| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Mallinckrodt Institute of Radiology, Washington University Medical Center, 510 S Kingshighway Blvd, St Louis, MO 63110-1076. From the 2001 RSNA scientific assembly. Received November 23, 2001; revision requested February 12, 2002; final revision received October 28; accepted October 31. Address correspondence to L.A.G. (e-mail: gilulal@mir.wustl.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: Standardized four-view radiographs obtained during 363 vertebroplasties in 181 treatment sessions in 152 patients were reviewed (121 patients with osteoporotic fractures, 30 with malignant disease, and one with hemangioma). Four types of PMMA leakage and other potential predictors (patient age and sex, treated region, number of vertebral levels injected, preprocedural pain, PMMA volume per vertebra) were related to postprocedural pain response and midterm outcome after vertebroplasty. 
2 and Kruskal-Wallis tests were used for statistical analysis. The mean follow-up period was 8.8 months (range, 1–24 months).
RESULTS: At the time of discharge after the procedure, pain was absent after 106 of the 181 sessions (58.5%), better after 50 (27.6%), and the same after 25 (13.8%). In 258 of the 363 treated vertebral levels, at least one type of leakage was found. None of the evaluated factors were related significantly to postprocedural pain response, including PMMA leakage. Pain response at midterm outcome was strongly related to postprocedural treatment success, however (P < .001).
CONCLUSION: Small to moderate amounts of PMMA may escape from the vertebral body with no significant effect on therapeutic success. Immediate postprocedural pain relief is the best predictor of midterm clinical outcome after vertebroplasty.
© RSNA, 2003
Index terms: Interventional procedures, complications, 30.1267 • Osteoporosis, 30.56 • Spine, diseases, 30.33, 30.362, 30.56 • Spine, fractures, 30.41 • Spine, interventional procedures, 30.1267 • Spine, secondary neoplasms, 30.33 • Spine, vertebroplasty, 30.1267
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
The purpose of our study was to evaluate different types of PMMA leakage and patient-related factors in relation to clinical midterm (1–24-month) outcome after vertebroplasty.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
One hundred twenty-one patients had been treated for osteoporosis-related fractures, 30 had malignant disease, and one had hemangioma. The number of 152 does not include those patients who were excluded from vertebroplasty because it was believed that they had pain or discomfort that was not related to a treatable disease of the vertebral body or because they had neurologic symptoms. No patient was excluded because of vertebra plana. In the initial vertebroplasties we performed, patients with extensive bone destruction and/or epidural extension of tumor were excluded.
In the 152 patients, 363 vertebral bodies were treated with vertebroplasty during 181 treatment sessions. One hundred twenty-six patients underwent a single treatment session, 23 underwent two sessions, and three underwent three sessions. Between one and five vertebral levels were injected with PMMA per session (mean, two levels). With one exception, the repeat injections were not performed at the same level but at additional levels that were newly symptomatic.
Standard anteroposterior and lateral radiographs were obtained in all patients prior to vertebroplasty. Magnetic resonance imaging and/or scintigraphy was performed when necessary to determine vertebral levels that required treatment, and computed tomography (CT) was performed when necessary to predict potential sources of PMMA leakage, such as defects in the posterior wall of the vertebral body. The exact number of specific imaging examinations performed and the results were not evaluated in this investigation. Attempts were made to perform these examinations within 2–4 weeks prior to vertebroplasty. Platelet levels were determined prior to vertebroplasty in all patients. Prothrombin and partial prothrombin times were tested in all patients with potential bleeding problems, such as those who received warfarin sodium treatment (Coumadin; Bristol-Myers Squibb, New York, NY) or had any clinical bleeding problems as defined by the patient or referring clinician.
Preprocedural pain was assessed with use of a visual analogue scale. A value of 0 was defined as no pain and 100 as the worst pain the patient could imagine. In two instances, vertebroplasty was performed without the presence of acute pain to stabilize a fractured vertebra. A third patient underwent vertebroplasty in conjunction with surgical decompression. In all other patients, the range of preprocedural pain varied between 10 and 100. Mean preprocedural pain was 77.7.
