Vertebral Fracture Initiative

Part II

Radiological Assessment of Vertebral Fracture Authored by: Judith E Adams1, Leon Lenchik2, Christian Roux3 and Harry K. Genant4 1. Clinical Radiology, The Royal Infirmary, Oxford Road, Manchester, M13 9WL, UK and Imaging Science and Biomedical Engineering, University of Manchester 2. Department of Radiology, Wake Forest University School of Medicine, Medical Center Blvd., Winston-Salem, NC 27157-1088, USA 3. Paris Descartes University, Cochin Hospital, Rheumatology Department, Paris, France

4. Departments of Radiology, Medicine and Orthopedic Surgery, University of California, San Francisco

1

CONTENTS: Executive summary 1) Introduction 2) Indications for spinal radiographs 3) Acquisition of spinal radiographs a) Ideal protocol b) Problems which may arise 4) Vertebral fractures a) Radiographic appearance, severity (grading) 5) Fortuitous diagnosis of vertebral fractures Lateral chest radiographs, abdominal radiographs, barium studies, computed tomography (CT) scans, magnetic resonance imaging (MRI) scans and radionuclide scans (RNS) 6) Differential diagnosis

Differentiation from other causes of vertebral deformities 7) Reporting a) Clear and accurate terminology b) Importance to FRAX® calculator c) Suggest referral for central DXA 8) Conclusions 9) References 10) Appendices a) Standardized methods for fracture assessment b) Semi-quantitative (SQ) method c) Comparison between semi-quantitative and quantitative techniques d) Algorithm based qualitative (ABQ) assessment e) References

2

EXECUTIVE SUMMARY

Vertebral fractures are powerful predictors of future spine and hip fractures, so accurate

diagnosis and clear, unambiguous reporting are essential.

There is considerable evidence that vertebral fractures are under-reported, and when present

appropriate intervention may not occur.

The purpose of this document is to raise awareness of the relevance and importance of

identification of vertebral fractures, be it on spinal radiographs or fortuitously from other

images (lateral chest radiographs, mid-sagittal spinal reformations from multi-detector

computed tomography [MDCT] of thorax and abdomen, magnetic resonance imaging [MRI]

and radionuclide scans [RNS]).

Methods to differentiate vertebral fractures from other causes of vertebral deformities are

outlined. The aim is to improve the diagnosis and management of osteoporosis and so reduce

fractures and suffering.

3

1. INTRODUCTION

Osteoporosis-related vertebral fractures have important health consequences for older

women, including disability and increased mortality (1). As further fractures can be

prevented with appropriate medications, recognition and treatment of these high-risk patients

is warranted. Hence the early and accurate diagnosis of vertebral fractures is an important

factor in optimizing the clinical management of patients with osteoporosis.

Although osteoporotic vertebral fractures are common in men and women, and the presence

of these fractures indicates that patients are at substantially increased risk for new fractures of

the spine and hip (2), there is strong evidence of widespread under-diagnosis of vertebral

fractures (3-6). In particular, clinicians often fail to recognize or report mild and moderate

vertebral fractures, or use terminology that is not specific for fracture. There is therefore an

urgent need to improve evaluation of patients who have vertebral fracture.

The purpose of this document is to emphasize the importance of appropriate diagnosis of

vertebral fractures in osteoporosis, and to provide a basis for standardization of radiographic

acquisition and radiological interpretation that require no specialized equipment and can be

performed by any appropriately trained clinician. Improved and accurate diagnosis of

vertebral fractures will enhance patient evaluation and the ability to target appropriate

therapeutic intervention to those patients who would benefit most, and so reduce the risk of

future fracture.

4

2. INDICATIONS FOR SPINAL RADIOGRAPHS

Clinical indications for spine radiographs, in the absence of trauma or malignancy, include

acute back pain, focal tenderness, loss of height and known, or suspected, cases of

osteoporosis, either primary or due to secondary causes (7).

Spinal radiographs, and dual energy X-ray absorptiometry (DXA), would also be appropriate

in patients over 50 years of age who have other radiographic features suggesting osteoporosis

(thinned cortices, reduced density [radiographic osteopenia], reduced number of trabeculae)

in any skeletal site (8-10).

3. ACQUISITION OF SPINAL RADIOGRAPHS

A. Ideal Radiographic Technique

For the initial assessment of vertebral osteoporotic fracture, spinal radiographs are still the

most common imaging technique used. Separate antero-posterior (AP) and lateral

radiographic views of the thoracic and lumbar spine are used. For follow-up examination

lateral thoracic and lumbar spine radiographs generally suffice. Radiographs of the thoracic

and lumbar spine should be acquired using a standardized protocol so that there is consistent

technique and good quality radiographs are obtained (11, 12). The focus-to-film distance

(FFD) is generally 100 cm. Thoracic radiographs are centered at T7 and the lumbar

radiographs at L3.

5

Antero-posterior (AP) spinal radiographs

AP views of the spine are used to accurately define the number of vertebrae present and may

aid in the detection of vertebral fracture. On the AP views, all the relevant vertebrae should

be clearly visible on the radiograph; for the thoracic spine vertebrae C7-L1, and for the AP

lumbar spine T12 to S1, should be visible (Fig 1a and b). For the thoracic spine the top of the

X-ray cassette is placed 5 cm (2 inches) above the shoulders. For the lumbar spine the natural

lumbar lordosis has to be reduced so that the spine is flat on the X-ray table. This is achieved

by flexing the hips and knees, with a small supporting pad being placed under the knees. The

vertebral levels are accurately identified by counting down from the top of the thoracic spine.

With this method, anomalies in the number of thoracic and lumbar vertebrae can be

identified.

Adequate collimation of the X-ray beam is important so that radiosensitive organs such as the

breast and thyroid are not unnecessarily irradiated (13), and the radiation dose to the patient

is kept to a minimum (Table 1). Adequate collimation also reduces scattered radiation and

thus improves contrast. For the AP thoracic view, the collimation should not be too narrow;

if cervical ribs or vestigial ribs at T12 are present, they should be clearly visible (Fig 1a).

6

Table 1. Typical patient effective radiation doses: from spine, chest and other radiographic examinations. The average effective doses from the annual natural background radiation and from a return transatlantic flight are given for comparison.

Type of exposure Effective Dose (mSv)

Thoracic spine

AP 0.41,2

Lateral 0.31,2

Lumbar spine

AP 0.71,2

Lateral 0.31,2

PA Chest 0.021,2

Pencil beam DXA (spine) <0.0013

Fan beam DXA (spine) ~ 0.014

Quantitative computed tomography (QCT): spine 0.065

Annual natural background radiation (NBR) 2.46

Return transatlantic flight (16 hours total flight time) ~0.077 1 Wall BF, Hart D 1997 Revised radiation doses for typical X-ray examinations Report on a recent

review of doses to patients from medical X-ray examinations in the UK by NRPB. National Radiological Protection Board. Br J Radiol;70(833):437-9

2 Hart D, Wall BF 2002 Radiation Exposure of the UK Population from Medical and Dental X-ray Examinations. National Radiation Protection Board, Oxon. 3 Lewis MK, Blake GM, Fogelman I 1994 Patient dose in dual X-ray absorptiometry. Osteoporos

Int;4(1):11-5 4 Blake G, Naeem M, Boutros M 2006 Comparison of effective dose to children and adults from dual X-ray absorptiometry examinations. Bone 38:935-42 5 Kalender WA 1992 Effective dose values in bone mineral measurements by photon absorptiometry

and computed tomography. Osteoporos Int;2:82-7 6 United Nations Scientific Committee on the Effects of Atomic Radiation. UNSCEAR 2000 Report to the General Assembly, with scientific annexes. Volume I. Vienna, Austria, UNSCEAR 7 Saez Vergara J, Romero Gutierrez AM, Rodriguez Jimenez R, Dominguez-Mompell Roman R 2004 In-flight measured and predicted ambient dose equivalent and latitude differences on effective dose estimates.Radiat Prot Dosim 110:363-70

7

Lateral spinal radiographs

Because the lateral views of the thoracic and lumbar spine are the most important for

assessment of osteoporotic fracture, time and attention should be taken to correctly position

the patient (Fig 2a and b). The important factor is to have the patient in the true lateral

position with the spine parallel to the X-ray table to avoid rotation or scoliosis, as the latter

will cause ‘tilting’ of the vertebrae causing biconcavity of the endplates (‘bean can’ effect)

(Fig 3a and b). Visualization of the thoraco-lumbar junction on the lateral thoracic radiograph

is useful to identify the vertebral levels. Accuracy in marking the vertebral levels at the

thoraco-lumbar junction is aided by visualization of the posterior spinous processes, which

change shape at the levels of T12 and L1, and also by confirmation of the presence and size

of the lower ribs on the AP thoracic view.

