Vascular malformations of the orbit: classification and the role of imaging in diagnosis and treatment strategies*.

PURPOSE: To describe the authors' experience with orbital vascular malformations using the International Society for the Study of Vascular Anomalies (ISSVA) classification and the preferred radiologic techniques. METHODS: Review of clinical and radiologic experience from 1976 to 2012. This article presents the findings from several studies conducted on vascular malformations of the orbit, all of which received institutional review board approval when needed. RESULTS: The orbital vascular malformations can be evaluated, classified, and managed according to the ISSVA classification to provide a common language of communication between specialties, which takes into account flow dynamics. CONCLUSIONS: The ISSVA can be applied for vascular malformations of the orbit.

Malignant degeneration of a lumbar osteochondroma into a chondrosarcoma which mimicked a large retropertioneal mass.

We report a case of a lumbar spinal osteochondroma that transformed into a large chondrosarcoma in a 39-year-old male who presented with an abdominal mass and back pain. This mass was also associated with a fracture of the stalk, which on cross-sectional imaging mimicked a mass of retroperitoneal origin. The diagnosis of chondrosarcoma transforming from a lumbar osteochondroma became apparent when comparison was made with previous studies.

Plain radiograph assessment of spinal hardware.

The frequency and variety of spinal instrumentation has increased tremendously over the past 100 years, and imaging plays an important role in evaluating the postoperative spine. Although assessment of spinal hardware often involves a multimodality approach, plain radiographs are the most commonly used modality, given accessibility, cost, relatively low radiation dose compared with computed tomography, and provision of positional information. An approach to assessment of plain radiographs of the postoperative spine is discussed, and examples of common postoperative complications are provided, including infection, hardware failure, incomplete fusion, and junctional failure.

Imaging in adult scoliosis: preoperative assessment and postoperative complications.

Adult scoliosis rates range from 2 to 32%. Surgery for scoliosis is common. Accurate and surgically relevant information should be provided to the referring surgeon from pre- and postoperative imaging. There are various methods to correct scoliosis surgically with the end points correction of the curve and relief of symptoms. This is achieved through the placement of spinal instrumentation with a goal of osseous fusion across the instrumented levels. There are many potential postoperative complications. The initial and postoperative imaging, types of surgery, and hardware are reviewed along with the common early and late complications with relevant illustrations.

Accuracy and safety of pedicle screw fixation in thoracic spine trauma.

OBJECT: The authors evaluated the accuracy of placement and safety of pedicle screws in the treatment of unstable thoracic spine fractures. METHODS: Patients with unstable fractures between T-1 and T-10, which had been treated with pedicle screw (PS) placement by one of five spine surgeons at a referral center were included in a prospective cohort study. Postoperative computed tomography scans were obtained using 3-mm axial cuts with sagittal reconstructions. Three independent reviewers (C.B., V.S., and D.G.) assessed PS position using a validated grading scale. Comparison of failure rates among cases grouped by selected baseline variables were performed using Pearson chi-square tests. Independent peri- and postoperative surveillance for local and general complications was performed to assess safety. Twenty-three patients with unstable thoracic fractures treated with 201 thoracic PSs were analyzed. Only PSs located between T-1 and T-12 were studied, with the majority of screws placed between T-5 and T-10. Of the 201 thoracic PSs, 133 (66.2%) were fully contained within the pedicle wall. The remaining 68 screws (33.8%) violated the pedicle wall. Of these, 36 (52.9%) were lateral, 27 (39.7%) were medial, and five (7.4%) were anterior perforations. No superior, inferior, anteromedial, or anterolateral perforations were found. When local anatomy and the clinical safety of screws were considered, 98.5% (198 of 201) of the screws were probably in an acceptable position. No baseline variables influenced the incidence of perforations. There were no adverse neurological, vascular, or visceral injuries detected intraoperatively or postoperatively. CONCLUSIONS: In the vast majority of cases, PSs can be placed in an acceptable and safe position by fellowship-trained spine surgeons when treating unstable thoracic spine fractures. However, an unacceptable screw position can occur.

Clinical exacerbation of multiple sclerosis following radiotherapy.

BACKGROUND: Radiation of the central nervous system in patients with demyelinating disease may have deleterious effects. OBJECTIVE: To describe a 30-year-old woman with multiple sclerosis who developed an attack of demyelination 2 months following radiotherapy for a parotid malignancy. RESULTS: Magnetic resonance imaging demonstrated new hyperintense lesions that corresponded to both the localization of the patient's symptoms and to the area defined by the 50% isodose radiation field. CONCLUSION: Radiation treatment likely triggered an exacerbation of multiple sclerosis.

Relationship of neural axis level of injury to motor recovery and health-related quality of life in patients with a thoracolumbar spinal injury.

