This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Quantifying the Extent of Femoral Head Involvement in Osteonecrosis

Sebastian F. Cherian, Alan Laorr, Khaled J. Saleh, Michael A. Kuskowski, Robert F. Bailey and Edward Y. Cheng
J Bone Joint Surg Am. 2003;85:309-315.

This information is current as of December 26, 2010

Supplementary Material http://www.ejbjs.org/cgi/content/full/85/2/309/DC1

Reprints and Permissions
Publisher Information
Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.
The Journal of Bone and Joint Surgery
20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org
 
COPYRIGHT © 2003 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

Quantifying the Extent of

Femoral Head Involvement
in Osteonecrosis
BY SEBASTIAN F. CHERIAN, MD, ALAN LAORR, MD, KHALED J. SALEH, MD, MSC, FRCSC,
MICHAEL A. KUSKOWSKI, PHD, ROBERT F. BAILEY, LPN, AND EDWARD Y. CHENG, MD
Investigation performed at the Department of Orthopaedic Surgery, University of Minnesota, Minneapolis, Minnesota

Background: There are numerous methods for quantifying the extent of osteonecrosis of the femoral head. However,
there is no consensus regarding which method is the most reliable. The purpose of this study was to determine the
reliability and prognostic accuracy of three commonly used methods for quantifying the extent of osteonecrosis of the
femoral head.
Methods: Thirty-nine hips in twenty-five patients who had stage-I or II osteonecrosis of the femoral head, according
to the grading system of the Association Research Circulation Osseous, were independently examined on two sepa-
rate occasions by three observers of different specialty backgrounds and experience. Each observer used three meth-
ods to quantify the extent of osteonecrosis of the femoral head: (1) the index of necrotic extent, (2) the modified
index of necrotic extent, and (3) the percentage of femoral head involvement. The interobserver and intraobserver
agreement was determined for each method, and the ability of each method to predict the time to subchondral col-
lapse was analyzed statistically.
Results: There was significantly valid agreement among the observers for all three methods (p < 0.001 for all
three). The correlation coefficients demonstrated substantial agreement among raters when they measured the in-
dex of necrotic extent and the percent involvement and nearly perfect agreement when they measured the modified
index of necrotic extent. Survivorship analysis revealed that the percent involvement (p < 0.05), index of necrotic ex-
tent (p < 0.007), and modified index of necrotic extent (p < 0.04) were prognostically significant predictors of sub-
chondral fracture.
Conclusions: Our results indicate that the index of necrotic extent, modified index of necrotic extent, and estimation
of the percentage of involvement of the femoral head are reproducible and reliable methods for quantitatively evaluat-
ing the extent of osteonecrosis of the femoral head. We believe that the three methods can be utilized with confi-
dence. Furthermore, they are clinically useful for identifying hips at greatest risk for subchondral collapse.

