New Vertebral Compression Fractures After Prophylactic Vertebroplasty in Osteoporotic Patients

Transcription

1 Musculoskeletal Imaging Original Research Kamano et al. Vertebral Compression Fractures After Prophylactic Vertebroplasty Musculoskeletal Imaging Original Research Hironori Kamano 1,2 Akio Hiwatashi 2 Nobuo Kobayashi 1 Sokun Fuwa 1 Osamu Takahashi 3 Yukihisa Saida 1 Hiroshi Honda 2 Yuji Numaguchi 1 Kamano H, Hiwatashi A, Kobayashi N, et al. Keywords: prophylactic vertebroplasty, subsequent compression fracture DOI: /AJR Received September 30, 2010; accepted after revision February 10, Supported in part by research grant from the Japan Society. 1 Department of Radiology, St. Luke s International Hospital, Tokyo, Japan. 2 Department of Clinical Radiology, Graduate School of Medical Sciences, Kyushu University, Maidashi, Higashi-ku, Fukuoka , Japan. Address correspondence to H. Kamano (kamano@med.kyushu-u.ac.jp). 3 Department of Internal Medicine, St. Luke s International Hospital, Tokyo, Japan. AJR 2011; 197: X/11/ American Roentgen Ray Society New Vertebral Compression Fractures After Prophylactic Vertebroplasty in Osteoporotic Patients objective. Previous studies have shown the possible efficacy of prophylactic cement injection for nonfractured vertebrae during percutaneous vertebroplasty for compression fractures. The purpose of this study was to investigate risk factors for subsequent fractures after prophylactic percutaneous vertebroplasty. MATERIALS AND METHODS. This retrospective study included 116 patients with osteoporotic compression fractures who underwent prophylactic percutaneous vertebroplasty. The patients were monitored with physical examinations and radiographs at 1 day and at 3 and 12 months after percutaneous vertebroplasty, and if back pain recurred. We analyzed the following multiple covariates to determine whether they were associated with recurrence: age, sex, steroid use, and the preoperative number of unhealed or chronic compression fractures. RESULTS. Subsequent fractures in any vertebra occurred within 3 months after the procedure at 26 vertebrae in 21 patients (18.1%), and 36 occurred in 28 patients (24.1%) within 12 months. The occurrence of subsequent fractures within 12 months depended on the preoperative number of unhealed vertebrae: the occurrence rate was 16.9% (11/65) for one vertebra, 27.0% (10/37) for two vertebrae, and 50.0% (7/14) for three or more vertebrae. The incidence of subsequent fractures was significantly higher in patients with three or more fractures than in those with one fracture (p < 0.05). There were no statistically significant differences for the other factors. CONCLUSION. Patients with three or more fractures tended to have subsequent fractures, despite undergoing prophylactic percutaneous vertebroplasty. However, there was no increased risk of subsequent fractures related to prophylactic percutaneous vertebroplasty. P ercutaneous vertebroplasty has been shown to provide symptomatic relief from the pain caused by vertebral compression fractures refractory to medical therapy [1 20]. However, some patients complain of recurrent pain after percutaneous vertebroplasty, and in such cases the pain is often caused by subsequent compression fractures. Therefore, patients often request repeat procedures [2, 11, 13, 21]. Subsequent compression fractures after percutaneous vertebroplasty may not be a complication of the procedure itself but can be a part of the natural course of osteoporosis [1, 22]. Previous studies have shown that 41 67% of subsequent fractures occur at the level adjacent to the augmented vertebra [2, 6, 9]. Although it is not recommended in the guidelines of the American College of Radiology [23], Kobayashi et al. [20] recently reported that prophylactic treatment in nonfractured vertebrae can decrease the incidence of subsequent fractures. Several risk factors for subsequent fractures after percutaneous vertebroplasty have been reported, including the involvement of multiple fractures [9], intradiscal cement leakage [15, 16], the presence of an intravertebral cleft [7, 19] or solid pattern cement filling [18], a low body mass index [17], steroid medication [10, 14], the severity of the wedge angle [13], and well-achieved vertebral height [5] or kyphosis correction [13]. However, to the best of our knowledge, there has been no previous study identifying the risk factors for subsequent fracture after prophylactic percutaneous vertebroplasty. In our clinical experience, we often experience subsequent fractures after prophylactic percutaneous vertebroplasty in patients with multiple unhealed fractures. Therefore, the AJR:197, August