Procedure
The same musculoskeletal staff radiologist (L.A.G.) who was experienced in vertebroplasty either performed the procedure personally or closely supervised a musculoskeletal resident or fellow who was trained to perform the procedure.
Vertebroplasty was performed with intravenous conscious sedation with fentanyl (Duragesic; Janssen Pharmaceutica, Titusville, NJ) and midazolam (Versed; Hoffmann-La Roche, Nutley, NJ) with the patient prone. A C-arm fluoroscopic unit was used. In fewer than six cases, a costovertebral approach was used. In the remaining cases, a transpedicular approach was used to place an 11- or 13-gauge disposable Jamshidi-type needle (such as the MD TECH bone marrow biopsy set; Medical Devices Technology, Gainesville, Fla).
Initially, with the transpedicular approach, the needle was passed through the pedicle with the needle tip placed in the anterior half of the vertebral body. For the last 233 of 363 injections, however, a modification of this technique was consistently attempted. With increased obliquity of needle entry through the pedicle, the needle tip was placed in the anterior one-fourth to one-third of the vertebral body in the lateral view and in the midline of the vertebral body in the frontal view. Unilateral injections were performed in 244 vertebral bodies, and bilateral injections were performed in 118. In one instance, a fractured vertebra was filled from the adjacent vertebral body through an anterior bone bridge. Bipedicular injections were used when early leakage was detected during the first injection or when only part of the vertebral body could be filled by means of unilateral injection.
Cranioplastic (Plastics One, Roanoke, Va) was used in 45 of the first 46 treatment sessions, and Osteobond (Zimmer, Warsaw, Ind) was used in the remaining 136 sessions. One slightly heaping teaspoon (7 g) of dry heat–sterilized barium and 1.2 g of tobramycin (Nebcin; Eli Lilly, Indianapolis, Ind) were added to each bag of methylmethacrylate powder. Powder volume greater than that of the original methylmethacrylate was poured off. One-half of a bag was mixed with one-half of the liquid monomer for each injection. Injection of the liquid PMMA was stopped when the vertebral body was filled; when PMMA passed to the posterior fourth of the vertebral body; when PMMA entered the disk, epidural space or veins, foramen, or paravertebral space or veins; or when patients experienced radicular pain. During the procedure, the radiologist who performed the procedure also reported any recognized embolization of PMMA. Although the types of leakage were recognized by the person performing the vertebroplasty, the musculoskeletal radiologist (J.H.) not involved with the procedure later recorded the presence of these types of leakage.
In the 363 vertebroplasties performed, there were 101 thoracic (T1–10), 182 thoracolumbar (T11–L2), and 80 lumbar (L3–S1) vertebral levels injected. Between the third thoracic and the first sacral (a lumbarized variant) vertebra, each level was injected at least once. The most commonly injected vertebrae were L1 (n = 56), T12 (n = 55), L2 (n = 41), L3 (n = 33), and L4 (n = 32). One injection filled two vertebra through an anterior bone bridge. The mean injected volume of PMMA per vertebra was 8.3 mL (range, 1.25–20.25 mL). For the thoracic spine, the mean injected volume was 6.4 mL (range, 1.25–12.75 mL); for the thoracolumbar spine, 8.9 mL (range, 1.75–20.25 mL); and for the lumbar spine, 11.1 mL (range, 3.75–18.50 mL).
