Biomechanical evaluation of the craniovertebral junction after odontoidectomy with anterior C1 arch preservation: A finite element study.

OBJECTIVE: Odontoidectomy with preservation of the anterior C1 arch can be increasingly achieved by an endoscopic endonasal approach. It is controversial whether preservation of the anterior C1 arch after odontoidectomy can prevent instability of the craniovertebral junction （CVJ） and avoid posterior fixation. The aim of this research was to investigate the biomechanical effect of the preserved anterior C1 arch after odontoidectomy. METHODS: A validated finite element model of a whole cervical spine (occipital bone to T1) was constructed to study the biomechanical changes due to traditional odontoidectomy (TO) and odontoidectomy with preservation of the anterior C1 arch (OPC1). RESULTS: The greatest biomechanical changes in the cervical spine model after TO and OPC1 occurred at C0-C1 and C1-C2. At C0-C1 and C1-C2, the motion changes of the TO and OPC1 models had no significant difference in flexion, extension and lateral bending. Compared with the intact model, motion increases of the two surgical models were both extremely significant at C1-C2 in extension (128.2% vs. 128.1%) and lateral bending (178% vs. 156%). In axial rotation, the TO approach produced more motions than the OPC1 approach, especially at C1-C2(90.3° under TO approach, and 74.6° under OPC1 approach). CONCLUSIONS: Preservation of the anterior C1 arch after odontoidectomy can preserve the axial rotational motion at C0-C1 and C1-C2, whereas the motions in extension and lateral bending continue to have an extremely abnormal increase at C1-C2. Thus, instability of the CVJ still exists, and posterior internal fixation may also be required after OPC1.

Enlarged External Occipital Protuberance in young French individuals' head CT: stability in prevalence, size and type between 2011 and 2019.

External Occipital Protuberance (EOP) enlargement has been recently reported to increase in young adults, with a putative link with postural factors such as the use of smartphones. This study aims to analyze finely the changes in prevalence and size of EOP enlargement in millennials, throughout the smartphone era (2011 - 2019). Anonymized head Computerized Tomography (CT) examinations from patients aged 18-30 in 2011 (n = 205) or 2019 (n = 240), were reviewed to assess the type of EOP and to measure its volume in case of enlargement. Additional CT analyses were performed on two ancient skulls, from a XVIth century young male and a young female Egyptian mummy. There was no significant evolution in the prevalence of EOP enlargement between 2011 (92/205, 44.9%) and 2019 (106/240; 44.2%) (P = 0.92). There was no significant evolution either in the distribution of enlarged EOP volumes (P = 0.14) or of EOP types (P = 0.92) between 2011 and 2019. In the meantime, rates of smartphone ownership in millennials rose from 35% to 98%. Compared to 2019 volumes, the Egyptian mummy displayed an EOP enlargement corresponding to the 85th percentile for young women, and the XVIth century skull to the 73rd percentile for young men. In conclusion, on a population scale, prevalence and volume of enlarged EOP in millennials remain stable between 2011 and 2019, which makes the impact of rapidly growing modern environmental factors on EOP changes unlikely. EOP enlargement was also already present in ancient skulls from young individuals, with measurements within today's upper ranges.

Taxonomy of the Lateral Strain Patterns at the Sphenobasilar Synchondrosis for Osteopathic Cranial Manipulative Medicine.

Lateral strain is a type of nonphysiologic cranial dysfunction that occurs at the sphenobasilar synchondrosis. In this dysfunction, the sphenoid and occiput rotate in the same directions along 2 vertical axes. There is currently no consensus on the nomenclature for this cranial dysfunction. In this article, the authors provide a standard nomenclature for lateral strains using the historical writings of pioneers in osteopathic medicine, including William Gardner Sutherland, DO, Anne L. Wales, DO, and Harold Magoun, DO. The authors establish the following consensus: (1) Lateral strains are named for the side to which the basisphenoid shifts; (2) The more prominent greater wing of the sphenoid is on the same side to which the basisphenoid shifts; (3) In vault and fronto-occipital holds, the holds form a parallelogram shape, with the index fingers pointing to the same side as the more prominent greater wing; and (4) The hand that is on the side of the prominent greater wing will shift anteriorly while the hand on the opposite side will shift posteriorly.

