AJR 2000; 174:1699-1705
© American Roentgen Ray Society
Dual X-Ray Absorptiometry
Recognizing Image Artifacts and Pathology
Jon A. Jacobson1,
David A. Jamadar and
Curtis W. Hayes
1
All authors: Department of Radiology, University of Michigan Medical Center,
1500 E. Medical Center Dr. TC-2910G, Ann Arbor, MI 48109-0326.
Received August 9, 1999;
accepted after revision November 16, 1999.
Address correspondence to J. A. Jacobson.
Introduction
Dual X-ray absorptiometry (DXA) is considered the method of choice for
measurement of low bone mineral density (BMD) associated with osteoporosis
[1]. Although the diagnosis of
osteopenia and osteoporosis with DXA relies on analysis of numeric data, the
importance of the DXA scan cannot be overemphasized. With current technology,
the resolution of the DXA scan allows detailed visualization of osseous
structures. This visualization enables correct region-of-interest (ROI)
placement for BMD measurements and confirms correct patient positioning. In
addition, the image can display artifacts that may adversely affect the BMD
measurement. Staron et al. [2]
have shown that operator-dependent errors related to incorrect ROI placement
or artifacts result in incorrect analysis more than 2% of the time.
Recognition of abnormalities and disorders on the DXA scan is also inherently
important and may affect BMD measurements. Therefore, a detailed assessment of
the DXA scan should be integrated into every DXA scan interpretation.
ROI and Patient Position Assessment
The DXA image should be assessed on several levels. Initially during DXA
scan acquisition, the technologist uses the image to confirm adequate
positioning and to ensure appropriate placement of ROIs, including appropriate
intervertebral designations (Fig.
1A,1B).
The image should be assessed for patient motion because motion produces
unpredictable alteration in bone mineral content measurements
[3]. Identification of
overlying hardware or retained barium may obscure the osseous structures,
making BMD measurements at a site invalid
(Fig. 2).
View larger version (165K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B. —Standard regions of interest (ROI) in dual X-ray absorptiometry
(DXA) in 60-year-old woman. DXA image of left hip shows ROI for femoral neck
(open arrows) and Ward's triangle (wavy arrow). Note
trochanteric ROI (arrowheads) and intertrochanteric ROI (solid
arrows). Total ROI comprises these four ROIs.
|
|
View larger version (104K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2. —Lumbar spine fixation in 78-year-old man. Dual X-ray absorptiometric
image of lumbar spine shows metal spinal rods (solid arrows)
precluding bone mineral density measurement. Note electric stimulating device
(open arrow).
|
|
At the time of the DXA result interpretation, the physician must also
critically assess the DXA image. First, correct patient positioning and ROI
placement are confirmed. Improper patient positioning and scan analysis may
result in poor precision or poor reproducibility, both of which will limit the
ability of the machine to measure small changes over time
[4]. Correct numbering of
lumbar vertebral levels and correct ROI placement are especially important
when performing serial BMD measurements to minimize scan result variability.
In fact, Staron et al. [2]
reported that incorrect placement of the intervertebral disk spaces on the
image for ROI placement was the most common operator-dependent error in spine
BMD measurements. The inclusion of unwanted calcified or osseous structures
(Fig.
3A,3B),
sclerotic abnormalities, or other dense objects in the ROI can spuriously
increase the measured BMD. Similarly, excluding osseous structures from the
ROI can affect the BMD (Fig.
4A,4B,4C,4D).
In patients with prior surgery such as laminectomy, the BMD also may be
affected [5] (Fig.
5A,5B).