For the lateral thoracic spine the cassette is positioned with the top 5 cm (2 inches) above

the patient’s shoulders (Fig 2a). The important factor is to have the patient in the true lateral

position with the spine parallel to the X-ray table to avoid rotation or scoliosis. With the

patient’s shoulders, hips, knees and ankles superimposed, and with padding between the

elbows and knees, this position can be maintained. With the spine parallel to the film/X-ray

table the vertebral endplates are vertical to the film and parallel to the X-ray beam which

avoids the parallax effect of the divergent X-ray beam. The latter falsely causes apparent

biconcave endplates which must not be erroneously identified as endplate fractures (Fig 3a).

A radiolucent pad may be required under the lumbar spine at waist level to straighten the

lower thoracic spine so that it is parallel to the X-ray table and avoids sagging of the spine

8

towards the film at the waist. If this cannot be achieved due to fixed scoliosis, then

angulation of the X-ray tube towards the head may be used (11).

To visualize the upper thoracic vertebrae, the arms should be raised so that the scapulae are

not superimposed on the vertebra. If the arms are raised too high above the head, the scapulae

may be superimposed on the upper thoracic vertebral bodies, making it difficult to visualize

the vertebral endplates. This can be overcome by placing the patient’s arms at right angles to

the body.

With the breath-hold technique in the thoracic spine, the margins of the ribs may obscure the

vertebral body endplates. This may be overcome by the breathing, or long exposure,

technique. This causes movement blurring of the overlying ribs and lung parenchyma so that

the vertebral bodies are more clearly visualized. This technique may be difficult in elderly

patients, as the patient has to remain still during the longer exposure time (usually 2-4

seconds) associated with this technique. This technique is not possible on X-ray equipment

that relies on automatic exposure time. The radiation dose is a little higher when the

breathing technique is used (Table 1). The quality of the radiograph is improved further by

placing a sheet of lead rubber on the X-ray table posterior to the spine so that backscatter is

reduced.

For the lateral lumbar spine view, T12 to S1 should be visualized on the radiograph in the

true lateral position without rotation or obliquity (Fig 2b). The presence of the last 12th rib

and the lower thoracic vertebrae on the lateral lumbar spine view help to accurately define

9

the vertebral levels. This is especially so in cases of lumbarization of S1 or sacralization of

L5, where the identification of T12 is important.

As with the lateral thoracic view, it is imperative that the shoulders, hips, knees and ankles be

superimposed, and padding should be used between the elbow, knees and ankles to assist in

maintaining the true lateral position without rotation. The lumbar spine must also be parallel

to the film to avoid the biconcave endplates (‘bean can’ effect) due to tilting of the vertebrae

(Fig 3b). With the patient in the lateral decubitus position, the long axis of the lumbar spine

tends to run obliquely in direction from L1 to L5. This can be corrected by placing

radiolucent pads under the upper part of the lumbar spine. Alternatively, the X-ray tube can

be angled toward the feet so that the X-ray beam is perpendicular to the spine.

B. Technical Problems

The thickness of the shoulders overlying the upper thoracic vertebrae makes it difficult for

the X-ray beam to penetrate through the shoulder girdle. The upper lateral thoracic region is

one of the most difficult regions of the body to radiograph successfully (11). Fortunately

isolated osteoporotic fractures rarely occur at levels T1-3 (14, 15). Similarly, L5 can

sometimes be difficult to see on the lateral view because of the thickness of the pelvis (13).

Furthermore, because of the parallax effect, L5 may be difficult to visualize clearly in the

lateral position, and a coned view may be useful (16).

10

Because of the differences in shape of the chest and the presence of the radiodense heart

overlying the lower thoracic spine, it may be difficult to visualize the whole of the thoracic

spine on the AP position. This effect is a particular problem in large or kyphotic patients.

With mild scoliosis, it may be useful to obtain the radiograph as the patient lies on the side of

the convexity of the scoliosis. With the scoliotic curvature of the spine away from the X-ray

table and with the use of the parallax effect, the vertebral bodies and inter-vertebral disc

spaces may be seen more clearly.

4. VERTEBRAL FRACTURES

Clinical Identification of Vertebral Fractures

Although vertebral fractures are common in postmenopausal women, they are difficult to

identify clinically (i.e. without spinal radiographs). Large-scale prospective studies indicate

that only about one in four vertebral fractures are clinically recognized (17), and it is

relatively uncommon for patients to be referred for radiographs in the course of investigation

of osteoporosis. The lack of clinical recognition of fractures is due to both the absence of

symptoms and difficulty in determining the cause of symptoms, which may have a variety of

origins. For example, it has been estimated that less than 1% of episodes of back pain are

related to vertebral fractures (18). As a result, vertebral fractures are not commonly suspected

in patients reporting back pain, unless the back pain is associated with trauma. Trauma-

related fractures are not considered classical (atraumatic) osteoporotic fractures. Height loss,

another indicator of vertebral fractures, is also difficult to assess clinically. Some height loss

11

is expected with ageing due to the dessication and compression of intervertebral discs, and

postural changes. Studies have concluded that height loss is an unreliable indicator of

fracture status until it exceeds 4 cm (19). Unfortunately, a loss of 4 cm could also be due to

multiple vertebral fractures, by which time significant and irreparable damage may have

occurred.

The quantitative definition of a vertebral fracture is also contentious, and in epidemiology

and pharmaceutical efficacy studies a variety of morphometric measurements have been used

(Fig 4). In these six points are placed on the vertebral body: at the anterior, middle and

posterior point of the upper and inferior endplates. These points define reductions in the

anterior (wedge) and mid (endplate) vertebral heights in relation to posterior heights to

determine change in vertebral shape, or posterior height in relation to such height in adjacent

vertebrae to determine degree of crush fracture, or variations of these parameters (Appendix)

(20,21). However, in a clinical setting more simple methods for the accurate diagnosis and

classification of vertebral fractures are required. Also, if six-point morphometry alone is used

to define vertebral fractures then other pathologies which change the shape of vertebra (e.g.

Scheuermann’s disease, spondylosis, etc) will erroneously be classified as fractures (22).

a) Radiographic Identification of Vertebral Fractures

Because it is often unsuspected clinically, the diagnosis of vertebral fracture relies upon

accurate radiographic detection and a succinct, unambiguous radiographic report of fracture

(23). Yet in a single-center retrospective study of hospitalized elderly women, 50% of

radiographic reports failed to report the presence of moderate or severe vertebral fractures

12

and many patients (over 90%) remained untreated (3). There is other evidence that vertebral

fractures are under-reported (4-6).

