What does the research say?
In a study of 6 people with ocular hypertension (high eye pressure) or early-stage glaucoma, participants took THC, CBD at two different doses, or an inactive drug (placebo) under the tongue on different days. THC temporarily lowered the pressure inside the eye. CBD, however, did not lower pressure at any measurement; at the high dose, pressure rose temporarily at the fourth hour from 23.2 to 25.9 mmHg. In a 2022 meta-analysis that re-analyzed this study against placebo, the difference between CBD and placebo was not significant (p = 0.08); the US National Academies' 2017 report also states that no difference from placebo was found.
The results of animal studies are mixed. A study in mice found that CBD had opposing effects on eye pressure and blocked the pressure-lowering effect of THC; in another mouse study, pressure decreased or did not change; and in a study of horses given CBD by mouth, which had no comparison group, it fell. In other words, the data on their own are not enough to say either 'CBD raises eye pressure in animals' or 'CBD lowers it.'
THC lowered eye pressure for about 4 hours, but it also caused a high or mental fogginess; in the study, one patient had a mild, panic-like reaction. In two older studies of patients with glaucoma using smoked marijuana or inhaled THC, the fall in eye pressure went hand in hand with a fall in blood pressure; a drop in blood pressure on standing up and a racing heart were common. The European Glaucoma Society's 2025 guideline states that there is no evidence that cannabinoids have a place in treating glaucoma; the American Glaucoma Society also does not recommend them, citing their short duration of action, side effects and the lack of evidence that they change the course of the disease.
THC given as eye drops did not lower the pressure inside the eye in a double-masked study of 6 patients with glaucoma or in two controlled studies of healthy volunteers (one of them in 28 people, four times a day for one week). The drops' oily vehicle could cause burning and eyelid swelling. A new drop that turns into THC in the eye was compared with placebo for two weeks in 56 patients with glaucoma or ocular hypertension: pressure fell only 0.3–1.2 mmHg more than with placebo, according to the company the difference was not significant, and development of the drug was stopped. The results of this study are only in the ClinicalTrials.gov record; there is no peer-reviewed publication.
Synthetic THC (dronabinol), as a single dose, increased blood flow at the head of the optic nerve by about 10% without changing the pressure inside the eye, in 24 healthy volunteers (5 mg) and at 10 mg in a 23-person study of patients with glaucoma (12 patients); in the patients with glaucoma, 5 mg had no effect. Whether this slows vision loss is not known; there are no long-term studies.
In a 2026 meta-analysis that pooled THC studies (5 studies, 99 people), the largest drop averaged about 15%, but in the four studies with a comparison group the difference from control was not significant. Genetic data found no cause-and-effect link between cannabis use and glaucoma; in a large UK database, glaucoma was also not more common among cannabis users, but in those who had used it 100 times or more, glaucoma had been diagnosed on average 10 years earlier (borderline significant; the authors note that the study has fundamental weaknesses). These observational findings do not show cause and effect.
Strength of evidence: ○○○○ Insufficient evidence. Finding: No effect found. What does this mean?
Ongoing research on this topic
No ongoing CBD trial on this topic appears on ClinicalTrials.gov right now. All ongoing trials
Possible side effects and drug interactions
In glaucoma, the most important concern is that CBD does not lower the pressure inside the eye: if it is used instead of treatment, pressure may go uncontrolled. In the single small study, the temporary rise in pressure seen after the high dose was not significant when compared with placebo. A mild, panic-like reaction was reported in one patient with THC. In studies of marijuana and THC, a marked rise in heart rate, a drop in blood pressure on standing up, a racing heart and mental changes were common.
Drug interactions were not studied in the glaucoma studies. CBD taken by mouth can slow how the body breaks down some medicines; people who take medicines regularly should keep this in mind.
To check interactions with the medicines you take: CBD drug interaction checker.
How was CBD used in studies?
In the only human study, CBD was given under the tongue as a single dose of 20 mg and 40 mg (study dose; not a personal recommendation); THC was used as a single dose of 5 mg (study dose; not a personal recommendation). At none of these doses did CBD lower eye pressure. There is no human study of long-term CBD use. The longest data on THC come from an uncontrolled program: all 9 patients who used oral THC or marijuana for 1–9 months stopped treatment. The longest controlled study with eye drops lasted 2 weeks (a drop that turns into THC in the eye, 0.5% or 1%, twice a day; study dose, not a personal recommendation) and found no clear difference from placebo.