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B. —Calcified costocartilage in 71-year-old woman. Note increase in T
score of L1 ROI (arrow) relative to remaining spinal levels. (T score
represents number of standard deviations from mean bone mineral density [BMD]
of young adult reference population).Z=number of standard deviations from the
mean BMD of an age-matched reference population.
|
|
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4A. —Osseous structures excluded from region of interest (ROI) because of
extreme osteopenia in 77-year-old man. Dual X-ray absorptiometric (DXA) image
of lumbar spine shows osteopenic bone (arrows) not included in
ROI.
|
|
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B. —Osseous structures excluded from region of interest (ROI) because of
extreme osteopenia in 77-year-old man. Note L2-L4 bone mineral density (BMD)
measurements (arrow). T = number of standard deviations from mean BMD
of young adult reference population, Z = number of standard deviations from
the mean BMD of an age-matched reference population.
|
|
View larger version (111K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4C. —Osseous structures excluded from region of interest (ROI) because of
extreme osteopenia in 77-year-old man. DXA image of lumbar spine shows
osteopenic bone (arrows) is now included in ROI.
|
|
View larger version (17K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4D. —Osseous structures excluded from region of interest (ROI) because of
extreme osteopenia in 77-year-old man. Resulting bone BMD (arrow) is
now decreased. Omission of osteopenic bone in A caused artifactual
increase of BMD measurement in B (arrow). T = number of
standard deviations from mean BMD of young adult reference population, Z =
number of standard deviations from the mean BMD of an age-matched reference
population.
|
|
View larger version (17K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5B. —Lumbar laminectomy in 64-year-old woman. Prior laminectomy causes
decrease in T score at L4 (arrow) compared with other spinal levels.
BMD = bone mineral density, T = number of standard deviations from mean BMD of
young adult reference population, Z = number of standard deviations from the
mean BMD of an age-matched reference population.
|
|
Artifacts Potentially Affecting BMD
The DXA scan should be assessed for visible artifacts. If included in the
ROI, dense artifacts typically increase the measured BMD. Staron et al.
[2] reported that the most
severe errors in densitometry analysis occurred when radiopaque artifacts were
included in the ROI. Examples include cardiac pacemaker leads (Fig.
6A,6B),
an inferior vena cava filter (Fig.
7), retained intestinal barium
(Fig. 8), retained
iophendylate myelographic contrast agent (Pantopaque; Lafayette Pharmacol,
Lafayette, IN) (Fig.
9A,9B),
and metal bullet fragments. If an object lies in a ROI, that particular ROI
may be excluded from the measurement. For example, the L1 vertebra affected by
an overlying artifact may be excluded, and the L2-L4 vertebrae may be used for
BMD measurement. The technologist may complete this measurement manually,
although software features can assist in this task. If the affected ROI cannot
be excluded from measurement, an alternate site of BMD measurement should be
chosen. If not included in the ROI (and not affecting the BMD measurement),
artifacts such as cholecystectomy clips
(Fig. 7) may be noted as
incidental findings.
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6A. —Pacemaker lead in 85-year-old woman. Dual X-ray absorptiometric
image of lumbar spine shows motion of cardiac pacemaker leads (solid
arrows) included in L1 vertebra region of interest. Note degenerative
sclerosis at L4 level (open arrow).
|
|
View larger version (13K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 6B. —Pacemaker lead in 85-year-old woman. Pacemaker lead causes
factitious increase in T score at L1 (solid arrow) when compared with
other spinal levels. The degenerative sclerosis at L4 also increases T score
(open arrow). BMD = bone mineral density, T = number of standard
deviations from mean BMD of young adult reference population, Z = number of
standard deviations from the mean BMD of an age-matched reference
population.
|
|
View larger version (137K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 7. —Spinal artifacts in 73-year-old woman with inferior vena cava
filter. Dual X-ray absorptiometric image of lumbar spine shows inferior vena
cava filter (solid arrows) that may increase bone mineral density
measurement (BMD) included in L3 vertebral region of interest (ROI).
Incidental note is made of cholecystectomy clips (open arrow) not
affecting BMD measurement outside ROI.
|
|
View larger version (147K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 8. —Spinal artifacts in 68-year-old woman with retained barium. Dual
X-ray absorptiometric image of lumbar spine shows retained intestinal barium
(arrows) that may artifactually increase bone mineral density
measurement in L4 vertebral region of interest.
|
|
View larger version (133K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9A. —Spinal artifact of myelographic contrast agent in 82-year-old woman.
