Send to

Choose Destination
See comment in PubMed Commons below
Reg Anesth Pain Med. 2013 Nov-Dec;38(6):515-9. doi: 10.1097/AAP.0000000000000009.

An analysis of the safety of epidural and spinal neuraxial anesthesia in more than 100,000 consecutive major lower extremity joint replacements.

Author information

From the *Center for Musculoskeletal Surgery, Charité-Universitätsmedizin Berlin, Berlin, Germany; †Department of Anesthesiology, Perioperative Medicine and Intensive Care Medicine, Paracelsus Medical University, Salzburg, Austria; ‡Departments of Anesthesiology, §Radiology, and ║Orthopedic Surgery, Hospital for Special Surgery, Weill Medical College of Cornell University, New York, NY.



A feared complication of spinal or epidural anesthesia is the development of epidural or spinal hematoma with subsequent neural element compression. Most available data are derived from the obstetric literature. Little is known about the frequency of hematoma occurrence among patients undergoing orthopedic joint arthroplasty, who are usually elderly and experience significant comorbidities. We sought to study the incidence of clinically significant lesions after spinal and epidural anesthesia and further describe their nature.


We retrospectively analyzed a database of all patients who underwent total hip or total knee arthroplasty under neuraxial anesthesia at our institution between January 2000 and October 2010. Patients with radiographically confirmed epidural lesions were identified and further analyzed.


A total of 100,027 total knee and hip replacements under neuraxial anesthesia were performed at our institution. Ninety-seven patients underwent imaging studies to evaluate perioperative neurologic deficits (0.96/1000; 95% confidence interval, 0.77-1.16/1000). Eight patients were identified with findings of an epidural blood or gas collection (0.07/1000; 95% confidence interval, 0.02-0.13/1000). No patients receiving only spinal anesthesia were affected. All patients diagnosed with hematoma took at least 1 drug that potentially impaired coagulation (5 nonsteroidal anti-inflammatory agents, 1 a tricyclic antidepressant, and 1 an antiplatelet drug). No patient incurred persistent nerve damage.


The incidence of epidural/spinal complications found in this consecutive case series is relatively low but higher than previously reported in the nonobstetric population. Further research using large data sets could quantify the significance of some of the potentially contributing factors observed in this study.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Full text links

    Icon for Lippincott Williams & Wilkins
    Loading ...
    Support Center