O
steonecrosis of the femoral head often occurs in young
active adults and frequently progresses to a subchon-
dral fracture of the femoral head, eventually leading to
complete joint destruction. Joint-preserving surgical interven-
tions generally have been more successful at earlier stages of
bone involvement than they have been after the occurrence of
a subchondral fracture. Therefore, the optimal time to evalu-
ate and treat osteonecrosis of the femoral head is before a sub-
chondral fracture has occurred1.
Various classification systems for staging osteonecrosis
of the femoral head, in order to assess the severity of the
disease and establish prognosis, have evolved over time. The
staging system of Steinberg et al. introduced the concept of
quantifying femoral head involvement as important in deter-
mining prognosis2,3. The international Association Research
Circulation Osseous (ARCO) incorporated the variables of lo-
cation and size of the necrotic area into the most recent stag-
ing system 4-6. While a subchondral fracture is the most
prognostically important variable indicating progression to
osteoarthritis, several studies have demonstrated that the size
of the necrotic lesion is important in determining whether a
subchondral fracture will occur3,7-16.
There are numerous published methods for quantifying
the extent of femoral head involvement by osteonecrosis of the
femoral head, including those that evaluate variables such as
necrotic volume or the surface area or arc of subchondral
bone affected. The plethora of methods in the literature is, in
part, due to the lack of a rigorous statistical evaluation of the
reliability and predictive ability of these techniques. Some are
less practical clinically, as they require specialized software to
analyze a digital image from either radiographs or magnetic
resonance imaging in two planes. At the current time, there is
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003
no consensus regarding which method is most accurate or
prognostically valuable3,11,17,18. The interobserver and intraob-
server reliability of even technically complex methods, as well
as of simpler methods, remains unclear. The purpose of this
study was to determine the reliability and prognostic ability of
three commonly used methods for quantifying the extent of
osteonecrosis of the femoral head.
Materials and Methods
atients with ARCO stage-I or II osteonecrosis of the fem-
P oral head were identified from an institutional review
board-approved, randomized, prospective trial of the surgical
treatment of osteonecrosis of the femoral head. All of the pa-
tients in the surgical trial were eligible for this study; however,
only those for whom a magnetic resonance image was avail-
able were included. Thirty-nine hips in twenty-five patients
were included in the present study. There were ten women
and fifteen men whose ages ranged from twenty to sixty
years (mean, thirty-nine years). The etiology of the osteone-
crosis of the femoral head was steroid-related in twenty-nine
hips (eighteen patients), ethanol-related in four hips (three
patients), and idiopathic in six hips (four patients). The size of
the lesion was estimated exclusively with the use of magnetic
resonance imaging, which is considered the most accurate im-
aging modality for this purpose3,19-24.
Magnetic Resonance Imaging Technique
The magnetic resonance imaging was performed with two
1.5-T superconducting magnets (Vision; Siemens, Erlangen,
Germany). Patients were imaged in the supine position, with
use of a phased array coil. T1-weighted magnetic resonance
images with use of spin-echo pulse sequences were made in
the coronal and sagittal planes of the femoral heads. T1-
weighted imaging in the coronal plane was performed with a
repetition time of 650 msec, an echo time of 15 msec, a 90°
flip angle, a 333 × 515 matrix, a 4-mm section thickness, a
0.2-mm intersection gap, and a square field of view of 320 ×
320 mm. T1-weighted imaging in the sagittal plane was per-
formed with a repetition time of 750 msec, an echo time of 15
msec, a 90° flip angle, a 219 × 512 matrix, a 3-mm section
thickness, a 0.2-mm intersection gap, and a square field of
view of 320 × 320 mm.
Reviewers of the Magnetic Resonance Images
The three raters in this study were a third-year radiology resi-
dent, an attending staff musculoskeletal radiologist, and an
attending staff orthopaedic surgeon with a subspecialty inter-
est in adult reconstructive surgery. The observers were blinded
to both their previous evaluations and the evaluations of the
other observers. Prior to the study, the three evaluators re-
viewed pertinent literature describing the specific methods,
and they discussed these methods together to clarify the evalu-
ation techniques before they performed the ratings. A study
coordinator who was not a rater and who had no knowledge
of prior evaluations recorded the data. The three raters exam-
ined each magnetic resonance image on two separate occa-
QUANTIFYING THE EXTENT OF FE M O R A L HE A D
I NVO L VE M E N T IN O S T E O N E C RO S I S
sions. The time interval between the interpretations was one
month for two raters and ten days for the third. The images
were randomly assembled and were shuffled between the first
and second interpretations.
Calculation of the Extent of Femoral Head
Involvement by Interpretation of the
Magnetic Resonance Images
The first method that was used to estimate the extent of the
osteonecrosis of the femoral head was estimation of the per-
centage of head involvement, which was first described in
198425. The extent of osteonecrosis of the femoral head was
defined by the abnormal signal on T1-weighted images. Spe-
cifically, a low-signal line on T1-weighted images formed the
border of the necrotic lesion. Osteonecrosis generally has a
more well-defined and distinct border than does bone edema,
which has a diffuse low signal. The necrotic area within the
femoral head was visually estimated on the basis of serial
coronal and sagittal images; it was not specifically measured
with use of point-counting or computer techniques. The esti-
mated values were grouped into three categories, according to
the estimated percentage of the area involved compared with
the area of the entire femoral head, as <15%, 15% to 30%,
and >30%.
The second method for quantifying the extent of osteo-
necrosis of the femoral head was the index of necrotic extent
developed by Koo and Kim in 199517. This index is determined
by measuring the angle created by the extent of subchondral
involvement as a proxy for the lesion size. The rater first deter-
mined which images represented the midsagittal and mid-
coronal slices and then assessed the abnormal T1 signal within
the femoral head on these slices. The lesion size was estimated
by measuring the angle of the arc of the necrotic segment from
the center of the femoral head, which can be defined as the ne-
crotic arc angle18. The necrotic arc angles on the midcoronal
and midsagittal images were designated A and B, respectively.
The index of necrotic extent was calculated as (A/180) × (B/
180) × 100 (Figs. 1-A and 1-B).
The third method was a variant of the index of necrotic
extent, as modified by one of us (E.Y.C.). The modification
was undertaken to account for a deficiency of the Koo method,
with which the osteonecrotic lesion is measured on the mid-
sagittal and midcoronal images. This tends to underestimate
the true extent of the osteonecrosis of the femoral head be-
cause the lesions usually are not centered in the superior, cen-
tral portion of the head; instead they are concentrated in the
anterior-superior segment of the head. The modified index of
necrotic extent is calculated with use of the same formula as is
used for the original index, but the necrotic arc angle is mea-
sured on the image that demonstrates the maximal lesion size
in the sagittal (A) and coronal (B) planes rather than on the
midcoronal and midsagittal images. This method has not been
reported in literature.
In order to correlate and assess the prognostic ability of
these three methods, we analyzed the relationships between
the measurements made with each method and the time be-
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG QUANTIFYING THE EXTENT OF FE M O R A L HE A D
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003 I NVO L VE M E N T IN O S T E O N E C RO S I S

