pipette
ESEspañol

Effects of saffron and crocin on glycemic control: an exploratory systematic review and meta-analysis of randomized trials

J. Kim, M. Schaffer, G. Kim, B. A. Nguyen, P. Batra, A. T. Nguyen, P. Potluri, K. Kim

PreprintReal-world use

In the authors' words

Background: Saffron and its constituent crocin have been investigated as adjuncts for metabolic dysfunction. Their effects on longer-term glycemic exposure and fasting indices may differ. Objectives: To assess the effects of saffron and its constituents on glycated hemoglobin (HbA1c), fasting glucose, and insulin resistance. Search methods: We searched PubMed Central on June 11, 2025, and PubMed on October 5, 2025. Selection criteria: Randomized controlled trials in humans comparing affron or its constituents with a control for at least four weeks and reporting a relevant glycemic outcome were eligible. Comparisons were selected to isolate the effect of saffron or crocin. Data collection and analysis: We synthesized continuous outcomes as mean differences (MDs) with 95% confidence intervals (CIs), using inverse-variance random-effects models. HbA1c was the primary outcome. Fasting glucose and the homeostatic model assessment of insulin resistance (HOMA-IR) were secondary outcomes. Main results: Twelve trials contributed 570 participants to the analyzed comparisons. Populations included adults with type 1 or type 2 diabetes, metabolic syndrome or coronary artery disease, and adolescents with obesity and prediabetes. Seven trials (391 participants) reported HbA1c: the pooled MD was -0.25 percentage points (95% CI -0.58 to 0.08; I2 = 61%). Fasting glucose was lower with saffron or crocin across 12 trials (570 participants; MD -16.16 mg/dL, 95% CI -27.02 to -5.31; I2 = 74%). Five trials (212 participants) contributed to HOMA-IR (MD -1.19, 95% CI -2.25 to -0.12; I2 = 80%). Fixed-effect sensitivity analyses gave smaller estimates across the three outcomes. Authors' conclusions: Saffron and crocin were associated with lower fasting glucose and HOMA-IR, whereas the HbA1c estimate remained compatible with both a reduction and little difference. Variation in populations, preparations, and trial sizes limits interpretation of the average effects. Larger trials with standardized preparations and adequate follow-up are needed to establish the persistence and clinical relevance of these findings. Funding: None

Main resultLimitation the authors admit

Appeared: Tuesday, November 3. medRxiv. Preprint, not yet peer-reviewed.

DOI: 10.64898/2026.10.30.26366889