2 Kamano et al. purpose of this study was to investigate risk factors for subsequent fractures after prophylactic percutaneous vertebroplasty. Materials and Methods Patients This retrospective study was approved by the review board at our institution, and written informed consent was obtained from all patients. We reviewed the medical records of 194 consecutive patients who underwent percutaneous vertebroplasty with polymethylmethacrylate (PMMA) for osteoporotic compression fractures at our hospital from January 2007 through March In this study, we defined unhealed fractures as fractured vertebrae with pain on palpation and bone marrow edema or abnormal enhancement on MRI. The inclusion criteria in this study were as follows: initial treatment of painful osteoporotic compression fracture, treatment of at least one nonfractured vertebra adjacent to the unhealed fracture, follow-up examination at least 3 months after the treatment or on an as-needed basis at our office, and repeated performance of spine radiographs or MRI. Patients were excluded if they met any of the following criteria: repeat treatment of the subsequent fractures after percutaneous vertebroplasty (32 sessions in 28 patients), compression fractures with Schmorl nodules (three patients), lack of follow-up (15 patients), or no prophylactic injection at all (32 patients). As a general rule, we did not inject more than 30 g (one pack) of PMMA at any 1-day session of treatment. We therefore did not perform prophylactic percutaneous vertebroplasty in 32 patients who had many unhealed fractures or if the adjacent vertebrae were healed fractures (Fig. 1). Consequently, this study included 116 patients with 190 unhealed painful osteoporotic compression fractures (96 women and 20 men; mean age, 76.2 years; range, years) treated in 118 sessions. The locations and numbers of the treated unhealed vertebrae were as follows: T6 (n = 1), T7 (n = 3), T8 (n = 3), T9 (n = 4), T10 (n = 7), T11 (n = 23), T12 (n = 45), L1 (n = 36), L2 (n = 31), L3 (n = 19), L4 (n = 14), and L5 (n = 4). Prophylactic cement injection was performed at the adjacent vertebrae above the unhealed fractures if the total amount of cement injected was less than 30 g, or one pack (the mixture of cement and barium was about 40 ml). Uemura et al. [24] have reported that the treatment of multiple vertebrae in one session could cause decreased partial pressure of oxygen in arterial blood (PaO 2 ). The amount of the cement used tended to increase by increasing of the number of vertebrae treated. We therefore limited the maximum use of cement (up to 1 pack) at each 1-day session. Because of this policy, we could not treat all adjacent vertebrae, Repeat treatment: 28 Patients who underwent percutaneous vertebroplasty: 194 Prophylactic treatment: 162 Compression fractures with Schmorl nodules: 3 No prophylactic treatment: 32 Lack of follow-up: 15 The other: 116 (Patients in this study) Fig. 1 Patients who underwent percutaneous vertebroplasty. Diagram shows how patients were included in study. Fractured vertebrae: 190 (2) Levels of adjacent vertebrae Fractured (treated) vertebrae: 105 (0) Sacrum: 4 (0) Prophylactically treated vertebrae: 180 (1) Nonprophylactically treated vertebrae (Untreated vertebrae): 91 (14) Level of vertebrae two above or below preoperatively unhealed fractured vertebrae Untreated vertebrae: 157 (8) Treated vertebrae: 20 (0) Adjacent to other treated vertebrae: 1 (0) Sacrum: 2 (0) Untreated vertebrae: 74 (4) Treated vertebrae: 1 (0) Adjacent to other treated vertebrae: 4 (0) Sacrum: 12 (0) Fig. 2 Number of vertebrae treated and follow-up. Numbers in parentheses denote number of subsequent fractures at 12 months. even if they were immediately above the fractured vertebra, in cases with multiple unhealed fractures. Prophylactic treatment was also performed for the adjacent vertebra below the unhealed fractures when there was intradiscal cement leakage to the disk below the fractured vertebra or the presence of an intravertebral cleft, both of which are considered to be risk factors for subsequent fractures [7, 15, 16, 19]. A total of 