Secondary Logo

Journal Logo

Surgical Decision-Making

Instrumentation-Related Complications of Multilevel Fusions for Adult Spinal Deformity Patients Over Age 65

Surgical Considerations and Treatment Options in Patients With Poor Bone Quality

DeWald, Christopher J. MD; Stanley, Thomas MD

Author Information
doi: 10.1097/01.brs.0000236893.65878.39
  • Free

Multilevel spinal instrumented constructs are frequently necessary for the treatment of adult scoliosis. The effect of increasing segmental fixation has been shown to increase the stiffness of the overall construct; however, multilevel fusions can have detrimental effects on levels adjacent to the construct. Although often assumed or anecdotal, we are unaware of clinical data showing the failure rates of multilevel spinal fixation in spinal deformity patients with poor bone stock. The purpose of this study is to review our perioperative experience with multilevel segmental fixation in patients with poor bone quality and to review the surgical techniques used in these patients.

Osteoporosis contributes to the deformity seen in degenerative scoliosis and senile spinal deformities.1–3 Spinal fixation in osteoporotic bone is becoming more frequent as the population in the U.S. ages. Osteoporosis is a common metabolic disease affecting approximately 30% of postmenopausal white females with a 15% lifetime risk of vertebral fracture.5–7 In the United States alone, there are 700,000 osteoporotic fractures of the spine per year.32 Furthermore, the incidence of vertebral fractures in men approaches that of women, contrary to the more traditional belief that osteoporosis is primarily a disease of women.9

There are multiple risk factors obtained from the medical history that can suggest poor bone stock. The National Osteoporosis Foundation clinical guidelines identify the following factors: history of fracture as an adult or fracture in a first-degree relative, white race, advanced age, smoking, low body weight, female gender, dementia, poor health, or fragility.10 In the presence of these risk factors, particularly with a history of previous fracture, a preoperative bone density studies should be considered. This will not only assist in the surgical decision-making but also for medical management of osteoporosis.

Spinal deformity and adult scoliosis can be associated with decreased bone density. Compression fractures in apical vertebra or adjacent to a curve can cause progression of a preexisting scoliosis. Scoliotic deformities are present in 36% to 48% of osteoporotic women. De novo degenerative scoliosis deformities can develop from fractured vertebrae and spinal collapse leading to spinal stenosis and mechanical back pain. Adult scoliosis patients with large spinal deformities typically present with poorer bone stock. Furthermore, patients with previous fusions can develop adjacent level degeneration and kyphosis. Adult scoliosis patients that develop progressive thoracolumbar kyphosis along with their scoliosis have been noted to develop pulmonary deterioration.3

The bone quality of the spine plays an important role in the planning of surgical intervention for spinal deformities. The indications for instrumentation include severe mechanical pain due to spinal imbalance, spinal instability, multilevel spinal stenosis, and neurologic deterioration (i.e., senile burst fracture). Techniques in instrumenting a spine with poor bone qualities require multiple points of fixation above and below the apex of the spinal deformity, the use of iliac fixation, and achieving a stable and balanced spine accepting modest correction.35

Multiple techniques to improve fixation in the osteoporotic spine have been developed. These include sublaminar wiring, cement augmentation of pedicle screw fixation, pediculolaminar hook fixation, conical screws, hydroxyapatite-coated screws, and expandable screws. Pedicle screw fixation alone has been shown to be less effective in osteoporotic spines by multiple studies.11–13 Analysis of failure mechanisms by Zindrick et al28 and Law et al27 suggest that translational motion causes a “wind-shield wiper” effect and loosening of the screws. Although there is no definite consensus, these studies would suggest that cement augmentation is the most likely to resist failure. This study will highlight instrumentation failures of long constructs in patients over 65 in an attempt to direct methods of spinal instrumentation in spinal deformity patients with decreased bone quality.

