Charles Fisher and Juliet Batke
John Street, Brian Lenehan and Charles Fisher
Criteria for methodological quality have been widely accepted in many fields of surgical practice. These criteria include those of Velanovich and Gill and Feinstein. No such analysis of the spine surgery literature has ever been reported. This study is a systematic review of the quality of life (QOL) publications to determine if the recent interest in QOL measurements following spinal surgery has been accompanied by an improvement in the quality of the papers published.
The archives of the journals Journal of Neurosurgery: Spine, Spine, Journal of Spinal Disorders & Techniques, European Spine Journal, and The Spine Journal, for the years 2000–2004 inclusive, were examined, and all publications reporting QOL outcomes were analyzed. Each paper was scored according to the criteria of Velanovich and Gill and Feinstein, and the methodological quality of these manuscripts—and any time-dependent changes—were determined.
During the study period, the total number of articles published increased by 36%, while the number of QOL articles increased by 102%. According to the criteria of Velanovich, there was a statistically significant improvement in the quality of the publications over the study period (p = 0.0394). In 2000, only 27% of outcome measures were disease specific, 77% were valid, and 77% were appropriate for the study design. In 2004, 43% were disease specific, 88% were valid, and 89% were appropriate. In 2000, 53% of studies used appropriate statistical analysis compared with 100 and 96% for 2003 and 2004, respectively. There was no demonstrable improvement in the fulfillment of the more rigorous Gill and Feinstein criteria for any of the 5 journals over the period of the study.
The authors' study illustrates a moderate improvement in the quality of these publications over the study period but much methodological improvement is required.
Stephen E. Natelson
Despite extensive published research on thoracolumbar burst fractures, controversy still surrounds which is the most appropriate treatment. The objective of this study was to evaluate the scientific literature on operative and nonoperative treatment of patients with thoracolumbar burst fractures and no neurological deficit.
In their search of the literature, the authors identified all possible relevant studies concerning thoracolumbar burst fracture without neurological deficit. Two independent observers performed study selection, methodological quality assessment, and data extraction in a blinded and objective manner for all papers identified during the search. In a synthesis of the literature, the authors obtained evidence for both operative and nonoperative treatments.
There is a lack of evidence demonstrating the superiority of one approach over the other as measured using generic and disease-specific health-related quality of life scales. There is no scientific evidence linking posttraumatic kyphosis to clinical outcomes. The authors found that there is a strong need for improved clinical research methodology to be applied to this patient population.
Marcel F. Dvorak, Michael G. Johnson, Michael Boyd, Garth Johnson, Brian K. Kwon and Charles G. Fisher
Object. The primary goal of this study was to describe the long-term health-related quality of life (HRQOL) outcomes in patients who have suffered Jefferson-type fractures. These outcomes were compared with matched normative HRQOL data and with the patient's perceptions of their HRQOL prior to the injury. Variables that potentially influence these HRQOL outcomes were analyzed.
No standardized outcome assessments have been published for patients who suffer these fractures; their outcomes have long thought to be excellent following treatment. Determining the optimal surrogate measure to represent preinjury HRQOL in trauma patients is difficult.
Methods. A retrospective review, radiographic analysis, and cross-sectional outcome assessment were performed. The Short Form (SF)—36 and the American Academy of Orthopaedic Surgeons/North American Spine Society (AAOS/NASS) outcome instruments were filled out by patients at final follow-up examination (follow-up period 75 months, range 19–198 months) to represent their current status as well as their perceptions of preinjury status.
In 34 patients, the SF-36 physical component score and the AAOS/NASS pain values were significantly lower than normative values. There was no significant difference between normative and preinjury values. Spence criteria greater than 7 mm and the presence of associated injuries predicted poorer outcome scores during the follow-up period.
Conclusions. Long-term follow-up examination of patients with Jefferson fractures indicated that patients' status does not return to the level of their perceived preinjury health status or that of normative population controls. Those with other injuries and significant osseous displacement (≥ 7 mm total) may experience poorer long-term outcomes. Limitations of the study included a relatively low (60%) response rate and the difficulties of identifying an appropriate baseline outcome in a trauma population with which the follow-up outcomes can be compared.