Study details (for health professionals)
In a small study of 6 patients, a CBD-dominant cannabis extract given under the tongue did not lower the pressure inside the eye at all; 4 hours after the 40 mg dose, pressure rose temporarily (23.2 → 25.9 mmHg; p = 0.028), but in a 2022 meta-analysis comparing the same data with placebo, the difference was not significant (p = 0.08). Animal data are mixed: CBD eye drops showed opposing effects in mice, lowered pressure or left it unchanged in other mice, and in a study of horses given CBD by mouth with no comparison group, pressure fell. Animal findings do not show the effect in humans. THC's lowering of eye pressure is short-lived and comes with psychoactive effects: in older studies, the fall in eye pressure went hand in hand with a fall in blood pressure and lasted 3–4 hours; a racing heart, a drop in blood pressure on standing up and mental changes were common; all 9 patients with advanced glaucoma stopped treatment within 1–9 months. A single dose of synthetic THC (dronabinol) increased blood flow at the head of the optic nerve without changing eye pressure: at 10 mg in a 23-patient study of glaucoma (5 mg had no effect), and at 5 mg in 24 healthy volunteers; the long-term effect is unknown. In a placebo-controlled study of 120 cancer patients receiving chemotherapy, pressures inside the eye stayed within normal limits after oral THC; an unexpected rise was seen only in the 5 mg group. THC given as eye drops did not lower pressure in three controlled studies, one of them in patients with glaucoma; a drop that turns into THC in the eye was also not clearly different from placebo after two weeks in 56 patients with glaucoma or ocular hypertension (results only in the ClinicalTrials.gov record; no peer-reviewed publication). In a 2026 meta-analysis that pooled THC studies (5 studies, 99 people), the largest drop averaged 14.7%, but in the four studies with a comparison group the difference from control was not significant (6.9%; p = 0.41). The European Glaucoma Society's 2025 guideline (low level of evidence, strong recommendation) and the American Glaucoma Society do not recommend cannabinoids for treating glaucoma; the US National Academies' 2017 report concluded that there is limited evidence that cannabinoids are ineffective in improving the pressure inside the eye in glaucoma. A genetic analysis found no cause-and-effect link between cannabis use and the risk of glaucoma. CBD is not a treatment recommendation for glaucoma. October 2026 search (PubMed, Cochrane, ClinicalTrials.gov, Europe PMC): 12 controlled human studies of cannabinoids in glaucoma or on the pressure inside the eye and optic nerve blood flow were found: 6 in patients with glaucoma or ocular hypertension (one with results only in the ClinicalTrials.gov record; for one, the sources conflict on whether it was placebo-controlled), 5 in healthy volunteers and 1 in cancer patients receiving chemotherapy. Only 1 looked at CBD (6 patients, single dose). There are also 2 uncontrolled studies, 2 systematic reviews and meta-analyses, 1 expert committee report, 2 guidelines or professional society position statements and 2 observational studies; there is no Cochrane review. All that met the criteria, together with 3 animal studies, are in the table below.
Evidence assessment: Insufficient (in a single small study, CBD did not lower pressure; the temporary rise after 40 mg is not significant in the comparison with placebo)
CBD alone
There is no human glaucoma study of pure CBD. In the only human study, a CBD-dominant extract derived from cannabis, given under the tongue, did not lower the pressure inside the eye; the temporary rise seen 4 hours after the 40 mg dose was not significant in the comparison with placebo.
Products containing THC
THC lowered the pressure inside the eye temporarily (for about 4 hours) and brought psychoactive effects; it also caused a drop in blood pressure and a racing heart; in an uncontrolled program, all 9 patients stopped within 1–9 months. THC given as eye drops did not lower pressure; a drop that turns into THC in the eye, used for two weeks by patients with glaucoma, was also not clearly different from placebo (results only in the registry). In the 2026 meta-analysis pooling THC studies, the difference from control was not significant in the studies with a comparison group; the European Glaucoma Society (2025) and the American Glaucoma Society do not recommend it for treating glaucoma. There is no study showing that it keeps pressure low over the long term or slows vision loss.
Evidence: Insufficient (THC: drops lasting hours in small, short studies; the longest controlled study lasted 2 weeks and found no clear difference from placebo; no long-term studies; guidelines do not recommend it)
Dose range in studies
In humans, a single dose of 20 and 40 mg CBD (6 patients)
Starting dose and titration
Not reported in detail in the studies.
These are doses researchers gave under medical supervision; they are not a recommended dose for you. "mg/kg/day" means milligrams per kilogram of body weight per day.
| Study | Participants | Product | Dose and titration | Duration | Result | Source |
|---|---|---|---|---|---|---|
| Tomida 2006 Double-masked, placebo-controlled, 4-arm crossover pilot RCT | 6 Ocular hypertension / early glaucoma | Sublingual THC and CBD extract | Single dose: 5 mg THC, 20 mg CBD, 40 mg CBD | Single dose | 2 hours after THC, eye pressure was lower than with placebo (23.5 vs 27.3 mmHg; p = 0.026) and returned to baseline after the 4th hour; CBD did not lower it at any measurement; 4 hours after 40 mg CBD, pressure rose from 23.2 to 25.9 mmHg (p = 0.028; the abstract does not say what it was compared against). Reviews read this trial differently: according to Whiting 2015 (as reported by NASEM 2017), there was no difference in eye pressure between placebo and cannabinoids; according to Bilbao 2022, temporary improvement in the THC arm (p = 0.02) and a non-significant temporary worsening in the CBD arm compared with placebo (p = 0.08) Side effects: A mild panic-like reaction in one patient with THC. According to one review, side effects were seen in 3 patients with THC, 2 with 20 mg CBD, 5 with 40 mg CBD and 2 with placebo; most were mild (mouth discomfort, dizziness, throat irritation), with no serious side effects | PMID 16988594, doi:10.1097/01.ijg.0000212260.04488.60 |
