DOI:10.2214/AJR.07.2236
AJR 2007; 189:1165-1168
© American Roentgen Ray Society
MRI Quantitative Morphologic Analysis of Patellofemoral Region: Lack of Correlation with Chondromalacia Patellae at Surgery
Yoshimi Endo1,2,
Mark E. Schweitzer1,
Marcelo Bordalo-Rodrigues1,3,
Andrew S. Rokito4 and
James S. Babb5
1 Department of Radiology, New York University, Hospital for Joint Diseases
Orthopaedic Institute, New York, NY.
2 Present address: Department of Radiology, State University of New York
Downstate Medical Center, 450 Clarkson Ave., Brooklyn, NY 11203.
3 Present address: Radiology Department, Hospital das Clinicas, University of
São Paulo School of Medicine, São Paulo, Brazil.
4 Department of Orthopaedic Surgery, New York University, Hospital for Joint
Diseases Orthopaedic Institute, New York, NY.
5 Department of Radiology, New York University School of Medicine, New York,
NY.
Received March 13, 2007;
accepted after revision May 19, 2007.
Address correspondence to Y. Endo
(yoshimi.endo{at}downstate.edu,
yoendo{at}gmail.com).
Abstract
OBJECTIVE. In numerous studies, the morphologic features of the
patellofemoral joint have been analyzed on radiographs. The objective of this
study was to assess patellofemoral measurements on MR images and to correlate
the measurements with the presence or absence of chondromalacia patellae
confirmed at surgery.
MATERIALS AND METHODS. Axial and sagittal MR images of 98 knees (97
patients) were evaluated. Lateral and medial patellar facet lengths,
lateral-to-medial facet length ratio, and interfacet angle were measured at
three levels through the patella. Trochlear depth was measured on an axial
slice. Patella and patellar tendon lengths, patellar tendon-to-patella ratio,
and overlap of the patellar and trochlear articular cartilages were measured
on sagittal slices. These measurements in knees with chondromalacia patellae
were compared with those in knees without chondromalacia patellae. For
assessment of reproducibility, axial measurements were repeated by a second
observer.
RESULTS. There was no statistically significant difference in any of
the axial and sagittal slice measurements between knees with and those without
chondromalacia patellae. Interobserver reliability was excellent for
measurements of trochlear depth and measurements in the superior and middle
aspects of the patella. Measurements through the inferior patella were
slightly less reproducible.
CONCLUSION. The results of our study with MRI confirmed many
previous radiographic findings. Although we did not find correlation between
the presence of chondromalacia patellae and the patellofemoral indexes we
analyzed, it is possible that the results of further investigations
incorporating different grades of chondromalacia and different locations along
the patellar articular surface may lead to further insight regarding the
morphologic risk factors for chondromalacia patellae.
Keywords: anatomy • chondromalacia patellae • patella • patellofemoral joint
Introduction
Chondromalacia patellae is a relatively common condition characterized by
fissuring within the articular cartilage of the patella and, in severe cases,
erosion of the cartilage to the subchondral bone. Various mechanisms
contributing to the development of chondromalacia patellae have been
suggested, including trauma to the knee, vascular insufficiency within the
subchondral bone [1], and
several structural factors. Structural abnormalities, such as patella alta,
genu valgum, and deformities of the femoral condyle
[2], are believed to injure the
articular surface of the patella because of patellar instability and increased
stress on parts of the articular cartilage.
Numerous studies have been conducted to investigate the correlation between
chondromalacia patellae and structural indexes assessed on radiography. The
ratio of the length of the patellar tendon to the length of the patella
[3–7],
the Q angle [3], the sulcus
angle [4,
5], and the congruence angle
[4] are some of these
measurements. Despite the finding of a degree of association between certain
indexes and chondromalacia patellae, studies have shown variable results, and
thus these indexes in general have not been clinically useful. Consequently,
there is a need to evaluate other structural measurements that may have a
stronger correlation with chondromalacia patellae and may prove to be an
adequate predictor of development of the condition.
Compared with the number of studies of the relative position of the patella
with respect to the distal femur, relatively few studies have been focused on
the shape of the patella itself as a correlate of chondromalacia patellae. In
a classic article, Wiberg [8]
described three types of patellae on the basis of the length of the medial and
lateral facets on axial radiographs and studied the incidence of these types
among knees with chondromalacia patellae. The findings have subsequently been
recognized as the Wiberg classification of patellar shape. Another
classification of patellar shape developed by Ficat and described by Fulkerson
[9] recognized four distinct
types based on the angle between the medial and lateral facets.
