To the Editor: We read with great interest the comments on our article1 by Chen et al. The authors indicate a potential immortal time bias in our study,1 which necessitates a more detailed elaboration as subsequently outlined. First, our study is a retrospective observational study including patients from > 20 centers spanning over a period of nearly 20 years. This methodology is obviously associated with a potential analysis bias particularly in the light of the limited number of patients. However, in the context of such a rare disease, prospective studies are virtually impossible to conduct. Second, the authors state that the main results of our study, the pivotal role of anticoagulation in critically ill patients with antiphospholipid syndrome (APS), might have been unjustifiably extrapolated to other institutions. However, anticoagulation2 and triple therapy3 are the recommended treatment for APS and catastrophic antiphospholipid syndrome (CAPS), respectively. Third, the zero time point for the mortality analysis was ICU admission, and we took into account all APS-specific treatments received during the hospital stay before ICU admission. Indeed, most of the patients received anticoagulation as an element of their usual treatment and its interruption was frequently a precipitating event for the APS episode under investigation. Fourth, in practice, nonsurvivors had the chance to receive anticoagulation/corticosteroids because most of them did not die early. Their hospital stays before ICU admission and ICU stay were longer than those of the survivors (11 vs 5 days, P = .003; 16 vs 9 days, P = .025, respectively). Moreover, that duration was similar for those patients who did or did not receive anticoagulation: 6 vs 8.5 days (P = .60) and 10.5 vs 6 days (P = .40), respectively. Fifth, the most relevant finding of our work is that nonsurvivors were less likely to receive anticoagulation. In these patients, anticoagulation was frequently contraindicated because of severe bleeding complications (eg, intracerebral hemorrhage). In our experience, considering anticoagulation as being contraindicated in patients with APS/CAPS may cumulate into a dreadful therapeutic decision associated with poor outcome. Finally, our results are in agreement with previously published literature on APS and CAPS.1,4 In conclusion, we acknowledge that the methodology of our study is inevitably associated with an inherent degree of potential bias; however, we think the issue of immortal time bias does not substantially impede the internal validation of our study and therefore allows to underscore our principal findings on the relevance of anticoagulation in critically ill patients with APS.