180 nonfractured vertebrae were treated, and the majority of prophylactically injected vertebrae were located between T11 and L1. The locations and numbers of the prophylactic vertebrae were as follows: T6 (n = 2), T8 (n = 2), T9 (n = 2), T10 (n = 18), T11 (n = 40), T12 (n = 32), L1 (n = 41), L2 (n = 23), L3 (n = 15), L4 (n = 3), and L5 (n = 2). These included vertebrae above the unhealed fracture (n = 107), those below the unhealed fracture (n = 53), and sandwich vertebrae between unhealed fractures (n = 20). The mean number of treated fractured vertebrae per session was 1.6 (range, 1 7), and the mean number of treated vertebrae per patient, including prophylactic vertebrae, was 3.2 (range, 2 8). Two patients were treated dividing one session on the second. Vertebroplasty Procedure Three board-certified radiologists performed percutaneous vertebroplasty. They had experience with percutaneous vertebroplasty for 15, 10, and 5 years, respectively. Percutaneous vertebroplasty was performed with biplane fluoroscopy under local anesthesia and conscious sedation in all patients. Blood pressure, ECGs, and arterial oxygen saturation were monitored during the procedure. Eleven- or 13-gauge bone biopsy needles were used. During needle insertion, diamond-cut stylets were predominantly used, and when minor adjustment was required, bevel-cut stylets were used. Two needles were used for all patients through the left and right pedicles to achieve adequate cement filling in the vertebra. An average of 4 minutes was required for placement of two needles in one vertebral body. A liquid monomer (17 ml) 452 AJR:197, August 2011

3 Vertebral Compression Fractures After Prophylactic Vertebroplasty and 30 g of PMMA powder were mixed. To ensure radiopacity, sterile powdered barium sulfate (10 g) was added (30% weight/volume). This mixture produced a volume of approximately 40 ml. The cement was manually injected with 1-mL syringes under biplane fluoroscopic observation. We tried to obtain cement filling from superior to inferior endplates. The injection was continued until the vertebral body was filled toward its posterior 25%, and it was terminated if cement extravasation was encountered in the vein, disk, or epidural space. Up to six levels were treated in 1-day sessions in our practice, because a previous investigation of changes in alveolar partial pressure confirmed that an increase in the number of treated vertebrae correlates with decreased PaO 2 [24]. The amount of cement mixed with the injected barium sulfate was ml per vertebrae (mean, 6.2 ml/vertebrae) and ml per session (mean, 19.5 ml/session). Patients were instructed to remain supine for 3 hours after the procedure to allow cement hardening and resolution of sedation. After 3 hours of bed rest, they were evaluated at the bedside and were allowed to ambulate. The patients were routinely discharged 2 days after the treatment, after careful postoperative observation and instruction regarding various physical exercises to be completed at home. There were no complications in any patients at the time of discharge. Imaging In all patients, preoperative imaging assessment consisted of thoracic and lumbar radiographs in frontal and lateral view, MRI of the thoracic and lumbar spine, and CT of the thoracic and lumbar spine. MRI of the thoracic and lumbar spine was performed with 1.5-T scanners using a spine-array surface coil. The imaging protocol included sagittal T1-weighted images (TR/TE, / ), STIR images (inversion time, /70 93/150 ms), and fat-saturated gadolinium-enhanced T1-weighted images (flip angle, /6.2 12/80 90 ) with a 4-mm section thickness and 0.5-mm slice spacing. The matrix was , with an FOV of 480 mm, or with an FOV of 280 mm. As a general rule, the contrast material was used to evaluate the intravertebral cleft accurately. In six patients, gadolinium-enhanced T1-weighted images were not obtained because of renal dysfunction or asthma. CT of the thoracic and lumbar spine was also performed using either a 64-section or a 16-section scanner. Reconstructed axial, transverse, and sagittal images were obtained with a 3-mm section thickness. The FOV was mm, and the matrix was Immediately after the procedure, CT scans of the treated vertebrae and radiographs of the treated vertebrae were obtained to assess the cement distribution. The patients were routinely followed up with physical examinations