Postprocedural Evaluation
At the end of the procedure, each patient was monitored for 1 hour in an outpatient medical procedures observation unit. Pain levels were then determined after the patient could walk before discharge from the hospital. The only exceptions to this immediate discharge were a small number (<12) of patients who had already been hospitalized (most had metastastic disease, and one drove to our institution from out of state for treatment). Immediately after the procedure, the same visual analogue scale was applied by either a treating physician or a nurse in the medical procedures observation unit, as was applied before the procedure. During this study, these postprocedural visual analogue scales were converted to evaluate if pain was worse, the same, better, or completely gone compared with that prior to the procedure. This assessment related to the original pain and not to additional pain, such as transient nerve root symptoms or symptoms relating to rib fractures. Ordinal numbers were attributed to this postprocedural response (worse, 0; same, 1; better, 2; and completely gone, 3). This newer response score of 0–3 was used because it was found that for many patients, it was difficult to quantify pain levels of 0–100. Patients could respond more readily and objectively to the 0–3 scoring system.
Two years after the first vertebroplasties were performed, a cross-sectional follow-up study was performed. A research technologist administered a telephone questionnaire to ask each patient if the pain or other symptoms were worse, the same, better, or completely gone compared with those prior to the procedure. The same ordinal numbers used in the postprocedural assessment were applied. In 121 of the 181 treatment sessions, such follow-up data were evaluated. Thirty-one patients had died, and 29 vertebroplasty sessions were followed by an additional procedure. Only the final treatment session for each of these patients who had multiple treatment sessions was evaluated at follow-up. The mean follow-up period was 8.8 months, with a range of 1–24 months. The term midterm was believed to best represent this type of follow-up with a mixture of long (24-month) and relatively short (1-month) periods.
Image Evaluation
All of the standard four-view radiographs obtained during vertebroplasty were reviewed by the same musculoskeletal radiologist who had been familiarized with the institution’s vertebroplasty techniques but who was not involved in patient care. A classification scheme was established prior to formal review of the radiographic images. The following types of PMMA leakage were determined for all 363 vertebroplasties: intradiskal, epidural, posterolateral (foraminal region), and paravertebral (anterior, anterolateral, and lateral). There was a choice of no, minor, moderate, or major leakage. Maximum diameter of the leakage was related to adjacent normal structures to compensate for noncontrolled projectional effects (Fig 1). Minor leakage was considered to be present when the longest diameter of the extravertebral PMMA collection was less than the longest diameter of the closest pedicle; moderate, when the leakage was greater than the longest diameter of the pedicle but less than the craniocaudal diameter of the nearest normal vertebral body; and major, when the leakage was larger than the craniocaudal diameter of the nearest normal vertebral body.
|
Statistical Evaluation
For statistical evaluation, all cases were classified as "no leakage present" or "leakage present." Minor and moderate leakages were combined into the "leakage present" category for each of the four types of leakage (intradiskal, epidural, posterolateral, and paravertebral). The case of hemangioma was added to the malignancy category for the purpose of statistical calculations.
Predictors for postprocedural response (grade 0, pain worse; 1, pain the same; 2, pain better; and 3, pain gone) were determined for the last vertebroplasty session in each of the 152 patients. The predictors assessed were patient age and sex, treated region, underlying vertebral body abnormality, number of vertebral levels injected, preprocedural pain (percentage), PMMA volume injected per vertebra (in milliliters), and leakage type (intradiskal, epidural, posterolateral, or paravertebral). 2 and Kruskal-Wallis tests were used for evaluation of these parameters.
For description of midterm outcome, the same type of statistical analysis was applied in 121 of the 181 treatment sessions that were not excluded because of either patient death or the performance of additional vertebroplasty sessions. The effects for this model were the same as those mentioned previously, in addition to follow-up period (months) and postprocedural pain assessment. Significance levels were set at P = .05. JMP software (version 3.1; SAS Institute, Cary, NC) was used for calculations.
Institutional review board approval was obtained for the review of patient charts and images and for follow-up investigation, including telephone interviews. Informed consent was obtained from patients prior to vertebroplasty.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
|
Eighty-one (66.9%) of 121 patients gave the same assessment after the procedure and at follow-up. Thirty-seven (30.6%) had a difference of one grade (10 [8.3%] with further improvement and 27 [22.3%] with worsening of pain compared with the postprocedural response). Three (2.5%) had a difference of two grades (all three had no pain after the procedure but recurrence of their original symptoms at follow-up).
