Transcript Bone Densitometry Power Point

Bone Densitometry
NOMENCLATURE:
 DXA NOT DEXA Dual-energy X-ray Absorptiometry
 BMD bone mineral density
 VFA vertebral fracture analysis (not DVA, LVA, IVA,
RVA, etc.)
Bone Densitometry
 T-score number of standard deviations the patient’s
BMD is above or below average peak BMD of young
adult reference population.
 Z-score number of standard deviation the patient’s
BMD is above or below age- and sex-matched mean
reference value. Z-score should be population specific
where adequate reference data exist, including
ethnicity.
T-score
WHO CLASSIFICATION
 Normal T-score ≥ -1.0
 Low bone mass (osteopenia)
T-score between -1.0 and -2.5
 Osteoporosis T-score ≤ -2.5
 Severe osteoporosis
T-score ≤ -2.5 with history of fragility fracture
Z-score
Z-scores are used instead of T-scores for children, premenopausal women and men younger than age 50!
A Z-score ≤ -2.0 is defined as "below the expected
range for age."
A Z-score > -2.0 is "within the expected range for age."
Z-score
Pediatric patients (ages 5-19)
Require dedicated software package and need to scan
lumbar spine and total body minus head. NOT hip.
Do not use T-scores.
A Z-score ≤ -2.0 is defined as "below the expected range for
age."
A Z-score > -2.0 is "within the expected range for age."
Bone
Cortical or compact = dense cortex
Trabecular or cancellous = marrow area
Bone
 Spine: about 2/3 trabecular/cancellous bone on PA
view, remainder cortical/compact bone.
 Hip: 25% trabecular bone at neck and 50% trabecular
bone at trochanter
 Forearm: almost entirely cortical bone and only 1%
trabecular bone
Osteoporosis
 Skeletal disorder characterized by compromised bone
strength predisposing to an increased risk of fracture.
 Bone strength reflects the integration of two main
features: bone density and bone quality.
 There are no symptoms from low bone mass unless
fracture occurs.
Why do we care?
 Fractures painful
 Debilitating
 Decrease mobility
 Decrease independence
 Expensive
Why do we care?
2/3 of vertebral fractures are clinically silent –i.e.
morphometric, discovered by X-ray or imaging!
Only 25-30% of fractures seen on X-ray are diagnosed
clinically.
Why do we care?
Complications of hip fractures --especially nursing
home patients:
24-30% excess mortality within 1 year
50% of survivors are permanently incapacitated
20% of survivors require long-term nursing home care
Bone density on average:
 Males > Females
 Blacks > Whites
 Peak bone mass achieved in teens or early 20's.
 Then relative plateau until 35 years old.
Bone density on average:
 Then age-related bone loss occurs at a rate of 0.5%-
1.0%/year.
 Bone loss accelerates with menopause (1.0-
2.0%/year) and accelerated phase lasts 5-10 years.
 Age-related bone loss continues, with bone loss
eventually going back down to pre-adolescent levels.
www.arthritisresearchuk.org
Central DXA
Current gold standard for DIAGNOSTIC
classification of BMD and osteoporosis.
However, multiple other techniques are
well-validated for fracture risk assessment
but not diagnosis of osteoporosis.
Central DXA
Excellent reproducibility (but it’s up to us!)
Low radiation dose (1-10 µSv) or 1/10 the dose
of a CXR, about one day’s worth of natural
daily background exposure.
Most epidemiologic studies and clinical
pharmaceutical trials backing it up.
Indications for BMD Testing
 Women ≥ 65 y/o.
 Postmenopausal women (natural or surgical) < 65 y/o




with risk factors for fx
Women during the menopausal transition
(perimenopausal) with clinical risk factors for fx, e.g.
low body weight, prior fx, or high-risk medication use.
Men ≥ 70 y/o.
Men < 70 y/o with clinical risk factors for fracture.
Adults with a fragility fracture*.
Indications for BMD Testing
 Adults with a disease or condition associated with low




bone mass or bone loss.
