The authors reported that 23 studies were included in the review (n=2,495, range 30 to 242). Six studies included control groups and seven were consecutive series. Study quality was described as suboptimal, mainly due to poor reporting.
Relationships between clinical and MRI diagnoses and findings:
Overall there was no clear evidence of a relationship between clinical and MRI diagnoses (eight studies). Evidence of a relationship between clinical findings and MRI findings was inconsistent (15 studies).
Odds ratios and predictive values, using MRI as a criterion standard:
Among patients with symptomatic temporomandibular disorders, the odds ratio (OR) for disc displacement was 12.2 (one study). Among those with symptomatic temporomandibular disorders and pain, odds ratios ranged from 0.88 to 5.15 (three studies). Among patients with osteoarthrosis, odds ratios for clinical diagnosis and negative predictive values were both high (OR 5.29 and 10.26, negative predictive value 92% and 94%; two studies). Among patients with pain, odds ratios for joint effusion ranged from 1.22 to 1.93 (four studies). Among patients with clinical pain, the odds of internal derangement or osteoarthritis ranged from 1.36 to 2.04 (five studies). The odds ratio for disc displacement without reduction among patients with provoked pain was 4.82 (one study). Among patients with crepitation the odds of disc displacement without reduction were 3.71 (one study).
Other findings were reported in the review.