Toenail fungus, called onychomycosis, grows in the nail bed underneath the nail plate, and an antifungal cream cannot cross that dense keratin to reach it, which is why those creams are labeled against use on nails. Oral terbinafine, 250 mg daily for 12 weeks for toenails and 6 weeks for fingernails, travels through the bloodstream into the nail bed and clears the fungus in roughly 70% to 80% of dermatophyte infections. Prescription nail solutions cross the plate better than a cream but produce complete cure in about 7% to 18% after 48 weeks. Because only about half of abnormal nails turn out to be fungal, Fishtown Medicine confirms the diagnosis on a nail specimen before starting months of therapy.
TL;DR: Fungus in a toenail lives in the nail bed, under the hard part of the nail. A cream sits on top of that hard part and cannot get through it, so months of rubbing cream on the nail usually change nothing. Pills work because the medicine travels in your blood and reaches the fungus from underneath. Before you start pills, have the nail tested, because about half of thick, discolored nails turn out to be something other than fungus. If it is fungus, the usual treatment is terbinafine, 1 pill a day for 12 weeks for a toenail. You get a blood test for your liver first, and you go over every medicine you take, because terbinafine changes how some of them work. When the pills are done the nail will still look bad, and that is expected. A big toenail takes 12 to 18 months to grow all the way out, so you watch the new nail coming in at the base. To keep the fungus from coming back, treat the athlete's foot on your skin at the same time, dry between your toes, keep the nails short, and disinfect or replace the shoes you wore while it was active.
Why does the cream keep failing?
The cream keeps failing because the fungus is not on the nail. It is underneath it, growing in the nail bed and in the underside of the plate, and the hard nail you are rubbing the cream onto is the roof over its head.
That roof is built to keep things out. The nail plate is compacted keratin, dense and layered, and it behaves less like skin and more like a shingle. Drug penetration work on nail plates taken from people with onychomycosis shows how poorly medication crosses that barrier, and it gets worse rather than better once the nail is infected and its structure is disturbed. A molecule has to be small enough and bind keratin loosely enough to travel through, and most cream formulations fail one test or the other before they ever get near the fungus. Whatever does not travel through simply wipes off on the sock.
This is not a subtle point that manufacturers have overlooked. Over-the-counter antifungal creams for athlete's foot carry a plain instruction in their labeling: do not use on nails or scalp. So someone who has spent 8 months applying an athlete's foot cream to a thick yellow toenail has been doing careful, disciplined work against the wrong target, and the lack of progress says nothing about their effort.
What reaches the fungus, and how well?
An oral antifungal reaches it, and it gets there the only way anything can, which is from below. Terbinafine taken by mouth is absorbed, carried in the blood, and deposited into the nail matrix and nail bed, so the drug arrives at the living tissue where the fungus is feeding rather than sitting on the dead plate above it. That is the whole reason the pill works where the cream does not.
Dermatophytes cause about 90% of toenail infections, most often Trichophyton rubrum, and terbinafine is most active against that group, which is why it is first-line. The FDA labeling for terbinafine tablets sets the adult course at 250 mg once daily for 12 weeks for toenails and 6 weeks for fingernails; toenails take longer because they grow more slowly and carry a bigger infected surface.
The published numbers are good without being perfect. Mycologic cure, meaning the fungus is gone on testing, runs roughly 70% to 80% across trials and meta-analyses of the 12-week course. Complete cure, which requires both a negative test and a nail that looks normal, is lower, commonly reported around 38%, and the gap between those 2 figures is mostly about how long a toenail takes to replace itself rather than about the drug failing. The American Family Physician rapid evidence review puts it plainly: oral terbinafine is preferred over topical therapy for effectiveness and for the shorter course.
Is it definitely fungus?
Roughly half of it is. Onychomycosis accounts for about 50% of nail disease, which means the other half of thick, yellowed, crumbling nails turn out to be something other than fungus, and no length of antifungal therapy will touch those.
Nail psoriasis is the main impostor, and it can look nearly identical from across the room. Pitting in the surface, an oil-drop stain of orange-brown under the plate, separation of the nail from the bed, and thickened debris underneath all show up in both conditions, so the tell is usually elsewhere: the scalp, the elbows, the knees, the gluteal cleft, a family history. Repeated trauma is the other common answer, and it hides in plain sight in runners and in anyone whose shoes are half a size too short, where the second toe or the great toe takes the impact thousands of times a week until the nail thickens and yellows on its own. Lichen planus of the nail, chronic paronychia, and a handful of rarer conditions round out the list, and a single dark streak or a nail that bleeds and will not heal needs an in-person look for reasons that have nothing to do with fungus.
