Seventy-Seven Percent, Attributed to a Paper That Does Not Contain It

A brown stripe running the length of a nail is called longitudinal melanonychia. It is usually nothing. Occasionally it is a melanoma that will take the finger, and sometimes more.

Everything about how that stripe gets handled — reassure, photograph, biopsy — depends on how common it is in the person sitting there. Which is why the number matters, and why it is worth knowing that the number is not real.

Following the citation home

The triad appears in almost every review: about 1 percent in white people, 10 to 20 percent in Japanese and Asian people, 77 to 100 percent in African Americans.

A 2020 review in the Indian Dermatology Online Journal prints it in exactly those words and hangs it on a single reference — a 2006 chapter on pigmented nail disorders in Dermatologic Clinics.

That 2006 piece is itself a narrative review. No sample size, no population, no survey. Its own source appears to be a paper from 1972, for which no abstract exists in the database and which I could not obtain.

So the chain is: review citing review citing something from half a century ago that nobody can read.

It gets worse, and this part is a matter of public record rather than inference. A 2023 peer-reviewed case series states that darker skinned ethnicities have a higher prevalence of longitudinal melanonychia in the US population — African Americans with 77 percent prevalence — and cites, for that figure, a specific paper in the Indian Journal of Dermatology from 2021.

That paper is a prospective dermoscopy study of 108 lesions in 35 children and 73 adults, conducted in Iran. I pulled its complete text — nearly 23,000 characters — and searched it. The word “African” appears zero times. The string “77” appears zero times.

The figure has been attributed, in print, in 2023, to a source that does not contain it.

What has actually been counted

The only true population-prevalence survey I could verify examined the fingernails and toenails of 2,457 Chinese patients attending a clinic, excluding anyone of mixed background.

Melanonychia striata was not detected at all in patients under twenty. It was present in 0.6 percent of those aged 20 to 29, and rose gradually to 1.7 percent at fifty and over. Thumbs were most often affected, then index fingers. It was bilateral in 40 percent of cases, and nobody had more than two digits involved.

The authors’ conclusion: melanonychia striata is uncommon in Chinese individuals, especially those under twenty.

Set that against “10 to 20 percent in Japanese and Asians.” One of those two statements comes from counting 2,457 people. The other comes from the citation chain above.

The rest of the literature counts people who already have a band, which is a different question with a different denominator and is constantly reported as if it were the same thing.

Sixty-eight consecutive patients in Mexico City presenting with longitudinal melanonychia, all biopsied: 68 percent racial melanonychia, 5.7 percent benign melanocytic hyperplasia, 5.7 percent a nail apparatus malignancy. That 68 percent is the proportion of bands that turned out to be constitutional, among people who came in with a band. It is not the proportion of Mexican people who have one.

Two hundred and seventy-five Korean patients with melanonychia: the commonest cause was subungual hemorrhage at 29.1 percent, then nail matrix naevus at 21.8, trauma at 14.5, lentigo at 11.6 — and ethnic-type nail pigmentation at 8.0. Melanoma in 6.2 percent.

Eight percent. In a Korean melanonychia clinic, constitutional pigment was the fifth commonest explanation, not the first.

And the largest cohort located, 1,556 patients over eighteen years in Singapore, 80.7 percent Chinese: of 274 referred, 198 were biopsied, and eight nail apparatus melanomas were found — 4.0 percent — including three in situ. Age fifty and over and blue-collar occupation were associated with malignancy. One in five of those biopsies produced a complication, most often nail dystrophy.

The rule that did not survive its test

The standard clinical mnemonic is ABCDEF, published in 2000 — in a paper whose actual title is “The ABC rule for clinical detection of subungual melanoma,” which is a small thing but tells you how casually this has all been handled.

A is age, with peak incidence in the fifth to seventh decades, and the populations named as higher risk. B is a brown to black band 3 millimeters or more across with variegated borders. C is change, or the absence of change despite treatment. D is the digit involved. E is extension of pigment onto the nailfold — Hutchinson’s sign. F is family or personal history of melanoma or dysplastic naevi.

The authors describe their method plainly: they reviewed the world literature and arranged the available information into a system. It was not derived from a cohort and it has never been validated against one prospectively.

It was tested retrospectively, once, in 84 biopsied cases — eight subungual melanomas and 76 benign. Patients with melanoma were younger, had had the band longer, and presented with a wider band and a greater width percentage.

