Background: Little is known on the long-term association between low-carbohydrate diets (LCD) and mortality.

Purpose: We aimed to evaluate the link between LCD with total and cause-specific mortality by applying on both individual data and pooling prospective studies.

Methods: Data from National Health and Nutrition Examination Survey (NHANES) (1999–2010) were collected. We used adjusted Cox regression to determine the risk ratio (RR) and 95% confidence interval (95% CI), as well as random effects models and generic inverse variance methods to synthesize quantitative and pooling data, followed by a leave-one-out method for sensitivity analysis.

Results: Based on the data from NHANES with 24825 participants (mean age of 47.6 years, comprising 48.6% men and 51.4% women), after adjustment, participants in the top quartile (Q4) of LCD had the highest risk of total (32%; hazard ratio [HR] 1.32 [1.14–2.01], p<0.001), cardiovascular (CVD) (50%; 1.50 [1.12–2.31], p<0.001), cerebrovascular (51%; 1.51 [1.19–1.91], p<0.001) and cancer (36%; 1.36 [1.09–1.83], p<0.001) mortality. In the same model, the association between LCD and total mortality was stronger in the non-obese (0.48%) than in the obese (19%) participants. Findings based on the meta-analysis of 7 prospective cohorts with 447,506 participants and 39,326 mortality cases indicted a positive association between LCD and total (RR: 1.15, 95% CI: 1.07–1.22, p<0.001, I2=8.6) (figure 1A), CVD (RR: 1.13, 95% CI: 1.02–1.24, p<0.001, I2=11.2) (figure 1B) and cancer mortality (RR: 1.07, 95% CI: 1.01–1.14, p=0.02, I2=10.3). These findings were robust in sensitivity analyses.

Conclusions: Our study highlighted the unfavorable effect of LCD on total and cause-specific mortality, based on both individual data and by pooling previous cohort studies. Given the fact that LCDs may be unsafe, it would be preferable not to currently recommend these diets. Further studies to clarify the mechanisms involved in these associations and to support our findings are eagerly awaited.