A word of caution is due in the interpretation of this study. The absolute number of adverse outcomes was small, potentially limiting the precision and generalizability of the study. A number of patients did not have hs-cTnI levels measured. This was more common in patients with active malignancy, so caution should be taken in applying the results to this group of patients. Perhaps most importantly, 10 patients identified as low risk ultimately experienced an escalation in care. Although none of these patients underwent thrombolysis or CPR or died, three of them had legitimate reasons for ICU admission (hypoxemia requiring nonrebreather mask or noninvasive ventilation or hypotension requiring vasopressor support). Of these three, one of them was in the high-risk Pulmonary Embolism Severity Index group, and, thus, the combination of a clinical score and the hs-cTnI level may have prevented this miscategorization. Prior studies have confirmed the very high negative predictive value of N-terminal pro-BNP levels in patients with acute PE. Would the addition of this biomarker reduce the number of patients inaccurately classified as low risk?6 Further, would a panel of cardiac biomarkers (serum sodium, hs-cTnI or hs-cTnt, N-terminal pro-BNP, heart-type fatty acid binding protein, and D-dimer level) assist accurate discrimination? These are questions that deserve further investigation.