Congenital scoliosis due to a hemivertebra requires surgical stabilization prior to skeletal maturity if rapidly progressive curve growth occurs. Here the authors present the unique case of a man who, at the age of 12 years, had undergone Harrington rod placement for stabilization of progressive congenital scoliosis due to a T-11 hemivertebra and then, at the age of 53 years, presented with acutely progressive myelopathy due to spinal cord compression from an arachnoid web at T-11 despite a solid fusion mass at the prior surgical site. The patient underwent a posterior midline approach for resection of the T-11 pedicle at the level of the hemivertebra, intradural spinal cord detethering with resection of the arachnoid web at T-11, and T2–L2 instrumented fusion with deformity correction, leading to subsequent resolution of his acute myelopathic symptoms. In conclusion, arachnoid web formation superimposed on preexisting tension on the thoracic spinal cord from congenital scoliosis due to a T-11 hemivertebra caused acute myelopathy in an adult with a previously solid fusion mass from childhood. The resolution of acute myelopathy and halting of further progression occurred with pedicle resection, arachnoid web fenestration, and spinal deformity correction.
Molly E. Hubbard, Matthew A. Hunt, Kristen E. Jones and David W. Polly
David W. Polly Jr.
Timothy R. Kuklo, Michael K. Rosner and David W. Polly Jr.
Synthetic bioabsorbable implants have recently been introduced in spinal surgery; consequently, the indications, applications, and results are still evolving. The authors used absorbable interbody spacers (Medtronic Sofamor Danek, Memphis, TN) packed with recombinant bone morphogenetic protein (Infuse; Medtronic Sofamor Danek) for single- and multiple-level transforaminal lumbar interbody fusion (TLIF) procedures over a period of 18 months. This is a consecutive case series in which postoperative computerized tomography (CT) scanning was used to assess fusion status.
There were 22 patients (17 men, five women; 39 fusion levels) whose mean age was 41.6 years (range 23–70 years) and in whom the mean follow-up duration was 12.4 months (range 6–18 months). Bridging bone was noted as early as the 3-month postoperative CT scan when obtained; solid arthrodesis was routinely noted between 6 and 12 months in 38 (97.4%) of 39 fusion levels. In patients who underwent repeated CT scanning, the fusion mass appeared to increase with time, whereas the disc space height remained stable. Although the results are early (mean 12-month follow-up duration), there was only one noted asymptomatic delayed union/nonunion at L5–S1 in a two-level TLIF with associated screw breakage. There were no infections or complications related to the cages.
The bioabsorbable cages appear to be a viable alternative to metal interbody spacers, and may be ideally suited to spinal interbody applications because of their progressive load-bearing properties.
Report of two cases
David W. Polly Jr., Michael K. Rosner, William Monacci and Ross R. Moquin
Hemivertebrae are a common cause of congenital scoliosis. Depending on their location and the magnitude of the resultant deformity, they may be asymptomatic or require treatment. In the past, treatment has focused on prevention of deformity progression in growing children. Little has been written about congenital scoliosis presenting in adulthood. Because the aging of the spine is a kyphosing process and hemivertebrae often present with a local segmental kyphotic alignment, this can become symptomatic. Excision of hemivertebrae is well established as a safe and effective procedure when treatment is required. Initially this was conducted via a combined anterior–posterior approach. Recently some authors have indicated that in the lumbar spine hemivertebra resection can safely and effectively be achieved via a single posterior transpedicular approach. The authors report two adult cases in which they performed posterior transpedicular lateral extracavitary excision of a thoracic, fully segmented hemivertebrae. Essentially complete correction of the deformity was achieved. There were no neurological complications. The patients were spared a thoracotomy and no chest tubes were required.
