More hospitals are being careful about CT scans

Posted: June 21, 2011

Over 14 months, physicians at Crozer-Keystone Health System ordered a series of CT scans for a 22-year-old woman with massive internal injuries.

Every one of the scans, which use radiation, was deemed beneficial. But when a medical physicist reviewed her records, the number of scans gave her pause: 49.

Soon afterward, the health system joined a growing number of hospitals nationwide that are taking a second look at such scans, which have saved countless lives but also, according to some experts, are overused. And last week, Crozer-Keystone was among 20 health-care organizations attending a conference about a new regional effort to study radiation doses from CT scans.

The 12-month program is being led by ECRI Institute, a nonprofit health-care research organization in Plymouth Meeting. Hospitals and imaging centers in Philadelphia and its four suburban counties are invited to submit data on patient radiation doses so they can compare themselves with their peers. Participants also will take part in surveys and seminars and have access to a special website for discussion, program manager Kathy Shostek said.

The issue caught fire in 2009 and 2010, when the Food and Drug Administration identified at least 385 patients who received excess radiation at five hospitals in California and one in Alabama.

These patients received CT brain perfusion scans, which are used to confirm a stroke and have higher expected doses than typical CT scans. But the agency also found that the machines had been misused, leading to radiation doses up to eight times the intended level. Some patients experienced hair loss and reddening of the skin.

Since then, numerous professional groups have weighed in, identifying instances of possible overuse but nevertheless cautioning that in most cases, the scans are medically necessary and well worth the small risk from radiation.

When properly used, CT scanners never cause short-term effects such as hair loss, said Rohit Inamdar, a medical physicist at ECRI, which is running the 12-month program at the invitation of the nonprofit Health Care Improvement Foundation.

In the long term, there is a theoretical risk that radiation from one scan can cause cancer, but it is slight. A typical CT scan is estimated to increase a person's lifetime risk a small fraction of 1 percent, Inamdar said, though the risk is higher in children.

Crozer-Keystone already was taking steps to monitor its use of CT scans before the California cases came to light. The case of the 22-year-old trauma patient was presented at an education seminar in 2007.

A year or two later, the health system started to flag cases in which patients received five scans within two years so ultrasounds and other alternatives could be considered when further testing was needed.

"You might want to stop and consider: Is there a test that could give me the information as well as the CT scan that doesn't have the radiation exposure?" said Eric Dobkin, Crozer-Keystone's chief quality officer.

National groups and agencies have gotten into the act as well.

In 2009, a coalition of four medical societies established a campaign called Image Gently, which reminds radiologists to use smaller doses on children.

Last year, the federal Centers for Medicare and Medicaid Services began tracking two kinds of outpatient CT scans, evaluating hospitals by whether they used too many higher-dose "combination" scans.

Policy analysts have identified several reasons CT scans may be overused. One is defensive medicine, meaning that doctors prescribe the scans lest they miss something and be accused of malpractice. Another is financial, in cases where physicians have an interest in an imaging center. Still another is a lack of clarity over who is responsible for monitoring a patient's total dose: the family physician, the hospital, or the radiologist.

Recently, numerous researchers have weighed in on which kinds of scans are appropriate.

In a Harvard study this year of 584 patients with abdominal pain, CT scan results led physicians to change their initial diagnoses in nearly half of cases, significantly increasing their diagnostic certainty.

In November, a National Cancer Institute study found that a low-dose variety of CT scan known as spiral helped prevent lung-cancer deaths. In the trial of 53,000 current or former heavy smokers, researchers found 20 percent fewer deaths among those screened with spiral CT scans compared with those who received chest X-rays.

But this month, a study led by Children's Hospital Boston found that for children who had suffered minor head trauma, when physicians waited several hours to observe the patient, a significant number ended up not needing a CT scan.

Typical Radiation Exposures

Source Dose, in millisieverts

Dental bitewing X-ray . . . .005

Airplane flight from New York to Chicago* . . . .009

Chest X-ray . . . 0.1

Mammogram (2 view) . . . 0.36

CT scan, head . . . 2

CT scan, chest . . . 7

CT scan, abdomen . . . 10

* From naturally occurring cosmic radiation

SOURCES: Health Physics Society; New England Journal of Medicine

Contact staff writer Tom Avril

at 215-854-2430 or