Assessing Fracture Risk With DEXA

Dennis Black, PhD


March 18, 2003


If we mostly agree that total hip is the preferred ROI in assessing hip fracture risk, is there added value in addressing the femoral neck? If so, what is that value, and if not, why do so many physicians use femoral neck data from dual-energy x-ray absorptiometry (DEXA) reports to diagnose and follow patients?


David I Greenfield, MD

Response From Expert

Dennis Black, PhD
Professor of Epidemiology & Biostatistics, University of California, San Francisco


The question of whether a single bone mineral density (BMD) site should be used for diagnosis or whether multiple sites should be considered has not been definitively studied. Clinically, the simplest answer to the question is to follow the International Society for Clinical Densitometry (ISCD) recommendations. They recommend that the clinician take the lowest t-score calculated from the following 3 sites: total hip, femoral neck, and total spine. The BMD at the other regions of the hip and at the individual vertebrae are not considered in diagnosis according to this recommendation.

A recent review of clinical use of densitometry published in JAMA refers to this recommendation and supplies much additional detail about our current knowledge of densitometry as a predictor of fracture risk.[1]

Using this algorithm, you will find that, in general, approximately equal numbers of women in their 50s and early 60s will be identified. However, after about age 65, the femoral neck T-score is more commonly the lowest of the 3. There are a number of reasons why this tends to be the case. Most important, after about age 65, anteroposterior spine BMD tends to remain constant or increase, due to measurement artifacts such as aortic calcifications and other nonosteoporotic calcifications.

BMD is best thought of as a risk factor much like cholesterol or blood pressure. There is no real disease at any given threshold; it is simply that risk of diseases increases as the risk factor level increases. To draw the analogy to blood pressure, a patient does not necessarily require treatment if she has a single diastolic blood pressure over 90 mmHg. There are 2 points to be made from this analogy. First, for BMD, risk of fracture increases as BMD decreases. There is no exact threshold above which risk is zero and above which risk is very high. We find it convenient and clinically necessary to set arbitrary thresholds, but the clinician needs to take a broader view and consider BMD in the context of other risk factors such as age and history of fracture. The second point is that there is not 1 specific measurement that somehow defines truth about osteoporosis. The implication of this is that there is no such thing as "missing" a low t-score (eg, on a single vertebral measurement). Reliance on the 3 measurements (total hip, femoral neck, and total spine) will adequately characterize fracture risk and need for treatment.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.