Letters Section Editor: Robert M. Golub, MD, Senior Editor.
In Reply: Dr Hartig and colleagues speculate that the trend toward lower bleeding with fondaparinux may be due to inadvertent unblinding and consequent imbalance in the use of antithrombotics, such as LMWH. This is very unlikely for 3 reasons. First, the lower rate of bleeding with fondaparinux was also observed in individuals not receiving LMWH, who were the overwhelming majority of patients enrolled in the trial (93.7% of the control group and 94.7% of the fondaparinux group). Open-label LMWH was most likely used in response to a failure of the original randomized treatment, such as recurrent ischemia or reinfarction, and the lower rate of LMWH use in the fondaparinux group is likely a reflection of its superior efficacy in reducing these events, as well as lower mortality. Second, the entire difference in bleeding could be accounted for by lower rates of cardiac tamponade in the fondaparinux group, which has a very high fatality rate. Differences in reporting of cardiac tamponade are unlikely to be due to ascertainment biases, given the seriousness of the condition. Third, rates of bleeds other than from cardiac rupture were similar between the groups, indicating no differential in the ascertainment of bleeding events between the randomized groups. We did not assess the blinding of the placebo because it could have led to unblinding.
Yusuf S, Mehta SR. Fondaparinux in Patients With ST-Segment Elevation Myocardial Infarction—Reply. JAMA. 2006;296(17):2087-2089. doi:10.1001/jama.296.17.2088-b