Combined hip and knee strengthening compared to knee strengthening for individuals with lateral patellar dislocation: A single-blind, superiority, randomised controlled trial

Lucas Simões Arrebola, Toby O Smith, Vanessa Gonçalves Coutinho de Oliveira, Pedro Rizzi de Oliveira, Paloma Yan Lam Wun, Rogério Teixeira de Carvalho, Carlos Eduardo Pinfildi

Research output: Contribution to journalArticlepeer-review


Objective: To investigate whether a combined hip and knee muscle strengthening programme is superior to a knee strengthening programme for people following lateral patellar dislocation (LPD).  

Design: Single-blind, superiority, randomised controlled trial with 48 weeks follow-up.  

Setting: Physiotherapy out-patient clinic.  

Participants: Forty individuals aged 16 or older, with a history of non-traumatic LPD were randomised to a knee-based strengthening (KBSG) or quadriceps and hip strengthening exercise (HQSG) programme. Inclusion criteria included a positive apprehension sign, pain on palpation along the medial retinaculum and J sign. Exclusion criteria included restricted range of motion (<90° knee flexion), and traumatic or postsurgical LPD.  

Interventions: Concealed randomisation was performed using random permuted blocks of size 4. Individuals received their corresponding exercise programme according to randomisation and group allocation: knee-based strengthening (n=20) or combined hip and quadriceps strengthening (n=20) twice weekly for eight weeks over 16 appointments.  

Main Outcome Measures: Primary outcome was the Lysholm Knee Score. Secondary outcomes included: Numerical Pain Ratings Scale (NPRS) at rest and during effort, Norwich Patellar Instability Score (NPIS), Kujala Anterior Knee Pain Scale (AKPS), Lower Extremity Functional Scale (LEFS), four domains of the WHOQOL-Bref, and recurrence rate. Patient-reported outcome measures were assessed from the baseline to 48 weeks. Assessments were performed by a physiotherapist who was blinded to the group allocation. Data were analysed by using a repeated-measures ANOVA model with Tukey's post-hoc test following an intention-to-treat principle.  

Results: At the primary time-point of 8 weeks, there were no substantial between-group differences in the Lysholm Knee Score: mean difference= -6.8 (95% CI -14.3 to 3.7); NPIS: mean difference= 23.5 (95% CI 5.6 to 41.3); AKPS: mean difference= -1.54 (95% CI -8.6 to 5.6), NPRS at rest and during effort (mean difference= 0.32 (95% CI -0.37 to 1); and mean difference= 0.68 (95% CI -0.9 to 1.86); LEFS mean difference= -1.08 (95% CI -5.9 to 2.4), WHOQOL-Bref domains (physical health: mean difference= -0.12, (95% CI -1.26 to 1.02); psychological: mean difference= -0.32 (95% CI -2.04 to 1.4); social relationships: mean difference= -0.7 (95% CI -2.2 to 0.82); environment: mean difference= 0.44 (95% CI -1 to 1.9)), and recurrence rate (p=0.69).  

Conclusion: This study indicates that combined hip and knee muscle strengthening is not superior to knee-based strengthening for LPD treatment. The limitations stemming from the underpowered nature of the trial must be acknowledged, concerning the potential oversight of moderate intervention effects.
Original languageEnglish
Article number100334
JournalArchives of Rehabilitation Research and Clinical Translation
Publication statusE-pub ahead of print - 19 Mar 2024

Cite this