Oxymetazoline hydrochloride

What it is

Oxymetazoline hydrochloride is a manufactured imidazoline compound that narrows small blood vessels where it is applied. It has three quite different licensed uses in three different products: a 0.05% nasal spray sold over the counter for a blocked nose, a 1% cream for the persistent facial redness of rosacea, and 0.1% eye drops for a drooping upper eyelid. In the cream, each gram holds 10 mg of the hydrochloride salt, equal to 8.8 mg of oxymetazoline free base. Because it is applied to the surface rather than swallowed, each route has its own evidence and its own harms.

What the research says

It tightens blood vessels wherever it is put, and the three uses are simply three consequences of that. For rosacea redness, two identical vehicle-controlled trials in 885 people found a real but small effect: roughly 12 to 18 in 100 reached the success endpoint on the cream against 5 to 9 in 100 on the vehicle. For a drooping eyelid, the body of a 2025 meta-analysis of four randomised trials gives a pooled lift of the lid margin of 0.37 mm more than placebo (95% CI 0.13 to 0.60), measured only out to 14 days; that paper’s abstract prints a larger figure that its own results section does not support, which is set out in the findings below. For a blocked nose it works fast, but prolonged use is the problem: a randomised trial found measurable rebound swelling after 30 days whether the spray was used once or three times a day, and another trial measured a large fall in nasal airflow after 14 days that an intranasal steroid reversed. A third trial, also in healthy people, did not detect rebound after four weeks, so the rebound literature is genuinely split — and all three trials were in healthy volunteers rather than people with a cold.

Evidence grade: Well established.

How it works

Drug class: Imidazoline-derived alpha-adrenoceptor agonist, used as a topical vasoconstrictor; the rosacea cream label calls it an alpha1A adrenoceptor agonist, the eye-drop label only an alpha adrenoceptor agonist

Oxymetazoline is an alpha1A adrenoceptor agonist and acts as a vasoconstrictor: it narrows small blood vessels wherever it is applied. That one action explains all three products. In the nose, narrowed vessels mean a less engorged lining and a wider airway; on the face, less visible redness from dilated vessels; in the eyelid, a lift of a drooping lid. The eye-drop label describes that eyelid target separately and more cautiously, as a subset of adrenoreceptors in Mueller’s muscle, and does not name a receptor subtype at all. (Source 1)

What it is used for

• Two identical vehicle-controlled trials reported in the US prescribing information, and an independent meta-analysis, agree it works. The absolute benefit in those two trials is small: about 6 to 12 extra people in 100 reached the success endpoint. The meta-analysis reports relative risks only and prints no absolute rates, no risk differences and no participant count. Dermatitis where the cream goes is the main cost. Evidence: established. (Source 2)
• The meta-analysis’s own results section gives a pooled lift of the lid margin of 0.37 mm more than placebo (95% CI 0.13 to 0.60), graded moderate certainty, with no excess of serious adverse events. Upper visual field also improved. The gain is short-term only: the authors list the short follow-up of the existing trials, 14 days, as a limitation, call oxymetazoline a short-term treatment, and say long-term trials and direct comparisons against surgery are still needed. Evidence: established. (Source 3)
• It decongests quickly, but the trials on prolonged use conflict, and all of them were in healthy volunteers rather than people with a cold: one found rebound swelling after 30 days, another measured a large loss of nasal airflow after 14 days, and a third did not detect either after 4 weeks. The over-the-counter label limits use to 3 days. Evidence: limited. (Source 4)
• In a 2026 meta-analysis of two old randomised trials the effect was modest and inconsistent, RR 0.75 with an upper limit of 0.99, only just excluding 1, and much weaker than oral pseudoephedrine in the same review. The review pooled with a fixed-effect model even though heterogeneity for oxymetazoline was moderate, which makes that interval narrower than a random-effects analysis would give. Evidence: limited. (Source 5)

Interactions

• Monoamine oxidase inhibitors (MAOIs) (label): Unlike the oral decongestants, the prescription oxymetazoline labels treat MAOIs as a caution rather than an absolute contraindication, on the basis that MAO inhibition changes how circulating amines are handled. (Source 6)
• Beta-blockers, blood pressure medicines and cardiac glycosides such as digoxin (label): Because the whole alpha-agonist class can move blood pressure, the label advises caution when these are taken together. This is a class-based caution, not a measured interaction study. (Source 6)
• Alpha-blockers, including those used for an enlarged prostate or Raynaud's disease (label): These drugs block the same receptors oxymetazoline switches on, so the label advises caution in people taking them. (Source 6)
• Intranasal fluticasone propionate, a corticosteroid spray (clinical trial): In a randomised crossover trial in 19 healthy volunteers, adding fluticasone after 14 days of oxymetazoline reversed both the loss of response and the rebound congestion. The same authors say further studies are needed to see whether combined decongestant-and-corticosteroid sprays actually work as a strategy, so this is a promising mechanism rather than an established treatment. (Source 7)
• Dietary supplements and foods (theoretical): We found no study of oxymetazoline with any dietary supplement, food or alcohol. What exists is the label's class statement that alpha-adrenergic agonists may affect blood pressure, which is the basis for caution with any stimulant supplement that does the same. Treat this as an absence of evidence rather than evidence of safety. (Source 6)

Stopping it

• For the nasal spray, the label puts a hard limit on how long to continue, and names the reason: congestion that comes back or gets worse. (Source 8)
• The strongest measurement of what prolonged use does comes from a randomised double-blind study: after 30 days, rebound swelling of the nasal lining and stuffiness appeared whether the spray was used once a day at night or three times a day, with mean swelling of 0.8 mm and 1.1 mm respectively, together with increased histamine sensitivity. (Source 4)
• What this does to someone trying to stop is the core of rhinitis medicamentosa: the spray works for shorter and shorter periods, so people use more of it, which makes the congestion worse again. A 2025 pharmacology review describes that loop and the dependency it creates, and says the mechanism is still not fully understood. (Source 9)
• In the trial that produced the loss of response on purpose, adding intranasal fluticasone after 14 days of oxymetazoline brought peak nasal inspiratory flow back up by 45 L/min and shifted the dose-response curve back up. That was in 19 healthy volunteers in whom tachyphylaxis had been induced experimentally, not in people with rhinitis medicamentosa trying to stop, so it is a mechanism result rather than a treatment result. (Source 10)
• Beyond stopping plus a steroid spray, the treatment evidence is thin. A 2025 systematic review of rhinitis medicamentosa treatments found only 12 studies and 373 patients in total, and said their designs were too different to build a single treatment algorithm from. (Source 11)