tween the date of diagnosis (defined by the date of the first regard to their ability to predict the time to progression.
magnetic resonance imaging examination) and the date of
radiographic progression (defined by the date when a sub- Results
chondral fracture was first noticed on a radiograph). The ra- Reliability and Validity
diographs were obtained serially every three months. he interobserver agreement of the index of necrotic extent,

Statistical Analysis
The interobserver reliability of each of the three methods was
examined by determining the agreement among the three ob-
servers, and the intraobserver reliability was evaluated by
determining the agreement between the same observer’s inter-
pretations made at different times. Intraclass correlation coef-
ficients were generated for evaluation of the index of necrotic
extent and the modified index of necrotic extent, and Kendall
coefficients of concordance were generated for evaluation of
the percent involvement.
The log rank test was used to compare Kaplan-Meier sur-
vival curves for the time from diagnosis (the date of the magnetic
resonance imaging) to radiographic progression (subchondral
collapse) for separate groupings based on percent involvement
(<15% [mild], 15% to 30% [moderate], or >30% [severe]),
index of necrotic extent (<40 or >40), and modified index of
necrotic extent (<40 or >40). Cox proportional hazards re-
gression was used to compare the percent involvement, index
of necrotic extent, and modified index of necrotic extent with
T as defined by intraclass correlation coefficients, was 0.58
(p < 0.001) for the first observation and 0.70 (p < 0.001) for the
second observation. The interobserver agreement of the modi-
fied index of necrotic extent, as defined by intraclass correla-
tion coefficients, was 0.63 (p < 0.001) for the first observation
and 0.81 (p < 0.001) for the second observation. The interob-
server agreement of the percent involvement, as defined by
Kendall coefficients of concordance, was 0.71 (p < 0.001) for
the first observation and 0.79 (p < 0.001) for the second obser-
vation. These results are summarized in the Appendix.
The intraobserver agreement of the index of necrotic ex-
tent, as defined by intraclass correlation coefficients, was 0.91
(p < 0.005) for the first rater, 0.83 (p < 0.005) for the second
rater, and 0.93 (p < 0.005) for the third rater. The intraob-
server agreement of the modified index of necrotic extent, as
defined by intraclass correlation coefficients, was 0.65 (p <
0.005) for the first rater, 0.88 (p < 0.005) for the second rater,
and 0.91 (p < 0.005) for the third rater. The intraobserver
agreement of the percent involvement, as defined by Kendall
coefficients of concordance, was 0.90 (p < 0.005) for the first