BACKGROUND: Outcomes following traumatic conus medullaris and cauda equina injuries are typically predicted on the basis of the vertebral level of injury. This may be misleading as it is based on the assumption that the conus medullaris terminates at L1 despite its variable location. Our primary objective was to determine whether the neural axis level of injury (the spinal cord, conus medullaris, or cauda equina) as determined with magnetic resonance imaging is better than the vertebral level of injury for prediction of motor improvement in patients with a neurological deficit secondary to a thoracolumbar spinal injury. METHODS: Patients diagnosed with a motor deficit secondary to a thoracolumbar spinal injury, and who met the inclusion criteria, were contacted. Each patient had a magnetic resonance imaging scan that was reviewed by a spine surgeon and a neuroradiologist to determine the termination of the conus medullaris and the neural axis level of injury. Patient demographic data were collected prospectively at the time of admission. Admission and follow-up neurological assessments were performed by formally trained dedicated spine physiotherapists. RESULTS: Fifty-one patients were evaluated at a median of 6.2 years (range, 2.7 to 12.3 years) postinjury. The final motor scores differed significantly according to whether the patient had a spinal cord injury (mean, 62.8 points; 95% confidence interval, 55.4 to 70.2), conus medullaris injury (mean, 78.6 points; 95% confidence interval, 70.3 to 86.9), or cauda equina injury (mean, 88.8 points; 95% confidence interval, 78.9 to 98.7) (p = 0.0007). A univariate analysis showed the improvement in the motor scores after the cauda equina injuries (mean, 17.1 points; 95% confidence interval, 8.3 to 25.9) to be significantly greater than that after the spinal cord injuries (mean, 7.7 points; 95% confidence interval, 3.1 to 12.3) (p = 0.03). A multivariate analysis showed that an absence of initial sacral sensation had a negative effect on motor recovery by a factor of 13.2 points (95% confidence interval, 4.2 to 22.1). When compared with classifying our patients on the basis of the neural axis level of injury, reclassifying them on the basis of the vertebral level of injury resulted in a misclassification rate of 33%. CONCLUSIONS: The motor recovery of patients with a thoracolumbar spinal injury and a neurological deficit is affected by both the neural axis level of injury as well as the initial motor score. The results of this study can help the clinician to determine a prognosis for patients who sustain these common injuries provided that he or she evaluates the precise level of neural axis injury utilizing magnetic resonance imaging.

Utility of screening for blunt vascular neck injuries with computed tomographic angiography.

PURPOSE: To prospectively study the impact of implementing a computed tomographic angiography (CTA)-based screening protocol on the detected incidence and associated morbidity and mortality of blunt vascular neck injury (BVNI). METHODS: Consecutive blunt trauma patients admitted to a single tertiary trauma center and identified as at risk for BVNI underwent admission CTA using an eight-slice multi-detector computed tomography scanner. The detected incidence, morbidity, and mortality rates of BVNI were compared with those measured before CTA screening. A logistic regression model was also applied to further evaluate potential risk factors for BVNI. RESULTS: A total of 1,313 blunt trauma patients were evaluated. One hundred seventy screening CTAs were performed, of which 33 disclosed abnormalities. Twenty-three were evaluated angiographically, of which 15 were considered to have significant BVNIs, as were 4 of the 10 patients with abnormal CTAs and no angiogram. The incidence of angiographically proven BVNIs in our series was 1.1%. If four patients who were treated for BVNIs based on CTA alone are included, the incidence rises to 1.4%. This is significantly higher than the 0.17% incidence before screening (p < 0.001). In addition, the delayed stroke rate and injury-specific mortality fell significantly from 67% to 0% (p < 0.001) and 38% to 0% (p = 0.002), respectively. Overall mortality also fell significantly, from 38% to 10.5% (p = 0.049). Univariate logistic regression identified the presence of cervical spine injury as a significant predictor of BVNI (p < 0.001). CONCLUSION: CTA screening increases the detected incidence of BVNI 8-fold, with rates similar to angiographically based screening protocols. CTA screening significantly decreases BVNI-related morbidity and mortality in an efficient manner, underlying its utility in the early diagnosis of this injury.

Neuroimaging features of heroin inhalation toxicity: "chasing the dragon".

OBJECTIVE: Our objective was to illustrate the dramatic neuroimaging findings of toxic leukoencephalopathy caused by heroin vapor inhalation. CONCLUSION: Symmetric abnormality involving the cerebellar white matter and posterior limb of the internal capsule is characteristic of heroin vapor inhalation toxicity, although involvement may be more extensive, depending on the severity of the condition. MR imaging and CT appear to be essential for making this diagnosis because clinical history is often unreliable and findings at physical examination are nonspecific.