Magnetic resonance imaging (MRI) is the most commonly used and highly valued noninvasive imaging procedure for suspected injuries to the ACL (21).
The main factors in determining the detection of the ACL by MRI are field strength and plane of view.
Although Kwon et al,
(8) reported that additional oblique planes are useful for the evaluation of ACL tears,
most previous studies have concluded that it is not clear which planes and sections are most appropriate for accurate MRI evaluation of the ACL.
Moreover,
the separation of ACL into AM and PL fiber bundles has now been widely accepted as a basis for understanding ACL function.
On routine MR images,
the AM bundle is usually the one that can be evaluated and the PL bundle is less frequently observed (5).
Because oblique sagittal and oblique coronal imaging have been used to delineate the ACL more clearly (17,
22),
we used the oblique coronal plane for separate evaluations of AM and PL bundles.
In patients with a severely atrophic or attenuated ACL or when the ACL exhibited a prominent angulated or wavy course or a severely edematous condition,
the clear differentiation of these two bundles was difficult.
Nevertheless,
we were able to analyze both the AM and PL bundles in most cases.
Furthermore,
since these bundles have different biomechanical functions,
the histological appearance of each bundle of the ACL in osteoarthritic knee should be investigated,
and thus,
we investigated each bundle during our histologic examination.
The ability to accurately differentiate complete,
partial,
and isolated AM tears and isolated PL tears would undoubtedly aid preoperative planning (23,
24).
Several investigators have found MR to be useful for the assessment of the double-bundle anatomy of the intact,
normal ACL,
but data is lacking regarding its ability to detect isolated AM or PL bundle tears in patients with a history of ACL injury (25).
In addition,
isolated ruptures of the AM and PL bundles are difficult to diagnose during arthroscopy,
and a potential cause of controversy (6).
Relatively few MR studies have considered complete and partial tears separately,
but these were performed using 1.5T or lower field strength systems,
which have not been shown to be accurate enough to differentiate complete and partial ACL tears,
or to identify partial ACL tears (6,
7,
18).
According to our results,
histologic,
macroscopic in the operative field and radiologic grades were inconsistent,
despite being significantly correlated.
In some cases,
the ACL did not show high-grade partial or complete rupture on MR images or in the operative field,
despite a high grade pathologic condition.
These findings demonstrate that the functional role of a histologically degenerated but morphologically intact ACL is questionable,
and suggest that a severely degenerative state might not cause direct high-grade partial or complete disruption on standard MR images or in macroscopic view.
In the present study,
we used T2* mapping to objective evaluate of ACLs.
T2 relaxation time reflects the specificity of tissue by describing the attenuation of transverse magnetization of tissue.
Specifically,
it is a measure of MR signal intensity during different echo times,
and specific values are calculated using a formula (26).
T2 mapping has the potential to quantitatively evaluate deteriorations in molecular composition (27),
and T2 is sensitive to water content and the arrangement of the collagen network structure (28).
On the other hand,
T2* is the relaxation time obtained with a gradient-echo pulse sequence that comprises both T2 relaxation and coherent dephasing effects,
which arise from spins within a voxel having different precession frequencies due to local field variations,
within the net T2* decay (29).
To measure T2* relaxation,
a series of images with different TEs must be acquired.
TEs between 0.008 and 40 ms have been used to date (30).
In the present study,
we used 4.2 ms TE for T2* mapping images and the total scanning time for this sequence was 3 minutes 32 seconds.
According to the results obtained,
the pathologic grades and T2* values of the ACL bundles were significantly and linearly correlated.
In terms of the correlation between T2* values and operative and radiologic grades,
only the AM bundle showed a significant correlation.
Therefore,
these results suggest that T2* values well demonstrate the severity of ACL degeneration.
However,
in some cases,
ROI values and pathologic grades showed a negative linear correlation.
We supposed that in the severe fibrotic state,
ROI values are lower,
despite higher pathologic grades.
Hovis et al.
(31) found that severity of degeneration in the knee joint did not differ between complete and non-complete ACL groups,
and also,
found that usually mild and moderate OA was associated with abnormalities of the ACL,
and that symptomatic OA was associated more so with severe knee degeneration.
These findings are important for surgeons considering cruciate-conserving arthroplasty or unicompartmental arthroplasty in osteoarthritic knees.
In our study,
according to operative results,
79 % (42/53) of AM bundles and 81 % (43/53) of PL bundles were less than grade 2.
Regarding imaging results,
43 % (23/53) of AM bundles and 39 % (21/53) of PL bundles were more than grade 3.
These results are contrary to those of several previous reports (32),
in which,
it was suggested that ACL rupture probably occur a prior to the development of knee OA in most subjects.
However,
according to our pathologic results,
58 % (31/53) of AM bundles and 62 % (33/53) of PL bundles were more than grade 3,
which concurs with the findings of Mullaji et.
al.
(33),
who found a statistically significant positive correlation between radiologic grade of arthritis and histologic grade of ACL degeneration.
There were some limitations in the present study.
First,
the number of patients included was relatively small.
Second,
the differentiation of AM and PL bundles was problematic as in cases of severely degenerative or atrophic ACL states,
clear separation of the two bundles was difficult on all images,
in the operative field,
and during tissue preparation for pathologic examinations.
However,
vague boundary regions between bundles were excluded from pathologic and imaging analyses.
Third,
we could not perform the isotropic three-dimensional (3D) MR sequences used at 3T,
because T2* mapping software was only available for a 1.5 T MR system.
Isotropic 3D sequences enable multiplanar reconstructions along the ACL,
and could aid diagnosis,
especially for clinical applications of the double-bundle concept (34).
Further study using more recent techniques and a 3 T-MR system is needed.
Fourth,
all patients enrolled underwent total knee arthroplasty for severe osteoarthritis.
Thus,
ACL status was evaluated from an osteoarthritic perspective,
and additional study of ACL injuries with a history of acute trauma is needed.
In summary,
the finding that higher pathologic grades are associated with greater mean ROI values for AM and PL bundles suggests that T2* mapping reasonably reflects degree of bundle degeneration.
In addition,
for AM bundles,
operative and imaging grades were found to be linearly correlated with ROI values.
On the other hand,
relations between the grades obtained using imaging,
pathology and macroscopy in the operative field were highly variable,
despite the significance of the correlations between pathologic,
macroscopic and imaging grades for AM and PL bundles.
Furthermore,
for complete tears,
the sensitivity,
specificity and accuracy of oblique coronal T2WI and T2* mapping were high,
whereas for incomplete tears,
they were relatively low.
Therefore,
pathologic grades may differ from operative grades because pathologic grades represent degrees of bundle degeneration rather than the presence or absence of a tear.
However,
we found that T2* maps provide more accurate information regarding the degree of degeneration of ACL bundles.
In addition,
we found that the severity of degeneration by pathology does not reflect the completenesses of tears in the arthroscopic or operative field,
which indicates that elevated ROI values on T2* maps and increased signal intensities on T2W-images are not directly linked to tear status.
To differentiate partial and complete tears,
other imaging techniques should be combined.
However,
we suggest that the higher grades by pathology or an elevated ROI on a T2* map,
indicate a greater risk of a tear or a rapidly developing bundle tear.