The Effect of Low-Carbohydrate and Low-Fat Diets on Pain in Individuals with Knee Osteoarthritis

Strath LJ, Jones CD, Philip George A, Lukens SL, Morrison SA, Soleymani T, Locher JL, Gower BA, Sorge RE · Pain Medicine

Three arms, two of them dietary, and only one moved. Within the low-carbohydrate arm, pain interference, knee quality of life, evoked pain intensity on temporal summation and chair stands, serum oxidative stress (TBARS) and leptin all fell significantly from baseline, and the low-fat and control arms showed no such within-arm changes. Across arms, only oxidative stress showed a time-by-diet interaction (p = 0.002), and the post-hoc tests against control did not reach significance for any pain measure. Both diet arms lost weight, 8.9 and 6.7 kg, with no significant difference between them. The fall in oxidative stress tracked the fall in evoked pain (r² = 0.23). The low-fat arm, alongside the control, means the comparison is not only a test of dietary attention.

Population

Adults aged 65-75 with knee osteoarthritis; 24 randomised, 21 completed (8 low-carbohydrate, 6 low-fat, 7 control)

Intervention

A low-carbohydrate diet (20 g a day, rising to 40 g if needed) for 12 weeks

Comparator

A calorie-reduced low-fat diet as a weight-loss control, or eating as usual

Limitations

Described by its authors as a pilot: 21 completers at one centre over 12 weeks, with self-reported and evoked pain as outcomes and no structural measure. The pain and leptin results are within-arm changes — no between-arm pain difference was significant on the post-hoc tests — and the oxidative-stress fall was driven mainly by the women. Diet trials cannot be blinded. Whether weight loss accounts for the effect is not resolved; the authors argue not, because both diet arms lost similar weight. NIH career award and start-up funds; no conflicts.