Glutathione: research

An antioxidant with mixed oral evidence and important risks from compounded injections.

Also called: GSH, Reduced L-glutathione, L-glutathione, GSSG, oxidized glutathione.

Antioxidant peptides · Endogenous thiol antioxidant; chemically a gamma-glutamyl tripeptide.

Wellness evidence is early human; an oncology phase 3 trial was negative. At least 85 distinct human records include 18 studies without posted registry results; reviews are not counted again.

Studies

• Single oral dose (Witschi 1992). pk-study, n=7, Seven healthy volunteers. A single conventional oral dose did not raise plasma glutathione. (source)
• IV clearance (Aebi 1991). pk-study, n=10, Healthy volunteers. Blood glutathione rose sharply and cleared quickly. (source)
• IV metabolite study (Hong 2005). pk-study, n=7, Seven healthy men aged 22–23. The tested dose was a single 50 mg/kg infusion. (source)
• Oral 1 g for four weeks. rct, n=40, Healthy adults. Four weeks of oral dosing did not change oxidative-stress markers. (source)
• Oral 250 or 1,000 mg. rct, n=54, Healthy nonsmoking adults. Longer oral dosing raised glutathione stores, without proving disease prevention. (source)
• Liposomal 500 or 1,000 mg. open-label, n=12, Healthy nonsmokers aged 50–80. This small pilot found higher stores after liposomal dosing. (source)
• Swallowed vs under the tongue. crossover-rct, n=20, Adults with metabolic syndrome. Sublingual dosing improved a redox marker more than swallowed dosing. (source)
• Buccal tablet absorption. pk-study, n=15, Healthy volunteers. A small combination-tablet study found a short-term blood increase. (source)
• Three oral formulations (2026). crossover-rct, n=14, Healthy adults. Micellar dosing increased measured exposure more than standard dosing. (source)
• Liposomal vs plain, single dose. pk-study, n=12, Healthy adults. A 12-person study found formulation-dependent blood exposure. (source)
• Dissolving film vs tablet. crossover-rct, Healthy adults. Both formulations raised circulating levels during repeated dosing. (source)
• Skin tone, oral 500 mg. rct, n=60, Healthy medical students. Some skin sites lightened over four weeks. (source)
• Reduced vs oxidized, 250 mg. rct, n=60, Healthy women aged 20–50. Overall skin results were weaker than the abstract conclusion suggests. (source)
• Topical oxidized glutathione. rct, n=30, Healthy women aged 30–50. A small split-face trial found localized lightening. (source)
• Skin tone, buccal lozenge. single-arm, n=30, Filipino women, skin types IV–V. An uncontrolled lozenge study reported skin lightening. (source)
• Oral plus topical skin trial. rct, n=46, Adults seeking skin lightening. A small combined-route trial suggested greater skin lightening. (source)
• Oral antioxidant mixture. rct, n=90, Indonesian women aged 33–50. A larger mixture trial did not show significant overall lightening. (source)
• Cystine plus glutathione. rct, n=124, Asian women. A combination trial favored cystine plus glutathione. (source)
• IV skin lightening (Zubair 2016). cohort, n=50, Women aged 25–47; active group paid for injections. The IV mixture had uncertain benefit and substantial reported harms. (source)
• Melasma injection add-on (2026). rct, n=50, Chinese women with epidermal melasma. A new small trial supports an intradermal combination as a melasma add-on. (source)
• Acne, oral 500 mg. rct, n=40, People with mild to moderate acne. The acne trial did not establish a clear benefit. (source)
• Early Parkinson disease, IV. open-label, n=9, Untreated early Parkinson disease. An early uncontrolled study reported symptom improvement. (source)
• Parkinson disease, IV placebo trial. rct, n=21, Adults with Parkinson symptoms despite medication. IV glutathione did not significantly beat placebo. (source)
• Nasal Parkinson phase I/IIa study. rct, n=30, Adults with Parkinson disease. The safety pilot supported further study, not proven symptom benefit. (source)
• Nasal Parkinson phase 2b. rct, n=45, Adults with stage 1–3 Parkinson disease. The larger nasal trial also failed to beat placebo. (source)
• Nasal delivery to the brain. pk-study, n=15, Adults with mid-stage Parkinson disease. A nasal dose raised a brain-scan glutathione signal. (source)
• ALS, daily muscle injection. open-label, n=32, Adults with definite ALS. Daily IM dosing did not slow ALS progression. (source)
• Male infertility, IM pilot. single-arm, n=11, Men with abnormal semen parameters. An uncontrolled pilot reported better sperm movement. (source)
• Male infertility, IM crossover. crossover-rct, n=20, Infertile men with varicocele or sterile inflammation. A small crossover study favored some semen measures. (source)
• CF inhaled pilot. rct, n=19, People aged 6–19 with cystic fibrosis. A small inhaled pilot found a peak-flow signal. (source)
