Melvin D. Helgeson

Evaluation of Proximal Junctional Kyphosis in Adolescent Idiopathic Scoliosis Following Pedicle Screw, Hook, or Hybrid Instrumentation

Study Design. Retrospective review. Objective. To compare the incidence of and risk factors for proximal junctional kyphosis (PJK) in adolescent idiopathic scoliosis (AIS) following posterior spinal fusion using hook, pedicle screw, or hybrid constructs. Summary of Background Data. Proximal junctional kyphosis is a recently recognized phenomenon in adults and adolescents after AIS surgery. The postoperative effect on PJK with the use of hooks, hybrid constructs, or screws has not been compared in a multicenter study to date. Methods. From a multicenter database, the preoperative and 2-year follow-up radiographic measurements from 283 patients with AIS treated with posterior spinal fusion using hooks (group 1, n = 51), hybrid constructs (group 2, n = 177), pedicle screws (group 3, n = 37), and pedicle screws with hooks only at the top level (group 4, n = 18) were compared. Results. The average proximal level kyphosis at 2 years after surgery was 8.2° (range −1 to 18) in the all screw constructs, representing a significant increase when compared with hybrid and all hook constructs, 5.7° (P = 0.02) and 5.0° (P = 0.014), respectively. Conversely, average postoperative T5–T12 kyphosis was significantly less (P = 0.016) in the screw group compared with the all hook group. Of potential interest, but currently not statistically significant, was the trend towards a decrease in proximal kyphosis in constructs with all pedicle screws except hooks at the most cephalad segment, 6.4°. The incidence of PJK (assuming PJK is a kyphotic deformity greater than 15°) was 0% in group 1, 2.3% in group 2, 8.1% in group 3, and 5.6% in group 4 (P = 0.18). Patients with PJK had an increased body mass index compared with those who did not meet criteria for PJK (P = 0.013). Conclusion. Adjacent level proximal kyphosis was significantly increased with pedicle screws, but the clinical significance of this is unclear. A potential solution is the substitution of hooks at the upper-instrumented vertebrae, but further investigation is required.

Bioartificial dermal substitute: a preliminary report on its use for the management of complex combat-related soft tissue wounds.

Objective: To report our institutional experience with the use of a bioartificial dermal substitute (Integra) combined with subatmospheric pressure [vacuum-assisted closure (VAC)] dressings followed by delayed split-thickness skin grafting for management of complex combat-related soft tissue wounds secondary to blast injuries. Design: Retrospective review of patients treated December 2004 through November 2005. Setting: Military treatment facility. Patients/Participants: Integra grafting was performed 18 times in 16 wounds at our institution. Indications for Integra placement were wounds not amenable to simple split-thickness skin grafting, specifically those with substantial exposed bone and/or tendon. Intervention: Patients underwent an average of 8.5 irrigation and debridement procedures and concurrent VAC dressings prior to placement of the Integra. Following Integra grafting, all patients were managed with VAC dressings, changed every 3 to 4 days at the bedside or in clinic, with subsequent split-thickness skin grafting an average of 19 days later. Main Outcome Measurements: The mechanism and date of injury, size of residual soft tissue deficit, indication for Integra placement, number of irrigation and debridement procedures prior to Integra placement, days from injury to Integra placement, days from Integra placement to split-thickness skin grafting, and clinical outcome were recorded. Results: Integra placement and subsequent skin grafting was successful in achieving durable and cosmetic definitive coverage in 15 of 16 wounds with two of these patients requiring repeat Integra application. Two patients with difficult VAC dressing placement had early Integra graft failure but successfully healed following repeated Integra application and skin grafting. Conclusions: Bioartificial dermal substitute grafting, when coupled with subatmospheric dressing management and delayed split-thickness skin grafting, is an effective technique for managing complex combat-related soft tissue wounds with exposed tendon. This can potentially lessen the need for local rotational or free flap coverage.

