REMS Z-Score (Right Femur)

Bone density that's low for your age can be a reason to investigate undernutrition, hormone changes, medications or another medical condition. This score provides an age-matched comparison for your right hip using an ultrasound-based estimate.

REMS is an emerging bone assessment method with published validation studies. Its results can add information, but the evidence for using its Z-score to make individual diagnosis and treatment decisions is less established than for DXA.

What the Score Measures

REMS stands for radiofrequency echographic multi-spectrometry. The scan examines the narrow part of your thighbone just below the ball of the hip joint. This area is called the femoral neck. An ultrasound image helps the operator position the probe. Software then analyzes the raw echoes and compares them with reference signals to estimate bone mineral density. Age and body-size information also enter the calculation.

The Z-score expresses how far that estimated density lies from the average for people of your age and sex. It's expressed in standard deviations. A standard deviation describes how spread out the reference group's results are. Zero represents the reference average; the score isn't a percentage of bone lost or a probability of fracture.

A T-score uses a young-adult reference group. DXA reporting guidelines prefer Z-scores before menopause and in men younger than 50, and T-scores after menopause and in men aged 50 or older. These recommendations explain the different comparisons; they don't establish that a REMS score can replace a DXA score.

This result describes the right femoral neck. Your left hip and spine can differ, particularly after an injury or with disease affecting one side. A study finding that the two hips correlate closely doesn't establish that either hip represents every other skeletal site.

How It Compares With a DXA Scan

DXA stands for dual-energy X-ray absorptiometry. It uses low-dose X-rays and is the reference method for diagnosing and monitoring osteoporosis. A meta-analysis of 17 studies included 7,464 participants with femoral-neck measurements and found that REMS and DXA density estimates moved together strongly. Agreement varied substantially between studies and populations.

In a large European study of women, REMS femoral-neck T-score classifications identified about 90 of every 100 women classified as having osteoporosis by DXA and about 96 of every 100 without it. Those estimates came after quality checks excluded erroneous scans. They describe agreement between T-score classifications, not validation of this right-hip Z-score.

Strong correlation across a group can coexist with substantial disagreement for an individual. An Australian study of 69 people found that REMS and DXA density values weren't interchangeable. A difference between the two methods should therefore be investigated before it's interpreted as bone gain or loss.

Fracture Risk

The main REMS fracture studies examined T-scores and the separate Fragility Score. The Fragility Score is designed to estimate fracture susceptibility from ultrasound patterns. It isn't a direct image of bone structure, and it measures something different from an age-matched density comparison.

In the first study, the femoral-neck REMS T-score classification identified about 40% of women who later fractured and correctly classified about 80% who didn't. DXA performed similarly in that cohort. These findings support an association with fracture risk, but they don't provide a validated conversion from your right-femur Z-score to your personal fracture probability.

Pregnancy and Younger Adults

A study compared 40 pregnant women with 60 non-pregnant women of similar age and pre-pregnancy body mass index. The pregnant group was around the middle of pregnancy and had about 6% lower estimated right femoral-neck density. Their average right-hip Z-score was 0.15, compared with 1.45 in the controls.

This was a comparison between different women at one point in time. It couldn't establish how much bone any woman had lost during pregnancy, whether she would recover it afterward, or whether a particular result predicted a fracture. The findings show that scanning during pregnancy is feasible; they don't establish a benefit from routinely screening every pregnancy.

REMS has also been studied in younger adults with conditions that affect bone. Osteogenesis imperfecta is an inherited disorder that causes fragile bones. Adults with this condition had lower hip Z-scores than healthy controls using both REMS and DXA, although the methods didn't give identical results. In young women with anorexia nervosa, REMS hip Z-scores were lower than DXA Z-scores. These studies support further assessment of bone health in these groups, without making the two methods interchangeable.

When Results Can Be Misleading

• Entered age and body size: In an experiment involving five healthy volunteers, increasing only the weight entered into the software by 5 kg raised estimated femoral-neck density by about 4.3%. The volunteers hadn't gained weight or bone. Changes in entered age also changed the estimates. A real weight change between visits can therefore affect the result through both biology and the calculation.
• Scan acquisition: Probe positioning, scan depth, focus and the quality of the ultrasound signals affect the estimate. Published accuracy studies often excluded scans with acquisition or analysis errors.
• A right hip replacement: A standard femoral-neck result isn't valid for a hip whose native femoral neck has been removed. In a study of 50 people, REMS still produced similar density estimates after hip replacement, with an average decrease of about 2.8%. A numerical report from an operated hip doesn't establish that it represents native bone density.
• Kidney failure treated with peritoneal dialysis: In a study of 20 people followed for about 20 months, DXA detected a 3.2% femoral-neck density decline while the change measured by REMS wasn't statistically significant. Changes measured by the two methods didn't correlate. This small study raises concerns about using REMS alone to monitor bone loss in this setting.
• Bone loss after spinal cord injury: In one study's subgroup with disuse-related osteoporosis, REMS and DXA agreed on the diagnostic classification only 13% of the time, and REMS estimated higher femoral-neck density. This finding shouldn't be generalized to every cause of limited mobility.
• Lupus: A study including people taking glucocorticoids found fair to moderate agreement between REMS and DXA classifications. REMS generally gave lower density estimates. Similar discrimination of previous fractures in that study doesn't establish equivalent prediction of future fractures.

Understanding Changes Over Time

A small study of repeated femoral-neck scans found short-term density measurement errors of 0.32% with the same operator and 0.51% between operators. These percentages describe the underlying density estimate, not percentage changes in the Z-score. They also aren't universal limits for every device, operator or clinical population.

Repeatability means that repeated scans give similar results under similar conditions. It doesn't by itself establish that the method accurately detects treatment-related bone changes. The dialysis findings show why that distinction matters.

The age-matched reference also changes as you get older. Your Z-score can remain similar even while absolute density declines with age. Interpreting a trend therefore requires the underlying density estimates, scan quality and changes in age and body size, along with your fracture and treatment history.