Cardiometabolic trade-offs of carbohydrate- and fat-focused diets: a network meta-analysis of randomised clinical trials

PURPOSE: This study compared the cardiometabolic effects of carbohydrate- and fat-focused dietary patterns in adults at increased cardiovascular disease (CVD) risk, with emphasis on trade-offs across lipid profiles, adiposity, blood pressure, and glycaemic outcomes. METHODS: A systematic literature search (January 2013-October 2025) identified randomised clinical trials enrolling adults aged 35-75 years with at least one CVD risk factor and evaluating dietary interventions with explicit macronutrient targets. Diets were classified based on fat and carbohydrate distribution, with monounsaturated fat content considered a marker of dietary quality and the low-fat-moderate-carbohydrate diet serving as the reference category. A random-effects frequentist network meta-analysis was conducted to estimate mean differences (with 95% confidence intervals) in CVD markers between dietary patterns. RESULTS: Forty-seven studies (n = 3,450) were included. High-fat-low-carbohydrate diets were associated with modest reductions in body mass index compared with low-fat-moderate-carbohydrate diets [- 0.48 kg/m (- 0.74; - 0.22)]. Ketogenic diets showed the largest increases in LDL cholesterol [0.43 mmol/L (0.13; 0.74)] total cholesterol [0.55 mmol/L (0.13; 0.96)] and HDL cholesterol [0.12 mmol/L (0.06; 0.18)], compared with the reference diet, whereas triglyceride responses differed, with high-carbohydrate diets increasing them [0.24 mmol/L (0.04; 0.43)]. Macronutrient composition had limited effects on waist circumference, blood pressure and glycaemic markers. CONCLUSION: Findings demonstrate distinct cardiometabolic trade-offs between carbohydrate- and fat-focused dietary patterns, whereby carbohydrate restriction improves triglycerides, HDL cholesterol, and adiposity but is not consistently favourable for LDL cholesterol. These findings should be interpreted within the context of broad macronutrient targets rather than macronutrient quality and food-based dietary patterns. No single macronutrient distribution is universally optimal and dietary interventions should consider baseline metabolic profiles and macronutrient quality and quantity.