V. Pelvic pseudotumor and pseudoaneu- rysm in a pediatric patient with moder- ate hemophilia B: successful management with arterial embolization and surgical excision. Pediatr Blood Cancer 2011; 56: 484–7. 10 Espandar R, Heidari P, Rodriguez-Merchan EC. Management of haemophilic pseudotu- mours with special emphasis on radiother- apy and arterial embolization. Haemophilia 2009; 15: 448–57. Should ice be used in the treatment of acute haemarthrosis in haemophilia? M. RAJAMANICKAM,* R. MICHAEL,* V. SAMPATH,* J. A. JOHN,* A. VISWABANDYA † and A. SRIVASTAVA † *Department of Physical Medicine and Rehabilitation; and †Department of Haematology, Christian Medical College, Vellore, India Forsyth et al. [1] have reviewed the use of cryotherapy for acute haemarthrosis in haemophilia and advised against its use. Their con- clusion is based on two forms of evidence that they have reviewed and then discussed their potential impact on this form of therapy. Their conclusions may be summarized as follows: 1. There is in vitro data that primary haemostasis (platelet func- tion) is affected as evidenced by prolonged bleeding time or reduced platelet aggregations by temperatures below 33°C [2] and that coagulation (secondary haemostasis) is also affected by cooling below 27°C [3] as evidenced by prolongation of prothrombin time and activated partial thromboplastin time. 2. The second part of their argument is derived from the first that such cooling will affect haemostasis in the joint and will there- fore be harmful. We have several concerns with this approach in terms of the logic used, the models from which the data have been derived and the conclusion made. In any acute haemarthrosis, one would expect the vascular leak that resulted in the bleed to be sealed as soon as ade- quate factor replacement therapy has been administered. This may take longer among those who may not receive any factor replace- ment. The natural history of these bleeds even among those patients suggests that the bleed stops in most patients in the first 12–24 h and resolutions begin from 2nd to 3rd day. In any case, for the vast majority of patients, once the haemostatic plug has been established in the first few hours, the extended data and arguments regarding the effect of cooling on haemostasis become essentially irrelevant. Measuring intra-articular temperature is of limited relevance there- fore, because the haemostatic plug could be formed inside the blood vessel where the temperature of flowing blood is likely to be higher. In fact, the authors have drawn most of their data from models where the pre-existing temperature in the joint was normal. They have not considered the fact that in acute haemarthrosis, the resul- tant inflammatory response would create a higher than normal tem- perature within the joint [4, 5] and a reduction with cryotherapy therefore may not drop the temperature to as low as they have sug- gested. Furthermore, they have also not considered the implication of the technique of cryotherapy used by individual patients particu- larly in terms of the duration of each application (say 10 min or 20 min) [6] and whether the plastic containing ice is directly applied to the skin or through a layer of cloth [7]. These are critical issues in this kind of treatment. We would further argue that the clinical features of acute haem- arthrosis after initial haemostasis has been achieved are more related to the inflammatory response to the blood in the joint space rather than continued leakage from the vascular compartment. We there- fore need to consider more the impact that cooling of the joint space will have on the inflammatory response in the joint [8] rather than the haemostatic plug formation in the blood vessel. This aspect has been completely ignored in this review by the authors. There is evi- dence from some elegantly conducted randomized clinical studies that cooling can be beneficial in the management of injuries around the joint [9, 10]. Finally, we would like to submit that thousands of patients with haemophilia around the world have applied ice for years and con- tinue to do so. They do so because a vast majority of them find ben- efit from it. We have recently surveyed patients being treated at our own centre regarding their experience with cryotherapy for acute haemarthrosis. Thirty patients with severe haemophilia (>7 years old) were administered a questionnaire after informal consent. All patients (100%) reported regular use of cryotherapy during an acute bleed, with a majority using it for both pain and swelling (73.4%). Most of them applied ice in plastic bags or cold packs (89.3%), wrapped in one layer of thin towel (72%), for 10–20 min each time (66.7%), every 1–2 h (63%) after an acute bleed for 2–3 days. All patients (100%) experienced decrease in pain or swelling with 76% having reduction of both pain and swelling. An overwhelming majority (78.6%) of the persons with haemophilia also reported a reduction in their factor usage after the use of cryotherapy. No adverse effect was reported by 76.7% of the patients. The adverse effects noted by 23.3% of patients included increased pain (two patients, 6.7%), numbness (two patients, 6.7%) and erythema (three patients, 9%). The latter was reported by those who applied ice directly and for longer durations (  20 min.) We would therefore like to conclude that it is premature, even inappropriate, to conclude that cryotherapy for acute haemarthrosis in haemophilia can be harmful and should therefore not be used. We think that such conclusions have been based on extrapolation of data that were all irrelevant to the situation under consideration and a flawed logic of the impact of intraarticular cooling on acute haem- arthrosis. It also ignores the experience of benefit that the vast majority of patients and their carers have observed over many years. What is needed is further clinical evaluation of this modality of treatment through well designed studies and greater effort to opti- mize the technique of cryotherapy as well as to understand the basis of the benefit experienced by the patients and not arguments against this treatment based on simplistic extrapolation of data from inap- propriate models of the effect of low temperatures on haemostasis and therefore acute haemarthrosis in haemophilia. Correspondence: Merlyn Rajamanickam, MPT, Department of Physical Medicine and Rehabilitation, Christian Medical College, Vellore 632004, India. Tel.: +91 416 2282663; fax: +91 416 2232035; e-mail: merlyntilak@gmail.com Accepted after revision 2 April 2013 DOI: 10.1111/hae.12163 © 2013 John Wiley & Sons Ltd Haemophilia (2013), 19, e256--e269 LETTERS TO THE EDITORS e267