The short version

This page explains what fractional CO₂ laser does to a scar, and what the research behind it does and doesn't show. It's written for anyone with a scar they've been thinking about — from surgery, an injury, a burn or acne, whether it happened last year or decades ago.

If you only read this page:

A scar isn't damaged skin. It's repaired skin. Your body prioritised closing the wound quickly and strongly over matching what was there before. That trade-off is why a scar feels and looks different.

Fractional CO₂ laser doesn't remove a scar. It makes tiny columns of controlled injury through it, leaving healthy skin in between, and lets your body rebuild the tissue over the following months.

The changes most consistently reported in studies are in how thick the scar is, how flexible it is, its colour, and its surface texture.

Timing changes the odds, and that creates a genuine dilemma. For scars from surgery, the strongest evidence is for starting as early as possible once the wound has closed. The difficulty is that at two or three weeks nobody can tell you yet how your scar is going to settle — and most settle well on their own. Treating early means committing recovery time and money before you know whether you'd have needed to. It's a judgement call, and for plenty of people waiting is the right answer. For older surgical scars the evidence is weaker and mixed — which doesn't mean nothing can be done, but it does mean nobody should promise you a result.

No treatment erases a scar, and results vary and may differ for other patients. What's realistic for your scar is something we work out at consultation, looking at your actual skin — not from a photo or a general rule.

What a scar actually is

Skin is remarkably good at closing wounds. It is not good at rebuilding itself exactly as it was.

When skin is cut, burnt, or inflamed deeply enough, the body's priority is speed and strength — get the barrier closed before infection gets in.

It does that using collagen. Collagen is the material your skin is mostly built from: fine, strong fibres, not unlike the threads in a woven fabric. It's the same material tendons are made of. Collagen is what makes skin tough enough not to tear — and the way those fibres are arranged is what lets skin stretch and move.

In uninjured skin, the fibres lie in a loose tangle, crossing each other at every angle. Researchers call it a basketweave. Because the fibres point in all directions, skin can stretch whichever way you move, sit flat, and catch light evenly across its surface.

In a scar, the body lays collagen down in a hurry, and the fibres end up thicker and lying more in the same direction as each other — combed straight rather than tangled.

This has been measured, not just noticed by eye. One research group built a computer method for measuring which way fibres run in a tissue sample, and found that difference in all three scar types they tested — ordinary flat scars, raised scars, and keloid scars, the kind that spread beyond the edges of the original wound — compared with normal skin.¹ ²

That change in arrangement is most of why a scar behaves the way it does. Fibres lying the same way reflect light differently, which is the sheen. They don't flex the way a tangle does, which is the tightness.

The amount of collagen changes too, and that's largely what decides whether a scar sits raised above the surrounding skin or sunken below it.

Scars also lose some of what normal skin has: hair follicles, sweat glands, and the small ridges where the skin's layers interlock. That's part of why a scar can look and feel like a different material altogether.

Why some scars settle and others don't

Most scars quietly improve. The redness fades, the firmness eases, and over a year or two they stop being something you notice.

Some don't follow that path, and the reasons are reasonably well understood:

The repair process ran too long. Healing moves through overlapping phases — inflammation, then rebuilding, then remodelling. If the inflammatory phase drags on, through infection, tension on the wound, or the injury simply being deep, the body keeps laying down collagen past the point where it's needed.³

The wound was under tension. Skin that's being pulled while it heals produces a wider, more raised scar.

The scar sits in an area with less to work with. Some parts of the body scar more readily than others.

Individual healing tendency. Some people's skin produces more scar tissue than the injury warrants. At the far end of that spectrum is keloid scarring, where scar tissue grows beyond the boundaries of the original wound. This is something we ask about early, because it changes the approach.

Two broad types come up most often. Raised (hypertrophic) scars follow trauma, burns and surgical incisions, and involve too much collagen being deposited. Sunken (atrophic) scars more often follow inflammation in the skin — cystic acne, chickenpox — and involve a different problem: an imbalance between how much collagen and elastic fibre is built and how much is broken down, driven by the inflammation itself.³

What the laser actually does

The idea behind it - The technique is called fractional ablative treatment, and the principle was described in a landmark 2004 paper that changed how skin resurfacing is done.⁴

Before it, resurfacing meant treating the entire surface — effective, but with a long, difficult recovery and real risks of pigment change and scarring.³

The fractional approach does something less obvious. Instead of treating all of the skin, it treats a grid of microscopic columns — tiny, separate zones of controlled injury — and deliberately leaves the skin between them completely untouched.

That untouched skin is the point. It does two jobs: it closes over the treated columns quickly, and it carries the healing signal outward through the whole treated area. The result is a full healing response across the region with a fraction of the wound.

In laboratory studies of normal skin treated this way, the cells that rebuild the surface have already begun migrating across the treated columns within about 24 hours.³ That is a finding about cells under a microscope, not about how your skin looks or feels the next day — visible healing takes considerably longer, and your aftercare guide covers what to actually expect.

What the laser is doing at the tissue level

A CO₂ laser produces light at a wavelength of 10,600 nanometres. At that wavelength, water absorbs the light very strongly.

Skin is mostly water. So when the beam hits, the energy is absorbed almost immediately by water in the collagen fibres, blood vessels and cells — and that tissue vaporises on the spot, leaving a narrow column. Around the edge of each column is a thin rim of heated tissue.³

Two things follow from this:

• The light itself doesn't travel far. It's absorbed at the surface.
• The effect does reach deep. The column extends down into the deeper layer of the skin, which is where scar tissue lives.

That's why calling this a "surface" treatment is misleading. The absorption is shallow; the tissue effect is not.

The depth and spacing of those columns are set by the clinician — how much energy goes into each column, how long the pulse lasts, and how densely the columns are placed.³

That control matters for scars specifically. A thin, flat scar and a thick, raised one need the treatment to reach very different depths. Skin on the face tolerates settings that skin on the chest or the shoulder doesn't. And in deeper skin tones the settings are deliberately eased back to reduce the heat load, for reasons we come to further down. It also means the plan can change as you go: what your skin does after the first session informs the next one. None of this runs from a standard preset, and a conservative first treatment is the safest starting point.

What happens afterwards — the part that does the work

The vaporised tissue isn't the treatment. The rebuilding that follows is. It runs for months, in three overlapping stages.³

Days 1–7: clearing out.

Immune cells arrive at the treated columns and clear away damaged tissue. At the same time, the body switches on a family of enzymes called MMPs — think of them as demolition tools — which break down the old, densely packed collagen of the scar. The two that do most of the early collagen clearing run high through the first week and then drop off sharply; others in the family work to their own timetable, some peaking later. The columns of debris work their way up and out through the surface.

The early surge in collagen-clearing enzymes has been measured in treated human burn scars. The rest of the detail in this stage comes from ordinary skin and laboratory models.

Weeks 1–5: rebuilding.

New surface skin forms. Small blood vessels grow back into the area. Then the cells that make collagen — fibroblasts — move into the treated zones and start producing new collagen. Interestingly, collagen production is suppressed for the first day or two before it rises. Researchers think this early pause clears space for the new material to be laid down in an orderly way rather than simply piling on top of what's there.

That early dip has been measured in treated hypertrophic scars as well as in ordinary skin. The rebuilding timeline itself comes mostly from ordinary skin.

Month 1 onwards: remodelling.

This is the slow part, and it runs for at least three months. Three things change, and unlike much of the detail above these have been measured in treated human scars rather than inferred from ordinary skin: the thick, bunched collagen bundles are replaced by finer ones; the mix of collagen types shifts back toward the proportions found in normal skin; and the elastic fibre network improves.

What doesn't happen is a return to the basketweave. The research on treated scars describes the new collagen as finer and more evenly laid down — not as reverting to the random tangle of uninjured skin. It is the structural reason a treated scar can become softer, flatter and closer in colour without ever becoming indistinguishable from the skin around it. The building material improves. The underlying architecture is still scar.

Treated scars tend to end up more flexible, not just flatter. If it's the tightness that bothers you rather than the look of it, that distinction is the one that matters — and there may be a reason for it.

While most of those demolition enzymes are switched up, the one that specialises in breaking down elastin is switched down, and stays down. Researchers have proposed that this is the skin selectively protecting its elastic network while it clears out old collagen, which would account for the flexibility. The mechanism is a proposal rather than a settled fact: the enzyme evidence comes from laboratory skin models and sun-damaged skin, not from scars. What has been measured in scar tissue is the outcome rather than the cause — treated scars do show a better elastic network afterwards. So the effect has been observed; why it happens is still being worked out.³

This is also why results aren't immediate. At six weeks you're looking at a work in progress. The meaningful assessment point is several months out.

What the research shows

Four findings matter most. The detail behind each, including the limitations of every study, is in the reference list.

Surgical scars: it works, and it works best early

Fourteen controlled trials covering 492 participants or scars have been pooled together.⁵ Across all of them, fractional CO₂ improved surgical scar scores. The effect held up statistically, though it varied from trial to trial.

The gains were largest when treatment began after the wound had closed. In most of these trials that meant within the first month, usually around the time the sutures came out or in the fortnight afterwards. A single session in that window produced a measurable improvement, and most trials used two or three.

For scars treated more than three months on, the pooled result was not statistically significant. That is a narrower statement than it sounds: those trials were unable to show a benefit, not able to show there wasn't one. Only a few of them treated older scars, all were small, and their results were still consistent with a modest improvement. One treated caesarean scars more than a year old and did find benefit, though it had only 11 participants.

The usual limitations apply. The trials measured their starting point differently, several were not properly randomised, and studies that found nothing may never have been published.

