Preliminary biomechanical evaluation of prophylactic vertebral reinforcement adjacent to vertebroplasty under cyclic loading

The Spine Journal 9 (2009) 174–181

Preliminary biomechanical evaluation of prophylactic vertebral reinforcement adjacent to vertebroplasty under cyclic loading Robert J. Oakland, PhDa, Navin R. Furtado, MRCSa, Ruth K. Wilcox, PhDa, Jake Timothy, FRCSb, Richard M. Hall, PhDa,c,* a

School of Mechanical Engineering, University of Leeds, Leeds LS2 9JT, UK Department of Neurosurgery, Leeds General Infirmary, Leeds LS1 3EX, UK c Academic Unit of Orthopaedic Surgery, Leeds General Infirmary, Leeds LS2 9NS, UK b

Received 19 December 2007; accepted 19 May 2008

Abstract

BACKGROUND CONTEXT: Percutaneous vertebroplasty has become a favored treatment option for reducing pain in osteoporotic patients with vertebral compression fractures (VCFs). Short-term results are promising, although longer-term complications may arise from accelerated failure of the adjacent vertebral body. PURPOSE: To provide a preliminary biomechanical assessment of prophylactic vertebral reinforcement adjacent to vertebroplasty using a three-vertebra cadaveric segment under dynamic loads that represent increasing activity demands. In addition, the effects of reducing the elastic modulus of the cement used in the intact vertebrae were also assessed. STUDY DESIGN/SETTING: Three-vertebra cadaveric segments were used to evaluate vertebroplasty with adjacent vertebral reinforcement as an intervention for VCFs. METHODS: Nine human three-vertebra segments (T12–L2) were prepared and a compression fracture was generated in the superior vertebrae. Vertebroplasty was performed on the fractured T12 vertebra. Subsequently, the adjacent intact L1 vertebra was prophylactically augmented with cement of differing elastic moduli (100–12.5% modulus of the base cement value). After subfailure quasi-static compression tests before and after augmentation, these specimens were subjected to an incrementally increasing dynamic load profile in proportion to patient body weight (BW) to assess the fatigue properties of the construct. Quantitative computed tomography assessments were conducted at several stages in the experimental process to evaluate the vertebral condition and quantify the gross dimensions of the segment. RESULTS: No significant difference in construct stiffness was found pre– or postaugmentation (t51.4, p5.19). Displacement plots recorded during dynamic loading showed little evidence of fracture under normal physiological loads or moderate activity (1–2.5 BW). A third of the specimens continued to endure increasing load demands and were confirmed to have no fracture after testing. In six specimens, however, greater loads induced 11 fractures: 7 in the augmented vertebra (2T12, 5L5) and 4 in the adjacent L2 vertebra. A strong correlation was observed between the subsidence in the segmental unit and the incidence of fracture after testing (rSpearman’s5 0.88, p5.002). Altering the modulus of cement in the intact vertebra had no effect on level of segmental compromise. CONCLUSIONS: These preliminary findings suggest that under normal physiological loads associated with moderate physical activity, prophylactic augmentation adjacent to vertebroplasty showed little evidence of inducing fractures, although loads representing more strenuous activities may generate adjacent and peri-augmentation compromise. Reducing the elastic modulus of the cement in the adjacent intact vertebrae appeared to have no significant effect on the incidence or

FDA device/drug status: Not applicable. Nothing of value received from a commercial entity related to this article. Sources of Support: Funding for this work was received from Action Medical Research (Grant Ref: AP1028), Rosetrees Trust, EPSRC, and the Yorkshire Children’s Spine Foundation. The authors gratefully acknowledge the contribution of DePuy CMW who provided the cement. 1529-9430/09/$ – see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.spinee.2008.05.009

Ethics Approval was received from Leeds East Research Ethics Committee, St. James University Hospital, Leeds, UK (06/Q1206/149). * Corresponding author. School of Mechanical Engineering, University of Leeds, Leeds LS2 9JT, UK. Tel.: (44) 113-343-2132; fax: (44) 113-2424611. E-mail address: [email protected] (R.M. Hall)

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181

175

location of the induced fracture or the overall height loss of the vertebral segment. Ó 2009 Elsevier Inc. All rights reserved. Keywords:

Biomechanics; Osteoporosis; Vertebroplasty; Cement; Compression fracture; Cyclic loading