Methods

We retrospectively reviewed case logs from a single adult spinal deformity practice over a 5-year period. Inclusion criteria included all patients over the age of 65 who underwent instrumented posterior spinal stabilization of at least 5 levels. Of this series no patients were excluded. All the operative reports were obtained to search for specific techniques used to supplement pedicle screw fixation. Postoperative clinic notes were then reviewed for both early and delayed complications. Complications were defined as early if they occurred within the first 3 months after surgery, and late beyond 3 months. Complication rates for early failure were calculated with respect to the overall number of procedures, whereas late complications were calculated with respect to the number of patients.

Results

A total of 47 procedures in 38 patients met the inclusion criteria. The group consisted of 16 males, and 22 females. The average age at the time of procedure was 72.4 years. Underlying diagnoses include idiopathic scoliosis, adult degenerative scoliosis, and ankylosing spondylitis. Twenty-six cases were primary instrumented stabilizations. Twenty-one cases involved extension or revision of a previous fusion. The average follow-up period was 30 months with a range from 3 to 116 for early failure analysis. The average number of levels fused was 9 with a range of 5 to 17. There was 1 case in which cement was used to augment screw fixation. Iliac screws were used in 12 cases. Laminar hooks or transverse process hooks were used at the end of the spinal construct in 7 of the cases. In 12 cases, an anterior release was performed before posterior stabilization, and in 10 cases a pedicle subtraction osteotomy was performed.

Early Instrumentation Complications

Of the 47 procedures there were six early postoperative complications related to instrumentation, or a rate of 13%. The average time to early complication was 7 weeks, with a range of 1.5 to 12 weeks. One patient developed an epidural hematoma secondary to a pedicle fracture at the cephalad end of her fusion. This complication occurred on postoperative day 11 and presented as with acute onset of pain and a unilateral foot drop. Her initial procedure was a multilevel laminectomy and spinal fusion from T11 to the sacrum with iliac screws. She was revised with laminectomy, hematoma evacuation, and extension of her fusion to T9, resulting in full recovery of neurologic function. A second early complication involved a mild compression fracture of the adjacent cephalad vertebral body following an L1–S1 fusion. The patient was managed conservatively and did not require additional surgery.

Three patients developed compression fractures of the most cephalad instrumented vertebral body. The first was initially fused from T12 to the sacrum with the compression fracture occurring at T12. This patient was treated with extension of her fusion to the proximal thoracic spine. This patient then sustained another compression fracture the proximally instrumented level that did not progress and did not require additional surgery. The second patient developed a compression fracture of T11 following a fusion from T11 to the sacrum with iliac screws. This patient required extension of their spinal instrumentation to T2. Two other patients developed compression fractures of the adjacent vertebrae to the last instrumented level including 1 patient developing multiple mild contiguous compression fractures of T8–T10 following a fusion from T11–pelvis (Table 1).

Table 1
Table 1:
Early and Late Instrumentation Complications

Late Instrumentation Complications

Of the 38 patients included in the study, 4 patients developed pseudoarthroses with rod breakage requiring revision surgery, a rate of 11%. Two patients (7%) had loosening of the pedicle screws of the last instrumented level indicated by lucencies about the screw. One occurred at T10 at the top of an 8-level construct, and the other occurred at the S1 pedicle screws at the bottom of a 6-level construct without iliac fixation. One patient (4%) developed an acute disc herniation at the disc space directly above the construct at 8 months. One of the 9 patients (11%) with pelvic fixation had prominent iliac screws were treated with surgical removal of the screws 2 years postoperatively. Only 1 of 28 patients (4%) had painful instrumentation requiring removal of her 8-level instrumentation construct. All patients’ radiographs showed a solid fusion mass at the time of most recent follow-up.

Ten of 38 patients (26%) developed progressive junctional kyphosis at the cephalad extent of the construct. The construct lengths in these patients ranged from 6 to 8 levels, and the terminal cephalad level ranged from T10 to L1. Two developed late compression fractures of the most cephalad vertebral body included in the construct leading to kyphosis, and 1 patient developed a compression fracture 2 vertebral bodies cephalad from the last instrumented vertebrae. Six of these patients did not have radiographic evidence of any compression fractures but still seemed to continue to become progressively kyphotic. Overall compression fractures of the last instrumented vertebrae occurred in 5 patients, and adjacent levels in 3 patients (Table 1).