Fig. 1-A

Fig. 1-B

Measurement of the index of necrotic extent and the modified index of necrotic extent on coronal (Fig. 1-A) and sagittal (Fig. 1-B)
T1-weighted images of a femoral head. A is the necrotic arc angle in the coronal plane, and B is the necrotic arc angle in the sagittal plane.
The index of necrotic extent ([A/180] × [B/180]) × 100) is derived from the size of subchondral involvement at the midcoronal and mid-
sagittal planes, and the modified index of necrotic extent is derived from the maximum size of subchondral involvement in both planes.
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003
Fig. 2
Survivorship curve of elapsed time from the diagnosis to subchondral
collapse stratified by percentage of femoral head involvement (<15%
[mild], 15% to 30% [moderate], or >30% [severe]).
Fig. 3
Survivorship curve of elapsed time from the diagnosis to subchondral
collapse stratified by the index of necrotic extent (<40 or >40).
rater, 0.89 (p < 0.005) for the second rater, and 0.88 (p <
0.005) for the third rater (see Appendix).
Prognostic Ability
Log rank tests revealed that the time to subchondral collapse
was significantly related to the percent involvement (log rank =
5.81, p < 0.05; Fig. 2), index of necrotic extent (log rank = 7.17,
p < 0.007; Fig. 3), and modified index of necrotic extent (log
rank = 4.05, p < 0.04; Fig. 4).
When the Cox proportional hazards regression was used
to compare the percent involvement, index of necrotic extent,
QUANTIFYING THE EXTENT OF FE M O R A L HE A D
I NVO L VE M E N T IN O S T E O N E C RO S I S
and modified index of necrotic extent as predictors of time
from diagnosis to subchondral collapse (with the index of ne-
crotic extent and the modified index of necrotic extent aver-
aged across the three raters for each patient), the index of
necrotic extent was found to be a significant predictor of time
to subchondral collapse (relative risk = 1.06, 95% confidence
interval = 1.01 to 1.12). The percent involvement (p = 0.07)
and the modified index of necrotic extent (p = 0.12) showed a
trend toward significant predictive power.
Discussion
agnetic resonance imaging can detect and characterize
M osteonecrotic segments of the femoral head before such
abnormalities are evident on plain radiography, scintigraphy,
or computerized tomography19-24,26-28. There are methods for
calculating the true volume or surface area of necrosis with
use of image analysis software11,18,29,30, but although they may
accurately measure necrotic volume they are probably too
complicated for everyday clinical application. We evaluated
the reliability of three methods for quantifying the extent of
femoral head involvement that are simple and practical enough
to be used in routine clinical practice.
In 1984, Steinberg et al. included the size of the le-
sion in their system for staging osteonecrosis of the femoral
head25. Later, ARCO proposed quantifying the extent of os-
teonecrosis of the femoral head by categorizing the size of
the necrotic involvement into three groups4-6. In 1995, Koo
and Kim described a method for predicting subchondral
collapse by quantifying necrotic involvementthat is, by
measuring an index of the necrotic extent of the diseased
femoral head17. The Koo method is similar to the surface arc
method developed in 1972 by Kerboul et al. for evaluation
of osteonecrosis of the femoral head on radiographs31.
Fig. 4
The survivorship curve of elapsed time from the diagnosis to sub-
chondral collapse stratified by the modified index of necrotic extent
(<40 or >40).
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003
Both the ARCO and the Koo indexes are accepted meth-
ods for predicting subchondral collapse; however, until now
their intraobserver and interobserver variability was un-
known3-6,17,18. It has been reported that the interobserver repro-
ducibility of the Koo method may not be adequate because of
inconsistencies among observers with regard to estimating
which are the midsagittal and midcoronal images18. To address
this potential pitfall, we modified the Koo method by measur-