Measuring quality of life outcomes in spine clinical trials
Michael G. Fehlings and Anoushka Singh
Tobias Pitzen, Juay Seng Tan, Marcel F. Dvorak, Charles Fisher and Thomas Oxland
To avoid the cost of bone graft substitutes and the morbidity of iliac crest bone graft retrieval, locally harvested vertebral body bone has been used to fill interbody cages. When marginal hypertrophic osteophytes are used, there is little impact on the adjacent vertebrae, but when cancellous bone is removed from the central part of the vertebral body, it is not clear how significantly this procedure weakens the vertebra. The objective of this study was to investigate the immediate mechanical response of the cervical spine after removing bone from the central vertebral body.
Fourteen cervical functional spinal units (FSUs) (mean age 73.3 years, range 63–90 years) were used. For each FSU, bone mineral density (BMD) was determined using lateral-view dual-energy x-ray absorptiometry studies. The FSUs were assigned to 1 of 2 groups (test group or control group) with an equal distribution of BMD. All specimens received a cage placed into the cleaned disc space. The specimens from the test group had a 5-mm-diameter bone plug removed from the vertebral bodies superior and inferior to the cage-fitted disc. The specimens were loaded in flexion-compression until failure via an eccentric compressive force at 0.25 mm/second.
The yield compression strength was 1149 ± 523 N for the test group and 1647 ± 962 N for the control group (p = 0.25). The ultimate compression strength was 1699 ± 498 N for the test group and 2450 N ± 835 N for the control group (p = 0.06). Force at 4 mm displacement was 1064 N for the test group and 1574 N for the control group (p = 0.15). Displacement at yield compression strength was 4.4 mm for the test group and 4.2 mm for the control group (p = 0.78). There was no significant intergroup difference for any of the studied parameters.
There does not appear to be a significant early biomechanical weakening of adjacent vertebrae caused by aforementioned technique of local bone harvest.
Andrea M. Simmonds, Y. Raja Rampersaud, Marcel F. Dvorak, Nicolas Dea, Angela D. Melnyk and Charles G. Fisher
A range of surgical options exists for the treatment of degenerative lumbar spondylolisthesis (DLS). The chosen technique inherently depends on the stability of the DLS. Despite a substantial body of literature dedicated to the outcome analysis of numerous DLS procedures, no consensus has been reached on defining or classifying the disorder with respect to stability or the role that instability should play in a treatment algorithm. The purpose of this study was to define grades of stability and to develop a guide for deciding on the optimal approach in surgically managing patients with DLS.
The authors conducted a qualitative systematic review of clinical or biomechanical analyses evaluating the stability of and surgical outcomes for DLS for the period from 1990 to 2013. Research focused on nondegenerative forms of spondylolisthesis or spinal stenosis without associated DLS was excluded. The primary extracted results were clinical and radiographic parameters indicative of DLS instability.
The following preoperative parameters are predictors of stability in DLS: restabilization signs (disc height loss, osteophyte formation, vertebral endplate sclerosis, and ligament ossification), no disc angle change or less than 3 mm of translation on dynamic radiographs, and the absence of low-back pain. The validity and magnitude of each parameter’s contribution can only be determined through appropriately powered prospective evaluation in the future. Identifying these parameters has allowed for the creation of a preliminary DLS instability classification (DSIC) scheme based on the preoperative assessment of DLS stability.
Spinal stability is an important factor to consider in the evaluation and treatment of patients with DLS. Qualitative assessment of the best available evidence revealed clinical and radiographic parameters for the creation of the DSIC, a decision aid to help surgeons develop a method of preoperative evaluation to better stratify DLS treatment options.
Rowan Schouten, Peter Lewkonia, Vanessa K. Noonan, Marcel F. Dvorak and Charles G. Fisher
The aim of this study was to define the expected functional and health-related quality of life outcomes following common thoracolumbar injuries on the basis of consensus expert opinion and the best available literature. Patient expectations are primarily determined by the information provided by health care professionals, and these expectations have been shown to influence outcome in various medical and surgical conditions. This paper presents Part 2 of a multiphase study designed to investigate the impact of patient expectations on outcomes following spinal injury. Part 1 demonstrated substantial variability in the information surgeons are communicating to patients. Defining the expected outcomes following thoracolumbar injury would allow further analysis of this relationship and enable surgeons to more accurately and consistently inform patients.