• CF inhaled, 24-week RCT. rct, n=153, People aged ≥8 with cystic fibrosis. The larger inhaled trial found no clinically relevant benefit. (source)
• CF inhaled, 12-month RCT. rct, n=105, 54 adults and 51 children with cystic fibrosis. A year-long study missed its primary lung endpoint. (source)
• CF growth, first oral trial. rct, n=44, Children aged 18 months–10 years with cystic fibrosis. The first pediatric trial reported growth improvements. (source)
• GROW, oral replication. rct, n=60, Children aged 2–10 with pancreatic-insufficient cystic fibrosis. The better-powered replication found no growth benefit. (source)
• Asthma inhalation safety test. crossover-rct, n=8, People with mild asthma. Nebulized glutathione narrowed the airways in asthma. (source)
• Cisplatin gastric cancer trial. rct, n=50, Adults with advanced gastric cancer receiving weekly cisplatin. This older trial favored a combined IV and IM chemoprotection regimen. (source)
• Cisplatin ovarian cancer trial. rct, n=151, Women with stage I–IV ovarian cancer. More participants completed chemotherapy with glutathione. (source)
• Cisplatin, fixed 5 g pilot. rct, n=20, Adults with lung or head-and-neck cancer. A 5 g oncology dose was tested with supervised hydration. (source)
• Oxaliplatin nerve protection. rct, n=52, Adults with advanced colorectal cancer. A small trial found less oxaliplatin neuropathy. (source)
• FOLFOX4 pharmacology trial. rct, n=27, Adults receiving adjuvant FOLFOX4 for colorectal cancer. A small study favored neuropathy measures but did not establish survival safety. (source)
• N08CA, phase 3 chemotherapy. rct, n=185, Adults receiving paclitaxel and carboplatin. The phase 3 taxane/platinum trial found no neuropathy prevention. (source)
• Fatty-liver oral pilot. single-arm, n=34, Adults with ultrasound-defined fatty liver. ALT fell during oral dosing, but no placebo group tested causation. (source)
• Cirrhosis, GSH and vitamin B6. rct, n=61, Adults with liver cirrhosis. The controlled cirrhosis trial did not show the proposed antioxidant benefit. (source)
• Cirrhosis cohort (2026). cohort, n=466, Adults with chronic liver disease receiving oral GSH. The large uncontrolled cohort cannot establish benefit from glutathione. (source)
• Insulin sensitivity, oral 1 g. rct, n=20, Obese men with or without type 2 diabetes. A small trial suggested improved insulin sensitivity. (source)
• Diabetes, six-month oral trial. rct, n=354, Adults aged 30–78; 250 with diabetes and 104 without. Biomarkers improved, but overall HbA1c benefit was limited. (source)
• Dialysis, oral 1 g pilot. single-arm, n=26, Adults on twice-weekly maintenance hemodialysis. One inflammatory marker improved in an uncontrolled dialysis study. (source)
• Contrast kidney-injury trial. rct, n=505, Adults aged ≥75 undergoing coronary angiography or intervention. Adding glutathione did not prevent contrast-associated kidney injury. (source)
• Heart-attack reperfusion pilot. rct, n=50, Adults undergoing urgent angioplasty for STEMI. The heart-attack pilot changed biomarkers, not proven survival. (source)
• Exercise fatigue crossover. crossover-rct, n=8, Healthy men. A tiny crossover study suggested less fatigue during mild cycling. (source)
• Citrulline combination, one week. rct, n=60, Resistance-trained men. A short combination study changed exercise biomarkers. (source)
• Citrulline plus resistance training. rct, n=75, Resistance-trained men. The combination did not show a sustained eight-week advantage. (source)
• Vascular combination in women. rct, n=44, Healthy postmenopausal women. A combination improved vascular test results in a small trial. (source)
• Sublingual vascular crossover. crossover-rct, n=16, Men with cardiovascular risk factors. Overall vascular function did not improve. (source)
• Oral vs skin formulation in autism. open-label, n=26, Children aged 3–13 with autism. The study measured blood metabolites, not proven autism benefit. (source)
• Autism IV crossover pilot. crossover-rct, n=24, Children aged 5–16 with autism and severe behavior problems. The placebo-controlled pilot found no behavioral benefit. (source)
• Shock after cosmetic IV mixture. case-report, n=1, Woman using tirzepatide with prolonged low food intake. A severe infusion reaction occurred in a tirzepatide user. (source)
• Severe skin reaction after IV vitamins. case-report, n=1, 33-year-old woman. An IV mixture was followed by a severe blistering skin reaction. (source)
• Weekly cisplatin, relapsed ovarian cancer. rct, n=33, Women with relapsed ovarian cancer after prior cisplatin. This small trial suggested better chemotherapy dose delivery. (source)
• IV glutathione during dialysis. cohort, n=28, Adults on maintenance hemodialysis. A long dialysis cohort reported better anemia measures without a control group. (source)
• Uremic anemia, IV add-on. rct, n=42, Adults with uremic anemia receiving erythropoietin. An older small trial reported improved anemia measures. (source)
