Can Ghee Be Made Sterile for the Eye? A Biomedical Review of Netra Tarpana

Blog

Can Ghee Be Made Sterile for the Eye? A Biomedical Review of Netra Tarpana

November 22, 2026

Ghee can be made sterile for use on the eye. Pharmaceutical standards already cover sterile oily eye preparations, and laboratory work shows that cow ghee can change how a drug moves into ocular tissue. The human evidence is smaller: a few dozen patients, in studies built to measure benefit more than harm.

This review covers how ghee behaves on the ocular surface, what a sterile preparation requires, and what the clinical studies support.

What the eye meets during netra tarpana

In netra tarpana, a ring around the orbit holds liquefied ghee against the lids and, when the eye opens, the ocular surface. The tissues exposed are the corneal and conjunctival epithelium, the tear film and the lid margins. In one published dry-eye study, each session used a pool of about 10 to 15 mL with intermittent blinking over 30 minutes (1).

The word ghee also covers more than one material. Plain cow ghee is clarified milk fat. Goghrita manda is the liquid fraction that separates at the top of it. A medicated ghrita carries plant constituents taken up during processing. These differ in composition, viscosity and impurities, so a result for one does not transfer to another.

Two different purposes follow from this. The fat may act physically, by changing the oil-tear boundary or lubricating the lid as it moves across the eye. Or it may act as a vehicle, carrying a drug or plant constituent to a defined tissue. A preparation could do both.

The distinction decides what has to be proved. If the aim is less friction, an effect confined to the surface is the goal. If the aim is delivery, the investigator has to name the constituent, show how much is released, and show that enough reaches its target.

Ghee as a pharmaceutical vehicle

Clarification removes most of the water and nonfat solids from butter. What remains is almost entirely milk lipid, chiefly triglycerides, and the Codex food standard sets a minimum of 99.6% milk fat (2).

Two properties interest a formulator. Milk fat is a mixture of triglycerides that softens and melts over a range of temperatures, so its flow and the film it leaves change as it cools on the eye. It can also dissolve compounds that dissolve poorly in water, which a watery drop struggles to carry.

Dissolving a compound is only half the job. A molecule that is very comfortable in fat may have little tendency to leave it. Formulators therefore measure release into a tear-like medium as well as total content (3, 4).

Medicated ghrita adds an analytical problem. A plant name identifies a source. The finished product contains only the compounds that survived processing and moved into the fat. Water-soluble constituents may be poorly represented, and others may travel with fine particles.

An ocular dose therefore cannot be inferred from the weight of plant used at the start. A chemical fingerprint can show whether batches resemble one another. Measuring chosen marker compounds before and after processing shows what was lost along the way (5, 6).

Physical behavior needs the same attention. A fat that looks clear in its container may partly crystallize as it cools. Its flow when poured, how easily a blink redistributes it, and the thickness of the film left behind can all change within one session.

How ghee behaves on the ocular surface

An ordinary eye drop is spread by blinking and cleared by tear turnover and drainage. A retained pool changes that. For as long as the ring is filled, fresh vehicle stays in contact with the thin tear layer over the cornea.

The natural tear film has its own lipid layer, supplied by the meibomian glands and rich in wax esters and cholesteryl esters (7). Milk fat is mostly triglyceride, so a ghee film cannot be assumed to copy healthy meibum. The testable hypothesis is narrower: under defined conditions, a milk-fat film might reduce friction or slow evaporation for long enough to improve a measurable outcome.

There is a model for testing it. In a small randomized crossover study, low-concentration castor oil drops improved tear evaporation and tear break-up time against placebo in people with obstructive meibomian gland dysfunction (8). The same design could be applied to ghee, with measurements taken after the pool is removed and clear vision has returned.

A fat film also blurs vision. How long that lasts after a session is a safety outcome in its own right and needs to be measured.

An irregular film degrades the optical surface and interferes with reading and other close work. A patient who feels more comfortable but cannot read clearly for an hour has had two outcomes, and both count. A study that records only a symptom score at the next visit would miss the trade-off.

