Could the ketogenic diet induce a shift in thyroid function and support a metabolic advantage in healthy participants? A pilot randomized-controlled-crossover trial

Background The ketogenic diet (KD) has been shown to result in body mass loss in people with disease as well as healthy people, yet the effect of the KD on thyroid function and metabolism are unknown. Objective We aimed to determine the effects of a KD, compared with an isocaloric high-carbohydrate low-fat (HCLF) diet, on resting metabolic rate and thyroid function in healthy individuals. Design Eleven healthy, normal-weight participants (mean(SD) age: 30(9) years) completed this randomized crossover-controlled study. For a minimum of three weeks on each, participants followed two isocaloric diets: a HCLF diet (55%carbohydrate, 20%fat, 25%protein) and a KD (15%carbohydrate, 60%fat, 25% protein), with a one-week washout period in-between. Importantly, while on the KD, the participants were required to remain in a state of nutritional ketosis for three consecutive weeks. Crossover analyses and linear mixed models were used to assess effect of diet on body mass, thyroid function and resting metabolic rate. Results Both dietary interventions resulted in significant body mass loss (p<0.05) however three weeks of sustained ketosis (KD) resulted in a greater loss of body mass (mean (95%CI): -2.9 (-3.5, -2.4) kg) than did three weeks on the HCLF diet (-0.4 (-1.0, 0.1) kg, p < 0.0001). Compared to pre-diet levels, the change in plasma T3 concentration was significantly different between the two diets (p = 0.003), such that plasma T3 concentration was significantly lower following the KD diet (4.1 (3.8, 4.4) pmol/L, p<0.0001) but not different following the HCLF diet (4.8 (4.5, 5.2) pmol/L, p = 0.171. There was a significant increase in T4 concentration from pre-diet levels following the KD diet (19.3 (17.8, 20.9) pmol/L, p < 0.0001), but not following the HCLF diet (17.3 (15.7, 18.8) pmol.L, p = 0.28). The magnitude of change in plasma T4 concentration was not different between the two diets (p = 0.4). There was no effect of diet on plasma thyroid stimulating hormone concentration (p = 0.27). There was a significantly greater T3:T4 ratio following the HCLF diet (0.41 (0.27, 0.55), p < 0.0001) compared to pre-diet levels but not following the KD diet (0.25 (0.12, 0.39), p = 0.80). Conclusions Although the diets were isocaloric and physical activity and resting metabolic rate remained constant, the participants lost more mass after the KD than after the HCLF diet. The observed significant changes in triiodothyronine concentration suggest that unknown metabolic changes occur in nutritional ketosis, changes that warrant further investigation.