To the Editor:
We read with great interest the recent papers and the joint editorial reporting intraoperative tranexamic acid use in craniosynostosis surgical repair,1,2,3and we do agree with Holcomb's point of view: “It works, but how?” and in whom? In our pediatric craniofacial surgical center we, as many others involved in cranisosynostosis surgical management, are still facing two problems that could be finally considered incompatible: uncompressible large blood losses proportional to suture involvement and surgical techniques, that remain difficult to precisely evaluate,4and limited total blood volume in low-weight children. Reducing blood losses and transfusion requirements has been a major issue in the past 20 yr with active research on development of minimally invasive surgical techniques,5autologous transfusion methods,6and adjunctive medical treatments. Eliminating the need for intraoperative homologous blood transfusion could be considered as the final goal of these attempts, but remains, in daily practice, an inaccessible dream in malformations with multiple sutures, requiring complex surgical procedures. It could be only considered as a reasonable objective in children with scaphocephaly, requiring simple linear craniotomies.
To be relevant, clinical studies have some prerequisites. The first one is the need for a valuable evaluation of peroperative blood losses, based upon calculation of estimated erythrocyte volume lost, that was used by Goobie et al. , after others,1,4,6but not by Dadure et al. The second one is a need for strict hemodilution guidelines with commonly defined transfusion thresholds, eliminating the biases related to overhydratation with inadequate fluid loading7and underestimation of minimally required hemoglobin level. These issues are illustrated in the two papers that both concluded that tranexamic acid could reduce the need for blood transfusion. In Goobie et al. 's study, hematocrit threshold for blood transfusion was 30% (estimated hemoglobin 9 to 10 g/dl), all children were finally transfused, and calculated blood losses were significantly reduced with tranexamic acid. In Dadure et al. 's study, hemoglobin threshold was 7g/dl, 63% of the children did not have transfusions, and even though blood losses, which were not really estimated, did not differ between the two groups, the questioning conclusion was that tranexamic acid did decrease the need for transfusion. It could be concluded by readers either that tranexamic acid 50 mg/kg could reduce blood transfusion requirement by the means of reduced preoperative blood losses, or that tranexamic acid 15 mg/kg reduces blood transfusion requirement but not blood losses by the means of an unknown mechanism in children with acute normovolemic hemodilution. Despite well-designed, double-blinded, randomized studies, the authors should be aware that important issues for determining the efficiency of a medical treatment are the homogeneity of the study population, the size of the sample, and the reproducibility of the protocol in clinical practice. Both these studies included around 40 patients, with various surgical conditions ranging from severe syndromic faciosynostosis to simple scaphocephaly, and requiring various types of procedures. It has been previously clearly demonstrated that the type of malformation and the type of resulting surgical procedure is the main determinant of preoperative blood losses.4Including, in a small sample of patients, numerous subgroups requiring various surgical managements could significantly attenuate the statistical power of a study. It could be therefore more efficient to further determine the benefit of tranexamic acid in a large, homogenous series of patients with simple suture involvement, the population who may most benefit from efficient adjunctive techniques to reduce intraoperative blood losses in craniosynostosis surgical repair. Finally, as underlined by Holcomb, an important question remains unanswered: If tranexamic acid could be efficient to reduce intraoperative blood losses and transfusion requirements in infants, how does it work?