Introduction Percutaneous vertebroplasty (PVP) has become a treatment option for the management of painful osteoporotic vertebral compression fracture (VCF) that does not respond to more conservative interventions [1–3]. The procedure involves the injection of bone cement into the compromised vertebral body (VB) with the primary aim of reducing pain through stabilization of the fracture site. Patients have reported rapid pain reduction within 1 to 2 days of the intervention [4–7] with a return of lost function [8]. The incidence of serious complication is low [9]. There are, however, significant concerns surrounding the longer-term effects of the treatment. In particular, several studies have suggested that there may be an associated increased risk of fracture in the vertebrae adjacent to the PVP augmentation site [10–12]. This observation has been supported by in vitro experiments which have demonstrated a reduction in the failure strength of the adjacent, nonaugmented vertebrae [13,14]. Finite element studies have also suggested that PVP causes an altered load distribution within the augmented segment of the vertebral column [15–17]. In addition, augmentation of the VB without kyphotic angle restoration after a VCF is thought to be a factor in causing adjacent vertebral failure [18]. Alternatively, these accelerated adjacent vertebral fractures may arise from the clustered nature of these skeletal-related events that have been observed to occur adjacent to a VCF because of a regional weakening of the bone [19–21]. Given this clustering, prophylactic reinforcement of the VB adjacent to the augmented fracture site is a possible intervention which may reduce the risk of accelerated adjacent VB failure. The use of prophylactic PVP in the neighboring intact vertebrae will also allow the retention of the VB height, and thus spinal alignment with the kyphotic angle will be maintained. This ensures that the moment forces in the segment are not significantly increased, reducing the prospect of further fractures cascading along the spine [18]. Clearly, if the cement augmentation itself is the primary cause of adjacent vertebral fracture, then prophylactic treatment will only be successful if it can prevent a further cascade of altered load distribution. However, if the spatial clustering of the vertebrae as a result of osteoporotic compromise is indeed a major contributory factor, then augmentation of the adjacent vertebrae may prevent further fracture. With the vertebral shell not compromised by fracture, there is also a reduced chance of cement leakage outside the VB which has been reported in 30% to 93% of patients treated with PVP for vertebral collapse [22–27].

The clinical outcomes of the current PVP procedure are thought to be dependent, in part, on the mechanical properties of the bone cement used in the augmentation process [28]. Previous studies have suggested that lowering the elastic modulus of the cement, typically polymethylmethacrylate (PMMA), may reduce the adverse load distribution on the vertebral end plate and hence the risk of adjacent VB failure [15,17]. Several studies have investigated the effects of vertebroplasty using multilevel specimens under cyclic loads [29– 34]. Kettler et al. [31] and Wilke et al. [32] prefilled the vertebra adjacent to the fracture site to prevent vertebral subsidence and Huber et al. [33] encased the adjacent levels to restrict motion. Other investigations restricted the load in the demand cycle to below that often seen in physiological motion [29]. The objective of this pilot study was twofold: (1) to provide an initial biomechanical evaluation of prophylactic vertebral reinforcement adjacent to vertebroplasty on a three-level vertebral segment under increasingly demanding dynamic loads and (2) to assess the effect of altering the mechanical properties of the cement used in the intact vertebrae.

Materials and methods Specimen preparation After ethics committee approval, nine cadaveric spines from male and female donors (age range: 57–84 years) were acquired from the Leeds Tissue Bank (Leeds General Infirmary, Leeds, UK). Gross examination and plain lateral radiographs (1 exposure, 70 kV) confirmed that no fractures were present in the T12–L2 region of each spine. Threevertebra segmental units (TVSUs) of T12–L2 were then dissected free of soft-tissue attachments and disarticulated at the cranial T12 and caudal L2 intervertebral disc, facet joint, and the costovertebral junction. In all cases, the posterior elements and integrity of the spinal canal were maintained, and all major ligaments including the capsular attachments were preserved. Initial assessments were performed on all vertebral segments using quantitative computed tomography (QCT) (PQ 2000, Picker, USA) and image transfer software (e-film v1.5.3, Milwaukee, USA). Spatial resolution was set at 2 mm (65 mA, 140 kV). An average Hounsfield Unit was obtained from the largest circular region of interest of trabecular bone taken from the middle of each vertebra and bone mineral density (BMD) was calculated using a CT-specific algorithm calibrated to