Dual X-ray absorptiometric image of lumbar spine shows retained iophendylate
myelographic contrast agent (Pantopaque; Lafayette Pharmacol, Lafayette, IN)
(solid arrow) included in L2 vertebral region of interest (ROI). Note
sclerotic degenerative change at L4 (open arrow).
|
|
View larger version (18K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 9B. —Spinal artifact of myelographic contrast agent in 82-year-old woman.
Inclusion of myelographic contrast agent in ROI may be artifactually
increasing T score (solid arrow) relative to L3 level. Note
degenerative changes increasing T score of L4 ROI (open arrow). BMD =
bone mineral density, T = number of standard deviations from mean BMD of young
adult reference population, Z = number of standard deviations from the mean
BMD of an age-matched reference population.
|
|
Disorders Affecting BMD
The DXA image is also assessed for visible disease processes. In addition
to noting the presence of these abnormal conditions in the DXA report, it is
important to determine how these processes affect the BMD measurement. In the
lumbar spine, degenerative disk disease and facet osteoarthrosis (Fig.
10A,10B)
can artifactually increase the BMD measurement. The affected vertebral levels
should be excluded from the ROI, or another BMD measurement site should be
selected. Lateral DXA may be used to reduce the effect of sclerotic facet
osteoarthrosis on the BMD measurement because this area would be excluded from
the ROI [2]. When a
sclerotic-appearing vertebral body is recognized, possible conditions include
Paget's disease, lymphoma, metastasis, and compression fracture (Fig.
11A,11B).
The latter diagnosis is suggested by noting decreased vertebral body height on
the DXA scan. Correlation with radiography typically is required to define the
disease process further. Regardless of the cause, the ROI containing the
abnormal sclerotic process should not be used for BMD measurement because this
inclusion will cause an increase in the resulting values.
View larger version (112K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10A. —Degenerative changes of lumbar spine in 72-year-old woman. Dual
X-ray absorptiometric image of lumbar spine shows sclerotic degenerative disk
disease and facet osteoarthrosis (arrows).
|
|
View larger version (17K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 10B. —Degenerative changes of lumbar spine in 72-year-old woman. Sclerotic
degenerative changes increase T scores at affected levels (arrows).
BMD = bone mineral density, T = number of standard deviations from mean BMD of
young adult reference population, Z = number of standard deviations from the
mean BMD of an age-matched reference population.
|
|
View larger version (18K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 11B. —Compression fracture in 68-year-old woman. Compression fracture
causes artifactual increase in T score relative to other spinal levels
(arrow). BMD = bone mineral density, T = number of standard
deviations from mean BMD of young adult reference population, Z = number of
standard deviations from the mean BMD of an age-matched reference
population.
|
|
In the hip, sclerotic conditions such as bone islands (enostosis) (Fig.
12A,12B)
may increase the BMD measurement if they are included in the ROI. Significant
osteoarthrosis (Fig. 13) may
increase the BMD measurement if cortical thickening or buttressing extends
into the femoral neck ROI. Similarly, Paget's disease (Fig.
14A,14B),
calcific tendinitis (calcium hydroxyapatite deposition)
(Fig. 15), and vascular
calcifications (Fig. 16) may
increase the BMD measurement if included in the ROI. If a disease process is
affecting the BMD measurement in a particular ROI, another ROI or,
alternatively, another site may be used. Other disease conditions such as
avascular necrosis of the femoral head
(Fig. 17) and developmental
dysplasia of the hip (Fig. 18)
may be visible on the DXA image but will not affect the BMD measurement
because these conditions are outside the ROI boundaries. It is essential,
however, that a suspected abnormality on the DXA image be evaluated further
with correlative imaging to confirm a diagnosis and to exclude potential
aggressive or malignant causes.