Exploring the association between number, severity, location of fracture, and occiput-to-wall distance.

This study of women with a suspected vertebral fracture determined the association between vertebral fracture characteristics and posture. The number of fractures was associated with posture. Severity of fracture was associated with posture when adjusting for pain. Fracture characteristics explain some variability in posture in women with a suspected vertebral fracture. PURPOSE: Osteoporotic vertebral fractures are associated with increased morbidity and mortality. An accumulation of vertebral fractures may lead to forward head posture, which has been independently associated with mortality. It is unclear how fracture characteristics, including the number, severity, and location of fracture, contribute to occiput-to-wall distance (OWD). METHODS: This was a cross-sectional secondary data analysis using baseline data from a randomized controlled trial, in community-dwelling women aged 65 years and older with a suspected vertebral fragility fracture. Lateral thoracic and lumbar spine radiographs were used to determine the number, location, and severity of fracture. Occiput-to-wall distance (OWD) was used to assess forward head posture. Pain during movement (0-10 scale) and age were considered as confounding variables. Multivariable regression models were used to evaluate relationships between fracture variables and OWD. RESULTS: Participants (n = 158) were of mean age 75.9 (SD 6.5) years with a mean (SD) BMI = 26.7 (5.3) kg/m2, OWD = 5.7 (4.6) cm, and number of fractures = 2.4 (2.4). In unadjusted analyses, the number of fractures (B = 0.82, 95% CI = 0.04, 1.59) was associated with OWD. When adjusting for pain, severity of fractures (B = 1.08, 95% CI = 0.001, 2.15) was independently associated with OWD. Location was not associated with OWD in any of the models. CONCLUSIONS: The number of fractures was significantly associated with OWD in the unadjusted model, explaining more of the variability in OWD than other fracture characteristics. Severity of fracture was associated with OWD in the adjusted model. However, pain may confound the relationship between OWD and fracture characteristics.

Occipital Condyle Syndrome: A Red Flag for Malignancy. Comprehensive Literature Review and New Case Report.

OBJECTIVE: To perform a literature review of the epidemiology, clinical presentation, diagnostic evaluation, and clinical course of occipital condyle syndrome, including a new case report. BACKGROUND: Occipital condyle syndrome (OCS) is a rare clinical syndrome, consisting of unilateral occipital headache accompanied by ipsilateral hypoglossal palsy. This headache typically radiates to the temporal region, and is triggered by contralateral head rotation. It is usually associated with skull base metastasis, often unrevealed in basic neuroimaging studies. OCS might be the first manifestation of malignancy, and its unfamiliarity can lead to a delay in the diagnosis. METHODS: We performed a systematic literature review using PubMed and Embase for OCS, along with a new case report. RESULTS: A total of 35 cases (mean age 59 years, range 25-77), 24 (70%) men, presented typical unilateral headache followed by ipsilateral hypoglossal palsy from 0 to 150 days after headache presentation. In 16 patients (46%), initial neuroimaging studies were normal. OCS was due to skull base metastasis in 32 cases (91%). In 18 patients (51%), OCS was the first symptom of disease. CONCLUSIONS: OCS represents a warning sign and requires an exhaustive search for underlying neoplasm. An appropriate clinical evaluation can lead to an earlier diagnosis in patients with consistent headache.

A New Method of Measuring the Occipitocervical Angle That Could be Applied as an Intraoperative Indicator During Occipitocervical Fusion.