What goes wrong

The same meta-analysis reported more adverse events on oxymetazoline than on vehicle, driven almost entirely by dermatitis where the cream was applied; the interval it prints for the overall adverse-event result is internally impossible, so only the dermatitis result can be relied on. (Source 12)

• Meta-analysis, Moderate certainty.
• Size: number of participants not stated in the abstract.
• Who: adults with rosacea.
• How long: up to 12 hours per assessment day in the pooled trials.
• Result: treatment-emergent adverse events reported as RR 1.34 with the interval printed as 95% CI: 1.10-1.2, which cannot be right because the upper limit is below the point estimate; application-site dermatitis RR 8.91 (95% CI 1.76 to 45.23) on few events; no other individual adverse event differed significantly.
• Funding: not stated.

Limit of this finding: The published paper prints the overall adverse-event result as “RR = 1.34, 95% CI: 1.10-1.2”. An upper limit of 1.2 sits below the point estimate of 1.34, so that interval cannot be right. Both PubMed and the publisher’s own deposited abstract carry it that way and the true upper limit is not reachable anywhere we could look, so the overall adverse-event figure should not be treated as a sound estimate. The separate dermatitis result, RR 8.91 (95% CI 1.76 to 45.23), does hold together, but it rests on very few events and its interval is so wide that it should not be reported as an 8.9-fold risk.

the pooled results show that the incidence of TEAEs after treatment with oxymetazoline is significantly higher than that of vehicle (RR = 1.34, 95% CI: 1.10-1.2). However, our analysis of specific adverse events found that the oxymetazoline group was only significantly higher than the vehicle group in the incidence of application-site dermatitis (RR = 8.91, 95% CI: 1.76-45.23), and there was no statistical significance in the difference in the incidence of other adverse events.

In the pivotal trials the commonest harm was dermatitis where the cream was applied, in about 2 in 100 people against none on vehicle. (Source 13)

• Randomized trial, Moderate certainty.
• Size: 489 subjects on oxymetazoline cream and 483 on vehicle cream in 3 controlled trials, plus 440 subjects in a one-year open-label trial.
• Who: adults with persistent facial erythema of rosacea.
• How long: 4 weeks in the controlled trials; up to one year open-label.
• Result: application site dermatitis 9 (2%) versus 0; worsening inflammatory lesions of rosacea 7 (1%) versus 1 (<1%); application site pruritus 5 (1%) versus 4 (1%); application site erythema 5 (1%) versus 2 (<1%); application site pain 4 (1%) versus 1 (<1%). Over one year, rates were 3% or less for each.
• Funding: not stated; reported in the manufacturer's US prescribing information.

Application site dermatitis 9 (2%) 0 Worsening inflammatory lesions of rosacea 7 (1%) 1 (<1%) Application site pruritus 5 (1%) 4 (1%) Application site erythema 5 (1%) 2 (<1%) Application site pain 4 (1%) 1 (<1%)

A pharmacovigilance analysis of five years of FDA adverse event reports found signals for the eye drops that are not on the label, including retinal and vitreous events, mydriasis, raised blood pressure and tachyphylaxis. (Source 14)

• Survey study, Low certainty.
• Size: 306 patients reporting 658 adverse drug events.
• Who: real-world users of oxymetazoline 0.1% eye drops in the United States.
• How long: reports from approval on 7 January 2020 through 30 April 2025.
• Result: 30 significant safety signals. Events already on the drug label, with the counts and percentages exactly as the paper prints them: ocular surface disease 129 (34.5%), conjunctival hyperaemia 43 (11.5%), temporary visual blurring 33 (8.8%), instillation site pain 27 (7.2%), headache 32 (8.6%). Events not on the label: vitreoretinal complications 12 (3.2%), mydriasis 21 (5.6%) with reporting odds ratio 175.4 (112.4 to 273.7), hypertension 6 (1.6%) ROR 3.2 (1.4 to 7.2), tachyphylaxis 5 (1.3%) ROR 5.1 (2.1 to 12.4). Spontaneous reports cannot give an incidence and do not establish cause.
• Funding: not stated.

Limit of this finding: The percentages in this paper do not divide by either total it states. It reports 306 patients and 658 adverse drug events, yet every percentage is consistent with a denominator of about 374 instead: 129 out of 306 would be 42%, not 34.5%, and 129 out of 658 would be 20%. Read the counts rather than the percentages, and do not read 34.5% as “a third of patients”. The study is also a disproportionality analysis of spontaneous reports, which cannot measure how often anything happens.

Novel ADEs (not on the drug label) were vitreoretinal complications (N = 12, 3.2%, multiple RORs, all p < 0.001)-retinal detachment (N = 3, 0.8%), vitreous detachment (N = 3, 0.8%), vitreous floaters (N = 3, 0.8%), and vitreous hemorrhage (N = 3, 0.8%)-mydriasis (N = 21, 5.6%, ROR 175.4 [112.4-273.7], p < 0.001), hypertension (N = 6, 1.6%, ROR 3.2 [1.4-7.2], p = 0.009), and tachyphylaxis (N = 5, 1.3%, ROR 5.1 [2.1-12.4], p < 0.001).

