Your Patella Dislocated: Will It Happen Again? Key Magnetic Resonance Imaging Differences Between Patients With...

Identifying anatomic risk factors for recurrent patellar dislocations can help guide clinical decision making and counseling on operative management for patients at the time of the initial dislocation.

To determine if there are significant differences in anatomic factors between patients with a one-time patellar dislocation (OTPD) and those with multiple patellar dislocations (MPDs).

Case-control study; Level of evidence, 3.

Patients aged 9 to 21 years who had at least one patellar dislocation between 2012 and 2023 were retrospectively categorized into 1 of 2 groups (verified by telephone communication with at least 2 years' follow-up): MPDs (≥2 patellar dislocations) and OTPD. Patient characteristics and several magnetic resonance imaging measurements, including proximal tibial tubercle-trochlear groove (pTT-TG) distance, Caton-Deschamps index (CDI), relative tibial external rotation (rTER), and patellar tilt (PT), were collected. Trochlear dysplasia was also assessed using the 2-image lateral trochlear inclination (LTI), sulcus angle (SA), and trochlear depth (TD). Multiple logistic regression was used to determine the association of anatomic risk factors with the likelihood of experiencing MPDs.

In total, 177 knees (104 female, 73 male) were included in analyses: 108 had MPDs, and 69 had an OTPD. Patients with MPDs had a significantly less TD (P < .001), greater SA (P < .001), greater CDI (P = .03), and greater PT (P = .001). LTI (P = .65), rTER (P = .72), and pTT-TG distance (P = .12) were not strongly associated with having MPDs. A multiple logistic regression model of SA and CDI found an area under the curve of 0.72 for having MPDs; a patient with an SA of 170° and a CDI of 1.5 had a .78 probability of experiencing MPDs.

Measures of trochlear dysplasia (SA and TD), patella alta (CDI), and PT may be useful to distinguish between patients with MPDs and those with an OTPD. pTT-TG distance, LTI, and rTER were not found to have statistically significant differences between the groups. The combination of SA and CDI distinguished between the groups with reasonable accuracy (area under the curve = 0.72). Applying this model, a patient with an SA of 170° and a CDI of 1.5 would have a .78 probability of experiencing MPDs. This may be helpful in counseling patients after an initial dislocation event on the likelihood of experiencing MPDs.