Adults taking medications associated with low bone
mass or bone loss.
Anyone being considered for pharmacologic therapy.
Anyone being treated, to monitor treatment effect.
Anyone not receiving therapy in whom evidence of
bone loss would lead to treatment.
FRAGILITY FRACTURE
Fracture from falling from standing height or lower, at
walking speed or slower, often vertebral, hip or wrist.
Fracture occurring in the absence of obvious trauma or
minimal trauma not usually expected to cause a
fracture.
Excludes pathologic fractures (underlying
abnormal bone, e.g. bone metastasis), and
fractures of digits, hands, ankles & feet & skull
fractures.
FRAGILITY FRACTURE
Although not part of the WHO classification, the
presence of a fragility fracture
-regardless of T-score–
should be considered diagnostic of osteoporosis !
(provided other causes for the fracture have been
excluded)
Spine
 Use positioning block to straighten lordosis.
 Centered
 Straight and not tilted or leaning
 If scoliotic, try to center curvature in the middle, off
set pelvis if necessary
Spine

Both iliac crests visible

Mid T12 to mid L5 included

12th rib (optional)

L3 usually has the longest transverse processes
Spine Example
GE
Spine Example
Hologic
Spine
Look at raw data and histograms to
confirm disk levels. Angle as
appropriate.
Check ROI and bone margins
-excluding large osteophytes?
-too narrow and excluding bone?
Spine
If have difficulty numbering due to rib variability
and transitional vertebrae, count from below,
with L4-L5 considered at the iliac crests if possible.
Usually L4-5 is at the iliac crests or slightly lower.
1 in 6 patients have variants! 4,5 or 6 lumbar
vertebrae and lowest ribs T11, T12, or L1.
Spine
 Exclude vertebrae with fracture --check VFA, X-rays
too!
 Exclude if hardware, vertebroplasty or kyphoplasty
(=vertebral augmentation) or laminectomy.
 Check for artifact, stent grafts, surgical clips, barium,
calcification, foreign bodies, bra hooks and
underwires, external to patient? If external to patient,
remove and repeat!
Spine
Careful because DXA does subtract soft tissue!
So even if outside the spine ROI it can affect the
measurement, especially if within global ROI.
Remove, remove, remove if possible and repeat scan.
If not then neutralize it.
Spine QA
QA IMPORTANT!!!
Look at BMDs and T-scores!
L1 usually lowest BMD.
BMC and area progressively increase from
L1 to L4.
BMD tends to increase from L1 to L3.
BMD of L4 similar or less than L3.
Spine QA
Exclude vertebrae if clearly abnormal and nonassessable within resolution of the system.
Exclude vertebra if there is > 1.0 T-score difference
between the vertebra in question and adjacent
vertebrae.
Assess for cause, often increased T-score due to subtle
fracture or increased degenerative change & sclerosis
(so usually drop high T-score vertebra)
Spine QA
Occasionally see spurious decrease in T-score due to
interval surgery or laminectomy –drop low T-score.
Sometimes T-scores change > 1.0 between every
vertebrae.
"Art of medicine" may be invoked on challenging cases.
e.g. if dropping a vertebra doesn't change diagnosis
may not want to drop it.
Spine QA
Use all evaluable vertebrae and only exclude vertebrae
that are affected by structural change or artifact.
Spine QA
Use all evaluable vertebrae and only exclude vertebrae
that are affected by structural change or artifact.
However, if only one evaluable vertebra remains,
then must use other skeletal sites and exclude
spine. Must have ≥ 2 vertebrae to use spine BMD.
Spine QA
Use all evaluable vertebrae and only exclude vertebrae
that are affected by structural change or artifact.
However, if only one evaluable vertebra remains,
then must use other skeletal sites and exclude
spine. Must have ≥ 2 vertebrae to use spine BMD.
If lots of variation between vertebral bodies, likely will
need additional imaging of hips/dominant forearm in
the future or now!
Spine QA
COMPARISON QA
Check and see if positioning and numbering is
the same.
