Costus — Benefits Deep Dive
The main Costus page already does the hardest job — sorting Cheilocostus speciosus out from three other plants that share its trade name, and debunking the myth that eating its diosgenin turns into steroid hormones in your body. These four pages go deeper into what is actually claimed for the plant itself: real animal and cell-culture evidence for blood sugar, inflammation and pain, and infection, plus the genuine industrial-chemistry story behind diosgenin. One finding here corrects the main page: this is one of the few herbs on this site with an actual human signal behind its best-known claim — just not the kind anyone is selling.
Deep-Dive Articles
Blood Sugar: The Real Human Evidence
Four independent rodent and cell mechanisms, read against the ceiling already set by acarbose — and then the part the main page understates: two Sri Lankan diabetes-clinic surveys (2014 and 2018, the second in a thousand patients) that linked crepe ginger to real hypoglycaemia in people who actually eat it, and a third (2015) showing how widely it is eaten. That is evidence of a genuine glucose-lowering effect. It is also a safety warning, not a treatment plan.
Read the deep dive →Diosgenin, Costunolide, and the Industrial Chemistry
Not a re-run of the hormone myth already debunked on the main page. The real Marker-degradation history in more depth, crepe ginger’s own documented place in it as an Indian alternative to the Mexican yam, why diosgenin content varies by an order of magnitude between batches, a second compound family in the rhizome nobody markets, and what isolated diosgenin and costunolide actually do to cancer cells in a dish — which is a different question from what the plant does to a person.
Read the deep dive →Anti-Inflammatory and Analgesic Evidence
A genuinely substantial rodent literature — carrageenan paw oedema, cotton-pellet granuloma, Freund’s arthritis, writhing, hot-plate, a full cytokine panel — run by several independent groups over fifteen years, with the named sesquiterpene lactones behind it. Also the page that catches a real part-substitution problem hiding in plain sight, and handles one paper carefully enough not to accidentally resurrect the hormone myth from a different angle.
Read the deep dive →Antimicrobial and Antiviral Evidence
Real minimum inhibitory concentration (MIC) values against dermatophytes and bacteria, printed and scaled the way this site scales any in-vitro potency figure — and the arithmetic points somewhere specific: toward the plant’s traditional topical use, not toward drinking it. One study went further than a Petri dish and confirmed reduced bacterial load in infected mice. Two different viruses have now been screened, under two different names for the same plant.
Read the deep dive →Table of Contents
Evidence Ledger
Claims sorted by what actually supports them, not by how they are marketed. Harms sit in the same table as benefits so the balance is visible rather than argued.
Human, observational, and this is the surprising one
- Crepe ginger causes real hypoglycaemia in real-world diabetic users. Evidence: three cross-sectional surveys of Sri Lankan diabetes patients, two of them sharing a first author. Two concern low blood sugar: at Teaching Hospital Peradeniya (2014, 252 valid answers) eating crepe ginger was associated with a higher incidence of hypoglycaemia (P = 0.01), and at a private diabetes clinic (2018, 1,000 patients) Costus speciosus was among the two most common native foods blamed for it. The third (2015, 220 herbal users) recorded only which herbs people use, not hypoglycaemia. Verdict: this is genuine human evidence of a real pharmacological effect — and it is a harm signal, not a demonstrated treatment benefit. No controlled trial has tested whether it lowers hemoglobin A1c (HbA1c). See the blood sugar page.
- How often native foods were blamed. In the 1,000-patient clinic survey, 26.1% of patients had symptomatic hypoglycaemia in the preceding month. A sudden change in diet (46.7%) was the commonest recognised cause, and non-prescribed native foods accounted for hypoglycaemia in 16.9% of patients; among those, bitter gourd was named in 54.5% and crepe ginger in 52.3%, ahead of Salacia prinoides (11.4%) and ivy gourd (6.8%). Dissanayake and colleagues, BMC Endocrine Disorders, 2018. PubMed PMID: 29884151
Established industrial chemistry — not a claim about the body
- Diosgenin from plants including crepe ginger was a real feedstock for early steroid-hormone drug synthesis. Evidence: documented twentieth-century pharmaceutical history (the Marker degradation, Syntex), with crepe ginger specifically identified as an Indian source as early as 1970 (Dasgupta and Pandey, “A new Indian source of diosgenin (Costus speciosus)”, Experientia, 1970, PubMed PMID: 5462855). Verdict: established chemistry, and it says nothing about what happens when a person eats the rhizome. See the diosgenin page.
