Background/Purpose:

Clinical guidelines recommend that “every patient with knee osteoarthritis should receive advice concerning appropriate footwear”, yet the recommended content of this advice is not specified. Some studies suggest that highly flexible shoes can protect the tibiofemoral (TF) joints against excessive load during gait, while other studies underscore the importance of stable or supportive shoes as a means of protecting the patellofemoral (PF) joint against maltracking brought about by foot pronation. The purpose of this observational study was to determine the relationship between the stability characteristics of a person’s usual walking shoe and the 2-year risk of worsening cartilage damage in the medial TF, lateral TF, and PF knee compartments.

Methods:

The Multicenter Osteoarthritis Study (MOST) includes middle aged and older adults that have or are at risk of knee OA. Subjects were asked to bring their usual walking shoes and, adapting the methods of Barton et al., examiners at the 60-month visit scored the sagittal, torsional, and heel counter stability of each subject’s shoe as 0= flexible or 1= stable / supportive (kappa ≥ 0.69). A Summative Shoe Stability Score (0-3) was calculated as the sum of the three component test scores. 1.0T MRIs were obtained at the 60 and 84-month exams, and one knee per subject was scored using Whole Organ MRI Scores (WORMS) to indicate the extent of cartilage damage (0-6) in each of 5 sub-regions of the medial and lateral TF compartments, and each of 4 sub-regions of the PF compartment (kappa ≥ 0.63). Using separate logistic regression models for each compartment, we estimated the relative odds of worsening knee cartilage damage in categories of increasing shoe stability, while adjusting for covariates. Generalized estimating equations accounted for non-independence between sub-regions of a compartment.

Results:

1126 subjects (mean +/- sd age 66.8 +/- 7.5 yrs,  BMI 29.6 +/- 4.8 kg/m²; 61.7% female, 89.8% white) contributed 1124, 1123, and 1116 knees to the analysis of cartilage damage in the medial TF, lateral TF, and PF compartments, respectively. A majority of shoes (64.3%, 68.3%, and 55.0%, respectively) were scored as stable / supportive during sagittal, torsional, and heel counter stability tests, with 47.8% of shoes obtaining the maximum Summative Shoe Stability Score of 3 and only 25.5% obtaining the minimum score of 0. Relative odds of worsening cartilage damage in the medial TF, lateral TF, and PF knee compartments did not change across categories of increasing shoe stability (p > 0.05 for all comparisons).

Conclusion:

These observational findings do not confirm an association between usual walking shoe stability and 2-year risk of worsening cartilage damage in medial TF, lateral TF, or PF knee compartments. Future studies are needed to clarify the shoe characteristics that are most relevant for persons with knee OA.