Spine QA
COMPARISON QA
Check and see if positioning and numbering is the
same.
If numbering is questionable, keep consistent
with prior study unless prior study clearly
mislabeled.
Spine QA
COMPARISON QA
Check and see if positioning and numbering is the
same.
If numbering is questionable, keep consistent with
prior study unless prior study clearly mislabeled.
Some DXA units will let you correct and reanalyze
prior data.
Spine QA
COMPARISON QA
Are ROIs similar? Check edges.
Spine QA
COMPARISON QA
Are ROIs similar? Check edges.
Compare prior BMDs and T-scores at EVERY level.
Spine QA
COMPARISON QA
Are ROIs similar? Check edges.
Compare prior BMDs and T-scores at EVERY level.
Is one level or two significantly different?
Spine QA
COMPARISON QA
Are ROIs similar? Check edges.
Compare prior BMDs and T-scores at EVERY level.
Is one level or two significantly different?
Review images for cause. e.g. interval fracture,
procedure, overlying artifact, etc..
Spine QA
COMPARISON QA --we depend on YOU!!
Consistency and precision are critical!
--Spine BMD has the best precision and is the most
responsive to therapy. Patient care is impacted.
--Significant BMD changes affect therapy decisions.
“Significant” is often just a few or several percent
difference.
Spine QA
COMPARISON QA --we depend on YOU!!
Consistency and precision are critical!
Precision errors affect health care costs, may result
in unnecessary change in treatment (more
expensive or more side effects) or lack of treatment
or unnecessary specialist referral or additional
diagnostic work-up to assess for secondary
osteoporosis, patient anxiety, etc..
Spine QA
COMPARISON QA
A national bone metabolism and bone health expert
reported : #1 reason for referral for significant bone
loss, despite appropriate medical therapy…
???
Spine QA
COMPARISON QA
A national bone metabolism and bone health expert
reported : #1 reason for referral for significant bone
loss, despite appropriate medical therapy…
POOR COMPARISON TECHNIQUE!!
HIP
 Positioning
-remove shoes and foot in positioning
device
Shoes can vary year to year, depending on the
season, etc.
HIP
 Positioning
-remove shoes and foot in positioning
device
-femoral neck centered in the image
HIP
 Positioning
-remove shoes and foot in positioning
device
-femoral neck centered in the image
-minimal lesser trochanter apparent
(internally rotated)
HIP
 Positioning
-vertical orientation of diaphysis/shaft
HIP
 Positioning
-vertical orientation of diaphysis/shaft
-total hip includes ischium and greater
trochanter
HIP
 Positioning
-vertical orientation of diaphysis/shaft
-total hip includes ischium and greater
trochanter
-verify bone margins!
HIP
Positioning
Rule of thumb for outer margins:
 5 mm superior to femoral head
 5 mm medial to femoral head
 5 mm lateral to the greater trochanter
 10 mm inferior to lesser trochanter
HIP
 Positioning
appropriate ROI box placement--slightly different by
manufacturer
e.g. femoral neck rectangular ROI should not include
ischium if possible (usually can auto-subtract
"neutralize" ischium vs. 2nd option narrow the box or
3rd option, move the box)
HIP
 Femoral Neck ROI
GE Lunar: finds narrowest point of femoral
neck and all 4 corners in soft tissue
HIP -GE example
HIP -GE example
HIP
 Femoral Neck ROI
Hologic Discovery: anchors it to the edge of greater
trochanter with 3 corners in soft tissue, one corner in
bone, touches superomedial margin of greater
trochanter
15 mm DO NOT change
HIP -Hologic example
HIP
 Femoral Neck ROI
Neither femoral neck box should include greater
trochanter or ischium --neutralize ischium if
necessary.
HIP
 Lower inferior margin
GE auto selects it based on greater trochanter triangle
Hologic supposed to find inferior margin of lesser
trochanter and go 1 cm inferior.
HIP
Check for artifacts, surgical clips, metal,
calcification, keys, coins, buttons, snaps,
zippers, objects external to patient. Exclude
hip if prior fracture or surgery.