Checked and refuted
- Eating diosgenin converts to progesterone or other steroid hormones in the body. Verdict: false. Refuted on the main page by basic enzymology, alongside a randomised, placebo-controlled crossover trial of a wild yam (Dioscorea villosa) cream rubbed on the skin — not eaten diosgenin, and not costus — in which 3 months of the cream left serum and salivary progesterone unchanged in 23 menopausal women (Komesaroff and colleagues, Climacteric, 2001, PubMed PMID: 11428178); the claim about eating it rests on the enzymology — no human enzyme performs the multi-step industrial conversion. See the main page’s full treatment.
Laboratory and animal only — inflammation and pain
- Anti-inflammatory, analgesic and antipyretic activity. Evidence: a substantial and consistent rodent literature across at least five independent groups and six different pain/inflammation models (carrageenan oedema, cotton-pellet granuloma, Freund’s arthritis, acetic-acid writhing, hot-plate, yeast pyrexia), plus a named eight-compound cytokine-level mechanism. Verdict: unusually well replicated for a preclinical literature, still entirely preclinical — no human trial exists. Also carries a genuine part-substitution problem: several of the strongest studies used aerial parts, not the rhizome that tradition and commerce actually use. See the inflammation page.
Laboratory and animal only — blood sugar mechanism
- Lowers glucose via at least four separate mechanisms. Evidence: streptozotocin and alloxan rat models, alpha-glucosidase inhibition, an isolated raffinose/PPAR pathway, and independent confirmation from an 18-plant functional-food screen. Verdict: mechanistically plausible and repeatedly positive, capped by what the same enzyme mechanism achieves as a licensed drug (acarbose) — modest. See the blood sugar page.
Laboratory only — cancer cell lines
- Cytotoxic and apoptotic activity against cancer cell lines. Evidence: breast (MCF-7, MDA-MB-231), liver (HepG2) and prostate (PC-3) cell lines, from at least six independent groups using different extracts and the isolated compounds diosgenin and costunolide. Verdict: preclinical only — no animal tumour model, no human trial — and a very large fraction of tested plant extracts show some cytotoxicity against a cancer cell line at high enough concentration, so this is a weaker signal than the citation count suggests. See the diosgenin page.
Laboratory, plus two in-vivo animal-infection studies — infection
- Antibacterial, antifungal and antiviral activity. Evidence: multiple in-vitro groups with real minimum inhibitory concentration values against dermatophytes and resistant bacteria, two different viruses screened (influenza A H1N1 and human adenovirus, under two different taxonomic names for the same plant), and one study that went further and confirmed reduced bacterial load in infected mice. The influenza work also went beyond a screen: a 2023 study gave a leaf extract to mice by mouth (see below). Verdict: an unusually solid screening tier for this kind of literature, with the in-vivo infection papers the strongest evidence anywhere in this Benefits set. The minimum inhibitory concentration arithmetic points toward the plant’s traditional topical use, not toward an oral systemic strategy. See the antimicrobial page.