Clinicians who interpret spine radiographs generally analyze radiographs of the thoraco-

lumbar spine in the lateral projection to identify vertebral fractures. Vertebral fractures

usually cause change in shape of the vertebrae, but not all vertebral deformities are due to

fractures. To differentiate fracture from deformity the interpreter takes into account not only

shape but also other features, such as the appearance of the endplate (24-28). The

interpretation can be aided by additional radiographic projections such as oblique views, or

by complementary examinations such as CT, MRI, or radionuclide scans (25). As with other

fractures, vertebral fractures have characteristic features that allow description and

classification, e.g. gradations in severity, the permanent nature of the deformity and the

possibility of a refracture at the same vertebral level from serial radiographs. However, there

is lack of standardization of radiologic assessment of vertebral fractures in routine clinical

practice, especially when attention is not focused specifically on the issue of fracture

identification. In this setting, the interpreting clinician often fails to recognize or report many

mild, and some moderate, fractures, or uses terminology that is non-specific and does not

adequately alert the referring clinician to the presence of a vertebral fracture and its

consequent importance in osteoporosis diagnosis and management.

Standardized Approach

Among the diagnostic protocols to diagnose vertebral fractures, the method proposed by

Genant et al (29) seems to be the most suitable for clinical applications, since the severity of

13

all vertebral fractures is assessed in a semi-quantitative fashion. The severity of a fracture is

assessed solely by visual determination of the extent of vertebral height reduction and

morphological change, and vertebral fractures are differentiated from other, non-fracture

deformities. The approximate degree of height reduction determines the assignment of grades

to each vertebra. Unlike the other approaches, the type of deformity (wedge, biconcavity or

compression) is no longer linked to the grading of a fracture in this approach.

Using the Genant et al (Fig 5) (29) semi-quantitative (SQ) method, thoracic and lumbar

vertebrae are graded on visual inspection of lateral spinal images and generally without direct

vertebral measurement as normal (grade 0) (Fig 6a); mildly deformed (grade 1:

approximately 20-25% reduction in anterior, middle, and/or posterior height and 10-20%

reduction of the projected vertebral area) (Fig 6b); moderately deformed (grade 2:

approximately 25-40% reduction in anterior, middle, and/or posterior height and 20-40%

reduction of the projected vertebral area) (Fig 6c); and severely deformed (grade 3:

approximately 40% or greater reduction in anterior, middle, and/or posterior height and in the

projected vertebral area) (Fig 6d). There is less consistency in diagnosis of mild (grade 1)

fractures, than with moderate (grade 2) and severe (grade 3) fractures (30, 31).

In addition to height reductions, careful attention is given to alterations in the shape and

configuration of the vertebrae relative to adjacent vertebrae and expected normal

appearances. These features add a strong qualitative aspect to the interpretation and also

render this method less readily definable as either qualitative or quantitative. Jiang et al (26)

have described an algorithm-based qualitative (ABQ) method in which the vertebral endplate

14

is scrutinized for features which are useful in the differentiation of fractures from other

causes of vertebral deformities.

Assessing the severity of the deformation as the reduction of vertebral height has the effect

(especially for the interpretation of incident fractures) that refractures of pre-existing

vertebral fractures can be assessed using the SQ method. This is an advantage of the SQ

method over the other standardized visual approaches, since it considers the continuous

nature of vertebral fractures and enables a meaningful interpretation of follow-up

radiographs.

Visual qualitative assessment of vertebral fractures using standardized grading schemes has

been found to be more reproducible than inspection of radiographs without specific criteria

for fracture diagnosis. Thus, standardized approaches have been found to be valid research

tools in epidemiological research and in clinical therapeutic trials. In contrast to the purely

morphometric analysis using digitization techniques, a visual assessment considers the

differential diagnosis of vertebral deformities. This is of great importance for the reliability

of prevalence and incidence data of vertebral fractures.

With respect to incident fractures a reader can, for example, adjust for magnification effects

or different centering of the X-ray beam, whereas these technical effects may actually have a

negative influence on assessments that are based solely on morphometric analysis of the

vertebral dimensions.

15

Ensuring the reliability of interpretation of incident vertebral fractures on serial radiographs

requires close attention to the radiographic procedure used. Serial radiographs of a patient

should always be viewed together in temporal order so as to accomplish a reliable analysis of

all new fractures.

The strength of standardized visual approaches is their use of a reader’s expertise in the

interpretation of vertebral deformities to differentiate fracture from non-fracture deformities,

or technical artifacts. However, this also constitutes their potential weakness, since there is

room for subjectivity in the interpretation. The reader’s training and experience are therefore

of utmost importance for valid use of standardized visual techniques; with trained,

experienced readers it has been shown that SQ grading of vertebral fractures can be applied

reliably (29, 32, 33).

5. FORTUITOUS DIAGNOSIS OF VERTEBRAL FRACTURES

Fortuitous diagnosis of vertebral fractures merits special attention. These vertebral fractures,

although frequently asymptomatic, still increase the risk of future vertebral and hip fractures.

They may also be used as an indication for further patient evaluation with DXA bone

densitometry and clinical investigation. In some cases patients may be candidates for

pharmacologic therapy to reduce future fracture risk, based on these fractures. For these

reasons it is important that reports from such diverse imaging studies as lateral chest

radiographs (Fig 7a), abdominal radiographs, barium studies, CT scans (Fig 7b), MRI

studies, and radionuclide scans include the presence of incidental vertebral fractures. With

16

multi-detector computed tomography (MDCT) of the thorax and abdomen, which are

examinations that are widely performed for various clinical indications, it is useful for

midline sagittal reformations to be obtained routinely, particularly in women over 65 and

men over 70 years. Vertebral fractures will be identified on the sagittal reformations which

are not evident on the transverse axial images and which may be clinically silent (34-37) (Fig

7b).

6. DIFFERENTIAL DIAGNOSIS

There are various normal variants, congenital abnormalities, degenerative changes, fractures

and other pathologies which can change the shape of vertebrae (Table 2). Differential

diagnosis of vertebral fractures includes all types of osteoporosis, both primary and

secondary, trauma, infection and malignancy (pathologic fractures).

Table 2. Differential diagnoses of changes in shape of vertebral bodies

Vertebral fractures Vertebral deformities Osteoporotic (low trauma)

Traumatic

Pathological (neoplastic, hemopoietic diseases and infections)

Developmental (short vertebral height, ‘butterfly’ vertebra and other abnormalities of spinal segmentation, ‘block’ vertebrae)

Normal variants (‘cupid’s bow’, anterior step deformity)

Scheuermann’s disease (juvenile osteochondritis)

Spondylosis (degenerative disc disease)

Metabolic (osteomalacia, Paget’s disease)

17

Deformities can be related to developmental anomalies (e.g. short vertebral height Fig 8a;

cupid’s bow Fig 8b; anterior step deformity), Scheuermann’s disease (Fig 8c); degenerative

disc disease (spondylosis Fig 8d) which changes the modeling of the vertebra and Schmorl’s

node (Fig 8e). The ABQ method suggests a pathway (Table 3) aimed at assisting the

interpreting clinician in differentiating fractures from deformities (26). This is accomplished

by careful scrutiny of the endplate; in vertebral fracture the endplate is irregular, with texture

change adjacent and below, due to micro-fractures, whereas in non-fracture deformities the

endplate is more distinct with characteristic abnormalities (Fig 8a-e). In congenital anomalies

the endplate is generally distinct (Fig 8a and b). In Scheuermann’s disease there is endplate

irregularity and slight wedging of several adjacent vertebrae, most commonly in the mid and

lower thoracic spine (Fig 8c). In spondylosis there is slight wedging and elongation (increase

in AP diameter of the vertebral body) with narrowing of the disc space and marginal

osteophytes (Fig 8d). Schmorl’s nodes, in which there is prolapse of disc material into the

vertebral body, there are bone defects with sclerotic margins in the anterior and posterior

aspect of the vertebra adjacent to the endplate (Fig 8e).