In the pivotal eye-drop trials, one to five people in a hundred had punctate keratitis, red eye, dry eye, blurred vision, stinging, eye irritation or headache. (Source 15)

• Randomized trial, Moderate certainty.
• Size: 360 subjects treated with the eye drops across three controlled phase 3 trials.
• Who: adults with acquired blepharoptosis.
• How long: 6 weeks in 203 subjects and 12 weeks in 157 subjects.
• Result: adverse reactions in 1 to 5% of treated subjects were punctate keratitis, conjunctival hyperemia, dry eye, blurred vision, instillation site pain, eye irritation and headache; the label gives no separate vehicle rates here.
• Funding: not stated; reported in the manufacturer's US prescribing information.

Adverse reactions that occurred in 1-5% of subjects treated with UPNEEQ were punctate keratitis, conjunctival hyperemia, dry eye, blurred vision, instillation site pain, eye irritation and headache.

Children who have swallowed oxymetazoline drops or sprays have been hospitalised with serious effects including coma. (Source 16)

• Case series, Low certainty.
• Size: number of cases not stated.
• Who: children with accidental oral ingestion of imidazoline-containing topical solutions.
• How long: acute.
• Result: reported effects requiring hospitalisation include nausea, vomiting, lethargy, tachycardia, decreased respiration, bradycardia, hypotension, hypertension, sedation, somnolence, mydriasis, stupor, hypothermia, drooling and coma.
• Funding: not applicable; a prescribing-information safety statement.

Accidental oral ingestion of topical intended solutions (including ophthalmic solutions and nasal sprays) containing imidazoline derivatives (e.g., oxymetazoline) in children has resulted in serious adverse events requiring hospitalization, including nausea, vomiting, lethargy, tachycardia, decreased respiration, bradycardia, hypotension, hypertension, sedation, somnolence, mydriasis, stupor, hypothermia, drooling, and coma. Keep UPNEEQ out of reach of children.

A randomised double-blind trial found rebound swelling and stuffiness after 30 days of oxymetazoline nasal spray, whether used once or three times a day. (Source 4)

• Randomized trial, Low certainty.
• Size: 20 healthy volunteers, 10 per group.
• Who: healthy volunteers with no rhinitis.
• How long: 30 days of spray, with rhinostereometry and histamine challenge before and after.
• Result: mean rebound swelling 0.8 mm (p < 0.01) with once-daily use and 1.1 mm (p < 0.01) with three-times-daily use; mean evening stuffiness score 43 on a 0-100 visual analogue scale in both groups (p < 0.05); increased histamine sensitivity in both groups; no significant difference between the two regimens.
• Funding: not stated.

After 30 days, rebound swelling and nasal stuffiness were found in both groups. In the group receiving oxymetazoline nasal spray once daily at night, the mean rebound swelling was 0.8 mm (p < 0.01) and the estimated mean symptom score for nasal stuffiness in the evening was 43 (p < 0.05). In the group receiving the same nasal spray three times daily, the mean rebound swelling was 1.1 mm (p < 0.01) and the mean evening symptom score was 43 (p < 0.05). The finding of an increase in histamine sensitivity in both groups was taken to indicate nasal hyperreactivity. There was no significant difference in the investigated variables between the two groups. It is concluded that the risk of developing rebound swelling and nasal hyperreactivity remains, whether oxymetazoline nasal spray is used once or three times a day for 30 days.

A crossover trial measured a large loss of nasal airflow after 14 days of oxymetazoline, and showed an intranasal steroid reversed it. (Source 10)

• Randomized trial, Moderate certainty.
• Size: 19 healthy subjects.
• Who: healthy volunteers.
• How long: oxymetazoline 200 mcg three times a day for 14 days, then fluticasone 200 mcg twice a day added for 3 more days.
• Result: peak nasal inspiratory flow fell by 47.9 L/min (95% CI -63.9 to -31.9; P < 0.001) by Day 14 and the dose-response curve shifted down by 24.8 L/min (20.3 to 29.3; P < 0.001). After fluticasone, flow rose 45 L/min (30 to 61; P < 0.001) and the curve shifted back up 26.2 L/min (21.7 to 30.7; P < 0.001)
• Funding: not stated; registered as NCT00487032.

On Day 14 versus Day 1, inspiratory flow decreased (mean difference, 95% confidence interval) (-47.9 L x min(-1); -63.9 to -31.9; P < 0.001) and the DRC shifted downward (24.8 L x min(-1); 20.3-29.3; P < 0.001).

A pharmacology review describes rhinitis medicamentosa as chronic congestion that responds less and less to the spray, driving people to use more of it. (Source 9)

• Expert review, not systematic, Low certainty.
• Size: not applicable.
• Who: people using over-the-counter topical nasal decongestants.
• How long: prolonged or excessive use.
• Result: no rates given; the review states the pathophysiology is not fully elucidated, and lists structural, functional and inflammatory mucosal changes and altered receptor sensitivity as hypotheses.
• Funding: not stated.

RM is characterised by the presence of chronic nasal congestion which responds less effectively and for progressively shorter periods to nasal vasoconstrictor treatments. Consequently, patients try to relieve symptoms by taking the nasal vasoconstrictor more frequently or at higher doses, which ultimately exacerbates the congestion further and creates a dependency on the nasal vasoconstrictor.

What the evidence supports

In two identical vehicle-controlled trials, a small minority of people using oxymetazoline cream reached the label’s endpoint of at least a 2-grade improvement in redness on both the clinician and the self-assessment scale, roughly twice as often as with the vehicle cream. (Source 2)

• Randomized trial, Moderate certainty.
• Size: 885 subjects aged 18 and over across two trials (222 and 224 on active, 218 and 221 on vehicle)
• Who: adults scoring moderate or severe for persistent facial erythema of rosacea on both clinician and self-assessment scales; 90% were Caucasian and 79% female.
• How long: once daily for 29 days.
• Result: composite success (at least a 2-grade improvement on both scales) on Day 29: at hour 12, 15% versus 6% in Trial 1 and 12% versus 6% in Trial 2; at hour 9, 18% versus 6% and 16% versus 9%. So roughly 6 to 12 extra people in 100 reached the endpoint.
• Funding: not stated; reported in the manufacturer's US prescribing information.

