Why Cold Picks Fat: Cryolipolysis Selectivity, Crystallisation and PAH

Cold doesn't melt fat. It picks it out. That one distinction carries most of the weight when you're comparing a cooling platform against a cavitation head, because selective is doing real work in the sentence — the fat layer takes the injury, the skin sitting on top of it walks away.

Below is the mechanism file behind that claim. Why adipocytes lose the cold contest. Why the gap between the appointment and the visible change is measured in weeks, not days. And what has to appear on your consent form about paradoxical adipose hyperplasia before anyone signs. Platform numbers stay where they belong, on the CR-01 specification page. This is technical reading for buyers and trained operators, not medical advice.

Why fat loses the cold contest

The clue came from paediatrics, not aesthetics. Epstein and Oren described popsicle panniculitis in the New England Journal of Medicine in 1970: an infant's cheek, a frozen sweet held against it, fat necrosis underneath skin that looked untouched. Nobody was contouring anything. It simply showed that fat could be injured selectively through an intact surface.

Turning that observation into a method took another thirty-odd years. Manstein, Laubach, Watanabe, Farinelli, Zurakowski and Anderson published selective cryolysis in Lasers in Surgery and Medicine in 2008, demonstrating fat loss through skin that survived the exposure. Our own technical documentation compresses the chain into one line: subcutaneous fat cells are more vulnerable at low temperature, cooling causes crystallisation that triggers natural cell death, and the debris clears through ordinary metabolism without disturbing liver enzymes or blood lipids.

Mechanistically the working account runs like this. Triglycerides packed inside an adipocyte begin to solidify while the mostly aqueous cytoplasm of a keratinocyte or fibroblast is nowhere near freezing. Crystal formation stresses the cell from within, inflammation follows, and the cell dies over days rather than on the treatment bed. Water-rich neighbours sit the whole event out. That's the trick, and it's why the useful treatment window is narrow enough to print on a control panel.

The setpoint is a plate temperature, not a tissue temperature

Operators get this wrong constantly. The number on screen is what the applicator is driving its plates toward — not what the fat inside the fold is experiencing. Our device manuals put the range at 0 to 15 °C in 1 °C steps, with 14 °C the warmest step that still actively cools; pick 15 °C and the cooling stage stands down to ambient, which is the parking position between clients. An archived protocol sheet in the same documentation runs a routine abdominal cycle at 1 to 4 °C for 30 minutes, suction power III–IV, IR IV–V.

Between plate and fat sit skin, a membrane, a petrolatum layer, and whatever coupling the vacuum managed to achieve. Suction is not a comfort setting. It pulls tissue against the cooling plates and holds it there, and a slack fold means the screen reading and the tissue reading have quietly parted company. Watch the first ten minutes of any cycle. A bad seal announces itself early or not at all.

Why the antifreeze membrane isn't optional

Skin survives because the exposure is controlled. Remove the control and you have frostbite, not contouring. Our manuals are blunt about the pre-treatment sequence: cleanse client and handle, apply petrolatum to the treatment area, then an antifreeze membrane, every single session. The same documentation fixes suction frequency at level I — continuous hold — throughout cryo work, leaving the pulsed levels for other handpiece functions. Two consumables and one menu setting. Skip any of them and you've moved from a disappointing result to a genuine injury.

From crystallisation to clearance: why the calendar runs in weeks

Cell death here isn't an event you can show the client on the bed. It's a process their immune system carries out over the following months — commercially awkward, which is why some suppliers blur it.

Zelickson and colleagues published the Yucatan pig work in Dermatologic Surgery in 2009, treating 22 sites and reporting significant reduction of the superficial fat layer with no damage to overlying skin and no deleterious change in blood lipids. Klein's group ran the human version in Lasers in Surgery and Medicine the same year and found no meaningful change in any blood lipid or liver test across 12 weeks of follow-up. Macrophages arrive, clear the damaged cells, the lymphatics carry the contents away. That takes as long as it takes.

Our archived treatment notes describe the clinical shape of the same curve: visible change from roughly three weeks, more of it by the two-month mark, continued clearance for four to six months afterwards, and a repeat cycle on the same site considered around two months out. Different evidence, same slope.