Discussion

In this series, we delineated early and late complications to attempt to find a specific mechanism of failure that could be addressed intraoperatively in older patients or patients with poorer bone quality. The two most common mechanisms of failure from this series were failure of the cephalad or last instrumented vertebral level with fracture or late screw loosening, and late progressive kyphosis cephalad to the spinal construct. Vertebral body failure was thought to be due to disruption of the trabecular bone by pedicle screw fixation plus the additional stresses seen following instrumentation of the lumbar spine. Augmentation or adjunctive fixation of the most cephalad vertebral body has been suggested to be effective at preventing late screw loosening or cutout. In this series, adjunctive fixation of the cephalad most vertebral body did not always prevent adjacent level fracture or late progressive kyphosis presumably due to transition of forces to the adjacent cephalad levels.

Late progressive kyphosis does not seem to be preventable with instrumentation as 32% of our patients developed this complication. Extension of the fusion to more proximal balanced vertebral levels within the thoracolumbar junction did not prevent progressive kyphosis above that level. Although only 1 of 10 patients required reoperation for progressive kyphosis without an associated fracture, longer term follow-up is needed to confirm any conclusions. A potential approach to this problem is to perform limited fusions with the intention of staging proximal extension as the junctional kyphosis progresses.

Assessment and treatment of the patient’s bone quality include a general assessment of the patient’s plain radiographs based on a subjective review. The standard method of measuring bone density is via a dual-energy radiograph absorptiometry (DEXA) scan. DEXA scans are typically measured from the spine, hip, or radius. Patients are given a T-score indicating the range of bone density compared to a young adult woman. The normal range is considered a T-score less than 1.0 standard deviation from the mean of young adult women. T-scores between 1.0 and 2.5 standard deviations below the mean are considered in the osteopenic range. T-scores less than 2.5 standard deviations below the mean are considered osteoporotic.

The workup of a diagnosis of osteoporosis includes the exclusion of other causes of decreased bone density including infectious, malignant, or metabolic diseases, especially bone marrow diseases such as multiple myeloma. Secondary causes resulting in decreased bone density must be excluded, including osteomalacia and alcoholism. The secondary causes of poor bone density require treatments in addition to the recommended osteoporosis medical management. Standard laboratory workup includes serum white blood count, erythrocyte sedimentation rate, urinalysis, urine for the collagen breakdown product, N-telopeptide, BUN/Cr, serum protein electrophoresis, serum calcium and phosphorous, alkaline phosphatase, and serum thyroid and liver function tests.33

Medical management of osteoporosis is effective at both preventing fractures and improving bone density. Improvements in bone density can be seen as early as 6 months after the initiation of treatment, with both hormone replacement therapy and bisphosphonates.7 The role of preoperative medical management to improve bone stock must be weighed against disability caused by delayed treatment, since decreased activity has been shown to decrease bone density.10 Furthermore, treatment of osteoporosis has only been shown to increase bone density by 1% to 2% per year.29,30 The clinical significance in the operative setting of these incremental increases has not been determined.

Surgical Treatment

Spinal surgery on adults with spinal deformities can be a significant challenge to the spinal surgeon, especially when using spinal instrumentation. Typically, spinal deformities require extensive spinal instrumentation constructs and prolonged surgeries. The bone quality of the patient is a major factor to the success of the surgical procedure. Older patients also have multiple comorbidities, including osteoporosis, making them at a higher risk for perioperative complications. Careful preoperative planning is necessary to shorten the overall surgical time and perioperative complications.

Planning includes a clear objective of whether a large or a smaller surgical procedure is required for the defined patient. If an extensive spinal surgery is necessary but the patients’ medical condition does permit this approach, then not operating may be the better option. Extensive decompression alone in an adult with a scoliosis is contraindicated as this can lead to further destabilization and worsening of their deformity. On occasion, a single nerve root can be decompressed in an older or medically debilitated patient without leading to an unstable deformity. Additionally a small 1- or 2-level instrumentation may be required to stabilize a decompressed area without instrumenting the entire deformity. This typically can be achieved with smaller deformities or below the apex of the curve or within the fractional curve of the lower lumbar spine.

