Eldin E. Karaikovic

The load sharing classification of spine fractures

Study Design. A 3 to 4 year follow-up was performed on a consecutive series of 28 patients who had three-column spinal fractures surgically stabilized by short-segment instrumentation with first generation VSP (Steffee) screws and plates and autograft fusion. The follow-up revealed 10 patients with broken screws. Background Data. Retrospective examination of preoperative radiographs and computed tomographic axial and sagittal reconstruction images clearly demonstrated that the screw fractures all occurred in patients with a disproportionately greater amount of injury to the vertebral body. Results. A point system (the load sharing classification) was developed that grades: 1) the amount of damaged vertebral body, 2) the spread of the fragments in the fracture site, and 3) the amount of corrected traumatic kyphosis. Conclusions. This point system can be used preoperatively to: 1) predict screw breakage when short segment, posteriorly placed pedicle screw implants are being used, 2) describe any spinal injury for retrospective studies, or 3) select spinal fractures for anterior reconstruction with strut graft, short-segment-type reconstruction.

Morphologic characteristics of human cervical pedicles

Study Design. Cervical pedicle morphology was investigated using manual and computed tomography measurements. Objectives. Normal anatomic variations of the cervical pedicles were measured to evaluate their safety as anchors for posterior cervical fixation systems. Summary of Background Data. There have been no cervical pedicle measurements on a large number of specimens. No study has ever measured the inner pedicle diameter. Methods. Fifty‐three spinal columns (C2‐C7) of Euro‐American origin identified by age, sex, and height (318 vertebrae or 636 pedicles) were measured using a digital caliper, a goniometer, and computed tomography scanning. Results. The pedicle axis lengths were similar from C3 to C7 (except for shorter C2 pedicles). In the horizontal plane, the medial inclination of the pedicles followed the cervical spinal cord enlargement. In the sagittal plane, the pedicles were directed superiorly in the upper spine and inferiorly in the lower cervical spine. Some pedicles had no medullary canal (i.e., were solid cortical bone: 0.9% C2, 2.8% C3 and C4, and 3.8% C5 pedicles). The outer pedicle width was smaller than the height in most of the pedicles. The inner pedicle width was equal to or smaller than 2 mm in 13.2% C2, 72.6% C3, 67.0% C4, 62.3% C5, 51.9% C6, and 16.0% C7. The outer pedicle width was equal to or smaller than 4 mm in 8.5% C2, 75.5% C3, 35.8% C4, 13.2% C5 and C6, and 6.6% C7 pedicles. The thinnest pedicle cortex was always the lateral cortex, which protects the vertebral artery. Measurements of the posterior pedicle projection also were taken. Conclusions. These data provide anatomic limitations to pedicle screw use in the cervical spine.

Accuracy of cervical pedicle screw placement using the funnel technique.

Study Design. This was a cadaver study assessing the accuracy of cervical pedicle screw placement. Objective. To evaluate the accuracy of the funnel technique of screw placement. Summary of Background Data. Although excellent results have been reported in clinical studies, with no major neurovascular injuries, several cadaveric studies have shown a high pedicle perforation rate during screw placement. Methods. Ten fresh frozen cervical spines (C2–C7) were used (120 pedicles, 20 pedicles per level). The average specimen age was 79.6 years (range 65–97); the average height was 159 cm (range 155–175). The male-to-female ratio was 3:7. Pedicle width and angulation were measured on preoperative axial computed tomography (1-mm slices). By use of four bony landmarks and the funnel technique, screws were placed under direct vision. Critical perforations (documented contact of a screw with, or an injury to, a spinal cord, nerve root, or vertebral artery) and noncritical perforations (a perforation with no critical contact) were recorded. Results. In seven pedicles (5.8%) the procedure was aborted because of a small or nonexistent pedicle medullary canal. Ninety-four pedicle screws (83.2%) were placed correctly, whereas 11 pedicles (9.7%) had noncritical perforations and 8 pedicles (7.1%) had critical perforations. The majority of the critical and noncritical perforations were at C3, C4, and C5. Conclusions. Axial computed tomography is necessary for the preoperative planning. Because of the small diameter and steep angulation of cervical pedicles, every spine surgeon who intends to use pedicle screws should first master the technique on cadavers.

