There are few data in the literature regarding the safety of neuraxial techniques in patients with the most common bleeding diatheses, including hemophilia, von Willebrand's disease (vWD), and idiopathic thrombocytopenic purpura (ITP). Neuraxial techniques are not widely used in these populations because of concerns of potential hemorrhagic and/or subsequent neurologic complications. In this article, we review the available literature describing neuraxial techniques in patients with hemophilia, vWD, or ITP with the aim to assist anesthesiologists considering neuraxial techniques in these populations. After a systematic Pubmed, MEDLINE, and EMBASE search, we reviewed 30 articles published between January 1, 1975 and October 1, 2008 in which neuraxial techniques were performed in patients with hemophilia, vWD, or ITP to determine the perioperative management and evaluate the frequency of hemorrhagic complications. We identified 507 neuraxial techniques (482 patients) performed in patients with hemophilia (107 neuraxial techniques, 85 patients), vWD (74 neuraxial techniques, 72 patients), or ITP (326 neuraxial techniques, 325 patients). Among the 507 neuraxial techniques performed, there were 371 lumbar epidural anesthetics, 78 spinal anesthetics, 53 lumbar punctures, 2 combined spinal epidural analgesia, 2 paravertebral blocks, and 1 thoracic epidural anesthetic. Four hundred six neuraxial techniques were placed in the obstetric population, 53 were performed in the emergency room for diagnostic lumbar puncture, 46 were performed for lower limb orthopedic surgery, 1 was performed for postoperative analgesia, and 1 was performed for an obstetric patient undergoing non-obstetric surgery. Factor replacement to normal levels (>0.5 IU mL(-1)) was initiated before block performance, though treatment was not standardized, in 105 of 107 patients with hemophilia and 10 of 74 with vWD. Sixty-four of the 74 patients with vWD had spontaneous normalization of factor levels before block performance. No hemorrhagic complications were reported when the diagnosis of hemophilia or vWD was known before the neuraxial technique. A single case of spinal hematoma (resulting in permanent paraplegia) was identified when the presence of hemophilia was not known before needle insertion and factor replacement had not been given. In all 326 cases of ITP, with or without systemic treatment of platelet transfusion, there were no reports of hemorrhagic complications associated with neuraxial techniques. Among the 326 neuraxial techniques placed in the setting of ITP, 9 patients had platelet counts of <50 x 10(9) L(-1), 19 had a platelet counts of 50-75 x 10(9) L(-1), 204 had a platelet counts of 75-100 x 10(9) L(-1), and 94 had a platelet count more than 100 x 10(9) L(-1) before needle insertion. There is a paucity of published data regarding the provision and safety of neuraxial techniques in patients with common bleeding diatheses. The minimum "safe" factor levels and platelet count for neuraxial techniques remain undefined in both the obstetric and general populations, and evidence-based recommendations in the setting of hemophilia, vWD, or ITP cannot be offered.