Abstract
Background:
Patella alta is a significant risk factor for patellofemoral instability (PFI). Correction of patella alta by distalization procedure is frequently recommended for PFI, so that the patella can engage into the trochlea during early flexion. Radiographic measurements for patellar height - Caton Deschamps Index (CDI) and Insall Salvati index (ISI), are commonly used indices to define patella alta. It is not known if these indices accurately define the relationship between patella and trochlea.
Hypothesis:
The purpose of the study was to inquire if the radiographic indices of patellar height represent patello-trochlear relationship.
Methods:
A deidentified radiographic database of 80 adolescent patients with (n=40) and without (n=40) PFI with knee radiographs and MRI performed prior to any surgical intervention was established, guided by a priori power analysis. Three fellowship-trained pediatric orthopaedic surgeons measured the CDI and ISI on radiographs. The patella-trochlear engagement on corresponding MRI was also measured using patella-trochlear index (PTI), which represents the overlap between the articular surface of patella and trochlea. Intraclass correlation coefficients (ICC) were calculated to assess interrater reliability for each measurement. Spearman rank correlation coefficients were calculated to determine correlation between CDI / ISI and PTI. Subgroup analysis was performed based on skeletal maturity status and for PFI vs controls.
Results:
The study cohort was predominantly female (59%) with a mean age of 14.0±3 years (36% skeletally immature). ISI and PTI had excellent reliability (ICC Range: 0.81-0.99) and CDI had very good to excellent reliability (ICC Range: 0.72-0.98). There was strong correlation between CDI and ISI (ρ=0.58, p=0.00), but there was no correlation between radiographic indices (CDI, ISI) and PTI on MRI (ρ=-0.02, p=0.8 and ρ=0.17, p=0.03 respectively), irrespective of skeletal maturity or diagnosis (PFI, controls). When evaluating patients with patella alta on radiographs using CDI>1.2 as cut-off value, only 1.7% and 4.2% had MRI evidence of insufficient patella-trochlear engagement, based on PTI cut-off values of <0.125 and <0.28, respectively. Similarly, using ISI>1.2 as cut-off value for patella alta, only 3% and 5.4% had MRI evidence of insufficient patella-trochlear engagement, based on PTI of < 0.125 and <0.28, respectively.
Conclusion:
Radiographic measurements (CDI, ISI) have acceptable reliability and can be used to screen PFI patients for patella alta. However, a cautious approach is recommended when these indices are used in isolation for surgical decision-making process to address patella alta, as they do not accurately represent patello-trochlear relationship.