176

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181

a hydroxyapatite phantom. All vertebrae were shown to have a BMD value below 80 mg/cm3 which has been previously used in defining osteoporotic vertebrae [35]. This is a value lower than that previously determined by Cann et al. whom recommended a value of 110 mg/cm3 [36]. Image analysis software (ImageJ, National Institutes of Health, Maryland, USA) was used to quantify the height of the intervertebral disc space from the QCT images. Quasi-static testing The stiffness of each segment prefracture and postaugmentation was characterized by quasi-static loading in the materials testing machine (AGS-10kNG, Shimadzu Corp., Kyoto, Japan). VB dimensions were taken using digital calipers accurate to 0.01 mm (Mitutoyo MTI Corp, Aurora, IL). Impressions of the cranial T12 and caudal L2 end plate surfaces were made in PMMA cement to provide a flat surface for mechanical testing. Each TVSU was then subject to a single-point compressive load applied at 1 mm/min through a steel end plate and ball located at the mid-point of the T12 cranial end plate. This point of load application was considered most appropriate in evaluating the segmental effects of augmentation in the construct. Three repeat tests were completed for each TVSU, with termination of the test defined as 50% of the predicted failure load of L1. Prediction of failure load was determined with QCT images using the method previously described in which a high level of agreement was observed between the actual failure load and the product of caudal end plate area and BMD for nonaugmented vertebrae [37,38]. From the load-deformation data, the stiffness of the segment was determined over a 0.6-mm range finishing at the termination of the test. This region was repeatedly found to be the most linear. T12 fracture generation Vertebrae L1 and L2 were enclosed in aluminum foil and the anterior margin of the TVSU spinal canal was clamped to a stainless steel rod to maintain sagittal and coronal alignment. This also acted as a reference point from which to locate the dimensions of the cranial end plate of the T12 VB. The L1 and L2 vertebrae were fixed in PMMA cement up to the cranial end plate of L1 to support their structural integrity under compression, and the cranial end plate surface of T12 was also secured in PMMA bone cement to provide a flat surface for mechanical testing (Fig. 1). The TVSU, held in custom-made rig to prevent lateral translation, was placed in a materials testing machine and subject to a single-point load applied at a location 25% of the VB length from the anterior margin on the T12 cranial end plate in the mid-sagittal plane (Fig. 1). A similar method had previously been used to produce reliable VCFs in single vertebra [37]. The endpoint of compression was taken as deformation of the T12 VB height to 75% of the original and the initial failure strength was defined as the peak load causing

Fig. 1. Anterior compression fracture generated in the superior (T12) vertebrae with a single-point load.

fracture. After fracture, the encasing cement and aluminum foil were removed and each TVSU was wrapped in salinesoaked gauze, sealed in plastic bags, and stored at 20 C until 24 hours before augmentation and testing at which time they were thawed at room temperature (20 C) [39]. The specimens were then divided into three groups determined by the predicted fracture strength to ensure an even distribution of this parameter within each cohort. The fractured T12 vertebrae underwent augmentation, while L1 was prophylactically reinforced with cement of varying moduli. Vertebral augmentation Previous investigations have changed the properties of injectable bone cement with the addition of soluble fillers [40–43]. In this study, aqueous carboxymethylcellulose (CMC) was added to Vertebroplastic cement (CMW, Blackpool, UK) to reduce the elastic modulus of the cement. Using data from a previous study [34], the Vertebroplastic:CMC ratio was varied to produce three materials having an elastic modulus of 100% (ie pure cement), 25%, and 12.5% that of the original cement. Augmentation of the T12 fractured and L1 intact vertebral bodies was achieved using a unilateral extrapedicular approach with a 13-Gauge bone biopsy needle advanced to the anterior third of the VB under fluoroscopic guidance (Fig. 2). Vertebroplastic cement was mixed following

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181

177

and to identify vertebral fracture using a primarily pathomorphological criteria [46]. Statistical analysis Comparisons within groups to investigate differences between pre– and postaugmentation were undertaken using paired t-tests after appropriate analyses had shown that the Central Limit Theorem applies and that parametric statistics could be used. Analysis of variance was used to investigate differences between groups numbering more than two. To investigate association between variables, Spearman’s rank correlation statistic was used. A nominal significance level, a, of 5% was used for all tests performed. Fig. 2. Fluoroscopic image indicating the advance of the bone biopsy needles to the anterior third of the vertebral bodies.

Results Fracture strength and segment stiffness

manufacturer’s recommendations, with the addition of the CMC for the 25% and 12.5% moduli materials. The cements were injected using a V-Max cement delivery system (CMW, Blackpool, UK). T12 was injected with conventional cement and the intact L1 vertebra was injected with one of the modified cements. The defined endpoint of the procedure was an estimated 20% volume fill [44,45] based on the gross dimensions (heightend plate surface area) taken from the QCT images. All augmented vertebrae were wrapped in saline-soaked gauze, placed in sealed plastic bags, and floated in a water bath at 37 C for 24 hours to simulate physiological conditions and to allow adequate time for cement curing [39]. QCT scans were then performed to evaluate the T12 wedge fracture and cement placement.