Update on the evidence for adjacent segment degeneration and disease

BACKGROUND CONTEXT The evidence surrounding the topic of adjacent segment degeneration and disease has increased dramatically with an abundant amount of literature discussing the incidence of and techniques to avoid it. However, this evidence is often confusing to discern because of various definitions of both adjacent segment degeneration and disease. PURPOSE To organize and review the recent evidence for adjacent segment degeneration and disease. RESULTS Although multifactorial, three distinct causes of adjacent segment disease in both the lumbar and cervical spine have been discussed: the natural history of the adjacent disc; biomechanical stress on the adjacent level caused by the fusion; and disruption of the anatomy at the adjacent level with the initial surgery. The incidence of adjacent segment degeneration in the lumbar spine has been widely reported in the literature from 0% to 100%; conversely, the reported incidence in the cervical spine is less variable. Similarly, strategies at avoiding adjacent segment disease in the lumbar spine include arthroplasty, dynamic fixation, and percutaneous fixation, whereas in the cervical spine the focus has remained on arthroplasty. CONCLUSIONS Adjacent segment disease and degeneration remain a multifactorial problem with several techniques being developed recently to minimize them. In the future, it is likely that the popularity of these techniques will be dependent on the long-term results, which are currently unavailable.

Adjacent vertebral body osteolysis with bone morphogenetic protein use in transforaminal lumbar interbody fusion.

BACKGROUND CONTEXT Recent studies have demonstrated cases of adjacent vertebral body osteolysis when assessing the effect of bone morphogenetic protein (BMP) on fusion rates. However, no study to date has evaluated the course of osteolysis at different periods. PURPOSE To determine the incidence and resolution of osteolysis associated with BMP used in transforaminal lumbar interbody fusions (TLIF). STUDY DESIGN Retrospective review. PATIENT SAMPLE All TLIF cases using BMP performed at one institution with routine postoperative computed tomography (CT) scans at defined intervals. OUTCOME MEASURES Area of osteolysis and fusion as determined by CT scan. METHODS We performed a retrospective analysis of all patients at our facility who underwent TLIF with BMP. Included were all patients who had obtained a CT scan within 48 hours of surgery, 3 to 6 months postoperatively, and 1 to 2 years postoperatively. Areas of osteolysis were defined as lucency within the vertebral body communicating with the interbody spacer that was not present on the immediately postoperative CT scan. Areas of osteolysis were measured in all three planes and the volume used for comparison of the 3 to 6 months CT scans with the greater than 1 year CT scan. RESULTS Twenty-three patients who underwent TLIF with BMP had obtained CT scans at all time periods required for evaluation. Seventy-eight vertebral bodies/end plates were assessed for osteolysis (39 levels). The incidence of osteolysis 3 to 6 months postoperatively in the adjacent vertebral bodies was 54% compared with 41% at 1 to 2 years. The mean volume of osteolysis was at 0.216 cm(3) at 1 to 2 years compared with 0.306 cm(3) at 3 to 6 months (p=.082). The area/rate of osteolysis did not appear to significantly affect the rate of fusion or final outcome with an overall union rate of 83%. CONCLUSIONS The rate of osteolysis decreased at 1 year compared with 3 to 6 months, but only 24% of the vertebral bodies with evidence of osteolysis at 3 to 6 months completely resolved by 1 year.