| Miller 2018 Animal (mouse), topical | Mouse | THC, CBD | CBD has two opposing effects on eye pressure and blocks the pressure-lowering effect of THC | PMID 30550613, doi:10.1167/iovs.18-24838 | ||
| Rebibo 2022 Animal (mouse), topical nanoemulsion | Mouse | CBD (0.4–1.6%) | 2 weeks | Lowered eye pressure or had no effect on it | PMID 35245638, doi:10.1016/j.ijpharm.2022.121627 | |
| Bazzano 2025 Animal (healthy horse), oral; no placebo, compared with the same horses' measurements from the previous day | 8 Healthy adult horse (no glaucoma) | CBD (20% oil) | 1 mg/kg | Single dose, 8 hours of follow-up | Eye pressure fell | PMID 40435047, doi:10.1371/journal.pone.0325191 |
| Crawford and Merritt 1979 Controlled clinical study (compared with placebo) | 16 Open-angle glaucoma; 8 patients with normal blood pressure, 8 with high blood pressure | Inhaled THC (2.8%) | Single inhalation | Single administration | Eye pressure fell by 6–21 mmHg; in every patient the drop paralleled the drop in blood pressure and lasted 3–4 hours Side effects: Heart rate increase of 20–62 beats/min; systolic pressure drop of 12–52 mmHg | PMID 468444 |
| Merritt 1980 Labeled in PubMed as a randomized controlled trial; the abstract does not mention a placebo and the full text could not be read. According to one review, patients smoked marijuana with 2% THC or a placebo cigarette; according to another, it was a single-arm study | 18 Various types of glaucoma | Smoked marijuana | not reported (in the abstract); according to the review, a single cigarette, 900 mg, 2% THC (about 18 mg THC) | Single administration | Eye pressure and blood pressure fell at 60–90 minutes; the drop in eye pressure followed the drop in blood pressure Side effects: Drop in blood pressure on standing up, tachycardia, palpitations and mental changes, frequent enough to prevent routine use according to the authors | PMID 7053160, doi:10.1016/s0161-6420(80)35258-5 |
| Flach 2002 Uncontrolled, open, non-randomized compassionate-use program | 9 Advanced open-angle glaucoma not responding to treatment | Oral THC capsule or inhaled marijuana (added to existing treatment) | not reported (in the abstract) | 1–9 months | Eye pressure initially fell in all patients, and the target was reached in 4 patients; the drop did not last, and all 9 patients stopped treatment within 1–9 months (tolerance and systemic side effects) Side effects: Marked systemic toxicity according to the author | PMID 12545695, PMC1358964 |
| Lindner 2026 Randomized, double-masked, placebo-controlled crossover pilot study | 25 enrolled; 23 analyzed (5 mg: 11, 10 mg: 12) Patients with primary open-angle glaucoma under treatment (mean age 68) | Oral synthetic THC (dronabinol) | Single dose of 5 or 10 mg | Single dose (two days) | 10 mg increased blood flow at the head of the optic nerve for up to 4 hours (about 11%; p = 0.018 vs placebo); 5 mg had no effect on blood flow. Eye pressure and mean blood pressure did not change. The long-term effect is unknown. Supported by the Austrian Science Fund (FWF) Side effects: Mild to moderate side effects possibly or probably related to the drug in 9 of 25 patients; all resolved the same or the next day. No serious side effects and no stopping because of side effects | PMID 40772417, doi:10.1111/aos.17573, NCT04596826 |
| Hommer 2020 Randomized, double-masked, placebo-controlled, two-way crossover RCT (healthy volunteers) | 24 (12 women) Healthy young adults (aged 20–35, Austria); no glaucoma | Oral synthetic THC (dronabinol) capsule | Single dose 5 mg | Single dose (two separate days) | Primary outcome: blood flow at the head of the optic nerve at rest rose by 9.5 ± 8.1% 1 hour after dronabinol and did not change with placebo (0.3 ± 7.4%; p < 0.001 between days). Eye pressure, blood pressure and ocular perfusion pressure did not change; the regulation that keeps blood flow steady when pressure rises with exercise (autoregulation) also did not change. In healthy volunteers, not in patients with glaucoma; the long-term effect is unknown. Supported by the Austrian Science Fund (FWF); no conflicts of interest were reported Side effects: No side effects or psychoactive effects reported; heart rate did not change | PMID 31977076, doi:10.1002/cpt.1797, NCT03907163, design, number of people and primary outcome, no registration number in the paper, no results in the registry). The same group's paper measuring retinal blood flow with the same design (24 healthy volunteers, 5 mg): Hommer 2021, PMID 33999707, whether it is from the same trial could not be verified |
| Merritt 1981 Randomized, balanced, double-masked controlled study (abstract); according to the review, THC in one eye and vehicle oil alone in the other | 6 Primary open-angle glaucoma | THC eye drops (in light mineral oil) | 0.05% and 0.1% THC drops | not reported (in the abstract); single drop according to the review | THC given as eye drops did not lower eye pressure at either concentration. The authors note that smoked or oral THC lowers pressure but comes with a marked drop in systolic blood pressure Side effects: not reported (in the abstract) | PMID 6271841, doi:10.1002/j.1552-4604.1981.tb02626.x, The same group's second THC eye drop study in 8 patients: Merritt 1981, PMID 6114151, no abstract, could not be read, no difference from placebo according to the Passani 2020 review |
| Jay and Green 1983 Randomized, vehicle oil-controlled study (healthy volunteers) | 28 men randomized; 23 completed Healthy male volunteers; no glaucoma | 1% THC eye drops (in light mineral oil) or vehicle oil alone | 1% THC drops, four times a day | 1 week (four times a day) | In the 23 people who completed, there was no difference in eye pressure between eyes receiving THC drops and controls Side effects: 5 people stopped because of burning and eyelid swelling; 4 of them were using the vehicle oil alone | PMID 6301412, doi:10.1001/archopht.1983.01040010591012 |