and radiographs of the treated vertebrae at 1 day and at 3 and 12 months after percutaneous vertebroplasty, or at any point A C when there was recurrent back pain. If patients did not visit our office for follow-up, a physician interviewed them by telephone to determine whether there was recurrent pain. Even if there was no subsequent fracture on radiographs, MRI of the thoracic and lumbar spine was performed if subsequent fractures were suspected on the basis of the physical examinations. Fig year-old woman with osteoporotic compression fracture at L1. A, Fat-suppressed contrast-enhanced T1-weighted image shows unenhanced portion in fractured L1 vertebral body (arrow) indicating intravertebral cleft. B, Lateral fluoroscopic image shows 11-gauge needles inserted into fractured vertebrae at L1 and nonfractured vertebra at T12 and L2. C, Sagittal reconstructed CT shows cement in fractured vertebrae at L1 as well as in adjacent nonfractured vertebra at T12 and L2. D, Twelve months after treatment, there is no new fracture or bone marrow edema seen on STIR image. B D AJR:197, August

4 Kamano et al. The images were reviewed by two board-certified radiologists in consensus. The criteria for diagnosis of unhealed fracture were the presence of bone marrow edema or abnormal enhancement in vertebrae on MRI. We defined chronic fracture as deformation of the vertebral body without bone marrow edema or abnormal enhancement on MRI. Subsequent vertebral fracture was defined as a deformation of the vertebral body on radiographs compared with the preprocedural images or new bone marrow edema or abnormal contrast enhancement on MRI, in any previously normal vertebral body. Statistical Analysis A Fisher exact test was used to estimate the occurrence of subsequent fractures in the vertebra-based analysis. A nonparametric trend test was used to estimate the positive trend of the proportion of subsequent fractures after percutaneous vertebroplasty as the number of treated vertebral compression fractures increased [25]. For the multivariate logistic analysis, we chose potential independent variables that had p values less than 0.25 using univariate analysis and that were of known clinical importance [26]. Using logistic regression analysis, we analyzed the following multiple covariates to determine whether they were associated with the recurrence of vertebral body fracture: age, sex, steroid use, and the preoperative number of unhealed or chronic compression fractures. Because colinearity had been seen regarding the number of unhealed fractures and the number of total compression fractures, we did not use the number of total compressions as a variable. All analyses were performed using commercially available software. A p value of less than 0.05 was regarded as statistically significant. Results Among the 116 patients, subsequent fractures occurred within 3 months after the procedure at 26 vertebrae in 21 patients (18.1%), and 36 subsequent fractures occurred in 28 patients (24.1%) within 12 months. The occurrence of subsequent fractures within 3 months depended on the number of unhealed vertebrae: the occurrence rate was 10.8% (7/65) in patients with one unhealed vertebra, 21.6% (8/37) in those with two unhealed vertebrae, and 42.9% (6/14) in those with three or more unhealed vertebrae. The numbers for subsequent fractures within 12 months were 16.9% (11/65) for one vertebra, 27.0% (10/37) for two vertebrae, and 50.0% (7/14) for three or more vertebrae. The subsequent fractures frequently occurred at vertebrae adjacent to ones treated by cement injection (22/36 [61.1%]). The locations of the subsequent fractures were as A C follows: immediately above the fractured vertebrae treated by conventional cement injection (4/22 [18.2%]), immediately below the fractured vertebrae treated by conventional cement injection (10/22 [45.5%]), immediately above the nonfractured vertebrae treated by prophylactic cement injection (5/22 [22.7%]), and immediately below the nonfractured vertebrae treated by prophylactic cement injection (3/22 [13.6%]). The locations of subsequent fractures ranged from T8 to L5. The locations and numbers of subsequent fractures within 12 months were as follows: T8 (n = 2), T9 (n = 5), T10 (n = 4), T11 (n = 2), T12 (n = 1), L1 (n = 6), L2 (n = 4), L3 (n = 6), L4 (n = 3), and L5 (n = 3). In the vertebra-based analysis (Fig. 2), there were 380 