View larger version (175K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12A. —Bone island in 86-year-old woman. Dual X-ray absorptiometric image
of hip shows sclerotic focus in intertrochanteric region of proximal femur
(arrow), representing enostosis (bone island).
|
|
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 12B. —Bone island in 86-year-old woman. Increased T score in
intertrochanteric region of interest (arrow) may be caused by bone
island. BMD = bone mineral density, T = number of standard deviations from
mean BMD of young adult reference population, Z = number of standard
deviations from the mean BMD of an age-matched reference population.
|
|
View larger version (165K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 13. —Hip osteoarthrosis in 74-year-old woman. Dual X-ray absorptiometric
image of hip shows narrowing of hip joint with sclerotic osteophytes
(arrows). Bone sclerosis does not extend into femoral neck region of
interest and, therefore, does not affect measured bone mineral density.
|
|
View larger version (170K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14A. —Paget's disease in 93-year-old man. Dual X-ray absorptiometric image
of hip shows trabecular thickening and disorganization, cortical thickening,
and osseous expansion (arrowheads) of acetabulum, compatible with
Paget's disease.
|
|
View larger version (16K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 14B. —Paget's disease in 93-year-old man. It is unclear if increased T
score of femoral neck (arrow) is caused by involvement of Paget's
disease or superimposed osteoarthrosis. BMD = bone mineral density, T = number
of standard deviations from mean BMD of young adult reference population, Z =
number of standard deviations from the mean BMD of an age-matched reference
population.
|
|
View larger version (185K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 15. —72-year-old woman with calcific tendinitis. Dual X-ray
absorptiometric image of hip shows sclerotic focus (arrow) indicating
calcium hydroxyapatite deposition in gluteus medius and minimus tendons
(calcific tendinitis). This calcification does not affect bone mineral density
measurement because it is excluded from region-of-interest boundary.
|
|
View larger version (158K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 16. —80-year-old woman with vascular calcifications. Dual X-ray
absorptiometric image of hip shows vascular calcifications (arrows).
If extensive and included in region of interest (ROI), such calcifications may
increase bone mineral density measurement. In this case, calcifications are
not included in total ROI boundary.
|
|
View larger version (153K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 17. —74-year-old woman with femoral head avascular necrosis. Dual X-ray
absorptiometric image of hip shows crescentic sclerotic superior margin of
femoral head (arrowheads) representing avascular necrosis. This
condition has no effect on bone mineral density measurement because sclerosis
does not extend into femoral neck region of interest.
|
|
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 18. —61-year-old woman with developmental dysplasia of hip. Dual X-ray
absorptiometric image of hip shows lateral upsloping of acetabulum and
uncovering of femoral head (arrows) compatible with developmental
dysplasia of hip. Superimposed osteoarthrosis does not extend to femoral neck
region of interest and does not affect bone mineral density measurement.
|
|
Conclusion
Critical assessment of the DXA scan by the technologist and interpreting
physician is essential to ensure accurate BMD measurement. Artifacts should be
identified and excluded from the ROI before BMD measurements. Many pathologic
conditions may also be identified on the DXA image. If included in the ROI,
the BMD measurement at that site should be disregarded and another site should
be selected.
References
-
Lenchik L, Sartoris DJ. Current concepts in osteoporosis.
AJR
1997;168:905
-911[Free Full Text]
-
Staron RB, Greenspan R, Miller TT, Bilezikian JP, Shane E, Haramati
N. Computerized bone densitometric analysis: operator-dependent errors.
Radiology
1999;211:467
-470[Abstract/Free Full Text]
-
Cawkwell GD. Movement artifact and dual X-ray absorptiometry.
J Clin Densitom
1998;1:141
-147
-
Lenchik L, Rochmis P, Sartoris DJ. Optimized interpretation and
reporting of dual X-ray absorptiometry (DXA) scans.
AJR
1998;171:1509
-1519[Free Full Text]
-
Spencer RP, Szigeti DP, Engin IO. Effect of laminectomy on measured
bone density. J Clin Densitom
1998;1:375
-377
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
|
|
 
References
J. ICRU,
April 1, 2009;
9(1):
115 - 131.
[PDF]
|
|
|
Top
Introduction
ROI and Patient Position...
