked PE in the absence of active cancer.

In total, 371 patients who had been treated initially for 6 uninterrupted months with vitamin K antagonist, were randomized to receive, for a further 18 months, either warfarin with a target INR of 2.0-3.0, or placebo with sham INR. After the end of the treatment period, all patients were followed up for 24 additional months without anticoagulants. The primary outcome was the composite of fatal and nonfatal symptomatic recurrent venous thromboembolism (VTE) and major bleeding at 18 months after randomization. Secondary outcomes were the composite at 42 months (18 month treatment period plus 24 months of follow-up), individual components of the composite, as well as death unrelated to PE or major bleeding at 18 and 42 months.

During the 18 month treatment period, the primary outcome occurred in 3.3% of patients in the warfarin group and in 13.5% of patients in the placebo group, resulting in a relative reduction of 78% in favour of warfarin (hazard ratio [HR], 0.22 ; 95%CI, 0.09-0.55 ; P=0.001). This result was driven by an 85% reduction in the risk of recurrent venous thromboembolism in the warfarin group (1.7% in the warfarin group vs. 13.5% in the placebo group ; P<0.001) at the cost of a minimal increase in the risk of bleeding in this group (2.2% vs. 0.5% respectively ; P=0.22). However, the benefit was not maintained once anticoagulation was discontinued, with a rebound in recurrent VTE events observed early after discontinuation of active treatment. During follow-up after treatment discontinuation, 25 patients (13.6%) in the warfarin group presented symptomatic recurrent VTE, vs 14 patients (7.5%) in the placebo group, resulting in a non-significant difference in the rate of composite outcome over the entire 42-month study period (HR, 0.75 ; 95%CI, 0.47-1.18 ; P=0.22). Finally, rates of recurrent VTE, major bleeding and unrelated death did not differ between groups.

Commentary

In light of these results from the PADIS-PE study, patients with a first episode of unprovoked PE could benefit from secondary prophylaxis with extended oral anticoagulation using vitamin K antagonist. Moreover, the place of direct oral anticoagulants, which are as effective as warfarin with a lower bleeding risk [8-10], remains to be defined. In particular, direct oral anticoagulants at reduced dose have shown promise for long-term treatment, with a significant reduction in recurrent events shown in the AMPLIFY-Extension trial, without an excess of bleeding events [11]. Further studies are ongoing, in particular with rivaroxaban, notably the EINSTEIN-Choice study. The selection of candidates for long-term anticoagulant therapy after a first unprovoked PE is highly complex. A number of variables have been identified as predictors of recurrent VTE in this setting. In particular, the D-dimer level at the time of discontinuation of anticoagulant therapy has been evaluated with a view to identifying patients in whom anticoagulation may safely be stopped [12].

Why is this paper important and what does this paper add compared to previous evidence?

This paper is important for two reasons. Firstly, it confirms the increased risk of recurrent PE after a first unprovoked episode, particularly during the first 6 months after interruption of therapy.

Secondly, it shows that once the decision to administer long-term anticoagulation has been made, then this treatment should not be interrupted, as there is a rebound in recurrent events after discontinuation of therapy. It is advisable to review the risk-benefit ratio of this strategy at regular intervals to decide on whether or not to pursue therapy, in particular taking into account the patient’s bleeding risk.

How is this paper likely to change practice?

It is likely that, in light of these results, prescription of anticoagulant therapy could be of 3 different durations, depending on the clinical context, namely:

• 3 months for a first PE occurring in the presence of a transient risk factor.
• 6 months for a first unprovoked PE in patients with a high risk of bleeding.

Long-term treatment for patients with a first unprovoked PE and a low bleeding risk. However, the duration of therapy, the appropriate drug and dose remain to be defined. Hopefully further studies will confirm these findings and provide answers regarding the optimal duration, drug and dose.

Authors of the report: Dr. Romain Chopard, Prof. Nicolas Meneveau, Fiona Ecarnot.
EA3920, University Hospital Besancon, 25000 Besancon, France