Vascular injury during lumbar spine surgery is a relatively rare complication but can have devastating outcomes. The injury may not be apparent during surgery and can present acutely or late in various manners, and some injuries can be asymptomatic. This report discusses the unusual case of a 35-year-old woman who underwent a right L4–5 microdiscectomy for disc herniation and 4 days postoperatively presented with a pulmonary embolus. A subsequent CT scan revealed a pseudoaneurysm and arteriovenous fistula of the right common iliac vein and artery, which gave rise to the embolus. The patient received a right iliac artery stent, and at 4 months after surgery she continues to be symptom free. This report describes the atypical presentation of vascular injury after lumbar microdiscectomy and stresses the importance of cautiously using the pituitary rongeur when removing deeper disc fragments.
Daniel Huttman, Mathew Cyriac, Warren Yu and Joseph R. O'Brien
Lauren M. Burke, Warren D. Yu, Anthony Ho, Timothy Wagner and Joseph R. O'Brien
Anatomical variability of the C-2 pedicle poses a challenge for C-2 fixation. The use of multidimensional CT scanning is not widely used but might be an asset to preoperative planning. Careful preoperative planning is imperative for instrumentation at C-2. Fine-cut, noncontrast CT scanning is a useful tool for delineating anatomy; however, the axis of the images is not always along the anatomical axis of the vertebra in question. The authors evaluated the suitability of C-2 pedicles for screw placement by using OsiriX (Pixmeo) software to change the gantry angle of CT angiograms to measure the anatomical dimensions of the C-2 pedicle.
The authors conducted a retrospective review of CT angiograms of the head and neck from 47 trauma patients seen consecutively at George Washington University Hospital. For each patient, 3 independent observers determined length and width of each C-2 pedicle (94 samples) by using OsiriX. OsiriX is a DICOM viewer that enables navigation and visualization in multidimensional imaging, such as 3D imaging, which was used for this study. Sex-specific measurements were also determined. Vertebral anatomy was studied to determine whether aberrant anatomy would preclude pedicle fixation. Statistical analyses were performed.
Of the 47 patients, 27 were male. Overall mean C-2 pedicle widths and lengths were 8.272 ± 1.364 mm and 27.052 ± 3.471 mm, respectively. The average widths and lengths of the pedicle in female patients were 8.040 ± 1.262 mm and 27.241 ± 2.731 mm, respectively, and those in male patients were 8.444 ± 1.414 mm and 26.913 ± 3.933 mm, respectively. The sex difference was statistically significant for width (p = 0.012) but not for length (p = 0.41). On the basis of width, the percentages of pedicles that could tolerate a 3.5-mm and 4.0-mm screw were 98% and 97%, respectively. Vertebral anatomy precluded screw length greater than 14 mm for only 3 patients.
Using multidimensional CT or 3D imaging, the authors found that C-2 pedicles in over 90% of patients could tolerate 3.5-mm and 4.0-mm pedicle screws. Vertebral anatomy precluded use of screw lengths greater than 14 mm for only 3 (6%) of 47 patients. Therefore, the C-2 pedicle might be more tolerant of fixation than previously reported.
Charles Milchteim, Warren D. Yu, Anthony Ho and Joseph R. O'Brien
Cervical transfacet screw placement has been described in the literature. Although the technique shows promise for percutaneous application, parameters for screw placement have not been well delineated. This study used reconstructed CT scans with imaging software to assess the feasibility of percutaneous transfacet screw placement, analyzing potential entry angles, transfacet lengths, and sex differences at each subaxial level.
Fifty consecutive cervical CT scans (obtained in 26 males and 24 females [mean age 41.5 years]) were reformatted using OsiriX software, and transfacet lengths, entry angles, and potential occipital clearance were analyzed at all subaxial levels. Statistical analyses were used to determine the differences, if any, between transfacet lengths, entry angle, and occipital clearance across individual cervical levels. Repeatability was quantified by calculating the intraclass correlation coefficient and Cohen kappa value.
