Copyright © 2009 by The Journal of Bone and Joint Surgery, Inc.

Commentary & Perspective

Commentary & Perspective on
"Thromboprophylaxis in Patients with Acute Spinal Injuries: An Evidence-Based Analysis"
by A. Ploumis, MD, PhD, et al.

Commentary & Perspective by
Ronald W. Lindsey, MD*,
University of Texas Medical Branch, Galveston, Texas

Posted November 2009

The medical community has long recognized that patients who sustain acute spinal injuries are at substantial risk for venous thromboembolism, which includes deep venous thrombosis and pulmonary embolism. Studies have established not only that the basis for this risk can be attributed to Virchow's triad of stasis, hypercoagulability, and intimal injury1 but also that variations in some hemostatic factors and fibrinolytic systems have been specifically implicated as potential sources for venous thromboembolism susceptibility following spinal cord injury. Although the substantial morbidity and mortality associated with venous thromboembolism in patients with spinal trauma have been well established, and although a consensus exists among spine surgeons that prophylaxis is essential, the optimal selection, initiation, and duration of mechanical and/or pharmacological prophylaxis remain elusive. In fact, evidence-based practice guidelines for prophylaxis against venous thromboembolism after acute spinal injuries (with or without injury to the spinal cord) are nonexistent.

Spinal trauma, as compared with other elective and nonelective orthopaedic conditions that warrant the use of early preventive measures against venous thromboembolism, is associated with a number of special considerations that pose a challenge to the development of meaningful treatment standards and recommendations. The level and/or type of spinal injury, the nature and extent of concomitant spinal cord injury, the associated other organ injuries, the optimal time to initiate therapy and the duration of that therapy, and the influence of surgery (anticipated or recent) are just a few of the factors that complicate the decision-making with regard to thromboembolic prophylaxis in these patients. Moreover, establishing how best to select and apply the various types of currently approved mechanical and pharmacological modalities to any protective regimen further compounds the issue of prophylaxis against venous thromboembolism. Finally, aside from effectively decreasing the prevalence of deep venous thrombosis and pulmonary embolism, the current inability to accurately determine which prophylaxis-related complications are related to traumatic injury to the spine and which are related to acute injury to the spinal cord suggests that the development of evidence-based recommendations for prophylaxis against venous thromboembolism will be extremely challenging.

Ploumis et al. have performed a meta-analysis of the electronically available literature on thromboprophylaxis in patients with acute spinal injuries (with and without spinal cord injury) to determine the optimal preventive regimen. Clearly, apart from a large, randomized, well-controlled, multicenter prospective study, this paper may provide the most credible guidelines to date on this issue. Their methodology included querying all major databases; their study selection process required not only the utilization of preventive measures but also the use of an objective deep venous thrombosis or pulmonary embolism diagnostic test as well as randomization and control for heparin-based pharmaceutical agents. All papers were graded in accordance with their level of evidence with regard to the primary research question, and only studies that met at least 50% of the validity criteria were included in the analysis of unfractionated heparin as compared with low-molecular-weight heparin. The data were grouped according to the following categories: (a) control (no prophylaxis), (b) all mechanical methods, (c) vitamin K antagonists, (d) non-vitamin K antagonists, (e) mechanical plus pharmacological prophylaxis, (f) early prophylaxis (initiated within two weeks after spinal cord injury), and (g) late prophylaxis (initiated more than two weeks after spinal cord injury). All patients receiving unfractionated heparin were grouped together as "heparin"; all patients receiving low-molecular-weight heparin were grouped as such. Finally, similar data were pooled, and the pooled data were analyzed. Although 489 articles were identified, only twenty-one studies met the inclusion criteria.

The level of evidence of the studies varied greatly; only a few constituted randomized trials. In addition, most of the literature pertained to patients with spinal cord injuries. However, the authors were still able to provide meaningful insight into some questions while highlighting the issues that continue to elude us. The risk for deep venous thrombosis is higher in patients with spinal cord injury; mechanical measures may be as effective without supplemental pharmacological agents; vitamin K antagonists appear to be more effective (than non-vitamin K antagonists) for pulmonary embolism; and, in patients with acute spinal cord injury, low-molecular-weight heparin (as opposed to unfractionated heparin) is more effective prophylactically for deep venous thrombosis, while similar in effect with regard to pulmonary embolism. Furthermore, low-molecular-weight heparin (compared with unfractionated heparin) is associated with less bleeding. However, with regard to the patient with acute spinal cord injury, this paper (and the literature) does not recommend when to initiate pharmacological prophylaxis, nor does it recommend the duration of treatment. Ultimately, the study recommends that clinicians approach these decisions on the basis of clinical experience, expertise, and patient-specific risk factors.

The authors should be applauded for their comprehensive and methodologically sound review of the subject. They have, undoubtedly, strengthened and refined our existing knowledge regarding prophylaxis against venous thromboembolism in patients with spine injuries. However, they have also demonstrated that the existing literature is too heterogeneous and the clinical issues are too complex to resolve without additional data from future prospective, randomized, controlled clinical trials dedicated to answering only one question at a time.

*The author did not receive any outside funding or grants in support of his research for or preparation of this work. Neither he nor a member of his immediate family received payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity.


1. Virchow RLK. Thrombose und Embolie. Gefässentzündung und septische Infektion. Gesammelte Abhandlungen zur wissenschaftlichen Medicin. Frankfurt am Main: Von Meidinger und Sohn; 1856. p 219-732. Translation in: Matzdorff AC, Bell WR. Thrombosis and embolie. Canton, Massachusetts: Science History Publications; 1998. p 1846-56.