Compléments alimentaires, confrontés aux preuves

Cinnamon

Solidité des preuves: Non étayé

Mis à jour 2026-08-07

Ce guide n’est pas encore traduit dans votre langue ; il est affiché en anglais.

Réponse courte

Does cinnamon lower blood sugar?

Not reliably. A Cochrane review of ten randomised trials in 577 people concluded there is insufficient evidence to support cinnamon for diabetes, with no significant effect on HbA1c, insulin or post-meal glucose. A later meta-analysis found a fasting glucose reduction but again no HbA1c effect, with very high inconsistency between trials.

Avant d’en prendre

Parlez-en à votre prescripteur avant d’ajouter ceci à un médicament, en particulier hypoglycémiant. N’arrêtez jamais un traitement prescrit au profit d’un complément.

Interactions

  • Cassia cinnamon — the cheap supermarket kind, and what most capsules contain — is high in coumarin, which is hepatotoxic at sustained intake. If you take anything else that burdens the liver, this is additive.

    Prudence

  • Any glucose-lowering effect is additive with glucose-lowering medication. The effect here is small and inconsistent, but the arithmetic still applies.2

    Prudence

Qui ne doit pas en prendre

  • Avoid supplement-dose cassia cinnamon if you have liver disease. Culinary quantities are a different matter from a daily capsule taken for months.

    À éviter

  • Supplement doses are not advised in pregnancy; safety at those quantities has not been established.

    Prudence

Notre verdict

We can find no good reason to buy this. The systematic review evidence is a null, the positive findings do not survive to the measure that matters, and cassia cinnamon carries a liver caution the marketing never mentions.

Ce que nous pouvons dire et vendre chez vous

Les allégations de santé pour ces produits ne sont pas autorisées dans votre région. Cette page ne comporte donc aucun lien commercial : elle est purement informative.

Aucun lien commercial n’est affiché dans votre région.

Comment lire les étiquettes de preuve

Chaque affirmation de cette page porte une étiquette indiquant le poids qu’elle peut supporter. Elle décrit la solidité du résultat, non le prestige de la revue.

  • Preuves solidesUn grand essai randomisé, une cohorte suivie longtemps, ou la concordance de plusieurs d’entre eux.
  • Preuves émergentesRéel, mais mince — petit, court, observationnel ou non encore répliqué.
  • ContestéDes chercheurs compétents sont en désaccord. Nous ne tranchons pas à votre place.

What the systematic reviews found

Cochrane reviews are the most conservative reading of a literature available, and the cinnamon one is unambiguous. Ten randomised trials, 577 participants with type 1 or type 2 diabetes, doses averaging 2 g a day for four to sixteen weeks. The conclusion was that there is not enough evidence to support using it.

Risk of bias was high or unclear in all but two of those ten trials — which is itself the finding. A literature this small and this loosely run cannot establish a modest effect either way.

Ce que disent les preuves

  • A Cochrane review of 10 randomised trials in 577 people found insufficient evidence to support cinnamon for type 1 or type 2 diabetes, with no statistically significant difference in HbA1c, serum insulin or post-meal glucose.1

    Preuves solides

  • Risk of bias was high or unclear in eight of the ten trials included in that review.1

    Preuves solides

The finding that gets quoted, and what it leaves out

A separate meta-analysis of ten trials did find a fasting glucose reduction of about 25 mg/dL, and that is the number cinnamon marketing quotes. Read one line further and the same analysis reports no significant effect on HbA1c.

That combination should make you suspicious rather than hopeful. HbA1c integrates glucose over months; fasting glucose is one morning reading. An intervention that moves the second but not the first has probably not changed much. The analysis also reported heterogeneity between trials of up to 94%, which its own authors said limits applying the results to patient care.

Ce que disent les preuves

  • A meta-analysis of 10 trials found cinnamon reduced fasting plasma glucose by about 25 mg/dL but had no significant effect on HbA1c.2

    Preuves émergentes

  • Heterogeneity between the trials in that analysis reached 94%, which the authors said limits the ability to apply the results to patient care.2

    Preuves émergentes

The coumarin problem

Almost all cheap cinnamon is cassia, not Ceylon, and cassia is high in coumarin — a compound with established liver toxicity at sustained intake. European food safety guidance sets a tolerable daily intake that a few grams of cassia daily can approach or exceed.

Sprinkling it on porridge is not the concern. Taking a supplement dose every day for months, on an evidence base a Cochrane review called insufficient, is a poor trade.

Questions fréquentes

Is Ceylon cinnamon better than cassia?
For coumarin content, yes — Ceylon contains far less, which removes the liver concern. It does not improve the efficacy evidence, which is null either way.
Why does everyone recommend cinnamon for blood sugar?
Because a handful of small positive trials circulated widely and the fasting-glucose figure is quotable. The systematic reviews that weigh all the trials together, including the unflattering ones, do not support it.

Références

  1. 1.Leach MJ, Kumar S (2012). Cinnamon for diabetes mellitus. Cochrane Database of Systematic Reviews. 10.1002/14651858.CD007170.pub2
  2. 2.Allen RW, Schwartzman E, Baker WL, Coleman CI, Phung OJ (2013). Cinnamon use in type 2 diabetes: an updated systematic review and meta-analysis. Annals of Family Medicine. 10.1370/afm.1517

À lire ensuite