What that does to your booking calendar

• The follow-up photo is a scheduled appointment, not a favour. Book it at the point of treatment, twelve weeks out, same lighting and same marks. Nobody recalls their own abdomen accurately two months later.
• Your review call sits after the process, not inside it. Assess someone at three weeks and you're grading a job still in progress, which sours a conversation for no reason.
• One placement, one area, then wait. The machine runs the cycle unattended once you've verified the seal, so operator minutes per session are low while the revenue calendar per client is long. Staff and price around that gap.

Building a room list around this cadence alongside faster-turnaround devices? Our note on planning equipment from the treatment menu covers how the two schedules coexist.

PAH: the complication your consent form has to name

Paradoxical adipose hyperplasia is the treated area getting bigger instead of smaller. Firm, usually painless, well demarcated, often following the shape of the applicator. Onset is delayed by several weeks to months. It does not resolve on its own — the StatPearls review of PAH states plainly that no case of spontaneous resolution has been documented — and correction means liposuction or excision, sometimes more than once.

The incidence figure has moved a long way, and you should quote the movement rather than a single decimal. Jalian, Avram, Garibyan, Mihm and Anderson put it at roughly 0.0051%, about one in twenty thousand treatments, in JAMA Dermatology in 2014, working from a manufacturer adverse-event count. The StatPearls PAH review traces a manufacturer revision to 0.033% by 2021, a 2020 systematic review by Hedayati and colleagues landing near 0.91%, and current estimates running as high as 2%. Then in 2025 Mah and colleagues pooled 28 studies covering 13,078 patients in Aesthetic Surgery Journal Open Forum and found 29 cases — 0.22% (95% CI 0.10–0.47), about one in 455 patients. Only four of those 28 studies followed patients for 16 weeks or longer, which is shorter than the window in which PAH tends to declare itself.

So: rare, plausibly under-reported, and estimates spanning two orders of magnitude. Anyone handing you one confident number is selling.

The mechanism is still a list of hypotheses

Nobody has demonstrated the cause. The review literature offers candidates — hypoxic injury driving adipocyte hypertrophy and angiogenesis, sympathetic denervation removing the brake on lipolysis, preadipocyte recruitment through inflammatory cytokines, altered receptor expression, and mechanical stimulation of adipocyte growth by the negative-pressure suction itself. Sit with that last one for a moment, because suction is a parameter you set.

Train staff to say "the mechanism isn't established" instead of improvising. Confident nonsense about a complication is how a clinical problem turns into a legal one.

The risk pattern that keeps reappearing

Descriptive, not proven. Still, the same features recur. StatPearls lists Hispanic ethnicity, male sex, use of a single large applicator and abdominal treatment location among the reported risk factors, alongside possible hormonal and genetic contributions. Kelly, Rodriguez-Feliz and Kelly's 510-patient report in Plastic and Reconstructive Surgery in 2016 is one of the papers that put that pattern on record. Mah's 2025 pooling found sex-based risk differences that weren't statistically significant, so treat the pattern as a flag for your treatment log, not a rule.

Applicator choice is where this touches your daily practice. The box ships two cups at 165 × 81 mm and two at 128 × 57 mm. The small handle exists for a reason, and defaulting to the biggest cup in the case because it covers more ground is a habit worth breaking.

Writing consent and records that hold up

Most consent forms in this category are decorative. Fix yours. Six items belong in it, and none of them costs you a booking if the conversation is handled properly.

1. Name the complication in full. "Paradoxical adipose hyperplasia", described plainly as enlargement of the treated area instead of reduction.
2. State that it does not self-resolve. Correction requires surgical intervention and referral. Say it before, not after.
3. Give a range and name the uncertainty. Published estimates disagree by orders of magnitude; your form should reflect that rather than pick the flattering end.
4. Separate expected reactions from rare ones. Redness, tingling, bruising and transient numbness are the ordinary aftermath our documentation describes. PAH sits in a different category and shouldn't be buried in the same sentence.
5. Put the review dates in writing. A twelve-week photographic review plus a further check around six months, both booked at the moment of consent.
6. Rule out cold-related conditions explicitly. Cryoglobulinaemia, cold urticaria and paroxysmal cold haemoglobinuria are absolute exclusions, as is a hernia at the treatment site. Screening questions, not fine print.

The treatment log clinics skip

Record per cycle: applicator size and serial, setpoint, dwell time, suction level, IR level, anatomical site, operator name. Our documentation puts infrared output at 5 to 10 W across the 0–8 scale in 1 W steps, so "IR 4" means something specific and repeatable. Write the level down.

