Questions: Is magnetic resonance imaging (MRI) of the knee accurate
in detecting tears of the menisci and cruciate ligaments? Do study design
characteristics and magnetic field strength affect diagnostic performance?
Data sources: Studies were identified through a search of MEDLINE
(1991 to 2000) with use of the search terms "magnetic resonance
imaging," "knee," "meniscus," "cruciate
ligament," and "arthroscopy."
Study selection: Studies included were those in the English-language
literature that compared MRI with arthroscopy for detecting lesions of the
medial or lateral meniscus, anterior cruciate ligament (ACL), or posterior
cruciate ligament (PCL) in =30 patients; defined positivity criteria for
MRI; and reported magnetic field strength. Studies were excluded if patients
were infants or adolescents, MRI was used for postoperative evaluation, the
study design was case-control, results were reported only in terms of the
medial and lateral meniscus combined, various magnetic strengths were used, or
only specific features and indirect signs of knee lesions were assessed at the
time of MRI.
Data extraction: Two authors made use of standardized data
extraction sheets to extract data independently on study characteristics,
patient characteristics, magnetic field strength, number and type of MRI
sequences, and results.
Main results: 29 studies (27 on both menisci, 23 on ACL tears, and
12 on PCL tears) were included. For meniscal tears, sensitivity was higher for
the medial meniscus than for the lateral meniscus; specificity was higher for
the lateral meniscus than for the medial meniscus (Table). Sensitivity for ACL
tears did not differ between ACL and PCL ligaments; specificity was higher for
PCL tears than for ACL tears (Table). In a multivariate analysis of summary
receiver operating characteristics for all lesions combined, mean age and
magnetic field strength were significant predictors of diagnostic performance.
In sensitivity analyses, the magnetic field strength remained a consistent,
significant, modest predictor of diagnostic performance (odds ratio range,
1.80 to 2.16).
Conclusions: Magnetic resonance imaging has a higher sensitivity for
the detection of tears of the medial meniscus than for tears of the lateral
meniscus. Specificity is higher for tears of the lateral meniscus than for
tears of the medial meniscus and for anterior cruciate ligament tears than for
posterior cruciate ligament tears. A higher magnetic strength field modestly
improves diagnostic performance.
This meta-analysis by Oei and colleagues combines 29 studies that evaluate
the validity of MRI with respect to meniscal and cruciate ligament disorders
of the knee. All studies used arthroscopy as the gold standard, and the
authors found that MRI had excellent sensitivity and specificity for these
diagnoses overall.
The mean age of the patients in the studies ranged from 25 to 40 years, and
therefore the results are most applicable to this age group. The review did
not evaluate the usefulness of MRI in the detection of articular cartilage
disorders, which has gained importance with the advent of new pulse
sequences1.
Although sample size was limited, increased magnetic field strength (which
ranged from 0.1 to 1.5 T) improved the diagnostic accuracy overall. The
minimal field strength for acceptable diagnostic accuracy was not determined
by this review.
On the basis of this research, the MRI findings for patients in this age
group with a meniscal or cruciate ligament disorder should, in general, be
consistent with the diagnosis. The role of diagnostic arthroscopy of patients
with negative results on MRI (e.g., a patient with a clear history of repeated
mechanical locking) is still undefined. In general, when MRI findings are
negative, arthroscopy of the knee should be undertaken with the understanding
that there is a high probability that the structures will be found to be
normal at the time of surgery.
Potter HG, Linklater JM, Allen AA,
Hannafin JA, Haas SB. Magnetic resonance imaging of articular cartilage in
the knee. An evaluation with use of fast-spin-echo imaging. J Bone
Joint Surg Am.1998;80:
1276-84.801276
1998