- Influenza in cells and mice. In a 2023 study, a water extract of crepe ginger leaf — not the rhizome — blocked influenza A virus growth in mouse and human cell lines (in the mouse cells it was half-effective at 15.19 µg/mL and harmed half the cells only at 117.12 µg/mL) and, given to mice by mouth, was safe and protected them against three flu strains (H1N1, H3N2 and H9N2). Cinnamic, caffeic and chlorogenic acids were identified as likely active compounds. Cell and animal work only. Senevirathne and colleagues, Viruses, 2023. PubMed PMID: 37376674
Plausible but unquantified
- Sesquiterpene-lactone contact-allergy potential. Evidence: the rhizome demonstrably contains costunolide, dehydrocostus lactone and related sesquiterpene lactones — a compound class with a well-established general record as contact allergens. Verdict: plausible by chemical class; a direct PubMed search for allergy, dermatitis or sensitisation specifically naming this species returned zero records. Absent, not refuted — and worth naming precisely because nothing on the main page currently flags it. See the diosgenin page.
Refused
- Any milligram-per-dose figure for diosgenin, costunolide or eremanthin in a standard preparation. Published diosgenin content alone varies with soil, season and plant part enough that a single figure would misrepresent every batch that differs from it.
- Any claim that the anti-inflammatory or antimicrobial rodent data has been tested in a person. It has not, for any of the claims on these four pages.
- Any framing of the rheumatoid-arthritis fertility study as evidence for a direct hormonal effect. Its own mechanism is anti-inflammatory protection of existing receptor expression, not steroid synthesis — see the inflammation page’s careful reading of it.
What the Ledger Shows
Read top to bottom, the ledger inverts the plant’s own marketing in one specific place, and confirms it almost everywhere else.
The inversion is blood sugar. Most herb pages on this site have to report that a popular efficacy claim has never been tested in a human being at all. Crepe ginger is the opposite case: nobody has run the controlled trial that would establish it as a treatment, but two Sri Lankan clinic surveys (2014 and 2018) found the same thing — people who eat this plant experience real hypoglycaemia at rates their doctors noticed and wrote up — and a third survey (2015) shows how widely it is eaten, by a quarter of 220 diabetes patients using household remedies. That is unusually strong evidence for a genuine biological effect, delivered through the least controlled possible study design, about a harm rather than a benefit. The honest summary is this herb probably does lower blood glucose in real use. The hypoglycaemia in these surveys was reported by diabetes-clinic patients, and in the 2015 survey all 220 herbal users took their regular diabetes medicines alongside the herbs — the same combination concern with insulin or a sulfonylurea that the main page raises about the confusable “insulin plant,” and this ledger shows applies to crepe ginger itself, on its own evidence, under its own name.
Everywhere else, the pattern is ordinary for a plant with this profile: a real, replicated, mechanistically coherent preclinical literature for inflammation, pain and infection, built by genuine laboratories using genuine methods, that has simply never been pointed at a person. That gap is not unique to this plant and it is not evidence the claims are wrong — it is evidence nobody has funded the next step. Two things distinguish crepe ginger’s version of that gap from most: the infection literature has two in-vivo animal-infection studies (one bacterial, one influenza, the latter using leaf rather than rhizome), which is rarer than it should be for this kind of claim across the whole herbal literature; and the diosgenin story carries a genuine, checkable, already-refuted myth sitting directly beside a genuine, unrelated, well-documented industrial-chemistry fact — two truths about the same molecule that point in completely different directions, and conflating them is the single most common way this plant gets misrepresented.
Key Research: Blood Sugar
The full citation list, with the human survey data, is on the blood sugar page. The human survey records and live searches:
- PubMed: Costus speciosus and diabetes/hyperglycaemia (title/abstract-scoped)
- Medagama AB, Bandara R, Abeysekera RA, Imbulpitiya B, Pushpakumari T. Use of Complementary and Alternative Medicines (CAMs) among type 2 diabetes patients in Sri Lanka: a cross sectional survey. BMC Complement Altern Med. 2014;14:374. PubMed PMID: 25280877 — 254 type 2 diabetes patients interviewed at Teaching Hospital Peradeniya; of 252 valid answers, 192 (76%) used a herbal remedy to lower blood glucose, crepe ginger by 92 people, and eating crepe ginger was associated with a higher incidence of hypoglycaemia (P = 0.01).