View larger version (152K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1A —41-year-old man (A and C) and 42-year-old man
(B) with chondromalacia patellae. Axial MR images show lengths of
lateral and medial facets (A) and interfacet angle (B) through
middle of patella. Similar measurements were made at superior and inferior
slices of patella for each knee.
|
|
View larger version (150K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1B —41-year-old man (A and C) and 42-year-old man
(B) with chondromalacia patellae. Axial MR images show lengths of
lateral and medial facets (A) and interfacet angle (B) through
middle of patella. Similar measurements were made at superior and inferior
slices of patella for each knee.
|
|
View larger version (191K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2A —42-year-old man with chondromalacia patellae (A) and
45-year-old man without chondromalacia patellae (B). Sagittal MR image
shows lengths of patella and patellar tendon through sagittal slice with
greatest patellar length. Patellar tendon-to-patella length ratio was
calculated.
|
|
View larger version (181K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —42-year-old man with chondromalacia patellae (A) and
45-year-old man without chondromalacia patellae (B). Sagittal MR image
shows overlap of articular surfaces of patella and trochlea.
|
|
To our knowledge, in no study has cross-sectional imaging been used to
analyze the relation between chondromalacia patellae and various patellar and
patellofemoral morphologic features. Compared with the quality of radiography,
the improved cross-sectional anatomic detail afforded by MRI allows more
accurate analysis of the patellofemoral joint and its contribution to
chondromalacia patellae. In this study, several measurements of the patella
and the patellofemoral articulation were made on axial and sagittal MR images
and correlated with the presence or absence of chondromalacia patellae at
surgery.
Materials and Methods
The cases of 97 patients (98 knees) who underwent MRI with a 1.5-T system
(Vision, Siemens Medical Solutions) were retrospectively and systematically
evaluated after the institutional review board approved the study. The
eligibility criteria were availability of axial and sagittal MR images (proton
density–, T1-, and T2-weighted) of the knee, no history of surgery on
the ipsilateral knee, and arthroscopic surgery on the ipsilateral knee after
MRI. Sixty-eight male and 29 female patients were included in the study. The
mean age was 45.4 years (range, 14–72 years) at the time of MRI. One of
the male patients underwent MRI and surgery on both knees. The mean interval
between MRI and surgery was 50.0 days (range, 4–485 days). The presence
of chondromalacia patellae was noted by the orthopedic surgeon at surgery.
The lengths of the lateral and medial facets, the ratio between lateral and
medial facet lengths, and the interfacet angle were measured on axial images
through the superior, middle, and inferior portions of the patella (Figs.
1A and
1B). The superior and inferior
levels were designated the most superior and most inferior images at which the
medial and lateral facets were distinguished. The middle level was designated
the midpoint between the superior and inferior levels. The depth of the
trochlea was measured on the axial image revealing the deepest measurement
(Fig. 1C). The lengths of the
patella (from the superoposterior margin to the inferior apex) and the
patellar tendon were measured on the sagittal image revealing the longest
patellar length (Fig. 2A), and
the patellar tendon-to-patella length ratio was calculated. The overlap of the
patellar and trochlear articular cartilages was measured on the sagittal image
revealing the greatest overlap (Fig.
2B). The measurements were made by one observer who was blinded to
the presence or absence of chondromalacia. For assessment of the degree of
interobserver reliability, the axial measurements were repeated by a second
blinded observer.
The measurements (mean 
SD) in the patients with chondromalacia
patellae were compared with those in the patients without chondromalacia
patellae. The p value for each variable was derived with a
Mann-Whitney test. Binary logistic regression analysis was conducted to assess
statistical correlation between any of the variables, alone or in combination,
and the presence of chondromalacia patellae. Interobserver reliability was
calculated with basic linear correlation.
Results
Chondromalacia patellae was detected in 44 knees and was absent in 54 knees
at arthroscopy. Table 1
compares the various patellar and patellotrochlear measurements of knees with
and knees without chondromalacia patellae. Analysis of the axial slices
through the superior and middle portions of the patella showed no
statistically significant difference in length of the lateral facet, length of
the medial facet, ratio of facet lengths, or interfacet angle between the
knees with and those without chondromalacia patellae. At the inferior level of
the patella, there was no significant difference between the two groups in
lateral facet length, ratio of facet lengths, or interfacet angle. The only
significant difference found on binary logistic regression was the length of
the medial facet of the inferior aspect of the patella. The medial facet was
slightly shorter in the knees with (16.0 2.9 mm) than in those
without (17.7 3.8 mm) chondromalacia patellae (p = 0.03).
However, after control for familywise type 1 error rate, this apparent
difference did not maintain statistical significance. One other measurement
made on axial slices, depth of the trochlea, was essentially identical in the
two groups of knees.