VCFs in the superior T12 vertebrae of all TVSUs were confirmed with QCT scans, showing a reduction in the anterior height and no compromise of L1 or L2 vertebra (Fig. 3). The mean failure strength was 2.2860.89 kN S.D. Comparison of the subfailure quasi-static compression tests before fracture and after augmentation showed no significant difference in the stiffness of the TVSU (t51.4, p5.19; paired t-test) (Fig. 4). Construct biomechanics under cyclic loading Each TVSU demonstrated an overall height reduction (mean514.9%66.6%) which corresponded with an increase in segment stiffness during loading (rSpearman’s50.77–0.99) (Fig. 5). In the first 95,000 cycles under maximum loads of

Cyclic loading After postaugmentation quasi-static loading, the cranial end plate of T12 and caudal end plate of L2 were secured within simulator fixtures using PMMA bone cement, with the central axis located in the mid-point of the T12 VB. Dynamic cyclic compression was applied at 1 Hz in a fatigue-testing machine (Simulation Solutions Ltd., Stockport, UK) under escalating loads proportional to the patient’s body weight (BW): 80,000 15,000 10,000 10,000

cycles cycles cycles cycles

from from from from

1 to 2 BW 1.25 to 2.5 BW 1.5 to 3 BW 1.75 to 3.5 BW

Force-displacement data were collected for every 100th cycle and the stiffness of the segment was derived from the complete loading protocol. QCT scans were performed after cyclic loading to quantify the intervertebral disc space

Fig. 3. A typical QCT scout image showing a reduction in the anterior height of the T12 superior vertebrae and no compromise of the L1 or L2 vertebral body.

178

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181

Fig. 4. Stiffness of the specimens pre– and postaugmentation.

2 and 2.5 BW, most specimens demonstrated a steady decrease in overall height (Fig. 6). Specimen 2 from the 12.5% E cement group, however, demonstrated step changes in the levels of subsidence under maximum loads of 2 and 2.5 BW indicating the phase in which fracture in the TVSU may have initially occurred. QCT post-testing images confirmed an L2 fracture in this specimen. Under more demanding profiles in which maximum loads were 3 and 3.5 BW, specimens demonstrated two general patterns of subsidence: 1. A small initial decrease in the overall TVSU height at the start of each loading phase followed by a steady state height decrease. This was observed in three specimens (TVSU 2: 100% E cement, TVSU 1: 25% E cement, and TVSU 3: 12.5% E cement) that were found to have no evidence of fracture from post-testing QCT scans and showed a mean height reduction of 7.8%61.8%. 2. A rapid decrease in the overall height of the TVSU during loading was observed in six specimens (TVSU 1 and 3: 100% E cement, TVSU 2 and 3: 25% E cement, and TVSU 1 and 2: 12.5% E cement). These height reductions included sudden step changes in the level of subsidence midway through loading or

Fig. 6. Overall TVSU height reductions shown across the incrementally increasing dynamic load application for the groups of different cement modulus tested: (Top) 100% E cement, (Middle) 25% E cement, and (Bottom) 12.5% E cement. The load applied to each specimen was in proportion to original patient’s body weight.

a severe decrease in height at the start of a loading profile. These specimens were found to have adjacent, peri-augmentation fractures, or a combination of both and had a mean loss in overall TVSU height of 18.6%64.7%. A poor correlation was found between the segment height loss and 1) the predicted failure strength of the L1 vertebra (rSpearman’s50.41, p5.24), 2) the mean TVSU bone density (rSpearman’s50.33, p5.19), and 3) the modulus of cement (one-way analysis of variance: F50.35, p5.72). QCT analysis

Fig. 5. Typical plots of the overall TVSU height reduction during dynamic loading and the corresponding stiffness.

A total of 11 fractures in six specimens were identified from QCT images after cyclic loading (Table 1). Two of these fractures occurred around the cement in the T12 vertebrae, whereas five were noted in the prophylactically augmented L1 vertebra. Four specimens showed fracture in the adjacent L2 vertebra. Altering the modulus of cement in the prophylactically augmented vertebrae appeared to have no

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181 Table 1 Vertebral fracture after cyclic loading Specimen

T12

L1

L2

100% E cement 1 2 3

d d d

# d #

# d #

25% E cement 1 2 3

d d #

d # #

d d d

12.5% E cement 1 2 3

# d d

# d d

# # d

The symbol # denotes fracture.

effect on the incidence of fracture, with four fractures occurring in both the 100% E and 12.5% E groups, and three in the 25% E group. There was a good correlation between the incidence of fracture in a segment and overall TVSU height loss (rSpearman’s5 0.88, p5.002). Reductions in the overall intervertebral disc space were noted in all specimens after dynamic loading (mean532.7%69.7%) (Table 2). In the specimens with no evidence of fracture, the loss in intervertebral disc space (mean54.861.4 mm) indicates a mean 30.2%610.5% creep in the discs during testing. QCT images confirmed no significant difference in the mean VB height pre– and postcyclic testing (t50.004, p5.98; paired t-test) in these specimens.

