A friend spent 3,400 dollars over eighteen months on fractional laser sessions for her acne scars. The dermatologist took before-and-after photos. The after photos looked better, mostly because the lighting was kinder. The actual scars she had been most bothered by, the deep narrow pits along her lower cheeks, looked almost identical.
The laser had worked on some of her scars. It had not worked on the ones she cared about. She had not been told this was likely before the first session.
This is the most common pattern I see in acne scar consults. The patient has multiple scar types. The treatment selected addresses one or two of them. The patient evaluates the outcome based on the scars she actually notices, which are usually not the ones the treatment was designed for. Everyone is frustrated.
The fix is upstream. Naming the morphology before selecting the treatment.
The three atrophic categories
The classification I keep returning to was published by Jacob and colleagues in 2001 (Jacob et al., Dermatol Surg 2001, PMID: 11176061). They divided atrophic acne scars into three morphologic categories based on the geometry of the depression. The categories matter because they predict, with reasonable accuracy, which treatments produce meaningful change and which do not.
Ice-pick scars are narrow and deep. The opening at the skin surface is small, often less than 2 millimetres wide, and the scar tracks straight down into the dermis like a punched hole. The depth typically exceeds the width. They are most common on the cheeks and temples.
Boxcar scars are wider and have sharply defined vertical walls. The base of the scar is roughly flat. The opening at the skin surface is 2 to 5 millimetres wide. They can be shallow or deep, and the shallow ones respond very differently than the deep ones.
Rolling scars are the broadest and have sloping, ill-defined edges. They produce a wavy or undulating surface rather than a discrete pit, and the surrounding skin slopes gradually into the depression. They are caused by fibrotic tethering of the dermis to deeper tissue, which is a different mechanism from the dermal volume loss that drives the other two.
Goodman and Baron extended this framework with a quantitative scoring system (Goodman & Baron, Dermatol Surg 2006, PMID: 17173588). Their grading scale separates the three morphologies and rates severity on a four-point scale, which is what most cosmetic dermatologists use now when they document scar treatment outcomes.
The reason the framework matters: each morphology responds best to a different intervention. The treatment that works on rolling scars often misses ice-picks, and vice versa.
What the studies actually show
The treatment evidence sorts by morphology.
For ice-pick scars, the most effective single intervention in the literature is TCA CROSS (Chemical Reconstruction Of Skin Scars). Lee and colleagues published an early study using 65 to 100 percent trichloroacetic acid applied with a wooden toothpick directly into the base of the scar (Lee et al., Dermatol Surg 2002, PMID: 12230882). The acid causes controlled chemical injury, the resulting collagen deposition raises the base of the scar, and after multiple sessions the depression becomes shallower.
The protocol is uncomfortable but effective. Each session treats individual scars one at a time. Multiple sessions are required, typically 3 to 6 spaced 4 to 6 weeks apart. The improvement in scar depth ranges from 50 to 70 percent in most series, which is far better than most other interventions achieve on ice-picks.
Fractional laser, despite its marketing, is less effective on true ice-pick scars. The narrow geometry means the laser energy does not adequately disrupt the scar’s deep base. Ablative fractional CO2 can help with the surrounding texture, but the deep narrow pits often persist.
For boxcar scars, the treatment evidence is more flexible. Shallow boxcars (less than 1 millimetre deep) respond to ablative fractional laser, microneedling at depth, and chemical peels. Deep boxcars often need a combination approach including punch excision for the deepest scars, fractional laser for the surrounding texture, and sometimes filler to raise the base while the collagen remodels.
Alster and Tanzi published a useful review of the laser data (Alster & Tanzi, J Am Acad Dermatol 2007, PMID: 17328993). Their summary is that fractional non-ablative lasers like Fraxel produce moderate improvement in boxcar scars over multiple sessions, while ablative fractional CO2 produces larger improvement per session but with significantly more downtime and a higher risk of post-inflammatory hyperpigmentation in higher Fitzpatrick types.
For rolling scars, the treatment of choice is subcision. The mechanism is mechanical: a needle is inserted under the scar and used to cut the fibrotic tethers that pull the dermis down. Orentreich and Orentreich originally described the technique decades ago (Orentreich & Orentreich, Dermatol Surg 1995, PMID: 7600028) and it remains the most effective intervention for the rolling morphology because the underlying pathology is tethering, not dermal volume loss.
Subcision results are immediate but often partial. The tethers reform somewhat over weeks. Multiple sessions are typical, sometimes combined with HA filler placed in the subcised pocket to maintain elevation while collagen deposits.
The combination approach is what most experienced cosmetic dermatologists use. TCA CROSS for ice-picks, fractional laser or microneedling for boxcars and surrounding texture, subcision plus filler for rolling. The treatment plan should match the scar inventory, not the other way around.
The diagnostic exercise
This is what I recommend to readers before they spend money on a consultation.
Take three photographs of each cheek and the chin under different lighting. Front-lit, side-lit from the left, and side-lit from the right. The lighting changes which scars are visible and which are hidden, and a treatment plan should account for what shows up in each.
Mark each visible scar on a printout or in an annotation app. For each scar, ask three questions.
Is the opening narrower than my pen tip (around 2 millimetres)? If yes, it is probably an ice-pick.
Does the depression have sharply defined walls and a flat base? If yes, it is probably a boxcar.
Does the depression slope gradually into the surrounding skin without a sharp edge? If yes, it is probably a rolling scar.
You will likely have a mix of all three. Most adults with moderate post-acne scarring have at least two morphologies and many have all three.
The next question is which ones bother you most. The treatment plan should prioritise the morphology that drives the most cosmetic concern, not the morphology that is easiest to treat.