If foreign body or object external to patient,
remove and rescan!
HIP
For diagnosis, use femoral neck or total
proximal femur, whichever is lowest.
Either hip or bilateral mean hip may be
used.
HIP
COMPARISON QA
Total hip is preferred for monitoring and followup because of better precision than femoral neck.
Mean total hip may be used.
Femoral neck only as a final alternative site for
follow-up.
HIP
COMPARISON QA
We depend on you!
HIP
COMPARISON QA
We depend on you!
 Check for similar positioning.
HIP
COMPARISON QA
We depend on you!
 Check for similar positioning.
 Are ROIs similar? Check edges.
HIP
COMPARISON QA
We depend on you!
 Compare prior BMDs and T-scores at total hip and
femoral neck.
HIP
COMPARISON QA
We depend on you!
 Compare prior BMDs and T-scores at total hip and
femoral neck.
 Is one significantly different? Review images for
cause. e.g. interval fracture, procedure, overlying
artifact, etc..
FOREARM
Alternate site and not primary preferred site
as it does not respond well to therapy
Use if another site is not evaluable (e.g. hip
replacements, spinal surgery, significant
degenerative change/scoliosis, obesity
exceeding table weight limit, etc.)
FOREARM
Use only 33% radius (GE)
=
1/3 radius BMD (Hologic)
FOREARM
Use only 33% radius = 1/3 radius BMD
Non-dominant forearm is
recommended for diagnosis.
FOREARM
Use only 33% radius = 1/3 radius BMD
Non-dominant forearm is
recommended for diagnosis.
If history of fracture of
nondominant wrist or forearm as an
adult, then use dominant forearm.
FOREARM
However, if patient has
hyperparathyroidism always do
forearm.
Hyperparathyroidism targets cortical
bone and remember forearm is 99%
cortical bone, so more sensitive.
FOREARM
QA
 Nondominant forearm
 Centered
 Radius & ulna straight, parallel to long axis of table
 Distal cortex of radius & ulna and proximal
carpal row visible
 Evaluate for artifacts , jewelry, watch, buttons, etc.
 Evaluate margins and edges
FOREARM GE example
FOREARM Hologic example
FOREARM
COMPARISON QA --we depend on you!
 Compare positioning
FOREARM
COMPARISON QA --we depend on you!
 Compare positioning
 Compare ROIs and edges for consistency
FOREARM
COMPARISON QA --we depend on you!
 Compare positioning
 Compare ROIs and edges for consistency
 Check old and new BMDs and T-scores
Comparison
Cannot accurately compare between different
machines due to different:
-dual energy methods
-calibration and least significant change
-detectors
-edge detection software
-data bases
-ROI techniques, maintenance, etc.
Comparison
Cannot accurately compare between different
machines—even if the same model.
So generalize…
If prior study showed normal BMD and today low
bone mass, then trend is worsening.
If low bone mass on prior study and today shows low
bone mass as well, relatively stable.
Comparison
So that is why it is so important to be imaged at the
same site, same exact DXA unit each year.
And why it is so important to be conscientious and
careful about positioning and critically reviewing your
work! Aim for high precision!
Comparison
Write notes and document in the patient’s chart about
special modifications or positioning tricks that worked to
assist the technologist next time and to improve
consistency and precision.
e.g. had the patient hold up panniculus during scanning of
the hip.
or tilted pelvis to better center and straighten spine.
Reminders about doing forearm or bilateral hips.
Comparison
The manufacture’s provided precision error and least
significant change (LSC) are unrealistic--calculated under
optimal conditions, using ideal patients and experienced
expert technologists. Very low error and small LSC.
So using their precision error and LSC overestimates
change!
Which means, we may diagnosis significant change even
when there really is not! Affects Rx decisions!
Calculating Precision Error and LSC
Please strongly consider measuring and calculating
your site’s precision error and least significant
change so you know your precision and what change
is significant for your specific site.
Must calculate the precision error for lumbar spine and
total hip separately.