In pathological fractures related to such etiologies as bone metastases or myeloma (Fig 9a

and b) there is often cortical destruction and bulging into the spinal canal of the posterior

vertebral margin, and other imaging techniques such as MRI (Fig 9c) are required for

diagnosis (25).

18

Table 3. Algorith-based Qualitative (ABQ) Assessment of vertebral fracture (Drawn from reference 26)

Depression of endplate?

Short vertebral height?

Scheuermann’s, childhood fx, scoliosis, variants

Normal

Variants: anterior: step-like endplate in thoracic vertebrae, posterior : Cupid’s bow or balloon disc in lumbar vertebrae

Close to centre of endplate?

Concave depression?

Check for oblique projection or scoliosis

Whole endplate depressed within

ring?

Trauma, tumour, metabolic disease?

Osteoporotic fracture

Focused area: Schmorl’s node

Non-fracture deformity, developmental variant, non-osteoporotic fracture, other disease / condition

Start

YesNo

Yes

Yes

Yes

Yes

No

Yes

No

No

No

No

7. REPORTING

a) Clear and accurate terminology must be used

If there is a fracture, then the term ‘fracture’ must be used, and terms such as ‘collapse’ and

others should be avoided (23). It is relevant to give the grades (1-3; mild, moderate or

severe) (29) and number of fractures present, as the more severe the grade, and the more

vertebral fractures which are present, indicate that the patient is at higher risk of future

fracture (38). Even mild fractures must be clearly identified as they are determinants of

19

further fracture risk (39), and combined with BMD can enhance prediction of fracture risk

(40).

Other features to suggest spinal osteoporosis (prominent vertical trabecular striations, as the

horizontal trabecular are lost preferentially; thinned cortices, radiographic osteopenia) should

also be commented upon (8-10). The report should indicate if the fracture is osteoporotic,

traumatic or pathological in origin, and suggest further appropriate imaging, if relevant. If the

change in vertebral shape is not due to fracture, then the term ‘deformity’ should be used and

the cause of the deformity suggested (normal variant, congenital anomaly, Scheuermann’s

disease, spondylosis, Schmorl’s nodes, etc).

b) Importance of WHO FRAX® calculator

In 1994 the World Health Organization (WHO) defined osteoporosis in postmenopausal

women, in terms of bone mineral densitometry (BMD) as performed by DXA in the lumbar

spine (L1-L4), proximal femur and distal forearm. A T-score (standard deviation [SD] score

related to mean BMD of young [20-29 years] normal Caucasian women) equal to, or below, -

2.5 was defined as osteoporosis in postmenopausal women (41). However, this was never a

satisfactory BMD level for intervention as age is such a strong and independent determinant

of fracture. In 2008 the WHO published a tool (FRAX®) to calculate 10-year fracture risk for

individual patients aged between 40 and 90 years using clinical risk factors (with or without

femoral neck DXA BMD) (42,43). The clinical risk factors used in the calculator include

age, gender, height, weight, previous low trauma fracture over age 50, parental hip fracture,

oral glucocorticoid therapy (for more than 3 months at a dose 5mg daily or more),

rheumatoid arthritis, current smoking, alcohol consumption (more than 3 units per day) and

20

secondary causes of osteoporosis (including type I [insulin dependent] diabetes, osteogenesis

imperfecta in adults, untreated long-standing hyperthyroidism, hypogonadism or premature

menopause (<45 years), chronic malnutrition, or malabsorption and chronic liver disease).

Subsequently national guidelines for appropriate treatment interventions were launched in

several countries (44,45). The presence of vertebral fracture therefore influences the

calculation of fracture risk in the FRAX® tool and consequently affects management of

patients. However, FRAX® underestimates the risk when more than 1 vertebral fracture is

diagnosed; both severity and number of vertebral fractures are strong determinants of risk of

further fractures (46) Fracture risk prediction is also enhanced by combining vertebral

fracture status and BMD. Thus FRAX® may not be useful in patients with spinal osteoporosis

with multiple and severe vertebral fractures.

c) Suggest referral for central DXA

If a low trauma vertebral fracture is present on imaging studies then the report must include

advice to consider further relevant investigation (e.g. performing central DXA) and

management to referring clinicians. This should be compatible with national guidelines,

which may vary between countries.

This would ensure that patients found to have vertebral fractures on imaging studies are

further evaluated, resulting in reduction in the suffering related to osteoporotic fractures.

21

8. CONCLUSIONS

Vertebral fractures are the most common consequence of osteoporosis, occurring in a

substantial proportion of post-menopausal women and elderly men. Most vertebral fractures,

however, are not clinically recognized and can occur and accumulate silently. It is

established that the presence of a vertebral fracture is a strong risk factor for subsequent

osteoporotic fractures, and that patients with low BMD and vertebral fractures are at highest

risk. Large-scale clinical trials have demonstrated that osteoporosis therapies can increase

BMD (by 4-12%) and reduce fracture rates (vertebral fractures can be reduced by 40-70%)

(47,48). These benefits are most pronounced in patients with low BMD and vertebral

fractures. Clinical guidelines promulgated by the International Osteoporosis Foundation,

National Osteoporosis Foundation, and others around the world, recognize the importance of

vertebral fractures, along with BMD, as key risk factors for use in patient evaluation.

However, while BMD is widely used in patient evaluation, radiologic assessment of vertebral

fractures is not commonly performed, or if performed, is inadequately standardized and

interpreted. By understanding the clinical principles of osteoporosis diagnosis and

management provided in this document, by adopting the radiological guidelines for assessing

vertebral fractures provided herein and by clearly indicating “vertebral fracture” in the

patient’s report clinicians worldwide can contribute substantially to reducing the

consequences of osteoporosis.

22

REFERENCES 1. Esrud K, Thompson D, Cauley J, Nevitt M, Kado D, Hochberg M, Santora An,

Black D 2000 Prevalent vertebral deformities predict mortality and hospitalization

in older women with low bone mass. Fracture Intervention Trial Research Group. J

Am Geriatr Soc. 48:241-9

2. Melton LJ 3rd, Atkinson E, Cooper C, O'Fallon W, Riggs B 1999 Vertebral fractures

predict subsequent fractures. Osteoporos Int 10:214-21.

3. Gehlbach S, Bigelow C, Heimisdottir M, May S, Walker M, Kirkwood J 2000

Recognition of vertebral fracture in a clinical setting. Osteoporos Int 11:577-82.

4. Mui LW, Haramati LB, Alterman DD, Haramati N, Zelefsky MN, Hamerman D

2003 Evaluation of vertebral fractures on lateral chest radiographs of inner-city

postmenopausal women. Calcif Tissue Int 73:550-4

5. Kim N, Rowe BH, Raymond G, Jen H, Colman I, Jackson SA, Siminoski KG,

Chahal AM, Folk D, Majumdar SR 2004 Underreporting of vertebral fractures on

routine chest radiography. AJR Am J Roentgenol 182:297-300.

