Frank’s Sign Images: What This Earlobe Crease Indicates

A diagonal crease running across the earlobe, known as Frank’s sign, is a physical marker linked to an increased likelihood of coronary artery disease, stroke, and metabolic conditions. First described in 1973 by the American physician Sanders T. Frank, the crease typically runs at roughly a 45-degree angle from the tragus (the small pointed flap near the ear canal) toward the outer edge of the earlobe. If you have been looking at images of this crease and wondering whether yours matches, what matters is not just whether a line exists but how deep and complete it is, and what the broader evidence says about its clinical meaning.

What Frank’s Sign Looks Like

When you search for images of Frank’s sign, you will see creases that vary enormously in depth and length. Not every wrinkle on the earlobe qualifies. The classic crease is diagonal, cutting across the lobe rather than following the natural curve of the ear. It is distinct from the fine wrinkling that comes with general skin aging or from the small indentations left by years of wearing heavy earrings.

Clinicians grade the crease on a scale that helps standardize what they are looking at:

Grade 1: A small amount of wrinkling on the earlobe, barely noticeable.
Grade 2a: A superficial crease running across the earlobe.
Grade 2b: A crease extending more than halfway across the earlobe.
Grade 3: A deep crease spanning the entire earlobe from the tragus to the outer edge.

The grading matters because research consistently shows that deeper, more complete creases carry a stronger statistical association with cardiovascular disease. A Grade 3 crease on both ears is the presentation most strongly tied to underlying arterial problems, while a faint Grade 1 crease on one side is far less concerning. Whether the crease appears on one ear or both also matters: bilateral creases show a stronger link to disease than unilateral ones.

The Link to Coronary Artery Disease

The evidence connecting Frank’s sign to coronary artery disease has been building for five decades. A study of patients undergoing coronary angiography found that bilateral creases appeared in 67% of those with obstructive coronary artery disease, compared to 33% of those without blockages. After adjusting for traditional risk factors like smoking, diabetes, and abnormal cholesterol, the bilateral crease remained independently associated with coronary disease.¹

In patients under 65, where you might think the crease is less relevant, the association actually appears quite strong. One study found that Frank’s sign was independently associated with severe coronary artery disease with roughly triple the odds, even after accounting for sex and cholesterol levels.² That finding is striking because younger patients are not typically expected to show external markers of heart disease.

An autopsy-based study offered even more dramatic numbers. Frank’s sign was present in about two-thirds of people who died from cardiovascular causes, compared to roughly a third of those who died from other causes. The crease was found in over three-quarters of cases with critical coronary artery blockages. After adjusting for age, sex, and ethnicity, the sign remained an independent predictor of cardiovascular death.³ Autopsy data is useful here because it eliminates the uncertainty of imaging; the actual state of the arteries is directly visible.

Another observational study found that estimated cardiovascular mortality risk climbed as the crease became more prominent. Among people with no crease, about a quarter fell into the moderate-to-very-high cardiovascular risk category. That proportion jumped to roughly a third for those with a unilateral crease and to well over half for those with bilateral creases.⁴

The Crease and Stroke Risk

The connection is not limited to heart disease. A study of 241 patients hospitalized for acute stroke found Frank’s sign in nearly 79% of them. The crease was significantly more common among those with full cerebrovascular accidents (about 89%) than among those with transient ischemic attacks (about 73%).⁵ That difference matters because full strokes involve more extensive damage to the brain’s blood supply, suggesting the crease may track with the severity of vascular disease rather than just its presence.

A separate study found a significant association between ischemic strokes and Frank’s sign even after excluding patients who already had known cardiovascular disease.⁶ That is a meaningful detail, because it suggests the crease is not merely a secondary marker of conditions you would already know about. It may flag vascular deterioration before other symptoms appear.

Metabolic Connections Beyond the Heart

Frank’s sign has also been linked to a cluster of metabolic problems that go well beyond blocked arteries. A cross-sectional study found that after adjusting for the major coronary risk factors, people with the crease had roughly double the odds of metabolic syndrome. The effect was somewhat stronger in women, where the adjusted odds ranged from about 2.4 to nearly 3 times higher than in women without the crease.⁷

Other research has found associations between the crease and individual risk factors including high blood pressure and elevated blood glucose, independent of body mass index.⁸ The crease appears more often in men and becomes more common with age, which overlaps heavily with the risk profile for cardiovascular disease generally. Disentangling the crease from the other risk factors that tend to accompany it is the central challenge in this field of research.