Timing matters for tear-film measurements too. Readings taken while a visible layer of ghee remains describe the ghee. Readings taken later, after normal blinking has resumed, describe the eye (9).

From the pool to the tissue

For a dissolved compound to reach the corneal stroma, it has to leave the fat at the oil-tear boundary, cross the outer epithelium with its cell membranes and tight junctions, and then enter the water-rich stroma. Size, charge and affinity for fat or water govern each step (3).

One caution applies to every permeation result. A formulation that injures the epithelium lets more material through, which can look like good delivery. Barrier integrity has to be measured alongside the drug. In patients, that means fluorescein staining and a check for new epithelial defects.

It also matters where a compound is measured. Three findings are often treated as one:

  • A marker found in tears shortly after treatment shows that it reached the tear sample.
  • A marker found in corneal tissue, once surface residue is accounted for, shows uptake into that tissue.
  • A marker found on the far side of an intact excised cornea shows transport across that barrier in the laboratory.

None of the three, alone, establishes a therapeutic concentration at a target in a living human eye. Reports are far easier to interpret when the tissue, the sampling time, the method of removing surface residue and the amount recovered are given together.

The strongest delivery case for tarpana is the ocular surface. The retina and optic nerve are far harder targets, with clearance, tissue binding and vascular barriers in the way (4). No pharmacokinetic evidence shows that a ghee bath delivers a herbal constituent to the back of the human eye.

What laboratory studies of ghee show

Three experimental studies used a dairy fat in a defined eye formulation.

StudyFormulationModelReported
Gupta et al., 2019 (10)Fluocinolone acetonide microemulsion fortified with cow gheeEx vivo permeation, irritation test, rat pharmacokineticsEnhanced permeation, non-irritant in the test used, drug retained in the posterior eye of the rat
Raval et al., 2018 (11)Triamcinolone acetonide microemulsion with butter oil and chitosanPermeation and ocular distributionDairy lipid studied as a permeation enhancer
Dumore et al., 2016 (12)Ketoconazole ointment in a cow-ghee base, compared with petrolatumExcised goat corneaTransport across the cornea compared between bases

Together they show that a dairy fat can affect the performance of a defined formulation. They do not describe traditional ghrita in a retention ring. The microemulsions contained water and surfactants, which alter the oil-water boundary, and each carried one identified drug that could be tracked.

For a botanical preparation, the next useful experiment is a direct comparison of the finished ghrita with its matched plain-ghee vehicle. It would run in three stages:

  1. A release test, asking whether an identified constituent leaves the fat into a tear-compatible medium.
  2. A tissue test, comparing deposition in, and passage across, intact epithelium over the real exposure time.
  3. A retention test, measuring whether the tissue still holds the compound after the pool is removed.

A result showing strong surface retention and negligible deeper penetration would still be valuable. It would point research toward surface conditions and close off unsupported claims about the interior of the eye.

Why ghee can be manufactured sterile

Sterile oily eye products already exist. The World Health Organization's International Pharmacopoeia recognizes oily ophthalmic drops and describes sterilization methods for fats and oils (13, 14). US regulation addresses the sterility of ophthalmic preparations and their dispensers (15).

Marketed eye products include a sterile castor-oil emulsion, a sterile mineral oil and petrolatum ointment, and a sterile water-free liquid, perfluorohexyloctane (16, 17, 18). None uses milk fat. They do establish that an oily or water-free product can be made and sold sterile.

Ghee's low water content limits microbial growth in storage, but growth and survival are different things. Viable organisms, including spores, can persist, and warming ghee until it flows does not sterilize it. Two pharmaceutical routes can:

  • Terminal sterilization. The final container is filled and sealed, then treated with a validated cycle. Dry heat is an established option for heat-stable, water-free materials. For ghee, the cycle must be shown to reach every part of the filled unit without oxidising the fat or degrading its constituents.
  • Aseptic processing. Sterile ingredients and components are combined and filled under monitored conditions (19). Sterilizing filtration may suit a clear, flowable lipid if the filter is shown to work with it.