A total of 200 transfacet lengths and 200 entry angles in 50 patients were analyzed. The mean transfacet lengths were 17.9 ± 2.6, 17.6 ± 3.2, 16.3 ± 3.6, and 13.1 ± 2.2 mm at C3–4, C4–5, C5–6, and C6–7, respectively, with mean entry angles at 52.7° ± 7.8°, 56.5° ± 8.0°, 55.0° ± 8.8°, and 53.0° ± 8.7°, respectively. Analysis of variance revealed a significant difference between the mean transfacet lengths, while post hoc analysis revealed significantly larger transfacet lengths in the upper 2 cervical levels (C3–4 and C4–5) than in the lower 2 cervical levels (C5–6 and C6–7). Analysis of variance demonstrated no significant difference between the entry angles. Males had significantly larger transfacet lengths at C5–6 (17.4 vs 15.1 mm) and C6–7 (13.7 vs 12.4 mm) than females. The occiput would have blocked percutaneous screw placement in 86%, 78%, 54%, and 20% of the cases at C3–4, C4–5, C5–6, and C6–7, respectively. Transfacet lengths may accommodate longer screws in the upper cervical spine, but potential screw sizes decrease in the lower subaxial levels. A transfacet entry angle of approximately 50° or greater was associated with a higher incidence of occipital clearance. Additionally, the occiput may pose a significant obstruction to percutaneous transfacet fixation in upper subaxial levels. Interrater reliability was poor for screw angle and length measurements, but was satisfactory in intrarater analysis in 6 of 8 measurements. There was moderate to good agreement of occipital clearance in all but one measurement.
Cervical transfacet screw placement is possible from C-3 to C-7. Because occipital clearance can be difficult at C3–4 and C6–7, the use of curved or flexible instruments may be necessary to obtain the appropriate screw trajectory. Screw lengths varied with spinal level and the sex of the patient.
Matthew Mantell, Mathew Cyriac, Colin M. Haines, Manasa Gudipally and Joseph R. O’Brien
Insufficient biomechanical data exist from comparisons of the stability of expandable lateral cages with that of static transforaminal lumbar interbody fusion (TLIF) cages. The purpose of this biomechanical study was to compare the relative rigidity of L4–5 expandable lateral interbody constructs with or without additive pedicle screw fixation with that of L4–5 static TLIF cages in a novel cadaveric spondylolisthesis model.
Eight human cadaver spines were used in this study. A spondylolisthesis model was created at the L4–5 level by creating 2 injuries. First, in each cadaver, a nucleotomy from 2 channels through the anterior side was created. Second, the cartilage of the facet joint was burred down to create a gap of 4 mm. Light-emitting-diode tracking markers were placed at L-3, L-4, L-5, and S-1. Specimens were tested in the following scenarios: intact model, bilateral pedicle screws, expandable lateral 18-mm-wide cage (alone, with unilateral pedicle screws [UPSs], and with bilateral pedicle screws [BPSs]), expandable lateral 22-mm-wide cage (alone, with UPSs, and with BPSs), and TLIF (alone, with UPSs, and with BPSs). Four of the spines were tested with the expandable lateral cages (18-mm cage followed by the 22-mm cage), and 4 of the spines were tested with the TLIF construct. All these constructs were tested in flexion-extension, axial rotation, and lateral bending.
The TLIF-alone construct was significantly less stable than the 18- and 22-mm-wide lateral lumbar interbody fusion (LLIF) constructs and the TLIF constructs with either UPSs or BPSs. The LLIF constructs alone were significantly less stable than the TLIF construct with BPSs. However, there was no significant difference between the 18-mm LLIF construct with UPSs and the TLIF construct with BPSs in any of the loading modes.
Expandable lateral cages with UPSs provide stability equivalent to that of a TLIF construct with BPSs in a degenerative spondylolisthesis model.
Lauren E. Matteini, Khaled M. Kebaish, W. Robert Volk, Patrick F. Bergin, Warren D. Yu and Joseph R. O'Brien
Multiple techniques of pelvic fixation exist. Distal fixation to the pelvis is crucial for spinal deformity surgery. Fixation techniques such as transiliac bars, iliac bolts, and iliosacral screws are commonly used, but these techniques may require separate incisions for placement, leading to potential wound complications and increased dissection. Additionally, the use of transverse connector bars is almost always necessary with these techniques, as their placement is not in line with the S-1 pedicle screw and cephalad instrumentation. The S-2 alar iliac pelvic fixation is a newer technique that has been developed to address some of these issues. It is an in-line technique that can be placed during an open procedure or percutaneously.
Report of 3 cases
Richard M. Young, Jonathan H. Sherman, Joshua J. Wind, Zachary Litvack and Joseph O'Brien
The object of this study was to demonstrate that a posterior-only approach for craniocervical junction pathology is feasible with intraoperative reduction. The authors reviewed 3 cases of craniocervical instability. All patients had craniocervical instability according to radiological imaging and various methods of measurement, with results outside the normal range. Posterior instrumentation aided the intraoperative reduction techniques while maintaining structural integrity and the desired fusion construct. No anterior approach was necessary in any of the patients. Neurological symptoms resolved in two patients and significantly improved in another. Follow-up imaging demonstrated stable constructs.
There are many approaches to anterior cervical pathology at the craniocervical junction. Posterior instrumented reduction and stabilization of the occipitocervical spine can be safely achieved, obviating the need for a transoral approach in the setting of craniocervical junction settling.