Expert opinion was assembled by distributing questionnaires comprising 4 cases representative of common thoracolumbar injuries to members of the Spine Trauma Study Group (STSG). The 4 cases included a thoracolumbar junction burst fracture treated nonoperatively or with posterior transpedicular instrumentation, a low lumbar (L-4) burst fracture treated nonoperatively, and a thoracolumbar junction flexion-distraction injury managed with posterior fusion. For each case, 5 questions about expected outcomes were posed. The questions related to the proportion of patients who are pain free, the proportion who have regained full range of motion, and the patients' recreational activity restrictions and personal care and social life limitations, all at 1 year following injury, as well as the timing of return to work and length of hospital stay. Responses were analyzed and combined with the results of a systematic literature review on the same injuries to define the expected outcomes.
The literature review identified 38 appropriate studies that met the preset inclusion criteria. Published data were available for all injuries, but not all outcomes were available for each type of injury. The survey was completed by 31 (57%) of 53 surgeons representing 24 trauma centers across North America (15), Europe (5), India (1), Mexico (1), Japan (1) and Israel (1). Consensus expert opinion supplemented the available literature and was used exclusively when published data were lacking.
For example, 1 year following cast or brace treatment of a thoracolumbar burst fracture, the expected outcomes include a 40% chance of being pain free, a 70% chance of regaining pre-injury range of motion, and an expected ability to participate in high-impact exercise and contact sport with no or minimal limitation. Consensus expert opinion predicts reemployment within 4–6 months. The length of inpatient stay averages 4–5 days.
This synthesis of the best available literature and consensus opinion of surgeons with extensive clinical experience in spine trauma reflects the optimal methodology for determining functional prognosis after thoracolumbar trauma. By providing consistent, accurate information surgeons will help patients develop realistic expectations and potentially optimize outcomes.
Y. Raja Rampersaud, Paul A. Anderson, John R. Dimar II, Charles G. Fisher and on behalf of the Spine Trauma Study Group and Degenerative Spine Study Group
Reporting of adverse events (AEs) in spinal surgery uses inconsistent definitions and severity grading, making it difficult to compare results between studies. The Spinal Adverse Events Severity System, version 2 (SAVES-V2) aims to standardize the classification of spine surgery AEs; however, its inter- and intraobserver reliability are unknown. The objective of this study was to assess inter- and intraobserver reliability of the SAVES-V2 grading system for assessing AEs in spinal surgery.
Two multinational, multicenter surgical study groups assessed surgical case vignettes (10 trauma and 12 degenerative cases) for AE occurrence by using SAVES-V2. Thirty-four members of the Spine Trauma Study Group (STSG) and 17 members of the Degenerative Spine Study Group (DSSG) participated in the first round of case vignettes. Six months later, the same case vignettes were randomly reorganized and presented in an otherwise identical manner. Inter- and intraobserver agreement on the presence, severity, number, and type of AE, as well as the impact of the AE on length of stay (LOS) were assessed using intraclass correlation (ICC), Cohen's kappa value, and the percentage of participants in agreement.
Agreement on the presence of AEs ranged from 97% to 100% in the 2 groups. Severity classification showed substantial interobserver (ICC = 0.75 for both groups) and intraobserver (ICC = 0.70 in DSSG, 0.71 in STSG) agreement. Judgments on the number of AEs showed high interobserver agreement and moderate intraobserver agreement in both groups. Both the STSG and DSSG had high intraobserver agreement on the type of AE; interobserver agreement for AE type was high in the STSG and fair in the DSSG. Agreement on impact of the AE on LOS was excellent in the DSSG and fair in the STSG.
There was good agreement on the presence, severity, and number of AEs in both trauma and degenerative cases in using the SAVES-V2. This grading system is a simple, reliable tool for identifying and capturing AEs in spinal surgery.