Paul A. Glazer

Ophthalmic complications after spinal surgery

Study Design. A retrospective review of 3450 spinal surgeries was performed. Objectives. To review ophthalmic complications and their etiologies, as well as treatments and outcomes, in patients who have undergone spinal surgery. Summary of Background Data. Ophthalmic complications after major spinal reconstructive surgery are rare and have not been adequately addressed in the orthopedic literature. Methods. In a series of 3450 spinal surgeries at three institutions, the authors identified seven patients (incidence = 0.20%) whose postoperative course was complicated by loss of visual acuity. These perioperative ophthalmic complications included posterior optic nerve ischemia, occipital lobe infarcts, and central retinal vein thrombosis. Operative time, estimated blood loss, and medical history of peripheral vascular, cardiovascular, or ophthalmic disease were obtained from the charts, as were follow-up data. Results. Three patients recovered completely, and one had partial return of visual function. In the remaining three patients, significant visual loss persisted. Conclusions. The risk of ophthalmic complications with spinal surgery has not been fully appreciated. Because ophthalmic complications in spinal surgery may be reversed with prompt recognition and intervention, it is important for clinicians to be aware of their possible occurrence.

Biomechanical analysis of multilevel fixation methods in the lumbar spine.

Study Design The authors measured and compared the stiffness of cadaveric lumbar spines stabilized with several anterior interbody fusion devices. The information obtained provides a foundation for determining how methods of anterior lumbar fixation can maximize rigidity and promote development of bony fusion. Objectives To compare the utility of three anterior spinal instrumentation systems for stabilizing the lumbar spine. Summary of Background Data Anterior spinal instrumentation is used to prevent progressive spinal deformity and maintain correction after spinal fusion surgery. Newer instrumentation systems developed for anterior interbody fusions can be inserted by minimally invasive procedures. The stability of these systems has not been tested adequately in human cadaveric specimens. Methods Fusion constructs were evaluated in 12 human cadaveric specimens sequentially loaded in axial compression and torsion, flexion and extension, and lateral bending. The fusion constructs used were 1) two anterior bilateral threaded interbody fusion devices, 2) lateral hollow interbody screws (Texas Scottish Rite Hospital-B screws), and 3) femoral allograft and conventional anterior Texas Scottish Rite Hospital instrumentation. Results The construct with Texas Scottish Rite Hospital-B screws connected by a rod produced stiffness comparable with that produced by conventional Texas Scottish Rite Hospital instrumentation with femoral ring allografts. The threaded interbody fusion device stiffness tested in axial rotation was comparable with that achieved with Texas Scottish Rite Hospital instrumentation. Conclusions Our data demonstrate the effectiveness of threaded interbody fusion device and the Texas Scottish Rite Hospital-B screw in immobilizing the L3-L4 and L4-L5 disc spaces. Rigidity of fixation in the lumbar spine may aid in the maintenance of lordosis.

Patient outcomes after decompression and instrumented posterior spinal fusion for degenerative spondylolisthesis.

STUDY DESIGN A retrospective study of patient outcomes after decompression and fusion for degenerative spondylolisthesis, using the SF-36 survey and a functional questionnaire. SUMMARY OF BACKGROUND DATA In recent studies, patient outcomes have been examined more specifically; however, detailed functional outcomes are not available nor have widely used outcomes instruments been administered. METHODS Thirty patients aged more than 40 years (average, 60.1 years) who had degenerative spondylolisthesis were evaluated after decompression and instrumented posterior fusion. Charts and radiographs were also reviewed. Questionnaires were administered by telephone, and consisted of the Medical Outcomes Study short form (SF-36) and 27 questions designed to evaluate function, quality of life, medication usage, and satisfaction with surgical results. RESULTS Ninety-three percent of patients were satisfied with their outcomes. Patients improved significantly in their ability to perform heavy and light activities, participate in social activities, sit, and sleep (P < 0.001) and also improved in pain, depression, and medication usage (P < 0.0001). SF-36 data showed significantly better overall assessment of health in all categories than that in a published cohort of patients with low back pain. The current study group also showed no difference in seven of eight categories when compared with the general population. Fusion rate was 93% at an average of 128 days. Three patients required reoperation: two for pseudarthrosis and one for a deep infection. A poorer outcome, scored by the SF-36, was associated with greater preoperative stenosis (P < 0.05) or occurrence of a complication (P < 0.05). CONCLUSIONS Patients treated with decompression and fusion for degenerative spondylolisthesis had improved functional outcomes, when measured by a disease-specific questionnaire and by widely used instruments.