Occasionally, the spinal surgeon is faced with the task of instrumenting a spine with poor bone quality. The indications for instrumentation include severe mechanical back pain due to spinal imbalance, spinal instability, multilevel spinal stenosis, and neurologic deterioration (i.e., senile burst fracture).

When spinal instrumentation is required in a patient with osteoporosis, the spinal surgeon needs to use certain techniques to improve on the success of the procedure. Techniques in instrumenting a spine with poor bone quality require multiple points of fixation, including adjunct fixation and/or augmentation of pedicle screws. Multiple fixation points should include three sets of fixation points above and below the deformity’s apex. The use of bilateral iliac screws to anchor the spinal instrumentation in long constructs has greatly added to the strength of spinal instrumentation constructs and should be considered another site of fixation.33 A stable balanced spine should be the overall goal of the spinal surgery accepting a modest correction. A lesser correction will lessen the stresses seen on the instrumentation protecting against instrumentation pullout and disaster.35 The surgeon should plan for a limited correction fixing the spine “as is” on the spinal surgery frame. Moderate correction can be obtained simply by placing the patient on the spinal frame in the prone position under anesthesia as many of the degenerative de novo deformities seen in older patients have flexible deformities. Often when rigidity persists, generous facetectomies will allow moderate correction without the need for a formal anterior release.

Generally, it is better to instrument longer initially, if medically appropriate, to decrease the risk of a salvage revision surgery in a nutritionally depleted patient. However, if the medical condition prohibits an extensive surgery, a staged or delayed staged surgery can be undertaken. Instrumentation stopping at T11, for example, can be extended up to T2 1 week or 3 months after the initial surgery as a staged procedure. If an anterior interbody release or strut support is required in addition to the larger posterior procedure, it too can be staged. Parenteral nutrition can be considered when staging these spinal procedures in the adult.

It is important not to end the spinal construct within the kyphotic portion of the spine to prevent “falloff.” In our series, the most common long-term complication was progressive kyphosis. Instead, the instrumentation should end caudal to or extending above the kyphotic portion of the spine typically to the upper thoracic vertebrae.35 Anterior-posterior spinal surgery procedures should be considered when large anterior interbody defects are created when correcting the lumbar loss of lordosis. When possible, the surgeon should attempt to balance the spinal construct between the cephalad and caudal ends. Having five sites of fixation in the cephalad portion of the construct does not improve the caudal fixation when only two sites of fixation have been used.

Multiple Fixation Points

Multiple fixation sites are needed both at the cephalad and caudad aspect of a spinal construct in a patient with poor bone quality. The cephalad end of spinal instrumentation should end above the kyphotic aspect of the spine, whether at the thoracolumbar junction or above the natural thoracic kyphosis. This is most important in a kyphotic spinal deformity. The three sets of fixation required include either laminar hooks in a claw configuration or three sets of pedicle screws. It is much easier to obtain more fixation sites using multiple screws versus hooks. Adjunctive fixation to the end pedicle improves the pullout strength. This can be the addition of sublaminar wires or supralaminar hook at the top most instrumented vertebrae (Figure 1).

Figure 1
Figure 1:
AP (A) and lateral (B) radiographs of the cephalad spinal fixation using adjunct supralaminar claws.