STUDY DESIGN: A method for measuring occipitocervical angle. OBJECTIVE: To develop a new method of measurement for assessing the occipitocervical angle using intraoperative fluoroscopic imaging, and to examine its reliability. SUMMARY OF BACKGROUND DATA: To avoid postoperative complications following occipitocervical fusion, it is vital to obtain a suitable fusion angle between the occipital bone and the upper cervical spine. MATERIALS AND METHODS: The subjects were 30 cases with occipito-atlanto-axial lesions and 30 healthy volunteers. Lateral plain radiographs of the cervical spine in neutral position were used to draw the McGregor line, the line between the external occipital protuberance and the most caudal point on the midline occipital curve (Oc line), the tangential line of the inferior endplate of the C2 vertebra (C2 line), and the posterior longitudinal line of the C2 vertebra (Ax line). The angles formed by these 4 lines and the horizontal line were measured. The O-C2 angle and the Oc-Ax angle, the new indicator, were measured by 3 doctors and reliability was evaluated. RESULTS: In the disease group, mean intraobserver variances of the McGregor, Oc, C2, Ax, O-C2, and Oc-Ax angles were 0.7, 1.3, 1.5, 1.2, 1.6, and 1.9 degrees. Mean intraobserver intraclass correlation coefficients were 0.997, 0.994, 0.994, 0.997, 0.989, and 0.988, showing high intraobserver reliability for all angles. Mean interobserver intraclass correlation coefficients were 0.998, 0.996, 0.994, 0.997, 0.988, and 0.990, showing high interobserver reliability for all angles. The same reliability was obtained in the healthy group. CONCLUSIONS: The Oc-Ax angle is as reliable an indicator as the conventional O-C2 angle, and could be used as a new intraoperative indicator for occipitocervical fusion. It may be particularly useful in cases where it is difficult to identify the McGregor line and/or the inferior endplate of the C2 vertebra. LEVEL OF EVIDENCE: Level 3-diagnostic study.

Biomechanical Comparison of Modified TARP Technique Versus Modified Goel Technique for the Treatment of Basilar Invagination: A Finite Element Analysis.

STUDY DESIGN: A finite element analysis. OBJECTIVE: The aim of this study was to determine the biomechanical differences between atlantoaxial fusion cage combined with transoral atlantoaxial reduction plate fixation (TARP + Cage, modified TARP technique) and that combined with C1 lateral mass screw and C2 pedicle screw fixation (C1LS + C2PS + Cage, modified Goel technique) in the treatment of basilar invagination (BI) by finite element analysis. SUMMARY OF BACKGROUND DATA: Clinical studies have shown that transoral anterior atlantoaxial release followed by TARP fixation can achieve reduction, decompression, fixation, and fusion of C1-C2 through a transoral-only approach. Although cage has been used to reduce the BI through posterior approach, there are no studies referred to the cage combined with TARP for C1-C2 fusion. METHODS: A finite element model was used to investigate and compare the stability between TARP + Cage fixation and C1LS + C2PS + Cage fixation in the treatment of BI. Vertical load of 40  N was applied on the C0, to simulate head weight, and 1.5  Nm torque was applied to the C0 to simulate flexion, extension, lateral bending, and rotation. RESULTS: In comparison with the C1LS + C2PS + Cage model, the TARP + Cage model reduced the ROM by 44.7%, 30.0%, and 10.5% in extension, lateral bending, and axial rotation, while the TARP + Cage model increased the ROM by 30.0% in flexion, and the TARP + Cage model also led to lower screw stress in all motions with one exception (anterior C2PS stress in extension). CONCLUSION: Our results indicate that the TARP + Cage fixation may offer higher stability to C1LS + C2PS + Cage in extension, lateral bending, and axial rotation but lower stability in flexion. Compared with modified Goel technique, the modified TARP technique not only has the capability of transferring the load and distributing the stress but also can provide neural decompression, stabilization and fusion, and restore C1-C2 normal fusion angle. LEVEL OF EVIDENCE: N/A.