Revision of Failed Lumbar Fusions: A Comparison of Anterior Autograft and Allograft

Study Design. The radiographic and clinical results of two different anterior structural grafts were compared in 38 patients who had combined anterior‐posterior revision surgery for failed lumbar fusion. Objectives. Failed lumbar fusion surgery, such as pseudarthrosis or flatback deformity, may result in disabling pain. The optimum revision technique has yet to be defined. The authors of the current study sought to determine which of two different types of anterior graft yields the best results. Summary of Background Data. Posterior procedures for revision of a failed lumbar fusion have not yielded reliably successful results. A combined anterior‐posterior approach may be effective in restoring sagittal balance and enhancing fusion rates. Recent studies have shown femoral ring allografts to be effective in lumbar fusion revision, but no studies have compared these with other types of structural grafts. Methods. Thirty‐eight patients with pseudarthrosis were treated with combined anterior‐posterior lumbar spine fusion using either femoral ring allografts (26 patients) or tricortical iliac autografts (12 patients). Radiographic follow‐up examination and retrospective patient self‐assessment questionnaires were used to evaluate outcomes. Results were assessed by independent reviewers after a mean follow‐up period of 35 months. Results. Radiographic follow‐up examination revealed acceptably low pseudarthrosis rates for structural autografts (0%) and allografts (6%). The questionnaires revealed significant improvement in pain for both groups. Allograft patients showed greater improvement in function, less pain medication usage, and higher overall success rates (83%) than autograft patients (64%). Conclusions. Femoral ring allografts are as effective, clinically and radiographically, as tricortical iliac autografts when used as an anterior structural element in revision lumbar spine fusion in patients who have undergone multiple surgical procedures for pseudarthrosis or flatback deformity. The slightly greater improvement for the allograft group needs to be confirmed in a larger study.

The use of bone allografts in the spine.

Bone allografts are commonly used in spinal surgery. Structural allografts placed anteriorly in the spine may be used as interbody grafts or as strut grafts spanning multiple segments. Posterior allografts are used to supplement autologous bone for spinal fusions in patients who lack sufficient host bone and to avoid significant donor site morbidity. In this paper the authors review their experience with allograft bone in spine surgery and the results reported in the literature. In the anterior cervical spine, interbody allografts have been used most successfully in single-level fusions. For thoracolumbar deformity, posterior allograft with instrumentation gives satisfactory results in pediatric patients but yields inferior results in adults unless it is combined with an anterior fusion. Fresh-frozen allograft bone has been shown to have higher fusion rates than freeze-dried allograft; ethylene oxide-sterilized allograft has shown uniformly poor results. Structural allografts have been useful for thoracolumbar deformity in both interbody and strut-grafting procedures. In the lumbar spine, allograft has a limited role in posterolateral fusion. For anterior interbody fusions, structural allografts, such as femoral ring allografts, have been used successfully to maintain intervertebral distraction, despite delayed incorporation. Successful use of allograft bone in the spine is dependent on the type of allograft bone used, its anatomic site of fusion, and the age of the patient.

In vivo evaluation of coralline hydroxyapatite and direct current electrical stimulation in lumbar spinal fusion

STUDY DESIGN An animal model of posterolateral intertransverse process lumbar spinal fusion using autologous bone, coralline hydroxyapatite, and/or direct current electrical stimulation. OBJECTIVES To evaluate the effect of an osteoconductive bone graft substitute and direct-current electrical stimulation on the rate of pseudarthrosis in a rabbit spinal fusion model. SUMMARY OF BACKGROUND DATA Conventional techniques for the surgical treatment of degenerative conditions in the lumbar spine have a substantial failure rate and associated morbidity. Bone graft substitutes and electrical stimulation are alternative techniques to enhance fusion rates and limit the morbidity associated with posterolateral intertransverse process fusion using autologous iliac crest bone graft. METHODS Fifty-three adult female New Zealand White rabbits underwent single-level lumbar posterolateral intertransverse process fusion. Animals were assigned to one of four groups using either autologous bone (Group I), coralline hydroxyapatite with autologous bone marrow aspirate (Group II), coralline hydroxyapatite with a 40-microA implantable direct current electrical stimulator and bone marrow aspirate (Group III), or coralline hydroxyapatite with a 100-microA implantable direct current electrical stimulator and bone marrow aspirate (Group IV). Animals were killed at 8 weeks, and fused motion segments were subjected to manual palpation, mechanical testing, and radiographic and histologic analysis to assess the fusion mass. RESULTS Successful fusion was achieved in 57% (8/14) of animals in Group I, 25% (3/12) in Group II, 50% (6/12) in Group III, and 87% (13/15) in Group IV. Mean stiffness and ultimate load to failure were significantly higher in Group IV than in all other groups (P < 0.05). Histologic analysis demonstrated a qualitative increase in fusion mass in Group IV versus all other groups. CONCLUSIONS Direct-current electrical stimulation increased fusion rates in a dose-dependent manner in a rabbit spinal fusion model. Coralline hydroxyapatite is an osteoconductive bone graft substitute, and thus requires an osteoinductive stimulus to ensure reliable fusion rates. Furthermore, coralline hydroxyapatite and direct current electrical stimulation can be used together to increase fusion rates in a rabbit spinal fusion model while avoiding the morbidity associated with harvesting iliac crest bone.

Comparison of methods for determining the presence and extent of anterior lumbar interbody fusion.

Study Design. Titanium alloy interbody fusion devices with autogenous bone were placed in the L5–L6 disc space of 31 adult pig-tailed monkeys through an anterolateral (retroperitoneal) approach. Anteroposterior and lateral radiographs, CT imaging, and histologic analysis of the specimens were performed. Objectives. This study compared the accuracy of plain film radiographs and CT imaging for determining bony fusion of a titanium interbody device implanted in a non-human primate model. The accuracy of the assessments was determined by comparison to histologic analysis. Summary of Background Data. Interbody fusion assessment is often difficult to compare in clinical studies because of differences in definition of fusion criteria. In addition, the accuracy of plain film radiographs and CT imaging assessments of fusion are debated because of device material radiopacity and introduction of artifacts. Methods. A uniform grading system evaluating both the presence and extent of bony fusion was applied to all evaluation techniques. Matched-pair nonparametric t tests were used to determine differences in scoring. Results. The radiographic and histologic presence of fusion grades was equivalent in only 13 of 29 cases (45%), while the CT imaging was equivalent to histologic assessment in 24 of 29 cases (83%). However, the extent of bony fusion in CT imaging and histologic assessment was equivalent in only 4 of 29 cases (14%). Grading of CT images significantly overestimated the extent of fusion. Conclusions. This study demonstrated CT imaging techniques to be superior to plain film radiographs in determining the presence of bony fusion. However, CT imaging did not accurately determine the extent of bony fusion present as confirmed by histologic analysis.

Prevention of postlaminectomy epidural fibrosis using bioelastic materials.

Study Design. The use of elastic protein-based polymers for the prevention of epidural fibrosis following lumbar spine laminectomy was investigated in a rabbit model. Objectives. To determine the safety and efficacy of two bioelastic polymers in matrix and gel forms as interpositional materials in preventing postlaminectomy epidural fibrosis. Summary of Background Data. Postlaminectomy epidural fibrosis complicates revision spine surgery and is implicated in cases of “failed back syndrome.” Materials employed as mechanical barriers to limit tethering of neural elements by the fibrosis tissue have met with little success. A recent family of protein-based polymers, previously reported to prevent postoperative scarring and adhesions, may hold promise in treating this condition. Methods. Sixteen female New Zealand White rabbits underwent laminectomy at L4 and L6. Two polymer compositions, each in membrane and gel forms, were implanted at a randomly assigned level in four rabbits each, with the remaining level serving as an internal control. The animals were killed at 8 weeks, and qualitative and quantitative histology and gross pathologic examination were performed for both the control and the experimental sites to assess the polymers’ efficacy in preventing dorsal epidural fibrosis. Results. The use of the polymers caused no adverse effects. Compared to the control sites, both polymers in either gel or membrane form significantly reduced the formation of epidural fibrosis and its area of contact with the dura postlaminectomy. However, no significant difference in efficacy was detected between either the polymers or their respective forms in preventing epidural fibrosis. Conclusions. The selected compositions of biosynthetic, bioelastic polymers were safe and effective in the limiting the direct contact and consequent tethering of the underlying neural elements by the postlaminectomy epidural fibrosis in rabbits.

