DOI:10.2214/AJR.07.2098
AJR 2008; 190:17-21
© American Roentgen Ray Society
Prevalence and MRI-Anatomic Correlation of Bone Cysts in Osteoarthritic Knees
Caroline Pouders1,2,
Michel De Maeseneer3,
Peter Van Roy1,
Jan Gielen4,
Annietta Goossens5 and
Maryam Shahabpour2
1 Department of Experimental Anatomy, Vrije Universiteit Brussels, Laarbeeklaan
103, 1090 Brussels, Belgium.
2 Department of Radiology, AZ-Vrije Universiteit Brussels, Brussels,
Belgium.
3 Division of Radiological Sciences, Wake Forest University, Winston-Salem,
NC.
4 Department of Radiology, University of Antwerp, Antwerp, Belgium.
5 Department of Pathology, AZ-Vrije Universiteit Brussels, Brussels,
Belgium.
Received February 20, 2007;
accepted after revision July 6, 2007.
Address correspondence to C. Pouders
(caroline.pouders{at}vub.ac.be).
Supported by Willy Gepts Foundation, Vrije Universiteit Brussels,
Belgium.
FOR YOUR INFORMATION
Realize Your Potential with AJR Integrative Imaging
Each quarterly issue of AJR Integrative Imaging offers you a
series of peer-reviewed learning opportunities, many with CME credit, which
will challenge and enhance your diagnostic and treatment skills. Whether it is
the printed version or the online version, AJR Integrative Imaging is
an important tool for your professional development. For more information or
the latest issue of AJR Integrative Imaging, visit
www.ajronline.org.
Abstract
OBJECTIVE. The objectives of this study were to determine the
prevalence of cysts in the weight-bearing areas and interspinous bone cysts in
tibial plateau specimens derived from knees with advanced osteoarthritis, to
perform MRI-anatomic correlation of these cysts, and to define their
histologic characteristics.
MATERIALS AND METHODS. Forty-two tibial plateau specimens were
recovered from patients undergoing total knee replacement surgery for severe
osteoarthritis (14 men and 28 women; mean age, 74 years; age range, 58-87
years). The tibial specimens underwent MRI with T1- and T2-weighted MR
sequences. No radiographic data were available before MRI. Tibial specimens
were sectioned using a high-speed rotating diamond disk into 3-mm-thick
slices. MR images and anatomic specimens were analyzed for the presence of
cysts in the interspinous and weight-bearing areas. Histologic staining
methods included routine Harris hematoxylin stain, trichrome stain, and Alcian
blue-PAS stain.
RESULTS. Twenty-three (54%) of 42 specimens contained one or more
cystic areas, with a total of 30 cystic areas. The cysts were distributed in
the weight-bearing area (14/30) and interspinous area (16/30). All cysts seen
on the anatomic slices could also be depicted on both MRI sequences.
Histologic findings were identical for all noncommunicating cysts and revealed
necrotic bone fragments with dead denuclearized cells. The cavities were
surrounded by a layer of fibrous connective tissue containing adipocytes and
osteoblasts. No evidence of epithelial components was found in the lining of
the cavities.
CONCLUSION. Interspinous and weight-bearing tibial cysts are common
in severe knee osteoarthritis. The cysts contain necrotic bone fragments and
are lined by a nonepithelial fibrous wall. Our findings support the hypothesis
that interspinous cysts could result from repetitive bone stresses through the
cruciate ligaments. Our findings do not support the use of the term
"herniation cyst."
Keywords: bone cysts • interspinous cysts • knee osteoarthritis
Introduction
Subchondral cystic areas are quite common in degenerative disease of
the knee and can be depicted on both radiography and MRI. Characteristic
radiologic changes associated with the degeneration of cartilage include
narrowing of the joint space; subchondral sclerosis; and the appearance of
osteophytes, intraarticular osteochondral bodies, bone marrow edema, and
subchondral cysts. Although the latter are associated with osteoarthritis,
they can develop as primary lesions independent of it
[1].