Salvage of C2 Pedicle and Pars Screws Using the Intralaminar Technique: A Biomechanical Analysis

Study Design. Human cadaveric biomechanical analysis. Objective. The purpose of this study is to evaluate the ability of using 1 of the remaining 2 methods of instrumenting C2, should the initial method fail. Summary of Background Data. Although 3 different methods of C2 fixation (pedicle, pars, and laminar screws) are possible, occasionally an attempt at screw insertion fails. In such cases, the surgeon needs a viable alternative to salvage/obtain fixation to obviate the need to instrument an additional motion segment. Methods. Eleven fresh-frozen cadaveric specimens (Occ-C4) were DEXA scanned for bone mineral density. On the left side, pedicle screws were first inserted, then pulled out. Then, “salvage” pars screws were inserted, then pulled-out, followed by laminar screws. On the right, a similar sequence was repeated, except that a pars screw was followed by a pedicle screw, then a laminar screw. All screws were placed by experienced cervical spine surgeons. Insertional torque (IT) was measured in Newton-meters (Nm). Tensile loading to failure was performed “in-line”with the screw axis at a rate of 0.25 mm/s using a MTS 858 MiniBionix II System with data recorded as peak pull-out strength (POS) in newtons “N”. Results. Pedicle screws generated statistically greater IT and POS than other techniques as the initial fixation type (P < 0.0001). Similar trends were observed with transpedicular fixation as a salvage procedure (P > 0.05). Laminar screws yielded consistently higher POS values than pars fixation when applied in a salvage scenario (POS range: LS = 146–707 N; PrS = 8–548 N); however, high standard deviation precluded statistical significance (P > 0.05). Significant predictive relationship was established between IT and POS for all screws using Pearson correlation coefficient and bivariate linear regression analysis (r = 0.75 and r2 = 0.511, respectively; P < 0.01). Conclusion. Our results suggest that pedicle screws provide the strongest fixation for both initial and salvage applications. If they should fail, lamina screws appear to provide stronger and more reproducible fixation than pars screws.

Does superior-segment facet violation or laminectomy destabilize the adjacent level in lumbar transpedicular fixation? An in vitro human cadaveric assessment.

Study Design. This is an in vitro biomechanical study. Objective. The current investigation was performed to evaluate adjacent level kinematic change following unilateral and bilateral facet violation and laminectomy following 1-, 2-, and 3-level reconstruction. Summary of Background Data. The incidence of superior-segment facet violation with lumbar transpedicular fixation has been reported as high as 35%; however, its contribution to biomechanical instability at the supradjacent level is unknown. In addition, superior-segment laminectomy has been implicated as a risk factor for the development of adjacent level disease. The authors assess the acute biomechanical effects of proximal facet violation and subsequent laminectomy in an instrumented posterior fusion model in 10 cadaveric specimens. Methods. Biomechanical testing was performed on 10 human cadaveric spines under axial rotation (AR), flexion-extension (FE), and lateral bending (LB) loading. After intact analysis, pedicle screws were inserted from L5-S1 and testing repeated with: (1) preserved L4–L5 facets, (2) unilateral facet breach, (3) bilateral breach, and (4) L5 laminectomy. Following biomechanical analysis, instrumentation was extended to L4, then L3 and biomechanical testing repeated. Full range of motion (ROM) at the proximal adjacent levels were recorded and normalized to intact (100%). Results. Supradjacent level ROM was increased for all groups under all loading methods relative to intact (P < 0.05). However, AR testing revealed progressive instability at the adjacent level in groups 3 and 4, relative to group 1, following 1-, 2- and 3-level fixation (P < 0.05). During FE, supradjacent level ROM was significantly increased for group 4 specimens compared with group 1 after L5-S1 fixation (P < 0.05), and was greater than all other groups for L3-S1 constructs (P < 0.05). Interestingly, under lateral bending, facet joint destabilization did not change adjacent segment ROM. Conclusion. There were significant changes in proximal level ROM immediately after posterior stabilization. However, an additional increase in supradjacent segment ROM was recorded during AR after bilateral facet breach. Subsequent complete laminectomy at the uppermostfixation level further destabilized the supradjacent segment in FE and AR. Therefore, meticulous preservation of the cephalad–most segment facet joints–is paramount to ensure stability.

