Dr. Kleerekoper has posted about DXA issues before. Now it is my turn.

I review the previous DXA of all patients that come to see me regarding osteoporosis and bone disease. I continue to be surprised by the range in quality of interpretations I see. The DXAs below were from the same patient conducted at an outlying facility between 2008 and 2010.

Despite the poor quality of the images, it is obvious that the levels on one study were not the same on the other. What is “L1-L2” on one study is “L3-L4” on the other. Technologists are human as we all are and humans occasionally make mistakes. Nevertheless, it is the responsibility of the interpreting physician to review and catch such an error — if it occurs.

The interpretation was: “Mild to moderate osteopenia with mild to moderately increased fracture risk.” The mistake in the labeling of the lumbar spine was missed — I wonder if the images were reviewed at all.

There is no such thing as “mild to moderate” osteopenia. If the T-score is between -1.0 and -2.5, the diagnosis is osteopenia. Indeed, the preferred term per the International Society of Clinical Densitometry (ISCD) is low BMD. When I calculated the patients estimated 10–year fracture risk per the WHO FRAX calculator, her risk for major osteoporotic fracture and of a hip fracture were not increased beyond that of other 50-year-old women.

Poor quality DXA interpretations are not uncommon and in some facilities they even appear to be the rule rather than the exception. I remember another case of a 17-year-old who was diagnosed with osteoporosis and begun on oral bisphosphonate because of a Ward’s triangle T-score of less than -2.5. She had no history of skeletal fragility or fracture. DXA was ordered because she had been on Depo-Provera for several months.

When I reviewed the images myself, my interpretation was bone mass within the expected range for age. Per the guidelines of the ISCD, Ward’s triangle should never be used in determination of low BMD or osteoporosis. I advised her to stop the bisphosphonate as there was no indication for it.

In pre-menopausal females, we use Z-scores, not T-scores in the evaluation of BMD. A Z-score of less than -2 would be “bone mass below the predicted range for age.” This young woman’s Z-score was greater than -2; therefore she was normal. In women who are pre-menopause as well as men younger than age 50, osteoporosis cannot be diagnosed unless there is evidence of skeletal fragility — no matter what the T-score or Z-score is.

I have forwarded copies of the ISCD Position Statement to centers which have had problems in performing and interpreting DXA. Thus far, I have received no response. I have seen no change in the quality of their DXA. This is a problem particularly for primary care providers who may not understand the intricacies and limitations of DXA. They may make treatment decisions based on DXA interpretations that are not accurate.

I strongly believe in certification of physicians and technologists by the ISCD. If more physicians and technologists were certified and trained in performing and interpreting DXA, errors such as these would be much less common.

  Comments (0)

To my recollection, non-invasive measurement of bone mineral density began several decades ago with single photon absorptiometry of the radius. At that time of the publication of the 1994 WHO guidelines, more of the data were based on forearm measurements than on spine or hip measurements. There is no doubt that BMD of the spine and hip provides a better prediction of major fragility fracture risk than does measurements at the forearm. That does not mean that forearm BMD should be ignored entirely since it has been documented to predict future spine or hip fractures. Remember, the Colles (wrist) fracture is the earliest clinical manifestation of postmenopausal osteoporosis. It is also the most missed fracture in terms of making a clinical diagnosis of osteoporosis.

The one-third radius site is the only forearm site that is approved for reimbursement, but measurement at the ultra–distal radius site can easily be measured during a forearm DXA study without adding to the cost or time taken to conduct the test. (Why does my field continue to use unnecessary terminology? Why not simply call it the distal radius site?) In my practice, I always order spine, hip and forearm BMD — there are some circumstances where forearm BMD is most helpful.

• Primary hyperparathyroidism: While this is a systemic disease affecting the entire skeleton, the lowest BMD is most often the forearm, which may be the only site at which BMD is less than –2.5. In this case, the result should influence the decision regarding surgical cure of primary hyperparathyroidism (PHPT).
• Hyperthyroidism: The effect of thyroxine on the skeleton is similar to that of parathyroid hormone, but at the kidney the effect is quite different. Hypercalcemia is seen in hyperthyroidism, but it is usually accompanied by a normal or elevated serum phosphate as opposed to the low serum phosphate in hyperparathyroidism. Rumor has it that one of Friedrich Daniel von Recklinghausen's original 13 cases of osteitis fibrosa was, in fact, Iod Basedow's disease (better known now as Grave's disease). I am not suggesting that patients with hyperthyroidism need a BMD study, but if you do routinely perform forearm density and it is the lowest of the three measured sites — thyroid function tests are in order.
• Scoliosis: I am never comfortable interpreting BMD of the spine in patients with scoliosis. The more pronounced the scoliosis, the less comfortable I am. Not only are there technical difficulties with spine DXA in scoliosis, but the loading of the vertebral bodies is quite different from normal, and we simply do not know the relationship between spine BMD and fracture risk in this condition. If scoliosis is detected clinically before the study is ordered, omit the spine measurement and measure the forearm. If scoliosis is detected by the technologist performing the density study, the spine study should be completed and a forearm study should also be done. Three sites are better than two — two sites are better than one.

  Comments (0)

One of the most difficult challenges of clinical endocrinology practice is inspiring our patients to adhere to recommendations for dietary modification, lifestyle change and pharmacotherapy.

The traditional information-based approach focuses on educating the patient so that they will understand why and how to change. The physician told the patient what was to be done and the patient was expected to follow this advice — without question. Those who did not follow recommendations were labeled “non-compliant.” Some providers use scare tactics in an attempt to frighten or bully patients into complying. Not surprisingly, this approach usually failed.

More recently, researchers have evaluated what successfully motivates people to change. This has led to a patient-centered, goal-oriented model of provider-patient interaction, known as motivational interviewing. This technique has been shown to be effective in improving adherence to recommendations for chronic disease management including behavioral modification and pharmacotherapy.

Motivational interviewing is based on three basic assumptions: Collaboration, evocation, and autonomy. By understanding and respecting these assumptions, practitioners establish rapport, reduce resistance, support autonomy, and elicit “change talk” — with the ultimate goal of improving clinical outcomes.

Patients appreciate when we try to understand their situation instead of simply telling them what to do. By showing that we value their autonomy, understand that it is their choice and are not trying to manipulate them into change, they are more likely to consider following our recommendations. By exploring ambivalence and barriers for adherence to therapy, we have a better understanding of why non-adherence may occur and how to avoid it. Finally, by involving the patient in developing their treatment plan, they have greater ownership and as a result are more likely to follow through.

Times have changed and I am glad they have.

Butterworth. J Manag Care Pharm. 2008;14:S21-S25.

Rollnick S. Motivational Interviewing in Health Care: Helping Patients Change Behavior. New York: Guilford Press; 2007.

  Comments (2)