The awkward part. The best-evidenced window closes before you can tell whether you need it. At two or three weeks a healing incision gives almost nothing away, no clinician can look at it and tell you how it will end up, and most surgical scars settle perfectly well with no help at all. The only thing carrying real weight that early is history: your own previous scars, and your family's. So treating early can mean spending on a scar that was never going to need it, while waiting means the window has gone by the time you can see whether you want it. There is no clean answer. Plenty of people wait and see, and that is a reasonable choice.

Acne scarring: better results than the main alternative

Eight randomised trials, 249 patients, comparing fractional CO₂ against microneedling radiofrequency.⁶ CO₂ came out ahead on both things they measured: how much the scars improved, and how satisfied patients were afterwards. Randomised trials are the strongest single study design, so this is good evidence by the standards of the field.

The trade-off is recovery. CO₂ carried more than four times the risk of the treated skin darkening afterwards, more discomfort during the session, and a longer stretch of redness. So the choice depends on your skin type and how much downtime you can take.

New scars: a strong early signal

A randomised trial of 60 people with scars two to four weeks old compared CO₂ laser against hyaluronidase injection.⁷ Scar volume fell by around 45% in the laser group against around 33% with the injection, and tissue samples afterwards showed better collagen arrangement and stronger elastic fibre networks. The caveats are real: one centre, sixty people, and anyone prone to raised scarring was excluded. It settles nothing on its own. What it adds is consistency, because it points the same way as the larger analyses above.

Skin tone: what we plan around

Darkening of the treated skin afterwards, called post-inflammatory hyperpigmentation, is the most common complication, and the risk is substantially higher in deeper skin tones. It usually fades, though it can take months, and it has been reported to persist for years in some cases. In the deepest skin types it has been reported in up to 100% of patients in some series — a figure quoted inside a 2025 review rather than one we have checked at its original source, so read it as how high the risk can run rather than an exact rate.⁸

This is managed rather than avoided. Preparing the skin beforehand, easing the laser settings back, and strict sun protection either side of treatment all reduce it.

What this treatment doesn't do

It doesn't erase a scar. No treatment does. The realistic goal is a scar that behaves more like the skin around it.

It doesn't work equally on every scar type. Deep, narrow "ice pick" acne scars are generally understood to respond less well to any resurfacing treatment than broader, shallower ones. That is widely held clinical experience rather than a finding from a specific trial.

It isn't the right choice for everyone. A tendency toward keloid scarring, a very recent scar, active infection in the area, and pregnancy are all reasons we'd wait, adjust the plan, or take a different route.

It isn't a single session. In the surgical scar trials pooled above, most protocols used two to three sessions or more. The number that would suit your scar is worked out at consultation and confirmed before you commit to anything.

How to weigh any of this

Most of us were never taught how to read medical research. Here is the short version. It applies well beyond scars.

Not all studies carry the same weight. One person's result tells you almost nothing. A single small study tells you a little. A randomised trial tells you more, because deciding by chance who gets which treatment removes the thing you can't see from the outside: clinicians steering the more promising patients toward the treatment they already believe in. Pooling several randomised trials tells you most of all. The treatment findings in this document come from those top two rungs. The biology comes from laboratory work, which is a different kind of evidence again, and we've said so wherever it's used.

Small studies mislead in both directions. They can find effects that aren't there, and miss effects that are. So when you read "no significant difference," it often means the study wasn't big enough to tell — not that there was nothing to find. That distinction does real work in the section on older surgical scars above.

Who paid matters. A study funded by the company selling the product isn't worthless, but it has to be read differently. There is one of those in this document, and we've named it in the reference list.

And this particular field has four soft spots.

Most of the biological detail comes from the wrong tissue — animal models, lab-grown skin, and biopsies of normal or sun-damaged human skin. Research looking directly inside human scar tissue is, in the words of a 2026 review, critically scarce.³

Laser settings differ so much between studies that comparing them properly is difficult.³

Remodelling runs for months, but most studies measure it once.³

And there is very little Australian or New Zealand data. The trials above ran in China, Korea, Egypt, Europe, Brazil and the United States. The one Australian study was partly funded by the manufacturer of the product it tested. Wound healing doesn't work differently here, but the gap is real and we won't paper over it.

So what is all of it worth? Not a prediction. No study can tell you what your scar will do. What this evidence does is narrow the field: which treatments have earned their place, roughly when they work best, what they cost you in recovery, and where nobody actually knows yet. That is enough to make a decision with. It is more than guessing, and a good deal less than certainty.

The honest summary

Fractional CO₂ laser has a well-described biological mechanism and a reasonable body of clinical evidence behind it, most consistently for improving scar thickness, flexibility, colour and texture.

The evidence is strongest for treating surgical scars early, once the wound itself has closed. It is good for acne scarring, with a clear pigmentation trade-off that has to be weighed against your skin type. For established surgical scars the evidence is thinner rather than negative: the studies that looked at them were too few and too small to settle the question. That is a reason for a proper assessment, not a reason to assume nothing can be done.

Results vary and may differ for other patients. Whether treatment suits you, what it could realistically achieve, and how many sessions that would take are things we assess and discuss at consultation.

Research and references can be found here on our Clinical Evidence Research Library.