| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Papageorgiou General Hospital Thessaloniki 54453, Greece
Peh et al. [1] have described an association of intraosseous vacuum phenomenon in vertebral compression fractures with preexisting vacuum phenomenon in the adjacent disk, in their excellent study on "Percutaneous Vertebroplasty: Treatment of Painful Vertebral Compression Fractures with Intraosseous Vacuum phenomena." Moreover, a relationship was postulated between polymethyl methacrylate (PMMA) intradiskal leakage and the presence of an intraosseous vacuum cleft leading to the adjacent disk. These findings are in agreement with those of previous authors [2] reporting a higher incidence of an intradiskal vacuum phenomenon in cases with intraosseous vacuum cleft, which was attributed to communication between intervertebral and intravertebral gaseous collections through a fractured endplate.
An unexpected accumulation of air in the epidural space after percutaneous vertebroplasty, which we observed in one of our patients with osteoporotic fractures, may also provide evidence that intradiskal gas may migrate into the neighboring sites. The 78-year-old woman was referred for percutaneous vertebroplasty with osteoporotic subacute L1 and L3 compression fractures that were causing progressive severe lower back pain associated with motor deficiency. Radiographs and MR images revealed vacuum phenomena in the fractured L1 vertebral body and the neighboring intervertebral disks. L1 and L3 percutaneous vertebroplasty was performed under the guidance of CT and C-arm fluoroscopy with injection of 4.5–5 mL of PMMA through a 16-guage vertebroplasty trocar needle. During the injection in the mid portion of L3 vertebral body, a PMMA leakage into the basivertebral and epidural venous plexus appeared on fluoroscopy. The injection was immediately halted, a negative pressure was applied on the injector, and reinjection of the cement was attempted. CT of the spine revealed a 1.5-cm PMMA opacification in the epidural venous plexus at the level of L3, and minimal paravertebral soft-tissue and intradiskal leakage of PMMA. Some gas in the epidural space at the level of L1 vertebra was also observed on CT (Fig. 1A). The patient did not experience any adverse reactions or new symptoms, and the neurologic examination did not reveal anything new. The patient was discharged 12 hr later in a painless state. One week later, a CT scan showed that the epidural gas had been absorbed completely. During the next 5 months the patient remained free from back pain, and her motor function was remarkably improved.
|
To our knowledge, the presence of gas in the epidural space immediately after percutaneous vertebroplasty has not been previously reported. It is unlikely that room air would be injected during vertebroplasty needle insertion to the anterior central part of the vertebral body. A compressed vertebral body commonly fractures through the endplate; therefore, it is possible that the intraosseous gas came from gas in the disk rather than occurring as an isolated development. Moreover, PMMA in the vertebral body and intervertebral disk (Fig. 1B) suggests osseous and disk communication or damage during the procedure, facilitating intradiskal gas release into the epidural space. The smaller volume of intradiskal gas under pressure might compensate for the larger volume of gas in the epidural space, which may have been under lower pressure. The condition was self-limited and was without any obvious clinical significance.
|
Intradiskal gas may migrate to the intravertebral cleft as the result of an osteoporotic compression fracture, or it may migrate to the epidural space during percutaneous vertebroplasty. It is more likely for intraosseous gas to originate from leakage of preexisting intradiskal gas than for it to result from vascular insult leading to secondary bone necrosis, as has been previously assumed [3].
References
Mallinckrodt Institute of Radiology Washington University School of
Medicine St. Louis, MO 63110
Singapore Health Services Pte Ltd., Singapore 169608
Mallinckrodt Institute of Radiology Washington University School of
Medicine St. Louis, MO 63110
We thank Dr. Souftas for recognizing our article describing vacuum phenomena associated with vertebrae being treated with vertebroplasty [1]. The author implies that we believe all gas in vertebral bodies comes from intradiskal gas. We did not try to indicate that belief. We believe that gas may enter the collapsed vertebral body from the disk in many cases, but in other cases with intravertebral gas, no intradiskal gas is evident. We believe that in some cases of vertebral body fracture, a nonunion may develop between superior and inferior fracture fragments. Extension of the body at that fracture level may separate fracture fragments and allow gas present in solution in the patient's body to pass into the fracture cleft, just as radiologists see "vacuum phenomena" elsewhere in a patient's body. It is common to see such a vacuum phenomenon caused by distraction in normal glenohumeral joints of small children who are undergoing frontal chest radiography with their arms raised over their heads.
In the case described by Souftas, we believe that the gas could have come either from the disk or the vertebral body. The epidural gas could have come from either or both sites because it seems most likely that the polymethyl methacrylate (PMMA) pushed the gas outward, and PMMA was stated to be present in both the patient's disk and the vertebral body.
The author did not mention use of intraosseous venography before the procedure. We routinely use "blush venography" [2], which shows when a direct connection to the epidural venous plexus is present. When such a direct venous connection is shown, we usually advance the needle more anteriorly or allow a tiny amount of the PMMA to set up before injecting the rest [2]. In our experience, this technique diminishes the amount of intravascular filling needed and may also decrease the amount of epidural venous plexus filling.
The author presents an interesting finding that logically should present no problem to a patient. Publishing this report should help alleviate concerns of others who may see this finding in the future.
References
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