Either route has to be validated for the exact formulation, fill volume and container. Straining ghee, or pouring it into a sterile vessel, does not establish sterility of the filled product.

Heat brings a trade-off. A cycle may kill microbes while speeding oxidation, shifting the balance of botanical markers or changing how the fat crystallizes on cooling. The material that passed the sterilization study could then differ from the one first tested for comfort or drug release. Chemistry, release and ocular compatibility should be measured on units that have been through the complete process (20).

What would persuade a reviewer that the process works?

RouteEvidence expected
Terminal sterilizationValidation on the filled package, proof of performance under the least favorable conditions, and records placing each lot within the validated process
Aseptic processingControls on incoming materials, qualified staff and equipment, environmental monitoring, and process simulations that include interventions during filling
BothA closure that holds the sterile barrier through the claimed shelf life

The finished-product sterility test adds evidence about the units sampled. Its meaning depends on everything that happened before sampling (19, 21).

What a sterile vial still has to prove

Sterility is one specification among several.

  • A valid sterility test. Fat disperses poorly and can interfere with culture methods, so the laboratory must show its test can detect organisms in the product (22).
  • Endotoxin. Fragments of bacterial cell walls can remain after the bacteria are killed. FDA does not generally require endotoxin testing for routine topical eye drugs, though a limit may apply for use on an abraded eye (23).
  • Particles. Fine debris, or crystals that form on cooling, may be invisible to the unaided eye and felt with every blink (24).
  • Oxidation. Milk fat oxidises in storage, driven by light, oxygen and temperature (25). A food freshness limit is not an ocular tolerability limit.
  • Packaging. A single-patient, single-use unit limits contamination, provided the contents stay protected during opening and transfer.
  • Milk protein. Clarification removes most milk protein but not necessarily all. Allergy guidance lists ghee as a product of concern in cow's milk protein allergy (26).

Some familiar tests for watery eye drops do not translate. pH and osmolality describe a water-based solution, and a value assigned to neat fat is hard to interpret. For an oily product, viscosity, melting and crystallization behavior, spreading and particle load tell a clinician more about what the eye will meet.

Oxidation needs more than one measure. Peroxide value tracks the first products of oxidation, but those break down in turn, so a low value alone cannot describe the full history of a fat. Measures of secondary oxidation products and assays of botanical markers fill the gap.

The container is part of the product. When a lipid sits against plastic, rubber or adhesive, chemicals can migrate into it, and that has to be studied for the actual materials, temperature and contact time. If a sealed unit is meant to be warmed before use, the warming belongs in those studies too.

On allergy, two conditions are often confused. Lactose intolerance is a digestive problem with milk sugar. Milk protein allergy is an immune reaction, and it is the one that matters for a ghee preparation near the eye.

What the human studies show

StudyParticipantsInterventionWhat it tells us
Timmapur and Fiaz (1)30 with dry eye, 15 per groupTarpana with a medicated ghee or with goghrita mandaBoth groups improved on symptoms and tear tests. Allocation alternated, and there was no non-ghee comparison
Dastapure et al. (27)10 with dry eyeTarpana with Jeevantyadi GhritaAll ten completed treatment
Poonam et al. (28)54 with myopia, in two groupsGhee retention40 completed. The reasons for the 14 withdrawals matter to any safety reading

These studies show that selected patients have undergone repeated ghee retention and completed their courses. They were built to measure benefit, and they give little systematic account of blur, burning, new corneal staining or delayed problems.

Small numbers also limit reassurance. If 30 people were followed completely and no adverse event occurred, the true rate could still be as high as about 10%, by the statistical rule of three. It would take about 300 such participants to bring that ceiling down to 1% (29).

Counting also needs care. Two eyes in one person, and repeated sessions in one eye, are not independent observations. Thirty people treated in both eyes over several sessions are still thirty people. Reporting them as hundreds of treatments makes the uncertainty look smaller than it is.

The harm report most often cited is a 2017 case of severe disease in both eyes, after a person with red eyes self-treated with ghee, goat's milk and other drops (30). The existing illness and the mix of agents prevent attributing it to ghee, and it says nothing about a sterile, standardized preparation. It does support one firm rule: household ghee should not be put in the eye.