Effect of Teriparatide [rhPTH(1,34)] and Calcitonin on Intertransverse Process Fusion in a Rabbit Model

Study Design. Randomized, double-blinded, placebo controlled animal study. Objective. To evaluate the effect of teriparatide and calcitonin after an intertransverse process spinal fusion in a rabbit model. Summary of Background Data. It is widely recognized that some osteoporosis medications, including bisphosphonates, can interfere with bone healing. Although prescribed frequently in the treatment of osteoporosis, the effect of teriparatide and calcitonin on spinal fusion has not been fully elucidated. We hypothesized that teriparatide, being the only anabolic medication for osteoporosis treatment, would have a beneficial effect on spine fusion. Methods. Fifty-one New Zealand white rabbits underwent a posterolateral L5–L6 intertransverse process arthrodesis using autogenous iliac crest bone graft. The rabbits were randomly divided into 3 groups. All animals received daily subcutaneous injections of group I (n = 17) 1 mL of saline placebo; group II (n = 17) 10 &mgr;g/kg/day of teriparatide; group III (n = 17) 14 IU/animal of calcitonin during the 8-week postoperative period. Postmortem analyses included manual palpation, radiographic, biomechanical, and histologic assessment. Three random 10× fields were examined/graded within the cephalad, middle, and caudal regions of each section (810 fields). Fusion quality was graded using the Emery histologic scale (0–7 based on fibrous/bone content of the fusion mass). Results. Histologic fusion rates for teriparatide averaged 86.7% and was significantly greater than the autograft control group (50%) (P = 0.033). Radiographically, there was a strong trend towards teriparatide being superior to the calcitonin group (85.7% vs. 56.3%, respectively; P = 0.07). The average Emery grading score was 5.99 ± 1.46 SD for the autologous group and 6.26 ± 0.93 SD for the teriparatide group (P = 0.031). Although not significant, the teriparatide group showed less motion in flexion/extension, lateral bending, and axial rotation. Conclusion. Our results suggest that teriparatide enhances spinal fusion while calcitonin has a neutral effect. The teriparatide group had the best histologic fusion rate and Emery scores, while the calcitonin group was similar to the saline controls. Although not significant, the teriparatide group had a strong trend towards superior radiographic fusion over the calcitonin group.

Low lumbar burst fractures

BACKGROUND CONTEXT The most common location for burst fractures occurs at the thoracolumbar junction, where the stiff thoracic spine meets the more flexible lumbar spine. With our current military conflicts in Iraq and Afghanistan, we have seen a disproportionate number of low lumbar burst fractures. PURPOSE To report our institutional experience in the management of low lumbar burst fractures. STUDY DESIGN Retrospective review. METHODS We performed a retrospective review of medical records and radiographs for all patients treated at our institution with combat-related injuries and thoracolumbar fractures. We included all patients who had sustained a burst fracture from T12 to L5 and had at least 1-year clinical follow-up. RESULTS Thirty-two patients sustained burst fractures. Nineteen patients (59.4%) had low lumbar (L3-L5) burst fractures, and 12 patients (37.5%) had thoracolumbar junction (T12-L2) burst fractures as their primary injury. Additionally, seven patients sustained less severe burst fractures at an additional level. One patient sustained burst fractures at both upper and lower lumbar levels. Of the low lumbar fractures, 52.6% had evidence of neurologic injury, two of which were complete. Similarly, in the upper lumbar group, 58.2% sustained a neurologic injury, two of which were complete. Twenty-two patients underwent surgical intervention, complicated by infection in 18%. At most recent follow-up, all but one patient with presenting neurologic injury had persistent deficits. CONCLUSION Low lumbar burst fractures are the predominant combat-related spine injury in our current military conflicts. The rigidity offered by current body armor may effectively lower the transition zone that normally occurs at the thoracolumbar junction, thereby, transferring forces into the lower lumbar spine. Increased awareness of this fracture pattern is warranted by all surgeons because of unique clinical challenges associated with its treatment. Although the incidence is increased in the military population, other surgeons may be involved with long-term care of these patients on completion of their military service.