vertebrae adjacent to painful unhealed fractures. In vertebrae adjacent to preoperative unhealed fractures, subsequent Fig year-old woman with osteoporotic compression fractures at T11, L1, and L2. A, STIR image shows bone marrow edema at T11, L1, and L2. B, Lateral fluoroscopic image shows 11-gauge needles inserted into fractured vertebrae at T11, L1, and L2 as well as nonfractured vertebra at T10 and T12. C, Sagittal reconstructed CT shows cement in fractured vertebrae at T11, L1, and L2 as well as in upper adjacent vertebra at T10 and sandwich adjacent vertebra at T12. D, Three months after treatment, bone marrow edema (arrow) is seen in superior endplate of L3, indicating new fracture. B D 454 AJR:197, August 2011

5 Vertebral Compression Fractures After Prophylactic Vertebroplasty fractures within 12 months occurred at one of the 180 prophylactically treated vertebrae (0.6%) and at 14 of the 91 nonprophylactically treated vertebrae (15.4%). There was a statistically significant difference (p < 0.001) in the incidence of subsequent fractures in adjacent vertebrae between those treated and those not treated prophylactically. Of the 271 vertebrae two above or below preoperatively unhealed fractured vertebrae, subsequent fractures occurred at eight of the 157 vertebrae adjacent to prophylactically treated vertebrae (5.1%) and at four of the 74 vertebrae adjacent to nonprophylactically treated vertebrae (5.4%). There was no statistically significant difference (p = 1). We divided the total of 116 patients into three groups according to the number of unhealed fractures: (group 1, one vertebra [n = 65]; group 2, two vertebrae [n = 37]; and group 3, three or more vertebrae [n = 14]). There was a significant difference among these groups in the occurrence of subsequent compression fractures among unhealed vertebral fractures (Figs. 3 and 4). We found a significant difference between groups 1 and 3 (p = 0.01), but there were no significant differences between groups 1 and 2 (p = 0.2) or groups 2 and 3 (p = 0.1). We found a significant difference between group 1 and group 2 plus group 3 (p = 0.04). Using a test for trend, we found a positive trend for the probability of subsequent fractures in relation to the number of unhealed vertebral fractures (p = 0.04). We confirmed our finding by two-sided analysis using multivariate analysis (Table 1). In multivariate logistic regression after adjusting for age, sex, the use of steroids, and the number of chronic vertebral fractures, the number of unhealed vertebral fractures was still highly significant (p = 0.02; odds ratio, 5.2 [95% CI, ]) in relation to the number of fractures. Discussion Kobayashi et al. [20] have recently described that prophylactic treatment of nonfractured vertebrae adjacent to fractured vertebrae could decrease the incidence of subsequent fractures. However, they did not discuss the risk factors after prophylactic percutaneous vertebroplasty. In this study, we found a relationship between the number of unhealed fractures and the occurrence of subsequent compression fractures after prophylactic percutaneous vertebroplasty. Patients with three or more fractures had a greater incidence of subsequent fracture (odds ratio, 5.2) than did patients with one Table 1: Results of Multivariate Logistic Analysis Variable Odds Ratio 95% CI p No. of unhealed fractures.04 1 Vertebra Vertebrae Vertebrae Age Sex (male) Steroid use No. of chronic fractures Note Dashes indicate that comparison was not made. fracture. Voormolen et al. [9] have also reported that the presence of multiple fractures is one of the risk factors for subsequent fractures [22]. Our results are in correspondence with their findings. In the level of vertebrae adjacent to fractured vertebrae, prophylactically treated vertebrae adjacent to unhealed fractures had a lower incidence of subsequent fracture (p < 0.001) than did nonprophylactically treated vertebrae adjacent to unhealed fractures. Prophylactic treatment was found to confer effective protection against subsequent fracture in adjacent vertebrae. In addition, there was no statistically significant difference between the incidence of subsequent fractures in the level of vertebrae two above or below fractured vertebrae in vertebrae adjacent to prophylactically treated vertebrae and the incidence in those adjacent to nonprophylactically treated vertebrae. Therefore, there