- Dissanayake HA, Keerthisena GSP, Gamage KKK, Liyanage JH, Ihalagama IRHS, Wijetunga WMUA, Tillekaratne TAD, Katulanda GW, Katulanda P. Hypoglycaemia in diabetes: do we think enough of the cause? An observational study on prevalence and causes of hypoglycaemia among patients with type 2 diabetes in an out-patient setting in Sri Lanka. BMC Endocr Disord. 2018;18(1):35. PubMed PMID: 29884151 — 1,000 patients at a private diabetes clinic; native foods accounted for hypoglycaemia in 16.9% of patients, crepe ginger named in 52.3% of those.
- Medagama AB, Senadhira D. Use of household ingredients as complementary medicines for perceived hypoglycemic benefit among Sri Lankan diabetic patients; a cross-sectional survey. J Intercult Ethnopharmacol. 2015;4(2):138-42. PubMed PMID: 26401401 — a survey of which herbs 220 patients use, not of hypoglycaemia: ivy gourd 32%, crepe ginger 25%, bitter gourd 20%, eaten mostly as salads (72.8%).
- Medagama AB, Bandara R. The use of complementary and alternative medicines (CAMs) in the treatment of diabetes mellitus: is continued use safe and effective? Nutr J. 2014;13:102. PubMed PMID: 25331834 — a review that concludes: “Crepe ginger has not been studied adequately in humans to draw conclusions.”
- PubMed: Costus speciosus and alpha-glucosidase inhibition
- PubMed: Acarbose and HbA1c — the ceiling set by the same drug class
Key Research: Diosgenin, Costunolide and Chemistry
The full citation list is on the diosgenin page. Live searches:
- PubMed: Costus speciosus and diosgenin (title/abstract-scoped)
- PubMed: Costus speciosus and anticancer/apoptosis (title/abstract-scoped)
- PubMed: Costunolide and breast cancer
- PubMed: Sesquiterpene lactones as a contact-allergen class
- PubMed: The Marker degradation and industrial progesterone synthesis
Key Research: Inflammation and Pain
The full citation list is on the inflammation page. Live searches:
Key Research: Antimicrobial and Antiviral
The full citation list is on the antimicrobial page. Live searches:
External Resources
- PubMed — the index behind every search link on these pages.
- LiverTox (U.S. National Institutes of Health) — reference database for drug- and herb-induced liver injury.
- National Center for Complementary and Integrative Health (U.S. National Institutes of Health) — plain-language summaries of herb and supplement evidence.
- Plants of the World Online (Kew) — the botanical authority for Cheilocostus speciosus, its synonyms and distribution.
- CITES species checklist — confirms Saussurea costus, not this plant, is the Appendix I “costus.”
- American Diabetes Association — evidence-based glucose management, the standard of care behind every caution on the blood sugar page.
Connections
- All Herbs
- Blood Sugar: The Real Human Evidence — the surveys that correct the main page.
- Diosgenin, Costunolide and the Industrial Chemistry
- Anti-Inflammatory and Analgesic Evidence
- Antimicrobial and Antiviral Evidence
- Costus (Cheilocostus speciosus) — the main page, with the four-plant naming problem and the diosgenin hormone-myth debunk in full.
- Ginger — the true ginger this plant is confused with at the market stall.
- Fenugreek and Fenugreek Benefits — another major dietary diosgenin source carrying the identical hormone-precursor myth.
- Turmeric — the Zingiberaceae comparison the main page uses to show what crepe ginger’s chemistry is not.
- Type 2 Diabetes — the condition most readers arrive from, and what has outcome evidence behind it.
- Hypoglycemia Awareness and Prevention — the actual documented human harm from this plant.
- Rheumatoid Arthritis — the model condition behind the fertility study on the inflammation page.
- Psoriasis — the mouse model behind the 2025 immunomodulator study.
- Cissus Benefits — another herb on this site whose strongest evidence is old, weak and positive rather than simply absent.