A keloid crosses the line the wound made. A hypertrophic scar does not.
Both are raised scars, and early on they look alike. The difference is at the edges. A hypertrophic scar stays inside the borders of the original injury and usually softens over months. A keloid pushes out into skin that was never cut. A review of keloid management in richly pigmented skin puts it plainly: these lesions grow over time, often recur after treatment, and do not regress spontaneously.
Both begin in the same layer. Rei Ogawa describes keloids and hypertrophic scars as chronic inflammation in the reticular dermis, the deep layer of skin, set off by an injury that reaches that far down. His trigger list includes skin piercing, alongside surgery, vaccination, burns and acne. One line matters before any procedure: superficial injuries that do not reach the reticular dermis never cause keloidal or hypertrophic scarring. A post through the earlobe goes the full depth of the dermis, and tattoo needles place pigment there by design.
Keloids run in families, and one family can look nothing alike
A 2001 study of 14 keloid families, most of them African American, covered 341 relatives, 96 of whom had keloids. The pattern fit autosomal dominant inheritance with incomplete penetrance, so a parent can pass on the tendency without ever showing it. Severity within one family ranged from a single small earlobe keloid to keloids across large areas of the body. An aunt with a keloid is real information about you. A parent with clear skin is not a clearance.
How much more common on Black skin? Clearly more, and the number everyone repeats is softer than the confidence around it suggests. The line about keloids occurring roughly 20 times more often in people of African than European descent is the opening sentence of a 2014 admixture-mapping study of 478 African Americans. It is that paper's background framing, not its finding. What the study actually measured was genetics, and what it found was a keloid-associated region at 15q21.2-22.3 plus a stronger signal at the myosin gene MYO1E. Nobody counted keloids across two populations to arrive at 20. A separate US health-records study of 5,457 adults carrying a keloid diagnosis, written by authors at Pfizer and the health-data firm Oracle Life Sciences with no conflict statement on PubMed, found those patients concentrated in non-Hispanic Black, Hispanic and Asian groups, mostly female, mean age 34. That describes who gets diagnosed inside a health system, not how often keloids happen. Both are background. Neither is your personal probability.
The high-risk sites are the ones people most want decorated
A one-year series at a Nigerian tertiary centre documented 192 keloids in 120 patients with darkly pigmented skin. The chest was the commonest site at 19.3 percent, then the earlobe at 14.1 percent, then the face at 11.0 percent. Buttocks and feet were least affected. Trauma, including ear piercing, shaving and cuts, accounted for 56.2 percent of triggers.
Lay that over a flash sheet. Sternum and chest pieces, shoulder caps, upper back and back-of-neck work, and earlobe and cartilage piercings all sit in high-risk territory. Forearm, calf and thigh sit lower. If the design can move, moving it is the most useful thing here.
Age at ear piercing tracked with keloids in the one study that asked
In 2005, dermatologists at the Medical College of Georgia surveyed 32 consecutive patients who already had keloids from ear piercing. Those pierced at 11 or older were more likely to have developed a keloid, 80 percent, than those pierced before 11, at 23.5 percent. The authors' advice was direct: people with a family history should consider not piercing, and if piercing happens anyway, early childhood may be preferable. Read it for what it is. Every person surveyed already had a piercing keloid and there was no unaffected comparison group, so those percentages are proportions inside a keloid clinic, not the chance that a piercing scars you. A signal about timing, not a rate.
What a personal or family history should change before you book
- Move the placement. Forearm or calf instead of sternum or shoulder does more than any aftercare product on the shelf.
- Go one at a time. A single lobe piercing that heals clean is information. Six in one sitting turns one decision into six wounds at a high-risk site.
- Prefer the lobe over cartilage. A review of head and neck piercing complications notes infections after ear piercing spread rapidly through the auricular cartilage and often need surgery. Anything that drags inflammation out works against you, though that is reasoning from mechanism, not a measured rate.
- Book the dermatologist first. Ask what happens if a firm bump appears at week six, and whether they will inject early rather than watch and wait. If you have none, search our directory.
- Let no be an answer. A different placement, or none, is a defensible reading of the evidence.
The test patch idea is intuitive and does not hold up
The logic is appealing: one small dot of ink or one piercing, watch, then commit. Four problems.
- Depth. Injuries that do not reach the reticular dermis never produce keloids, so a shallow test says nothing about a full-depth one.
- Time. Keloids build over weeks to months and keep growing. A four-week look rules nothing out.
- Site. A clean forearm test says little about your sternum.
- Cost of a positive. If the test scars, you have a permanent keloid at a spot you picked casually.
We could not find a published study evaluating test tattoos or trial piercings for predicting keloid risk. What is defensible is a staged decision: one small piece at a lower-risk site, watched across several months, with a plan to treat at the first sign of thickening.