was no increased incidence of subsequent fractures related to prophylactic treatment. Uemura et al. [24] have reported that the treatment of multiple vertebrae in one session could cause decreased PaO 2. The amount of the cement used tended to increase by increasing of the number of treatment vertebrae. We therefore limited the maximum usage of cement (up to one pack) at one session. Because of this policy, we could not treat all adjacent vertebrae, even if they were immediately above the fractured vertebra, in cases with multiple unhealed fractures. According to previous studies with cadavers, cemented vertebrae can cause increased stress on adjacent vertebrae [27]. On the other hand, on the basis of a study with cadavers, Chiang et al. [28] have indicated that prophylactic percutaneous vertebroplasty may reduce the risk of fatigue injury in adjacent intact vertebrae. Becker et al. [29] conducted a pilot randomized controlled trial for prophylactic kyphoplasty in 50 patients and concluded that the procedure does not lower the risk of subsequent fractures. However, they did not include patients with multiple compression fractures involving more than two vertebrae. According to Lindsay et al. [22], the incidence of subsequent compression fractures within 12 months of the initial fracture was 19.2% among patients with a history of spinal compression fracture and 24% in those with more than two involved vertebrae. In addition, as we discussed earlier, the presence of multiple fractures can be a risk factor for subsequent fractures [9]. Therefore, the failure to observe an improvement by prophylactic treatment in the study by Becker et al. may have been related to the low risk of subsequent fractures in their patients. To our knowledge, there has been no randomized control study in cases with multiple fractures. In the present study, patients with three or more fractures tended to have subsequent fractures, even though prophylactic percutaneous vertebroplasty was performed. However, additional investigations will be needed to confirm this tendency. The limitations of this study include its retrospective nature. To reveal the usefulness of prophylactic percutaneous vertebroplasty, a prospective randomized trial will be needed. In addition, we could not investigate other risk factors for fracture in patients with osteoporosis, including body mass index, smoking, and collagen disease [30]. In the multivariate logistic analysis, the p value for steroid usage was not significant (p = 0.06) (Table 1). This finding might be influenced by the fact that there were only nine steroid users. Because this was a retrospective study, we could not fully investigate the actual dose and duration of the steroid used. The lack of routine use of MRI to diagnose subsequent fractures may be another limitation. At our institution, it is difficult to postoperatively scan AJR:197, August

6 Kamano et al. all patients with prophylactic percutaneous vertebroplasty on a routine basis. However, we found no significant difference between the incidence of subsequent fracture in patients with one and two fractures and that in those with two and three or more fractures. We found a positive trend for the probability of subsequent fractures in relation to the number of unhealed vertebral fractures. Further studies will be needed to examine these issues and are currently ongoing. In conclusion, patients who had three or more unhealed fractures tended to have subsequent fractures even though prophylactic percutaneous vertebroplasty was performed. There was a correlation between the number of unhealed fractures and the occurrence of subsequent fractures after prophylactic percutaneous vertebroplasty. However, there was no increased risk of subsequent fractures after prophylactic percutaneous vertebroplasty compared with patients treated with percutaneous vertebroplasty without added prophylactic percutaneous vertebroplasty. References 1. Kallmes DF, Jensen ME. Percutaneous vertebroplasty. Radiology 2003; 229: 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: Evans AJ, Jensen ME, Kip KE, et al. Vertebral compression fractures: pain reduction and improvement in functional mobility after percutaneous polymethylmethacrylate vertebroplasty: retrospective report of 245 cases. Radiology 2003; 226: Grados F, Depriester C, Cayrolle G, Hardy N, Deramond H, Fardellone P. Long-term