Optimal trajectory for the occipital condyle screw.

STUDY DESIGN: Retrospective analysis. OBJECTIVE: To understand what may constitute an optimal trajectory for an occipital condyle (OC) screw. SUMMARY OF BACKGROUND DATA: OC screws are an alternative to standard occipital plates as a cephalad fixation point in occipitocervical fusion. An optimal trajectory for placement of OC screws has not been described. METHODS: We conducted a computed tomography-based study of 340 human occipital condyls. All computed tomographies were negative for traumatic, degenerative, and neoplastic pathology. On the basis of the current literature, linear measurements of distances were made based on a constant entry point. Medial angulations of 10, 20, and 25 degrees relative to the sagittal midline were used. In addition, 10-, 5-degree cranial, 10- and 30-degree caudal angulations were studied to evaluate the incidence of hypoglossal canal and atlantooccipital joint compromise. RESULTS: Average distances were 17.1±2.8, 20.4±2.8, and 22.2±2.9 for 10, 20, and 25 degrees of medial angulation, respectively. Right-sided and left-sided measurements for each category were not significantly different. However, the difference in the measured distances between 10 versus 20 degrees, 10 versus 25 degrees, and 20 versus 25 degrees was all significantly different (P<0.01). Hypoglossal canal compromise incidence was 0% and 7.1% for 5- and 10-degree cranial angulation, respectively. Atlantooccipital joint compromise incidence was 21.8% and 99.1% for 10- and 30-degree caudal angulation, respectively. CONCLUSIONS: The condylar entry point should be medial to the condylar fossa, midcondylar, and ≥2 mm caudal to the skull base. An optimal trajectory for the OC screw should have a medial angulation of ≥20 degrees relative to the sagittal midline, trying to stay parallel to the skull base. Minor adjustments in angulation can be made, but any adjustment approaching 10 degrees cranial or caudal leads to an increased risk of hypoglossal canal cranially or atlantooccipital joint compromise caudally.

Evaluation of occipitocervical subluxation in rheumatoid arthritis patients, using coronal-view reconstructive computed tomography.

STUDY DESIGN: Reconstructive computed tomography (CT) study of occipito-atlanto and atlantoaxial joints in RA patients. SUMMARY OF BACKGROUND DATA: The occipitocervical region is one of the most common sites of rheumatoid arthritis (RA). Although lateral radiography has been used for the diagnosis of atlantoaxial subluxation and vertical subluxation, reconstructive CT imaging of the occipito-atlanto and atlantoaxial joints is more sensitive in detecting morphologic changes in this region. We investigated this region in RA patients, using coronal-view reconstructive CT images, and examined the relationship between the morphology and other radiographic parameters. METHODS: The occipitocervical region was examined in 58 female RA patients by reconstructive CT, plain radiography, and MRI. The degree of destructive change on reconstructive CT was compared to that on other radiographic evaluations. RESULTS: Coronal-view reconstructive CT revealed primary destructive changes before detection by lateral radiography, using Redlund-Johnell or Ranawat values. A Redlund-Johnell value less than 34 mm was diagnostic for occipitocervical subluxation in female RA patients. CONCLUSION: Coronal-view reconstructive CT is useful for the diagnosis of occipitocervical joint subluxation in RA.

The kinematic relationships of the upper cervical spine.