Andrew M. Gardeck, Xuan Pu, Qiuyu Yang, David W. Polly and Kristen E. Jones
Residency work-hour restrictions necessitate efficient, reproducible training. Simulation training for spinal instrumentation placement shows significant benefit to learners’ subjective and objective proficiency. Cadaveric laboratories are most effective but have high cost and low availability. The authors’ goal was to create a low-cost, efficient, reproducible spinal instrumentation placement simulation curriculum for neurosurgery and orthopedic surgery residents using synthetic models and 3D computer-assisted navigation, assessing subjective and objective proficiency with placement of thoracolumbar pedicle screws.
Fifteen neurosurgery and orthopedic surgery residents participated in a standardized curriculum with lecture followed by two separate sessions of thoracolumbar pedicle screw placement in a synthetic spine model utilizing 3D computer-assisted navigation. Data were collected on premodule experience, time and accuracy of screw placement, and both subjective and objective ratings of proficiency.
Fifteen of 15 residents demonstrated improvement in subjective (Physician Performance Diagnostic Inventory Scale [PPDIS]) and 14 in objective (Objective Structured Assessment of Technical Skills [OSATS]) measures of proficiency in navigated screw placement with utilization of this curriculum (p < 0.001 for both), regardless of the number of cases of previous experience using thoracolumbar spinal instrumentation. Fourteen of 15 residents demonstrated decreased time per screw placement from session 1 to session 2 (p = 0.006). There was no significant difference in pedicle screw accuracy between session 1 and session 2.
A standardized curriculum using synthetic simulation training for navigated thoracolumbar pedicle screw placement results in significantly improved resident subjective and objective proficiency. Development of a nationwide competency curriculum using simulation training for spinal instrumentation placement should be considered for safe, efficient resident training.
Doniel Drazin, Terrence T. Kim, David W. Polly Jr and J. Patrick Johnson
Image-guided surgery (IGS) has been evolving since the early 1990s and is now used on a daily basis in the operating theater for spine surgery at many institutions. In the last 5 years, spinal IGS has greatly benefitted from important enhancements including portable intraoperative CT (iCT) coupled with high-speed computerized stereotactic navigation systems and optical-based camera tracking technology.
Sharon C. Yson, Edward Rainier G. Santos, Jonathan N. Sembrano and David W. Polly Jr.
In this paper the authors sought to determine the segmental lumbar sagittal contour change after bilateral transforaminal lumbar interbody fusion (TLIF).
Between March 2007 and October 2010, 42 consecutive patients (57 levels) underwent bilateral TLIF. Standard preoperative and 6-week postoperative standing lumbar spine radiographs were examined. Preoperative and postoperative segmental lordosis was determined by manual measurements using the Cobb method. The difference between the preoperative and postoperative values were calculated and analyzed for statistical significance.
The mean preoperative segmental alignment was 8.1°. The mean postoperative alignment was 15.3°, with a mean correction of 7.2° per segment. The largest gain in lordosis was obtained at the L5–S1 level (10.1°). There was a significant difference between the preoperative and postoperative values (p = 5 × 10−9). There was no significant difference in mean segmental correction between levels. Improvement in lordosis was higher in multilevel fusions (9.8°) than in single-level fusions (5.2°) (p = 0.047). There was an inverse correlation between preoperative sagittal lordosis measurement and change in lordosis (r = −0.599).
A significant improvement in lumbar lordosis can be gained by preforming bilateral facetectomies in TLIF with posterior compression. This procedure provides an additional option to a spine surgeon's armamentarium in dealing with significant lumbar sagittal plane deformities.
Michael K. Rosner, David W. Polly Jr., Timothy R. Kuklo and Stephen L. Ondra
Techniques to improve segmental fixation have advanced the ability to correct complex spinal deformity. The purpose of instrumentation is to correct spinal deformity or to stabilize the spine to enhance the long-term biological fusion. The ultimate goal of spinal deformity surgery is the creation of a stable, balanced, pain-free spine centered over the pelvis in the coronal and sagittal planes. The minimum number of segments should be fused. These concepts remain challenging in the setting of deformity and instability. Successful results can be obtained if the surgeon understands the technology available, its capabilities, biological limitations, and the desired solution.