ing the necrotic arc angles on the images that demonstrated
the area of maximal involvement in the sagittal (A) and coro-
nal (B) planes, as opposed to measuring them on the midsag-
ittal and midcoronal images. While no previous investigator
has described measuring the maximum necrotic arc angles in
both the coronal and the sagittal plane, we showed that this
method better estimates the true extent of osteonecrosis of the
femoral head.
There was significant agreement (p < 0.001) among the
observers for all three of the methods that we evaluated (index
of necrotic extent, modified index of necrotic extent, and per-
cent involvement). The correlation coefficients demonstrated
substantial agreement among raters when they measured the
index of necrotic extent and nearly perfect agreement when
they measured the modified index of necrotic extent. Epide-
miologists consider interobserver agreement of 0.6 to 0.8 to be
substantial and 0.8 to 1.0 to indicate nearly perfect agree-
ment32. With the sample size used in this study (thirty-eight
hips), a correlation coefficient of 0.44 could be detected as
significant at the p = 0.05 level with 80% power. Since all ob-
served correlations far exceeded that value, this study had
substantial statistical power to detect significant effects. No
statistical test was performed to determine the difference be-
tween the original and modified indexes of necrotic extent;
nonetheless, it is evident that the reliability of the modified
index is greater than that of the original index. The Kendall
coefficient of concordance indicated substantial agreement
for the method of estimating the percentage of femoral head
involvement; however, this is a noncontinuous categorical
variable. The threshold for establishing agreement among
raters for a noncontinuous categorical variablee.g., one or
two groupingsis much lower than that for establishing
agreement among raters for a continuous noncategorical
variable, in which there is an infinite number of groupings.
Therefore, it is more meaningful that agreement was estab-
lished among the raters for the index of necrotic extent and
the modified index of necrotic extent methods than for the
percent involvement.
All three observers had nearly perfect intraobserver
agreement for the index of necrotic extent and for the percent
involvement. Two of the three observers had nearly perfect in-
traobserver agreement and the third had substantial intraob-
server agreement for the modified index of necrotic extent.
These findings were to be expected as the interobserver varia-
tion provides an upper limit; it is always greater than intraob-
server variation32. For this reason, interobserver variation is
the key variable in an assessment of reliability. Our statistical
evaluation supports the finding that these methods can be
QUANTIFYING THE EXTENT OF FE M O R A L HE A D
I NVO L VE M E N T IN O S T E O N E C RO S I S
used reproducibly by a single rater or by multiple raters.
Interobserver agreement improved on the second eval-
uation of the images of each hip. This finding suggests that
reliability improves as observers become more familiar with
the specific methods. This was to be expected, although the
“learning curve” for these methods is very short. The agree-
ment among observers with different types of training im-
proved after only two evaluation sessions. This suggests that
the methods do not require subspecialty training and can be
successfully adopted within the general radiologic and ortho-
paedic communities.
These results refute earlier work by Kim et al.18, who, in
1998, indicated that none of the methods commonly employed
to evaluate osteonecrosis of the femoral head were reliable for
universal usage. Kim et al. found the reliability of the Koo in-
dex of necrotic extent to be unacceptable. The reasons for the
disparity between our findings and those of Kim et al. are not
clear. However, Kim used a much smaller sample (six hips) as
well as different statistical analysis (coefficient of variation).