6. Delmas PD, van de Langerijt L, Watts NB, Eastell R, Genant H, Grauer A, Cahall

DL; IMPACT Study Group 2005 Underdiagnosis of vertebral fractures is a

worldwide problem: The IMPACT study. J Bone Miner Res;20(4):557-63

7. Roux C, Priol G, Fechtenbaum J, Cortet B, Liu Léage S, Audran M 2007 A clinical

tool to determine the necessity of spine radiography in postmenopausal women with

osteoporosis presenting with back pain Ann Rheum Dis;66:81–5

8. Quek ST, Peh WC Radiology of osteoporosis. Semin Musculoskelet Radiol

2002;6(3):197-206

23

9. Adams JE 2008 Osteoporosis. Chapter 75 In: Imaging of the Musculoskeletal System.

Eds: Pope TL, Bloem HL, Beltran J, Morrison WB, Wilson D. ISBN 978-1-4160-

2963-2 Saunders Elsevier, Philadelphia PA 2008 pp1489-508

10. Adams JE Osteoporosis. 2009 In: Imaging of Arthritis and Metabolic Bone Disease.

Editor Weissman B ISBN 978-0-323-04177-5 Saunders Elsevier, Philadelphia 2009:

pp 601-21

11. Clark’s Positioning in Radiography 2005 (12th edition) Whitley AS, Sloan C,

Hoadley G (Eds) Hodder Arnold, ISBN 0-340-76390-6

12. Banks LM, van Kuijk C, Genant HK. 1995 Radiographic technique for assessing

osteoporotic vertebral fracture. In Vertebral fracture in osteoporosis. Genant HK,

Jergas M, van Kuijk (Eds.) San Francisco, CA, University of California

Osteoporosis Research Group pp 131-47.

13. Curry TS, Dowdey JE, Murry RE 1990 Christensen’s Physics of Diagnostic

Radiology (4th edition) Lippincott Williams & Wilkins

14. Davies KM, Recker RR, Heaney RP 1989 Normal vertebral dimensions and normal

variation in serial measurements of vertebrae. J Bone Miner Res 4(3):341-9.

15. De Smet AA, Robinson RG, Johnson BE, Lukert BP 1988 Spinal compression

fractures in osteoporotic women: patterns and relationship to hyperkyphosis.

Radiology 166(2):497-500.

16. Gallagher JC, Hedlund LR, Stoner S, Meeger C 1988 Vertebral morphometry:

normative data. Bone Miner 4(2):189-96.

24

17. Ensrud KE, Nevitt MC, Palermo L, Cauley JA, Griffith JM, Genant HK, Black DM

1999 What proportion of incident morphometric vertebral fractures are clinically

diagnosed and vice versa? J Bone Miner Res 14(S1):S138.

18. Ettinger B, Cooper C 1995 Clinical Assessment of Osteoporotic Vertebral Fractures.

In: Genant HK, Jergas M, van Kuijk C (eds.) Vertebral Fracture in Osteoporosis.

Radiolgy Research and Education Foundation, San Francisco.

19. Ettinger B, Black DM, Nevitt MC, Rundle AC, Cauley JA, Cummings SR, Genant

HK 1992 Contribution of vertebral deformities to chronic back pain and disability.

The Study of Osteoporotic Fractures Research Group. J Bone Miner Res 7(4):449-

56.

20. Guglielmi G, Diacinti D, van Kuijk C, Aparisi F, Krestan C, Adams JE, Link TM

2008 Vertebral morphometry: current methods and recent advances. Eur

Radiol;18(7):1484-96

21. Jiang G, Ferrar L, Barrington NA, Eastell R 2007 Standardised quantitative

morphometry: a modified approach for quantitative identification of prevalent

vertebral deformities Osteoporosis Int;18(10):1411-9

22. Ferrar L, Jiang G, Adams J, Eastell R 2005 Identification of vertebral fractures: an

update. Osteoporos Int;16(7):717-28

23. Lenchik L, Rogers LF, Delmas PD, Genant HK 2004 Diagnosis of osteoporotic

vertebral fractures: importance of recognition and description by radiologists. AJR

Am J Roentgenol;183(4):949-58.

24. Guermazi A, Mohr A, Grigorian M, Taouli B, Genant HK 2002 Identification of

vertebral fractures in osteoporosis. Semin Musculoskelet Radiol;6(3):241-52

25

25. Link TM, Guglielmi G, van Kuijk C, Adams JE 2005 Radiologic assessment of

osteoporotic vertebral fractures: diagnostic and prognostic implications Eur

Radiol;15(8):1521-32

26. Jiang G, Eastell R, Barrington NA, Ferrar L. 2004 Comparison of methods for the

visual identification of prevalent vertebral fracture in osteoporosis. Osteoporos

Int;15(11):887-96

27. Genant HK, Jergas M. 2003 Assessment of prevalent and incident vertebral

fractures in osteoporosis research. Osteoporos Int;14 Suppl 3:S43-55

28. Grigoryan M, Guermazi A, Roemer FW, Delmas PD, Genant HK 2003 Recognizing

and reporting osteoporotic vertebral fractures. Eur Spine J;12 Suppl 2:S104-12.

29. Genant HK, Wu CY, van Kuijk C, Nevitt MC 1993 Vertebral fracture assessment

using a semiquantitative technique. J Bone Miner Res;8(9):1137-48

30. Genant HK, Jergas M, Palermo L, Nevitt M, Valentin RS, Black D, Cummings SR 1996

Comparison of semiquantitative visual and quantitative morphometric assessment of

prevalent and incident vertebral fractures in osteoporosis The Study of Osteoporotic

Fractures Research Group. J Bone Miner Res;11(7):984-96.

31. Ferrar L, Jiang G, Schousboe JT, DeBold CR, Eastell R. 2008 Algorithm-based

qualitative and semiquantitative identification of prevalent vertebral fracture: agreement

between different readers, imaging modalities, and diagnostic approaches. J Bone Miner

Res;23(3):417-24

32. Grados F, Fechtenbaum J, Flipon E, Kolta S, Roux C, Fardellone P 2009 Radiographic

methods for evaluating osteoporotic vertebral fractures. Joint Bone Spine. 76(3):241-7.

26

33. Genant HK, van Kuijk C, Jergas M 1995 Vertebral fracture assessment in

osteoporosis. Radiology Research and Education Foundation, San Francisco

34. Bauer JS, Müller D, Ambekar A, Dobritz M, Matsuura M, Eckstein F, Rummeny

EJ, Link TM 2006 Detection of osteoporotic vertebral fractures using multidetector

CT. Osteoporos Int;17(4):608-15

35. Müller D, Bauer JS, Zeile M, Rummeny EJ, Link TM. 2008 Significance of sagittal

reformations in routine thoracic and abdominal multislice CT studies for detecting

osteoporotic fractures and other spine abnormalities. Eur Radiol;18(8):1696-702

36. Woo EK, Mansoubi H, Alyas F 2008 Incidental vertebral fractures on multidetector

CT images of the chest: prevalence and recognition. Clin Radiol;63(2):160-4

37. Williams AL, Al-Busaidi A, Sparrow PJ, Adams JE, Whitehouse RW 2009. Under-

reporting of osteoporotic vertebral fractures on computed tomography. Eur J

Radiol;69(1):179-83

38. Delmas PD, Genant HK, Crans GG, Stock JL, Wong M, Siris E, Adachi JD. 2003

Severity of prevalent vertebral fractures and the risk of subsequent vertebral and

nonvertebral fractures: results from the MORE trial. Bone;33(4):522-32

39. Roux C, Fechtenbaum J, Kolta S, Briot K, Girard M. 2007 Mild prevalent and incident

vertebral fractures are risk factors for new fractures Osteoporos Int.;18(12):1617-24.