Even before the advent of MRI, these cysts received great attention. They
are thought to be formed either by synovial intrusion or through bone
contusion. According to the intrusion theory
[2-7],
severely damaged cartilage allows intrusion of synovial fluid into the bone,
resulting in cyst formation. This type of cyst communicates with the joint
surfaces and has been widely investigated in the setting of knee
osteoarthritis. According to the bone contusion theory
[1,
6,
8-13],
noncommunicating cysts arise in foci of bone necrosis resulting from impact at
opposing bone surfaces. Fewer investigations have addressed noncommunicating
cysts.
In our clinical experience with MRI, we observed that cystic areas are
quite commonly seen at the interspinous area in the tibial plateau. The bone
in this location is not covered by cartilage and is not located in a
weight-bearing area. The absence of cartilage in the interspinous area makes
the intrusion theory unlikely. In addition, this is not a weight-bearing area;
therefore, the contusion theory also seems not applicable. In clinical
practice, these interspinous cysts are commonly reported as mucoid cysts or
herniations of synovial tissue. However, to our knowledge, no histologic
evidence is available to support this assumption. To our knowledge, the
prevalence, histology, and significance of these interspinous tibial cysts
have not been further investigated previously.
View larger version (51K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 1 —87-year-old man with osteoarthritis. Sagittal T1-weighted spin-echo
image of weight-bearing area of tibial specimen. Note surface of cartilage
(white arrows) and cartilage-subchondral bone interface (black
arrows). S = subchondral bone.
|
|
View larger version (67K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 2B —Weight-bearing area of tibial specimen in 82-year-old man with
osteoarthritis. In sagittal T2-weighted gradient-echo image, hyperintense
cystic structure in weight-bearing area is seen (arrow).
|
|
The objective of our work was to determine the prevalence of interspinous
cysts in osteoarthritic tibial plateau specimens, to perform MRI-anatomic
correlation of these cystic areas, and to define histologic characteristics of
these cysts.
Materials and Methods
Forty-two tibial plateau specimens were obtained from patients undergoing
total knee replacement for severe knee osteoarthritis. The study was approved
by the ethics review board of the hospital in which the patients underwent the
surgical procedure. Specimens of patients with known infectious vectors were
not used for the study. No further patient data or existing radiographic
images were available. After harvesting, the specimens were frozen until MRI
could be scheduled. A 1.5-T clinical system was used (Intera, Philips Medical
Systems). The sequences performed included sagittal T1-weighted spin-echo with
the following imaging parameters: TR/TE, 450/15; slice thickness, 3 mm;
acquisition time, 6 minutes 21 seconds; field of view, 80 mm; one repetition.
Sagittal T2-weighted fast-field echo (FFE) sequences were performed with the
following imaging parameters: 500/14; slice thickness, 3 mm; acquisition time,
6 minutes 59 seconds; field of view, 80 mm; one repetition.
After MRI, all tibial specimens were sectioned using a high-speed rotating
diamond-covered disk (SG Elastic EHT 115-1, 0 A60 R SG-Inox, PFERD) into
3-mm-thick slices. MR images and slices were photographed and placed side by
side for inspection by consensus of an anatomist with 4 years of experience in
cartilage gross anatomy and a radiologist with 10 years of experience in
musculoskeletal radiology. The MR images were interpreted at the same time as
the anatomic slices.
The presence of intraosseous cysts in the weight-bearing areas and
interspinous areas was analyzed on the MR and anatomic slices, as was the
presence of cartilage damage. Intact cartilage was defined as an uninterrupted
cartilage line above the subchondral bone. Damaged or breached cartilage was
defined as a disturbance in the cartilage integrity, with exposure of an area
of subchondral bone to the joint space cavity. For cysts in the weight-bearing
area, the presence of macroscopic communication between the cyst and joint
space was determined to differentiate communicating from noncommunicating
cysts. The dimensions of the cysts were recorded as the largest diameters on
the sagittal sections.
On MRI, cysts were defined as well-circumscribed rounded or oval areas of
signal abnormality with low signal on T1-weighted images and high signal on
T2-weighted images. The abnormal areas on MRI were characterized with regard
to signal intensity.