The clinical utility of these methods is demonstrated by
their ability to predict disease outcome. We showed that the
three methods are capable of predicting which patients are at
higher risk for subchondral collapse (Figs. 2, 3, and 4). The in-
dex of necrotic extent had the most power to predict disease
progression, but the modified index of necrotic extent was the
most reliable of the methods. We believe that, from a clinical
standpoint, either method has clinical utility and can be used
by treating physicians to identify patients who are at the high-
est risk for disease progression. In addition, comparison of
different treatment interventions will be more equitable if this
disease stratification is taken into account.
Although our study demonstrated significant findings, it
had limitations. The sample size is larger than those of other
studies, but only three observers were used and each performed
two evaluations. The configuration of some lesions does not
Fig. 5
T1-weighted coronal image of a femoral head, demonstrating the
patchy and diffuse pattern of some osteonecrotic lesions that are diffi-
cult to evaluate by measuring the index of necrotic extent.
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003
lend itself readily to an arc measurement. Such measurement is
usually difficult when there is either extensive femoral head in-
volvement or patchy areas of disease with several independent
subchondral locations (Fig. 5). When the raters encountered
those situations, they attempted to combine all of the subchon-
dral locations to estimate the total area of involvement. Those
situations are less frequent than the more common crescent
configuration seen in Figures 1-A and 1-B.
We conclude that both the original and the modified in-
dexes of necrotic extent are reliable, reproducible measure-
ments that accurately quantitate the extent of femoral head
involvement by osteonecrosis; we also conclude that they are
clinically useful for identifying hips at greatest risk for disease
progression. These indexes should be incorporated into stag-
ing criteria in order to provide a more valid means of compar-
ing the outcomes of different treatments, to more accurately
establish prognosis, and to help to ascertain the optimal treat-
ment for this diverse group of patients.
Appendix
Tables showing the interobserver and intraobserver
agreement of the three methods for quantifying the extent
of osteonecrosis of the femoral head are available with the elec-
tronic versions of this article, on our web site at www.jbjs.org
(go to the article citation and click on “Supplementary Mate-
rial”) and on our quarterly CD-ROM (call our subscription
department, at 781-449-9780, to order the CD-ROM).
References
1. Ficat RP. Idiopathic bone necrosis of the femoral head. Early diagnosis and
treatment. J Bone Joint Surg Br. 1985;67:3-9.
2. Steinberg ME, Brighton CT, Steinberg DR, Tooze SE, Hayken GD. Treat-
ment of avascular necrosis of the femoral head by a combination of bone
grafting, decompression, and electrical stimulation. Clin Orthop. 1984;
186:137-53.
3. Steinberg ME, Hayken GD, Steinberg DR. A quantitative system for staging
avascular necrosis. J Bone Joint Surg Br. 1995;77:34-41.
4. Gardeniers JWM. ARCO committee on terminology and staging (report on the
committee meeting at Santiago De Compostela). ARCO Newsletter. 1993;
5:79-82.
5. Gardeniers JWM, ARCO Committee on Terminology and Staging. The ARCO
perspective for reaching one uniform staging system of osteonecrosis. In:
Schoutens A, Arlet J, Gardeniers JWM, Hughes SPF, editors. Bone circulation
and vascularization in normal and pathological conditions. New York: Plenum
Press; 1993. p 375-80.
6. Gardeniers JWM. ARCO committee on terminology and staging (report from
the Nijmegen meeting). ARCO Newsletter. 1991;3:153-9.
7. Shimizu K, Moriya H, Akita T, Sakamoto M, Suguro T. Prediction of collapse
with magnetic resonance imaging of avascular necrosis of the femoral head.
J Bone Joint Surg Am. 1994;76:215-23.
8. Takatori Y, Kokubo T, Ninomiya S, Nakamura S, Morimoto S, Kusaba I. Avas-
