Identification of the Vascular and Avascular Zones of the Human

Meniscus Using Magnetic Resonance Imaging:

Correlation with Histology

Peter S. H. Chan, M.D., J. Bruce Kneeland, M.D., Francis H. Gannon, M.D.,

Wayne T. Luchetti, M.D., and Richard J. Herzog, M.D.

Summary: Since the initial employment of magnietic resonance imaging (MRI)

to diagnose meniscal tears, a characteristic low-signal intensity, triangular-shaped
structure has been interpreted as representing the entire meniscus. The difficulty in
diagnosing meniscocapsular separations with MRI has brought attention to our
lack of understanding of the appearance on MRI of the outer third of the meniscus
and the meniscocapsular junction. We correlated MRIs of the meniscus in cadaver
knees with histological sections and found that the low-signal, wedge-shaped
structure corresponds only to the avascular (white) zone of the meniscus, whereas
the high-signal zone peripheral to it corresponds to the vascularized (red) zone.
Key Words: Magnetic resonance imaging—Knee—Meniscus—Histology.

M agnetic resonance imaging (MRI) has become

widely accepted as an accurate, noninvasive
imaging technique for detecting meniscal tears. Since
nature of the two different types of tissue seen with
MRI.

the earliest MRI examinations of the knee, it has been

believed that the characteristic wedge-shaped low-
signal structure seen on the MRIs represents the entire
meniscus. Only in one recent study1 did the investiga-
tors question this belief and, in an attempt to account
for the difficulty they observed in the MR diagnosis of
meniscocapsular separation, suggest that at least part
of the higher signal tissue that is peripheral to the low
signal wedge might represent the highly vascularized
peripheral red zone of the meniscus and not the joint
capsule. We undertook this study to determine the
From the Departments of Orthopaedic Surgery (P.S.H.C., W.T.L.),
Radiology (J.B.K., R.J.H.), and Pathology (F.H.G.), the Hospital of
the University of Pennsylvania, Philadelphia, Pennsylvania, U.S.A.
Supported in part by National Institutes of Health Grant No. RR
02305.
Address correspondence and reprint requests to J. Bruce Knee-
land, M.D., Department of Radiology, Silver-1, Hospital of the
University of Pennsylvania, 3400 Spruce St, Philadelphia, PA
19104, U.S.A.
r 1998 by the Arthroscopy Association of North America
0749-8063/98/1408-1731$3.00/0
MATERIALS AND METHODS
Six adult fresh-frozen cadaveric knees were thawed
at room temperature for 24 hours and studied with
MRI, performed on a 1.5-Tesla clinical MRI system
(Signa, General Electric Medical Systems, Milwau-
kee, WI) using the standard extremity coil. Images
were obtained using a conventional spin echo se-
quence with TR ⫽ 2,500 msec and TE ⫽ 20 msec
(proton-density weighted) and 80 msec (T2-weighted),
a 12-cm field of view, a 256 ⫻ 192 acquisition matrix,
3-mm thick sections with a 1-mm gap and 1 signal
averaged. (This sequence was chosen because it is
widely used for the detection of meniscal tears clini-
cally.)
Measurements of the length (i.e., the radial thick-
ness) of the low-signal, triangular portion of the
meniscus from its apex to its peripheral border were
made on films of the proton-density weighted MRIs at
selected locations in both the sagittal and coronal
planes (Fig 1) using calipers and the measurement grid

820 Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 14, No 8 (November-December), 1998: pp 820–823
 MRI OF MENISCAL RED/WHITE ZONES 821

FIGURE 1. Magnified view of sagittal sections obtained through the medial compartment of a cadaveric knee specimen obtained using
spin-echo sequences with (A) TR ⫽ 2,500 msec and TE ⫽ 17 msec, and (B) TE ⫽ 80 msec. Straight, white arrows indicate the locations
chosen for the measurement of the length of the low signal portion of the meniscus. Black arrows indicate the joint capsule. The higher signal
peripheral zone of the meniscus is indicated by the curved black arrow.

