Non-linear associations between protein intake and lean and femur bone mass in individuals with rheumatic diseases: findings from the NHANES 2007–2018

<h2>Abstract</h2><h3>Background</h3> Individuals with rheumatic diseases are at higher risk of muscle and bone mass loss, and increased protein intake could reduce this risk. We investigated associations between protein intake and lean and bone mass in individuals with rheumatic diseases from the NHANES. <h3>Methods</h3> A sample of 3,972 individuals with psoriatic arthritis, rheumatoid arthritis, osteoarthritis and gout was obtained. Outcomes were lean and bone mass, measured at the whole-body, femoral neck and lumbar spine by DXA, with the exposure being daily protein intake in absolute and relative to body mass values. Generalized linear models with linear and non-linear associations adjusted for confounding variables were used. Directed acyclic graphs aided in visualizing variables and determining an adjustment set. <h3>Results</h3> There was a linear association between absolute protein intake (g/day) and lean mass (1.54 kg lean mass per 50 g/day protein [95% CI 0.78 to 2.30], <i>p</i><0.001) and a non-linear association for relative protein intake (g/kgBM/day, <i>p</i><0.001), with intakes higher than ∼1.6 g/kgBM/day indicating no further increase in lean mass. Non-linear associations between absolute (<i>p</i><0.001) and relative (<i>p</i>=0.005) protein intake and femoral neck BMD were shown, with intakes higher than ∼150 g/day or ∼1.2 g/kgBM/day suggesting no further increase in femoral neck BMD. No significant associations were shown for whole-body and spine BMD. <h3>Conclusion</h3> Protein intakes higher than 0.8 g/kgBM/day may have a relevant role in improving lean and bone mass in individuals with RD. Assessing and ensuring an adequate protein should be a nutritional goal in individuals with RD.