Author + information
- Rhian E. Davies,
- Andrew Foy and
- Ian Gilchrist
Background: Trial sequential analysis (TSA) is a technique used to establish the threshold when firm evidence is reached in meta-analyses. We performed a TSA of randomized controlled trials (RCTs) comparing thrombolysis to anticoagulation alone for all-cause mortality in patients with intermediate-risk pulmonary embolism (PE).
Methods: Eight RCTs were identified that specified therapy for hemodynamically stable PE with objective assessment of RV function for inclusion. The Mantel-Haenszel method was used to calculate fixed and random effects odds ratios for mortality. Testing for heterogeneity and publication bias were performed using standard methods. TSA was performed with 5% risk of type I error based on the random effects model. Monitoring boundaries were constructed according to the required information size to detect or reject an intervention effect equal to that observed in the included trials with a type II error of 20% (power of 80%).
Results: Thrombolysis reduced mortality compared to anticoagulation alone according to the fixed effects (OR 0.51; 0.27-0.97) but not the random effects model (OR 0.59; 0.30-1.18). Publication bias is likely based on funnel plot assessment. TSA cannot confirm or exclude a relative risk reduction of 41%. This would require an additional trial(s) with 1,918 patients.
Conclusions: There remains considerable uncertainty regarding the effect of thrombolysis on mortality in patients with intermediate-risk PE. TSA shows that additional studies are required.
Poster Hall, Hall C
Friday, March 17, 2017, 3:45 p.m.-4:30 p.m.
Session Title: Updates on Pulmonary Embolism Management in 2017
Abstract Category: 35. Pulmonary Hypertension and Pulmonary Thrombo-embolic Disease
Presentation Number: 1141-007
- 2017 American College of Cardiology Foundation