• Liposomal glutathione in HIV. rct, n=15, Adults with HIV; ten healthy reference volunteers. A small HIV study changed laboratory markers, not proven infection outcomes. (source)
• S-acyl cream and UV redness. rct, n=10, Healthy adults with skin phototypes II–III. A ten-person study tested a modified topical form, not injectable glutathione. (source)
• Leg circulation, five-day infusion trial. rct, n=40, Adults with Fontaine stage II peripheral artery disease. A short infusion trial reported improved walking distance. (source)
• Skin evidence review (2019). systematic-review, Published human studies; counts here refer to trials, not participants. Four studies; three placebo-controlled and one single-arm. Results were inconsistent and mostly limited to sun-exposed sites. (source)
• Skin and melasma review (2025). systematic-review, Published human studies; counts here refer to trials, not participants. Oral and topical effects were modest and often temporary. The single controlled IV study had weak efficacy and important harms. (source)
• Topical evidence review (2025). systematic-review, Published human studies; counts here refer to trials, not participants. Five clinical trials met inclusion criteria; formulations and outcomes differed, and safety data were limited. (source)
• Skin review through March 2026. review, Published human studies; counts here refer to trials, not participants. Small studies, mixed ingredients and short follow-up prevent a reliable cosmetic IV benefit-risk judgment. (source)
• Cochrane platinum neuropathy review. meta-analysis, Published human studies; counts here refer to trials, not participants. Seven glutathione trials, 387 participants. Subjective grade ≥2 neuropathy RR 0.29, 95% CI 0.10–0.85; objective evidence remained insufficient. (source)
• Cochrane CF antioxidant review. meta-analysis, Published human studies; counts here refer to trials, not participants. Twenty antioxidant trials, 924 participants overall. Glutathione findings were mixed; most studies had bias concerns. (source)
• Parkinson evidence review (2026). systematic-review, Published human studies; counts here refer to trials, not participants. Nine NAC/GSH studies, 196 participants overall. Glutathione clinical benefits remained inconclusive. (source)
• Post-polio registry results. single-arm, n=20, Adults with post-polio syndrome. Results were posted in the registry; no controlled efficacy conclusion is possible. (source)
• Autism oral registry results. single-arm, n=6, Children with autism. Results were posted in the registry; no controlled efficacy conclusion is possible. (source)
• Cold-season healthy days, conference report. rct, n=124, Generally healthy adults aged 50–75. The conference abstract did not show fewer illness symptoms or more healthy days. (source)
• Liposomal pilot registry. rct, n=11, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Inflammation, GSH vs NAC. rct, n=78, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Parkinson IV registry. rct, n=20, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Contrast kidney-injury registry. rct, n=1000, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Glutathione vs curcumin. rct, n=75, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Marathon immune markers. rct, n=84, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Diabetes registry, 217 enrolled. rct, n=217, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Nonrandomized three-arm blood-level study. cohort, n=56, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Vitamin C/GSH absorption. rct, n=20, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Vitamin C/GSH efficacy study. rct, n=30, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Lung cancer combination trial. rct, n=80, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Cardiac-surgery kidney trial. rct, n=242, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• S-acetyl derivative in soccer. rct, n=20, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Effervescent skin formulation. rct, n=62, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Ultrasound melasma mixture. rct, n=35, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Microneedling melasma study. rct, n=42, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• Gingival pigmentation study. rct, n=15, Registered participants; see registry for eligibility. Registered study without public results; no efficacy or safety conclusion. (source)
• N08CA posted results. rct, n=185, Same N08CA cohort; not an additional trial. Posted results confirm the lack of neuropathy benefit. (source)
• Ovarian cisplatin trial, review extraction. rct, n=54, Women receiving first-line cisplatin for advanced ovarian cancer. The older open trial did not establish a clear objective neuropathy benefit. (source)