Retrospective Review of Lumbosacral Dissociations in Blast Injuries

Study Design. Retrospective review of medical records and radiographs. Objective. We assessed the clinical outcomes of lumbosacral dissociation (LSD) after traumatic, combat-related injuries, and to review our management of these distinct injuries and report our preliminary follow-up. Summary of Background Data. LSD injuries are an anatomic separation of the pelvis from the spinal column, and are the result of high-energy trauma. A relative increase in these injuries has been seen in young healthy combat casualties subjected to high-energy blast trauma. Methods. We performed a retrospective review of inpatient/outpatient medical records and radiographs for all patients treated at our institution with combat-related lumbosacral dissociations. Twenty-three patients met inclusion criteria of combat-related lumbosacral dissociations with one-year follow-up. Patients were treated as follows: no fixation (9), sacroiliac screw fixation (8), posterior spinal fusion (5) and sacral plate (1). All patients with radiographic evidence of a zone III sacral fracture, in addition to associated lumbar fractures indicating loss of the iliolumbar ligamentous complex integrity were included. Results. In 15 patients, the sacral fracture were an H or U type zone III fracture, whereas in the remaining nine, the sacral fracture was severely comminuted and unable to classify (six open fractures). There was no difference in visual analog scale (VAS) between treatment modalities. Two open injuries had residual infections. One patient treated with an L4-ilium posterior spinal fusion with instrumentation required instrumentation removal for infection. At a mean follow-up of 1.71 years (range, 1–4.5), 11 patients (48%) still reported residual pain and the mean VAS at latest follow-up was 1.7 (range, 0–7). Conclusion. Operative stabilization promoted healing and earlier mobilization, but carries a high-postoperative risk of infection. Nonoperative management should be considered in patients whose comorbidities prevent safe stabilization.

Acute and Long-term Stability of Atlantoaxial Fixation Methods: A Biomechanical Comparison of Pars, Pedicle, and Intralaminar Fixation in an Intact and Odontoid Fracture Model

Study Design. In vitro human cadaveric biomechanical study. Objective. The aims of this project were to evaluate the acute segmental fixation and long-term screw stability afforded by 3 C2 fixation techniques: intralaminar, pars, and pedicle. Summary of Background Data. C2 intralaminar screws offer the advantages of avoiding the vertebral artery; however, direct biomechanical comparison of this technique to the other methods of instrumenting C2 has not been performed. Methods. Fourteen cadaveric specimens were dual energy radiograph absorptiometry scanned and segregated into 2 groups (n = 7/group) matching the C2 bone mineral density. All specimens were instrumented with C1 lateral mass and C2 intralaminar screws while measuring the insertional torque (IT). In group 1 C2 pars screws were inserted while in group 2 pedicle screws were placed. Nondestructive testing was performed in axial rotation, flexion/extension (FE), and lateral bending. The odontoid was then resected and loading repeated. Subsequently, specimens were disarticulated about C2 and individually loaded for 2000 cycles in the cephalocaudad plane. The screws were then failed by a tensile load directed in the parasagittal plane. Full range of motion over C1–C2 and peak screw pull-out force was quantified. Results. Transpedicular technique generated significantly higher IT than the pars screws and marginally greater IT than intralaminar screws. With the intact atlantoaxial ligamentous complex, intralaminar fixation was superior to pars and similar to pedicle instrumentation at limiting axial torsion. After odontoid destabilization, however, this technique was less effective at reducing the lateral bending range of motion. Destructive loading revealed the highest pull-out forces with the pedicle screws, followed by intralaminar and pars screws. Conclusion. Our results suggest that C2 intralaminar fixation provides a viable alternative to pedicle screws and is superior to pars instrumentation in cases with preserved atlantoaxial ligamentous attachments. In the presence of a traumatic dens fracture, however, intralaminar fixation may not be the optimal choice.