The procedure adds its own questions

A vial is sterile until it is opened. During tarpana its contents touch a warming vessel, the ring, skin, lashes and tears. Traditional dough can shed particles and carry microbes even when the ghee was made sterile. A purpose-made ring would need its own evidence on cleanliness, seal and behavior in warm fat.

Temperature should be defined where the fat meets the eye, including the peak just after pouring. Comfort on the skin does not establish what a cornea tolerates over a long exposure.

The eye itself changes the risk. An intact epithelium is a barrier. A persistent epithelial defect, suspected infectious keratitis, recent corneal surgery or reduced corneal sensation changes the consequences of the same exposure, and results from early dry eye do not carry over.

In an eye with reduced sensation, a patient can report little discomfort despite real tissue injury (31). New or worsening pain, light sensitivity, discharge or loss of vision after any ocular exposure calls for urgent ophthalmic assessment (32). The clinical priority is to tell a passing optical film apart from epithelial injury or infection.

The living eye and the skin around it are not sterile either (33). Aseptic handling does not aim to make them so. Its purpose is to avoid adding contamination that was preventable.

What a proper safety study would look like

A credible study starts with a fully identified product, in its intended package, with a clearly described method of retention. Findings can then be tied to a composition, a lot, a device and an exposure.

In the laboratory. Epithelial viability and barrier recovery can be examined across the intended contact time, including after the fat is removed. Reconstructed human cornea-like epithelium models offer a way to screen for irritation (34). They help reject unsuitable formulations. They do not qualify a medicine for repeated use on the eye.

In people. Baseline slit-lamp findings and visual acuity make later changes interpretable. The outcomes to follow are surface staining, epithelial defects, inflammation, allergic signs, discomfort and the time until vision is clear. Follow-up has to extend beyond the session to catch delayed problems.

In the reporting. People, eyes, sessions and completed examinations should be counted separately, with reasons for every withdrawal and event rates per participant. A masked examiner reduces bias when grading staining or inflammation.

The comparison depends on the question being asked.

ComparisonQuestion it answers
Medicated ghrita against its plain-ghee vehicleWhat do the botanical ingredients add?
Ghee retention against an established sterile lubricantDoes the whole procedure beat ordinary lubrication?
The same preparation as drops and as a retained poolDoes the pool add enough benefit to justify the larger exposure?

Data of this kind could support a tolerability statement for the population and exposure actually studied. That is the standard a safety claim for netra tarpana should be held to.

What the evidence supports

  • Ghee can be made into a sterile ophthalmic preparation, using methods that pharmacopoeias already describe for oils.
  • A dairy fat can affect drug delivery in a defined laboratory formulation.
  • Small human studies show that examined, selected patients complete courses of ghee retention and report improvement in dry-eye measures.
  • Safety for a specific product rests on that product: its sterility validation, its handling, the ring, the exposure and the patients it is used in.
  • Household ghee is not an eye preparation.

At Netra Eye Institute, netra tarpana is one component of Netra Restoration Therapy and not a stand-alone treatment. The program also draws on Traditional Chinese Medicine for eye disorders.

For a patient, this turns into three questions for any clinic: what preparation is used, how it was made sterile, and who examines the eye first. Our netra tarpana guide covers the procedure itself.