Representative specimens of interspinous cysts (n = 2) and
noncommunicating cysts from the weight-bearing area (n = 3) were
selected for further histologic analysis. After decalcification, the specimens
were embedded in paraffin. Staining methods included routine Harris
hematoxylin stain (nuclear), trichrome stain (connective tissue), and Alcian
blue-PAS stain (mucoproteins, proteoglycans). Histologic findings were
interpreted by an experienced pathophysiologist who was aware of the
macroscopic and MRI findings.
Results
The patients were 14 men and 28 women with a mean age of 74 years (age
range, 58-87 years). All tibial specimens (n = 42) showed findings of
moderate to advanced cartilage damage. Twenty-three (54%) of 42 specimens
contained one or more cystic areas, with a total of 30 cystic areas. The cysts
were distributed in the weight-bearing area (14/30) (Figs.
1,
2A,
2B,
3A,
3B,
3C) and interspinous area
(16/30) (Fig. 4A,
4B). In seven (30%) of 23
positive specimens, cysts were only found in the weight-bearing areas; in 11
(48%) of 23 positive specimens, cysts were only found in the interspinous
area; and in five (22%) of 23 positive specimens, cysts were found in the
weight-bearing and interspinous areas.
View larger version (74K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3A —Weight-bearing area of tibial specimen in 69-year-old man with
osteoarthritis. Sagittal T1 spin-echo image shows hypointense intraosseous
cyst with hypointense margins (arrow). No joint space communication
is seen.
|
|
View larger version (80K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3B —Weight-bearing area of tibial specimen in 69-year-old man with
osteoarthritis. Sagittal T2 gradient-echo image shows hyperintense
intraosseous cyst surrounded by hypointense margins (arrow).
|
|
View larger version (128K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 3C —Weight-bearing area of tibial specimen in 69-year-old man with
osteoarthritis. Sagittal anatomic slice of same region shows rounded cystlike
structure containing amorphous fragments (arrow).
|
|
View larger version (64K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 4B —Interspinous area of tibial specimen in 82-year-old woman with
osteoarthritis. Sagittal T1 spin-echo image shows hypointense cystlike
structure surrounded by margin with lower signal intensity than cyst
(arrows).
|
|
Most cysts in the weight-bearing areas were located below intact cartilage
and lacked a communication with the joint space cavity. Only two cysts were
located below breached cartilage. The cortical bone overlying the cystic areas
at the interspinous area was intact and no communication with the knee joint
was seen.
All cysts seen on the anatomic slices could be depicted on both MRI
sequences. On MRI, cysts presented as well-demarcated areas of high signal
intensity on T2-weighted images and low signal intensity on T1-weighted
images. The interspinous cysts ranged in size from 0.15 to 0.75 cm (mean, 0.38
cm).
Histologic findings were identical for all noncommunicating cysts below the
weight-bearing areas and interspinous cysts
(Fig. 5). Histologic
correlation showed that the noncommunicating cysts contained necrotic bone
fragments with dead denuclearized cells. The cavities were surrounded by a
layer of fibrous connective tissue containing adipocytes and osteoblasts. No
evidence of epithelial components was found in the lining of the cavities.
View larger version (125K):
[in this window]
[in a new window]
[as a PowerPoint slide]
|
Fig. 5 —Histologic stain of cyst in 82-year-old woman with osteoarthritis.
Fibrous connective tissue layer with adipocytes and osteoblasts is seen (F).
Cyst cavity (C) contains necrotic bone fragments with dead denuclearized
cells. (Harris hematoxylin)
|
|
Discussion
The spectrum of subchondral bone cysts seen on radiographs encompasses a
heterogeneous group of conditions. Various terms have been used to describe
subchondral lucencies, including geode, mucoid cyst, degenerative cyst,
intraosseous ganglion, and synovial herniation. Although some of these
conditions may represent different radiologic and histologic entities, it is
often unclear what the distinctive features may be, and it seems likely that
overlap is present. Although the term "bone cyst" is most
appropriate to describe the implied subchondral radiolucency, it is not
entirely correct because it suggests the presence of a cellular lining, which
is not always the case, as was again confirmed by our observations. Cysts
occurring in the interspinous area of the tibia in osteoarthritic knees were
the focus of this study.