Calculating Precision Error and LSC
Measure both spine and hip on 30 patients twice.
e.g. first patient or two of the day over several weeks or
every patient until get to 30, extremely low radiation.
Measure hip and spine and then it is critical that you
get the patient up and off the table completely
before rescanning. You do a separate second scan
and, as always, try to perform the scan exactly the
same way.
Calculating Precision Error and LSC
Calculate the root mean square standard deviation for
the group.
Calculate least significant change at 95% confidence
interval.
??
No big deal, you loved math in school, right?
Calculating Precision Error and LSC
It really is simple, just go to the internet!
Precision Calculating Tool or Excel program free at
http://www.iscd.org/resources/calculators/
Calculating Precision Error and LSC
The least significant change (LSC) is calculated separately
for spine and total hip.
LSC of your site is used to calculate whether the interval
change in the patient’s BMD is significant or
nonsignificant.
e.g. the change in BMD must exceed the least significant
change of your facility, otherwise it is simply normal
variation in precision.
Calculating Precision Error and LSC
For example
Most recent prior total hip BMD
0.866 gm/cm²
Today’s follow-up total hip BMD
- 0.832 gm/cm²
Difference (decrease) between two 0.034 gm/cm²
Calculated least significant change (LSC) for our site is 0.022 gm/cm²
or expressed as a percentage x 100 = 2.2%
The difference between the 2 studies is greater than the LSC, so the
0.034 gm/cm² change is significant!
Calculating Precision Error and LSC
For example
Most recent prior total hip BMD
Today’s follow-up total hip BMD
Difference between two
0.866 gm/cm²
- 0.832 gm/cm²
0.034 gm/cm²
Calculated least significant change (LSC) for our site is
0.022 gm/cm²
or
expressed as a percentage x 100 = 2.2%
The difference between the 2 studies is greater than the LSC, so the
0.034 gm/cm² change is significant!
Calculating Precision Error and LSC
For example
Most recent prior total hip BMD
Today’s follow-up total hip BMD
Difference between two
0.866 gm/cm²
- 0.832 gm/cm²
0.034 gm/cm²
Calculated least significant change (LSC) for our site is 0.022 gm/cm²
or expressed as a percentage x 100 = 2.2%
The difference between the 2 studies is greater than the LSC, so
the 0.034 gm/cm² decrease since the prior exam is significant!
Calculating Precision Error and LSC
For example:
Most recent prior total hip BMD
Today’s follow-up total hip BMD
Difference between two
0.866 gm/cm²
- 0.832 gm/cm²
0.034 gm/cm²
Calculated least significant change (LSC) for our site is 0.022 gm/cm²
or expressed as a percentage x 100 = 2.2%
The difference between the 2 studies is greater than the LSC, so the
0.034 gm/cm² change (decrease) is significant!
If there is not a significant change, then you are done.
Calculating Precision Error and LSC
If there is a significant change, then the change can also be reported as
percentage change ± LSC. Here you do have to use your math skills.
Example:
(change in BMD ÷ prior BMD ) x 100 = significant percentage change.
Change in BMD
Prior BMD
x 100 = significant percentage change
Then ± LSC expressed as a percentage.
0.034 gm/cm²
0.866 gm/cm²
x 100 = 3.9 % significant decrease since prior exam.
or
3.9 % ± 2.2 %
Calculating Precision Error and LSC
If there is a significant change, then the change can also be reported as percentage change ± LSC. Here
you do have to use your math skills.
Example:
(change in BMD ÷ prior BMD ) x 100 = significant percentage change.
Then ± least significant change expressed as a percentage.
For example:
0.034 gm/cm²
0.866 gm/cm²
x 100 = 3.9 % significant decrease since prior exam.
or
3.9 % ± LSC or 3.9 % ± 2.2%
Calculating Precision Error and LSC
If there is a significant change, then the change can also be reported as
percentage change ± LSC. Here you do have to use your math skills.
Example:
(change in BMD ÷ prior BMD ) x 100 = significant percentage change.
Then ± LSC expressed as a percentage.