| Green and Roth 1982 Randomized, vehicle oil-controlled study (applied to one eye; healthy volunteers) | not reported (in the abstract); 16 according to the Passani 2020 review Healthy paid volunteers; no glaucoma | THC eye drops (in light mineral oil) or vehicle oil alone | Single drop (concentration not in the abstract; 1% according to the review) | Single drop | There was no drop in eye pressure. A small (1 mm) pupil dilation was seen in both eyes, not related to the drug Side effects: Mild eye redness, mild burning or watering lasting less than 60 minutes; similar with the drug and the vehicle | PMID 6279061, doi:10.1001/archopht.1982.01030030267006 |
| Flom 1975 Controlled clinical study (PubMed publication type); double-blind with placebo cigarettes according to the review | not reported (in the abstract); 15 according to the Passani 2020 review Volunteers with normal eye pressure (marijuana users with different frequencies of use) | Smoked marijuana | Single cigarette, 12 mg THC | Single cigarette | Eye pressure fell only in light-to-moderate users who felt a marked high and relaxation from the experimental dose; the authors suggested this might be an indirect effect due to relaxation rather than a direct effect of the drug. According to the review, the drop was significant only at minute 80 (an average of 2.1 mmHg from 14.6 mmHg) and was small or absent in frequent users Side effects: not reported (in the abstract) | PMID 1089090 |
| Cooler and Gregg 1977 Controlled clinical study (compared with placebo and diazepam; randomized according to the review) | 10 People with normal eye pressure | Intravenous THC | Single dose 0.022 or 0.044 mg/kg | Single administration | Eye pressure fell by up to 51% from baseline, 37% on average (according to the review, 29% at the low dose and 37% at the high dose; 2% with placebo, 10% with diazepam). Heart rate rose 22–65% above resting values; anxiety increased with THC compared with placebo and diazepam; no pain-relieving effect was seen. In people without glaucoma; intravenous administration is not a route used in treatment Side effects: 22–65% rise in heart rate, increased anxiety; according to the review, marked drops in blood pressure and feeling faint in 3 people | PMID 329423, doi:10.1097/00007611-197708000-00016 |
| NCT06144918 (SBI-100 eye drops, registry results) Multicenter, double-masked, placebo-controlled, randomized parallel phase 2 study; results only in the registry (no peer-reviewed publication); sponsor is the drug's developer (Skye Bioscience) | 56 randomized (19 / 19 / 18); all completed Adults with primary open-angle glaucoma or ocular hypertension (eye pressure 21–36 mmHg; 42 of the 56 people were 65 or older) | SBI-100 emulsion eye drops (a synthetic THC prodrug that turns into THC in the eye) or placebo drops without the active ingredient | 0.5% or 1%, 1 drop in each eye twice a day | 14 days | Primary outcome (change in eye pressure from baseline at day 14, least squares mean): −1.99 mmHg with 0.5%, −1.12 mmHg with 1%, −0.83 mmHg with placebo; that is, only 0.3–1.2 mmHg more drop than with placebo, and less at the higher dose. No statistical test in the registry; according to the company's June 2024 announcement, a significant drop compared with placebo was not achieved and development of the drug was stopped Side effects: No serious side effects; non-serious side effects in 14 of 19 patients in the 0.5% group, 19 of 19 in the 1% group and 8 of 18 with placebo (most common burning and stinging on instillation); vivid dreams in one patient in the 0.5% group | NCT06144918, November 2023–February 2024, results in the registry, January 14, 2026, no publication in PubMed as of October 2026) |
| Levitt 1981 Randomized controlled study (6 arms: no treatment, placebo, prochlorperazine, 3 THC doses); main aim chemotherapy nausea, with eye pressure as a physiological measure | 120 randomized (number per arm not in the abstract) Cancer patients about to receive potentially nausea-causing chemotherapy for the first time; no glaucoma | Oral THC | 5, 10 or 15 mg THC, 4 times 4 hours apart (starting 2 hours before chemotherapy) | 1 day (4 doses 4 hours apart) | All eye pressures stayed within normal limits before and after treatment; the abstract reports no drop in pressure with THC; on the contrary, an unexpected, statistically significant rise was seen in the 5 mg THC group (the numbers and what the comparison was made against are not in the abstract). Full text could not be read Side effects: According to the abstract, no marked physiological side effects were seen in patients (blood pressure and lung function were also measured) | PMID 6271814, doi:10.1002/j.1552-4604.1981.tb02583.x |
| Newell 1979 Uncontrolled clinical study (no comparison group in the abstract); plus a rabbit experiment | not reported (in the abstract); 18 according to the Passani 2020 review (14 open-angle glaucoma, 4 ocular hypertension) Open-angle glaucoma | Oral nabilone (a synthetic THC-like cannabinoid) | Single dose 0.5–2 mg | Single dose (3 weeks in one patient, according to the review) | Eye pressure fell by 10–54%, 28% on average. In rabbits, nabilone given as eye drops lowered pressure, which returned to baseline in 3 hours, and tolerance developed within a week. No control group; the only human data with nabilone in glaucoma Side effects: not reported (in the abstract); few and self-limiting according to the review | PMID 530560, doi:10.1016/s0161-6420(79)35539-7 |
| Dhivagaran 2026 Systematic review and meta-analysis (PROSPERO CRD420251007916; MEDLINE, EMBASE, Web of Science; random-effects model) | 5 studies; 99 people (69 received THC) Studies reporting the drop in eye pressure after THC | THC (intravenous, oral, eye drops) | not reported (in the abstract) | The largest drop in eye pressure after THC averaged 14.7% (95% CI 3.4–25.9). By route: intravenous 33.3% (95% CI 20.4–46.2); oral 10.7% (−7.6 to 28.9) and eye drops 9.4% (−8.9 to 27.6), not significant. In the 4 studies with a comparison group, the drop after THC (6.9%; 95% CI −9.6 to 23.3; p = 0.41) was not different from control. Because the full text could not be read, which 5 studies were pooled is not known | PMID 42069343, doi:10.1016/j.jcjo.2026.04.003 | |
