In this paper the authors review spine trauma and spinal cord injury (SCI) in the geriatric population. The information in this study was compiled through a literature review of clinical presentation and management of SCI in the elderly population. This was done to define, identify, and specify treatment algorithms and management strategies in this unique patient population.
Pascal Jabbour, Michael Fehlings, Alexander R. Vaccaro and James S. Harrop
George M. Ghobrial, David W. Cadotte, Kim Williams Jr., Michael G. Fehlings and James S. Harrop
The use of intrawound vancomycin is rapidly being adopted for the prevention of surgical site infection (SSI) in spinal surgery. At operative closure, the placement of vancomycin powder in the wound bed—in addition to standard infection prophylaxis—can provide high concentrations of antibiotics with minimal systemic absorption. However, despite its popularity, to date the majority of studies on intrawound vancomycin are retrospective, and there are no prior reports highlighting the risks of routine treatment.
A MEDLINE search for pertinent literature was conducted for studies published between 1966 and May 2015 using the following MeSH search terms: “intrawound vancomycin,” “operative lumbar spine complications,” and “nonoperative lumbar spine complications.” This was supplemented with references and known literature on the topic.
An advanced MEDLINE search conducted on May 6, 2015, using the search string “intrawound vancomycin” found 22 results. After a review of all abstracts for relevance to intrawound vancomycin use in spinal surgery, 10 studies were reviewed in detail. Three meta-analyses were evaluated from the initial search, and 2 clinical studies were identified. After an analysis of all of the identified manuscripts, 3 additional studies were included for a total of 16 studies. Fourteen retrospective studies and 2 prospective studies were identified, resulting in a total of 9721 patients. A total of 6701 (68.9%) patients underwent treatment with intrawound vancomycin. The mean SSI rate among the control and vancomycin-treated patients was 7.47% and 1.36%, respectively. There were a total of 23 adverse events: nephropathy (1 patient), ototoxicity resulting in transient hearing loss (2 patients), systemic absorption resulting in supratherapeutic vancomycin exposure (1 patient), and culture-negative seroma formation (19 patients). The overall adverse event rate for the total number of treated patients was 0.3%.
Intrawound vancomycin use appears to be safe and effective for reducing postoperative SSIs with a low rate of morbidity. Study disparities and limitations in size, patient populations, designs, and outcomes measures contribute significant bias that could not be fully rectified by this systematic review. Moreover, care should be exercised in the use of intrawound vancomycin due to the lack of well-designed, prospective studies that evaluate the efficacy of vancomycin and include the appropriate systems to capture drug-related complications.
George M. Ghobrial, Thana Theofanis, Bruce V. Darden, Paul Arnold, Michael G. Fehlings and James S. Harrop
Unintended durotomy is a common occurrence during lumbar spinal surgery, particularly in surgery for degenerative spinal conditions, with the reported incidence rate ranging from 0.3% to 35%. The authors performed a systematic literature review on unintended lumbar spine durotomy, specifically aiming to identify the incidence of durotomy during spinal surgery for lumbar degenerative conditions. In addition, the authors analyzed the incidence of durotomy when minimally invasive surgical approaches were used as compared with that following a traditional midline open approach.
A MEDLINE search using the term “lumbar durotomy” (under the 2015 medical subject heading [MeSH] “cerebrospinal fluid leak”) was conducted on May 13, 2015, for English-language medical literature published in the period from January 1, 2005, to May 13, 2015. The resulting papers were categorized into 3 groups: 1) those that evaluated unintended durotomy rates during open-approach lumbar spinal surgery, 2) those that evaluated unintended durotomy rates during minimally invasive spine surgery (MISS), and 3) those that evaluated durotomy rates in comparable cohorts undergoing MISS versus open-approach lumbar procedures for similar lumbar pathology.
The MEDLINE search yielded 116 results. A review of titles produced 22 potentially relevant studies that described open surgical procedures. After a thorough review of individual papers, 19 studies (comprising 15,965 patients) pertaining to durotomy rates during open-approach lumbar surgery were included for analysis. Using the Oxford Centre for Evidence-Based Medicine (CEBM) ranking criteria, there were 7 Level 3 prospective studies and 12 Level 4 retrospective studies. In addition, the authors also included 6 studies (with a total of 1334 patients) that detailed rates of durotomy during minimally invasive surgery for lumbar degenerative disease. In the MISS analysis, there were 2 prospective and 4 retrospective studies. Finally, the authors included 5 studies (with a total of 1364 patients) that directly compared durotomy rates during open-approach versus minimally invasive procedures. Studies of open-approach surgery for lumbar degenerative disease reported a total of 1031 durotomies across all procedures, for an overall durotomy rate of 8.11% (range 2%–20%). Prospectively designed studies reported a higher rate of durotomy than retrospective studies (9.57% vs 4.32%, p = 0.05). Selected MISS studies reported a total of 93 durotomies for a combined durotomy rate of 6.78%. In studies of matched cohorts comparing open-approach surgery with MISS, the durotomy rates were 7.20% (34 durotomies) and 7.02% (68), respectively, which were not significantly different.
