To the Editor: We thank Marcy et al. [1] for their interest and detailed comments on our manuscript evaluating the performance of totally implantable venous access ports (TIVAPs) placed in the forearm [2]. Literature on arm-port placement is rare, and in most of the published studies, patients received antibiotic prophylaxis before implantation as a routine, although arguments for this strategy seem weak: A meta-analysis found no statistically significant benefit from antibiotic prophylaxis before insertion of long-term tunnelled central venous catheters [3]. The Centers For Disease Control and Prevention does not recommend administration of oral or parenteral antibacterial drugs to prevent catheter-related blood stream infection [4]. Insofar the study of Marcy et al. aims at the lack of evidence regarding infectious prophylaxis in patients undergoing port implantation. There is obvious disagreement about what is believed to be the right way to prevent infections and what is proposed by evidence-based medicine [5]. Nevertheless, in patients undergoing TIVAP implantation, one should differentiate between those with normal risk of device infection and those with increased risk of device infection. In addition to the groups mentioned by Marcy et al. (i.e., immunocompromised and diabetic patients), we would also consider systemic prophylaxis for haematological patients receiving induction chemotherapy, for those who are neutropenic at the time of catheterinsertion, and for those undergoing bone marrow transplantation. It would be useful to evaluate whether systemic antibiotic prophylaxis would be of value for this subgroup of high-risk patients. Furthermore, it would be beneficial to have guidelines, based on additional published data, for interventional radiologists regarding systemic infectious prophylaxis in patients undergoing TIVAP implantation. Own experience coincides with that of Marcy et al. regarding the predominance of Staphylococcus aureus and Candida species as being the most common pathogens among infected TIVAPs demanding explantation, and any peri-interventional prophylaxis should cover these pathogens [6]. With interest we noted Marcy’s statement regarding mechanical complications in the arm-port subgroup: They found a mechanical complication rate of 7.3%, which is much higher than our own experience (0.9% mechanical complication). To our knowledge, Marcy et al. implanted the TIVAPs in the upper arm, whereas we implanted these devices in the proximal third of the forearm. We use a tunnelling device, which is included in the device set, to create a small tunnel for the port catheter to cross the elbow joint. As described earlier by Yamamoto et al., this technique has a low rate of mechanical complications [7]. We believe that compared with upper-arm placement, the location of the port chamber in the forearm is the reason for the lower rate of mechanical complications in our study. Implantation in the forearm facilitates a longer and more harmonic course of the tunnelled port catheter, thus resulting in less flexibility and lower rates of catheter fractures, slits, and disconnections as well as preventing malrotation of the port chamber. In addition, the forearm has less subcutaneous fat tissue than the location used for implantation in the upper arm; this allows more accurate sizing of the port pocket, which therefore appears less extendable. Furthermore, it is our subjective impression that a forearm port is easier to access because it can be felt, J. P. Goltz (&) R. Kickuth Institute of Radiology, University of Wurzburg, 97080 Wurzburg, Germany e-mail: goltz@roentgen.uni-wuerzburg.de