Confirming it takes a nail specimen, and it is a small ask before a 3-month prescription. A potassium hydroxide preparation read under the microscope is fast, cheap, and specific enough that a positive result is sufficient to begin treatment. A nail clipping sent for PAS staining is the most sensitive of the routine options and is often the answer when the KOH is negative but the picture still fits. Fungal culture is the slowest and misses a fair number of infections, though it is the one that names the organism, which matters when the culprit turns out to be a mold or a yeast rather than a dermatophyte. PCR is faster and more sensitive than culture and is increasingly how these get answered. The sampling technique carries more weight than most people expect: the useful material is the debris from under the nail near where healthy nail begins, rather than the crumbly free edge everyone reaches to clip.
What about the prescription nail solutions?
Prescription nail solutions are a different animal from an athlete's foot cream, and they deserve an honest accounting. Ciclopirox 8% lacquer, efinaconazole 10% solution, and tavaborole 5% solution were each engineered to cross the nail plate, and each one does cross it to some degree. The trial results tell you how much.
Applied daily for 48 weeks, efinaconazole 10% produced complete cure in 15% to 18% of patients against 3.3% to 5.5% for the vehicle, tavaborole 5% in 6.5% to 9.1% against 0.5% to 1.5%, and ciclopirox 8% lacquer in about 7% against 1%. The 2020 Cochrane review of topical and device-based treatments found the same shape of result: better than nothing, and low in absolute terms.
Those numbers still buy something. A topical makes sense when only a small share of the nail is involved and the matrix at the base is spared, when the pattern is the white superficial kind that sits on the surface of the plate, when a liver condition or a medication list rules out the oral drug, and when someone would rather accept a lower chance of cure than take a systemic medication for 3 months. Topicals also earn a steady place after an oral course as maintenance, which is where some of their best data lives.
Laser treatment is the other question people bring in, and the honest answer is that the FDA clearances for these devices are for a temporary improvement in the appearance of the nail rather than for curing the infection. Reviews of the improvement rates and the Cochrane analysis both fall well short of what oral therapy delivers, with low-quality evidence showing little or no difference in mycologic cure against sham.
How risky is oral terbinafine for the liver?
Serious liver injury from terbinafine is rare, and its reputation runs ahead of its record. The National Institutes of Health LiverTox database puts clinically apparent liver injury at roughly 1 in 50,000 to 120,000 prescriptions, and other estimates put it at 0.5 to 3 cases per 100,000 people exposed. Mild, transient bumps in the liver enzymes are more common than that and rarely require stopping.
The precautions that follow from those numbers are proportionate. Terbinafine is contraindicated in chronic or active liver disease, and the FDA labeling directs that ALT and AST be checked before the first tablet, which is a single blood draw that also catches the person who did not know they had a fatty liver or a quiet hepatitis. Whether to repeat those labs mid-course is debated among dermatologists; analyses of routine monitoring have found that abnormalities meaningful enough to change management are uncommon, and that no scheduled draw reliably catches injury before symptoms do.
What does the work instead is knowing the warning signs, since injury usually appears within the first 6 weeks. Persistent nausea, loss of appetite, unusual fatigue, pain in the upper right side of the abdomen, dark urine, pale stools, or any yellowing of the eyes or skin means stop the medication and call the same day. That instruction is worth more than a calendar of blood draws, and it is the part of the conversation I refuse to rush.
What medications interact with terbinafine?
Terbinafine inhibits CYP2D6, one of the liver enzymes that clears a long list of common drugs, and it can turn a normal metabolizer into a poor one for the duration. The classes named in the labeling and in the pharmacology literature are tricyclic antidepressants, selective serotonin reuptake inhibitors, beta blockers, and class 1C antiarrhythmics, with nortriptyline, desipramine, perphenazine, metoprolol, and propafenone named specifically. In pharmacokinetic study, desipramine exposure rose about 5-fold when terbinafine was added, and the effect was still measurable 4 weeks after the terbinafine was stopped.
So the medication review happens before the prescription is written, and it includes anything picked up from another prescriber. In most cases the answer is careful monitoring or a temporary dose adjustment rather than abandoning the plan, and the lingering effect after the last tablet is the piece people are most often not told about.
Itraconazole is the usual alternative when the organism turns out to be a yeast or a non-dermatophyte mold, or when terbinafine is off the table. It carries its own constraints: the FDA labeling says it should not be used for onychomycosis in anyone with ventricular dysfunction or a history of congestive heart failure, and its interaction footprint is wider than terbinafine's because of how strongly it affects CYP3A4.