Discussion A growing elderly population in Western countries means that the appropriate management of osteoporotic Table 2 Measurements of the intervertebral disc space pre– and post-dynamic loading Precyclic loading disc height (mm) Fracture Specimen sites T12–L1

L1–L2

Postcyclic loading disc height (mm) T12–L1

L1–L2

Mean height reduction (%)

100% E cement 1 Yes 2 No 3 Yes

8.72 7.35 8.51

9.58 7.77 9.49

6.31 4.29 5.65

7.30 4.64 6.28

25.73 40.97 33.75

25% E cement 1 No 2 Yes 3 Yes

7.82 8.40 8.57

7.99 7.29 7.70

4.80 5.90 5.39

6.31 6.43 5.38

29.75 20.769 33.60

12.5% E cement 1 Yes 2 Yes 3 No

8.85 6.78 8.69

10.33 8.04 8.86

4.76 4.46 6.06

4.87 4.57 8.01

49.54 38.70 19.93

179

VCFs now poses a significant challenge to the clinical community. Vertebroplasty has shown extensive promise as a treatment modality for these fractures, although the serious complication of accelerated adjacent VB failure may limit the current procedure to the management of short-term, serious unresolved pain rather than long-term structural augmentation of the vertebra. The purpose of this preliminary study was to provide an initial biomechanical evaluation of prophylactic reinforcement in the VB adjacent to the augmented fracture site. In addition, the effect of augmenting the intact vertebrae with modified injectable bone cement was assessed. The response of the segment to quasi-static compression appeared to be governed by the intervertebral discs with no significant change in the stiffness after augmentation. This concurs with previous research [47]. Under cyclic loads representing normal physiological motion and moderate activity, the subsidence was characterized by a small and continuous segment height reduction during loading and there was little evidence of fracture in most cases. This behavior is in concordance with previous vertebroplasty research that recorded similar levels of subsidence in the segment under conservative loads [32]. This pattern of steady height loss continued in a third of specimens under more demanding load profiles that represented more strenuous physical activity. These specimens were found to have no induced fractures and the overall height reduction was because of intervertebral disc height loss. In most cases, however, the more demanding conditions appeared to induce combinations of adjacent and peri-augmentation fractures, characterized by either step changes in the subsidence midway through loading or sudden and large changes in the height of the TVSU at the start of a new loading profile. This corresponds with previous dynamic assessments on a prophylactically treated vertebral segment, where loads approaching two to three times patient BW–generated fracture [33]. Vertebral reinforcement adjacent to vertebroplasty may therefore be a useful procedure in strengthening vertebra when subject to moderate physiological loads, but may not prevent fracture under more strenuous activities. Indeed, under more demanding load conditions, the cement in the intact vertebra may still cause an altered load distribution within the spinal column that contributes to fractures around the augmentation site and compromise of the adjacent vertebra. However, the effect of spatial clustering, where vertebrae in the region of a fracture site are intrinsically at greater risk of fracture even in the absence of a treated vertebra [20,21], was not assessed in this study. Control groups of similar bone quality undergoing the same loading protocols may provide the necessary indications in this area and should be considered in the design of future biomechanical investigations of vertebroplasty. The segments in this study were subject to an incrementally increasing load regime in proportion to the patient’s BW to represent various activity demands for this cohort. The loading protocol of 115,000 cycles represented an

