Abdominal, thoracic CT scans reliably detect incidental low lumbar BMD

View on the News

Extracting more data from existing images

Dr. Pickhardt and his associates "have laid all the groundwork needed to justify using conventional CT imaging to detect incidental osteoporosis," said Dr. Sumit R. Majumdar and Dr. William D. Leslie.

Given the large number of such CT scans performed every year, "the idea of extracting more information from imaging data collected for other purposes holds merit," they said.

"It is now up to the rest of us to safely and cost-effectively translate this new knowledge into everyday clinical practice," Dr. Majumdar and Dr. Leslie said.

Dr. Majumdar is with the University of Alberta, Edmonton. Dr. Leslie is with the University of Manitoba, Winnipeg. Neither reported any financial conflicts of interest. These remarks were taken from their editorial, which accompanied Dr. Pickhardt’s report (Ann. Intern. Med. 2013;158:630-1).



Abdominal and thoracic CT scans obtained for a variety of reasons, such as to assess pain, GI symptoms, or urinary tract complaints, also can be used "opportunistically" to examine lumbar bone mineral density and screen for occult osteoporosis, according to a report in the April 16 issue of the Annals of Internal Medicine.

Abdominal and thoracic CT scans done in routine practice happen to include imaging of the L1 level, which can easily be identified because it is the first non–rib-bearing vertebra. Such scans readily yield data on lumbar bone mineral density (BMD), which is a clinically useful way to diagnose or rule out osteoporosis, said Dr. Perry J. Pickhardt of the department of radiology and his associates at the University of Wisconsin, Madison.

It is important not to confuse this standard CT scanning with quantitative CT (QCT) scanning. QCT "is more labor-intensive; requires an imaging phantom or angle-corrected [region-of-interest] measurement of bone, muscle, and fat at multiple levels; and involves additional money, time, and radiation exposure," they explained.

Unlike dual-energy x-ray absorptiometry (DXA) screening or QCT assessment, "the method that we used requires a negligible amount of training and time; could be applied prospectively by the interpreting radiologist or retrospectively by a radiologist or even nonradiologist; adds no cost; and requires no additional patient time, equipment, software, or radiation exposure," the investigators wrote.

Such incidental CT scans can be assessed retrospectively because they are almost always stored indefinitely in electronic medical records, they noted.

Dr. Pickhardt and his colleagues evaluated CT-derived BMD assessment and compared it against DXA scanning of the hips and spine by identifying 1,867 adults who had undergone the two types of scanning within a 6-month period during the 10-year study interval. They retrieved and reviewed the images, paying particular attention to obvious moderate or severe compression deformities on the CT images, rather than to milder ones, "to avoid ambiguity related to more subjective borderline or mild compression deformities."

The study subjects had a total of 2,063 pairs of CT and DXA assessments that had been performed a median of 67 days apart. A total of 81% of these subjects were women, and the mean age was 59 years.

These patients had undergone abdominal or thoracic CT for a variety of clinical indications, most often for a suspected mass or an oncologic work-up (414 subjects), genitourinary problems (402 subjects), gastrointestinal symptoms (398 subjects), and/or unexplained abdominal pain or symptoms (374 subjects).

Approximately 55% of the CT scans involved intravenous contrast. The use of contrast had no effect on the interpretation of lumbar data on the scans.

The DXA screening identified 22.9% of the study subjects as osteoporotic, 44.8% as osteopenic, and 32.3% as having normal BMD. The CT scans were significantly more sensitive than DXA at distinguishing these three states.

In particular, CT scans identified 119 patients as having osteoporosis, with readily identifiable moderate or severe vertebral fractures, when DXA had classified 62 of these patients as having normal BMD (12 subjects) or only osteopenia (50 subjects).

"Our observations are consistent with prior studies documenting that many patients without osteoporosis diagnosed by DXA will sustain fragility fractures, and suggest that CT attenuation may be a more accurate risk predictor," Dr. Pickhardt and his associates wrote (Ann. Intern. Med. 2013:158:588-95).

If their findings are confirmed in other studies, it may become routine for all abdominal and thoracic CT scans performed for any reason to be used for lumbar BMD assessment as well. "In the future, it may even be possible to incorporate CT ... data into fracture risk assessment tools," they added.

This should result in substantial savings in health care costs since osteoporosis will be diagnosed and treated earlier, before fractures occur, and since it also will reduce the number of costly DXA studies performed.

More than 80 million CT scans were performed in the United States in 2011, "most of which carry potentially useful information about BMD," the researchers noted.

The investigators are now turning their attention to using pelvic CT scans that were obtained for various clinical indications to assess hip BMD. "We are currently investigating the potential for deriving a DXA-equivalent T-score for the hips from standard pelvic CT scans by using a dedicated software tool," Dr. Pickhardt and his associates said.

This study was funded by the National Institutes of Health. None of the investigators reported having any financial conflicts of interest.

Next Article: