| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Vascular and Interventional Radiology |
1 From the Mallinckrodt Institute of Radiology, Washington University School of Medicine, 510 S Kingshighway Blvd, Campus Box 8131, St Louis, MO 63110. Received March 4, 2003; revision requested May 23; final revision received December 18; accepted January 29, 2004. Address correspondence to J.S.S. (e-mail: shimonyj@mir.wustl.edu).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
MATERIALS AND METHODS: PVP was performed in 50 patients with metastatic disease or multiple myeloma between June 1998 and April 2002. Twenty-five women (mean age, 62.3 years; range, 38–85 years) and 25 men (mean age, 63.1 years; range, 37–92 years) were included. Cases were retrospectively reviewed. Patients who had undergone cross-sectional imaging were classified into three groups. First group had no epidural involvement; second group, mild epidural involvement without contact with spinal cord or nerve roots; third group, moderate involvement and contact with spinal cord or nerve roots. Procedural safety was evaluated with review of all post-PVP complications and their treatment. Effectiveness was evaluated with follow-up phone calls for assessment of change in pain level and activity after PVP. Follow-up calls were performed at 1 day; 2 weeks; 1, 3, and 6 months; and 1 and 2 years. Differences between groups were assessed with singly ordered Kruskal-Wallis test.
RESULTS: Fourteen patients were classified in the first group, 18 in the second, and 18 in the third. There were no significant differences in pain or mobility outcomes among groups. At the last follow-up call, 41 (82%) of 50 patients reported improvement in pre-PVP pain. Six (12%) reported no change, and three (6%) reported increased pain. After PVP in 26 (52%) patients, there was a period of increased mobility; in 19 (38%), no improvement in mobility occurred; and in five (10%), decreased mobility was reported. Complications included acute increased pain or new areas of pain in seven (14%) patients. None of these required surgery; four were treated with nerve root block; two, with central epidural injection; and one, with overnight intravenous steroids.
CONCLUSION: PVP can be performed safely and effectively with conscious sedation in patients with malignant compression fractures and epidural involvement.
© RSNA, 2004
Index terms: Spine, CT, 30.1211 • Spine, fractures, 30.416 • Spine, interventional procedures, 30.1267 • Spine, MR, 30.121411 • Spine, secondary neoplasms, 30.33 • Spine, vertebroplasty, 30.1267
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The risk of complications from PVP performed in patients with metastatic disease and multiple myeloma is increased, compared with that in patients with osteoporosis (3,5–7,9,12,18,19). In these patients there are likely to be osteolytic areas with destruction of the bone cortex, and this destruction increases the risk for symptomatic leakage of copolymer bone cement (polymethylmethacrylate [PMMA] copolymer powder and methylmethacrylate monomer) into the spinal canal and neural foramina; hereafter, the bone cement mixture will be referred to as PMMA.
Patients in whom performance of PVP is contraindicated include asymptomatic patients, surgical candidates, patients with pain unrelated to the fracture, and patients with infection. Several authors (20) describe symptomatic spinal cord compression as a contraindication to the performance of PVP because of the risk of increasing these symptoms. In the absence of direct compressive symptoms, destruction of the posterior vertebral body wall or epidural extension of tumor has been described as a relative contraindication to the procedure (20–24). Several researchers reported that they performed the procedure in a series of patients with destruction of the posterior vertebral body wall for analgesia and spinal stabilization (3,6,25), in those in whom conservative therapy failed, in those in whom no other pain control options were available, and in those in whom anticipated survival was limited. To our knowledge, no articles have been published in which safety and effectiveness of performance of PVP was compared in patients who had cancer with different degrees of involvement in the epidural space.
Thus, the purpose of our study was to evaluate the safety and effectiveness of performance of PVP in patients with malignant compression fractures and involvement of the epidural space.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The remaining 50 patients, 25 women and 25 men (age range, 37–92 years; mean age, 62.7 years ± 14.0 [standard deviation]) were included in our sample. Among the women, the age range was 38–85 years, with a mean age of 62.3 years, and among the men, the age range was 37–92 years, with a mean age of 63.1 years. Of the 50 patients, 36 had metastatic disease to the spine and 14 had multiple myeloma. The primary cancers among the patients with metastatic disease included the following: lung cancer in 13; breast cancer in eight; prostate cancer in four; lymphoma in three; unknown in three; and adenocarcinoma, pancreatic cancer, neuroendocrine cancer, bladder cancer, and uterine cancer in one each. These patients were treated in 60 sessions (in 10 patients the procedure was performed twice on different levels) on a total of 129 levels. Table 1 includes various patient characteristics.