Time-point (Hour) Trial 1 Trial 2 Oxymetazoline HCl Cream (N=222) Vehicle Cream (N=218) Oxymetazoline HCl Cream (N=224) Vehicle Cream (N=221) 3 12% 6% 14% 7% 6 16% 8% 13% 5% 9 18% 6% 16% 9% 12 15% 6% 12% 6%

An independent meta-analysis confirmed oxymetazoline beat vehicle cream on both clinician and self-assessed rosacea redness at every time point. (Source 12)

• Meta-analysis, Moderate certainty.
• Size: number of participants not stated in the abstract; randomised oxymetazoline-versus-vehicle trials to May 2021.
• Who: adults with persistent facial erythema of rosacea.
• How long: assessments at 3, 6, 9 and 12 hours.
• Result: clinician erythema assessment success RR 1.76 (95% CI 1.53 to 2.03) at 3 hours falling to RR 1.41 (95% CI 1.18 to 1.67) at 12 hours; subject self-assessment success RR 1.65 to 1.78 across the same time points. These are relative risks; this abstract prints no absolute success rates, no risk differences and no participant count, so the absolute numbers must be read from the pivotal-trial entry.
• Funding: not stated.

Limit of this finding: The efficacy intervals quoted here all hold together and all exclude 1. Two cautions about the same paper, both the paper’s own: the adverse-event result in the next finding has an impossible interval, and this abstract prints no absolute success rates and no participant count at all, so the figures for how many people in a hundred benefit have to come from the pivotal-trial entry rather than from here.

The pooled results show that the 3 (RR = 1.76, 95% CI: 1.53-2.03), 6 (RR = 1.71, 95% CI: 1.47-2.00), 9 (RR = 1.63, 95% CI: 1.40-1.90), 12 (RR = 1.41, 95% CI: 1.18-1.67) -hours CEA success rate and the 3 (RR = 1.65, 95% CI: 1.34-2.03), 6 (RR = 1.75, 95% CI: 1.43-2.14), 9 (RR = 1.63, 95% CI: 1.33-2.00), 12 (RR = 1.78, 95% CI: 1.45-2.18) -hours SSA success rate after oxymetazoline treatment for rosacea is significantly higher than that of vehicle.

A small single-arm study in mild to moderate rosacea reported improvement with no side effects, but it had no control group. (Source 17)

• Case series, Very low certainty.
• Size: 15 patients.
• Who: patients with mild to moderate facial rosacea.
• How long: twice daily for 4 weeks.
• Result: clinician erythema assessment, patient self-assessment and erythema index all fell significantly at 2 and 4 weeks; the source prints these as “p-value: 0.001” with no inequality sign. No side effects were experienced by any patient. With no comparator, none of this separates a drug effect from natural variation.
• Funding: not stated.

CEA and PSA indexes decreased significantly 2 and 4 weeks after the treatment (p-value: 0.001). The erythema index also decreased significantly in both 2 weeks (p-value: 0.001) and 4 weeks (p-value: 0.001) after the treatment. The capillaroscopic pattern of lesions showed a decrease in telangiectasia and a reddish background of skin. No side effects of treatment were experienced by any of the patients.

Pooled randomised trials found oxymetazoline eye drops lifted a drooping upper eyelid by 0.37 mm more than placebo, a fraction of a millimetre, on moderate-certainty evidence. (Source 3)

• Meta-analysis, Moderate certainty.
• Size: 448 participants across 4 randomised controlled trials.
• Who: adults with acquired blepharoptosis.
• How long: pooled from baseline to day 14; the primary outcome was change in marginal reflex distance 1.
• Result: pooled weighted mean difference in marginal reflex distance 1 of 0.37 mm (95% CI 0.13 to 0.60), z = 3.04, p<0.01, graded moderate certainty under GRADE; pooled Leicester peripheral field test 4.72 points (95% CI 3.37 to 6.08), z = 6.84, p<0.01, also moderate certainty.
• Funding: independent of commercial sponsorship on the paper’s own statement: it says the research did not receive any specific grant from commercial, public funding agencies or non-profit sectors.

Limit of this finding: This paper disagrees with itself about the size of the effect. Its abstract prints a lift of 0.65 mm (95% CI 0.44 mm to 0.86 mm), p=0.012, a figure that appears nowhere else in the article; the body of the paper and its Table 1 both give a single pooled weighted mean difference of 0.37 mm (95% CI 0.13 to 0.60). The z statistic of 3.04 and the p-value below 0.01 come from the body text only; Table 1 prints no z statistic and gives the p-value for this outcome as 0.001. We publish the body figure, which is the one the paper’s own analysis supports. The same paper also prints the placebo figure as 0.26 mm with an interval of 0.30 mm to 0.48 mm, in both the abstract and Table 1. That interval does not contain its own point estimate, so it cannot be right, and no reader should take the placebo figure as a measured effect. The abstract also writes the result as though those were two groups’ average changes; a weighted mean difference is a single between-group quantity, not an arm-level one. One further figure printed in this passage should be disregarded. The paper reports a “mortality rate” of 0.9% for the oxymetazoline group against 1.3% for placebo, p = 0.73. A death rate near one in a hundred across four 14-day trials of an eye drop is not credible, and 0.73 is the same p-value the paper reports for serious adverse events, so this row is almost certainly serious adverse events mislabelled as mortality. Nothing we publish rests on it, and no reader should read it as a death rate.

The overall pooled WMD for the MRD1 outcome [0.37 (95% CI: 0.13–0.6), z = 3.04] showed that oxymetazoline significantly increases MRD1 compared to placebo (p<0.01; Figure 5). Using the GRADE criteria, the quality of evidence for this outcome yielded moderate certainty of evidence (Table 2).

