Improving Muscle Function Through a Multimodal Behavioural Intervention for Knee Osteoarthritis and Obesity: The POMELO Trial

Abstract

Background: Treatments aimed at improving physical function and body composition, including reducing fat mass (FM) and increasing muscle mass, may benefit individuals with advanced knee osteoarthritis (OA) and obesity. We investigated the feasibility and efficacy of a multimodal behavioural intervention compared to usual care to enhance physical function and muscle mass in this population.

Methods: The POMELO (Prevention Of MusclE Loss in Osteoarthritis) study is a two-arm pilot randomized controlled trial; NCT05026385. Participants aged 40-75 years, with a BMI ≥ 35 kg/m² and knee OA were randomized 1:1 to either the intervention group (POMELO) or usual care (UC). The 3-month POMELO intervention incorporated progressive resistance exercise (3 sessions/week), individualized nutrition counselling targeted for OA, and 12 group education sessions on nutrition and arthritis self-management. The UC group received standard clinical care. After the 3-month supervised intervention, both groups were followed for 6 months without support. Assessments at baseline, 3 months and 9 months (primary endpoint) included body composition (DXA, measuring FM and appendicular lean soft tissue [ALST]), physical function (chair-sit-to-stands [CSTS], 6-min walk [6MWT], maximal handgrip strength [HGS]), and health-related quality of life (Euroqol visual analog scale [EQ-5D VAS]). Co-primary outcomes were feasibility (intervention completion ≥ 80% and per-protocol adherence ≥ 60% [i.e., attendance at 12 education sessions and exercise 3 ×/week]) and acceptability (4-item Likert-scale satisfaction survey, and open-ended questions). Secondary outcomes included changes in physical function and ALST.

Results: Fifty participants were randomized (POMELO = 25, UC = 25), with 32 completing the study (69% female, mean age 64.9 ± 1.2 years, BMI 42.1 ± 1.0 kg/m²). The POMELO intervention group had 80% completion and 74% adherence, confirming feasibility. Higher satisfaction rates were observed in POMELO compared to UC (3.5 vs. 2.2, p < 0.001) indicating greater acceptability. The POMELO group had improvements in CSTS (mean difference [MD] 3.96, ES 1.2, p < 0.001), 6MWT (MD 31.6 m, ES 0.4, p = 0.039) and EQ-5D VAS (MD 7.9 points, ES = 0.4, p = 0.01) compared to UC. Both groups experienced FM loss, but only the UC group lost ALST and HGS.

Conclusion: The POMELO intervention, combining personalized nutrition, resistance exercise and self-management support, was feasible, acceptable and showed greater efficacy than usual care to improve physical function in patients with knee OA and obesity. Our pilot study of this intervention showed potential benefits on body composition and quality of life without focusing on weight reduction. A larger study is needed to confirm these results, as this approach may offer advantages over usual care, potentially leading to better mobility and health outcomes.

Keywords: body composition; exercise; intervention; knee osteoarthritis; multimodal intervention; nutrition intervention; obesity; physical function; sarcopenic obesity.

FIGURE 1

CONSORT flow diagram of study participants with advanced knee osteoarthritis and a body mass index ≥ 35 kg/m². DXA = dual‐energy x‐ray absorptiometry, POMELO = intervention group, UC = usual care group (control).

FIGURE 2

Mean changes in secondary and select exploratory outcomes (2.a‐c, physical function and strength, 2.d. appendicular lean soft tissue, 2.e. fat mass and 2.f., health‐related quality of life) in the POMELO intervention and usual care groups after the 9‐month study period.