The aim of our study was to determine the six-month adverse event rate following the use of pulmonary magnetic resonance angiography (PE-MRA) for the primary diagnosis of pulmonary embolism (PE).This was a retrospective HIPAA-compliant, IRB-approved study. We reviewed the electronic medical records (EMR) of all patients who underwent PE-MRA scans performed between 4/1/2008-3/31/2013 as a first line diagnostic test for PE. These exams all used a previously described method.(Ref 1, 2) We excluded patients with current anti-coagulation, pre-existing IVC filter, or atrial fibrillation. The final radiology report was used to determine if the exam was technically limited, based upon inclusion of the word “limited” or a recommendation for additional imaging to rule out PE. If technically successful, the exam was categorized as positive, negative, or equivocal, again based upon the final report. Using the standard definitions of major adverse events derived from the pulmonary computed tomographic angiography (CTA) literature (Ref 3,4), we reviewed the EMR and recorded the following adverse events: venous thromboembolism (VTE), major bleeding and all-cause mortality that occurred within 6 months following the index PE-MRA exam.A total of 675 patients underwent PE-MRAs during the inclusion period (Figure 1). Of these, 56/675 (8.3%) were excluded and 2/675 (0.3%) had incomplete EMR reviews. This resulted in 617/675 (91.4%) being included in the analysis (Figure 1). Of the included cases, 500/617 (81%) were negative for PE, 17/617 (2.8%) were equivocal, 46/617 (7.5%) were positive for PE (Figure 2), and 54/617 (8.8%) were technically limited. Overall, the major adverse event rate following a technically successful PE-MRA was 32/563 (5.7%). Outcomes broken down by subgroup based on the results of the PE-MRA are reported in Table 1. Following a negative PE-MRA scan, only 3/500 (0.6%) experienced a VTE within 6 months. When all-cause mortality is included, only 15/500 (3%) patients experienced an adverse event that possibly could have been related to a false negative test – a 97% negative predictive value (NPV). Following a positive PE-MRA exam, 2/46 (4.3%) patients experienced a major bleeding event and 4/46 (8.7%) experienced adverse events that may have been related to anti-coagulation for treatment of PE. The 6-month major adverse event rate following PE-MRA observed in this study was slightly greater than the reported 3-month adverse event rate following CTA (~3% for those with positive CTA) ( Ref 3) ; ~1.7% for those with negative CTA (Ref 4). The inclusion of all-cause mortality as an adverse event likely overstates the risk of adverse events related to PE. Notably, only 8.8% of the analyzed exams were reported as technically limited in some way. This is far lower than the 25% uninterpretable technical failure rate reported in the PIOPED III study (Ref 5). This improvement in technical success likely reflects the maturation of PE-MRA methodology since the time of the PIOPED III scans nearly a decade ago (2006-2008). Limitations of this retrospective outcomes analysis include: (1) Potential selection bias related to ED physician ordering preferences for younger and female patients; (2) 8.8% of analyzed exams were reported to be limited within the radiology report, which is higher than a previously reported technical failure rate of 2.6%, ( Ref 2) which is likely due to the more conservative criteria used in the present study; and (3) Incomplete data regarding the cause of patient deaths. Using a very conservative outcome metric that includes all-cause mortality and all interval VTE events observed within 6 months of a negative PE-MRA, we found that the NPV of PE-MRA was 97%. This value compares favorably to the reported NPV of CTA (98%) for this indication. These clinical outcomes demonstrate that pulmonary magnetic resonance angiography is a safe non-ionizing alternative to computed tomographic angiography for the primary diagnosis of pulmonary embolism.The use of Gadolinium based contrast material for pulmonary magnetic resonance angiography is an off label use of these agents.