The first six weeks are where the leverage is
From about week three, watch for the scar getting firmer instead of flatter, itching that is not settling, darkening that creeps past the original hole or line, and a raised edge at six to eight weeks that has not begun to come down. Do not wait it out. Keloids do not regress spontaneously, so a few more months is a delay, not a plan. The 2014 international advisory panel on scar management called for more aggressive initial management, including earlier 5-fluorouracil. That is a consensus recommendation from an expert panel, not a trial result.
Silicone gel sheeting is the cheapest thing worth starting while you wait for an appointment, and you should know exactly how thin its evidence is. A Cochrane review of 20 trials in 873 people found that in scar-prone people it lowered the incidence of hypertrophic scarring compared with no treatment (risk ratio 0.46, 95 percent CI 0.21 to 0.98). The reviewers judged those prevention trials highly susceptible to bias and summed their own result up as weak evidence of a benefit, with poor research quality leaving a great deal of uncertainty. The outcome those trials counted was hypertrophic scarring, not keloids. Low cost, low risk, worth doing, not insurance. Pressure earrings need 12 to 20 hours a day for months to do anything, per the American Academy of Dermatology.
The treatments that work, and the one that backfires alone
Steroid injection is the standard first move. The AAD puts shrinkage at 50 to 80 percent of injected keloids and notes many regrow within five years. Adding 5-fluorouracil is where the strongest evidence sits: a 2024 systematic review of keloid and hypertrophic-scar trials pooled 13 studies, 12 randomised, and found the combination beat triamcinolone alone (odds ratio 3.45, 95 percent CI 2.22 to 5.35) and 5-fluorouracil alone (odds ratio 4.17), with less telangiectasia than steroid alone.
Surgery alone is the classic mistake. The AAD states nearly 100 percent of keloids return after excision by itself. A 2024 meta-analysis pooled 22 studies covering 608 keloids treated by intralesional excision, which takes the keloid out from inside its own shell, and put recurrence at 13 percent across an average 19.2 months of follow-up. The authors set that against the 45 to 100 percent reported in earlier work after complete excision, so the comparison runs across separate literatures rather than a head-to-head trial. For ears, a trial of 60 patients found 73.3 percent recurrence-free at six months after excision plus 5-fluorouracil and triamcinolone, versus 43.3 percent after excision plus radiotherapy. Six months is a short look at a scar the AAD says often comes back within five years.
Laser is oversold. A 2022 Cochrane review of 15 randomised trials in 604 participants found insufficient evidence to support or refute laser for these scars, with every trial at high risk of bias in at least one domain. Cryotherapy takes three or more sessions to give its best results, per the AAD. Our fuller piece on keloid scars on Black skin walks the whole ladder.
One honest gap: tattoo-specific keloid data is thin. Piercing appears on every standard trigger list, and we could not find a sound estimate of how often keloids follow tattooing. On removal, a chart review of 1,041 patients who had at least five Q-switched Nd:YAG sessions reported 0.28 percent hypertrophic scarring and zero keloids. Weigh who ran it. The authors practise at LaserAway, a commercial laser tattoo-removal chain, so this is a provider reporting its own patients' outcomes, and PubMed carries no conflict statement for it. It is also one protocol at one practice, with no breakdown by skin tone. Related: acne keloidalis nuchae and dark spots and hyperpigmentation.
Frequently asked questions
Can I get a tattoo if I already have a keloid somewhere else? ▼
Physically yes, and whether you should depends on how the existing one behaved. A small earlobe keloid from a childhood piercing carries different weight than a spreading chest keloid. If you go ahead, keep the placement off the chest, shoulder and upper back.
Does numbing cream, a particular ink, or a scar oil lower keloid risk? ▼
We found no evidence that numbing agents, ink brand, or over-the-counter scar oils change keloid risk. Depth of injury and how long the deep dermis stays inflamed are what matter. The only prevention with trial evidence behind it is silicone gel sheeting, and Cochrane rates that evidence weak.
If a keloid starts at a piercing, should I take the jewelry out? ▼
Removing an ongoing irritant is reasonable, but it will not fix anything by itself. Keloids do not regress spontaneously, so taking the jewelry out and waiting is the version that ends in a large keloid a year later. Book the dermatologist that week.
How long after a piercing or tattoo can a keloid still appear? ▼
Keloids grow in over time rather than arriving fully formed, so a clean site at two weeks proves little. MedlinePlus advises protecting the area for at least six months after an injury in adults and up to 18 months in children.
Should I have my child's ears pierced early to lower the risk? ▼
Among 32 people who already had piercing keloids, those pierced at 11 or older were more likely to have developed one, 80 percent, than those pierced earlier, at 23.5 percent. The authors suggested families with a keloid history consider not piercing. Everyone in that survey already had a keloid and there was no unaffected comparison group, so treat it as one input to a family conversation rather than a risk estimate.