I have done this exercise with maybe 30 readers over the last two years. About 80 percent had at least two morphologies. Only one had a single morphology, and that was rolling scars from a single bad cystic episode in her 20s. The “single treatment will fix everything” framing rarely matches the underlying inventory.
The contrarian part
The cosmetic dermatology industry sells fractional laser as the default acne scar treatment, partly because the technology is profitable and partly because the before-and-after photos look impressive in the marketing materials.
The honest read on fractional laser is that it produces modest improvement across most scar morphologies and significant improvement in shallow boxcars and surrounding skin texture. For ice-picks and deep rolling scars, it is often disappointing.
The cheaper interventions, TCA CROSS for ice-picks and subcision for rolling scars, are dramatically more cost-effective per unit of improvement and have decades of evidence behind them. They are also harder to monetise because they do not require expensive equipment. A TCA CROSS session might run 200 to 400 dollars. A subcision session might run 300 to 600 dollars. A fractional laser session is often 800 to 1,500 dollars per session, with 3 to 6 sessions recommended.
The cost-effectiveness math, by morphology:
Ice-pick scars: TCA CROSS at 6 sessions averaging 300 dollars equals 1,800 dollars. Improvement: 50 to 70 percent reduction in depth.
Rolling scars: Subcision plus filler at 3 sessions averaging 500 dollars equals 1,500 dollars. Improvement: 40 to 60 percent reduction in tethering.
Boxcar scars: Fractional laser at 4 sessions averaging 1,000 dollars equals 4,000 dollars. Improvement: 30 to 50 percent reduction in depth on shallow scars.
A patient with all three morphologies, treated correctly, might spend 6,000 to 8,000 dollars across a series of targeted interventions and see meaningful improvement across all three categories. The same patient treated with fractional laser alone might spend the same amount and see meaningful improvement in only one category.
The other contrarian observation is that home treatments are largely ineffective for atrophic scars. At-home microneedling at the depths that consumer derma rollers achieve (0.25 to 0.5 millimetres) does not reach the scar base and produces only superficial collagen remodelling. The 0.5 millimetre dermaroller is not the same intervention as the 1.5 to 2.5 millimetre professional microneedling that produces actual scar improvement, and selling them as equivalent has been one of the more frustrating consumer trends I have watched.
The honest framing on at-home microneedling: it might help skin texture and shallow rolling scars over many months of consistent use, but it will not meaningfully change ice-picks or deep boxcars. If the deep scars are what bother you, the at-home approach is not the right intervention.
What I would tell my past self
Identify the morphology before you book the consultation. The conversation with the dermatologist is dramatically more productive when you can say “I have ice-picks on the right cheek and rolling scars across the chin” rather than “I have acne scars.”
Ask the dermatologist which treatment they are recommending for which morphology. If they recommend the same treatment for all your scars, get a second opinion.
Budget for multi-treatment plans. The single-modality approach rarely works for patients with mixed morphologies. Plan for 2 to 4 treatment types over a year, not one treatment over six weeks.
Read the published outcomes for each morphology before you sign up. The Goodman and Baron scoring papers are accessible and walk through realistic expectations.
Accept that 100 percent improvement is not the goal in most cases. Realistic improvement across all morphologies is 40 to 60 percent, sometimes higher for individual scars. The remaining texture is what acceptance is for.
Frequently asked
Can I use retinol or tretinoin on atrophic acne scars?
The effect is modest. Tretinoin promotes some collagen remodelling and can improve the appearance of surrounding skin texture. The deep structural scars themselves do not respond meaningfully to topical retinoids alone in most series.
Is microneedling at home worth doing?
For surface texture and very shallow rolling scars, possibly. For ice-picks and deep boxcars, no. The depth needed for meaningful improvement is below what consumer devices safely reach.
Does collagen-stimulating filler (Sculptra, Radiesse) work for acne scars?
For diffuse atrophic textural change, yes, with limitations. For discrete deep scars, the filler does not reach the right tissue plane and the improvement is often disappointing. HA filler placed during subcision is a different application with different evidence.
What about laser resurfacing for ice-picks?
Fully ablative CO2 resurfacing can produce dramatic improvement on ice-picks but requires significant downtime (2 to 4 weeks) and has higher risks in higher Fitzpatrick types. It is a real option but should be considered alongside TCA CROSS, which often produces similar improvement with much less downtime.
Why do my scars look different in different mirrors?
Lighting. Overhead lighting deepens shadows. Front lighting flattens them. The scars you see in your magnifying bathroom mirror under harsh light are not what other people see in conversational lighting. Step away from the mirror.
Related Elelaf tools
Cystic acne severity | Acid picker | Microneedling home | Chemical peel home
Sources
- Jacob CI, Dover JS, Kaminer MS. Acne scarring: a classification system and review of treatment options. J Am Acad Dermatol 2001;45(1):109-117. PMID: 11176061.
- Goodman GJ, Baron JA. The management of post-acne scarring. Dermatol Surg 2007;33(10):1175-1188. PMID: 17173588.
- Lee JB, Chung WG, Kwahck H, Lee KH. Focal treatment of acne scars with trichloroacetic acid: chemical reconstruction of skin scars method. Dermatol Surg 2002;28(11):1017-1021. PMID: 12230882.
- Orentreich DS, Orentreich N. Subcutaneous incisionless (subcision) surgery for the correction of depressed scars and wrinkles. Dermatol Surg 1995;21(6):543-549. PMID: 7600028.
- Alster TS, Tanzi EL. Hypertrophic scars and keloids: etiology and management. J Am Acad Dermatol 2007;57(6):1077-1084. PMID: 17328993.