| Bilbao 2022 Systematic review and meta-analysis (PROSPERO CRD42021229932; RCTs of dronabinol, nabilone, CBD and nabiximols in 22 conditions; GRADE) | 152 RCTs (12,123 people); 1 RCT in glaucoma (6 people) All medical indications; the glaucoma section covers ocular hypertension or early glaucoma | Dronabinol, nabilone, CBD, nabiximols | In glaucoma, only a single small crossover RCT (Tomida 2006) was found: temporary improvement in the THC arm, which the review classed as dronabinol (SMD −1.28; 95% CI −2.36 to −0.20; p = 0.02); a non-significant temporary worsening in the arm the review classed as 'CBD containing a small amount of THC' (p = 0.08). Because there was only one study, the evidence in glaucoma was not graded. No conflicts of interest were reported | PMID 35982439, doi:10.1186/s12916-022-02459-1 | ||
| NASEM 2017 (US National Academies report) Expert committee report (assessment based on good-quality systematic reviews and later primary studies) | In glaucoma, 1 good-quality systematic review (Whiting 2015), which itself contained a single RCT (6 people) Glaucoma | Cannabinoids (sublingual THC and CBD) | In the single crossover RCT found by Whiting 2015 (Tomida 2006), no difference was found in eye pressure between placebo and cannabinoids. Studies reporting positive results showed only drops lasting hours. The committee's conclusion (4-14): there is limited evidence that cannabinoids are ineffective in improving glaucoma-related eye pressure | PMID 28182367, doi:10.17226/24625, NCBI Bookshelf NBK423845, source of the glaucoma finding Whiting 2015, PMID 26103030 | ||
| EGS 2025 (European Glaucoma Society guideline) Clinical guideline, 6th edition (evidence-based questions; level of evidence and strength of recommendation stated) | Glaucoma and ocular hypertension | Cannabinoids | There is no evidence supporting a place for cannabinoids, acupuncture or other alternative therapies in treating glaucoma (low level of evidence, strong recommendation). The treatment section also states that there is no evidence that nutritional supplements or cannabinoids have a place in glaucoma (based on Whiting 2015) | PMID 41026937, doi:10.1136/bjophthalmol-2025-egsguidelines, Br J Ophthalmol 2025, 109(Suppl 1) | ||
| Jampel 2010 (American Glaucoma Society position) Professional society position statement (approved by the board of directors on October 23, 2009; author H. Jampel) | Glaucoma | Marijuana and THC (smoked, oral, sublingual, eye drops) | Although marijuana lowers eye pressure, the effect lasts 3–4 hours (keeping it low all day would require smoking every 3 hours) and it affects mood and attention; 9 of 9 patients using oral THC or marijuana stopped within 9 months because of side effects; drops cannot get into the eye in sufficient amounts; lowering blood pressure may reduce blood flow to the optic nerve and cause harm. Because of its side effects, short duration of action and the lack of evidence that it changes the course of glaucoma, it is not recommended in any form | PMID 20160576, doi:10.1097/IJG.0b013e3181d12e39, Text on the society's website (americanglaucomasociety.net/about/statements, read on October 28, 2026) | ||
| Katsimpris 2023 (Mendelian randomization) Two-sample Mendelian randomization (an analysis that uses genetic predisposition like a natural experiment) | Lifetime cannabis use GWAS 184,765 people; cannabis use disorder 17,068 cases and 357,219 controls; glaucoma GWAS 16,677 cases and 199,580 controls Adults of European ancestry (genetic databases) | Cannabis use (genetic predisposition) | No cause-and-effect link was found between cannabis use and primary open-angle glaucoma: OR 1.04 for lifetime use (95% CI 0.88–1.23), OR 0.97 for cannabis use disorder (0.92–1.03). Sensitivity analyses pointed the same way. The genetic markers used explain only a very small part of the variation in cannabis use (0.08–0.09%) | PMID 37949880, doi:10.1038/s41598-023-45872-z, Preprint: Research Square (Europe PMC PPR585420) | ||
| Lehrer 2022 (UK Biobank) Cross-sectional analysis (UK Biobank) | 815 people with glaucoma and cannabis information; the paper gives two different numbers for the eye pressure analysis (38,816 and 29,388) Middle-aged adults in the UK (mean age 56) | Cannabis use (self-reported) | After accounting for age and smoking, cannabis use was not linked with primary open-angle glaucoma (OR 1.03; 95% CI 0.91–1.17; p = 0.66). In those who had used it 100 times or more, glaucoma had been diagnosed on average 10 years younger than in those who had never used it (p = 0.05). Eye pressure was lower in those who had used it 11–100 times. Participants had not used it for an average of 24 years; based on one-time information and simple analyses, it does not show cause and effect | PMID 35093258, doi:10.1016/j.jfo.2021.12.012 |
Reported side effects
- With THC-containing products: increased heart rate, a drop in blood pressure on standing up, a racing heart and mental changes (Crawford and Merritt 1979; Merritt 1980; Cooler and Gregg 1977)
Monitoring done in studies
- If someone with glaucoma is going to use CBD, their eye doctor should monitor the pressure inside the eye
- THC-containing products can make the pressure inside the eye read low for a few hours; if one was used before an eye exam, the eye doctor should be told
Who should be careful
- The statement 'CBD raises eye pressure in animals' is not correct on its own; the animal data are mixed
- CBD is not a treatment recommendation for glaucoma
Abbreviations: RCT, randomized controlled trial (a study in which participants are assigned to groups by chance); placebo, an inactive substance; SMD, OR and RR, effect measures used in analyses that pool studies; CYP, liver enzymes that break down medicines.
If you are considering CBD
- Do not stop your eye drops or your glaucoma treatment; CBD is not a treatment recommendation for glaucoma.
- If you are thinking about using CBD, tell your eye doctor; the pressure inside your eye needs to be monitored.
- If you take medicines regularly, talk with your doctor or pharmacist about drug interactions.
- Check the product's label and certificate of analysis (COA); find out whether it contains THC.