180

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181

estimated 2 weeks of in vivo movement in a young and healthy individual, and approximately 4 weeks in an elderly person [48,49]. A study of the longer-term treatment effects was not possible because of the challenges of tissue degradation and the associated changes in biomechanical behavior. Previous vertebroplasty research involving multilevel specimens under cyclic compression used an eccentric flexion load under modest physiological conditions [31–33]. Although this provides a more accurate representation of the physiological situation in vivo, this complex loading combination would have increased the intricacy of this initial study with a possible increase in variability in the results. The demand profile was therefore applied down the mid-point of the T12 VB to allow a meaningful comparison to be made between study groups. It should be noted, however, that the application of a single axial compressive load does not completely address how progressive vertebral collapse occurs in vivo, with augmented vertebrae subject to a complex range of spinal loads that are dependent on patient-related factors. This should be considered in future biomechanical studies that use cyclic loading profiles to evaluate spinal interventions. The adjacent levels in a multi-segmental model have previously been fixed to reduce the amount of subsidence and hence the variability in the study [31–33], although this may affect the failure characteristics of the treated VB because of the increased pressure in the nucleus pulposus and the deflection of the adjacent end plate [14,16]. The incidence of fracture also appears to be dependent on the load demands being placed on the segment and careful consideration should be given to the design of any future dynamic assessments to ensure the physiological motion and activity demands are appropriately represented. The high elastic modulus of conventional PMMA bone cement used in vertebroplasty is thought to play a crucial role in the clinical outcomes of the procedure [28]. It is the cement that is thought to transfer a greater proportion of the load through the central augmented trabeculae structure than would occur naturally [50], causing an altered load distribution within the spinal segment [15–17]. In this preliminary investigation, however, lowering the modulus of the cement injected into the adjacent intact vertebrae did not appear to have a significant effect on the incidence of adjacent or peri-augmentation fractures, or the overall TVSU height reduction when subject to cyclic loading. These findings are in agreement with previous experimental and computational work which suggested that augmentation with a lower modulus cement does not significantly affect the occurrence of fracture in the adjacent levels [51,52].

Conclusions This preliminary work has started to investigate the biomechanical effectiveness of augmenting the intact vertebra

adjacent to vertebroplasty, with particular attention to the segmental consequences of the treatment and the effects of different activity demands and cement properties. Adjacent and peri-augmentation fractures were induced in several treated segments, although these fractures appeared to occur under loads that represented more strenuous activities. The frequency or location of the induced fracture within the vertebral segment did not appear to be related to the elastic modulus of the cement in the adjacent vertebra. Further research is now required over a larger sample size with the necessary statistical power to assess enhancements to the vertebroplasty treatment under complex loading regimes that accurately represent physiological environment. Spatial clustering of these osteoporotic fractures also requires further investigation. In addition, the development of bone cements with mechanical properties that are specifically suited to the mechanical environment in which they operate are now required if the procedure is to be a long-term treatment option for painful vertebral fracture.

References [1] Diamond TH, Champion B, Clark WA. Management of acute osteoporotic vertebral fractures: a nonrandomized trial comparing percutaneous vertebroplasty with conservative therapy. Am J Med 2003;114: 257–65. [2] Hide IG, Gangi A. Percutaneous vertebroplasty: history, technique and current perspectives. Clin Radiol 2004;59:461–7. [3] McGraw JK, Lippert JA, Minkus KD, Rami PM, Davis TM, Budzik RF. Prospective evaluation of pain relief in 100 patients undergoing percutaneous vertebroplasty: results and follow-up. J Vasc Interv Radiol 2002;13:883–6. [4] McKiernan F, Faciszewski T, Jensen R. Quality of life following vertebroplasty. J Bone Joint Surg [Am] 2004;86:2600–6. [5] Winking M, Stahl JP, Oertel M, Schnettler R, Boker DK. Treatment of pain from osteoporotic vertebral collapse by percutaneous PMMA vertebroplasty. Acta Neurochir 2004;146:469–76. [6] Kobayashi K, Shimoyama K, Nakamura K, Murata K. Percutaneous vertebroplasty immediately relieves pain of osteoporotic vertebral compression fractures and prevents prolonged immobilization of patients. Eur Radiol 2005;15:360–7. [7] Liliang P-C, Su T-M, Liang C-L, Chen H-J, Tsai Y-D, Lu K. Percutaneous vertebroplasty improves pain and physical functioning in elderly osteoporotic vertebral compression fracture patients. Gerontology 2005;51:34–9. [8] Zoarski GH, Snow P, Olan WJ, Stallmeyer MJB, Dick BW, Hebel JR, et al. Percutaneous vertebroplasty for osteoporotic compression fractures: quantitative prospective evaluation of long-term outcomes. J Vasc Interv Radiol 2002;13:139–48. [9] Hu YC, Hart DJ. Complications of vertebroplasty and kyphoplasty. Tech Reg Anesth Pain Manage 2007;11:164–70. [10] Uppin AA, Hirsch JA, Centenera LV, Pfiefer BA, Pazianos AG, Choi IS. Occurrence of new vertebral body fracture after percutaneous vertebroplasty in patients with osteoporosis. Radiology 2003;226:119–24. [11] Kim SH, Kang HS, Choi JA, Ahn JM. Risk factors of new compression fractures in adjacent vertebrae after percutaneous vertebroplasty. Acta Radiol 2004;45:440–5. [12] Trout AT, Kallmes DF, Kaufmann TJ. New fractures after vertebroplasty: adjacent fractures occur significantly sooner. AJNR Am J Neuroradiol 2006;27:217–23.