|
Patients in whom cross-sectional images had been obtained were classified into three groups on the basis of the amount of epidural involvement. This classification scheme is illustrated in Figure 1. Group 1 had no epidural involvement and provided a comparison as a control group with the more severely affected groups. Group 2 had mild epidural involvement but no contact with the spinal cord or nerve roots on the basis of findings at cross-sectional imaging. Group 3 had moderate epidural involvement, with contact between the tumor and the spinal cord or nerve roots. For patients who had involvement of more than one level, the grading was based on the worst (highest grade) level.
|
|
|
|
|
|
|
|
|
A small skin incision was made, and an 11-gauge Jamshidi-type trocar (MDTech, Gainesville, Fla) for bone biopsy was advanced to the lamina posterior to the pedicle. With fluoroscopic guidance, the needle was advanced through the cortex and through the pedicle into the vertebral body. The needle was advanced until its tip was located at the junction of the anterior one-fourth to one-third of the vertebral body.
The stylette was removed from the trocar, and intraosseous "blush" venography was performed with 1–2 mL of iohexol (Omnipaque 180; Nycomed, Princeton, NJ) to evaluate needle positioning (26). The needle was adjusted so that bone trabeculae opacified before venous filling to decrease the risk of direct venous embolization. Seven grams of dry heat-sterilized barium sulfate (E-Z-Em, Westbury, NY) (27) was broken into fine particles and mixed with PMMA copolymer powder (Osteobond; Zimmer, Warsaw, Ind). During approximately the 1st year of performing PVP, 1.2 g of tobramycin sulfate (Nebcin; Eli Lilly, Indianapolis, Ind) was also added to the mixture for prophylaxis. Later, 1 g of cefazolin sodium (Ancef; SmithKline Beecham, Philadelphia, Pa) instead of the tobramycin was administered intravenously just prior to PVP. The methylmethacrylate monomer was added to the PMMA copolymer powder and mixed into a smooth liquid consistency.
The PMMA was loaded into a 20-mL syringe that in turn was loaded into a screw-type 10-mL syringe (LeVeen; Boston Scientific, Watertown, Mass) or a 10-mL syringe with hub adapter (28). The PMMA in the 20-mL syringe was placed in a container that held cold sterile saline. With the 10-mL syringe, the PMMA was injected into the vertebral body with lateral fluoroscopic guidance until the it reached the posterior fourth of the vertebral body or until it extended into the disk space or paravertebral tissues. The injection was stopped if the patient started complaining of any pain, such as radicular pain, that could be due to pressure on posterior neurologic structures. If the injection became difficult because of high resistance, the PMMA was back loaded into 1-mL strong-walled syringes (Medallion; Merit Medical System, South Jordan, Utah) to continue the injection to fill as much of the vertebral body and tumor as possible. If the injection filled only one side of the vertebral body, a second needle was used to enter the other pedicle, and the procedure was repeated. Figures 2–5 provide examples of post-PVP fluoroscopic images.
Safety and Outcome Evaluation
Retrospective review of the patients’ records was performed by one of the authors (J.S.S.). The safety of the procedure was evaluated by noting any acute post-PVP symptomatic complications and their treatment. These had been recorded prospectively on a patient data form in a special examination folder prepared for each patient. Asymptomatic local complications such as leakage of PMMA outside of the vertebral body were not counted for the purposes of this study.
The clinical outcome was assessed with a telephone questionnaire approved by the local institutional review board. The effectiveness was evaluated by means of follow-up phone calls to assess change in pain level, mobility, and activity after PVP. Patients were questioned about their pain level prior to and immediately following the procedure, and the patients used a visual pain intensity scale method to assess their pain level on a scale of 0–10, with 0 being no pain and 10 indicating the worst pain. After PVP, the patient’s pain level was evaluated with follow-up phone calls by using a scale of 0–10. The visual component could not be used over the phone. The calls were made after the procedure at 1 day; 2 weeks; 1, 3, and 6 months; and 1 and 2 years. The patients were also asked if their pain was absent, improved, the same, or increased compared with the pain prior to the procedure. The follow-up evaluation was performed in this manner, as many patients had difficulty in using the visual analog scale over the phone. When we evaluated pain level, the information from the last interview was used for the purposes of data analysis.
The patients were also asked about their activity level and mobility in performance of day-to-day tasks, such as moving around the house, personal care activity, and meal preparation. The patients were asked if their abilities in these tasks improved, remained the same, or decreased as compared with their abilities during the pre-PVP period. When we analyzed activity level, we used the best mobility level that the patients reported in any of the interviews following the procedure, since most patients reported decreased activity and mobility level at the last follow-up during the terminal stage of their disease. Use of only the results from the last interview would be a misrepresentation of the improvements that many patients had prior to their final decline.