Choose your age group, medicines and situation to prepare a short summary you can take to your doctor. Nothing you enter is sent anywhere.
Bottom line
In glaucoma, CBD has not been shown to lower eye pressure; the temporary rise seen at the high dose in the single small human study was also not significant when compared with placebo. The most important step to protect your vision is to keep up the treatment your eye doctor recommends and not to miss your checkups.
Frequently asked questions
Does CBD oil help with glaucoma?
There is no evidence showing that it does. In the only small human study, CBD did not lower the pressure inside the eye at all; the temporary rise seen four hours after the higher dose was also not significant when compared with placebo. That is why CBD is not a treatment recommendation for glaucoma; the European Glaucoma Society's 2025 guideline also states that there is no evidence that cannabinoids have a place in treating glaucoma.
Doesn't cannabis lower eye pressure?
THC in cannabis can lower the pressure inside the eye, but this effect lasts about 4 hours and also causes a high or mental fogginess. Keeping the pressure low all day would mean using it about every 3 hours; in the studies, a drop in blood pressure on standing up and a racing heart were common, and all 9 patients who used oral THC or marijuana stopped treatment within 1–9 months. For these reasons, the European and American glaucoma societies do not recommend it. CBD, on the other hand, did not show this effect.
Does CBD raise eye pressure?
In the only human study (6 patients), pressure rose from 23.2 to 25.9 mmHg 4 hours after 40 mg of CBD; but in a meta-analysis that re-analyzed the same study against placebo, the difference between CBD and placebo was not significant. So this single, small study does not prove that CBD raises pressure, nor does it show that it lowers it. The results of animal studies are mixed; in some, pressure fell or did not change. If someone with glaucoma is going to use CBD, it is important for their eye doctor to monitor the pressure.
Do CBD eye drops help with glaucoma?
CBD applied to the eye has been tested only in mice: in one study, pressure decreased or did not change; in the other, CBD showed opposing effects and blocked THC's pressure-lowering effect. There are no human studies of CBD eye drops. THC eye drops, meanwhile, did not lower pressure in three controlled studies of patients with glaucoma and healthy volunteers; a new drop that turns into THC in the eye was also not clearly different from placebo in a two-week placebo-controlled study of 56 patients with glaucoma or ocular hypertension.
In Turkey, can CBD oil be obtained with a prescription for glaucoma?
In Turkey, products derived from cannabis are regulated by the Kenevirden Elde Edilen Ürünlere Dair Yönetmelik (Regulation on Products Derived from Hemp) dated January 31, 2026. These products, including CBD, can be sold only in pharmacies. Those classed as medicines or health products are dispensed with a prescription; the regulation does not state a prescription requirement for supplement products and cosmetics with a THC content of no more than 0.3%. As of October 2026, there is no approved (licensed) CBD medicine in Turkey. If you are thinking about using it, talk with your doctor first.
Does using cannabis increase the risk of glaucoma?
Few studies have looked at this. An analysis using genetic data (Mendelian randomization) found no cause-and-effect link between cannabis use and glaucoma. In a large UK health database, glaucoma was also not more common among cannabis users; however, in those who had used it 100 times or more, glaucoma had been diagnosed on average 10 years younger. This finding is borderline significant, relies on one-time information and simple analyses, and does not show cause and effect. Cannabis is not recommended for preventing or treating glaucoma.
References
Show 24 references
- Tomida 2006. Double-masked, placebo-controlled, 4-arm crossover pilot RCT. PMID 16988594, doi:10.1097/01.ijg.0000212260.04488.60
- Miller 2018. Animal (mouse), topical. PMID 30550613, doi:10.1167/iovs.18-24838
- Rebibo 2022. Animal (mouse), topical nanoemulsion. PMID 35245638, doi:10.1016/j.ijpharm.2022.121627
- Bazzano 2025. Animal (healthy horse), oral; no placebo, compared with the same horses' measurements from the previous day. PMID 40435047, doi:10.1371/journal.pone.0325191
- Crawford and Merritt 1979. Controlled clinical study (compared with placebo). PMID 468444
- Merritt 1980. Labeled in PubMed as a randomized controlled trial; the abstract does not mention a placebo and the full text could not be read. According to one review, patients smoked marijuana with 2% THC or a placebo cigarette; according to another, it was a single-arm study. PMID 7053160, doi:10.1016/s0161-6420(80)35258-5
- Flach 2002. Uncontrolled, open, non-randomized compassionate-use program. PMID 12545695, PMC1358964
- Lindner 2026. Randomized, double-masked, placebo-controlled crossover pilot study. PMID 40772417, doi:10.1111/aos.17573, NCT04596826
- Hommer 2020. Randomized, double-masked, placebo-controlled, two-way crossover RCT (healthy volunteers). PMID 31977076, doi:10.1002/cpt.1797, NCT03907163, design, number of people and primary outcome, no registration number in the paper, no results in the registry). The same group's paper measuring retinal blood flow with the same design (24 healthy volunteers, 5 mg): Hommer 2021, PMID 33999707, whether it is from the same trial could not be verified
- Merritt 1981. Randomized, balanced, double-masked controlled study (abstract); according to the review, THC in one eye and vehicle oil alone in the other. PMID 6271841, doi:10.1002/j.1552-4604.1981.tb02626.x, The same group's second THC eye drop study in 8 patients: Merritt 1981, PMID 6114151, no abstract, could not be read, no difference from placebo according to the Passani 2020 review
- Jay and Green 1983. Randomized, vehicle oil-controlled study (healthy volunteers). PMID 6301412, doi:10.1001/archopht.1983.01040010591012