Surgical anatomy of the cervical pedicles: landmarks for posterior cervical pedicle entrance localization.

The posterior entrance to the cervical pedicle is described using quantitative and descriptive parameters. Fifty-three spines (C2-C7) were evaluated using a digital caliper and by visual inspection using four bony landmarks: the lateral vertebral notch and inferior articular process (C2-C7), the medial pedicle cortex at C2, and the transverse process at C7. Three distances were defined. (1) At C2, the average medial pedicle cortex-pedicle distance was 7.2 mm. (2) The lateral vertebral notch-pedicle distances showed that the entrances were located close to the notch at C2, almost at the notch at C3 and C4, and gradually moved medially away from the notch from C5 to C7. The pedicles were rarely located lateral to the lateral vertebral notch. (3) The inferior articular process-pedicle distance was large at C2, the shortest at C3, and gradually increased toward C7. Three relations were defined. (1) The pedicles were located mostly in the intermediate third of the inferior facet at C2; in the lateral third at C3, C4, and C7; or in the lateral or intermediate thirds at C5 and C6. Only C2 and C6 pedicles were located in its medial third. (2) The pedicles were located mostly below the lateral vertebral notch at C2, at C3-C6, or almost equally above and at the notch at C7. (3) Most of the C7 pedicles were located below the midline of the transverse process. The location of the pedicle entrance was unique at each cervical level. Their distribution followed the cervical spinal cord enlargement. These landmarks should assist with safe placement of pedicle screws.

The accuracy of pedicle screw placement in the thoracic spine using the "Funnel Technique": part 1. A cadaveric study.

A cadaveric study using the “funnel technique” to probe thoracic pedicles was conducted. The results (location, level, and perforation rate) of three spine surgeons of varying experience were compared. The objectives were to evaluate the reliability and accuracy of the funnel technique for the placement of thoracic pedicle screws and to describe the technique. Nine fresh cadavers (216 thoracic pedicles) were used for pedicle screw placement using the funnel technique. The study was conducted by three spine surgeons with a significantly different level of experience in thoracic pedicle screw placement (72 thoracic pedicles each). Critical and noncritical perforations were recorded. The perforation rate was 6% (13 of 216 pedicles). Of this, only 0.4% (1 of 216) was a critical perforation (a contact with T8 nerve root). The junior spine surgeon who had no previous experience with thoracic pedicle screw placement had a 12.5% (9 of 72) perforation rate, the surgeon very familiar with the technique had a 5.5% (4 of 72) perforation rate, and the senior author who originated this technique had a 1.4% (1 of 70) perforation rate. All perforations made by the junior spine surgeon occurred in his first 24 pedicles; none occurred in his last 48 pedicles. The reliability of the funnel technique in placement of thoracic pedicle screws was proven in our cadaveric study. It provided even an entry-level surgeon with a safe way to identify and place thoracic pedicle screws. The funnel technique is a simple, safe, and cost-effective alternative to any other currently recommended techniques for pedicle screw placement.

Thoracic pedicle screw instrumentation using the "Funnel Technique": part 2. Clinical experience.

This study is a retrospective review of the clinical results of patients treated with thoracic pedicle screws using the “funnel technique.” The objective is to report the clinical results of patients treated with thoracic pedicle screws using the funnel technique for screw placement. A total of 115 patients treated with the use of at least one thoracic pedicle screw were retrospectively identified. All patients were treated at a single medical center, under the senior authors supervision. Twenty-five different residents were responsible for placing 50–60% of these screws, and five different fellows and the senior author placed the remainder. The accuracy of screw placement and the complications related to the use of thoracic pedicle screws were analyzed by assessing intraoperative and postoperative charts and radiographs. There were 115 patients (age range 9–82 years) with the average follow-up period of 17 months. The total number of screws used was 348; the screw diameter ranged from 4.0 to 7.75 mm. There were no vascular or pulmonary complications. There was no iatrogenic neurologic injury, except for one patient who developed a transient anterior thigh numbness from intraoperative positioning. In fracture patients, 35% (10 of 28) had documented neurologic improvement postoperatively. There were no obviously misplaced pedicle screws on detailed review of intraoperative and postoperative radiographs. No screws had to be electively removed for complications related to misplacement. There were four broken screws, one broken rod, two loose screws, and three connector disengagements. Two patients had deep infections and four patients had pseudarthrosis requiring additional surgery. The clinical results proved that thoracic pedicle screws can be safely placed with the funnel technique.