Why bother? Because when a client comes back at month five with a firm, enlarging area, the only useful questions are which applicator, which setpoint, how long, and where. No log and you're guessing, your distributor is guessing, and any conversation with the manufacturer stalls at question one.

Where this leaves your shortlist

Cold buys you fat-cell reduction that persists, at the price of a slow calendar and one uncommon complication you're obliged to disclose. Ultrasonic cavitation trades differently — broader field, faster visible circumference change, course-based revenue — which is why the SM-01 cavitation platform usually sits beside a cooling unit rather than beneath it. The full comparison across cold, ultrasound and RF lives on our body contouring solutions page.

Want the parameter tables, the applicator drawings, or the consent checklist above as a document before you commit? Ask us. Clinical decisions stay with the treating practitioner, where they belong.

Frequently asked questions

Does a colder setpoint give a better result?

No, and chasing the bottom of the range is how someone gets hurt. The mechanism depends on holding tissue inside a window long enough for lipid crystallisation to proceed, not on reaching the lowest number the panel allows. Our archived protocol sheet starts an abdominal cycle at 1 to 4 °C for 30 minutes, and that's a baseline to adjust for fold thickness. Dwell time and contact quality move the outcome further than shaving degrees off the setpoint.

Why does skin survive when fat doesn't?

Different composition, different response. Adipocytes are packed with lipid that starts solidifying at temperatures the water-rich dermis, vessels and nerves above them tolerate. Epstein and Oren's 1970 report of cold-induced fat necrosis under intact cheek skin was the first published hint; Manstein's group formalised it as selective cryolysis in 2008. The controlled version still depends on petrolatum and an antifreeze membrane — remove those and selectivity goes with them.

How should staff answer a client who asks about PAH?

Straight. It's a known, uncommon complication where the treated area enlarges rather than shrinks, it appears weeks to months later, it doesn't go away by itself, and fixing it means surgery. Published incidence estimates disagree widely, with the 2025 pooled figure sitting around one in 455 patients and older manufacturer counts far lower. Don't improvise a mechanism, because the mechanism isn't settled. Refer anything that grows.

Can we shorten the wait by retreating the same area sooner?

Not usefully. Clearance is still running, and retreating early stacks a second injury onto tissue that's already working through the first. Roughly two months between cycles on the same site is the spacing our documentation supports, and it also gives you a clean photographic comparison instead of a muddled one.

Which applicator suits a small flank?

The 128 × 57 mm cup. A 165 × 81 mm handle draws more tissue and belongs on a broad abdominal roll; on a modest fold it produces uneven contact and awkward edges. Reports describing PAH have repeatedly involved abdominal treatment with a single large applicator — not proof of causation, but another argument for matching the handle to the anatomy.

References

1. Manstein D, Laubach H, Watanabe K, Farinelli W, Zurakowski D, Anderson RR. Selective cryolysis: a novel method of non-invasive fat removal. Lasers Surg Med. 2008;40(9):595-604.
2. Epstein EH Jr, Oren ME. Popsicle panniculitis. N Engl J Med. 1970;282(17):966-7.
3. Zelickson B, Egbert BM, Preciado J, Allison J, Springer K, Rhoades RW, Manstein D. Cryolipolysis for noninvasive fat cell destruction: initial results from a pig model. Dermatol Surg. 2009;35(10):1462-70.
4. Klein KB, Zelickson B, Riopelle JG, Okamoto E, Bachelor EP, Harry RS, Preciado JA. Non-invasive cryolipolysis for subcutaneous fat reduction does not affect serum lipid levels or liver function tests. Lasers Surg Med. 2009;41(10):785-90.
5. Jalian HR, Avram MM, Garibyan L, Mihm MC, Anderson RR. Paradoxical adipose hyperplasia after cryolipolysis. JAMA Dermatol. 2014;150(3):317-9.
6. Mah AE, Razeghi P, Li C, Datta S, Tao BK, Ahmad J, Austin RE. Incidence of paradoxical adipose hyperplasia after cryolipolysis: a systematic review and meta-analysis. Aesthet Surg J Open Forum. 2025;7:ojaf142.
7. Paradoxical Adipose Hyperplasia. StatPearls, NCBI Bookshelf - reported incidence range, proposed mechanisms, risk factors and management.
8. Kelly E, Rodriguez-Feliz J, Kelly ME. Paradoxical adipose hyperplasia after cryolipolysis: a report on incidence and common factors identified in 510 patients. Plast Reconstr Surg. 2016;137(3):639e-640e.