Complications, though initially considered to be low and reported as such, unfortunately are higher for inexperienced physicians or those who attempt the procedure without adequate image guidance or cement opacification. Appropriate training needs to be completed before the procedure is attempted. Recommendations can be obtained from the American College of Radiology Standards of Practice on Percutaneous Vertebroplasty.
In osteoporosis-induced vertebral fractures, clinical reports of complications are around 1%.16,17,22,27 Many of these are transient and include increase in local pain after cement introduction (nonradicular and not associated with neurological deficit). This is usually easily treated with nonsteroidal anti-inflammatory drugs and resolves within 24 to 48 hours. Uncommonly, cement leaking from the vertebra adjacent to a nerve root will produce radicular pain. Analgesics combined with local steroid and anesthetic injections usually provide adequate relief. A trial of this type of therapy is warranted as long as there are no associated motor deficits. The discovery of a motor deficit (or bowel or bladder dysfunction) should initiate an immediate surgical consultation. This type of severe complication will almost always be associated with large-volume leaks that have resulted in neurological compression.
Cement leaks have also been implicated in producing pulmonary embolus.16 These are usually not symptomatic but rarely have produced the clinical symptoms accompanying pulmonary infarct. With a right-to-left shunt, this can result in cerebral infarct.52 Likewise, infection has been rare with PV.
The complication rate found when treating compression fractures resulting from malignant tumors is considerably higher.22,26,29,30,53 This occurs because there are frequently lytic areas involving the vertebral cortex and a greater propensity for cement to leak into the surrounding tissues or vessels. Cement leaks causing symptoms in this setting occur in up to 10% of patients (again most are transient).
Until recently, death had not been a complication associated with PV. Now, however, in two multilevel procedures patients have died. Though the exact details are not known, there was pulmonary compromise, which is suspected to have been due to fat (from the vertebral marrow) or cement emboli. A safe number of vertebrae to treat at one time has yet to be definitely established. Mathis et al. reported treating seven vertebrae in a 35-year-old patient with multiple fractures associated with steroid use for lupus.46 This patient's therapy occurred in three treatment sessions. Because the introduction of cement is a hydraulic event with as much marrow pushed out of the intertra-becular space as cement injected, there is concern about fat emboli in large-volume cement injections. For reasons described earlier, I recommend treating no more than three vertebrae in any one session. Additionally, there are no data that support the prophylactic use of PV to treat vertebra that are believed to be at risk of fracture. Except for pro phylactic use, there is little conceivable reason to perform PV on large numbers of vertebrae at one time.
Any deviation from an expected good result (such as increased pain or neurological compromise) should initiate an immediate imaging search with CT to look for a cause of the clinical change. Unremitting or progressive symptoms may require surgical or aggressive medical intervention, and outpatients should be hospitalized and monitored.
Was this article helpful?
This guide will help millions of people understand this condition so that they can take control of their lives and make informed decisions. The ebook covers information on a vast number of different types of neuropathy. In addition, it will be a useful resource for their families, caregivers, and health care providers.