In the two pivotal eye-drop trials the gain in upper visual field was statistically significant and lasted through 6 hours. (Source 18)

• Randomized trial, Moderate certainty.
• Size: 140 subjects in Trial 1 (94 active, 46 vehicle) and 164 in Trial 2 (109 active, 55 vehicle)
• Who: adults with acquired blepharoptosis; mean age 64 years in Trial 1 and 63 in Trial 2.
• How long: once daily to each eye for 42 days, with the primary endpoints at hour 6 on Day 1 and hour 2 on Day 14.
• Result: the increase in points seen on the Leicester Peripheral Field Test was statistically significant against vehicle at both time points; the prescribing information's own table of mean points is present but its caption and footnote are the only text extractable here, so the point values should be read from the label itself.
• Funding: not stated; reported in the manufacturer's US prescribing information.

The increases in the number of points seen in the superior visual field in the study eye of the UPNEEQ group compared to the vehicle group were statistically significant at both time points, showing that the improvement in superior visual field was evident at the 2-hour time point and maintained at the 6-hour time point.

The same trial's authors conclude that the loss of response and the rebound are reversible with an intranasal steroid. (Source 7)

• Randomized trial, Moderate certainty.
• Size: 19 healthy subjects.
• Who: healthy volunteers in whom the loss of response had been induced experimentally over 14 days.
• How long: 3 days of added fluticasone after 14 days of oxymetazoline.
• Result: the authors state tachyphylaxis and rebound congestion are reversed by intranasal fluticasone, and in the same breath call for further studies to evaluate whether combination decongestant-and-steroid sprays are an effective strategy. This was experimentally induced tachyphylaxis in 19 healthy subjects over 17 days, not treatment of rhinitis medicamentosa in patients.
• Funding: not stated.

Oxymetazoline-induced tachyphylaxis and rebound congestion are reversed by intranasal fluticasone. Further studies are indicated to evaluate if combination nasal sprays of decongestant and corticosteroid are an effective strategy to obviate tachyphylaxis and rebound in rhinitis.

What the evidence does not support

A contradictory randomised trial did not detect rebound congestion or loss of response after four weeks of oxymetazoline spray three times daily in healthy people. (Source 19)

• Randomized trial, Low certainty.
• Size: 30 normal adult subjects.
• Who: healthy adults with no rhinitis — not people with rhinitis or habitual decongestant users, so the trial cannot rule out rebound in the group that actually overuses the spray.
• How long: 4 weeks of spray plus 2 weeks of follow-up after stopping.
• Result: no significant increase in subjective nasal blockage across the 6 weeks in either group; no significant between-group differences in peak nasal inspiratory flow, airway resistance or volume; a highly significant decongestant response to challenge was still present at each visit. With 30 subjects split between two arms and no stated non-inferiority margin, this is a failure to detect an effect rather than a demonstration that there is none.
• Funding: not stated.

Limit of this finding: This is an absence of a detected effect, not a demonstrated absence. Thirty healthy volunteers were split between two arms and no non-inferiority margin was set, so the trial was not built to prove that rebound does not happen. The authors themselves write only that they identified no significant nasal blockage or impaired decongestant response.

Results demonstrated no significant increases in subjective nasal blockage throughout the 6 weeks study period in either oxymetazoline- or placebo-treated subjects. No significant differences were observed between groups for baseline measurements of nasal peak inspiratory flow, airway resistance or volume at each clinic visit. A highly significant decongestant effect of oxymetazoline was observed at each clinic visit with changes in all 3 measurements for both treatment groups, again with no significant differences between groups.

Where the evidence is mixed

Pooled trials show intranasal oxymetazoline only modestly and inconsistently reduces ear barotrauma, unlike oral pseudoephedrine in the same review. (Source 5)

• Meta-analysis, Low certainty.
• Size: 490 participants across 4 double-blind randomised trials conducted 1992 to 1998; oxymetazoline contributed 2 trials.
• Who: adults exposed to environmental pressure change.
• How long: short-term prophylaxis.
• Result: RR 0.75 (95% CI 0.57 to 0.99; P = 0.05), an upper limit that only just excludes 1, with moderate heterogeneity, I squared = 53%, and inconsistent benefits across individual symptom subgroups.
• Funding: not stated.

Limit of this finding: The review pooled with a fixed-effect inverse-variance model, but heterogeneity for the two oxymetazoline trials was moderate (I squared 53%). A fixed-effect model on heterogeneous data gives a narrower interval than a random-effects model would, so this estimate is less robust than an upper limit of 0.99 suggests.

Oxymetazoline (2 trials) showed a modest overall effect (RR: 0.75; 95% CI: 0.57-0.99; P =0.05) but inconsistent benefits across individual symptom subgroups and moderate heterogeneity ( I ²=53%).

A systematic review of treatments for rhinitis medicamentosa found only 12 small studies and could not produce a treatment algorithm. (Source 20)

• Systematic review, Very low certainty.
• Size: 12 studies including 373 patients.
• Who: patients diagnosed with rhinitis medicamentosa.
• How long: varies by study.
• Result: seven studies used topical intranasal steroids such as budesonide or fluticasone, one used dexamethasone drops, and five used surgery; all reported high efficacy for their own method, but the review says differences in design and assessment prevent systematising the data.
• Funding: not stated.

Twelve studies, including 373 patients, met the search criteria. Out of these, seven studies used topical intranasal steroids like budesonide or fluticasone propionate sprays as a conservative treatment. One study used dexamethasone nasal drops. Five studies involved surgical treatment for patients; three of these studies used radiofrequency ablation to reduce the inferior turbinates, while the other two studies used diode laser and kinetic stimulation, respectively.

The same blepharoptosis meta-analysis prints two different pooled eyelid-lift results, and the figure in its abstract is not the figure in its own results section. (Source 21)

• Meta-analysis, Low certainty.
• Size: 448 participants across 4 randomised controlled trials.
• Who: adults with acquired blepharoptosis.
• How long: outcomes measured to day 14.
• Result: the abstract gives 0.65 mm (95% CI 0.44 mm to 0.86 mm) against a placebo figure of 0.26 mm (95% CI 0.30 mm to 0.48 mm), p=0.012; the body of the same paper and its Table 1 give one pooled weighted mean difference of 0.37 mm (95% CI 0.13 to 0.60), z = 3.04, p<0.01. The abstract also reports no visual-field result at all. Serious adverse events, 1.2% against 1.6%, p = 0.73, agree between abstract and body.
• Funding: not stated.