Sagittal MRI slices through the knee were used to measure length of the
patella, length of the patellar tendon, ratio of patellar tendon length to
patellar length, and overlap of the patellar and trochlear articular surfaces.
There was no statistically significant difference in any of these variables
between the knees with and those without chondromalacia patellae. Assessment
of interobserver reliability of the axial slice measurements showed excellent
reproducibility of trochlear depth and of the patellar facet measurements in
the superior and middle levels of the patella
(Table 2). The medial and
lateral facet length measurements through the inferior aspect of the patella
were slightly less reproducible.
Discussion
Previous investigations of the morphologic features of the patellofemoral
region and how they relate to chondromalacia patellae have been conducted with
radiographs. We theorized that the failure of these studies to conclusively
show morphologic risk factors for chondromalacia patellae was related to the
inherent inadequacy of the two-dimensional nature of radiography in accurately
representing three-dimensional anatomic features. We believed that
cross-sectional MR images would yield more accurate anatomic measurements and
provide new insight regarding risk factors for chondromalacia patellae that
were not obtainable by analysis of radiographs. In particular, cross-sectional
imaging allows analysis of patellar shape at multiple discrete levels along
its craniocaudal axis compared with the single view allowed by an axial
radiograph.
In 1941, Wiberg [8]
described three types of patellae on the basis of lengths of the medial and
lateral facets and proposed that a disproportionately short medial facet
predisposed the patella to chondromalacia patellae. He explained that the
small area of contact between the medial facet of the patella and the medial
femoral condyle led to increased stress on the medial articular cartilage of
the patella during physiologic motion of the knee. Others
[3,
10,
11] refuted this theory.
Outerbridge [10] did not find
significant correlation between chondromalacia patellae and any particular
Wiberg type. In our study conducted with cross-sectional imaging, the lengths
of the medial and lateral facets and the ratio of the facet lengths measured
at three levels through the patella showed no significant correlation with the
presence of chondromalacia. Other investigators
[9,
10] focused on the angle
formed by the lateral and medial facets of the patella. Outerbridge
[10] found no association
between interfacet angle and the presence of chondromalacia patellae among 168
medial meniscectomy patients. We also did not find significant correlation
between the presence of chondromalacia patellae and interfacet angle at any of
the three levels through the patella.
In numerous studies, the sulcus angle, the angle formed by the lateral and
medial condyles as seen on tangential radiographs of the flexed knee, has been
analyzed as a risk factor for chondromalacia patellae. Aglietti and Cerulli
[5] found a correlation between
a wide sulcus angle and the presence of chondromalacia patellae, whereas Dowd
and Bentley [4] found a
correlation between wide sulcus angle and patellar instability but not with
the presence of chondromalacia patellae. In this study, we measured the depth
of the trochlea to assess the femoral aspect of the patellofemoral
articulation and how it may correlate with the development of chondromalacia
patellae. We found that the mean depth in the patients with chondromalacia
patellae was identical to that in patients without chondromalacia
patellae.
Patella alta, based on the ratio of the length of the patellar tendon to
that of the patella, now commonly called the Insall-Salvati ratio
[12], has been the focus of
numerous investigations as a possible risk factor for chondromalacia patellae
[3–7].
Those who found an association between patella alta and chondromalacia
patellae reasoned that a high-riding patella contributes to malalignment and
incongruity between the articular surfaces of the patella and trochlea
[3,
5,
6]. Other studies have shown no
significant correlation between patella alta and chondromalacia patellae
[4,
7]. Dowd and Bentley
[4] did not find an increased
incidence of patella alta in knees with chondromalacia patellae, but they did
find an association between patella alta and clinical patellar instability. In
another study, Outerbridge
[10] analyzed the length of
the patellar tendon itself rather than the Insall-Salvati ratio and found no
relation between patellar tendon length and the presence of chondromalacia. In
agreement with these findings, our study did not show an association between
the presence of chondromalacia patellae and patellar length, patellar tendon
length, or patella alta.
Outerbridge [10] observed
the position of the patella relative to the femoral condyles as seen on
lateral radiographs but did not perform a detailed analysis regarding its
relation to chondromalacia patellae. He did mention, however, that error would
be introduced depending on whether the patient was actively contracting or
relaxing the quadriceps during radiography. There is a paucity of studies
focused on the patellofemoral articular surfaces as seen on lateral
radiographs. To our knowledge, no previous investigation has correlated the
presence of chondromalacia patellae with the degree of overlap of the patellar
and trochlear articular surfaces. We found no correlation between degree of
cartilage overlap and presence of chondromalacia patellae.