Statistical Analysis
To avoid interactions in the statistical analysis among study participants who received treatment at multiple levels, only one entry was allowed per patient. Patients who received treatment at multiple levels were classified on the basis of their worst level, and this was not changed during follow-up. Statistical testing was performed with statistical software (JMP, version 5.0, SAS Institute, Cary, NC; StatXact, version 5.0, Cytel Software, Cambridge, Mass). Statistical analysis was performed to compare distribution differences between the groups on the basis of sex with the 
2 test and those between groups on the basis of age with analysis of variance. Differences in outcomes between the groups were tested with the singly ordered Kruskal-Wallis test, since the groups were not ordered; however, the pain relief and mobility scales were ordered by severity. In all cases, the significance of the test was P = .05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
2 test, there was no statistically significant difference in the distribution of patients between the groups on the basis of sex (P = .45). With analysis of variance, no significant differences were demonstrated in the distribution of patients between the groups on the basis of age (P = .70). The time distribution of the follow-up phone calls among the 50 patients was skewed toward short periods because of the short life expectancy in this population. Only five of the 50 patients were alive for the last follow-up call at 2 years. The final phone calls to the patients’ homes prior to their death were made at the following times: five at 1 year, nine at 6 months, eight at 3 months, 18 at 1 month, and five at 2 weeks. The mean and median follow-up times were 23.8 weeks and 3 months, respectively.
Table 2 summarizes the post-PVP change in pain level at the time of the last follow-up for patients classified into groups according to the degree of epidural involvement. Among all patients, 18 (36%) had complete resolution of pre-PVP pain, and 23 (46%) had some pain improvement, for a total of 41 (82%) with pain improvement. Six (12%) of the patients had no change in pain level, and three (6%) reported increased pain. Two of the three who reported increased pain after PVP had periods of improvement at earlier follow-up phone conversations and reported increased pain, which may have been related to disease progression, only at their most recent follow-up. It is important to recognize that the increased pain reported in Table 2 was chronic and not related to acute or subacute increased pain that was also monitored for evaluation of the safety of PVP. Statistical evaluation of data in Table 2 with the singly ordered Kruskal-Wallis test demonstrated no statistically significant difference between the groups (asymptotic confidence level, P = .46).
|
|
The remaining four patients reported increased pain several weeks after PVP and were considered in the subacute complication group. All patients in this group were treated with neuroforaminal epidural nerve root block, with some immediate relief. One of these patients had increased pain at the last follow-up call and was included in the "pain increased" column of Table 2. Given the delay between PVP and the reported complications in this subacute group, the increase in pain could have been from other causes such as progression of disease. In two of these patients, the nerve root block was performed at one level away from the level at which PVP was performed, with relief of symptoms.
Of the seven patients with acute or subacute complications, none were in group 1, three were in group 2, and four were in group 3. Although these findings are suggestive of an increased risk for complications in patients with epidural involvement, a comparison of the number of patients with complications with the number of those without complications among the three groups with the 2 test demonstrated no statistically significant difference among the groups (P = .18). This result was likely due to the small numbers involved.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
The presence of neurologic symptoms from nerve root or spinal cord compression, other than localized pain, is accepted by most authors as a contraindication to the performance of PVP (20,30,31). The presence of destruction of the posterior vertebral wall or epidural retropulsion without direct compressive symptoms has been described as a relative contraindication to the procedure (20–24), since there is increased risk of symptomatic extravasation of PMMA into the spinal canal or neural foramina.
Our experience with performance of PVP in patients with metastatic disease or multiple myeloma with epidural involvement does not confirm these fears. When the patients are screened, when those with neurologic deficits that relate to spinal cord and nerve root compression are excluded, and when PVP is performed as described previously, the number of complications in this group of patients and the degree of reported pain relief are not substantially different from those reported in other studies that include series of patients with metastatic disease and multiple myeloma. For example, Cotten et al (3) reported a decrease in pain level in 75% of patients at 6-month follow-up and a complication rate of 8% in a series of 37 patients with osteolytic metastases and myeloma.
Similarly, Weill et al (6) reported a decrease in pain level in 73% of patients at 6 months and a complication rate of 14% in a series of 37 patients with spinal metastases. Our results of 82% pain relief or pain improvement are similar to those of Weill et al and to those of other researchers who reported 80%–90% pain relief in patients with benign osteoporotic compression fractures treated with PVP (32). These results are confirmed in our study on the basis of a comparison of the results obtained in group 1, which included patients who would be considered by most practitioners to be good candidates for PVP, with the results obtained in groups 2 and 3, which included patients who would be excluded by some practitioners on the basis of current guidelines (21).
