Dear Editor,We thank Sugiyama et al. [1] for their comments in relation to our study [2], in which we investigated potential predictors of bone mineral density trajectory in children exposed to warfarin for at least 1 year.In their letter, they elegantly summarize the current knowledge on the effect of warfarin on bone health, highlighting the complexity of this research area in adult patients. Yet, research in the pediatric population faces additional difficulties. Beyond the obvious limitation of the far smaller number of children exposed to warfarin as compared to adults, there are a number of issues about the assessment of bone health in children that remain to be solved. Current official guidelines [3,4] address the uncertainties surrounding the use of bone densitometry in pediatrics, the discussion on the best method for body size adjustment, and the limited but growing understanding of how bone density assessment relates to important clinical outcomes such as bone fractures. An added challenge is that the risk of bone fracture in children differs from that of adults, being influenced by the interacting effects of bone size, bone mass, and activity level [5].Despite the above issues, the health of a growing skeleton exposed to warfarin remains a concern that cannot be ignored. Our study focused on a practical perspective: understanding what additional variables a pediatric hematologist should consider when following a child who is already on warfarin. We considered that this approach would give our results more immediate application in clinical practice.Because the study only enrolled patients on warfarin, comparing the bone density of the warfarin-exposed to warfarinunexposed bone was beyond the scope. Nevertheless, our longitudinal analysis revealed an interesting point: once major risk factors were accounted for, the size-adjusted bone density Z-scores of patients exposed to warfarin did not significantly change over time (mean Z-score change 0.033/year on warfarin, 95 % confidence interval −0.026 to +0.092). In contrast, a methodologically comparable longitudinal study conducted at our institution showed that the size-adjusted bone density Z-scores of 68 children with systemic lupus erythematosus deteriorated over time, decreasing by 0.06/year. Interestingly, weight, pubertal status, and prednisone predicted bone density trajectory in the study [6]. None of those patients were exposed to warfarin.It could be argued that interventions must have occurred to maintain the bone health of our patients. Even if we disregard the fact that over two thirds of the children were not compliant to vitamin D/calcium supplementation, or that one third of them were not physically active, the finding still does not suggest an overt deleterious effect of warfarin on the growing bone among monitored patients (≥5 years of age). Moreover, we did not find unusual patterns in the frequency and location of fractures, as noted by Sugiyama et al. Nevertheless, we must emphasize that,