The quadriceps active ratio: a dynamic MRI-based assessment of patellar height.

Abstract

Patella alta (PA) is one of the primary correctable risk factors for patellofemoral instability (PFI). Both an accurate diagnosis of PA and a clinically relevant target for correction are necessary for optimal treatment. An ideal test for PA should relate the position of the patella to the femur rather than tibia, should do so with the quadriceps contracted and the patellar tendon under tension and should have good sensitivity and specificity. None of the currently used radiographic tests PA meet these criteria, most of which are based on the position of the patella relative to the tibia with diagnostic cutoffs based on 2 standard deviations from the mean rather than optimal sensitivity and specificity. The authors describe the quadriceps active ratio (Q+R), an MRI-based assessment of PA based on patellofemoral contact under quadriceps activated with a cutoff based on optimal sensitivity a specificity for PFI. Ninety-four participants investigated for knee pain or instability with a clinically indicated MRI were recruited. Routine MRI sequences were obtained, with the addition of a quadriceps contracted sagittal T1-weighted sequence. Participants presenting with PFI were identified. Those with trochlear dysplasia were identified and excluded from analysis so that patellar height could be assessed against PFI without being confounded by trochlear dysplasia. Q+R and patellotrochlear index (PTI) were calculated from the remaining 78 scans by 3 consultant orthopaedic surgeons at three time points. In 54 of these cases, a lateral radiograph was available from which the Insall-Salvati, modified Insall-Salvati, Caton-Deschamps and Blackburn-Peel ratios were also calculated. Intra- and inter-observer reliability was assessed for the Q+R. A cutoff value for the Q+R based on optimal sensitivity and specificity for the diagnosis of PFI was calculated from receiver-operator characteristic (ROC) curves and compared to the PTI. The cutoff for the Q+R was compared for sensitivity and specificity for the diagnosis of PFI against the radiographic ratios. The Q+R had satisfactory or better ICC values across time points and surgeons. The Q+R was superior to the PTI on area under curve ROC analysis (0.76 vs 0.74). A cutoff value of 0.12 for the Q+R gave sensitivity of 79% and specificity of 55% for the diagnosis of PFI. The radiographic indices were generally insensitive for this diagnosis of PFI with sensitivities ranging from 0-66%. TheQ+R is a reliable diagnostic test for patellar height assessment, showing good intra- and inter-rater consistency, and greater diagnostic accuracy than the PTI. A Q+R value of 0.12 is a good test for clinically significant PA. Of the radiographic indices, the Insall-Salvati ratio had the best diagnostic accuracy.