- Green and Roth 1982. Randomized, vehicle oil-controlled study (applied to one eye; healthy volunteers). PMID 6279061, doi:10.1001/archopht.1982.01030030267006
- Flom 1975. Controlled clinical study (PubMed publication type); double-blind with placebo cigarettes according to the review. PMID 1089090
- Cooler and Gregg 1977. Controlled clinical study (compared with placebo and diazepam; randomized according to the review). PMID 329423, doi:10.1097/00007611-197708000-00016
- NCT06144918 (SBI-100 eye drops, registry results). Multicenter, double-masked, placebo-controlled, randomized parallel phase 2 study; results only in the registry (no peer-reviewed publication); sponsor is the drug's developer (Skye Bioscience). NCT06144918, November 2023–February 2024, results in the registry, January 14, 2026, no publication in PubMed as of October 2026)
- Levitt 1981. Randomized controlled study (6 arms: no treatment, placebo, prochlorperazine, 3 THC doses); main aim chemotherapy nausea, with eye pressure as a physiological measure. PMID 6271814, doi:10.1002/j.1552-4604.1981.tb02583.x
- Newell 1979. Uncontrolled clinical study (no comparison group in the abstract); plus a rabbit experiment. PMID 530560, doi:10.1016/s0161-6420(79)35539-7
- Dhivagaran 2026. Systematic review and meta-analysis (PROSPERO CRD420251007916; MEDLINE, EMBASE, Web of Science; random-effects model). PMID 42069343, doi:10.1016/j.jcjo.2026.04.003
- Bilbao 2022. Systematic review and meta-analysis (PROSPERO CRD42021229932; RCTs of dronabinol, nabilone, CBD and nabiximols in 22 conditions; GRADE). PMID 35982439, doi:10.1186/s12916-022-02459-1
- NASEM 2017 (US National Academies report). Expert committee report (assessment based on good-quality systematic reviews and later primary studies). PMID 28182367, doi:10.17226/24625, NCBI Bookshelf NBK423845, source of the glaucoma finding Whiting 2015, PMID 26103030
- EGS 2025 (European Glaucoma Society guideline). Clinical guideline, 6th edition (evidence-based questions; level of evidence and strength of recommendation stated). PMID 41026937, doi:10.1136/bjophthalmol-2025-egsguidelines, Br J Ophthalmol 2025, 109(Suppl 1)
- Jampel 2010 (American Glaucoma Society position). Professional society position statement (approved by the board of directors on October 23, 2009; author H. Jampel). PMID 20160576, doi:10.1097/IJG.0b013e3181d12e39, Text on the society's website (americanglaucomasociety.net/about/statements, read on October 28, 2026)
- Katsimpris 2023 (Mendelian randomization). Two-sample Mendelian randomization (an analysis that uses genetic predisposition like a natural experiment). PMID 37949880, doi:10.1038/s41598-023-45872-z, Preprint: Research Square (Europe PMC PPR585420)
- Lehrer 2022 (UK Biobank). Cross-sectional analysis (UK Biobank). PMID 35093258, doi:10.1016/j.jfo.2021.12.012
Points that could not be verified from the full text
- The full texts of studies from the 1970s and 1980s (Flom 1975, Cooler and Gregg 1977, Newell 1979, Crawford and Merritt 1979, Merritt 1980, Merritt 1981, Levitt 1981, Green and Roth 1982, Jay and Green 1983) and of Tomida 2006 could not be read. Details come from the PubMed abstract; participant numbers and design details not in the abstract were taken from the Passani 2020 review. Reviews describe differently whether Merritt 1980 was placebo-controlled (according to one, patients also smoked placebo cigarettes; according to the other, it was a single-arm study). For Tomida 2006, the THC content of the extracts used, the authors' ties to the drug company, and whether the rise after 40 mg of CBD was calculated against baseline or against placebo could not be verified.
- The Canadian Ophthalmological Society's position statement on medical marijuana in glaucoma (2010) and its 2019 editorial could not be read because the publisher's site required bot verification. The official assessment text of the American Academy of Ophthalmology on cannabis in glaucoma could not be accessed. The full text of the 2026 meta-analysis (Dhivagaran 2026) could not be read; which 5 studies it pooled is not known.
- A small randomized study of smoked cannabis in healthy volunteers (Mosaed 2020; 11 people cannabis, 3 people placebo) is not in PubMed and could not be read because the journal page required a login; according to one review, pressure fell from baseline but there was no consistent difference from placebo. An Israeli registry tracking the pressure inside the eye in medical cannabis users (NCT02080676) has not been updated since 2015 and has no results (October 2026).
Trusted sources from around the world
Show 3 references
- European Glaucoma Society guideline, 6th edition (2025)European Glaucoma Society (EGS); British Journal of Ophthalmology supplement, English, September 2025
The European Glaucoma Society's guideline states that there is no evidence supporting a role for cannabinoids in the treatment of glaucoma (low level of evidence, strong recommendation).
- American Glaucoma Society: Position statement on marijuana and the treatment of glaucomaAmerican Glaucoma Society (AGS), English, October 2009 (pediatric glaucoma statement 2019)
The Society writes that marijuana lowers eye pressure for only 3–4 hours, and that because of its side effects, short duration of action and the lack of evidence that it changes the course of the disease, it cannot currently be recommended in any form for treating glaucoma. The 2019 pediatric glaucoma statement on the same page also does not support the use of cannabinoids in childhood glaucoma.
- American Academy of Ophthalmology: Does marijuana help treat glaucoma? (patient page)American Academy of Ophthalmology (AAO), English, January 2026
The Academy does not recommend marijuana or other cannabis products for treating glaucoma until more research is done. It explains that because eye pressure must be kept under control 24 hours a day, cannabis products, including CBD, are not a practical treatment, and that there is no convincing research showing CBD is effective in glaucoma.