Limit of this finding: This paper disagrees with itself about the size of the effect. Its abstract prints a lift of 0.65 mm (95% CI 0.44 mm to 0.86 mm), p=0.012, a figure that appears nowhere else in the article; the body of the paper and its Table 1 both give a single pooled weighted mean difference of 0.37 mm (95% CI 0.13 to 0.60), z = 3.04, p<0.01. We publish the body figure, which is the one the paper’s own analysis supports. The same paper also prints the placebo figure as 0.26 mm with an interval of 0.30 mm to 0.48 mm, in both the abstract and Table 1. That interval does not contain its own point estimate, so it cannot be right, and no reader should take the placebo figure as a measured effect. The abstract also writes the result as though those were two groups’ average changes; a weighted mean difference is a single between-group quantity, not an arm-level one.

The meta-analysis yielded a statistically significant higher difference in the WMD in the oxymetazoline group [0.65 mm (95% CI: 0.44 mm to 0.86 mm)] compared to the placebo group [0.26 mm (95% CI: 0.30 mm to 0.48 mm) [p=0.012].

The same review’s authors list the short follow-up of the existing trials, only 14 days, as a limitation, along with the small number of trials and their English-only search. (Source 22)

• Meta-analysis, Moderate certainty.
• Size: 4 randomised controlled trials, 448 participants.
• Who: adults with acquired blepharoptosis.
• How long: no trial followed outcomes beyond 14 days.
• Result: no result beyond day 14 exists in the pooled evidence; the authors also flag the limited number of available trials and their inclusion of English-language studies only as sources of selection bias.
• Funding: independent of commercial sponsorship on the paper’s own statement: it says the research did not receive any specific grant from commercial, public funding agencies or non-profit sectors.

However, our findings have certain limitations, including the short follow-up duration of existing studies, which evaluated outcomes over only 14 days.

The review’s authors state that the long-term efficacy and safety of oxymetazoline are still unevaluated and that larger trials, including direct comparisons against surgery, are needed before its place in care is clear. (Source 23)

• Meta-analysis, Moderate certainty.
• Size: 4 randomised controlled trials, 448 participants.
• Who: adults with acquired blepharoptosis.
• How long: no trial in the review followed outcomes beyond 14 days.
• Result: no pooled result beyond day 14 exists and no included trial compared oxymetazoline with surgical correction; the authors’ own conclusion describes the drug as a short-term treatment for acquired blepharoptosis.
• Funding: independent of commercial sponsorship on the paper’s own statement: it says the research did not receive any specific grant from commercial, public funding agencies or non-profit sectors.

While these findings are promising, further research is warranted to evaluate the long-term efficacy and safety of oxymetazoline. Larger, high-quality randomized controlled trials and comparative studies against surgical interventions are needed to clarify their role in clinical practice.

Where the research disagrees

Whether four weeks of oxymetazoline nasal spray causes rebound congestion and loss of response

• Graf and colleagues, Acta Oto-Laryngologica (1995), randomised double-blind parallel trial in 20 healthy volunteers, measured by rhinostereometry and histamine challenge: After 30 days, rebound swelling and stuffiness appeared on both regimens, with increased histamine sensitivity taken to show nasal hyperreactivity. (Source 4)
• Watanabe and colleagues, Rhinology (2003), randomised double-blind placebo-controlled trial in 30 normal (healthy) adults rather than people with rhinitis or habitual spray users, measured by diary scores, peak nasal inspiratory flow, rhinomanometry and acoustic rhinometry; 30 subjects split between two arms with no stated non-inferiority margin: After four weeks of the spray three times daily, they identified no significant nasal blockage and no impaired decongestant response compared with matched placebo. (Source 19)
• Vaidyanathan and colleagues, American Journal of Respiratory and Critical Care Medicine (2010), randomised double-blind placebo-controlled crossover trial in 19 healthy subjects with a prazosin probe: They measured a fall in peak nasal inspiratory flow of 47.9 L/min after 14 days and a downward shift in the dose-response curve, and reversed both with fluticasone. (Source 10)

How much

• Reference intake: No reference intake exists, because this is a medicine and not a nutrient. Dosing is set by each product's label or by a prescriber, and differs completely between the three products. As a dated position, the Mayne Pharma Rhofade US prescribing information published 15 July 2026 describes the cream as 1%, each gram containing 10 mg of oxymetazoline hydrochloride, equal to 8.8 mg of oxymetazoline free base. (Source 24)
• Upper limit: No tolerable upper intake level exists. The relevant ceiling for the over-the-counter nasal spray is one of duration rather than amount: as a dated position, the CVS oxymetazoline 0.05% nasal spray US Drug Facts label published 6 October 2026 states the product should not be used for more than 3 days, and that frequent or prolonged use may cause nasal congestion to recur or worsen. (Source 8)
• Studied: The rosacea pivotal trials had subjects apply either the cream or vehicle once daily for 29 days. (Source 2)
• Studied: The rebound trial gave healthy volunteers oxymetazoline nasal spray either once daily at night or three times daily for 30 days. (Source 4)
• Studied: The tachyphylaxis crossover trial gave 19 healthy subjects intranasal oxymetazoline 200 mcg three times a day for 14 days, then added fluticasone 200 mcg twice a day for 3 days. (Source 25)
• Studied: The eye-drop pivotal trials administered the drops once daily to each eye for 42 days, randomised about 2:1 against vehicle. (Source 18)

A common belief, and what the research shows

The belief: Decongestant nasal sprays are addictive, so a few days of one will hook you.

