| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Cardiopulmonary Imaging |
1 Department of Radiology, Box 57, Division of Diagnostic Imaging, The
University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd.,
Houston, TX 77030.
2 Present address: Department of Radiology, Korea Cancer Center Hospital, Seoul,
Korea.
Received July 30, 2003;
accepted after revision March 22, 2004.
Address correspondence to E. M. Marom
(emarom{at}di.mdacc.tmc.edu).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Chest radiographs were obtained after 69 percutaneous vertebroplasty procedures in 64 patients. Chest radiographs were reviewed retrospectively for the presence of pulmonary emboli of cement, and findings were assessed. The frequency was calculated from the cases treated. Medical records were reviewed for procedure-related complications.
RESULTS. The emboli of cement were noted radiographically in three (4.6%) of 65 procedures performed in our institution. All patients with cement emboli had multiple myeloma. The chest radiographic findings were multiple radiographically dense opacities with a tubular and branching shape that were scattered sporadically or distributed diffusely throughout the lungs. All patients with cement pulmonary embolism remained asymptomatic. A correlation of embolism of cement to lungs was found with paravertebral venous cement leak (p < 0.001) but not with the number of vertebral bodies treated (p = 0.185) or with the type of procedure performed—kyphoplasty versus vertebroplasty (p = 0.98).
CONCLUSION. Pulmonary embolism of cement is seen in 4.6% of patients after percutaneous vertebroplasty or kyphoplasty. The characteristic radiographic findings should be recognized by radiologists.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cement leaks are common, although studies evaluating vertebroplasty procedures failed to find any systemic complications associated with these leaks [3–6, 8]. Recent descriptions of cement pulmonary embolization in the clinical literature are limited to a few case reports and focus on the clinical aspects of this complication [9–13]. The radiographic appearance of cement pulmonary emboli can be impressive and alarming and may be encountered on routine chest radiographs, although to the best of our knowledge, their appearance has not been evaluated in the radiology literature.
The purpose of this study was to determine the frequency, the radiographic findings, and the clinical significance of pulmonary embolism caused by cement that occurs during percutaneous vertebroplasty and kyphoplasty procedures, as depicted on chest radiographs.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Chest radiographs were obtained after 65 procedures in 62 of the patients (three patients were treated twice). In the other 34 patients, no periprocedural chest radiograph was obtained, and they remained asymptomatic. Of these 34, 25 received follow-up of their primary malignancies in our institution using other imaging techniques, and nine were followed up at outside institutions because they were referred to our institution for vertebroplasty only. Two additional patients who underwent vertebroplasty at an outside institution and had chest radiographic follow-up in our institution were included in our study group for descriptive purposes only. Thus, the final study group was 64 patients, including 33 women and 31 men who ranged in age from 31 to 79 years (mean age, 62 years).
Primary malignancies for which they were followed up in our institution included multiple myeloma (n = 23), non–small cell lung cancer (n = 12), breast cancer (n = 8), leukemia (n = 2), ampulla of Vater cancer (n = 1), ethesioneuroblastoma (n = 1), lymphoma (n = 5), renal cell carcinoma (n = 4), prostate cancer (n = 3), pancreatic islet cell tumor (n = 1), rectal cancer (n = 1), small cell lung cancer (n = 1), thyroid cancer (n = 1), and tongue cancer (n = 1).
Vertebroplasty Technique
A detailed description of the vertebroplasty technique used has appeared
elsewhere [14]. A similar
technique was also used in the two patients from an outside institution.
Briefly described, vertebroplasty was performed with the patient receiving
either general anesthestic or local anesthetic combined with IV narcotics or
sedative drugs at the discretion of the anesthesiologist, and kyphoplasty was
always performed with the patient under general anesthesia. The patients were
in a prone position for both procedures. Vertebroplasty and kyphoplasty were
both performed under biplane fluoroscopic control in the neurointerventional
angiography suite by an experienced interventional radiologist. The
transpedicular route was preferred. Although a unilateral approach was usually
sufficient, it was occasionally necessary to access both pedicles to achieve
sufficient vertebral body-filling with PMMA. Bilateral access to the vertebral
body was uniformly obtained for kyphoplasty.