Possible complications of anterior perforation of the vertebral body using cervical pedicle screws

No previous studies have analyzed the possible complications of anterior perforation of the cervical vertebral body with pedicle screws. The objective of this study was to identify the possible implications of an anterior vertebral body perforation. Ten consecutive Euro-American cadavers (C2–C7) were used. The male-to-female ratio was 3:7. The average specimen age was 79.6 years (range: 65–97 years), and average height was 159 cm (range: 155–175 cm). Axial computed tomography scans through the isthmus of pedicles were taken. Five millimeter and 10 mm margins anterior to the vertebral bodies were defined. Within 5 mm anterior to the anterior cortex of the vertebral body, we found mostly muscles (at C2: m. longus colli and pharyngeal constrictors; at C3 and C4: m. scalenus medius, longus colli, pharyngopalatinus and pharyngeal constrictors; at C5 and C6: m. longus colli and longus capitis; and at C7: m. longus colli), except at C3, C4, and C7, where the pharynx and esophagus were within the margin. Between 6 and 10 mm, we found mostly hollow organs (at C2: pharynx and small veins; at C3 and C4: the same muscles as within the 5 mm margin, with addition of the pharynx and some small veins; at C5 and C6: pharynx, pharyngeal constrictors and the thyroid cartilage; and at C7: the esophagus). Except C2, there is no safe zone anterior to the cervical vertebral bodies in the cervical spine, which would allow bicortical purchase of pedicle screws without being close to important surrounding structures.

A comparison of biomechanical stability and pullout strength of two C1–C2 fixation constructs

BACKGROUND CONTEXT Several fusion techniques are used to treat atlantoaxial instability. Recent literature suggests that intralaminar screw (LS) fixation and pedicle screw (PS) fixation offer similar stability and comparable pullout strength. No studies have compared these characteristics after cyclic loading. PURPOSE To compare the stability and pullout strength of intra-LSs and PSs in a C1-C2 instability model after 1,000 cycles of axial loading. STUDY DESIGN In vitro biomechanical study. OUTCOME MEASURES Stability in axial rotation and screw pullout strength after cyclic loading. METHODS Six fresh-frozen human cadaveric cervical spines (C1-C2) were used in this study. C1-C2 instability was mimicked via odontoidotomy at its base and posterior soft-tissue release, including the supraspinous ligaments and facet joint capsules. Specimens were tested to 1,000 cycles after stabilization with two fixation constructs: C1 lateral mass (LM) screws and C2 intra-LSs (C1LM-C2LS) and C1 LM screws and C2 PSs (C1LM-C2PS). Angular motion was recorded for right and left axial rotation using an Optotrak 3020 system (Northern Digital, Waterloo, Ontario, Canada). Tensile loading to failure was then performed collinear to the longitudinal axis of the screw, and the data were recorded as peak pullout strength in newtons. RESULTS There was no statistically significant difference in stability (measured in degrees of rotation) between the intra-LS and PS constructs at 250, 500, 750, and 1,000 cycles of axial rotation. Furthermore, there was no significant difference in stability at 250 cycles versus 1,000 cycles for the LS (1.30 vs. 1.49, p = .80) or PS (0.84 vs. 0.85, p = .96). Pedicle screws had higher pullout strength when compared with the intra-LSs (757.5 ± 239 vs. 583.4 ± 472 N); however, high standard deviation precluded statistical significance (p = .44). CONCLUSIONS Our data suggest that a C1LM and C2LS construct has similar biomechanical stability when compared with a C1LM and C2PS construct after 1,000 cycles of axial rotation. Furthermore, PSs had higher pullout strength when compared with LSs; however, this result was not statistically significant.