Our study results demonstrate that there are no statistically significant differences between the outcomes among the groups. Although the difference in complication rates between the groups is not statistically significant, the number of complications is small and there is a trend for more complications in the groups with epidural involvement. In general, our complications have been minor, and no patients required surgery. Our data indicate that the increased risk for these patients is small. In addition, the benefits of the procedure for patients with severe pain, few other treatment options, and a short life expectancy can be substantial, even in clinical situations that have been cited in the literature as contraindications to performance of PVP.
Caution should be applied to analysis of immediate and long-term follow-up results in this population, which are not always representative of outcomes of the procedure. Many patients report a pattern of early benefit, with a subsequent deterioration in their condition near the end of life. It is reasonable to assume that these patients benefited from the procedure until their advancing disease negated these improvements.
We believe that performance of the procedure with conscious sedation provides an extra measure of safety because patients are able to tell us if any pain develops during the injection of the PMMA. If any type of pain developed during the procedure, especially pain of a radicular type, the location of the PMMA was immediately checked on both the frontal and lateral views. Midline pain was thought to be related to the fracture pain during the injection, although if the PMMA was extending to the posterior aspect of the vertebral body, further injection of it was stopped.
Other than the PVP procedures performed in our first few patients, all were performed with the same PMMA, which was injected when very liquid. Injection of liquid PMMA allows it to pass easily through tumor tissue and, we believe, decreases the risk of pushing tissue posteriorly. This statement needs confirmation with findings of others who use a more viscous type of PMMA.
Several limitations of this study were inherent in the short life span and debilitated state of our patient population. The follow-up for many of the subjects was short, and the rapid progression of disease could mask both benefits and risks of the procedure. Because of the difficulty associated with performance of the telephone interviews with many of these debilitated patients, we did not use the visual analog scale for pain after they left the hospital. We instead used simplified pain and activity scores, which have not been validated in other studies. The small number of patients with complications makes it difficult to draw firm conclusions about the extra risk involved in performance of PVP in this population. Although there appears to be a trend for more complications in groups 2 and 3, this trend is not statistically significant. A final limitation relates to the limited number of CT scans, as compared with MR images, available to us. Ideally, it would be better to have both types of images available for every patient to help decide grouping of patients in equivocal cases and to better judge the degree of bone destruction of the posterior wall of the vertebral body as a possible additional risk criterion.
In conclusion, we believe epidural involvement should not be a contraindication to performance of PVP in patients who have cancer with intractable pain, few other treatment options, and a short life expectancy when they are properly screened and PVP is performed with conscious sedation.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: PMMA = polymethylmethacrylate, PVP = percutaneous vertebroplasty
Author contributions: Guarantor of integrity of entire study, J.S.S.; study concepts and design, J.S.S., L.A.G.; literature research, J.S.S.; clinical studies, all authors; data acquisition, all authors; data analysis/interpretation, J.S.S., L.A.G.; statistical analysis, J.S.S.; manuscript preparation, J.S.S.; manuscript definition of intellectual content and editing, J.S.S., L.A.G.; manuscript revision/review and final version approval, all authors
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODS RESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R.J. McDonald, A.T. Trout, L.A. Gray, A. Dispenzieri, K.R. Thielen, and D.F. Kallmes Vertebroplasty in Multiple Myeloma: Outcomes in a Large Patient Series AJNR Am. J. Neuroradiol., April 1, 2008; 29(4): 642 - 648. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Hiwatashi and P.-L.A. Westesson Vertebroplasty for Osteoporotic Fractures with Spinal Canal Compromise AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 690 - 692. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B.A. Georgy and W. Wong Plasma-Mediated Radiofrequency Ablation Assisted Percutaneous Cement Injection for Treating Advanced Malignant Vertebral Compression Fractures AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 700 - 705. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Calmels, J.-N. Vallee, M. Rose, and J. Chiras Osteoblastic and Mixed Spinal Metastases: Evaluation of the Analgesic Efficacy of Percutaneous Vertebroplasty AJNR Am. J. Neuroradiol., March 1, 2007; 28(3): 570 - 574. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. C. Wiggins, M. Sehizadeh, T. K. Pilgram, and L. A. Gilula Importance of Intravertebral Fracture Clefts in Vertebroplasty Outcome Am. J. Roentgenol., March 1, 2007; 188(3): 634 - 640. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Singh, D. Samartzis, A. R. Vaccaro, G. B. J. Andersson, H. S. An, and J. G. Heller Current concepts in the management of metastatic spinal disease: THE ROLE OF MINIMALLY-INVASIVE APPROACHES J Bone Joint Surg Br, April 1, 2006; 88-B(4): 434 - 442. [Full Text] [PDF]
|
|
|
|
 |
|
 P. H. Kuo and D. W. Cheng Artifactual Spinal Metastases Imaged by PET/CT: A Case Report J. Nucl. Med. Technol., December 1, 2005; 33(4): 230 - 231. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