Chondromalacia patellae can be categorized into several grades based on the
severity of cartilage involvement. Outerbridge
[13] proposed one such grading
scheme, which is now widely accepted. One limitation of our study was that we
did not compare grades of cartilage loss with the morphologic indexes. The
sample size in our study prohibited statistically meaningful analysis
correlating the morphologic measurements with the specific grade of
chondromalacia. Furthermore, the study of chondromalacia patellae is
complicated by the fact that chondromalacia within various areas of the
patellar cartilage can be governed by different mechanisms. Fulkerson
[9] stated that chondromalacia
along the lateral facet of the patella often relates to chronic patellar tilt
and excessive lateral pressure syndrome, whereas chondromalacia of the medial
facet is commonly related to patellar dislocation. This observation suggests
that future investigations focusing on chondromalacia within particular
locations of the patellar surface may identify morphologic risk factors that
are mechanism-specific.
Despite the use of cross-sectional imaging to more accurately characterize
the anatomic features of the patellofemoral region than on radiographs, we
were unable to identify risk factors for chondromalacia patellae. This finding
reaffirms the level of complexity governing the biomechanics of the
patellofemoral joint. Although we did not identify any morphologic correlate
to chondromalacia patellae among the particular indexes analyzed, it is
possible that a more comprehensive analysis would reveal certain morphologic
features of the patellofemoral articulation that predispose the patella to
chondromalacia. Our study focused on each index in isolation. Perhaps several
morphologic features in combination are needed to increase the risk of
chondromalacia patellae. It is likely that a cross-sectional analysis of the
patellofemoral articulation during its full range of motion, from maximal
flexion to full extension, with and without weightbearing, is necessary for
further understanding of this subject. As early as 1961, it was proposed
[13] that motion of the
patella over a bony rim present along the medial femoral condyle causes wear
of the cartilage of the medial facet of the patella. Results of cinematic
analysis of the knee in motion may validate this theory.
Our study with cross-sectional imaging showed no significant correlation
between chondromalacia patellae and various measurements of patellar facet
length and interfacet angle at three slices through the patella. We also found
no correlation between chondromalacia patellae and depth of the trochlea,
degree of overlap of the patellar and femoral articular cartilages, patellar
length, patellar tendon length, or patella alta. These findings confirm those
of many previous studies conducted with radiographs. With the advent of
cartilage-specific 3D MRI techniques that provide exquisite detail of the
anatomic and pathologic features of cartilage, further studies are needed to
investigate the role of morphologic features of the patellofemoral joint in
the development of chondromalacia patellae.
References
- Neusel E, Graf J. The influence of subchondral vascularization on
chondromalacia patellae. Arch Orthop Trauma Surg1996; 115:313
–315[Medline]
- Norkin C, Levangie P. Joint structure and function: a
comprehensive analysis. Philadelphia, PA: F. A. Davis,1983
: 327–329
- Insall J, Falvo KA, Wise DW. Chondromalacia patellae: a prospective
study. J Bone Joint Surg Am 1976;58
: 1–8[Abstract/Free Full Text]
- Dowd GS, Bentley G. Radiographic assessment in patellar instability
and chondromalacia patellae. J Bone Joint Surg Br1986; 68:297
–300[Medline]
- Aglietti P, Cerulli G. Chondromalacia and recurrent subluxation of
the patella: a study of malalignment, with some indications for radiography.
Ital J Orthop Traumatol 1979;5
: 187–201[Medline]
- Lancourt JE, Cristini JA. Patella alta and patella infera: their
etiological role in patellar dislocation, chondromalacia, and apophysitis of
the tibial tubercle. J Bone Joint Surg Am1975; 57:1112
–1115[Abstract/Free Full Text]
- Marks KE, Bentley G. Patella alta and chondromalacia. J
Bone Joint Surg Br 1978; 60:71
–73[Medline]
- Wiberg G. Roentgenographic and anatomic studies on the
femoro-patellar joint. Acta Orthop Scand1941; 12:319
–410
- Fulkerson JP. Disorders of the patellofemoral
joint, 4th ed. Philadelphia, PA: Lippincott Williams &
Wilkins, 2004: 129–142,
239–253
- Outerbridge RE. Further studies on the etiology of chondromalacia
patellae. J Bone Joint Surg Br 1964;46
: 179–190[Medline]
- Goodfellow J, Hungerford DS, Zindel M. Patellofemoral joint
mechanics and pathology: functional anatomy of the patello-femoral joint.
J Bone Joint Surg Br 1976;58
: 287–290[Medline]
- Insall J, Salvati E. Patella position in the normal knee joint.
Radiology 1971;101
: 101–104[Medline]
- Outerbridge RE. The etiology of chondromalacia patellae.
J Bone Joint Surg Br 1961;43
: 752–757[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