R.J. Oakland et al. / The Spine Journal 9 (2009) 174–181 [13] Ananthakrishnan D, Berven S, Deviren V, Cheng K, Lotz JC, Xu Z, et al. The effect on anterior column loading due to different vertebral augmentation techniques. Clin Biomech 2005;20:25–31. [14] Berlemann U, Ferguson SJ, Nolte LP, Heini PF. Adjacent vertebral failure after vertebroplasty. A biomechanical investigation. J Bone Joint Surg [Br] 2002;84:748–52. [15] Baroud G, Nemes J, Heini P, Steffen T. Load shift of the intervertebral disc after a vertebroplasty: a finite-element study. Eur Spine J 2003;12:421–6. [16] Polikeit A, Nolte LP, Ferguson SJ. The effect of cement augmentation on the load transfer in an osteoporotic functional spinal unit: finiteelement analysis. Spine 2003;28:991–6. [17] Wilcox RK. The biomechanical effect of vertebroplasty on the adjacent vertebral body: a finite element study. Proc Inst Mech Eng [H] 2006;220:565–72. [18] Rohlmann A, Zander T, Bergmann G. Spinal loads after osteoporotic vertebral fractures treated by vertebroplasty or kyphoplasty. Eur Spine J 2006;15:1255–64. [19] Kaplan FS, Scherl JD, Wisneski R, Cheatle M, Haddad JG. The cluster phenomenon in patients who have multiple vertebral compression fractures. Clin Orthop Relat Res 1993;161–7. [20] Wasnich RD. Vertebral fracture epidemiology. Bone 1996;18: 179S–83S. [21] Lunt M, O’Neill TW, Felsenberg D, Reeve J, Kanis JA, Cooper C, et al. Characteristics of a prevalent vertebral deformity predict subsequent vertebral fracture: results from the European Prospective Osteoporosis Study (EPOS). Bone 2003;33:505–13. [22] Jensen ME, Evans AJ, Mathis JM, Kallmes DF, Cloft HJ, Dion JE. Percutaneous polymethylmethacrylate vertebroplasty in the treatment of osteoporotic vertebral body compression fractures: technical aspects. AJNR Am J Neuroradiol 1997;18:1897–904. [23] Cortet B, Cotten A, Boutry N, Flipo RM, Duquesnoy B, Chastanet P, et al. Percutaneous vertebroplasty in the treatment of osteoporotic vertebral compression fractures: an open prospective study. J Rheumatol 1999;26:2222–8. [24] Cyteval C, Sarrabere MP, Roux JO, Thomas E, Jorgensen C, Blotman F, et al. Acute osteoporotic vertebral collapse: open study on percutaneous injection of acrylic surgical cement in 20 patients. AJR Am J Roentgenol 1999;173:1685–90. [25] Peh WCG, Gilula LA, Peck DD. Percutaneous vertebroplasty for severe osteoporotic vertebral body compression fractures. Radiology 2002;223:121–6. [26] Hodler J, Peck D, Gilula LA. Midterm outcome after vertebroplasty: predictive value of technical and patient-related factors. Radiology 2003;227:662–8. [27] Yeom JS, Kim WJ, Choy WS, Lee CK, Chang BS, Kang JW. Leakage of cement in percutaneous transpedicular vertebroplasty for painful osteoporotic compression fractures. J Bone Joint Surg [Br] 2003;85:83–9. [28] Lewis G. Injectable bone cements for use in vertebroplasty and kyphoplasty: state-of-the-art review. J Biomed Mater Res [B] Appl Biometer 2006;76:456–68. [29] Hitchon PW, Goel V, Drake J, Taggard D, Brenton M, Rogge T, et al. Comparison of the biomechanics of hydroxyapatite and polymethylmethacrylate vertebroplasty in a cadaveric spinal compression fracture model. J Neurosurg 2001;95:215–20. [30] Lu WW, Cheung KM, Li YW, Luk KD, Holmes AD, Zhu QA, et al. Bioactive bone cement as a principal fixture for spinal burst fracture: an in vitro biomechanical and morphologic study. Spine 2001;26: 2684–90; discussion 2690–1. [31] Kettler A, Schmoelz W, Shezifi Y, Ohana N, Ben-Arye A, Claes L, et al. Biomechanical performance of the new BeadEx implant in the treatment of osteoporotic vertebral body compression fractures:

[32]

[33]

[34]

[35]

[36]

[37]

[38] [39]

[40]

[41]

[42] [43]

[44]

[45]

[46]

[47]

[48]

[49] [50]

[51] [52]

181

restoration and maintenance of height and stability. Clin Biomech 2006;21:676–82. Wilke H-J, Mehnert U, Claes LE, Bierschneider MM, Jaksche H, Boszczyk BM. Biomechanical evaluation of vertebroplasty and kyphoplasty with polymethyl methacrylate or calcium phosphate cement under cyclic loading. Spine 2006;31:2934–41. Huber G, Muller-Bergen L, Heinze J, Eggers C, Puschel K, Morlock MM. Mechanical stability of augmented spinal segments. J Biomech 2006;39:S523. Oakland RJ, Furtado NR, Wilcox RK, Hall RM. The biomechanical effectiveness of prophylactic vertebroplasty: a dynamic cadaveric study. J Neurosurg Spine 2008;8:442–9. Homminga J, Weinans H, Gowin W, Felsenberg D, Huiskes R. Osteoporosis changes the amount of vertebral trabecular bone at risk of fracture but not the vertebral load distribution. Spine 2001;26:1555–61. Cann CE, Genant HK, Kolb FO, Ettinger B. Quantitative computed tomography for prediction of vertebral fracture risk. Bone 1985;6: 1–7. Furtado N, Oakland RJ, Wilcox RK, Hall RM. A biomechanical investigation of vertebroplasty in osteoporotic compression fractures and in prophylactic vertebral reinforcement. Spine 2007;32: E480–7. Brinckmann P, Biggemann M, Hilweg D. Prediction of the compressive strength of human lumbar vertebrae. Spine 1989;14:606–10. Belkoff SM, Mathis JM, Jasper LE. Ex vivo biomechanical comparison of hydroxyapatite and polymethylmethacrylate cements for use with vertebroplasty. AJNR Am J Neuroradiol 2002;23:1647–51. Bruens ML, Pieterman H, de Wijn JR, Vaandrager JM. Porous polymethylmethacrylate as bone substitute in the craniofacial area. [see comment]. J Craniofac Surg 2003;14:63–8. van Mullem PJ, Vaandrager JM, Nicolai JP, de Wijn JR. Implantation of porous acrylic cement in soft tissues: an animal and human biopsy histological study. Biomaterials 1990;11:299–304. Vaandrager JM, van Mullem PJ, de Wijn JR. Craniofacial contouring and porous acrylic cement. Ann Plastic Surg 1988;21:583–93. van Mullem PJ, de Wijn JR. Bone and soft connective tissue response to porous acrylic implants. A histokinetic study. J Craniomaxillofac Surg 1988;16:99–109. Liebschner MA, Rosenberg WS, Keaveny TM. Effects of bone cement volume and distribution on vertebral stiffness after vertebroplasty. Spine 2001;26:1547–54. Molloy S, Riley LH 3rd, Belkoff SM. Effect of cement volume and placement on mechanical-property restoration resulting from vertebroplasty. AJNR Am J Neuroradiol 2005;26:401–4. Magerl F, Aebi M, Gertzbein SD, Harms J, Nazarian S. A comprehensive classification of thoracic and lumbar injuries. Eur Spine J 1994;3:184–201. Kayanja M, Togawa D, Lieberman I. Biomechanics of prophylactic and post fracture augmentation of osteoporotic vertebral compression fractures in the thoracolumbar spine. Spine J 2004;4:S46. Goldsmith AAJ, Dowson D, Wroblewski BM, Siney PD, Fleming PA, Lane JM, et al. Comparative study of the activity of total hip arthroplasty patients and normal subjects. J Arthroplasty 2001;16:613–9. Wallbridge N, Dowson D. The walking activity of patients with artificial hip joints. Eng Med 1982;11:95–6. Kayanja MM, Togawa D, Lieberman IH. Biomechanical changes after the augmentation of experimental osteoporotic vertebral compression fractures in the cadaveric thoracic spine. Spine J 2005;5:55–63. Wilcox RK. Biomechanical study of vertebroplasty using finite element analysis. J Biomech 2006;39:S419. Kayanja MM, Evans K, Milks R, Lieberman IH. Adjacent level load transfer following vertebral augmentation in the cadaveric spine. Spine 2006;31:E790–7.