What actually treats toenail fungus? Oral pills clear it best; creams work slowly

A thickened, yellow, crumbling toenail is one of the most searched-for nuisances in foot care, and the honest answer to what fixes it is less cheerful than for athlete's foot on the skin: the drugs that work best are pills, they take months, and even the good options leave a fair share of nails uncured. Two Cochrane reviews — one of oral treatments, one of topical and device-based ones — put numbers on that trade-off [s1] [s2].

Why nails are hard to treat

Fungal infection of the toenails, called onychomycosis, damages the nail's structure and appearance and, when severe, can interfere with normal daily activities [s1]. The nail plate is the problem: its physical properties mean traditional topical treatments have low success rates, because the drug struggles to reach the fungus living under and within the nail [s1]. Oral treatments, by contrast, reach the nail bed through the bloodstream and tend to have shorter treatment times and better cure rates [s1].

What the oral evidence shows

The oral review pooled 48 randomised trials involving 10,200 participants, most with infection at the far end of the toenail [s1]. It found high-quality evidence that terbinafine beats placebo for clinical cure, with a risk ratio of 6.00 (95% confidence interval 3.96 to 9.08, across 8 studies and 1006 participants), and for mycological cure — clearing the fungus on laboratory testing — a risk ratio of 4.53 (2.47 to 8.33) [s1]. Azoles such as itraconazole also beat placebo, with a clinical-cure risk ratio of 22.18 (12.63 to 38.95, 9 studies, 3440 participants) [s1]. Head to head, there was moderate-quality evidence that terbinafine was probably more effective than azoles for achieving clinical cure [s1].

The reassuring part is safety. Among terbinafine-treated participants, side effects included gastrointestinal symptoms, infections and headache, but there was probably no significant difference in their risk compared with placebo (risk ratio 1.13, 95% CI 0.87 to 1.47) [s1]. Azoles showed slightly more adverse events — commonly headache, flu-like symptoms and nausea — though that difference was probably not significant either (risk ratio 1.04, 0.97 to 1.12) [s1]. The review is about averages in trials, not a licence to self-prescribe: oral antifungals interact with other medicines and are prescription drugs a clinician has to weigh for each person.

What the creams and devices show

The topical review, updated in 2020, included 56 studies and 12,501 participants, most with mild-to-moderate infection that had not reached the nail matrix [s2]. The medicated lacquers and solutions work, but the cure rates are humble and the courses long — most studies ran 48 to 52 weeks [s2]. Efinaconazole 10% solution was more effective than its vehicle at achieving complete cure, with a risk ratio of 3.54 (95% CI 2.24 to 5.60, 3 studies, 1716 participants), rated high-quality evidence, with only slightly more adverse events such as dermatitis [s2]. Ciclopirox 8% lacquer may also help, with a complete-cure risk ratio of 9.29 (1.72 to 50.14) against vehicle, though that estimate rests on low-quality evidence and a very wide interval [s2]. "Complete cure" here is a strict outcome: a normal-looking nail plus laboratory evidence the fungus is gone [s2].

What it adds up to

For a confirmed toenail infection, the evidence points one way: oral terbinafine has the strongest and best-quality support, azoles are an alternative, and topical solutions are a slower, lower-cure option better suited to milder disease or people who cannot take pills [s1] [s2]. None of it is fast, and none guarantees a clear nail — even the best oral figures leave a meaningful minority uncured [s1].

Two cautions sit underneath the numbers. First, the diagnosis matters: trials enrolled people whose infection was confirmed by culture, microscopy or examination of the nail [s1] [s2], and not every discoloured or thickened nail is fungal — psoriasis, trauma and other conditions mimic it, so treating blind can mean months of a drug aimed at the wrong problem. Second, the quality of the underlying trials varied: the oral review judged only one of 48 studies at low risk of bias across all domains, and rated 18 at high risk in at least one [s1]. That does not overturn the direction of the results, but it widens the uncertainty around their exact size. This article describes what the evidence shows, not what any individual should take; the choice between a pill and a lacquer — and whether treatment is warranted at all — is a clinical one.