Bone cement was prepared by mixing 40 mL of the PMMA powder (Simplex P, Stryker-How-medica-Osteonics) with 6 g of sterile barium sulfate powder (Biotrace, Bryan) for opacification and 1 g of powdered tobramycin (Nebcin, Eli Lilly) for antibiotic prophylaxis before adding 10 mL of the liquid monomer. Cement injection was monitored on continuous fluoroscopy in the lateral plane with intermittent evaluation in anteroposterior projection to detect early lateral venous leaks. When cement was visualized in the posterior fourth of the vertebral body or beyond the confines of the vertebral body, the procedure was usually terminated.
Chest Imaging and Follow-Up
All patients included in the study had posteroanterior and lateral chest
radiographs obtained before and after the procedure. Chest radiographs were
obtained using the standard technique. The time interval between the procedure
and postprocedural chest radiography varied (range, 0–365 days; median,
47 days). Chest radiographs were reviewed using hard-copy images (n =
60) if they were available and soft-copy images (n = 7) (iSite
Enterprise, Stentor) if necessary. Each postprocedural chest radiograph was
compared with a baseline chest radiograph obtained before the procedure.
Findings recorded included the presence of radiographically dense cement
material in the lungs; its size, shape, and distribution; new pulmonary
opacities; new pleural abnormalities; and leakage of cement material into
adjacent perivertebral structures. When anomalies were seen on the
postprocedural chest radiograph, all follow-up radiographs were examined for
interval changes. Chest CT scans were reviewed when they were available
(n = 23) and were obtained using either HiLight Advantage, HiSpeed
CT/i, or LightSpeed QX/i CT scanners (GE Healthcare), with a 3.75- or 7-mm
collimation and a pitch ranging from 1.0 to 1.5 from the lung apices to the
adrenal glands and using IV contrast material. Chest CT scans were reviewed
using soft-copy images, and findings were correlated to the chest radiographs
and recorded. All chest radiographs and chest CT scans were simultaneously
reviewed by three experienced radiologists. Differences were resolved by
consensus.
The frequency of cement pulmonary embolism was calculated from the procedures (n = 65) that were performed in our institution. Electronic medical records of all patients were reviewed for possible complications during the periprocedural time. In addition, hard-copy medical records were reviewed for all patients (n = 16) who had an abnormality found on their chest radiograph for evaluation of vital signs throughout the vertebroplasty or kyphoplasty procedure. Medical records and computerized images of the pretreatment vertebral bodies were retrospectively reviewed to assess the cause of vertebral body fracture.
Statistical Analysis
The Mann-Whitney test was used to examine the association between the
development of PMMA pulmonary emboli and the number of vertebral bodies
treated per session. The chi-square test was used to find whether development
of emboli correlated with visualized paravertebral PMMA venous leaks and the
type of procedure: kyphoplasty versus vertebroplasty.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several patients had risk factors for osteoporosis (advanced age, postmenopausal state, chronic steroid use, and medically debilitated state), and it was sometimes difficult to determine the extent to which osteoporosis was responsible for the vertebral body collapse compared with pure malignant involvement. The cause was determined by a combination of bone biopsy (n = 18), analysis of bone marrow aspirate (n = 28), MRI (n = 59), bone scanning (n = 3), or CT (n = 4). Causes thought to have been responsible for the vertebral body fractures in the patients treated were multiple myeloma (n = 23), osteoporosis (n = 17), metastatic disease from breast cancer (n = 6), lung cancer (n = 5), lymphoma (n = 3), prostate cancer (n = 3), renal cell carcinoma (n = 2), leukemia (n = 1), ampulla of Vater cancer (n = 1), ethesioneuroblastoma (n = 1), thyroid cancer (n = 1), and hemangioma (n = 1).
