We thannk Dr Collignon for his interest to our work,1 and for the opportunity he gives us to clarify some points. The major aim of our study was to determine the critical level of a simplified quantitative culture technique of catheter tips that would accurately identify clinically significant catheter-related sepsis (CRS), whether bacteremic or not. We found that a cutoff of ≥103 colony-forming units (CFUs) was adequate for this purpose, yielding a sensitivity and specificity for the diagnosis of CRS, both over 85%.
Dr Collignon questions the use of the positivity of the catheter tip culture as one of the criteria for the diagnosis of CRS in our study. Clinical criteria alone cannot be used in such a study population with central vein catheterization. Besides, the knowledge of a bacterial growth from the catheter is obviously necessary for the interpretation of bacteremic episodes. If one is to ascribe