And the number of ABCDEF criteria met did not differ between the groups.

The criterion that did separate them was the proportion of the nail plate the band occupies. The authors’ recommendation is that biopsy be performed for any concerning band, and especially where the width percentage is above 40.

What twelve nails did to 152 dermatologists

This is the study I keep coming back to.

Twelve cases of melanonychia — five melanomas, seven not — were shown to 152 dermatologists at two meetings, in four successive steps: clinical evaluation, then evaluation by the ABCDEF rule, then nail plate dermoscopy, then intraoperative dermoscopy. The audience included eleven nail experts, 53 senior dermatologists and 88 juniors.

Across the first three steps, the percentage getting the right answer ran from 46 to 55.

Only intraoperative dermoscopy — looking at the matrix directly, in the operating room, with the nail plate off — significantly improved accuracy. And the level of expertise did not statistically influence the correct diagnosis. Nail experts did not do better than juniors.

The authors’ own summary is that overall accuracy in diagnosing nail matrix melanoma in situ is low. That is not a comment on those 152 clinicians. It is a comment on what can be known by looking at a nail.

Children, where the bands look worst and almost never are

A meta-analysis pooled 24 studies and 1,391 children with longitudinal melanonychia.

Dark bands in 69.8 percent. Multiple colors in 47.6. Broad bandwidth in 41.1. Pseudo-Hutchinson sign in 41.0. Irregular patterns in 38.1. True Hutchinson sign in 23.7. Nail dystrophy in 18.2. Nearly a third widened or darkened over follow-up.

Eight cases of melanoma in situ. Zero invasive melanomas.

Every feature that would have an adult in the operating room, present in large fractions of these children, and not one invasive cancer in 1,391 of them.

The comparative work explains why. A dermoscopic analysis of nail matrix naevi found that in children they were darker and more multicolored than in adults, with pseudo-Hutchinson sign, triangular sign and dots and globules all significantly more common. Children’s benign lesions simply carry more melanoma-associated features.

The natural-history data says the same thing from the other end: in 703 lesions across 381 children, roughly 3 percent had fully regressed within three years, 5 percent within four and a half, and 10 percent within nine and a half. The authors’ recommendation is that clinicians follow children without intervention for several years even if the lesions grow darker or wider.

Two further series make the point about what biopsy actually delivers. Fifty-four children biopsied at one institution: no melanoma in situ, no invasive melanoma, and no association at all between physical examination features and high-risk pathology — with twelve going back for re-excision. Sixty-two children in Singapore: no subungual melanoma, and consensus among all three dermatopathologists in only half the cases biopsied.

Half. Even the reference standard wobbles.

What nobody has measured

I went looking for the study that measures time to biopsy or referral for nail pigmentation by skin color. I searched for it several different ways and did not find one.

What exists is one small series of 19 nail apparatus melanoma patients — Caucasian and African American patients in equal proportion — with a mean pre-diagnosis duration of 33 months. Nearly three years from noticing to diagnosis, not broken down by race, because with nineteen patients it could not be.

There is a large SEER analysis showing African American patients four times more likely than white patients to present with stage IV melanoma, and a 1.48-fold higher risk-adjusted stage-specific mortality. That is cutaneous melanoma overall. It is not a nail-referral figure and should not be recycled as one.

Where that leaves the stripe on your nail

If you are an adult with a single new band on one digit, the base rate that would tell you how worried to be does not exist for most of the world’s population. In the referral series that do exist, between 4 and 10 percent of biopsied adult bands turn out to be malignant.

The features that carry real weight are the ones that survived testing: how much of the nail plate the band occupies — above about 40 percent is the published threshold — whether it is changing, whether the nail itself is breaking down, and pigment spilling onto the surrounding skin. In the Singapore series, every invasive melanoma had all three of nail plate destruction, nail bed ulceration and a positive Hutchinson sign.

In a child, the same features mean something close to the opposite, and the published position is watchful waiting rather than surgery.

And if a clinician tells you a band is nothing because it is common in people like you, the honest version of that sentence is that nobody has ever counted. It might still be nothing. But the reassurance is being drawn from a number that was never measured, passed down through five decades of reviews, and attributed in 2023 to a paper about 108 lesions in Iran.

This article is educational and is not a substitute for care from someone who can look at your skin. Anything new, changing, painful or spreading should be assessed rather than researched.