Four patients, all of whom had multiple myeloma, developed multiple tubular or branching radiographically dense opacities to the segmental level of the pulmonary arteries consistent with PMMA embolism (Figs. 1A and 1B). Three of these four patients with emboli had undergone vertebroplasty, and one had undergone kyphoplasty. The chest radiographs of the patients with cement emboli were obtained 10–135 days (mean, 83 days) after their procedure. None were obtained because of any acute symptom. Indications for chest radiography included routine follow-up (n = 2) and postprocedural treatment change (n = 2). Emboli were either small in number and sporadically distributed (n = 3) (Figs. 1A and 1B) or numerous and widely distributed through both lungs (n = 1) (Figs. 2A and 2B). Radiographic follow-up (range, 69–154 days; mean, 112 days) showed stability of the appearance of these emboli. The emboli were detected after the treatment of one, two, three, and five vertebral body levels in each of the four patients. PMMA pulmonary embolism occurred during vertebroplasty in 4.6% (3/65) of the procedures performed at our institution. No statistical correlation was found between the risk of developing a cement embolus and the type of procedure performed (kyphoplasty vs vertebroplasty, p = 0.98) or the number of vertebral bodies treated per session (p = 0.185). One of these four patients had routine a follow-up chest CT scan obtained 2 months after the procedure that clearly showed the PMMA emboli. The rest of the chest CT scans studied (n = 22) did not show any additional cement pulmonary emboli.
|
|
|
|
PMMA leaks beyond the confines of vertebral bodies into the perivertebral veins were visualized in six patients (Figs. 3A and 3B). In two of these six, the leak was visualized only on CT. Four of the six patients with a perivertebral venous leak also had PMMA pulmonary emboli, but the perivertebral venous leak was only visualized on conventional chest radiographs in two patients (Figs. 3A and 3B). The presence of venous PMMA leaks as seen on conventional chest radiographs significantly increased the risk of PMMA pulmonary emboli (p < 0.001).
|
|
Pulmonary parenchymal opacities were noted in nine patients. None were thought to be related to pulmonary embolism. Six of nine cases presented with linear opacities in the lower lobes. These opacities were believed to represent subsegmental atelectasis, but they did not correlate with any clinical abnormality detected in the medical records, and the patients remained asymptomatic. Three patients showed heterogeneous pulmonary opacities in the left upper (n = 1), left lower (n = 1), or right lower lobe (n = 1). Clinically, none of these opacities were believed to correlate with the vertebroplasty. In one patient, the opacities were within a radiation port and believed to be due to radiation pneumonitis. The second patient developed opacities after spinal surgery performed the day after the vertebroplasty, and the opacities were believed to be due to aspiration pneumonia. The third patient was believed to have had neutropenic pneumonia not related to the vertebroplasty procedure because the opacities developed 2 weeks after vertebroplasty when the patient developed fever, sputum, and neutropenia.
Four patients developed either a left (n = 3) or bilateral (n = 1) pleural effusions. Pleural effusions were small, occupying less than one fourth of the hemithorax. The cause of the effusion was believed to be pneumonia (n = 1), radiation therapy (n = 1), or recent spinal surgery (n = 1), and in one patient, the cause remained unknown. None of the pulmonary parenchymal opacities or pleural effusions were seen in the patients who developed PMMA pulmonary emboli.
No major complications were related to the procedure. All patients had stable readings for blood pressure, heart rate, and oxygen saturation throughout the study. One patient developed transient pain at the site of the procedure without radiculopathy. In another patient, pain relief was not achieved, and he underwent spinal surgery for continued instability-related pain. All patients with PMMA pulmonary emboli remained asymptomatic with no respiratory symptoms during a 5- to 11-month (mean, 7 months) follow-up. Of the patients with perivertebral PMMA leak, one developed local transient pain and the rest remained asymptomatic.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
As an attempt to improve bone alignment and reduce cement extravasation, a new technique has recently been introduced: percutaneous balloon kyphoplasty. It involves inflation of a balloon within the collapsed vertebral body, before stabilization with PMMA [15–17]. Because the inflated balloon creates a void within the vertebral body, into which cement is injected under low pressure, theoretically, the risk of cement extravasation is reduced. Reduced risk of perivertebral extravasation has been shown in one series [14], although in our study this technique did not significantly reduce the rate of cement emboli.
