The diagnosis of a meniscal tear may require MRI, which is costly. Ultrasonography has been used to image the meniscus, but there are no reliable data on its accuracy. We performed a prospective study investigating the sensitivity and specificity of ultrasonography in comparison with MRI; the final outcome was determined at arthroscopy. The study included 35 patients with a mean age of 47 years (14 to 73).

There was a sensitivity of 86.4% (95% confidence interval (CI) 75 to 97.7), a specificity of 69.2% (95% CI 53.7 to 84.7), a positive predictive value of 82.6% (95% CI 70 to 95.2) and a negative predictive value of 75% (95% CI 60.7 to 81.1) for ultrasonography. This compared favourably with a sensitivity of 86.4% (95% CI 75 to 97.7), a specificity of 100.0%, a positive predictive value of 100.0% and a negative predictive value of 81.3% (95% CI 74.7 to 87.9) for MRI.

Given that the sensitivity matched that of MRI we feel that ultrasonography can reasonably be applied to confirm the clinical diagnosis before undertaking arthroscopy. However, the lower specificity suggests that there is still a need to improve the technique to reduce the number of false-positive diagnoses and thus to avoid unnecessary arthroscopy.

Total knee replacement and high tibial osteotomy are common orthopaedic operations with low complication rates. Such surgery is in close proximity to the popliteal artery (PA), the behaviour of which during flexion of the knee is poorly understood.

We used Duplex ultrasonography to determine the distance of the PA from the posterior tibial surface at 0° and 90° of flexion in 100 knees. When the knee was flexed the PA was closer to the posterior tibial surface at 1 to 1.5 cm below the joint line in 24% and at 1.5 to 2 cm below the joint line in 15%. There was a high branching anterior tibal artery in 6% of knees. We provide an anatomical account to help to explain our findings by using cadaver dissections, arteriography and static MRI studies.