These sources sell no products. All trusted sources
Research history
Here we record what changed on this page, when and why, with dates. The newest change is at the top.
The last sentence of the page's short summary was rewritten in a calmer, more guiding tone: instead of "Talk to your doctor before you start.", it now says whom you can talk with, and about what, if you are considering CBD. The conclusion and the strength of evidence did not change.
The finding label changed from 'May be harmful' to 'No effect found'; the strength of evidence stayed 'Insufficient'. The label had been based on the temporary pressure rise seen 4 hours after 40 mg of CBD in the single human study (Tomida 2006, 6 patients, single dose). In the October 2026 search, sources that re-analyzed this study against placebo (Bilbao 2022, NASEM 2017) found no significant difference between CBD and placebo; in most animal studies, CBD also did not raise pressure. Nor has CBD been shown to lower eye pressure. Because the evidence rests on a single, very small study, it is 'Insufficient'.
Studies that led to the change: Tomida 2006, published October 2006 (PMID 16988594, doi:10.1097/01.ijg.0000212260.04488.60); Bilbao 2022, published August 19, 2022 (PMID 35982439, doi:10.1186/s12916-022-02459-1); NASEM 2017 (US National Academies report), published January 2017 (PMID 28182367, doi:10.17226/24625, NCBI Bookshelf NBK423845, source of the glaucoma finding Whiting 2015, PMID 26103030)
The October 2026 search added 16 sources: two controlled studies of THC eye drops in patients with glaucoma or ocular hypertension (one of them the 2-week study of a new drop that turns into THC in the eye in 56 patients, with results only on ClinicalTrials.gov); studies of smoked, intravenous and oral THC and of THC eye drops in healthy volunteers, and an RCT measuring optic nerve blood flow; a placebo-controlled study of oral THC in cancer patients receiving chemotherapy that also measured the pressure inside the eye; an uncontrolled study with nabilone; two systematic reviews and meta-analyses; the US National Academies report; the European Glaucoma Society guideline and the American Glaucoma Society position statement; and two observational studies on cannabis use and glaucoma risk. The Tomida 2006, Lindner 2026 and Merritt 1980 rows were corrected from the primary source; we added that the temporary pressure rise seen after 40 mg of CBD in the single human study was not significant in sources that re-analyzed this study against placebo; we noted in the 'fell in horses' sentence that the study had no placebo; and the THC findings and search counts were worked into the text. The finding and the strength of evidence did not change.
Related studies: Hommer 2020, published July 2020 (PMID 31977076, doi:10.1002/cpt.1797, NCT03907163, design, number of people and primary outcome, no registration number in the paper, no results in the registry). The same group's paper measuring retinal blood flow with the same design (24 healthy volunteers, 5 mg): Hommer 2021, PMID 33999707, whether it is from the same trial could not be verified); Merritt 1981, published August–September 1981 (PMID 6271841, doi:10.1002/j.1552-4604.1981.tb02626.x, The same group's second THC eye drop study in 8 patients: Merritt 1981, PMID 6114151, no abstract, could not be read, no difference from placebo according to the Passani 2020 review); Jay and Green 1983, published April 1983 (PMID 6301412, doi:10.1001/archopht.1983.01040010591012); Green and Roth 1982, published February 1982 (PMID 6279061, doi:10.1001/archopht.1982.01030030267006); Flom 1975, published January 1975 (PMID 1089090); Cooler and Gregg 1977, published August 1977 (PMID 329423, doi:10.1097/00007611-197708000-00016); Newell 1979, published January 1979 (PMID 530560, doi:10.1016/s0161-6420(79)35539-7); Dhivagaran 2026, published October 2026 (online August 5, 2026) (PMID 42069343, doi:10.1016/j.jcjo.2026.04.003); Bilbao 2022, published August 19, 2022 (PMID 35982439, doi:10.1186/s12916-022-02459-1); NASEM 2017 (US National Academies report), published January 2017 (PMID 28182367, doi:10.17226/24625, NCBI Bookshelf NBK423845, source of the glaucoma finding Whiting 2015, PMID 26103030); EGS 2025 (European Glaucoma Society guideline), published September 8, 2025 (PMID 41026937, doi:10.1136/bjophthalmol-2025-egsguidelines, Br J Ophthalmol 2025, 109(Suppl 1)); Jampel 2010 (American Glaucoma Society position), published February 2010 (PMID 20160576, doi:10.1097/IJG.0b013e3181d12e39, Text on the society's website (americanglaucomasociety.net/about/statements, read on October 28, 2026)); Katsimpris 2023 (Mendelian randomization), published November 10, 2023 (PMID 37949880, doi:10.1038/s41598-023-45872-z, Preprint: Research Square (Europe PMC PPR585420)); Lehrer 2022 (UK Biobank), published April 2022 (PMID 35093258, doi:10.1016/j.jfo.2021.12.012); NCT06144918 (SBI-100 eye drops, registry results), published January 14, 2026 (registry results) (NCT06144918, November 2023–February 2024, results in the registry, January 14, 2026, no publication in PubMed as of October 2026)); Levitt 1981, published August–September 1981 (PMID 6271814, doi:10.1002/j.1552-4604.1981.tb02583.x)
Older changes (3)
We clarified that the CBD in the human study was not pure but a CBD-dominant extract derived from cannabis; added that the horse study had no placebo; added four THC studies to the table; and corrected the year of one source. The finding and the strength of evidence did not change.
The names of the finding labels were simplified: on this page, the "Possible harm" label became "May be harmful". The assessment did not change.
Page published. This first review examined 4 studies.