The rate of cement leaks reported in the literature is high and is seen in up to 73% of vertebral bodies treated [3, 6, 8, 14, 18], with venous leaks reported in up to 24% of vertebral bodies treated [3, 6, 8, 19]. The rate at which cement embolizes to the lung from this procedure is unknown because patients are not routinely screened with chest imaging before and after the procedure. Reviews in the literature and series of patients published specifically report no case of cement pulmonary embolism despite significant numbers of cement leaks into the venous system [3–6, 8, 14, 19]. Our study shows an increased risk of cement embolization to the lungs with venous cement leaks; perhaps cement embolization was not diagnosed in previous series because the patients were asymptomatic and chest radiographs were not routinely obtained to search for cement emboli. Reports of cement pulmonary embolism have appeared only recently in the literature [9–12]. Most patients with cement emboli remain asymptomatic, although in one case report [9], the outcome was catastrophic. In that patient, saddle cement emboli at the right and left pulmonary artery levels caused immediate respiratory distress, right heart failure, and renal failure, necessitating pulmonary embolectomy.
As vertebroplasty becomes more common, radiologists are more likely to encounter the typical cement emboli on routine chest radiographs. Conventional radiographs readily show the emboli because of their high density compared with lung parenchyma. The appearance of high-density opacities in a tubular branching pattern corresponding to pulmonary arterial distribution is suggestive of the diagnosis. The findings can be striking and alarming, as seen in our series and case reports [12], and can vary from profuse cement emboli to minimal numbers of emboli. Patients may remain asymptomatic and develop no known long-term sequelae. However, when emboli are discovered incidentally on a chest radiograph, reference should be made to the symptoms. Initial chest pain in patients with emboli may be attributed to the local or regional pain seen within 48 hours after vertebroplasty and the diagnosis of pulmonary emboli may be delayed. Dyspnea can occur immediately or lag, occurring days after the vertebroplasty and after discharge from the hospital [12]. A radiologist who is aware of these facts may be able to contribute to a correct and timely diagnosis. On the other hand, when cement emboli are encountered in an asymptomatic patient, they are probably of no clinical significance and have no known long-term sequelae.
The incidental finding of cement emboli in an asymptomatic patient should not alter medical treatment. Symptomatic patients with cement emboli have been previously treated with anticoagulation therapy, although it would be dangerous to recommend any therapeutic approach on the basis of these few cases. The rationale for anticoagulation is to reduce the risk of thrombus formation on the embolic material. No data support the formation of thrombi on the cement however. In three asymptomatic patients with cement emboli, no defect was found on ventilation–perfusion scanning, but a defect was seen in a symptomatic patient [10, 11]. The patients in our series received no anticoagulation therapy and remained asymptomatic during long-term follow-up. This finding supports the notion that plugging of a small percentage of the arterial pulmonary vessels does not result in respiratory symptoms.
Our study has several limitations. First, the retrospective nature of the study introduces inherent biases. We may have underestimated the incidence of cement emboli because minor symptoms of pulmonary emboli may have been missed or interpreted as local pain associated with the vertebroplasty procedure, resulting in no chest imaging. Likewise, we may have overestimated the pulmonary embolism rate because patients without chest symptoms who did not get routine chest radiographs were not included in this study.
Second, the relatively high incidence of cement emboli, 4.6%, may have been related to lack of experience because at the time that this article was written, fewer than 100 patients had been treated with vertebroplasty or kyphoplasty in our institution. We do not believe this to be true, however, because we use great caution and perform this procedure with an experienced staff. Also, the emboli occurred after we had more than 1 year of experience with these procedures. The percentage of cement emboli found in our institution may be high because we are a cancer center. Most of the vertebral bodies we treated were involved with malignancy and therefore may have been at higher risk for cement extravasation [1, 4–6]. Alternatively, it may be that our relatively high rate of detection of pulmonary emboli is due to our routine frequent imaging of asymptomatic patients.