Why the Crease Forms

The most common question people have after seeing Frank’s sign images is a simple one: why would the state of your arteries show up on your earlobe? The earlobe is one of the few body parts that gets its blood supply from a single small artery with no backup circulation. This makes it especially vulnerable when blood flow deteriorates.

A histopathological study examined earlobe tissue from cadavers and found distinctive changes at the base of the crease: thickening and scarring of arterial walls, widespread tissue fibrosis, and degeneration of small nerve fibers. These changes mirrored similar deterioration found in the heart muscle of the same individuals.⁹ The picture that emerges is that the crease is not random wrinkling. It reflects actual structural damage to tiny blood vessels and connective tissue, driven by the same process that narrows coronary arteries.

A forensic study added further support by showing that the crease grade correlated significantly with both the severity of coronary artery disease and the degree of general atherosclerosis throughout the body. People who died with high-grade creases were older, had heavier hearts, and more frequently had a cardiac cause of death.¹⁰

An alternative or complementary theory suggests that facial fat distribution plays a role. The idea is that visceral fat deposits around the face and ears may physically contribute to crease formation while also reflecting the kind of metabolic profile that drives cardiovascular disease.¹¹ This would help explain why the crease becomes more common with age and weight gain, even if the underlying arterial disease process is the primary driver.

How Reliable Is It as a Diagnostic Tool

Here is where the story gets more complicated, and where the gap between “associated with” and “useful for diagnosis” becomes important. A systematic review of diagnostic accuracy studies found that the crease’s sensitivity for detecting coronary artery disease ranged from 26% to 90%, and its specificity ranged from 32% to 96%. Those ranges are so wide that they are nearly meaningless on their own. Most of the positive likelihood ratios fell below 2, which in practical terms means the crease does not shift the probability of disease enough to be useful by itself.¹²

The autopsy-based study mentioned earlier reported better individual numbers, with sensitivity around 84-88% and specificity around 51-58% for critical coronary blockages and cardiovascular death.³ But even those researchers characterized the crease’s discriminative power as moderate. The angiography study that confirmed bilateral creases as an independent predictor explicitly noted that its diagnostic accuracy was comparable to, not better than, traditional risk factors like smoking or diabetes.¹

In other words, Frank’s sign is a real signal buried in a lot of noise. It is not a reliable standalone screening tool. Its value lies in being free, instantly visible, and additive: if you already have risk factors for heart disease and you also have a deep bilateral crease, that is one more piece of information pointing in the same direction. No doctor should order a cardiac workup based solely on an earlobe crease, but no doctor should ignore it either.

Frank’s Sign in Young, Apparently Healthy People

One of the more unsettling findings in recent research is that Frank’s sign does appear in younger people who have no symptoms of heart disease. A study of 853 healthy young individuals found the crease in about 15% of them. Among those who had it, roughly two-thirds also had a family history of at least one chronic disease, which was significantly more common than in those without the crease.¹³

This does not mean that a 25-year-old with a faint earlobe crease is heading for a heart attack. The crease’s predictive power strengthens with age, which makes sense if it reflects cumulative vascular damage. But the familial association raises the possibility that genetic factors influencing connective tissue or vascular health may predispose some people to developing the crease earlier. If you are young and notice a crease, it is worth mentioning to your doctor as context rather than treating it as a diagnosis.

Artificial Intelligence and Automated Detection

Researchers have begun training machine-learning models to detect Frank’s sign from images, which could eventually make screening faster and more consistent. One group created the first publicly available dataset of earlobe crease photographs and tested a range of deep learning architectures on it, achieving 97.7% accuracy in detecting the crease from annotated photos.¹⁴ The researchers found that a lightweight model called MobileNet offered the best balance between accuracy and computational cost, making it a candidate for smartphone-based screening tools.