The authors prefer to use thoracic pedicle screws when treating patients with spinal deformity because they provide greater corrective forces for realignment. This allows shorter-segment constructs and the possibility of true derotation in correction. In this article the authors focus on the use of thoracic transpedicular screw fixation in the management of complex spinal disorders and deformity.
Edward Rainier G. Santos, Jonathan N. Sembrano, Benjamin Mueller and David W. Polly
The authors performed a study to determine the optimal iliac screw size, length, and trajectory that produce the highest insertional torques.
Ten fresh cadavers were used and 7.5 × 140–mm and 9.5 × 140–mm iliac screws were placed using 3D image guidance in a randomized fashion in 1 of 2 trajectories. The screws were inserted from the posterior superior iliac spine (PSIS) to either 1) supraacetabular bone or 2) the anterior inferior iliac spine (AIIS). Insertional torque was measured for each full revolution, and the concomitant depth for each torque measurement was recorded. Insertional torque was correlated with detailed bony anatomy.
There was no difference in mean peak insertional torque between the 2 trajectories (25.6 ± 16.4 in-lb [supraacetabular], 26.3 ± 18.2 lb-in [AIIS]; p = 0.8). However, there was a difference between the 2 screw diameters (21.1 ± 10.9 lb-in [7.5-mm-diameter screw], 33.7 ± 19.4 lb-in [9.5-mm-diameter screw]; p = 0.0003). The greatest mean peak insertional torques were observed at depths greater than 80 mm (12.7 ± 9.6 lb-in [≤ 80 mm], 23.7 ± 15.7 lb-in [> 80 mm]; p = 2.6 × 10−7). Insertional torque peaks correlated with engagement of the lateral iliac cortex and the superior iliac fossa.
Although the trajectory had no effect on insertional torque, increased torques are achievable by placing larger-diameter and longer screws in proximity to bony landmarks, most of which are at distances greater 80 mm from the entry point at the PSIS. Iliac screws longer than those commonly used in clinical practice can be safely and accurately placed using image guidance, and reproducible screw paths can be achieved.
Osa Emohare, Alison Dittmer, Robert A. Morgan, Julie A. Switzer and David W. Polly Jr.
Recently published data make it possible to generate estimates of bone mineral density (BMD) by using CT attenuation; this innovation can save time and reduce costs. Although advanced age is associated with reduced BMD, especially in patients with a fracture of C-2, relatively few patients ever undergo formal dual x-ray absorptiometry studies. To the authors' knowledge, this is the first study to assess the utility of this technique in elucidating BMD in patients with an acute fracture of the cervical spine.
Patients who presented to a Level I trauma center with an acute fracture of the cervical spine and underwent abdominal (or L-1) CT scanning either at admission or in the 6 months before or after the injury were evaluated. Using a picture-archiving and communication system, the authors generated regions of interest of similar size in the body of L-1 (excluding the cortex) and computed mean values for CT attenuation. The values derived were compared with threshold values, which differentiate between osteoporotic and nonosteoporotic states; age-stratified groups were also compared.
Of the 91 patients whose data were reviewed, 51 were < 65 years old (mean 43.2 years) and 40 were ≥ 65 years old (mean 80.9 years). The overall mean CT attenuation values (in Hounsfield units [HU]), stratified according to age, were 193.85 HU for the younger cohort and 117.39 HU for the older cohort; the result of a comparison between these two values was significant (p < 0.001).
Using opportunistic CT scanning, this study demonstrates the relative frequency of osteoporosis in acute fractures of the cervical spine. It also objectively correlates overall BMD with the known higher frequency of C-2 fractures in older patients. This technique harnesses the presence of opportunistic CT scans of the abdomen, which potentially reduces the need for the extra time and cost that may be associated with dual x-ray absorptiometry scanning.