References

  1. Timmapur GM, Fiaz S. Efficacy of Triphala Ghrita and Goghrita Manda Tarpana in dry eye syndrome. AYU. 2020;41(1):52-57.
  2. Codex Alimentarius Commission. Standard for Milkfat Products. CXS 280-1973, amended 2022.
  3. Partitioning and spatial distribution of drugs in ocular surface tissues. Pharmaceutics. 2021.
  4. del Amo EM, et al. Pharmacokinetic aspects of retinal drug delivery. Progress in Retinal and Eye Research. 2017.
  5. US Food and Drug Administration. Botanical drug development. Guidance for industry, 2016.
  6. In vitro release testing method development for ophthalmic ointments. International Journal of Pharmaceutics. 2017.
  7. Willcox MDP, et al. TFOS DEWS II Tear Film Report. The Ocular Surface. 2017.
  8. Goto E, et al. Low-concentration homogenized castor oil eye drops for noninflamed obstructive meibomian gland dysfunction. Ophthalmology. 2002.
  9. TFOS DEWS III. Diagnostic methodology. American Journal of Ophthalmology. 2025.
  10. Gupta A, et al. Cow ghee fortified ocular topical microemulsion: in vitro, ex vivo, and in vivo evaluation. Journal of Microencapsulation. 2019.
  11. Raval N, Khunt D, Misra M. Microemulsion-based delivery of triamcinolone acetonide to posterior segment of eye using chitosan and butter oil as permeation enhancer. Journal of Microencapsulation. 2018.
  12. Dumore MN, et al. Formulation of ketoconazole ophthalmic ointment using cow ghee as a base and penetration enhancer. International Research Journal of Pharmacy. 2016.
  13. World Health Organization. Ophthalmic preparations. The International Pharmacopoeia, 12th ed. 2025.
  14. World Health Organization. Methods of sterilization. The International Pharmacopoeia, chapter 5.8.
  15. US Code of Federal Regulations. Title 21, section 200.50. Ophthalmic preparations and dispensers.
  16. Restasis (cyclosporine ophthalmic emulsion 0.05%). US prescribing information. DailyMed.
  17. Retaine PM (mineral oil and white petrolatum ophthalmic ointment). US Drug Facts label. DailyMed.
  18. Miebo (perfluorohexyloctane ophthalmic solution). US prescribing information. DailyMed.
  19. US Food and Drug Administration. Sterile drug products produced by aseptic processing: current good manufacturing practice. Guidance for industry, 2004.
  20. European Medicines Agency. Guideline on the sterilisation of the medicinal product, active substance, excipient and primary container. 2019.
  21. US Food and Drug Administration. Questions and answers on current good manufacturing practice regulations: production and process controls.
  22. World Health Organization. Test for sterility. The International Pharmacopoeia, chapter 3.2.
  23. US Food and Drug Administration. Questions and answers on quality-related controlled correspondence: bacterial endotoxin limits for topical ophthalmic drug products.
  24. US Food and Drug Administration. Quality considerations for topical ophthalmic drug products. Revised draft guidance, December 2023.
  25. Gosewade S, et al. A study on the physico-chemical changes occurring in ghee (butter oil) during storage. Indian Journal of Dairy Science.
  26. Australasian Society of Clinical Immunology and Allergy. Dietary guide for cow's milk protein allergy.
  27. Dastapure M, et al. A clinical study on efficacy of Jeevantyaadi Ghrita in Sushkakshipaaka (dry eye). International Journal of Ayurveda and Pharma Research. 2(6).
  28. Poonam, et al. A clinical study on the role of Akshi Tarpana with Jeevantyadi Ghrita in Timira (myopia). AYU. 2011.
  29. Hanley JA, Lippman-Hand A. If nothing goes wrong, is everything all right? Interpreting zero numerators. JAMA. 1983.
  30. Self-medication complicating pseudo membranous conjunctivitis. BMJ Case Reports. 2017.
  31. Expert consensus on the identification, diagnosis, and treatment of neurotrophic keratopathy. BMC Ophthalmology. 2021.
  32. Rhee MK, et al. Bacterial Keratitis Preferred Practice Pattern. American Academy of Ophthalmology. 2024.
  33. Current evidence on the ocular surface microbiota and related diseases. Microorganisms. 2020.
  34. Organisation for Economic Co-operation and Development. Test No. 492: Reconstructed human cornea-like epithelium test method.

Medical Disclaimer: This article is for general education and is not medical advice, diagnosis or treatment. Netra tarpana should only be performed after an eye examination, by a trained practitioner. Do not stop or change any prescribed eye treatment because of this article. New eye pain, redness, light sensitivity, discharge or loss of vision needs urgent eye care.

More From the Blog