STUDY DESIGN: A retrospective radiographic study. OBJECTIVE: To elucidate the kinematic relationships of the upper cervical spine. SUMMARY OF BACKGROUND DATA: To our knowledge, few reports have described the kinematic relationships of the upper cervical spine in patients with general age-related cervical spondylosis. METHODS: We performed Kinetic magnetic resonance imaging for 295 consecutive patients experiencing neck pain without neurologic symptoms. Subjects with rheumatoid arthritis, traumatic history, and severe degenerative changes in the upper cervical spine were excluded. Anterior atlantodens interval (AADI) and the cervicomedullary angle in 3 different postures were measured, and the variations in each value between flexion and neutral (F-N), neutral and extension (N-E), and flexion and extension (F-E) were calculated. The subjects were classified into 3 groups according to the space available for the cord values (A: or=15 mm). RESULTS: AADI significantly increased from extension to flexion posture, however, no significant differences were observed in every posture among the groups. F-N variation in AADI showed no significant differences among the groups; however, N-E variation between Groups A and C and between Groups B and C and F-E variation between Groups A and C showed significant differences. The cervicomedullary angle significantly increased from flexion to extension posture, however, no significant differences were observed in every posture among the groups. Angle variations among the groups showed no significant differences, except for F-N angle variation between Groups B and C. None of the variations in AADI and the cervicomedullary angle were significantly correlated. CONCLUSION: Our results suggest that only the kinematics of the atlantoaxial movement, especially the posterior movement, was greatly affected by the narrowing of space available for the cord. The central atlantoaxial joint may be closely related to the mechanisms for protecting the spinal cord by restriction of the atlantoaxial movement.

Biomechanical evaluation of parasagittal occipital plating: screw load sharing analysis.

STUDY DESIGN: Biomechanical evaluation of occipitocervical instrumentation techniques. OBJECTIVE: Compare methods of occipital instrumentation by quantifying load sharing of occipital screws and measuring motion across instrumented occipitocervical spines. SUMMARY OF BACKGROUND DATA: Newer occipitocervical plate/screw systems that attach to longitudinal rods have been developed to improve fixation. These devices place screws in the center of occipital bone or off-midline. Midline plates offer screw purchase in thicker bone. Off-midline systems may increase the effective moment arm for torsional and lateral bending control. Measurement of screw loads within occipital plates is useful for determining optimal plate configuration. METHODS: Ten cadaveric specimens (occiput-C4) were tested in flexion/extension (FE), lateral bending (LAT), and axial rotation (ROT) over +/-3 Nm pure moment. After intact testing, 4 occipitocervical fixation constructs were tested using washer load cells to assess loading across screws used to fix the plates to the occiput. Parasagittal occipital plates were positioned either convex or concave side facing medially. Each plate was first fixed using 3 screws (rostral, middle, caudal), then with the caudal screw eliminated (simulated failure). Range of motion (ROM) and peak screw loads are reported. RESULTS: ROM decreased from intact to any of the 4 fusion plate configurations in FE, LAT, and ROT (P << 0.05), but not between plate configurations. Screw load significantly decreased from medially convex to medially concave configurations in LAT, but no significant changes were observed in FE or ROT. With caudal screws removed, middle screws peak loads significantly increased in FE and LAT (P < 0.05), but not ROT. CONCLUSION: Occipital screw placement off-midline improves screw loads under lateral bending forces on occipitocervical constructs, though loads for FE and ROT are unchanged. As screws pullout, the loads may be redistributed, resulting in increased screw pullout forces above. Despite the improvement in screw loads for laterally based plates during lateral bending, overall ROM across the occipitocervical junction is unchanged.

Sleep apnea in rheumatoid arthritis patients with occipitocervical lesions: the prevalence and associated radiographic features.

Since sleep apnea is a risk factor for high mortality of rheumatoid arthritis (RA) patients, this study examined the prevalence in RA patients with occipitocervical lesions, and the associated radiographic features. Twenty-nine RA patients requiring surgery for progressive myelopathy due to occipitocervical lesions (3 males, 26 females, average age 65 years) were preoperatively evaluated. Twenty-three (79%) had sleep apnea defined as apnea-hypopnea index >5 events per hour measured by a portable monitoring device, and all of them were classified as the obstructive type. Among gender, age, bone mass index (BMI), and radiographic parameters related to occipitocervical lesions: atlantodental interval (ADI), cervical angles (O/C1, C1/2, and C2/6), and cervical lengths (O-C2 and O-C6), the ADI and cervical lengths were shown to be significantly associated with the presence of sleep apnea by parametric statistical analysis. Since there were positive correlations between the ADI and cervical lengths by Pearson's test, we performed a multivariate logistic regression analysis after adjustment for confounding factors and found that small ADI was the principle parameter associated with sleep apnea. We therefore conclude that the prevalence of sleep apnea is higher than that in a general RA population that was reported previously, and believe that occipitocervical lesions are an independent risk factor for this condition. Small ADI and short neck, secondary to the vertical translocation by RA, may cause obstructive sleep apnea, probably through mechanical or neurological collapse of the upper airway.