MRI is now commonly performed in osteoarthritic knees, revealing cartilage
lesions, osteophytosis, meniscoligamentous injury, bone marrow edema patterns,
and subchondral cysts. Subchondral cysts may be located in the weight-bearing
areas or interspinous area. Communicating cysts below breached cartilage have
been widely studied, with fewer studies addressing noncommunicating cysts in
the weight-bearing area. The precise nature of interspinous tibial bone cysts
has, to the best of our knowledge, not been addressed.
We found that the prevalence of interspinous cysts in severe osteoarthritic
tibial specimens was 38.5%. The prevalence of cysts below the weight-bearing
areas was 33%. Twenty-two percent of the specimens had interspinous cysts
combined with cysts in the weight-bearing area.
Previously reported theories for cysts in the weight-bearing area conform
to either the intrusion or the contusion model. In the intrusion model
[2-7],
a defect of the articular cartilage occurs, and joint fluid passes through the
breached cartilage, resulting in hydraulic destruction of the subchondral bone
[2,
3]. Synovial membranes can also
pass through the breached cartilage (synovial herniation) and become lodged in
the subchondral bone. However, in many cysts, a direct communication with the
joint is not shown [9]. This
finding forms the basis for the contusion model
[1,
6,
8-13].
In the contusion model, repetitive microtrauma is thought to lead to a
localized area of subchondral necrosis with secondary cyst formation. This
theory is based on histologic evidence of bone contusion, trabecular fracture,
and primary subchondral osteolysis. Other theories attempt to explain the
formation of intraosseous ganglions and synovial herniations. Theories for
development of intraosseous ganglions include postulation of the ingrowth of a
soft-tissue ganglion in the bone versus a concept based on repetitive
microtrauma quite similar to the contusion model
[14-16].
In synovial herniation, synovium becomes embedded into the bone surface
[7].
Our findings in noncommunicating cysts in the weight-bearing and
interspinous areas are consistent with the reported contusion theory. Our
observations showed cystic lesions that consisted of necrotic bone fragments
and lacked an epithelial lining. The findings were identical for
noncommunicating cysts in the weight-bearing and interspinous areas. The
mechanism for repetitive microtrauma in the interspinous area is not
immediately evident because this is not a weight-bearing area. Because
mechanical forces resulting from the articulating femoral condyles do not
directly affect the interspinous area, we hypothesize that the tensile forces
produced by the cruciate ligaments may cause sufficient bone stress to result
in bone microtrauma and cyst formation. McLaren et al.
[13] previously hypothesized
that chronic stress on the bone may produce bone resorption and focal necrosis
resulting in the secondary formation of a cyst.
The cyst wall, as shown by histologic examination, corresponds to a
pseudocapsule composed of fibrous tissue and does not contain an epithelial
cellular lining. The composition of this wall somewhat resembles the wall of
ganglion cysts, however, without a connective tissue matrix composed of
parallel fascicles of collagen, which is a typical feature of ganglion cysts
[12]. Our histologic studies
showed no evidence of embedded synovial tissues in the interspinous cysts and
hence provided no supporting evidence for the synovial herniation theory.
Our study has several limitations. First, because specimens were obtained
from patients undergoing total knee arthroplasty, degenerative disease was
severe at least in one compartment. The obtained prevalence of interspinous
cysts is only representative of such a population and may be lower in
less-advanced osteoarthritis. Second, the specimens were frozen and thawed for
this study, which may have influenced the MRI or histologic characteristics.
We did not observe any adverse effects on MRI, macroscopy, or histology,
however. Third, the number of specimens studied histologically was limited;
however, macroscopic findings in all specimens were similar as well as the
histologic findings in the studied specimens. Fourth, MR images and anatomic
slices were analyzed simultaneously, and this may have caused bias in
assessing detectability of cysts. Fifth, the use of specimens also implied the
absence of a standard for signal intensity on MRI. And last, because we used
specimens, it was also not possible to assess the condition of the anterior
cruciate ligament.