Spinal surgery for lumbar degenerative disease carries a significant rate of unintended durotomy, regardless of the surgical approach selected by the surgeon. Interpretation of unintended durotomy rates for lumbar surgery is limited by a lack of prospective and cohort-matched controlled studies.
James Guest, James S. Harrop, Bizhan Aarabi, Robert G. Grossman, James W. Fawcett, Michael G. Fehlings and Charles H. Tator
The North American Clinical Trials Network (NACTN) includes 9 clinical centers funded by the US Department of Defense and the Christopher Reeve Paralysis Foundation. Its purpose is to accelerate clinical testing of promising therapeutics in spinal cord injury (SCI) through the development of a robust interactive infrastructure. This structure includes key committees that serve to provide longitudinal guidance to the Network. These committees include the Executive, Data Management, and Neurological Outcome Assessments Committees, and the Therapeutic Selection Committee (TSC), which is the subject of this manuscript. The NACTN brings unique elements to the SCI field. The Network's stability is not restricted to a single clinical trial. Network members have diverse expertise and include experts in clinical care, clinical trial design and methodology, pharmacology, preclinical and clinical research, and advanced rehabilitation techniques. Frequent systematic communication is assigned a high value, as is democratic process, fairness and efficiency of decision making, and resource allocation. This article focuses on how decision making occurs within the TSC to rank alternative therapeutics according to 2 main variables: quality of the preclinical data set, and fit with the Network's aims and capabilities. This selection process is important because if the Network's resources are committed to a therapeutic, alternatives cannot be pursued. A proposed methodology includes a multicriteria decision analysis that uses a Multi-Attribute Global Inference of Quality matrix to quantify the process. To rank therapeutics, the TSC uses a series of consensus steps designed to reduce individual and group bias and limit subjectivity. Given the difficulties encountered by industry in completing clinical trials in SCI, stable collaborative not-for-profit consortia, such as the NACTN, may be essential to clinical progress in SCI. The evolution of the NACTN also offers substantial opportunity to refine decision making and group dynamics. Making the best possible decisions concerning therapeutics selection for trial testing is a cornerstone of the Network's function.
George M. Ghobrial, Christopher M. Maulucci, Mitchell Maltenfort, Richard T. Dalyai, Alexander R. Vaccaro, Michael G. Fehlings, John Street, Paul M. Arnold and James S. Harrop
Thoracolumbar spine injuries are commonly encountered in patients with trauma, accounting for almost 90% of all spinal fractures. Thoracolumbar burst fractures comprise a high percentage of these traumatic fractures (45%), and approximately half of the patients with this injury pattern are neurologically intact. However, a debate over complication rates associated with operative versus nonoperative management of various thoracolumbar fracture morphologies is ongoing, particularly concerning those patients presenting without a neurological deficit.
A MEDLINE search for pertinent literature published between 1966 and December 2013 was conducted by 2 authors (G.G. and R.D.), who used 2 broad search terms to maximize the initial pool of manuscripts for screening. These terms were “operative lumbar spine adverse events” and “nonoperative lumbar spine adverse events.”
In an advanced MEDLINE search of the term “operative lumbar spine adverse events” on January 8, 2014, 1459 results were obtained. In a search of “nonoperative lumbar spine adverse events,” 150 results were obtained. After a review of all abstracts for relevance to traumatic thoracolumbar spinal injuries, 62 abstracts were reviewed for the “operative” group and 21 abstracts were reviewed for the “nonoperative” group. A total of 14 manuscripts that met inclusion criteria for the operative group and 5 manuscripts that met criteria for the nonoperative group were included.
There were a total of 919 and 436 patients in the operative and nonoperative treatment groups, respectively. There were no statistically significant differences between the groups with respect to age, sex, and length of stay. The mean ages were 43.17 years in the operative and 34.68 years in the nonoperative groups. The majority of patients in both groups were Frankel Grade E (342 and 319 in operative and nonoperative groups, respectively). Among the studies that reported the data, the mean length of stay was 14 days in the operative group and 20.75 in the nonoperative group.
The incidence of all complications in the operative and nonoperative groups was 300 (32.6%) and 21 (4.8%), respectively (p = 0.1065). There was no significant difference between the 2 groups with respect to the incidence of pulmonary, thromboembolic, cardiac, and gastrointestinal complications. However, the incidence of infections (pneumonia, urinary tract infection, wound infection, and sepsis) was significantly higher in the operative group (p = 0.000875). The incidence of instrumentation failure and need for revision surgery was 4.35% (40 of 919), a significant morbidity, and an event unique to the operative category (p = 0.00396).