cular necrosis of the femoral head. Natural history and magnetic resonance
imaging. J Bone Joint Surg Br. 1993;75:217-21.
9. Beltran J, Knight CT, Zuelzer WA, Morgan JP, Shwendeman LJ, Chandnani
VP, Mosure JC, Shaffer PB. Core decompression for avascular necrosis of
the femoral head: correlation between long-term results and preoperative MR
staging. Radiology. 1990;175:533-6.
10. Lafforgue P, Dahan E, Chagaud C, Schiano A, Kasbarian M, Acquaviva P-C.
Early-stage avascular necrosis of the femoral head: MR imaging for prognosis
in 31 cases with at least 2 years of follow-up. Radiology. 1993;187:
199-204.
11. Steinberg ME, Bands RE, Parry S, Hoffman E, Chan T, Hartman KM. Does
QUANTIFYING THE EXTENT OF FE M O R A L HE A D
I NVO L VE M E N T IN O S T E O N E C RO S I S
Sebastian F. Cherian, MD
Department of Radiology, University of Minnesota, 420 Delaware
Street S.E., Minneapolis, MN 55455
Alan Laorr, MD
Suburban Radiologic Consultants, 4801 West 81st Street, Suite 108,
Minneapolis, MN 55437
Khaled J. Saleh, MD, MSc, FRCSC
Edward Y. Cheng, MD
Department of Orthopaedic Surgery, University of Minnesota, 420
Delaware Street S.E., MMC 492, Minneapolis, MN 55455. E-mail address
for E.Y. Cheng: cheng002@tc.umn.edu
Michael A. Kuskowski, PhD
Veterans Affairs Medical Center, GRECC (11G), One Veterans Drive,
Minneapolis, MN 55417
Robert F. Bailey, LPN
Department of Surgery, University of Minnesota, 420 Delaware Street
S.E., Box 195, Minneapolis, MN 55455
In support of their research or preparation of this manuscript, one or
more of the authors received grants or outside funding from the National
Institutes of Health (Grant 2P01DK13083-300219) and the Minnesota
Medical Foundation. None of the authors received payments or other
benefits or a commitment or agreement to provide such benefits from a
commercial entity. No commercial entity paid or directed, or agreed to
pay or direct, any benefits to any research fund, foundation, or educa-
tional institution, or other charitable or nonprofit organization with
which the authors are affiliated or associated.
lesion size affect the outcome in avascular necrosis? Clin Orthop. 1999;
367:262-71.
12. Kopecky KK, Braunstein EM, Brandt KD, Filo RS, Leapman SB, Capello WN,
Klatte EC. Apparent avascular necrosis of the hip: appearance and sponta-
neous resolution of MR findings in renal allograft recipients. Radiology.
1991;179:523-7.
13. Chan TW, Dalinka MK, Steinberg ME, Kressel HY. MRI appearance of femo-
ral head osteonecrosis following core decompression and bone grafting.
Skeletal Radiol. 1991;20:103-7.
14. Sugano N, Ohzono K, Masuhara K, Takaoka K, Ono K. Prognostication of
osteonecrosis of the femoral head in patients with systemic lupus erythema-
tosus by magnetic resonance imaging. Clin Orthop. 1994;305:190-9.
15. Holman AJ, Gardner GC, Richardson ML, Simkin PA. Quantitative MRI pre-
dicts clinical outcome of core decompression for osteonecrosis of the femo-
ral head. J Rheumatol. 1995;22:1929-33.
16. Ohzono K, Saito M, Takaoka K, Ono K, Saito S, Nishina T, Kadowaki T. Natu-
ral history of atraumatic avascular necrosis of the femoral head. J Bone Joint
Surg Br. 1991;73:68-72.
17. Koo K-H, Kim R. Quantifying the extent of osteonecrosis of the femoral head.
A new method using MRI. J Bone Joint Surg Br. 1995;77:875-80.
18. Kim YM, Ahn JH, Kang HS, Kim HJ. Estimation of the extent of osteo-
necrosis of the femoral head using MRI. J Bone Joint Surg Br. 1998;80:
954-8.
19. Mitchell MD, Kundel HL, Steinberg ME, Kressel HY, Alavi A, Axel L. Avascu-
lar necrosis of the hip: comparison of MR, CT, and scintigraphy. AJR. Am J
Roentgenol. 1986;147:67-71.
20. Mitchell DG, Rao VM, Dalinka MK, Spritzer CE, Alavi A, Steinberg ME,
Fallon M, Kressel HY. Femoral head avascular necrosis: correlation of MR
imaging, radiographic staging, radionuclide imaging, and clinical findings.
Radiology. 1987;162:709-15.
21. Coleman BG, Kressel HY, Dalinka MK, Scheibler ML, Burk DL, Cohen EK.
Radiographically negative avascular necrosis: detection with MR imaging.
Radiology. 1988;168:525-8.
 
