What it is
Glutathione is a tripeptide — glutamate, cysteine and glycine — that essentially every cell in your body makes for itself, continuously. It is not a signalling molecule like most of this library's entries. It is a working chemical, and it has two main jobs:
- The cell's principal antioxidant. It absorbs reactive molecules that would otherwise damage proteins, membranes and DNA, and is then recycled back into its active form.
- A central component of liver detoxification. A whole family of liver enzymes exists specifically to attach glutathione to drugs and toxins, which is how many substances are made water-soluble enough to be excreted.
Both roles are real, well characterised and genuinely important. Neither of them is an argument for injecting it, and the distance between those two statements is what this entry is about.
The bond that makes it unusual
Ordinary peptides are chains built by the ribosome, joining each amino acid's backbone carboxyl to the next one's amino group. Glutathione is not built that way. Its glutamate residue connects through its side-chain carboxyl instead — a gamma-peptide bond — and the molecule is assembled by two dedicated enzymes rather than translated from a genetic template.
That has a practical consequence worth knowing: most enzymes that chop up peptides recognise the ordinary backbone bond and cannot cut this one. Glutathione therefore survives inside the cell in a way an ordinary tripeptide would not. It is a peptide by composition and something slightly different by construction.
The record, and what reading it shows
Glutathione arrives with numbers that dwarf everything else in this library:
| Measure | Glutathione | For comparison |
|---|---|---|
| PubMed records | 207,068 | VIP, the previous largest here, 15,146 |
| Randomized controlled trials | 2,110 | VIP, 91 |
| Registrations on ClinicalTrials.gov | 269 | Mazdutide, 41 |
| Of those, interventional | 233 | Mazdutide, 41 of 41 |
| US drug-label records | 57 | VIP, 1 |
| Approved injectable product | none | — |
The last two rows are the entry. Fifty-seven records in the US drug label database is the kind of figure that gets quoted as legitimacy, and it takes about five minutes to find out what it means.
We read all 57. They fall into exactly two groups, and neither is a glutathione medicine.
Group one: homeopathic combination remedies
Products with names like NEPHROPLEX, KIDNEY TONIC I and KIDNEY TONIC II, HEPATAPLEX, HEPATATOX, APIZELEN, and two separate products called FREE RADICAL. In these, glutathione is one listed ingredient among dozens — alongside entries such as Berberis vulgaris, Uva ursi, Solidago virgaurea, Chelidonium majus, Taraxacum officinale and porcine organ preparations. There are also several plain products registered simply as glutathione liquid by homeopathic suppliers.
Homeopathic products appear in DailyMed because they are filed there. Appearing in the database is not the same as being reviewed and approved as effective.
Group two: topical skin-whitening cosmetics
The larger group. Soaps, creams, lotions and serums pairing glutathione with kojic acid, arbutin, niacinamide or ascorbic acid — products registered by cosmetics manufacturers, most of them based in China. Names in the set include glutathione and kojic acid soap, dark-spot serums, whitening face creams and body lotions.
These are cosmetics. They are topical. They tell a reader nothing whatsoever about an injected or intravenous preparation.
What is missing from the list
An approved injectable glutathione product. That is the form the compound is actually sold and administered in — the IV drips offered for skin lightening, the vials sold by compounding pharmacies and research suppliers — and it is the one thing the 57 records do not contain.
The pattern this library keeps finding
This is now the third distinct way a US drug-database hit can be misread, and collecting them makes the check easier to run:
- A bulk-substance registration by a chemical supplier. Our VIP entry found exactly one DailyMed record for vasoactive intestinal peptide, and it was raw powder listed by a supplier.
- Homeopathic combination remedies, where the compound is one name in a long ingredient list.
- Topical cosmetics, where the compound is in a cream and the claim being made is about an injection.
In all three cases the count is real and the inference is wrong. The database records filings; it does not certify them. The only reliable move is to open the records and read what they are — which anyone can do, for free, in a browser.
Why the trial registry looks different here
Glutathione also breaks this library's usual complaint about evidence ownership. Most entries here rest on research controlled by a single company: our survodutide entry found 25 of 26 registrations filed by one sponsor, and our cagrilintide entry found all 43 filed by one.
Glutathione is the opposite. Across a 200-record sample of its 269 registrations (233 of them interventional), the largest single sponsor accounted for 7 studies. The rest are spread across Baylor College of Medicine, the University of Florida, Cairo University, hospitals, children's clinics and small companies on several continents. Nobody owns this research.
That is a genuine strength, and it is worth being precise about what it does and does not establish. Independent evidence is harder to shape to a commercial conclusion. It still does not tell you that injecting glutathione does anything for skin tone or detoxification, because independence is a property of who did the studies, not of what they found.
What is actually sold
Injectable and intravenous glutathione, from compounding pharmacies, clinics offering drips, and research suppliers selling lyophilised powder — marketed principally for skin lightening, and secondarily for liver support and detoxification. Oral capsules and topical creams are sold under cosmetic rules.
None of those preparations has an approved label. The enormous research literature behind the molecule is about what glutathione does inside cells that make their own; it is not evidence for what a vial of it does when injected. Our entry on how to read this library sets out the distinction between a compound being real and a product being established.