Is the effectiveness of patellofemoral bracing modified by patellofemoral alignment and trochlear morphology?

This study was performed to determine if the effectiveness of patellofemoral bracing as a treatment for patellofemoral osteoarthritis is influenced by patellofemoral joint alignment and trochlear morphology. We hypothesized that those with more extreme patellar malalignment would benefit more from bracing.

Thirty-eight patients who had received bracing as part of a comprehensive treatment plan for patellofemoral osteoarthritis were selected for this study. Ten measures of patellar alignment were taken from X-rays. These alignment measures were divided into percentile groups (tertiles) for contingency table analysis. Treatment outcome was measured by Western Ontario and Macmasters Universities Osteoarthritis Index (WOMAC) scores and these were dichotomised into two groups according to "Improved" or "Not Improved" according to the minimum clinically important difference (MCID). Spearman's rho test was performed for continuous variables and Fisher's exact test was performed for correlation between tertile groups and MCID categories.

Thirty-eight patients (9 male and 29 female) between the ages of 51 to 89 were included in this study. WOMAC scores ranged from -25 to 41.67, with a mean change of -3.97, 31.6, 44.7 and 31.6% of patients falling into the "Improved" group for Global, Pain and Function scores respectively. We found a non-significant trend shown (p = 0.058, correlation coefficient 0.31) between bisect offset and change in WOMAC global, indicating a trend for higher change in WOMAC scores with increasing bisect offset. Statistically significant correlations were found between mean MCID categories for the WOMAC global and function groups when analysed against percentile groups for bisect offset (p < 0.01) and patellar subluxation distance (p < 0.05), indicating those in higher percentile groups were more likely not to improve after six months.

Higher bisect offset and patellar subluxation distance measures were associated with poorer outcomes. However, due to the limited sample size, more studies are required to fully examine this relationship.