provided with the software. In the sagittal plane, the
sections on which the measurements were made were
the central section (as determined by counting the
number of sections on which the meniscus was visible)
through the anterior and posterior horns of both the
medial and lateral meniscus as well as the adjacent
sections on either side of the central section. In the
coronal plane, the measurements were taken on the
central section in the body of both the medial and
lateral menisci as well as the adjacent anterior and
posterior sections. All measurements were made inde-
pendently by two musculoskeletal radiologists (J.B.K.,
R.J.H.) with many years of experience in the interpre-
tation of MRI examinations of the knee.
Immediately after the MRI, the knee was disarticu-
lated and the menisci and surrounding structures were
removed en bloc. Careful attention was paid to pre-
serve all capsular attachments and synovium lining the
menisci. The central sections in both sagittal and
coronal planes were tagged and adjacent sections at
4-mm intervals were marked. This allowed histologi-
cal sectioning to be performed that corresponded to the
MRIs. Specimens were embedded in paraffin and and
sectioned in 5-µm thick increments at the location
indicated by the markers on the meniscus. One slide of
each block was stained with H&E and additional slides
were then stained immunohistochemically using the
avidin-biotin technique for Factor VIII and CD 34
(both of which are vascular markers).2 The presence of
vessels was indicated by a brown pigment in the
cytoplasm of the vascular endothelial cells. The slides
were reviewed by a musculoskeletal pathologist and
the distance from the central apex of the meniscus to
the end of dense fibrocartilage was measured for each
section (Fig 2). The distribution of vessels with respect
to the end of the dense fibrocartilage was noted (Fig 3).
The size of the fibrocartilaginous portion of the
meniscus was measured under low magnification us-
ing a reticle (a distance scale embedded in the optics).
A correlation matrix was generated and an overall
correlation coefficient determined for the comparison
of radiological and histomorphometric measurements
for each of the two radiologists and for comparison of
the two radiologists’ measurements. Fisher’s r to z test
was used to determine significance.
RESULTS
The MRIs consistently showed a low-signal central
zone extending peripherally from the apex. Peripheral
to this lay a higher signal intensity zone that extended
to the edge of the joint (the capsule) (Fig 1). This
higher signal zone was much more evident on the
proton density than the T2-weighted images in which
the signal was only minimally increased over that of
the lower signal central zone. The thickness of this
higher signal zone varied depending on the location
within the meniscus but was greatest in both posterior
horns and smaller in the body and anterior horns of
both menisci.
On the H&E–stained sections, a zone of fibrocarti-
lage was shown to extend peripherally from the apex
(Fig 2). This represents the white zone of the menis-
cus.3 The immunohistochemical staining failed to
show any vessels within the fibrocartilage (Fig 3).
822 P.S.H. CHAN ET AL.
FIGURE 2. Low-power photo-
micrograph of a section of the
meniscus showing the typical
histological appearance. At the
junction of the areolar tissue (a)
and the fibrocartilage (f) is a
thin (⬍ 0.2 mm) transition zone
that contains both dense fibrosis
and a fibrocartilage-like tissue
(s, synovial reflection). (H&E,
original magnification ⫻ 60.)
Adjacent to the zone of fibrocartilage was a thin (less
than 0.2 mm) transitional zone of dense fibrosis and
fibrocartilaginous-like tissue. Immunohistochemical
staining showed several small-caliber vessels within
this zone. Peripheral to the transitional zone and
extending to the visible remnants of the capsule lay a
zone of loose, areolar connective tissue that was
shown to be highly vascularized by the immunohisto-
FIGURE 3. Medium-power view of a section of meniscus stained
immunohistochemically for Factor VII (an endothelial marker).
The reaction product is noted only in vascular walls (arrow). The
vessels extend from the loose areolar tissue (left end of section) into
the thin transition zone noted in Fig 2 and labeled here as the
vascular fibrocartilage (vf). The remainder of the fibrocartilage is
avascular (af).
chemical technique and that corresponds to the red
zone of the meniscus. The thickness of this zone of
areolar connective tissue varied depending on the
location but, similar to the higher signal zone on the
MRIs, was greatest in the posterior horns of both
menisci.
The correlation coefficient between the measure-
ments of radial distance of the low signal zone on the
MRIs and the zone of fibrocartilage on the histological
sections was 0.79 for one radiologist and 0.81 for the
second radiologist with P ⬍ .0001 for both values. The
correlation coefficient between the measurements per-
formed by the two radiologists was 0.83 with P ⬍
.0001.
By visual inspection and by measurement, the low
signal zone on the MRIs was seen to correlate with the
zone of fibrocartilage (white zone) and the higher
signal zone with the loose areolar connective tissue
(red zone). The small thickness of the transitional zone
seen on the histological sections (less than 0.2 mm)
made it difficult to assign to either the low or higher
signal zones at the resolution used for the imaging
studies.
DISCUSSION
An improved understanding of the importance of
the meniscus to the normal function of the knee has led
 MRI OF MENISCAL RED/WHITE ZONES 823

FIGURE 4. Sagittal sections through the medial meniscus obtained using spin-echo sequences with (A) TR ⫽ 2,500 and TE⫽17 msec, and (B)
TE ⫽ 80 msec. There is a linear focus of increased signal (arrows) located within the higher signal peripheral zone that was found at
arthroscopy to represent a red zone tear close to the meniscocapsular junction.

to a new emphasis on preserving the meniscus as much
as possible at surgery through the use of partial
meniscectomy and meniscal repair.4,5 The choice of
therapy is influenced considerably by the location of
the tear in relation to the peripheral vascularized red
zone and central avascular white zone of the meniscus.
Thus, there is a strong rationale for determining if MRI
can reliably distinguish between these zones.
In this investigation, we showed that MRI can
consistently identify the presence of two zones within
the meniscus on the basis of different signal intensities
and that both qualitatively and quantitatively there is a
good correlation between the size of the central,
low-signal zone on the MRIs and that of avascular,
fibrocartilaginous (white) zone of the meniscus as seen
on histological examination. We believe that this is an
important first step toward the use of MRI to differen-
tiate between white/white and red/white tears as well
as facilitating the diagnosis of meniscocapsular separa-
tions. Figure 4 illustrates a case in which the MRIs
showed a linear focus of bright signal in the peripheral
zone of the meniscus that at arthroscopy was found to
represent a tear within the red zone close to the
meniscocapsular junction.
REFERENCES
1. Rubin DA, Britton CA, Towers JD, Harner CD. Are the MR
imaging signs of meniscocapsular separation valid? Radiology
1996;201:829-836.
2. Hsu S-M, L Raine, H Fanger. The use of avidin-biotin-
peroxidase complex (ABC) in immunoperidase techniques: A
comparisopn between ABC and unlabeled antibody procedures.
J Histochem Cytochem 1981;29:577-580.
3. Arnoczky SP, RF Warren. Microvasculature of the human
meniscus. Am J Sports Med 1982;10:90-95.
4. Dehaven KE. Peripheral meniscal repair: An alternative to
meniscectomy. J Bone Joint Surg Br 1981;63:463-472.
5. Cassidy RE, AJ Shaffer. Repair of peripheral meniscal tears: A
preliminary report. Am J Sports Med 1981;9:209-215.