Due to the limited number of high-quality studies, conclusions related to complication rates of operative and nonoperative management of thoracolumbar traumatic injuries cannot be definitively made. Further prospective, randomized studies of operative versus nonoperative management of thoracolumbar and lumbar spine trauma, with standardized definitions of complications and matched patient cohorts, will aid in properly defining the risk-benefit ratio of surgery for thoracolumbar spine fractures.
Charles H. Tator, Robin Hashimoto, Annie Raich, Daniel Norvell, Michael G. Fehlings, James S. Harrop, James Guest, Bizhan Aarabi and Robert G. Grossman
There is a need to enhance the pipeline of discovery and evaluation of neuroprotective pharmacological agents for patients with spinal cord injury (SCI). Although much effort and money has been expended on discovering effective agents for acute and subacute SCI, no agents that produce major benefit have been proven to date. The deficiencies of all aspects of the pipeline, including the basic science input and the clinical testing output, require examination to determine remedial strategies. Where has the neuroprotective/pharmacotherapy preclinical process failed and what needs to be done to achieve success? These are the questions raised in the present review, which has 2 objectives: 1) identification of articles that address issues related to the translational readiness of preclinical SCI pharmacological therapies; and 2) examination of the preclinical studies of 5 selected agents evaluated in animal models of SCI (including blunt force trauma, penetrating trauma, or ischemia). The 5 agents were riluzole, glyburide, magnesium sulfate, nimodipine, and minocycline, and these were selected because of their promise of translational readiness as determined by the North American Clinical Trials Network Consortium.
The authors found that there are major deficiencies in the effort that has been extended to coordinate and conduct preclinical neuroprotection/pharmacotherapy trials in the SCI field. Apart from a few notable exceptions such as the NIH effort to replicate promising strategies, this field has been poorly coordinated. Only a small number of articles have even attempted an overall evaluation of the neuroprotective/pharmacotherapy agents used in preclinical SCI trials. There is no consensus about how to select the agents for translation to humans on the basis of their preclinical performance and according to agreed-upon preclinical performance criteria.
In the absence of such a system and to select the next agent for translation, the Consortium has developed a Treatment Strategy Selection Committee, and this committee selected the most promising 5 agents for potential translation. The results show that the preclinical work on these 5 agents has left numerous gaps in knowledge about their preclinical performance and confirm the need for significant changes in preclinical neuroprotection/pharmacotherapy trials in SCI. A recommendation is made for the development and validation of a preclinical scoring system involving worldwide experts in preclinical and clinical SCI.
James S. Harrop, Robin Hashimoto, Dan Norvell, Annie Raich, Bizhan Aarabi, Robert G. Grossman, James D. Guest, Charles H. Tator, Jens Chapman and Michael G. Fehlings
Using a systematic approach, the authors evaluated the current utilization, safety, and effectiveness of cellular therapies for traumatic spinal cord injuries (SCIs) in humans.
A systematic search and critical review of the literature published through mid-January 2012 was performed. Articles included in the search were restricted to the English language, studies with at least 10 patients, and those analyzing cellular therapies for traumatic SCI. Citations were evaluated for relevance using a priori criteria, and those that met the inclusion criteria were critically reviewed. Each article was then designated a level of evidence that was developed by the Oxford Centre for Evidence-Based Medicine.
The initial literature search identified 651 relevant articles, which decreased to 350 after excluding case reports and reviews. Evaluation of articles at the title/abstract level, and later at the full-text level, limited the final article set to 12 papers. The following cellular therapies employed in humans with SCI are reviewed: bone marrow mesenchymal and hematopoietic stem cells (8 studies), olfactory ensheathing cells (2 studies), Schwann cells (1 study), and fetal neurogenic tissue (1 study). Overall the quality of the literature was very low, with 3 Grade III levels of evidence and 9 Grade IV studies.
Several different cellular-mediated strategies for adult SCI have been reported to be relatively safe with varying degrees of neurological recovery. However, the literature is of low quality and there is a need for improved preclinical studies and prospective, controlled clinical trials.
Jeffrey D. Coe, Alexander R. Vaccaro, Andrew T. Dailey, Rick C. Sasso, Steven C. Ludwig, James S. Harrop, Joseph R. Dettori, Christopher I. Shaffrey, Sanford E. Emery and Michael G. Fehlings
Diana S. L. Chow, Yang Teng, Elizabeth G. Toups, Bizhan Aarabi, James S. Harrop, Christopher I. Shaffrey, Michele M. Johnson, Maxwell Boakye, Ralph F. Frankowski, Michael G. Fehlings and Robert G. Grossman
The aim of this paper was to characterize individual and population pharmacokinetics of enterally administered riluzole in a Phase 1 clinical trial of riluzole as a neuroprotective agent in adults 18–70 years old with acute spinal cord injury (SCI).
