A retrospective study comparing ultrasound and MRI for diagnosing quadriceps and patellar tendon ruptures, along with...

Accurate differentiation between partial and complete knee extensor tendon ruptures is crucial, generally accomplished with MRI or USS. Few studies provide detailed statistical analyses beyond basic sensitivity and specificity calculations when comparing USS and MRI. This study aims to compare the two modalities and perform a cost-benefit analysis.

We conducted a single-centre retrospective cohort study, identifying cases of surgical QT or PT repair from 2016 to 2022. The sample was divided into cohorts based on intraoperative findings and radiology reports. In addition to sensitivity and specificity tests, Cohen's Kappa (κ) statistic is used to determine the degree of agreement between imaging modalities and intraoperative findings, and is interpreted using the Landis & Koch method. Effect size was also calculated using the Phi coefficient (φ), and cost analysis was based on NHS tariffs.

Ninety-three patients were analysed: 49 underwent USS and 37 underwent MRI, with 7 receiving both. MRI was found to be the superior imaging modality, showing "almost perfect" agreement according to the Landis and Koch interpretation (κ = 0.812, p < .001), with a large effect size suggesting that MRI closely matches intraoperative findings (φ = 0.814, p < .001). In contrast, USS was found to have "slight agreement" (K = 0.132, p = .301) and to be weak-to-moderately associated. There were five false positives in the USS cohort and one in the MRI cohort, costing £7510 and £1,502, respectively. However, an additional cost of £2880 would be incurred if all patients underwent an MRI.

MRI is superior in characterizing extensor tendon ruptures in the knee. Despite higher costs, its accuracy minimises misdiagnoses and optimises the use of surgical resources.