observations of vertebral osteoporotic fractures treated by percutaneous vertebroplasty. Rheumatology (Oxford) 2000; 39: Kim SH, Kang HS, Choi JA, Ahn JM. Risk factors of new compression fractures in adjacent vertebrae after percutaneous vertebroplasty. Acta Radiol 2004; 45: Trout AT, Kallmes DF, Kaufmann TJ. New fractures after vertebroplasty: adjacent fractures occur significantly sooner. AJNR 2006; 27: Trout AT, Kallmes DF, Lane JI, Layton KF, Marx WF. Subsequent vertebral fractures after vertebroplasty: association with intraosseous clefts. AJNR 2006; 27: Trout AT, Kallmes DF, Layton KF, Thielen KR, Hentz JG. Vertebral endplate fractures: an indicator of the abnormal forces generated in the spine after vertebroplasty. J Bone Miner Res 2006; 21: Voormolen MH, Lohle PN, Juttmann JR, van der Graaf Y, Fransen H, Lampmann LE. The risk of new osteoporotic vertebral compression fractures in the year after percutaneous vertebroplasty. J Vasc Interv Radiol 2006; 17: Syed MI, Patel NA, Jan S, Shaikh A, Grunden B, Morar K. Symptomatic refractures after vertebroplasty in patients with steroid-induced osteoporosis. AJNR 2006; 27: Syed MI, Patel NA, Jan S, Harron MS, Morar K, Shaikh A. New symptomatic vertebral compression fractures within a year following vertebroplasty in osteoporotic women. AJNR 2005; 26: Layton KF, Thielen KR, Koch CA, et al. Vertebroplasty, first 1000 levels of a single center: evaluation of the outcomes and complications. AJNR 2007; 28: Lin CC, Chen IH, Yu TC, Chen A, Yen PS. New symptomatic compression fracture after percutaneous vertebroplasty at the thoracolumbar junction. AJNR 2007; 28: Hiwatashi A, Westesson PL. Patients with osteoporosis on steroid medication tend to sustain subsequent fractures. AJNR 2007; 28: Lin EP, Ekholm S, Hiwatashi A, Westesson PL. Vertebroplasty: cement leakage into the disc increases the risk of new fracture of adjacent vertebral body. AJNR 2004; 25: Komemushi A, Tanigawa N, Kariya S, et al. Percutaneous vertebroplasty for osteoporotic compression fracture: multivariate study of predictors of new vertebral body fracture. Cardiovasc Intervent Radiol 2006; 29: Lin WC, Cheng TT, Lee YC, et al. New vertebral osteoporotic compression fractures after percutaneous vertebroplasty: retrospective analysis of risk factors. J Vasc Interv Radiol 2008; 19: Tanigawa N, Komemushi A, Kariya S, et al. Relationship between cement distribution pattern and new compression fracture after percutaneous vertebroplasty. AJR 2007; 189:1516 [web]; W348 W Lin CC, Wen SH, Chiu CH, Chen IH, Yu TC. The clinical influence of fluid sign in treated vertebral bodies after percutaneous vertebroplasty. Radiology 2009; 251: Kobayashi N, Numaguchi Y, Fuwa S, et al. Prophylactic vertebroplasty: cement injection into non-fractured vertebral bodies during percutaneous vertebroplasty. Acad Radiol 2009; 16: Tanigawa N, Komemushi A, Kariya S, Kojima H, Shomura Y, Sawada S. Radiological follow-up of new compression fractures following percutaneous vertebroplasty. Cardiovasc Intervent Radiol 2006; 29: Lindsay R, Silverman SL, Cooper C, et al. Risk of new vertebral fracture in the year following a fracture. JAMA 2001; 285: American College of Radiology. Practice Guideline ACR. ACR-ASNR-ASSR-SIR-SNIS practice guideline for the performance of percutaneous vertebroplasty. American College of Radiology Website. percutaneous_vertebroplasty.aspx. Published Updated Accessed May 31, Uemura A, Numaguchi Y, Matsusako M, Kobayashi N, Saida Y, Rahman M. Effect on partial pressure of oxygen in arterial blood in percutaneous vertebroplasty. AJNR 2007; 28: Rosner B. Hypothesis testing: categorical data. Fundamentals of biostatistics, 6th ed. Belmont, CA: Thomson Brooks/Cole, 2006: Hosmer DW, Lemeshow S. Model-building strategies and methods for logistic regression. Applied logistic regression, 2nd ed. New York: Wiley-Interscience, 2000: 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: Chiang CK, Wang YH, Yang CY, Yang BD, Wang JL. Prophylactic vertebroplasty may reduce the risk of adjacent intact vertebra from fatigue injury: an ex vivo biomechanical study. Spine 2009; 34: Becker S, Garoscio M, Meissner J, Tuschel A, Ogon M. Is there an indication for prophylactic balloon kyphoplasty? A pilot study. Clin Orthop Relat Res 2007; 458: Kanis JA, Borgstrom F, De Laet C, et al. Assessment of fracture risk. Osteoporos Int 2005; 16: AJR:197, August 2011