Radiologists have long known that medical images can reveal problems entirely unrelated to the reason for which they were taken. The increased use of cardiac CT angiography by nonradiologists has raised concern that untrained eyes may miss problems outside the heart.
With that in mind, Dr. U. Joseph Schoepf and colleagues at the Medical University of South Carolina in Charleston retrospectively reviewed records of 556 consecutive patients who had been imaged for suspected coronary artery disease. The researchers used a 16-slice scanner until they purchased a 64-slice machine this year. Their analysis showed one-fifth of patients with significant noncardiac findings, including 9% pulmonary, 4% each extrathoracic and other vascular, and 2% each lymph and pericardial. The study was conducted for an in-house cardiac CT course, said Schoepf, director of CT research and development.
Dr. Smita Patel, an assistant professor of radiology, and colleagues at the University of Michigan reported at the American Roentgen Ray Society meeting in May that half of 98 patients scanned for CAD had significant or potentially significant extracardiac findings. Abnormalities in the chest included lung nodules larger than 4 mm (two were later found to be stage I cancer), signs of emphysema and other respiratory diseases, fluid buildup in the lungs, enlarged lymph nodes, and chest wall masses. One-fifth of patients showed signs of cardiac tamponade, eight had an aneurysm or dissection in the thoracic aorta, and one patient had pulmonary embolism. Researchers also found liver and pancreatic lesions.
"These data show just how important it is for trained radiologists to view cardiac CT scans, rather than having the scans read by heart specialists alone," Patel said.
Dr. Irfan Shafique and colleagues at Johns Hopkins Bayview Medical Center conducted a study of 75 patients with chest pain or known CAD who underwent coronary CTA. The group found that 16% of the patients had major noncardiac findings, including two pulmonary embolisms, two lung masses, three cases of bulky lymphadenopathy, and three large hiatal hernias. Minor noncardiac findings in the remaining 25 patients included a small hiatal hernia, esophageal wall thickening, liver cysts, atelectasis, and mild lymphadenopathy. The study was reported at the American Heart Association meeting last November.
In the June issue of Radiology, a group of researchers presented a "Clinical Statement on Noninvasive Cardiac Imaging" from the American College of Radiology. It calls for specialized training, education, and specific levels of experience for radiologists who read coronary CTA images. It also states that radiologists' knowledge of structures beyond the heart provides added value in cardiac imaging. But the fast adoption of cardiac CTA caught even the ACR by surprise. A caveat to the clinical statement mentions that the document must suffice until practice guidelines are written and approved through the usual ACR process.
The American College of Cardiology Foundation and other organizations published a Clinical Competence Statement in July that outlines the requirements for practicing cardiovascular CT and MRI (Circulation, 2005;112[4]:598-617). It does not deal with noncardiac findings.
The majority of published studies on extracardiac CTA findings involve electron-beam CT rather than multislice CT. That is likely to change as more centers purchase 64-slice scanners and as data from 16-slice scanners become mature, Schoepf said. Even so, EBCT findings mirror the emerging reports from MSCT.
University of Pittsburgh epidemiologists, led by Dr. Jeff G. Schragin, retrospectively reviewed records of 1356 patients scanned for coronary calcium measurements with EBCT (J Thorac Imaging 2004;19[2]:82-86). Twenty percent of patients had one or more extracardiac abnormality. Nearly 5% were recommended for CT follow-up, most for pulmonary nodules.
In another study, Dr. Karen M. Horton and colleagues at Johns Hopkins evaluated records of 1326 patients who were tested for coronary calcium with EBCT (Circulation 2002;106:532). Nearly 8% had significant extracardiac pathology requiring clinical or imaging follow-up.
Dr. Peter Hunold and colleagues at the University of Essen in Germany found significantly high rates of noncoronary pathological findings in patients scanned with EBCT for coronary artery disease. Of 1812 patients, 23% had aortic disease, 20% had lung abnormalities, and 15% had other pathology (Europ Hrt J 2001;22[18]:1748-1758).
In an interesting twist, Dr. Linda Haramati and colleagues from Albert Einstein College of Medicine report that radiologists looking at contrast-enhanced chest CT exams should also review the patient's heart to rule out heart attack. An evaluation of 59 chest CT patients showed reduced enhancement on CT in the one patient who had had a recent heart attack, Haramati said at the American Roentgen Ray Society meeting in May.
"The CT exam must be correlated with medical records and other cardiac tests to confirm a recent heart attack, because there are false positives with the CT exam," she said.