What the research shows: The effect is real but it is not addiction in the drug-dependence sense, and the trial evidence disagrees on how fast it appears. A randomised trial found measurable rebound swelling of the nasal lining after 30 days, 0.8 mm with once-daily use and 1.1 mm with three-times-daily use. Another crossover trial measured peak nasal inspiratory flow falling by 47.9 L/min after only 14 days. But a third randomised trial, using the spray three times a day for four weeks, did not detect any significant nasal blockage or loss of decongestant response — though with 30 healthy volunteers split between two arms it was not built to prove an absence. All three trials were in healthy volunteers. What reviewers describe is a self-reinforcing loop rather than a craving: the spray works for shorter and shorter periods, so people use more, which worsens the congestion. In the one trial that induced that loop deliberately, an intranasal steroid undid both the rebound and the loss of response, and the same authors say it still needs testing as a treatment strategy.

Questions and answers

What is it?

Oxymetazoline hydrochloride is a manufactured imidazoline compound that narrows small blood vessels where it is applied. Its chemical name is 3-[(4,5-Dihydro1H-imidazol-2-yl)methyl]-6-(1,1-dimethylethyl)- 2,4-dimethyl-phenol hydrochloride, with a molecular weight of 296.8. It is sold as an over-the-counter nasal spray, as a 1% cream for rosacea redness, and as 0.1% eye drops for a drooping eyelid. (Source 24)

What does it do in the body?

It switches on alpha adrenoceptors, which constricts blood vessels where the product is put. In the eye it acts on what the prescribing information calls a subset of adrenoreceptors in Mueller’s muscle, the small muscle that helps hold the upper eyelid up; the lift peaks about 2 hours after a dose and is still present at 8 hours. The rosacea cream’s own label describes the same drug simply as a vasoconstrictor, which is what reduces visible redness in the face and swelling in the lining of the nose. (Source 26)

Is it good or bad for you?

Both, and which one depends entirely on how long it is used and where. Used briefly it does what it is meant to do. Used on the nose for weeks it can produce the opposite of the intended effect: a randomised trial found rebound swelling and stuffiness after 30 days on both once-daily and three-times-daily regimens, with increased histamine sensitivity indicating nasal hyperreactivity. For the cream and the eye drops, both of which have fixed longer courses in their trials, the harms were mostly local and mild. (Source 4)

How do you get more of it?

The nasal spray is bought over the counter; the cream and the eye drops are, among products currently marketed in the United States, prescription-only. Getting more of the nasal spray is the thing the evidence warns against rather than recommends: the label states the product should not be used for more than 3 days, and that frequent or prolonged use may cause congestion to recur or worsen. (Source 8)

If it is harmful, what reduces it?

Stopping the spray is the first step, and intranasal steroids are what the treatment literature mostly uses. A 2025 systematic review of rhinitis medicamentosa treatments found twelve studies in 373 patients: seven used topical intranasal steroids such as budesonide or fluticasone propionate sprays, one used dexamethasone nasal drops, and five used surgery — three of those radiofrequency ablation to reduce the inferior turbinates, the other two a diode laser and kinetic stimulation. Separate trial evidence on adding a steroid spray after stopping is set out in the stopping section of this entry. (Source 20)

Why might someone be low in it or missing it?

The question does not apply in the nutrient sense: the body does not make or need oxymetazoline, so no one can be deficient in it. What can go missing is its effect. A pharmacology review describes exactly that: in rhinitis medicamentosa the congestion responds less effectively and for progressively shorter periods to the same spray, which is a loss of response rather than a loss of the substance. (Source 9)

Which whole foods contain it or feed it?

No food contains oxymetazoline. It is a manufactured imidazoline drug substance with a defined chemical formula, supplied in a spray, a cream and eye drops; there is no dietary source and nothing in food feeds it. (Source 24)

We searched: Searched Europe PMC for oxymetazoline with food, dietary, natural-product and plant terms, and reviewed the US prescribing information's DESCRIPTION section for any natural source. The literature describes it only as a synthetic compound; the prescribing information gives its molecular formula as C16H25ClN2O.

What happens if you do not have it?

Nothing happens to the body from not having it, because it treats symptoms rather than filling a gap. For the conditions it is used in, the untreated state is the condition itself: a blocked nose, visible facial redness, or a drooping lid. A pharmacology review puts nasal congestion in proportion, calling it a minor ailment that is nonetheless a bothersome symptom for many people with rhinitis. (Source 9)

How can you test for it?

There is no test for oxymetazoline itself, and no deficiency to test for. What the trials measure instead is the effect: a photographic measure called marginal reflex distance 1, the distance from the central pupillary light reflex to the central margin of the upper lid. That is a trial endpoint, not a test a member would have done. (Source 26)

We searched: Searched Europe PMC for oxymetazoline with assay, urine, blood level, monitoring and diagnostic terms, and read the pharmacokinetics sections of both prescription US labels. No clinical or self-administered test for oxymetazoline status was found; only trial endpoints and research pharmacokinetic assays.