Patient age and sex, treated region, number of injected vertebral levels, injected volume of PMMA per vertebra, type of abnormality, and type of disk leakage were not predictors of postprocedural pain response (Table 2). The duration of the follow-up period was 9.6 months for the patients without pain, 7.6 months for those who felt better, and 11.4 months for those who had the same symptoms before and after the procedure. Although these numbers are significantly different (P = .03), they do not indicate a consistent influence on midterm outcome (Table 2). However, there was a significant relationship between postprocedural response and midterm outcome (P < .001).
|
Four patients experienced clinically relevant rib fractures. One of them had no immediate improvement from vertebroplasty and died as a result of malignancy. One patient felt better immediately after vertebroplasty with regard to original pain, one felt better at 3-month follow-up, and one was pain free both immediately after the procedure and at follow-up. Two patients who underwent thoracic vertebroplasty (T3, T4, T7, and T11 in one patient; T10–12 and L1 in the other) had small pneumothoraces that were detected at the end of the procedure. Both patients were asymptomatic. One of these pneumothoraces at the T3 level occurred when the needle was placed in the pedicle by the senior radiologist and then advanced by a resident through the lateral wall of the vertebral body. It was uncertain how pneumothorax occurred in the other patient. When the case was reviewed, at least one of the needles placed at the T10-L1 levels may have been advanced into the vertebral body without sufficient medial obliquity of the needle. Neither of those patients needed additional treatment. Both were pain free immediately after the procedure and at 12 months.
One patient had the clinical symptoms of noninfectious diskitis after PMMA leakage into a disk. Biopsy was not performed to confirm the diagnosis. Vertebroplasty was not successful in this patient both after the procedure and at 24 months. As soon as PMMA passed into the disk, pain was produced. After the procedure, the patient could not flex her hip because of pain. The referring physician believed strongly that, in retrospect, this patient’s symptoms were due to disk-related symptoms rather than pain from her fracture. One patient sustained a pedicular fracture, which healed spontaneously. Vertebroplasty was successful with regard to treatment of original pain in this patient.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
Not all PMMA leakages are clinically relevant. Cotten et al (8) found nerve root compression that required surgery in two of eight foraminal leakages and transient femoral neuropathy in one of 21 paravertebral leakages. In the series of Heini et al (6), no clinical symptoms were related to leakages. Depriester et al (12) found nerve root compression that required surgery in 1.5% of injections, with radicular symptoms in up to 8%. Such symptoms appear to be more common in neoplastic disease. Spinal cord compression appears to be rare (9,12). In our series, no surgical decompression was required. One patient with metastatic disease had nerve root symptoms, which disappeared overnight after treatment with intravenous steroids. Ten patients underwent 10 fluoroscopically directed nerve blocks during the follow-up period for symptoms that were not necessarily related to vertebroplasty.
The increased number of intradiskal leakages found with bilateral pedicular injections may be related to the fact that when PMMA passed into the disk and it was believed that not enough PMMA was present to be satisfactory for that vertebral body, the opposite pedicle was entered. Moreover, such needle placement was more common in the severely collapsed vertebrae, which are technically more difficult to treat.
Interestingly, leakages did not occur more commonly in neoplastic disease when compared with those in osteoporosis in our investigation. A possible explanation for this somewhat unexpected result may relate to the preprocedural evaluation, in which CT was more commonly used for assessment of metastatic disease and may have shown more potential sites of PMMA leakage. Also, because of the fact that the radiologists were more concerned about leakage in patients with metastases, injection was stopped as soon as a patient started to develop even questionable symptoms or as soon as the PMMA approached a region known to be at risk for PMMA leakage. Also, although unproven, the use of conscious sedation rather than general anesthesia, as used occasionally (9), may provide an early warning as to the development of neurologic symptoms. Detection of early radicular symptoms is a strong indication that the position of PMMA should be checked immediately, and the injections may need to be stopped.