For example
0.034 gm/cm²
0.866 gm/cm²
x 100 = 3.9 % significant decrease since prior exam.
or
3.9 % ± 2.2 %
Calculating Precision Error and LSC
The minimum acceptable precision for an individual
technologist is:

Lumbar Spine: 1.9 % LSC = 5.3%

Total Hip: 1.8% LSC = 5.0%

Femoral Neck: 2.5% LSC = 6.9%
DXA
References and Sources:
 International Society for Clinical Densitometry (ICSD)
2007 Official Positions and Pediatric Official Positions
 ICSD Bone Densitometry Course for Clinicians 2012
 Osteoporosis Essentials: Densitometry, Diagnosis and
Management 2013 (ISCD and IOF)
 Bone Densitometry in Clinical Practice: Application
and Interpretation 3rd Ed. 2010 by Sydney Lou
Bonnick, MD
http://www.shef.ac.uk/FRAX/
The FRAX® tool has been developed by WHO
to evaluate fracture risk of patients.
It is based on individual patient models that
integrate the risks associated with clinical
risk factors as well as bone mineral density
(BMD) at the femoral neck.
FRAX® can help classify risk especially in
osteopenic patients and determine need
for treatment.
NOT to be used if patient is already
diagnosed as osteoporotic.
Not to be used if patient is ALREADY on
prescription treatment but Vitamin D or
calcium supplements-- okay to use FRAX®.
FRAX®
 1. Age: used for ages 40-90
 2. Race:
 3. Gender:
Caucasian / Black / Hispanic / Asian
Female / Male
 4. Weight: kg!
 5. Height: cm!
6. Previous fracture i.e. previous fracture after age 40
occurring spontaneously, or a fracture arising from
trauma which, in a healthy individual, would not have
resulted in a fracture (such as a ground level fall.)
 Do NOT include fractures of the skull, hands and
ankles/feet.
 However, DO include vertebral/spine fractures
detected by x-ray alone. Patient may be unaware!
FRAX®
 7. Mother or father fractured hip:
 8. Current smoking: Y or N
Y or N
FRAX®
 9. Glucocorticoids/STEROIDS/prednisone, etc. i.e.
currently using oral glucocorticoids/steroids or has
been exposed to oral glucocorticoids for more than 3
months at a dose of prednisolone/prednisone of 5 mg
daily or more (or equivalent doses of other
glucocorticoids): Y or N
FRAX®
 10. Rheumatoid arthritis (confirmed diagnosis): Y or
N
FRAX®
 11. Secondary Osteoporosis (Answer "yes" if the
patient has Type I diabetes (insulin dependent),
osteogenesis imperfecta as an adult, untreated longstanding hyperthyroidism, hypogonadism or
premature menopause (<45 years old), chronic
malnutrition, malabsorption or chronic liver disease)
FRAX®
 12. Alcohol 3 or more units per day (e.g. 1 unit = 1 glass
of beer, 1 medium-sized glass of wine, 1 shot/1 oz of
liquor): Y or N
FRAX®
 REMEMBER: GE LUNAR vs. HOLOGIC AND
FEMORAL NECK BMD!!
 REMEMBER: North America-->US Database and
select proper Race.
 IF NOT CERTAIN OF AN ANSWER, SELECT "NO."
Consider FDA-approved medical therapies
in postmenopausal women and men aged
50 years and older, based on the following:
•A hip or vertebral (clinical or
morphometric) fracture
•T-score ≤ -2.5 at the femoral neck or spine
after appropriate evaluation to exclude
secondary causes
Consider FDA-approved medical therapies in
postmenopausal women and men aged 50 years and
older, based on the following:
•Low bone mass (T-score between -1.0 and -2.5 at
the femoral neck or spine) and a 10-year probability
of a hip fracture ≥ 3% or a 10-year probability of a
major osteoporosis-related fracture ≥ 20% based on
the US-adapted WHO algorithm
•Clinicians judgment and/or patient preferences may
indicate treatment for people with 10-year fracture
probabilities above or below these levels
THANK YOU so much for your attention!
But wait, there’s more.