EP1674 - ECE_1157 - Metabolic effects of sucralose below the acceptable daily intake: pooled evidence from 14 randomised trials

Abstract Background The non-nutritive sweetener, Sucralose, is widely used in sugar-reduction strategies but may not be metabolically inert. Objective To assess the association of sucralose with insulin resistance (HOMA-IR), insulin sensitivity (Matsuda index), and OGTT-stimulated insulin response (tAUC120), and to examine dose and exposure duration effects. Data Sources Embase, Cochrane, and Scopus (final update March 14, 2025), plus trial registries and reference lists. Study Selection Randomized human trials reporting pre/post data for ≥1 prespecified metabolic outcome with sucralose exposure. Data Extraction, Synthesis and Appraisal Dual independent extraction; random-effects meta-analysis with Hartung-Knapp adjustments; meta-regression for association with dose and duration; RoB2, funnel plots and Egger tests for bias assessment; GRADE assessment for certainty of evidence. Results Fourteen trials (n = 754) were included (n = 383 for insulin response; n = 300 for HOMA-IR; n = 215 for Matsuda). Sucralose was associated with reduced insulin sensitivity (Matsuda pooled mean difference = 0.52, 95% CI 0.08 to 0.95; I2 = 0%) and increased insulin resistance (HOMA-IR mean difference = -0.42, CI −0.70 to −0.14; I2 = 78.5%). Overall effects on insulin tAUC120 were imprecise, but longer exposure (≥30 days) was significantly associated with higher OGTT insulin response (tAUC120 MD = −1632; 95% CI −2807 to −458; I2 = 0%); meta-regression supported a duration–effect relationship. No dose–response gradient was detected; small-study and sensitivity analyses did not materially change conclusions Conclusions and Implications In non-diabetic adults, sucralose exposure increases fasting insulin resistance, reduces insulin sensitivity, and—after ≥30 days—raises OGTT insulin response. Effects occurred at typical-use, sub-ADI exposures, indicating metabolic activity below toxicology-based thresholds. Clinical counselling and clearer labelling (mg per serving/container) may be warranted; diabetes-specific, ≥12-week RCTs with weight-normalised dosing are needed