In conclusion, the prevalence of interspinous cysts and noncommunicating
cysts in weight-bearing areas in osteoarthritic knees is 45%. Histologic
findings of interspinous tibial plateau cysts show necrotic bone and are
similar to cysts in the weight-bearing area without joint-space communication.
The interspinous cysts contain necrotic bone fragments and are lined by a
nonepithelial fibrous wall. We hypothesize that interspinous cysts could
result from repetitive mechanical stresses transmitted through the cruciate
ligaments. Because an epithelial lining is absent, we suggest that the term
"pseudocyst" may be more appropriate than "cyst" to
describe these changes. The designation "synovial herniation cyst"
seems incorrect for the description of these cystic areas.
References
- Crawford R, Sabokbar A, Wulke A, Murray DW, Athanasou NA. Expansion
from an osteoarthritic cyst associated with wear debris. J Bone
Joint Surg Br 1998; 80:990
-993[CrossRef][Medline]
- Crane AR, Scarano JJ. Synovial cysts (ganglia) of the bone: report
of two cases. J Bone Joint Surg Am 1967;49
: 355-361[Abstract/Free Full Text]
- Schmalzried TP, Akizuki KH, Fedenko AN, Mirra J. The role of access
of joint fluid to bone in periarticular osteolysis. J Bone Joint
Surg Am 1997; 79:44
-52[Abstract/Free Full Text]
- Landells JW. The bone cysts of osteoarthritis. J Bone
Joint Surg Br 1953; 35:643
-649[Medline]
- Kambolis C, Bullough PG, Jaffe HL. Ganglionic cystic defects of the
bone. J Bone Joint Surg Am 1973;55
: 496-505[Abstract/Free Full Text]
- Schajowicz F, Clavel Sainz M, Slullitel JA. Juxta-articular bone
cysts (intra-osseous ganglia): a clinicopathological study of eighty-eight
cases. J Bone Joint Surg Br 1979;61
: 107-116[Medline]
- Vanhoenacker FM, Van de Perre S, De Vuyst D, De Schepper AM.
Proceedings of the meeting of the KBVR-SRBR Osteo-Articular section, Brussels,
June 21, 2003: cystic lesions around the knee.
JBR-BTR, 2003; 86:302
-304
- Woods CG. Subchondral bone cysts. J Bone Joint Surg
Br 1961; 43:758
-766[Medline]
- Sabokbar A, Crawford R, Murray DW, Athanasou NA.
Macrophage-osteoclast differentiation and bone resorption in osteoarthrotic
subchondral acetabular cysts. Acta Orthop Scand2002; 71:255
-261[CrossRef]
- Rhaney K, Lamb DW. The cysts of osteoarthritis of the hip: a
radiological and pathological study. J Bone Joint Surg
Br 1955; 37:663
-675[Medline]
- Furgeson AB. The pathological changes in degenerative arthritis of
the hip and treatment by rotational osteotomy. J Bone Joint Surg
Am 1964; 46:1337
-1352[Free Full Text]
- Abdel-Wanis ME, Tsuchiya H, Minato H, Morinaga T, Yamamoto N,
Tomita K. Bilateral symmetrical cysts in the upper tibiae in a skeletally
mature patient: might they be simple bone cysts? J Orthop
Sci 2001; 6:59
-60[CrossRef][Medline]
- McLaren DB, Buckwalter KA, Vahey TN. The prevalence and
significance of cyst-like changes at the cruciate ligament attachments in the
knee. Skeletal Radiol 1992;21
: 365-369[Medline]
- McCarthy CL, McNally EG. The MRI appearance of cystic lesions
around the knee. Skeletal Radiol 2004;33
: 187-209[CrossRef][Medline]
- Janzen DL, Peterfy CG, Forbes JR, Tirman PFJ, Genant HK. Cystic
lesions around the knee joint: MR imaging findings.
AJR 1994; 163:155
-161[Abstract/Free Full Text]
- Feldman F, Johnston A. Intraosseous ganglion.
AJR 1973; 118:328
-343[Abstract]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
Top
Abstract
Introduction