At the caudal end of a deformity construct, the surgeon should also plan for three sets of fixation below the deformity, including pedicle screws or laminar hooks. Supralaminar-infralaminar claws can be used but need to be placed at every other segment due to the crowding of the implants. Pedicle screws have become the standard mode of instrumentation in the lumbar spine to obtain multiple sites of fixation. As with the cephalad end of a spinal deformity construct adjunctive fixation with sublaminar wires or infralaminar hooks will improve the pullout strength of the most caudal pedicle screws. With extension of the instrumentation to sacrum, improved pullout strength can be obtained with bicortical S1 screws. These screws should angle toward the sacral promontory. In our experience, it is better to be slightly within the disc space versus being distal to the sacral endplate. The addition of iliac screws has greatly increased the protection and pullout strength of the sacral screws.34

Coe et al noted that laminar bone was stronger than the cancellous bone of the vertebral body.25 Laminar hooks placed in a claw configuration was the standard of fixation in spines with poor bone quality in the past until pedicle screws became the standard. Claws can be configured as a laminar hook-pedicle hook claw or as a laminar hook-pedicle screw claw (Figure 2). A combination of implants provides a spread of stresses seen by the bone. Halverson et al determined that adding an adjunctive laminar hook improved the pullout strength of a pedicle screw.13 Sublaminar wires have for years been the mainstay for achieving multiple fixation points in neuromuscular or osteoporotic spinal constructs, and still provide excellent adjunctive fixation to the instrumented vertebrae. However, cement augmentation of pedicle screws has been shown to provide the greatest pullout strength.15 Augmentation has been clinically used widely using PMMA and less commonly using a calcium-phosphate compound.16,18–21 Most reports have used 2 to 3 mL of PMMA per screw with C-arm fluoroscopy confirmation of the placement of the cement to ensure that the cement does not flow posterior into the spinal canal. The advantage of the calcium phosphate cement is that it resorbs and is replaced with bone whereas the PMMA is permanent substance (Figure 3).

Figure 2
Figure 2:
AP radiograph of a multiple site fixation using transverse process–pedicle hook claws in the thoracic spine and pedicle screws in the lumbar spine. Note the adjunct infralaminar hooks below the last instrumented level.
Figure 3
Figure 3:
AP (A) and lateral (B) radiographs using cement augmentation for the cephalad fixation at the thoracolumbar junction and iliac screws to protect the caudal instrumentation.

The addition of adjunctive fixation whether wire, hook, or cement augmentation has added to the strength of a pedicle screw. In 2004, Tan et al separated three groups of pedicle screws augmented with different mechanisms: pedicle screw/laminar hook; pedicle screw/sublaminar wires; pedicle screw/calcium phosphate cement.26 All three groups were noted to increase the rigidity of fixation by improving the pullout strength of the pedicle screws. The improved results were similar among all three groups. Their conclusion was that any augmentation or adjunctive fixation provided better fixation strength over a pedicle screw alone. The strongest fixation was a combination of a cement-augmented screw and an adjunctive fixation of either a hook or a sublaminar wire.

The optimum pedicle screw size and shape have been debated without definitive conclusion being made. Zindrick et al demonstrated using a cadaver model that improved insertional torque resulted in greater pullout strength.28 Our experience has shown that nontapered screws with a conical inner core have an improved insertional torque. Larger diameter screws have been shown to have improved pullout strength with better “fill” within the pedicle. However, 24% to 40% of pedicle screws placed in an osteoporotic pedicle have been noted to cause a fracture of the pedicle. Zindrick et al28 also noted that the length of the screw improved pullout strength, particularly when the screws were bicortical. Same diameter tapping of screws has been shown to decrease the insertional torque and, thus, decrease the pullout strength of the pedicle screw. Thus, it has been recommended that either the surgeon should undertap the path of the pedicle screw or not tap the screw hole at all.

When an anterior release is required or anterior lumbar support is required, this should be performed for support and fusion versus for a complete correction. Stiff struts such as metal cages can cut into vertebral endplates and result in subsidence and loss of lordosis. PEEK cages or carbon fiber cages more closely match the modulus of bone and have been noted to have less subsidence or endplate cutout. Often when a stiff kyphotic deformity is noted in the osteoporotic spine, a pedicle subtraction osteotomy is a better option to obtain improved sagittal balance without performing an anterior-posterior surgical procedure. Pedicle subtraction osteotomies can provide a greater degree of sagittal alignment correction from a single approach.

When using a pedicle screw construct in patients with poorer bone quality, many questions still remain. What is the ideal rod diameter or rod stiffness for osteoporotic spine constructs? Does a smaller rod decrease the bone-implant interface? Are crosslinks necessary?

