Cedars-Sinai Medical Center, the prestigious Los Angeles healthcare institution known as the hospital to the Hollywood stars, has been jolted by an FDA alert indicating that perfusion CT performed during an 18-month period exposed more than 200 stroke patients to eight times the normal dose of ionizing radiation for the procedures.
A custom perfusion CT protocol developed at Cedars-Sinai is considered the prime suspect for elevated doses discovered in August when a patient complained of patchy hair loss -- a cardinal symptom of radiation poisoning -- after receiving a CT procedure.
For more on this topic: |
According to an Oct. 8 FDA alert, 206 patients were exposed to elevated doses for perfusion CT ordered to characterize ischemic stroke. Patients received doses of 3 Gy to 4 Gy instead of the standard 0.5 Gy for the procedure. When notified, eight additional patients reported hair loss and skin reddening. The symptoms appeared within six weeks of scanning, according to the hospital. Cedars-Sinai reported the incident to the California Department of Public Health.
Emergent perfusion CT is typically performed on stroke patients to select candidates for endovascular or intra-arterial clot removal therapy during the hours after symptom onset, according to Dr. Michael Lev, director of emergency neuroradiology at Massachusetts General Hospital.
"It identifies anatomic regions at risk and how complete the infarction is at the time of admission," he said in an interview.
Standard perfusion CT for stroke involves 80 kV and 200 mAs scans, according to Lev, a well-known developer of perfusion protocols. The typical 45 to 60-sec acquisition exposes the patient 2 to 3 mSv, about the same amount of radiation as a routine CT head scan, though concentrated in a smaller area, he said.
E-mail traffic among radiologists had them speculating that the Cedars-Sinai group was using 120 kV or 140 kV settings. A lower kV setting would have increased the conspicuity of the iodine(Drug information on iodine) in the contrast media but would have otherwise produced noisier images, Lev said. The mA setting was also apparently modified.
"They may have tried to use a higher mA setting to deliver more photons for less noisy images," he said, "But with CT perfusion what you really want to optimize is the iodine conspicuity."
The procedures were performed on a GE multislice scanner. The specific model number was not disclosed. In a written statement, GE Healthcare said it is cooperating with the FDA to investigate the incidents. It said no malfunctions or defects in any of the GE scanners used for perfusion CT at Cedars-Sinai were found.
Though the specific incident may be unique to Cedars-Sinai, the FDA characterized the magnitude of the overdoses and their impact on affected patients as "significant." During its initial assessment, the FDA identified shortcomings with the CT quality assurance program at the Los Angeles hospital that could arise at other hospitals and diagnostic imaging procedures. Such incidents could go undetected for the procedures, exposing patients to a level of radiation high enough to elevate their future health risks while not high enough to induce immediate symptoms, the FDA said.
In a written statement, Cedars-Sinai acknowledged that its CT perfusion protocol was modified in February 2008 to improve the quality of data for stroke diagnosis. Patients were exposed to elevated radiation doses because of "a misunderstanding about the embedded default setting applied by the machine in this brain perfusion scan, which is different from a regular CT brain scan."
Cedars-Sinai instituted additional "double-checks" in its operation of the CT scanner and additional equipment protocols to ensure that the problem is not repeated, according to the statement.
The FDA encouraged every facility performing CT imaging to review its protocols and to be aware of the dose indices normally displayed on the control panel before scanning. Key indices include the volume CT dose index (CTDIvol) and the dose-length product (DLP).
The agency urged users to carefully monitor the dose indices displayed on the control panel for each protocol selected before patient scanning. To prevent accidental overexposure, they were asked to make sure that the values displayed reasonably correspond to the doses normally associated with the protocol. The displayed values should be confirmed again after the patient has been scanned.
"The take home message is that if you do perfusion CT for stroke using the standard accepted parameters for mA and kV and deciding against doing serial scans, especially for patients who are receiving other CT and CTA studies, you should be safe relative to the risk of mismanaging these acute stroke patients," Lev said.