Vertebroplasty may successfully reduce or eliminate pain relating to vertebral abnormalities (4,6,8,11–13). Pain reduction has not been assessed uniformly in the literature. Most authors use qualitative gradings, as were used in our population. Patient assessments, in our experience, are commonly more consistent when broad categories are provided to choose from, such as no improvement, some improvement, and complete disappearance of pain, as opposed to the more sophisticated visual analogue scale. All investigators, however, agree that pain relief is quick and relevant in most patients (4,6,8,11–13), which has also been demonstrated in our patients.
Our statistical model did not allow us to identify predictors of postprocedural pain response for our study population. This includes data such as patient age, underlying abnormality, volume of injected PMMA, preprocedural pain, and leakage detected on radiographs. The patients’ postprocedural pain assessments may have been influenced by residual effects of conscious sedation and local anesthesia. Differences could have appeared after a few days but were not assessed in our study.
Several authors have indicated that the pain reduction achieved initially remains stable during follow-up periods of different duration. Weill et al (4) indicated that such stability was present in 73% of their patients after 6 months. Heini et al (6) found stable results after 1 year by using a visual analogue scale. Our results are in line with those in such publications, since pain assessment was not commonly changed at follow-up when compared with the immediate postprocedural assessment. This fact may be used for patient information and guidance. However, patients referred for vertebroplasty are in danger of developing additional fractures as a result of underlying disease, such as osteoporosis or malignancy.
In our investigation, the amount of PMMA injected per vertebra was not significantly related to follow-up results. A relationship with PMMA volume may be anticipated by those authors who stress mechanical stabilization of the vertebral body (2). Our results do not support or reject the question that the amount of PMMA filling is an important outcome factor because the variability of PMMA volume in our investigation may be explained by the variability of the vertebral body size (thoracic vs lumbar), different vertebral body composition (severe vs mild osteoporosis), or collapsed versus maintained shape rather than percentage filling of the vertebral body.
The various types of PMMA leakages did not impair midterm outcome. However, the leakages found in our investigation were relatively small, and the effects of leakages may be more pronounced in patients with major extravertebral PMMA collections, such as those demonstrated in the literature (8,12).
Study limitations include the fact that although data were collected systematically, the design of the follow-up investigation has to be considered retrospective. We also realize that inclusion of only the last vertebroplasty session for follow-up in patients with multiple sessions represents a potential selection bias. However, this did not exclude treatment failures from follow-up because the additional vertebroplasty sessions were performed in vertebral levels that were not treated during the prior session. A rigorous analysis of the data in all 181 vertebroplasty sessions should be performed by using a repeated measures analysis with a polytomous logistic regression model. This would require sophisticated methods that are appropriate for designs with use of only one procedure for 82.9% of patients (126 of 152) and use of up to three procedures for the remaining patients. Although our analysis was not performed by using such a procedure because of the lack of software accessability, appropriate measures, as stated, were taken to minimize problems that might arise by not using the repeated measures analysis.
In conclusion, on the basis of results in our experience with 363 injected vertebral levels, vertebroplasty is a successful pain treatment method in osteoporotic and metastatic disease of the spine. Small to moderate amounts of PMMA may escape from the vertebral body with no significant effect on therapeutic success. Immediate postprocedural pain relief is the best predictor of midterm outcome of vertebroplasty.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
