Outcome Predictors for Conservative Patellofemoral Pain Management: A Systematic Review and Meta-Analysis

Background Patellofemoral pain (PFP) is highly prevalent within both sporting and recreationally active populations. Multiple treatment approaches have been advocated for the management of PFP, attempting to address both intrinsic and extrinsic factors thought to contribute to the development and persistence of pain. A number of predictors of treatment success have been proposed, and evaluated, for directing intervention choice. Objective Our aim was to systematically review the literature that identifies outcome predictors of specific conservative interventions in the management of PFP, including quality of the current evidence, to guide clinical practice and future studies investigating outcome predictors within this population. Data Sources The AMED, CINAHL, EMBASE, MEDLINE and Web of Science databases were searched from inception to April 2013. Study Selection Randomized controlled trials (RCTs) and cohort studies. Study Appraisal and Synthesis Methods Following initial searching, all potential papers were assessed by two independent reviewers for inclusion using a checklist developed from the inclusion criteria. Cited, and citing, references were also searched in Google Scholar, but unpublished work was not sought. Methodological quality was assessed using a previously designed quality assessment scale. Definitions for levels of evidence were guided by recommendations made by van Tulder et al. Results Fifteen low-quality (LQ) cohort studies were included. No RCTs were found. This systematic review identified the evaluation of 205 conservative management outcome predictor variables. Of this large number of variables that have been assessed, 19 (9 %) were found to significantly predict a successful outcome. Where two or more outcome predictors and success determinants were consistent between studies, data were pooled. Within these studies, the low number of participants per output variable, and absence of controls, is likely to compromise the validity of the predictor’s accuracy. Very limited evidence identified higher functional index questionnaire scores (mean 0.82, 95 % confidence interval [CI] 0.18–1.46), greater forefoot valgus (mean 0.67, 95 % CI 0.05–1.28) and greater rearfoot eversion magnitude peak (mean −0.93, 95 % CI −1.84 to −0.01) to significantly predict improved outcomes with orthoses interventions. Shorter symptom duration ( p = 0.019), lower frequency of pain ( p = 0.012), younger age, faster vastus medialis oblique reflex response ti