Get Real Answers
Tired of being told your labs are 'normal'? Dr. Ash digs deeper.
Why does the nail still look bad when the pills are done?
Because the medication clears the fungus long before the nail can replace itself, and the two timelines are nowhere near each other. Toenails grow at roughly 1 to 1.5 mm a month, so a great toenail needs 12 to 18 months to grow out from the matrix to the tip. The damaged nail you are looking at on the last day of treatment was already damaged before the first tablet, and nothing can speed its exit.
This is why progress gets judged at the base rather than at the tip. A band of normal, smooth, ordinary-colored nail advancing out from the cuticle is the finding that tells you the treatment worked, even while the far half of the nail still looks terrible. In the topical trials, more than half of the people whose fungus had been eradicated still had visible changes in the target nail at week 48 for this reason alone, and plenty of people abandon effective treatment at month 4 because they are reading the wrong end of the nail.
How do I keep it from coming back?
Recurrence runs about 20% to 25% after successful treatment, usually within 2 years, and most of that traffic comes from the skin rather than from a treatment failure.
Athlete's foot on the sole and between the toes is the reservoir that reseeds the nail, so treating it at the same time as the nail is the highest-yield habit on this list, and it is the step most often skipped. Household members with tinea pedis are worth treating alongside you for the same reason. Shoes hold spores, so the pairs worn heavily during the infection get disinfected or replaced, and rotating 2 or 3 pairs so each dries fully between wearings does more than it sounds like it should. Dry between the toes after a shower or a swim, because that web space is where this starts. Keep the nails trimmed short and filed thin, which lowers the pressure on the nail bed and gives topical maintenance a thinner barrier to cross. Wear something on your feet in gym showers, locker rooms, and on pool decks; a Philadelphia winter spent in warm boots and a summer of humid days do the rest of the work for the fungus without any help.
Topical antifungal maintenance after an oral course lowers recurrence in the published data, and the prevention literature suggests that for people with a history of this, prophylaxis may need to continue for as long as 3 years to hold the ground you gained.
What if the nail is curling into the skin?
A nail curling into the flesh beside it is an ingrown nail, which is a mechanical problem rather than an infectious one, and no antifungal will correct it. The two travel together often enough to cause confusion, because a thickened, distorted nail presses on the fold and the tissue underneath it, and that pressure is its own source of pain and its own doorway for bacteria.
They get treated separately and usually at the same time. The fungus gets the medication and the ingrown edge gets a procedure: numbing the toe and removing the offending spike of nail, sometimes with treatment of the matrix so that portion does not grow back. That work belongs with a podiatrist, and it is reasonable to have it done while the antifungal course is underway rather than waiting to see whether the nail improves on its own.
Does diabetes or poor circulation change the plan?
Yes, and it raises the stakes enough that a cosmetic complaint becomes a medical one. Onychomycosis is more common in people with diabetes, and in that setting it is associated with secondary bacterial infection, cellulitis, foot ulceration, and in advanced cases amputation. The mechanism is mundane and preventable: a thickened dystrophic nail injures the nail bed and the skin around it, and the fissures that tinea pedis opens in the sole and between the toes give bacteria a way in.
Peripheral vascular disease and neuropathy compound it from both directions. Poor circulation slows nail growth and blunts the delivery of any medication to the tissue, while neuropathy removes the pain that would otherwise tell you something has gone wrong. For anyone with diabetes or known peripheral arterial disease, the sequence is to confirm the diagnosis, treat the skin along with the nail, keep the nails professionally trimmed if reaching them is difficult, and have the feet examined on a schedule rather than when something hurts.
Guidance from the Clinic
Actionable Steps in Philly
A working plan for a nail that has not answered to anything.
- Get it identified before you treat it. Ask for a nail specimen: a clipping taken near where healthy nail begins, along with the debris underneath, sent for KOH microscopy or PAS staining. A positive KOH is enough to start treatment, and a negative one with a convincing picture earns a PAS or PCR rather than a shrug.
- Bring the full medication list to the visit. Include anything from another prescriber and anything over the counter. Terbinafine inhibits CYP2D6, so antidepressants, beta blockers, and certain heart rhythm medications need a look before the first tablet, and the interaction outlasts the course by about a month.
- Have the liver checked once, at the start. ALT and AST before the first dose, then learn the warning signs: persistent nausea, poor appetite, dark urine, pale stools, right upper abdominal pain, or yellowing of the eyes. Any of those means stop and call the same day.
- Treat the skin along with the nail. Athlete's foot between the toes and on the soles is the reservoir. Treating it during and after the oral course is what keeps the nail from being reinfected within a year or two.