It could also be that if we had added more barium to the cement mixture, leaks would have been identified earlier, decreasing the rate of embolization. However, the addition of barium changes the characteristics of cement and could weaken it as it solidifies. In the future, these mixture variables should be resolved because the trend now is to market the cement together with the barium, thus standardizing the characteristics of the cement mixture.
Third, the cement emboli occurred in patients with advanced multiple myeloma, so mild symptoms may have been masked by the patients' conditions. We do not believe this to have been the case because none of the patients suffered any respiratory symptoms or chest pain and a thorough search of medical records, vital signs, and saturation at the time of the procedure showed no evidence of significant emboli.
Fourth, improved detection of cement emboli might have been obtained with postprocedural thin-section chest CT or with pre- and postprocedural perfusion scans. However, our goal was to study the emboli detected on conventional radiographs. We doubt that extra examinations are justified because of the lack of clinical consequence that small emboli have on patients' morbidity or mortality rates. Also, the cement used is of such high density compared with lung tissue that the visualization of these emboli on conventional chest radiographs is quite striking. None of the CT scans obtained in our study showed any cement pulmonary emboli that were not visualized on the chest radiographs.
In conclusion, the characteristic radiographic findings of multiple tubular or branching radiographically dense pulmonary opacities after cement vertebroplasty or kyphoplasty are consistent with cement pulmonary embolism and should be recognized by radiologists. Interpretation should be made in the correct clinical context and enable timely diagnosis for symptomatic patients and have little clinical significance in asymptomatic patients.
Acknowledgments
We thank Lyle Broemeling of The University of Texas M. D. Anderson Cancer
Center, Houston, TX, for performing the statistical analysis in this
manuscript.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  F.A. Marden and C.M. Putman Cement-Embolic Stroke Associated with Vertebroplasty AJNR Am. J. Neuroradiol., November 1, 2008; 29(10): 1986 - 1988. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. H. Lim, H. Kim, H. K. Kim, and M.-J. Baek Multiple cardiac perforations and pulmonary embolism caused by cement leakage after percutaneous vertebroplasty Eur. J. Cardiothorac. Surg., March 1, 2008; 33(3): 510 - 512. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M.E. Jensen, J.K. McGraw, J.F. Cardella, and J.A. Hirsch Position Statement on Percutaneous Vertebral Augmentation: A Consensus Statement Developed by the American Society of Interventional and Therapeutic Neuroradiology, Society of Interventional Radiology, American Association of Neurological Surgeons/Congress of Neurological Surgeons, and American Society of Spine Radiology AJNR Am. J. Neuroradiol., September 1, 2007; 28(8): 1439 - 1443. [Full Text] [PDF]
|
|
|
|
|
|
K.F. Layton, K.R. Thielen, C.A. Koch, P.H. Luetmer, J.I. Lane, J.T. Wald, and D.F. Kallmes Vertebroplasty, First 1000 Levels of a Single Center: Evaluation of the Outcomes and Complications AJNR Am. J. Neuroradiol., April 1, 2007; 28(4): 683 - 689. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. J. Lim, S. Z. Yoon, Y.-S. Jeon, J. H. Bahk, C. S. Kim, J.-H. Lee, and J. W. Ha An Intraatrial Thrombus and Pulmonary Thromboembolism as a Late Complication of Percutaneous Vertebroplasty Anesth. Analg., April 1, 2007; 104(4): 924 - 926. [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]
|
|
|
|
 |
|
 S. G. Memtsoudis, P. Rosenberger, and J. M. Walz Critical Care Issues in the Patient After Major Joint Replacement J Intensive Care Med, March 1, 2007; 22(2): 92 - 104. [Abstract] [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]
|
|
|
|
 |
|
 D. Togawa, R. Schlenk, I. H. Lieberman, and E. M. Marom Cement Emboli in Vertebroplasty and Kyphoplasty Am. J. Roentgenol., July 1, 2005; 185(1): 277 - 277. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