A separate effort went further, training deep learning models not on earlobe photographs but on brain MRI scans to detect Frank’s sign, achieving an area under the curve greater than 0.9.¹⁵ The logic of using MRI is that the crease might correlate with structural brain changes associated with vascular disease, providing an additional pathway for detection in patients already undergoing imaging. Both approaches are experimental and not in routine clinical use, but they suggest the crease could eventually be flagged automatically during health assessments.

Can Cosmetic Treatments Remove the Crease

Some people discover Frank’s sign not because of health concerns but because the deep crease bothers them aesthetically. Hyaluronic acid fillers, the same type used in lip and cheek augmentation, have been used to plump up the earlobe and reduce the crease’s appearance. Research on this approach found that fillers addressed volume loss, reduced crease depth, and improved overall earlobe appearance in ways that surgical correction alone may not achieve.¹⁶

There is an important caveat here that deserves emphasis: filling in the crease cosmetically does absolutely nothing to address the underlying vascular changes that caused it. The crease is a surface marker. Smoothing the surface does not reverse arterial stiffening or atherosclerosis. If you choose to treat the crease for cosmetic reasons, that is perfectly reasonable, but it should not substitute for cardiovascular risk assessment. Think of it the way you would think about covering a check-engine light with tape: the dashboard looks fine, but the engine has not changed.

When a Crease Is Just a Crease

Not every diagonal line on the earlobe is Frank’s sign, and not every Frank’s sign means disease is present. Several factors can produce earlobe creases that have nothing to do with vascular health. Sleeping on one side habitually can press a crease into the earlobe over years. Heavy earrings worn for decades can stretch and groove the tissue. Natural aging thins and loosens skin everywhere, including the earlobes, producing folds and lines that overlap visually with the diagnostic crease.

The specificity problem in the diagnostic accuracy research reflects this reality. A meaningful percentage of people with the crease turn out to have clean arteries, and some people with advanced coronary disease have smooth earlobes. Ethnicity also plays a role in baseline earlobe shape and skin folding patterns, which is why the better-designed studies adjust for it. Obesity independently affects earlobe anatomy, and the facial visceral adiposity hypothesis suggests that some creases may be driven more by fat distribution than by arterial damage.

The crease is most informative when it appears bilaterally, is deep and complete (Grade 2b or 3), and shows up in someone with other risk factors. A shallow unilateral crease in an otherwise healthy person without a family history of cardiovascular disease is much less likely to mean anything clinically significant. Context is everything with this sign, which is exactly why it resists being turned into a simple yes-or-no screening test.

1
PubMed. The relationship of diagonal earlobe crease (Frank’s sign) and obstructive coronary artery disease in patients undergoing coronary angiography
2
Medicina Clínica (English Edition). Frank’s sign associated with the severity of ischemic heart disease in patients under 65 years old
3
PubMed Central. Relationship Between Frank’s Sign and Cardiovascular Disease: An Autopsy-Based Study
4
The American Journal of Medicine. Frank’s Sign and Cardiovascular Risk: An Observational Descriptive Study
5
PubMed. Diagonal Earlobe Crease (Frank’s Sign): A Predictor of Cerebral Vascular Events
6
PubMed Central. Frank’s Sign: A Clinical Predictor of Ischaemic Strokes
7
Epidemiology and Health. Association Between Earlobe Crease and the Metabolic Syndrome in a Cross-sectional Study
8
Biomedical Journal of Scientific & Technical Research. Frank’s Sign as a Predictor of Cardiovascular Disease
9
PubMed Central. The Histological Basis of Frank’s Sign
10
PubMed Central. The prognostic value of the Frank sign
11
PubMed Central. Unified Anatomical Explanation of Diagonal Earlobe Creases, Preauricular Creases, and Paired Creases of the Helix
12
PubMed Central. Diagonal Earlobe Crease (Frank’s Sign) for Diagnosis of Coronary Artery Disease: A Systematic Review of Diagnostic Test Accuracy Studies
13
Springer Link / Forensic Science, Medicine and Pathology. Prevalence of Frank’s sign in healthy young individuals, morphological characteristics, and its association with family history of chronic disease
14
arXiv. Deep Learning for Diagonal Earlobe Crease Detection
15
Scientific Reports. Advancements in Frank’s sign Identification using deep learning on 3D brain MRI
16
PubMed. Rejuvenating earlobe esthetics with dermal fillers