3 Current address: Gem State Radiology, Boise, Idaho.
Abbreviation: PMMA = polymethylmethacrylate
Author contributions: Guarantor of integrity of entire study, L.A.G.; study concepts and design, J.H., L.A.G.; literature research, J.H.; clinical studies, D.P., L.A.G.; data acquisition, D.P., L.A.G.; data analysis/interpretation, D.P., J.H., L.A.G.; statistical analysis, J.H.; manuscript preparation and definition of intellectual content, J.H., L.A.G.; manuscript editing, L.A.G.; manuscript revision/review, L.A.G., D.P.; manuscript final version approval, L.A.G.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTS DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  N. Tanigawa, A. Komemushi, S. Kariya, H. Kojima, Y. Shomura, N. Omura, and S. Sawada Relationship Between Cement Distribution Pattern and New Compression Fracture After Percutaneous Vertebroplasty Am. J. Roentgenol., December 1, 2007; 189(6): W348 - W352. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. B. Morrison, L. Parker, A. J. Frangos, and J. A. Carrino Vertebroplasty in the United States: Guidance Method and Provider Distribution, 2001-2003 Radiology, April 1, 2007; 243(1): 166 - 170. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H.J. Cloft and M.E. Jensen Kyphoplasty: An Assessment of a New Technology AJNR Am. J. Neuroradiol., February 1, 2007; 28(2): 200 - 203. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T.J. Kaufmann, A.T. Trout, and D.F. Kallmes The effects of cement volume on clinical outcomes of percutaneous vertebroplasty. AJNR Am. J. Neuroradiol., October 1, 2006; 27(9): 1933 - 1937. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Tanigawa, A. Komemushi, S. Kariya, H. Kojima, Y. Shomura, K. Ikeda, N. Omura, T. Murakami, and S. Sawada Percutaneous Vertebroplasty: Relationship between Vertebral Body Bone Marrow Edema Pattern on MR Images and Initial Clinical Response Radiology, April 1, 2006; 239(1): 195 - 200. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. K. Singh, T. K. Pilgram, and L. A. Gilula Osteoporotic Compression Fractures: Outcomes after Single- versus Multiple-Level Percutaneous Vertebroplasty Radiology, January 1, 2006; 238(1): 211 - 220. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. M. Barragan-Campos, J.-N. Vallee, D. Lo, E. Cormier, B. Jean, M. Rose, P. Astagneau, and J. Chiras Percutaneous Vertebroplasty for Spinal Metastases: Complications Radiology, January 1, 2006; 238(1): 354 - 362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. G. Hadjipavlou, M. N. Tzermiadianos, P. G. Katonis, and M. Szpalski Percutaneous vertebroplasty and balloon kyphoplasty for the treatment of osteoporotic vertebral compression fractures and osteolytic tumours J Bone Joint Surg Br, December 1, 2005; 87-B(12): 1595 - 1604. [Full Text] [PDF]
|
|
|
|
|
|
N. Tanigawa, A. Komemushi, S. Kariya, H. Kojima, and S. Sawada Intraosseous Venography with Carbon Dioxide Contrast Agent in Percutaneous Vertebroplasty Am. J. Roentgenol., February 1, 2005; 184(2): 567 - 570. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Mckiernan, T. Faciszewski, and R. Jensen Quality of Life Following Vertebroplasty J. Bone Joint Surg. Am., December 1, 2004; 86(12): 2600 - 2606. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Y. Carlier, H. Gordji, D. M. Mompoint, N. Vernhet, A. Feydy, and C. Vallee Osteoporotic Vertebral Collapse: Percutaneous Vertebroplasty and Local Kyphosis Correction Radiology, December 1, 2004; 233(3): 891 - 898. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. H. Choe, E. M. Marom, K. Ahrar, M. T. Truong, and J. E. Madewell Pulmonary Embolism of Polymethyl Methacrylate During Percutaneous Vertebroplasty and Kyphoplasty Am. J. Roentgenol., October 1, 2004; 183(4): 1097 - 1102. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Perisinakis, J. Damilakis, N. Theocharopoulos, G. Papadokostakis, A. Hadjipavlou, and N. Gourtsoyiannis Patient Exposure and Associated Radiation Risks from Fluoroscopically Guided Vertebroplasty or Kyphoplasty Radiology, September 1, 2004; 232(3): 701 - 707. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Appel and L. A. Gilula Percutaneous Vertebroplasty in Patients with Spinal Canal Compromise Am. J. Roentgenol., April 1, 2004; 182(4): 947 - 951. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