Proximal Thoracic

Specific anatomic areas of concern for instrumentation cutout are located at the proximal thoracic spine, and the junctions of the thoracolumbar and lumbosacral spine. In the proximal thoracic region, the severity and location of the patients thoracic kyphosis determine the fixation used. Again, a minimum of three sets of fixation points using hooks in claw configurations are recommended to prevent pullout of the proximal fixation. Thoracic pedicle screws have become more commonly used in the thoracic spine and allow for more site of fixation than hook fixation. Fixation to the proximal thoracic spine should be extended to T1 or T2. Our experience has led to using a combination of a supralaminar hooks as adjunctive fixation at the vertebrae cephalad to the topmost pedicle screw, thus improving the end fixation pullout strength (Figure 1).

Thoracolumbar Spine

Instrumentation in a degenerative lumbar spinal deformity often requires extension cephalad into the thoracolumbar spine. The degenerative lumbar vertebrae typically have become arthritic and sclerotic in the degenerative environment and may provide improved fixation despite an osteopenic patient. The lower thoracic and thoracolumbar vertebrae cephalad to the degenerative segment do not become as sclerotic and arthritic as the lumbar vertebrae. Pedicle screw fixation in these relatively osteopenic vertebrae has decreased resistance to pullout. Instrumentation in these vertebrae also leads to increased stresses due to a longer construct. These most critical vertebrae in a lumbar spinal deformity have poorer fixation than the midapex vertebrae and thus require special attention to its fixation. Augmentation of the most cephalad pedicle screw with cement has decreased same level compression fractures and subsequent endplate cutout of the pedicle screw. Screw augmentation with adjunct fixation with either supralaminar hooks or sublaminar wires may improve pedicle screw pullout but did not appear to decrease same level vertebrae compression fractures (Cases 1 and 2).

Case 1. Lateral radiographs (A) of a 77-year-old woman with severe low back pain and neurogenic claudication after undergoing a multilevel decompression and spinal fusion. Although clinically doing well, she has developed a compression fracture of the last instrumented vertebral level at T12. (B) Follow-up radiographs show a progressive kyphosis due to this fracture with cutout of the screws into the disc space. (C and D) Extension of the posterior instrumentation provides partial correction of the deformity.
Case 2. Preoperative (A) and postoperative (B) lateral radiographs of a rheumatoid arthritis patient who had undergone an anteroposterior stabilization of a two-level contiguous compression fracture. (C) Lateral radiograph demonstrating failure of the caudal spinal construct with a distraction fracture through the pedicle of the last instrumented level. (D) This patient was revised with closure of the distraction injury using infralaminar hooks and cement augmentation of the caudal pedicle screws.

It is not known which level to stop the cephalad extension of a lumbar spinal deformity. Many surgeons may extend past the thoracolumbar junction with fixation to T12. However, further extension to T11 or T10 provide improved fixation by further extending beyond the thoracolumbar junction, as well as, including vertebrae with a more substantial rib attachment adding spinal construct support. It is important to stop the cephalad instrumentation within the sagittal plumb zone decreasing the forward pullout stresses. Additionally, hip flexion contractures can decrease the patient’s ability to extend the sagittal plumb line behind the hip joints. It may be necessary to address the patient’s hip pathologies before planning any surgical correction of an osteoporotic spinal deformity.

Lumbosacral Spine

The lumbosacral spine fixation is a particular difficult region to instrument due to the added stresses of fixation to the stable pelvis. Improved fixation with increased S1 pullout resistance using bilateral iliac screws has decreased S1 screw failures.13,16,33 Additional strength is achieved by adding L5–S1 anterior column support with an interbody strut graft or cage. As with the cephalad instrumentation, three sets of fixation can be used at the lumbosacral junction, including bilateral bicortical S1 pedicle screws angled to the sacral promontory and bilateral iliac screws. Iliac screws have been manufactured in different configurations from threaded screws to bolts with only minimal thread. Iliac screws have been shown to have greater strength with larger diameter screws with lengths from 70 to 80 mm (Figure 3).