- Fix the footwear. Disinfect or replace the shoes worn hardest during the infection, rotate 2 or 3 pairs so each dries out fully, change socks after a run along the Schuylkill, and wear something on your feet in gym showers and on pool decks.
- Measure from the base, monthly. Photograph the nail at the cuticle on the same day each month in the same light. A widening band of normal nail at the base is the result you are looking for, and it will be visible months before the tip looks presentable.
Key Takeaways
- Antifungal cream fails on nails because the fungus grows in the nail bed under a dense keratin plate that creams cannot cross; over-the-counter athlete's foot creams are labeled against use on nails.
- Oral terbinafine, 250 mg daily for 12 weeks for toenails and 6 weeks for fingernails, reaches the infection through the bloodstream and produces mycologic cure in roughly 70% to 80% of dermatophyte cases.
- Only about half of abnormal nails are fungal, so a nail specimen for KOH, PAS, culture, or PCR belongs before a 3-month prescription.
- Prescription nail solutions do cross the plate but produce complete cure in about 7% to 18% after 48 weeks of daily use; laser devices are cleared for appearance rather than for cure.
- Serious liver injury from terbinafine is rare at roughly 1 in 50,000 to 120,000 prescriptions; check ALT and AST before starting and know the warning signs, which usually appear in the first 6 weeks.
- Terbinafine inhibits CYP2D6, so tricyclic antidepressants, SSRIs, beta blockers, and class 1C antiarrhythmics need review, and the interaction persists about 4 weeks after the last dose.
- Recurrence runs 20% to 25% within 2 years, and treating the athlete's foot on the skin along with the nail is the single most effective thing that lowers it.
- Diabetes, peripheral vascular disease, or neuropathy raises the stakes, since onychomycosis in those settings is associated with cellulitis, foot ulceration, and amputation.
Related at Fishtown Medicine
- The Skin Playbook - how skin, hair, and nails read out what is happening underneath
- Acne in Teens and Adults - what happens when a look-alike condition gets the wrong treatment for years
- Dandruff and Seborrheic Dermatitis - the other common yeast-driven condition, and why contact time matters
- Eczema That Will Not Clear - when a stubborn skin problem points somewhere systemic
- Tattoo and Piercing Infections - what a spreading skin infection looks like and when it needs same-day care
Scientific References
- "Onychomycosis: Rapid Evidence Review." American Family Physician. 2021;104(4):359-367. AAFP
- Foley K, et al. "Topical and device-based treatments for fungal infections of the toenails." Cochrane Database of Systematic Reviews. 2020;1:CD012093. PubMed
- "Topical Antifungals for Treatment of Onychomycosis." American Family Physician. 2016. AAFP
- Terbinafine hydrochloride tablets, prescribing information. U.S. Food and Drug Administration. FDA label
- "Terbinafine." LiverTox: Clinical and Research Information on Drug-Induced Liver Injury. National Institute of Diabetes and Digestive and Kidney Diseases. NCBI Bookshelf
- Abdel-Rahman, et al. "Investigation of terbinafine as a CYP2D6 inhibitor in vivo." Clinical Pharmacology & Therapeutics. 1999. PubMed
- "An investigation of how fungal infection influences drug penetration through onychomycosis patients' nail plates." European Journal of Pharmaceutics and Biopharmaceutics. PubMed Central
- "Onychomycosis: An Updated Review." PubMed Central
- "Diagnosis of Onychomycosis: From Conventional Techniques and Dermoscopy to Artificial Intelligence." PubMed Central
- "Onychomycosis." StatPearls. National Library of Medicine. NCBI Bookshelf
- "Onychomycosis: Strategies to Minimize Recurrence." Journal of Drugs in Dermatology. 2016. PubMed
- "Onychomycosis: Treatment and prevention of recurrence." Journal of the American Academy of Dermatology. ScienceDirect
- "Utility of Laboratory Test Result Monitoring in Patients Taking Oral Terbinafine or Griseofulvin for Dermatophyte Infections." JAMA Dermatology. JAMA Network
- "A critical review of improvement rates for laser therapy used to treat toenail onychomycosis." PubMed
- "Onychomycosis in Diabetics: A Common Infection with Potentially Serious Complications." Life. 2025;15(8):1285. PubMed
- Itraconazole capsules, prescribing information. U.S. Food and Drug Administration. FDA label
Frequently Asked Questions
Common Questions
Deep-Dive Questions
Ready when you are
The chat is our AI assistant, answering from our published guides. To talk it through with Dr. Ash himself, schedule the free 20-minute call.