THE JOUR NAL OF BONE & JOINT SURGER Y · JBJS.ORG
VO L U M E 85-A · N U M B E R 2 · F E B R U A R Y 2003
22. Seiler JG 3rd, Christie MJ, Homra L. Correlation of the findings of magnetic
resonance imaging with those of bone biopsy in patients who have stage-I
or II ischemic necrosis of the femoral head. J Bone Joint Surg Am. 1989;
71:28-32.
23. Hauzeur JP, Pasteels JL, Schoutens A, Hinsenkamp M, Appelboom T, Cho-
chrad I, Perlmutter N. The diagnostic value of magnetic resonance imaging
in non-traumatic osteonecrosis of the femoral head. J Bone Joint Surg Am.
1989;71:641-9.
24. Markisz JA, Knowles RJ, Altchek DW, Schneider R, Whalen JP, Cahill PT.
Segmental patterns of avascular necrosis of the femoral heads: early detec-
tion with MR imaging. Radiology. 1987;162:717-20.
25. Steinberg ME, Hayken GD, Steinberg DR. A new method for evaluation and
staging of avascular necrosis of the femoral head. In: Arlet J, Ficat RP, Hun-
gerford DS, editors. Bone circulation. Baltimore: Williams and Wilkins; 1984.
p 398-403.
26. Bassett LW, Gold RH, Reicher M, Bennett LR, Tooke SM. Magnetic reso-
nance imaging in the early diagnosis of ischemic necrosis of the femoral
head. Preliminary results. Clin Orthop. 1987;214:237-48.
QUANTIFYING THE EXTENT OF FE M O R A L HE A D
I NVO L VE M E N T IN O S T E O N E C RO S I S
27. Beltran J, Herman LJ, Burk JM, Zuelzer WA, Clark RN, Lucas JG, Weiss LD,
Yang A. Femoral head avascular necrosis: MR imaging with clinical-pathologic
and radionuclide correlation. Radiology. 1988;166:215-20.
28. Cheng EY. The problem hip. Osteonecrosis of the femoral head: the North
American perspective. In: Bulstrode C, Buckwalter J, Carr A, Marsh L, Fair-
bank J, Wilson-MacDonald J, editors. Oxford textbook of orthopedics and
trauma. Oxford: Oxford University Press; 2002, p 9738-810.
29. Theodorou DJ, Konstantinos NM, Beris AE, Theodorou SJ, Soucacos PN.
Multimodal imaging quantitation of the lesion size in osteonecrosis of the
femoral head. Clin Orthop. 2001;386:54-63.
30. Hernigou P, Lambotte JC. Volumetric analysis of osteonecrosis of the femur.
Anatomical correlation using MRI. J Bone Joint Surg Br. 2001;83:672-5.
31. Kerboul M, Thomine J, Postel M, Merle d’Aubigné R. The conservative surgi-
cal treatment of idiopathic aseptic necrosis of the femoral head. J Bone Joint
Surg Br. 1974;56:291-6.
32. Norman GR, Streiner DL. Biostatistics: the bare essentials. St. Louis: Mosby;
1994.