Too Osteoporotic

The largest unanswered question is: when is a spinal deformity too osteoporotic for surgical intervention? Although this remains largely unknown, generalized suggestions have been made. There has been no documented evidence to recommend a specific or ultimate T-score for which surgical instrumentation should not be performed. In general, it is a combination of multiple factors that help the spinal surgeon to make this determination. These factors include the patients’ age/health, gross inspection of bone quality on the radiographs, DEXA score, the magnitude of the deformity, the length of the anticipated surgery, the presence or absence of neurologic deficit, and whether there was a need for anterior spinal surgery.

Multiple fixation sites, augmentation of pedicle screws, and use of iliac screws have significantly improved the spinal surgeon’s ability to instrument the osteoporotic spine. It is possible to achieve stable fixation in osteoporotic bone for immediate stability. Nonetheless, high risks of complications are seen in patients with poor bone stock. The spinal surgeon must use proper preoperative planning and bone density assessment to determine the degree of correction, and the levels to be instrumented. Unfortunately, long-term problems continue for these patients despite successful instrumentation, particularly progressive thoracic kyphosis. Recognizing this problem allows us to educate patients with regard to long-term results.

Key Points

  • Progressive kyphosis is an inevitable consequence of multilevel instrumentation in patients with poor bone stock.
  • Planning for incomplete correction of spinal deformity should be considered in the setting of osteoporosis.
  • Important surgical techniques for instrumentation in patients with poor bone stock include multilevel fixation and iliac screws.
  • Augmentation of caudal and cephalad vertebral fixation should be routinely used to prevent late instrumentation failure.
  • Cement augmentation of the cephalad vertebral body can help prevent common mechanisms of failure in multilevel fixation.