40. Siris ES, Genant HK, Laster AJ, Chen P, Misurski DA, Krege JH. 2007 Enhanced

prediction of fracture risk combining vertebral fracture status and BMD. Osteoporos

Int.;18(6):761-70

27

41. World Health Organisation Study Group 1994 Assessment of fracture risk and its

application to screening for postmenopausal osteoporosis. World Health Organisation,

Geneva, Switzerland (WHO Technical Report Series 843)

42. World Health Organisation (WHO) FRAXTM fracture risk calculator

http://www.shef.ac.uk accessed 14th February 2010

43. Kanis JA, Johansson H, Oden A, McCloskey EV. 2009 Assessment of fracture risk.

Eur J Radiol;71(3):392-7

44. Dawson-Hughes B, Tosteson AN, Melton LJ 3rd, Baim S, Favus MJ, Khosla S,

Lindsay RL; National Osteoporosis Foundation Guide Committee. 2008

Implications of absolute fracture risk assessment for osteoporosis practice guidelines

in the USA. Osteoporos Int;19(4):449-58

45. Compston J, Cooper A, Cooper C, Francis R, Kanis JA, Marsh D, McCloskey EV,

Reid DM, Selby P, Wilkins M; National Osteoporosis Guideline Group (NOGG).

2009 Guidelines for the diagnosis and management of osteoporosis in

postmenopausal women and men from the age of 50 years in the UK.

Maturitas;62(2):105-8

46. Kerkeni S, Kolta S, Fechtenbaum J, Roux C 2009 Spinal deformity index (SDI) is a

good predictor of incident vertebral fractures Osteoporos Int;20(9):1547-52.

47. Compston J. 2009 Clinical and therapeutic aspects of osteoporosis.

Eur J Radiol;71(3):388-91

48. Boonen S, Kay R, Cooper C, et al. 2009 Osteoporosis management: a perspective based

on bisphosphonate data from randomised clinical trials and observational databases. Int J

Clin Pract;63(12):1792-804

28

FIGURES: Figure 1: Good technique for spinal radiography: AP views a) thoracic spine: the levels from C7 to at least L1 need to be visualized; centering at T7, and field of view wide enough to include first (or cervical) and 12th ribs, to enable accurate counting of vertebral bodies b) lumbar spine: levels from at least T11 to sacrum need to be visualized; centering L3. Spine needs to be as flat as possible to the X-ray table, without rotation.

a) b)

29

Figure 2: Good technique for spinal radiography: lateral views a) thoracic spine: the arms are positioned so as to rotate the scapulae off the upper thoracic spine and centering at T7; the levels from at least T4 to L1 should be assessable b) lumbar spine: the levels from at least L11 to L4 should be assessable; centering is at L3. For both views it is essential that the spine is parallel to the X-ray film/table so as to avoid tilting of the vertebrae which causes apparently biconcave endplates.

a) b)

30

Figure 3: Technical problems: because of the divergent X-ray beam if centering is not ideal, the spine is not parallel to the X-ray table or if there is a spinal scoliosis then the vertebrae appear tilted and have apparently biconcave endplates as illustrated in lateral a) thoracic and b) lumbar spine radiographs, and which must not be erroneously diagnosed as vertebral fractures.

a) b)

31

Figure 4: Six point vertebral morphometry: this is a quantitative method of assessing vertebral shape by placing a) six points on the superior and inferior endplate at the front, mid and posterior margins. From these can be measured the b) anterior (A), middle (M) and posterior (P) heights and various ratios calculated. a)

b)

32

Figure 5: Semi-quantitative (SQ) assessment of vertebrae classifying them as normal or graded: 1 mild, 2 moderate and 3 severe vertebral fractures according to the degree of change in shape of the vertebra (Drawn from reference 29).

33

Figure 6: Vertebral fractures: a) normal appearance b) grade 1 mild upper endplate fracture c) grade 2 moderate vertebral fracture d) grade 3 severe vertebral fractures

a) b)

c) d)

34

Figure 7: Fortuitous diagnosis of vertebral fractures: a) vertebral fractures may be evident on lateral chest radiographs as evident in the lower thoracic spine; b) MDCT of the thorax and abdomen are frequently performed in radiology department. With routine reformatting of mid line sagittal images through the spine, particularly in women over 65 and men over 70 years of age, will demonstrate vertebral fractures which are not visible on transverse axial sections and which may be asymptomatic and so not clinically suspected (upper endplate grade 2 moderate fracture of L1).

a) b)

35

Figure 8: Vertebral deformities: etiologies other than vertebral fractures can change the shape of vertebrae. Developmental anomalies such as a) short vertebral height, as evident in two of these lower thoracic vertebrae; the endplate is crisp excluding fracture as cause; b) ‘cupid’s bow’ is a normal variant depicted by a smooth concavity in the posterior, inferior endplate as illustrated in the lumbar spine c) Sheuermann’s disease (juvenile osteochondritis) affecting several, adjacent thoracic vertebral endplates which are irregular, with slight wedging and elongation of the vertebral bodies d) spondylosis in which there has occurred remodeling of the vertebral body due to degenerative disc disease as evident by anterior marginal osteophytes e) Schmorl’s nodes in the endplates of T8 which may simulate fractures. These tend to occur in the anterior and posterior endplates and have sclerotic margins.

a) b)

c) d) e)

36

Figure 9: Pathological vertebral fractures: multiple myeloma: a) lateral thoracic spine radiograph and b) sagittal spine reformation from MDCT showing diffuse lytic areas with vertebral fractures and destruction of cortical margins, a sinister feature in vertebral fractures c) multiple bone metastases: T2 weighted sagittal MR scan showing heterogenous signal intensity of vertebral bodies and a pathological fracture of T11. The latter has posterior bulging of its posterior margin, another sinister feature in vertebral fracture.

a) b)

c)

37

10) APPENDICES

METHODS USED TO DEFINE CHANGE IN VERTEBRAL SHAPE

A) STANDARDIZED METHODS FOR FRACTURE ASSESSMENT

The first quantitative method to assess vertebral deformities was Fletcher’s “index of

wedging” in which normal variations in anterior heights were compared with posterior

heights (1). Barnett and Nordin used a “biconcavity index” in which the biconcavity of a

vertebra was measured as a quotient of the middle vertebral height and the anterior vertebral

height (2). This quotient was assessed from only one lumbar vertebra, and a value of less

than 0.8 was regarded as an indication of osteoporosis. Hurxthal was the first to describe in

detail the measurement of vertebral heights for the purpose of assessing anterior wedge

fractures (3).

Since then many epidemiologic studies and clinical trials have used various morphometric

methods to identify fractures (4-6). Melton’s method defined vertebral fractures using

percentage reductions in ratios of anterior, middle or posterior heights of vertebral bodies

compared with normal values for that particular vertebral body (5). Eastell et al. modified

this method, defining fractures on the basis of standard deviation reductions instead of fixed

percentages (7). McCloskey et al. modified Eastell/Melton methods by including the use of

predicted posterior heights (8). The Minne et al. method compared vertebral heights that have

been normalized for body size by dividing all values by the corresponding values of T4 and

comparing the results to values in healthy young women (6). Other groups have also used

different approaches to vertebral morphometry (9-11).

38

B. SEMI-QUANTITATIVE (SQ) METHOD

In addition to morphometric methods, semi-quantitative (SQ) methods for detecting vertebral

fracture have been used in various research settings. In these approaches there is the

assignment of numeric scores to vertebral fractures, or their assignment to distinct categories,

according to their shape or type and their severity, in a definable and reproducible manner

without making measurements of vertebral dimensions. Several SQ methods have been used.

The first standardized approach was proposed by Smith and colleagues (12). Like the

biconcavity index of Barnett and Nordin (2), the method grades only the vertebra with the

most severe deformity on the radiograph (12). In contrast to Smith et al., Meunier graded

each vertebra according to its shape (13). Grade 1 is assigned to a normal vertebra that has no

deformity; grade 2 to a biconcave vertebra; and grade 3 to an endplate fracture or a wedged

or crushed vertebra (14). Kleerekoper et al. modified Meunier’s method and introduced the

“vertebral deformity score” (VDS) (15). In the VDS each vertebra from T4 to L5 is assigned

an individual score from 0 to 3, depending on the type of deformity. This grading scheme is

based on the reduction of the anterior, middle, and posterior vertebral heights (ha, hm, and hp

respectively). A vertebral deformity (graded 1 to 3) is present when ha, hm, or hp is reduced

by at least 4 mm or 15% (15).