In vivo evaluation of calcium sulfate as a bone graft substitute for lumbar spinal fusion

BACKGROUND CONTEXT Posterolateral fusions of the lumbar spine have nonunion rates as high as 35%. The availability of autologous bone to promote fusion is limited, particularly for multilevel fusions. Bone substitutes have been proposed to augment or replace autologous bone for spinal surgery. Calcium sulfate offers high porosity, osteoconductivity, and high resorption rate. This material has been used successfully for treatment of long bone defects but has not been investigated as a bone graft substitute for spinal fusions. PURPOSE To determine whether the use of calcium sulfate granules in conjunction with an implantable electrical stimulator is a safe and effective means of attaining spinal fusion. STUDY DESIGN/SETTING A rabbit lumbar fusion model was used to assess a calcium sulfate bone graft substitute in combination with electrical stimulation for spinal fusion. METHODS Thirty-six adult New Zealand White female rabbits were divided into three groups. Each group underwent a single-level (L5-L6) fusion, receiving 3.0 cc calcium sulfate granules with bone marrow aspirate from the iliac crest. Group 1 had no electrical stimulator applied. Groups 2 and 3 received a 40-microA (Group 2) or a 100-microA (Group 3) implantable electrical stimulator. The animals were sacrificed at 8 weeks, and the rabbit spines were subjected to radiographic assessment, manual palpation, and mechanical testing. RESULTS Two rabbits died postoperatively. The radiographic assessment revealed no fusions occurred at the adjacent nonoperated control levels (L4-L5). There were no fusions observed within Group 1, containing the calcium sulfate and bone marrow aspirate alone. The sites with the implantable stimulators showed a dose-dependent increase in fusion stiffness. However, no fusion mass in Group 2 or 3 was graded as bilaterally complete. CONCLUSION This study found that calcium sulfate as a bone graft substitute was unsuccessful in promoting spine fusion in a rabbit model. There was radiographic evidence of rapid resorption of the calcium sulfate within 4 weeks after surgery. The use of electrical stimulation created a dose-dependent increase in mechanical competence of the bony mass. However, the addition of direct current (DC) current did not significantly alter fusion rates with calcium sulfate used as the bone graft substitute in this model.

Effects of direct current electrical stimulation on gene expression of osteopromotive factors in a posterolateral spinal fusion model.

Study Design. An in vivo model was used to determine levels of mRNA expression in response to direct current (DC) electrical stimulation in a rabbit posterolateral fusion model. Objectives. This study tested the possibility that DC stimulation at the surgery site would increase expression of genes related to bone formation relative to expression in autograft alone. Summary of Background Data. DC electrical stimulation as an adjunct treatment in spinal surgery has shown increased fusion rates when compared with autograft alone, yet the biology of such treatment is not fully understood. Methods. Thirty New Zealand White rabbits were entered into the study. A posterolateral, intertransverse process fusion was performed bilaterally at L4–L5, with autogenous bone graft. An implantable DC stimulator was placed across the decorticated transverse processes before placement of autograft. Animals were killed at 3, 7, 14, 21, and 28 days. mRNA levels of BMP-2, 4, 6, 7, VEGF, FGF-2, TGF-β, ALK-2, and ALK-3 were evaluated with real-time RT-PCR. Results. mRNA expression was significantly higher in the DC stimulated animals versus the control animals for several of the genes studied. In particular, levels of mRNA were elevated for BMP-2, BMP-6, and BMP-7. Conclusions. This study shows for the first time that DC stimulation results in a sustained increase of multiple osteogenic genes, suggesting that the biologic mechanism for the DC-induced increase in the rate and extent of bone formation observed clinically may be mediated by the up-regulation of these osteoinductive factors.