Influence of cranial morphology on the location of chronic subdural haematoma.

BACKGROUND: The purpose of the present study was to evaluate the relationship between cranial morphology and location of a chronic subdural haematoma (CSDH) in patients with and without intracranial vault asymmetry. METHOD: The study was conducted in 110 consecutive adult patients who underwent surgery for CSDH. The relationship between the following variables and CSDH was studied: sex, age, past medical history, history of trauma, interval between head injury and symptoms, clinical presentation, location of the CSDH, symmetry of the frontal and occipital intracranial vault on the CT scan and/or MR images, surgical treatment and outcome. Throughout the analysis, p < 0.05 was considered statistically significant. FINDINGS: The frontal cranial vault was symmetrical in 48 patients (43.6%) and asymmetrical in 62 patients (56.4%). CSDH was more commonly bilateral in patients with a symmetrical frontal cranial vault than those with an asymmetrical shape (41.7% vs 17.7% and this difference is statistically significant (p = 0.01). In 62 patients with an asymmetric frontal skull vault, the CSDH was bilateral in 11 patients. In the remaining 51 patients, the CSDH was located on the same side of the most curved frontal convexity in 34 patients and on the side of the less curved frontal convexity in 17 patients. The occipital cranial vault was symmetrical in 44 patients (40%) and asymmetrical in 66 patients (60%). CSDH was more commonly bilateral in patients with a symmetrical occipital cranial vault than those with an asymmetrical one (40.9% vs 19.7%) and this difference was also statistically significant (p = 0.019). In 66 patients with an asymmetric occipital skull vault, the CSDH was bilateral in 13 patients. In the remaining 53 patients, the CSDH located on the same side of the most curved occipital convexity in 39 patients and on the side of the less curved occipital convexity in 14 patients. CONCLUSIONS: Frontal and occipital intracranial vault morphology provides valuable information about location of CSDH. Bilateral CSDH is common in patients with symmetrical frontal and occipital cranial vault. In asymmetrical cranium, CSDH usually locates on the same side of the most curved frontal or occipital convexity. Identification of this relationship can be very useful to elucidate the origin and the pathogenesis of CSDH.

Growth and development of the cranial base in mice that spontaneously develop anterior transverse crossbite.

INTRODUCTION: Homozygous brachymorphic (bm/bm) mice are characterized by disproportionately short stature. Anterior transverse crossbite sometimes develops spontaneously in inbred BALB/c-bm/bm mice that produce undersulfated glycosaminoglycans. The reason for inferior growth of the craniofacial structures of BALB/c-bm/bm mice has not been clarified. We focused on the spheno-occipital synchondrosis (SOS)-the growth site of the cranium-and investigated the growth pattern in the SOS of the BALB/c-bm/bm mice. METHODS: The BALB/c-+/+ mice and BALB/c-bm/bm mice without malocclusion were used. Sagittal sections of the cranial bases were stained with hematoxylin and eosin, observed histologically, and used to measure the anteroposterior length and the percentage of active bone volume around the SOS. RESULTS: In the histologic study, compared with the cartilage in the BALB/c mice, a bipolar column was not seen in the cartilage of the SOS in the BALB/c-bm/bm mice. The histomorphometric study showed that the anteroposterior length of the posterior cranial base was significantly shorter and the percentage of active bone volume was significantly smaller in the BALB/c-bm/bm mice than in the BALB/c mice. CONCLUSIONS: The results suggested that normal endochondral growth is disturbed in the synchondrosis of the posterior base of the skull in BALB/c-bm/bm mice.