C. GENANT SEMI-QUANTITATIVE (SQ) METHOD

The strength of a Genant SQ method is that it makes use of the entire spectrum of visible

features that are helpful in identifying vertebral fractures (16,17). In addition to changes in

dimension, vertebral fractures are detected visually by the presence of endplate

39

abnormalities, lack of parallelism of the endplates, and general alterations in appearance

when compared with neighboring vertebrae (Fig. 5, 6b-d). These visual characteristics may

not be captured by the six digitized points used in morphometric methods (Fig. 4). Subtle

distinctions between a fractured endplate and deformity associated with Schmorl’s nodes can

be made visually only by an experienced observer (Fig. 8e). The same is true for the wedge-

shaped appearance caused by remodeling of the vertebral bodies in degenerative disc disease

(spondylosis) (Fig. 8d).

As with prevalent fractures, most incident fractures are easily identifiable on sequential

radiographs. The inevitable variation in positioning of patients and parallax effect of the

divergent X-ray beam may result in differences in point placement on follow-up radiographs.

This can result in the morphometric detection of an incident fracture that would be

interpreted visually as simply an alteration in radiographic projection. These sources of false-

positive or false-negative interpretations are particularly common when parallax problems

arise due to poor radiographic technique or improper patient positioning.

The reproducibility of Genant’s SQ method has been assessed in various studies (16,18-21).

In one study inter-observer agreement was 94% (kappa score 0.74) for the diagnosis of

prevalent fractures and 99% (kappa score 0.80) for the diagnosis of incident fractures (16). In

another study Li et al. (18) reported the inter-observer agreement was about 94% for the

dichotomous fracture/non-fracture diagnosis (the respective kappa scores were 0.80 to 0.81).

The agreement between the two readers using the whole grading scale to rate fractures was

90.6%, with a corresponding kappa score of 0.69. Wu et al. reported on the agreement of the

40

Genant SQ method for the assessment of incident vertebral fractures (21,22). Kappa scores

ranged from 0.80 to 0.84.

There are limitations of the Genant SQ method that may also apply to other standardized

approaches. For example, from morphometric data in normal subjects vertebrae in the mid-

thoracic spine and in the thoraco-lumbar junction are slightly more wedged than in other

regions of the spine (short vertebral height) (Fig. 8a) (8,23-25); consequently normal

variations may be misinterpreted as mild vertebral deformities (8,26,27). This may falsely

increase prevalence results for vertebral fractures from visual readings in the specific

anatomical regions. The same applies, to a lesser extent, in the lumbar spine, where some

degree of biconcavity is frequently seen (normal variants e.g. ‘cupid’s bow) (Fig 8b).

Accurate diagnosis of prevalent fractures, which requires that the reader distinguish between

normal variations and the degenerative changes resulting from true fractures, still depends on

the experience and training of the observer.

It has been argued that the diagnosis of mild vertebral fractures in particular may be quite

subjective, and that these fractures may be unrelated to osteoporosis (8). However, mild

grade 1 fractures detected with the SQ method are also associated with a lower BMD than

normal, and they also predict future vertebral fractures, although to a lesser extent than do

moderate (grade 2) or severe (grade 3) fractures (28).

For the diagnosis of incident fractures, other limitations may apply. Generally, incident

fractures are easily identified qualitatively on serial spinal radiographs, since a direct

41

comparison with baseline radiographs is possible. Using the Genant SQ method for the

assessment of incident fractures, however, the reader may sometimes feel that even though a

further height reduction is evident in a vertebra, it may not justify assigning a higher grade to

the incident fracture in comparison with the pre-existing prevalent fracture, since some

degree of settling or remodeling of the vertebral shape generally occurs following a fracture.

Therefore, in general, serial radiographs of a patient should be viewed together so that

incident fractures can be readily identified. Only those progressive changes that lead to a full

increase in deformity grade or an increase from a questionable deformity (grade 0.5) to a

definite fracture constitute designation as an incident fracture.

D. COMPARISON BETWEEN GENANT SQ METHOD AND MORPHOMETRIC

TECHNIQUES

Quantitative morphometric assessment of vertebral fracture was developed to obtain an

objective and reproducible measurement, using rigorously defined point placement and well-

defined algorithms for fracture definition. However, such an approach has some limitations.

In general, a substantial number of mild deformities detected by SQ method are missed by

morphometric methods. A significant number of false positives are found with morphometric

methods owing to the choice of point placement and threshold for defining vertebral

deformity. Although most moderate to severe fractures are detected by both techniques, only

SQ method can detect mild and subtle fractures, and appreciate anatomic, pathologic and

technical issues that influence the evaluation of fracture detection.

42

Leidig-Bruckner et al (29), compared Genant SQ method (16) with a morphometric approach

(6) and reported a good correlation (r=0.76) for baseline measurements and a moderate

correlation (r=0.57) for follow-up measurements. Li and colleagues compared Genant SQ

method (16,17) with vertebral morphometry for the detection of prevalent fracture (18).

Kappa scores for agreement with the consensus reading ranged from 0.84 to 0.87 for visual

interpretations, and from 0.54 to 0.75 for the morphometric approach. Wu et al. (21,22) also

compared Genant SQ method (16) with a morphometric approach for the detection of

incident fracture. There was only fair to moderate agreement between quantitative

morphometry and SQ method (the highest kappa score was 0.63). In a comprehensive study,

Black et al. (28) compared four different morphometric techniques (6-8, 24) in 3,013 spine

radiographs. In addition, Genant SQ method (16) was compared with the morphometric

approach in 502 cases. The agreement between the SQ method and the quantitative

approaches was moderate (kappa score of approximately 0.6). There was a high concordance

between quantitative morphometry and the SQ method for fractures defined as moderate or

severe by SQ method. There was, however, a significant discordance for fractures designated

mild in the SQ method.

E. ALGORITHM BASED QUALITATIVE (ABQ) ASSESSMENT

A structured algorithm-based qualitative method, with emphasis on scrutiny of the vertebral

endplates more than change is vertebral shape to differentiate between deformities and

fractures has been suggested (30,31). Recently, the ABQ method has been applied to research

cohorts (32-35). In one study ABQ, SQ, and morphometric methods of defining vertebral

fractures were compared (33). Among elderly men participating in the MrOs study the

43

prevalence of vertebral fracture ranged from 10% to 13%. Agreement between diagnostic

methods was moderate. Discordance related mainly to differential classification of mild

thoracic deformities or ABQ definition of vertebral fractures as traumatic and short vertebral

height identified by ABQ was common and not linked to low BMD. In another study (34)

using both radiographic and DXA vertebral fracture assessment (VFA) the prevalence of

radiographic vertebral fracture identified by ABQ and SQ was similar, but on VFA was 50%

higher for SQ. Mild ABQ vertebral fracture was associated with low BMD. Inter-observer

agreement for radiographic diagnosis was significantly better for ABQ than for SQ;

agreement between ABQ and SQ was moderate.

F) APPENDIX REFERENCES: 1. Fletcher H 1947 Anterior vertebral wedging-frequency and significance. Am J

Roentgenol 57(2):232-8.

2. Barnett E, Nordin BEC 1960 The radiological diagnosis of osteoporosis: a new approach.

Clinical Radiology 2:166-74.