Gaps: No published human SC dose, absorption study or reliable long-term safety dataset was found. No validated routine adult wellness titration, microdose or cycling schedule was found. Skin effects are usually small or temporary; long-term IV cosmetic benefit and safety remain unknown. Pregnancy, breastfeeding and elective use while trying to conceive lack adequate safety data. Most studies report biomarkers. Raising glutathione stores does not establish longer life or disease prevention. The human-study count is a documented minimum, not a complete census of every historical study or national registry. Many papers omit arm-level adverse-event rates; missing reporting must not be read as no adverse events. Drug-interaction studies and dose adjustments for kidney or liver impairment are not established. EU national and UK product authorization checks were incomplete; possession and import rules vary by product and region.

Related to Glutathione

• NAD+ and precursors: NAD+ helps cells use energy; wellness injection benefits remain unproven.

Glutathione: the short version

Sources

• 503B Bulk Drug Substances List, FDA · FDA · 2026
• Bulk Drug Substances Nominated for Use in Compounding Under Section 503A of the Federal Food, Drug, and Cosmetic Act, FDA · FDA · 2026
• FDA reminds compounders not to use dietary supplement grade glutathione for injectables, FDA · 2026
• Find product information about medicines, Medicines and Healthcare products Regulatory Agency · MHRA · 2026
• The 2026 Prohibited List, World Anti-Doping Agency · WADA · 2026
• Bulk Drug Substances Nominated for Use in Compounding Under Section 503B of the Federal Food, Drug, and Cosmetic Act, FDA · FDA · 2025
• Advertising guidance for businesses involved with intravenous (IV) vitamin and related therapies: Complying with therapeutic goods advertising requirements, Therapeutic Goods Administration · 2023
• Pharmacy Compounding Advisory Committee Meeting June 8, 2022 Summary Minutes, FDA · 2022
• 30ml Reduced-Glutathione 200mg/ml (2020-11-13), Health Canada · 2020
• FDA highlights concerns with using dietary ingredient glutathione to compound sterile injectables, FDA · 2019
• BSS PLUS- balanced salt solution enriched with bicarbonate, dextrose, and glutathione kit, Alcon Laboratories, Inc. · DailyMed
• EU/3/06/361 - orphan designation for treatment of cystic fibrosis, European Medicines Agency
• Product information, Health Canada
• Validation of an HPLC-UV method for the analysis of glutathione and its impurities, FDA
• A cluster analysis of 466 patients demonstrates that glutathione supplementation could preferentially benefit advanced unstable cirrhosis phenotype rather than stable cirrhosis., Philips CA et al. · Frontiers in pharmacology · 2026

Sources checked: 2026-10-18.

Information only, not medical advice. Peptifacts does not sell anything and does not link to sellers.