Coil embolization of a lumbar artery to control vascular injury during intradiscal surgery.

Study Design. Case study. Objective. To emphasize the role that interventional radiology can perform in stemming bleeding to vascular structures injured during spine surgery without altering patient position. Summary and Background Data. Injury to the lumbar artery or aorta may occur during lumbar disc surgery. Occasionally the site of bleeding may not be readily identifiable or accessible through the surgical incision. Interventional radiology techniques may be employed to help locate and stop these difficult to locate vascular structures without changing a patient position. Methods. A 48-year-old woman undergoing L4-L5 lumbar hemilaminectomy and discectomy secondary to a herniated disc sustained an injury to a right L3 lumbar artery. Several liters of blood were lost in an attempt to surgically locate and repair the injury to the lumbar artery. A literature search identified the potential severity and treatment options. Results. An interventional radiologist was called for and he was able to angiographically locate the source of bleeding and stem its source using coil embolization of the lumbar artery. Conclusion. Whenever there is bleeding from an inaccessible site, consultation with an interventional radiologist to perform an intraoperative coil embolization of the injured vessel should be done especially if a resort to an anterior abdominal approach would permit uncontrolled bleeding.

Decision making in thoracolumbar fractures.

The management of thoracolumbar fractures continues to evolve. Strong agreements exist in certain aspects of care but significant controversy remains in many other areas. This paper reviews our current diagnostic and therapeutic approach to treating these injuries as of the spring of 2005. Evaluation Initial assessment of a patient should include the history of an injury from as accurate a source as possible, a thorough physical examination, and an accurate assessment of the patient’s neurological status and spinal stability to identify all the associated major injuries that have occurred. Needs proper wording. Clearly , assessment of neurological status and spinal stability is independent of identification of ‘associated injuries’. Treatment priorities include resuscitation of patient, and treatment of life-threatening injuries before mechanical restoration of the injured osteoligamentous column and preservation or restoration of neurological function. Every spine surgeon has to answer three fundamental questions when facing a thoracolumbar fracture: First, how to treat a patient (non-operative or operative)? Second, how many segments should one instrument and fuse (short versus long segment operation)? Third, which approach should be used (anterior, posterior or both)? The three fundamental questions are 1. whether to operate, 2. When to operate (emer gent, next day, or later), and 3. How to operate (anterior or posterior or combined approach. The answers to the above questions begin with a complete evaluation. Patient Comprehensive assessment of the patient must be performed. The medical issues that have occurred in the past must be identified. Medically unfit, obese, demented or noncompliant patients have to be identified. Their pre-injury personality characteristics influence treatment choices and the successful use of short segment surgical reconstruction. Medical problems that determine the patient’s suitability for surgical reconstruction must be identified and assessed. Short segment reconstructive options—the most sophisticated reconstruction now available—are more appropriate for physically fit, intelligent, healthy patients who can understand the need for compliance with post-operative recommendations until their fracture heals. Non-compliant patients, patients with past psychological disturbances, drug abusers and alcoholics are especially vulnerable to surgical failures. The inability to co-operate with post-operative bracing makes long segment instrumentation and fusion the best reconstructive option for people who cannot be trusted to understand the importance of post-operative bracing. Clinical assessment Spine fractures usually result from blunt injuries, which can cause other long bones fractures too. A high index of suspicion must be maintained with palpation of all joints and bones during examination. F ull neurological e xamination (sensation, motor , anal tone, [Mention about signs of sacral sparing in complete paraplegia] etc) should be done and documented repeatedly to look for and pick up neurological deficits and deterioration. The patient’s spine must be palpated using log roll to look for tenderness, swelling, haematoma, gibbus or step off. These can indicate the existence of fracture translation. [The purpose for palpation of the spine after log roll is to look for evidence of posterior column injuiry, like wide gap between the spinous process, hematoma, ecchymoses etc] A seat belt bruise, facial fractures, pelvic hematoma and calcaneus fractures can suggest the possible existence of major organ injuries that should be addressed by the trauma and general surgeon first. Resuscitation