Sutural strain in orthopedic headgear therapy: a finite element analysis.

INTRODUCTION: The goal of this study was to analyze the strains induced in the sutures of the midface and the cranial base by headgear therapy involving orthopedic forces. Does the mechanical signal induced in the sutures sufficiently account for a growth-influencing effect? METHODS: A finite element model of the viscerocranium and the neurocranium was used. It consisted of 53,555 tetrahedral elements and 97,550 nodes. The strain induced in the sutures of the cranial base and the midface when applying orthopedic headgear forces of 5 and 10 N was computed and recorded with an interactive measurement tool. RESULTS: The magnitude and the distribution of the measured strains depended on the level and the direction of the acting force. Overall, the strain values measured at the sutures of the midface and the cranial base were moderate. The measured peak values at a load of 5 N per side were usually just below 20 microstrain irrespective of the force direction. A characteristic distribution of strain values appeared on the anatomical structures of the midface and the cranial base for each vector direction. The measurements based on the finite element method provided a good overview of the approximate magnitudes of sutural strains with orthopedic headgear therapy. The signal arriving in the sutures is apparently well below threshold, since the maximum measured strains in most sutures were about 100 fold lower than the minimal effective strain. A skeletal effect of the orthopedic headgear due to a mechanical effect on sutural growth cannot be confirmed from these results. CONCLUSIONS: The good clinical efficacy of headgear therapy with orthopedic forces is apparently based mainly on dentoalveolar effects, whereas the skeletal effect due to inhibition of sutural growth is somewhat questionable.

A detailed morphologic and functional magnetic resonance imaging study of the craniocervical junction in adolescent idiopathic scoliosis.

STUDY DESIGN: Conventional and phase-contrast magnetic resonance (MR) imaging were used to evaluate the morphology and cerebrospinal fluid (CSF) flow dynamics at craniocervical junction in adolescent idiopathic scoliosis (AIS). OBJECTIVES: To determine the morphology of cerebellar tonsil, foramen magnum, and dynamic flow of CSF at the craniocervical junction in AIS patients versus normal controls and their correlation with somatosensory cortical evoked potentials (SSEP). SUMMARY OF BACKGROUND DATA: Previous studies have documented obstructed CSF flow in patients with Chiari I malformation. Low-lying cerebellar tonsils and syringomyelia are also observed in AIS patients. We sought to investigate whether disturbed CSF flow is also evident in AIS subjects at the foramen magnum level and its association with level of cerebellar tonsils and dimensions of foramen magnum. METHODS: Conventional and phase-contrast MR were performed in 105 adolescent girls (69 AIS subjects and 36 age-matched controls). Measurements of cerebellar tonsillar level related to the basion-opsithion (BO) line, anteroposterior (AP), transverse (TS) diameter, and area of foramen magnum, and peak velocity of CSF flow in both the anterior and posterior subarachnoid space through foramen magnum were obtained. Correlations were made among different parameters and SSEP findings. RESULTS: A total of 42% of subjects in the AIS group had the cerebellar tonsillar tip positioned 1 mm below the BO line. The cerebellar tonsillar level in AIS subjects was significantly lower than the median tonsillar level in normal controls (P < 0.01). The AP diameter and area of foramen magnum were significantly larger in AIS subjects when compared with normal controls (P < 0.05), but the peak CSF velocities through foramen magnum showed no significant difference (P > 0.05). CONCLUSION: Peak CSF velocities through foramen magnum were not significantly different in AIS subjects despite the presence of low-lying cerebellar tonsils. This might be explained by the compensatory effect of larger foramen magnum in AIS subjects.