Thirty-five individuals with acute SCI, American Spinal Injury Association Impairment Scale Grades A–C, neurological levels from C-4 to T-12, who were enrolled in the Phase 1 clinical trial sponsored by the North American Clinical Trials Network for Treatment of Spinal Cord Injury, received 50 mg riluzole twice daily for 28 doses. The first dose was administered at a mean of 8.7 ± 2.2 hours postinjury. Trough plasma samples were collected within 1 hour predose, and peak plasma samples were collected 2 hours postdose on Days 3 and 14 of treatment. Riluzole concentrations were quantified by high-performance liquid chromatography assay. The data were analyzed for individual and population pharmacokinetics using basic structural and covariate models. The pharmacokinetic measures studied were the peak concentration (Cmax), trough concentration (Cmin), systemic exposure (AUC0–12), clearance (CL/F), and volume of distribution (V_F) normalized by the bioavailability (F).
The Cmax and AUC0–12 achieved in SCI patients were lower than those in ALS patients on the same dose basis, due to a higher CL and larger V. The pharmacokinetics of riluzole (Cmax, Cmin, AUC0–12, CL, and V) changed during the acute and subacute phases of SCI during the 14 days of therapy. It was consistently observed in patients at all clinical sites that Cmax, Cmin, and AUC0–12 (128.9 ng/ml, 45.6 ng/ml, and 982.0 ng × hr/ml, respectively) were significantly higher on Day 3 than on Day 14 (76.5 ng/ml, 19.1 ng/ml, and 521.0 ng × hr/ml, respectively). These changes resulted from lower CL (49.5 vs 106.2 L/hour) and smaller V (557.1 vs 1297.9/L) on Day 3. No fluid imbalance or cytochrome P 1A2 induction due to concomitant medications was identified during the treatment course to account for such increases in V and CL, respectively. Possible mechanisms underlying these changes are discussed.
This is the first report of clinical pharmacokinetics of riluzole in patients with SCI. The Cmax and AUC0–12 achieved in SCI patients were lower than those in ALS patients on the same dose basis, due to a higher clearance and larger volume of distribution in SCI patients. The finding in SCI patients of an increase in the clearance and distribution of riluzole between the 3rd and 14th days after SCI, with a lower plasma concentration of riluzole on the 14th day, stresses the importance of monitoring changes in drug metabolism after SCI in interpreting the safety and efficacy of therapeutic drugs that are used in clinical trials in SCI. Clinical trial registration no.: NCT00876889.
Robert G. Grossman, Ralph F. Frankowski, Keith D. Burau, Elizabeth G. Toups, John W. Crommett, Michele M. Johnson, Michael G. Fehlings, Charles H. Tator, Christopher I. Shaffrey, Susan J. Harkema, Jonathan E. Hodes, Bizhan Aarabi, Michael K. Rosner, James D. Guest and James S. Harrop
The aim of this multicenter, prospective study was to determine the spectrum, incidence, and severity of complications during the initial hospitalization of patients with spinal cord injury.
The study was conducted at 9 university-affiliated hospitals that comprise the clinical centers of the North American Clinical Trials Network (NACTN) for Treatment of Spinal Cord Injury. The study population comprised 315 patients admitted to NACTN clinical centers between June 25, 2005, and November 2, 2010, who had American Spinal Injury Association (ASIA) Impairment Scale grades of A–D and were 18 years of age or older. Patients were managed according to a standardized protocol.
The study population was 79% male with a median age of 44 years. The leading causes of injury were falls (37%) and motor vehicle accidents (28%). The distribution of initial ASIA grades were A (40%), B (16%), C (15%), and D (29%). Fifty-eight percent of patients sustained 1 or more severe, moderate, or mild complications. Complications were associated with more severe ASIA grade: 84% of patients with Grade A and 25% of patients with Grade D had at least 1 complication. Seventy-eight percent of complications occurred within 14 days of injury. The most frequent types of severe and moderate complications were respiratory failure, pneumonia, pleural effusion, anemia, cardiac dysrhythmia, and severe bradycardia. The mortality rate was 3.5% and was associated with increased age and preexisting morbidity.
Knowledge of the type, frequency, time of occurrence, and severity of specific complications that occur after spinal cord injury can aid in their early detection, treatment, and prevention. The data are of importance in evaluating and selecting therapy for clinical trials.