References

1. Mayne Pharma (US prescribing information via DailyMed, National Library of Medicine). RHOFADE (oxymetazoline hydrochloride) cream, 1% - US prescribing information, 12.1 Mechanism of Action and 12.2 Pharmacodynamics sections. 2026. Read the source
2. Mayne Pharma (US prescribing information via DailyMed, National Library of Medicine). RHOFADE (OXYMETAZOLINE HYDROCHLORIDE) CREAM [MAYNE PHARMA] - US prescribing information, 14 CLINICAL STUDIES section, including Table 2 in full (this section names the product only as “Oxymetazoline Hydrochloride Cream”). 2026. Read the source
3. Clinical Ophthalmology. Efficacy of Oxymetazoline 0.1% in Acquired Blepharoptosis: A Systematic Review and Meta-Analysis of Randomized Clinical Trials. (full text: Results, Intervention vs Placebo Outcomes). 2025. PMID 40881581, DOI 10.2147/OPTH.S529633. Read the source
4. Acta oto-laryngologica. Four-week use of oxymetazoline nasal spray (Nezeril) once daily at night induces rebound swelling and nasal hyperreactivity. (abstract). 1995. PMID 7762389, DOI 10.3109/00016489509133350. Read the source
5. Otology & neurotology : official publication of the American Otological Society, American Neurotology Society [and] European Academy of Otology and Neurotology. Efficacy of Pseudoephedrine and Oxymetazoline in Preventing Otic Barotrauma: A Systematic Review and Meta-Analysis. (abstract: Results). 2026. PMID 41931637, DOI 10.1097/mao.0000000000004905. Read the source
6. Mayne Pharma (US prescribing information via DailyMed, National Library of Medicine). RHOFADE (oxymetazoline hydrochloride) cream, 1% - US prescribing information, complete 7 DRUG INTERACTIONS section. 2026. Read the source
7. American journal of respiratory and critical care medicine. Fluticasone reverses oxymetazoline-induced tachyphylaxis of response and rebound congestion. (abstract: Conclusions). 2010. PMID 20203244, DOI 10.1164/rccm.200911-1701oc. Read the source
8. CVS Pharmacy (US prescribing information via DailyMed, National Library of Medicine). NASAL (oxymetazoline hydrochloride) spray - US Drug Facts label, complete Warnings section. 2026. Read the source
9. Frontiers in pharmacology. Part II - imidazolines and rhinitis medicamentosa: how can we tackle the rebound dilemma? (abstract). 2025. PMID 41394157, DOI 10.3389/fphar.2025.1655254. Read the source
10. American journal of respiratory and critical care medicine. Fluticasone reverses oxymetazoline-induced tachyphylaxis of response and rebound congestion. (abstract: Measurements and main results). 2010. PMID 20203244, DOI 10.1164/rccm.200911-1701oc. Read the source
11. European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery. Variants of rhinitis medicamentosa treatment: a systematic review. (abstract: Conclusion). 2025. PMID 40175814, DOI 10.1007/s00405-025-09344-6. Read the source
12. Journal of cosmetic dermatology. Efficacy and safety of oxymetazoline for the treatment of rosacea: A meta-analysis. (abstract: Results). 2023. PMID 37128814, DOI 10.1111/jocd.15747. Read the source
13. Mayne Pharma (US prescribing information via DailyMed, National Library of Medicine). RHOFADE (oxymetazoline hydrochloride) cream, 1% - US prescribing information, 6.1 Clinical Studies Experience section, including Table 1 in full. 2026. Read the source
14. Ophthalmic plastic and reconstructive surgery. Safety Analysis of Oxymetazoline Eye Drops for Blepharoptosis Using the FDA Adverse Event Reporting System. (abstract: Results). 2026. PMID 41949436, DOI 10.1097/iop.0000000000003226. Read the source
15. RVL Pharmaceuticals, Inc. (US prescribing information via DailyMed, National Library of Medicine). UPNEEQ (oxymetazoline hydrochloride ophthalmic solution), 0.1% - US prescribing information, 6.1 Clinical Trials Experience section. 2026. Read the source
16. RVL Pharmaceuticals, Inc. (US prescribing information via DailyMed, National Library of Medicine). UPNEEQ (oxymetazoline hydrochloride ophthalmic solution), 0.1% - US prescribing information, 10 OVERDOSAGE section. 2026. Read the source
17. Journal of cosmetic dermatology. Efficacy and Safety of Oxymetazoline 1% Cream for the Treatment of Mild to Moderate Facial Rosacea. (abstract: Results). 2025. PMID 40165547, DOI 10.1111/jocd.16782. Read the source
18. RVL Pharmaceuticals, Inc. (US prescribing information via DailyMed, National Library of Medicine). UPNEEQ (oxymetazoline hydrochloride ophthalmic solution), 0.1% - US prescribing information, 14 CLINICAL STUDIES section. 2026. Read the source
19. Rhinology. Oxymetazoline nasal spray three times daily for four weeks in normal subjects is not associated with rebound congestion or tachyphylaxis. (abstract). 2003. PMID 14579657. Read the source
20. European archives of oto-rhino-laryngology : official journal of the European Federation of Oto-Rhino-Laryngological Societies (EUFOS) : affiliated with the German Society for Oto-Rhino-Laryngology - Head and Neck Surgery. Variants of rhinitis medicamentosa treatment: a systematic review. (abstract: Results). 2025. PMID 40175814, DOI 10.1007/s00405-025-09344-6. Read the source
21. Clinical ophthalmology (Auckland, N.Z.). Efficacy of Oxymetazoline 0.1% in Acquired Blepharoptosis: A Systematic Review and Meta-Analysis of Randomized Clinical Trials. (abstract: Results). 2025. PMID 40881581, DOI 10.2147/opth.s529633. Read the source
22. Clinical Ophthalmology. Efficacy of Oxymetazoline 0.1% in Acquired Blepharoptosis: A Systematic Review and Meta-Analysis of Randomized Clinical Trials. (full text: Discussion, final paragraph, on limitations). 2025. PMID 40881581, DOI 10.2147/OPTH.S529633. Read the source
23. Clinical Ophthalmology. Efficacy of Oxymetazoline 0.1% in Acquired Blepharoptosis: A Systematic Review and Meta-Analysis of Randomized Clinical Trials. (full text: Conclusion section in full, its two paragraphs separated by a line break). 2025. PMID 40881581, DOI 10.2147/OPTH.S529633. Read the source
24. Mayne Pharma (US prescribing information via DailyMed, National Library of Medicine). RHOFADE (oxymetazoline hydrochloride) cream, 1% - US prescribing information, 11 DESCRIPTION section. 2026. Read the source
25. American journal of respiratory and critical care medicine. Fluticasone reverses oxymetazoline-induced tachyphylaxis of response and rebound congestion. (abstract: Methods). 2010. PMID 20203244, DOI 10.1164/rccm.200911-1701oc. Read the source
26. RVL Pharmaceuticals, Inc. (US prescribing information via DailyMed, National Library of Medicine). UPNEEQ (oxymetazoline hydrochloride ophthalmic solution), 0.1% - US prescribing information, 12.1 Mechanism of Action and 12.2 Pharmacodynamics sections. 2026. Read the source