References

1.Grubb SA, Lipscomb HJ, Coonrad RW. Degenerative adult onset scoliosis. Spine 1988;13:241–5.
2.Vanderpool DW, James JIP, Wynne-Davies R. Scoliosis in the elderly. J Bone Joint Surg Am 1969;51:446–55.
3.Healy JH, Lane JM. Structural scoliosis in osteoporotic women. Clin Orthop 1985;195:216–23.
4.McAfee PC, Farey ID, Sutterlin CE, et al. Device-related osteoporosis with spinal instrumentation. Spine 1989;14:919–25.
5.Melton LJ III. How many women have osteoporosis now? J Bone Miner Res 1995;10:175–7.
6.Dennison E, Cooper C. Epidemiology of osteoporotic fractures. Horm Res 2000;54(suppl 1):58–63.
7.Lane JM, Russell L, Khan SN. Osteoporosis. Clin Orthop 2004;425: 126–34.
8.Glassman SD, Alegre GM. Adult spinal deformity in the osteoporotic spine: options and pitfalls. Instr Course Lect 2003;52:579–88.
9.Melton LJ III. Epidemiology of spinal osteoporosis. Spine 1997;22(suppl 24):2–11.
10.Heineman DF. Osteoporosis: an overview of the National Osteoporosis Foundation Clinical Practice Guide. Geriatrics 2000;55:31–6.
11.Soshi S, Shiba R, Hidemaru K, et al. An experimental study on transpedicular screw fixation in relation to osteoporosis of the lumbar spine. Spine 1991;16:1335–41.
12.Okuyama K, Sato K, Abe E, et al. Stability of transpedicle screwing for the osteoporotic spine.
13.Halvorson TL, Kelley LA, Thomas KA, et al. Effects of bone mineral density on pedicle screw fixation. Spine 1994;19:2415–20.
14.Cook SD, Salkeld SL, Stanley T, et al. Biomechanical study of pedicle screw fixation in severely osteoporotic bone. Spine J 2004;4:402–8.
    15.Sarzier JS, Evans AJ, Cahill DW. Increase pedicle screw pullout strength with vertebroplasty augmentation in osteoporotic spines. J Neurosurg 2002;96:309–12.
    16.Turner AWL, Gillies RM, Svehla MJ, et al. Hydroxyapatite composite resin cement augmentation of pedicle screw fixation. Clin Orthop 2003;406: 253–61.
    17.McKoy BE, An YH. An expandable anchor for fixation in osteoporotic bone. J Orthop Res 2001;19:545–7.
    18.Rohmiller MT, Schwalm D, Glattes RC, et al. Evaluation of calcium sulfate paste for augmentation of lumbar pedicle screw pullout strength. Spine J 2002;2:255–60.
    19.Wuisman, PIJM, Van Dijk, M, Staal H, et al. Augmentation of (pedicle) screws with calcium apatite cement in patients with severe progressive osteoporotic spinal deformities: an innovative technique. Eur Spine J 2000;9:528–33.
    20.Renner SM, Lim T, Kim W, et al. Augmentation of pedicle screw fixation strength using an injectable calcium phosphate cement as a function of injection timing and method. Spine 2004;29:212–6.
    21.Spivak JM, Neuwirth MG, Labiak JJ, et al. Hydroxyapatite enhancement of posterior spinal instrumentation fixation. Spine 1994;19:955–64.
    22.Hilibrand AS, Moore DC, Graziano GP. The role of pediculolaminar fixation in compromised pedicle bone. Spine 1996;21:445–51.
    23.Hasegawa K, Takahashi HE, Seiji U, et al. An experimental study of a combination method using a pedicle screw and laminar hook for the osteoporotic spine. Spine 1997;22:958–63.
    24.Wittenberg RH, Lee K, Shea M, et al. Effect of screw diameter, insertion technique, and bone cement augmentation of pedicular screw fixation strength. Clin Orthop 1993;296:278–87.
    25.Coe JD, Warden KE, Herzig MA, et al. Influence of bone mineral density on the fixation of thoracolumbar implants.
    26.Tan J, Kwon B, Dvorak MF, et al. Pedicle screw motion in the osteoporotic spine after augmentation with laminar hooks, sublaminar wires, or calcium phosphate cement: a comparative analysis. Spine 2004;29:1723–30.
    27.Law M, Tencer AF, Anderson PA. Caudo-cephalad loading of pedicle screws: mechanisms of loosening and methods of augmentation. Spine 1993;18:2438–43.
    28.Zindrick MR, Wiltse LL, Widell EH. A biomechanical study of intrapedicular screw fixation in the lumbosacral spine. Clin Orthop 1986;203:99–112.
    29.Chesnut CH III, Silverman S, Andriano K, et al. A randomized trial of nasal spray salmon calcitonin in postmenopausal women with established osteoporosis: the Prevent Recurrence of Osteoporotic Fractures Study. PROOF Study Group. Am J Med 2000;109:267–76.
    30.Reid IR, Brown JP, Burckhardt P, et al. Intravenous zoledronic acid in postmenopausal women with low bone mineral density. N Engl J Med 2002;346:653–61.
    31.Brodke DS, Bachus KN, Mohr RA, et al. Segmental pedicle screw fixation or cross-links in multilevel lumbar constructs: a biomechanical analysis. Spine J 2001;1:373–9.
    32.Riggs BL, Melton LJ III. The worldwide problem of osteoporosis: insights afforded by epidemiology. Bone 1995;17(suppl):505–11.
    33.Dawson EG, DeWald CJ, Nattiv A. Osteoporosis: evaluation and pharmacologic treatment. In: Spinal Deformities: The Comprehensive Text. New York: Thieme, 2003:19.
    34.Tsuchiya K, Bridwell KH, Kuklo TR, et al. Minimum 5-yr analysis of L5–S1 fusion using sacropelvic fixation for spinal deformity. Spine 2006;31:303–8.
    35.Hu SS. Internal fixation of the osteoporotic spine. Spine 1997;245(suppl):43–8.
    Keywords:

    adult scoliosis; osteoporosis; spinal deformity; instrumented spinal fusion; instrumentation failure

    © 2006 Lippincott Williams & Wilkins, Inc.