3. Hurxthal LM 1968 Measurement of anterior vertebral compressions and biconcave

vertebrae. Am J Roentgenol Radium Ther Nucl Med 103(3):635-44.

4. Smith-Bindman R, Steiger P, Cummings SR, Genant HK 1991 The index of radiographic

area (IRA): a new approach to estimating the severity of vertebral deformity. Bone Miner

15(2):137-49.

5. Melton III LJ 1987 Epidemiology of vertebral fractures. In: Christensen C, Johansen JS,

Riis B (eds.) Osteoporosis, Copenhagen.

44

6. Minne HW, Leidig G, Wuster C, Siromachkostov L, Baldauf G, Bickel R, Sauer P, Lojen

M, Ziegler R 1988 A newly developed spine deformity index (SDI) to quantitate

vertebral crush fractures in patients with osteoporosis. Bone Miner 3(4): 335-49.

7. Eastell R, Cedel SL, Wahner HW, Riggs BL, Melton III LJ 1991 Classification of

vertebral fractures. J Bone Miner Res 6(3): 207-15.

8. McCloskey EV, Spector TD, Eyres KS, Fern ED, O’Rourke N, Vasikaran S, Kanis JA

1993 The assessment of vertebral deformity: a method for use in population studies and

clinical trials. Osteoporos Int 3(3):138-47.

9. Davies KM, Recker RR, Heaney RP 1989 Normal vertebral dimensions and normal

variation in serial measurements of vertebrae. J Bone Miner Res 4(3):341-9.

10. Hedlund LR, Gallagher JC 1988 Vertebral morphometry in diagnosis of spinal fractures.

Bone Miner 5(1):59-67.

11. Ross PD, Yhee YK, He YF, Davis JW, Kamimoto C, Epstein RS, Wasnich RD 1993 A

new method for vertebral fracture diagnosis. J Bone Miner Res 8(2):167-74.

12. Smith RW, Eyler WR, Mellinger RC 1960 On the incidence of senile osteoporosis. Ann

Int Med 52:773-81.

13. Meunier P 1968 La dynamique du remaniement osseux humain, étudiée par lecture

quantitative de la biopsie osseuse. Lyon.

14. Meunier PJ, Bressot C, Vignon E, Edouard C, Alexandre C, Courpron P, Laurent J 1978

Radiological and histological evolution of post-menopausal osteoporosis treated with

sodium fluoride-vitamin D-calcium. Preliminary results. In: Courvoisier B, Donath A,

Baud CA (eds.) Fluoride and Bone. Hans Huber Publishers, Bern, pp 263-76.

45

15. Kleerekoper M, Parfitt AM, Ellis BI 1984 Measurement of vertebral fracture rates in

osteoporosis. In: Christiansen C, Arnaud CD, Nordin BEC, Parfitt AM, Peck WA, Riggs

BL (eds.) Copenhagen International Symposium on Osteoporosis June 3-8, 1984, vol. 1.

Department of Clinical Chemistry, Glostrup Hospital, Copenhagen, pp 103-8.

16. Genant HK, Wu CY, van Kuijk C, Nevitt MC 1993 Vertebral fracture assessment using a

semiquantitative technique. J Bone Miner Res 8(9):1137-48.

17. van Kuijk C, Genant HK 1995 Radiology in osteoporosis. In: Riggs BL, Melton LJ (eds.)

Osteoporosis. Raven Press, New York.

18. Li J, Wu CY, Jergas M, Genant HK 1993 Comparison of semi-quantitative and

quantitative methods for assessment of vertebral fractures. In: Christiansen C (ed.) Fourth

International Symposium on Osteoporosis and Consensus Development Conference.

Gardiner-Caldwell, Hong Kong.

19. Cohen J 1960 A coefficient for agreement for nominal scales. Educ Psych Meas 10:37-

46.

20. Landis JR, Koch GG 1977 The measurement of observer agreement for categorical data.

Biometrics 33(1):159-74.

21. Wu CY, Li J, Jergas M, Genant HK 1994 Semi-quantitative and quantitative assessment

of incident fractures: comparison of methods (abstract). Journal of Bone and Mineral

Research 9(suppl 1):S157.

22. Wu CY, Li J, Jergas M, Genant HK 1995 Comparison of semi-quantitative and

quantitative techniques for the assessment of prevalent and incident vertebral fractures.

Osteoporos Int;5(5):354-70

46

23. Black DM, Cummings SR, Stone K, Hudes E, Palermo L, Steiger P 1991 A new

approach to defining normal vertebral dimensions. J Bone Miner Res 6(8):883-92.

24. Melton III LJ, Kan SH, Frye MA, Wahner HW, O’Fallon WM, Riggs BL 1989

Epidemiology of vertebral fractures in women. Am J Epidemiol 129(5):1000-11.

25. Ross PD, Davis JW, Epstein RS, Wasnich RD 1992 Ability of vertebral dimensions from

a single radiograph to identify fractures. Calcif Tissue Int 51(2):95-9.

26. Harrison JE, Patt N, Muller C, Bayley TA, Budden FH, Josse RG, Murray TM, Sturtridge

WC, Strauss A, Goodwin S 1990 Bone mineral mass associated with postmenopausal

vertebral deformities. Bone Miner 10(3):243-51.

27. Kleerekoper M, Nelson DA 1992 Vertebral fracture or vertebral deformity. Calcif Tissue

Int 50(1):5-6.

28. Black D, Palermo L, Nevitt MC, Genant HK, Epstein R, San Valentin R, Cummings SR,

and the Study of Osteoporotic Fractures Research Group 1995 Comparison of methods

for defining prevalent vertebral deformities: The study of osteoporotic fractures. J Bone

Miner Res 10(6):890-902.

29. Leidig-Bruckner G, Genant HK, Minne HW, Storm T, Thamsborg G, Bruckner T, Sauer

P, Schilling T, Soerensen OH, Ziegler R 1994 Comparison of a semiquantitative and a

quantitative method for assessing vertebral fractures in osteoporosis. Osteoporos Int

4(3):154-61.

30. Jiang G, Eastell R, Barrington NA, Ferrar L. 2004 Comparison of methods for the visual

identification of prevalent vertebral fracture in osteoporosis. Osteoporos Int;15(11):887-

96

47

31. Ferrar L, Jiang G, Adams J, Eastell R 2005 Identification of vertebral fractures: an

update. Osteoporos Int;16(7):717-28

32. Ferrar L, Jiang G, Armbrecht G, Reid DM, Roux C, Glüer CC, Felsenberg D, Eastell R.

2007 Is short vertebral height always an osteoporotic fracture? The Osteoporosis and

Ultrasound Study (OPUS).Bone;41(1):5-12.

33. Ferrar L, Jiang G, Cawthon PM, San Valentin R, Fullman R, Lambert L, Cummings SR,

Black DM, Orwoll E, Barrett-Connor E, Ensrud K, Fink HA, Eastell R 2007

Identification of vertebral fracture and non-osteoporotic short vertebral height in men: the

MrOS study. Osteoporotic Fractures in Men (MrOS) Study. J Bone Miner

Res;22(9):1434-41

34. Ferrar L, Jiang G, Clowes JA, Peel NF, Eastell R. 2008 Comparison of densitometric and

radiographic vertebral fracture assessment using the algorithm-based qualitative (ABQ)

method in postmenopausal women at low and high risk of fracture. J Bone Miner

Res;23(1):103-11

35. Ferrar L, Jiang G, Schousboe JT, DeBold CR, Eastell R.2008 Algorithm-based

qualitative and semiquantitative identification of prevalent vertebral fracture: agreement

between different readers, imaging modalities, and diagnostic approaches. J Bone Miner

Res;23(3):417-24

48

© International Osteoporosis Foundation