Effects of Cordyceps militaris on Blood Glucose in Experimental Diabetes: A Systematic Review and Meta-Analysis of Animal Studies
Pharmacological Research - Modern Chinese Medicine · 2026-10 · 100896
DOI: 10.1016/j.prmcm.2026.100896 ↗Kurz gefasst
Systematische Übersicht in der Zeitschrift Pharmacological Research - Modern Chinese Medicine, 2026-10. Verfasst von Palaniyappan K und 6 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Themen: Diabetes mellitus
Abstract (Originalsprache)
Background
Cordyceps militaris ( C. militaris ) and its constituents, including polysaccharides and cordycepin, have been investigated in a growing number of experimental diabetes models, but their effect on blood glucose has not previously been synthesised quantitatively.
Objective
To systematically evaluate and quantitatively synthesise the effect of C. militaris and its derived constituents on blood glucose in animal models of diabetes, and to characterise the risk of bias and certainty of the underlying evidence.
Methods
Two databases (PubMed and Scopus) were searched from inception to 20 July 2026 for controlled in-vivo animal studies evaluating C. militaris , its extracts, polysaccharides or eligible derived constituents in diabetic or hyperglycaemic rodents. The protocol was registered retrospectively on PROSPERO, after the search. Screening, data extraction and risk-of-bias assessment were performed by a single reviewer without independent verification, and some blood-glucose values were digitised from published figures. Final blood-glucose values were synthesised as Hedges' g because studies reported glucose in different units, with raw mean differences also reported. A random-effects restricted maximum-likelihood (REML) model with Hartung-Knapp confidence intervals was used, and heterogeneity was assessed with Cochran's Q, τ 2 , I 2 and a 95% prediction interval. Subgroup analyses were performed by intervention chemistry (purified cordycepin versus whole material, extracts or polysaccharides) and diabetes-model type, with an exploratory subgroup analysis by induction method. Sensitivity analyses comprised restriction to six studies designed to evaluate glycaemic efficacy, a post hoc exclusion of studies with extreme effect sizes, leave-one-out analysis and an illustrative scenario analysis for unretrieved reports. Small-study effects were explored with a funnel plot, Egger regression and trim-and-fill. Risk of bias was evaluated with the SYRCLE tool, and certainty of evidence with a GRADE approach adapted for preclinical animal studies.
Results
Ninety-nine records were identified (69 PubMed, 30 Scopus); after removal of nine duplicates, 90 records were screened, and 18 of the 44 reports sought for full text (41%) could not be retrieved. Sixteen reports describing 15 unique animal studies were included qualitatively, and 11 independent studies (156 animals) contributed comparable final blood-glucose data to the primary meta-analysis. On average, C. militaris -derived interventions were associated with lower blood glucose than diabetic controls (Hedges' g = −2.32, 95% CI −3.92 to −0.72), but this estimate is highly uncertain. Heterogeneity was very high (I 2 = 86.2%), and the 95% prediction interval crossed the null (−7.35 to 2.70): a future animal study could show a substantial benefit, no effect or higher glucose. The pooled estimate depended on three studies with extreme effect sizes (g = −4.74 to −7.05), each with six animals per group and figure-derived data; excluding them gave g = −1.40 (95% CI −2.87 to 0.08; k = 8), which was not statistically significant. In a sensitivity analysis restricted to six studies designed primarily to evaluate glycaemic efficacy, the effect remained significant (g = −2.73, 95% CI −4.50 to −0.96; I 2 = 72.6%), but not when its two extreme-effect studies were also excluded (g = −1.97, 95% CI −3.96 to 0.01; k = 4). Subgroup analyses identified no statistically significant moderator (between-subgroup p = 0.19 to 0.91). Egger's regression indicated funnel-plot asymmetry (p = 0.008), and trim-and-fill reduced the estimate to g = −2.01 (one study imputed) or g = −1.02 (95% CI −2.85 to 0.81; four imputed). Allocation concealment, random housing and blinding were rarely reported and were judged as unclear risk of bias in almost all studies, and the certainty of evidence was rated very low.
Limitations
Only two databases were searched (Embase, Web of Science, CNKI and Wanfang were not), and 18 of 44 sought full texts (41%) could not be retrieved, so availability and language bias are possible. Screening, extraction and risk-of-bias assessment were performed by one reviewer and were not independently verified; inter-rater agreement could not be calculated. Several effect sizes, including the three largest, were derived from digitised figures, and country, strain, sex and duration could not be verified against the full text for several studies. Most SYRCLE domains were unclear because the primary studies reported their methods poorly. The protocol was registered retrospectively. Heterogeneity was very high and was not explained by any subgroup analysis.
Conclusion
In this synthesis of 11 preclinical studies, C. militaris -derived interventions were associated with lower final blood glucose than diabetic controls, and the effect remained statistically significant in a sensitivity analysis restricted to six studies (n = 6) designed primarily to test glycaemic efficacy. The finding is fragile: it was not statistically significant when studies with extreme effect sizes were excluded, heterogeneity was very high, the prediction interval crossed the null, small-study analyses suggest that the estimate may be inflated, and the certainty of evidence is very low. These preclinical findings are hypothesis-generating only. They do not support clinical efficacy or human trials at this stage, and they justify only further standardised, independently verified and adequately powered preclinical research.
Zitieren
APA
K, P., A, G., A, M., S, S. K., S, S. N. M., V, R., & Ramachandiran, S. (2026). Effects of Cordyceps militaris on Blood Glucose in Experimental Diabetes: A Systematic Review and Meta-Analysis of Animal Studies. Pharmacological Research - Modern Chinese Medicine, 100896. https://doi.org/10.1016/j.prmcm.2026.100896
Vancouver
K P, A G, A M, S SK, S SNM, V R, et al. Effects of Cordyceps militaris on Blood Glucose in Experimental Diabetes: A Systematic Review and Meta-Analysis of Animal Studies. Pharmacological Research - Modern Chinese Medicine. 2026:100896. doi:10.1016/j.prmcm.2026.100896
BibTeX
@article{K2026,
author = {K, Palaniyappan and A, Gnanavel and A, Murugesan and S, Saravana kumar and S, Sathya Narayana Murthy and V, Rajimol and Ramachandiran, Sivaramakrishnan},
title = {Effects of Cordyceps militaris on Blood Glucose in Experimental Diabetes: A Systematic Review and Meta-Analysis of Animal Studies},
journal = {Pharmacological Research - Modern Chinese Medicine},
year = {2026},
pages = {100896},
doi = {10.1016/j.prmcm.2026.100896}
}
EUVIMED