The influence of sex and obesity on gait biomechanics in people with severe knee osteoarthritis scheduled for arthroplasty

Highlights

• Sex and obesity may affect biomechanics in severe knee osteoarthritis.

• We compared sexes, and class I obese and overweight with severe knee osteoarthritis.

• Men had a higher knee adduction moment after adjusting for body size.

• Further adjustment for the more varus knee of men removed this sex-difference.

• There were no differences in knee biomechanics between obesity groups.

Abstract

Background

Sex and body mass may influence knee biomechanics associated with poor total knee arthroplasty (TKA) outcomes for knee osteoarthritis (OA). This study aimed to determine if gait differed between men and women, and overweight and class I obese patients with severe knee OA awaiting TKA.

Methods

34 patients with severe knee OA (average age 70.0 (SD 7.2) years, body mass index 30.3 (4.1 kg/m²)) were recruited from a TKA waiting list. Three-dimensional gait analysis was performed at self-selected walking speed. Comparisons were made between men and women, and overweight (body mass index (BMI) 25.0–29.9 kg/m²) and class I obese (BMI 30.0–34.9 kg/m²) participants. Biomechanical outcomes included absolute and body size-adjusted peak knee adduction moment (KAM), KAM impulse, peak knee flexion moment, as well as peak knee flexion and varus-valgus angles, peak varus-valgus thrust, and peak vertical ground reaction force (GRF).

Findings

Men had a higher absolute peak KAM, KAM impulse and peak GRF compared to women, and this sex-difference in frontal plane moments remained after adjusting for body size. However, when additionally adjusting for static knee alignment, differences disappeared. Knee biomechanics were similar between obesity groups after adjusting for the greater body weight of those with class I obesity.

Interpretation

Men had greater KAM and KAM impulse even after adjustment for body size; however adjustment for their more varus knees removed this difference. Obesity group did not influence knee joint kinematics or moments. This suggests sex- and obesity-differences in these variables may not be associated with TKA outcomes.

Introduction

Knee osteoarthritis (OA) is a leading cause of disability amongst older adults (Cross et al., 2014). It is a progressive disease and total knee arthroplasty (TKA) for end-stage OA is common (Australian Orthopaedic Association, 2015). Despite overall improvements in pain and quality of life (Silva et al., 2014), poor outcomes are common, with between 10 and 35% of patients reporting moderate to severe chronic post-surgical pain (Beswick et al., 2012) and up to 20% reporting dissatisfaction with surgery (Williams et al., 2013). Abnormal pre-operative gait biomechanics, such as altered frontal plane knee loading patterns (Astephen Wilson et al., 2010), greater sagittal plane knee joint loading (Smith et al., 2004, van Jonbergen et al., 2014), and decreased maximum knee flexion and knee flexion range of motion (Turcot et al., 2013), have all been associated with poor outcomes following TKA. Furthermore, female sex and obesity have been shown to adversely affect some gait biomechanical variables (Kumar et al., 2015, Messier et al., 1996) and TKA outcomes (Dowsey et al., 2010, Mehta et al., 2015). This is relevant because the majority of patients with OA are female (Pereira et al., 2011), and those proceeding to TKA are also predominantly female (68%), obese (60%), or both female and obese (44%) (Dowsey et al., 2010). Consequently, a more comprehensive understanding of the influence of sex and obesity on gait biomechanics in those with severe knee OA awaiting TKA is important in identifying factors that might influence poor post-surgical outcomes.

Limited research has examined differences in knee biomechanics during gait between sexes in people with knee OA. Women with mild to moderate knee OA are reported to walk with lower relative loading across the medial tibiofemoral joint compartment (shown by a lower peak size-normalized external knee adduction moment; KAM) (Kumar et al., 2015, Sims et al., 2009) and either similar (Phinyomark et al., 2016) or greater peak knee flexion (Kaufman et al., 2001), when compared to men. Regarding severe OA, it has been reported that women awaiting TKA had a lower knee joint flexion range of motion and lower normalized peak KAM than men (Astephen Wilson et al., 2015). Collectively, these few studies suggest that women with OA experience a lower normalized KAM, but less clear sagittal plane range of motion differences, compared to men. However, this may be counter-intuitive given a higher normalized KAM has been associated with worse knee OA symptoms and joint structural decline (Bennell et al., 2011, Miyazaki et al., 2002), which are also related to progression to arthroplasty (Conaghan et al., 2010). Thus, given the higher rate of female patients with OA, and proceeding to TKA, as well as the association between female sex and poor outcomes following TKA, further research is needed to clarify sex differences in knee biomechanics such as the KAM in those awaiting TKA.

In addition to sex, obesity may also influence gait patterns however few studies have examined this relationship in individuals with knee OA, possibly due to difficulties in 3D gait analysis of the obese. Of the limited research to date, obese individuals (defined as having a BMI ≥ 30 kg/m²) with mild radiographic knee OA have been shown to walk with higher absolute peak vertical ground reaction force (Messier et al., 1996), lower normalized external knee flexion moments (KFM) (Kaufman et al., 2001), and a more extended knee (Messier et al., 1996) than non-obese individuals with similar severity of knee OA. Thus there may be differences between obese and non-obese knee OA patients in knee and other load-related biomechanical variables that have been associated with poor outcomes following TKA, including sagittal plane knee joint loading (Smith et al., 2004, van Jonbergen et al., 2014) and range of motion (Turcot et al., 2013). While this information is useful, a comparison between obese and overweight knee OA patients (rather than obese and non-obese) may be more relevant given most knee OA patients are either overweight or obese (Marks, 2007). Due to the paucity of studies comparing overweight and obese knee OA patients, further research examining differences in knee kinematics and kinetics between these groups is needed.

Furthermore, varus-valgus thrust identified during visual observation has been related to knee pain (Lo et al., 2012) and OA progression (Chang et al., 2004a), and also recently associated with knee kinematics relevant to knee OA (Chang et al., 2013). However, thrust quantified using 3-dimensional (3D) gait analysis has only been examined in severe OA in one study (Sosdian et al., 2016), and the effects of sex and obesity upon this parameter are unknown. Specifically, research has identified that knee OA patients with a varus thrust, quantified using 3D gait analysis, have a more varus static knee alignment, and walk with a greater knee varus angle, normalized peak KAM and KAM impulse, compared to knee OA patients with a valgus thrust. In addition, sex might also influence thrust as males without joint pathology have a more varus static alignment when compared to females (Bellemans et al., 2012), and static knee alignment (Chang et al., 2004b) and knee varus-valgus angles (Chang et al., 2013) have both been related to varus-valgus thrust. To date, the effects of obesity upon varus thrust in knee OA are unknown.

The purpose of this study was to examine sex- and obesity-related differences in knee biomechanics relevant to knee arthroplasty in a group of people with severe knee OA. It was hypothesised that female OA patients, and those with class I obesity, would walk with altered knee and other biomechanics compared to males and those classified as overweight.