What cold actually does to tissue

Start with what is measured rather than believed. Thermistor studies show that ice applied to the skin drops surface temperature quickly and intramuscular temperature slowly, with the depth of cooling depending heavily on subcutaneous adipose thickness. Twenty minutes of ice on a lean calf may drop tissue at one centimeter by several degrees; the same application over a thicker layer produces a fraction of that. Adipose is an excellent insulator, which means the dose of a standard ice protocol varies enormously between two athletes doing the same thing.

Blood flow is equally measurable. Using Doppler on the femoral artery, Gregson and colleagues showed cold water immersion at 8 degrees Celsius reduced limb blood flow substantially more than immersion at 22 degrees, confirming the dose-dependence of the vasoconstrictor response. Nerve conduction velocity studies show a roughly linear slowing with cooling, in the region of a couple of meters per second per degree Celsius of nerve temperature.

Where the literature gets messy is outcomes. For repeated same-day or next-day performance, cold water immersion has reasonable support: meta-analyses report small improvements in perceived recovery and in some performance measures within twenty-four hours, particularly after high-intensity intermittent work. For chronic adaptation, the direction reverses. Roberts and colleagues randomized cold water immersion against active recovery over twelve weeks of strength training and found the immersion group gained less muscle mass and strength, with blunted acute satellite cell and p70S6K signaling in a companion study.

For acute injury specifically, the evidence is thinner than the ubiquity of the practice suggests. Systematic reviews of cryotherapy for acute soft-tissue injury consistently conclude that the studies are small, heterogeneous, and rarely functional, and that the biochemical rationale is largely extrapolated. There is no human trial demonstrating that early icing shortens time to return to play after a strain or sprain.

Cold does exactly four things to tissue, and every claim about ice is one of them wearing a costume.

One: vasoconstriction. Cutaneous cooling triggers a sympathetic reflex releasing norepinephrine onto alpha-adrenergic receptors in vessel walls, and cooling additionally increases alpha-2c receptor trafficking to the smooth muscle membrane, so the same neural signal produces more constriction when cold. Reduced local perfusion means less plasma extravasation into the interstitium, which is a genuine reason cold limits how much swelling forms if applied early. It is not a reason cold removes swelling that already exists: drainage is lymphatic and lymphatics need muscular pumping, which cold and immobility both reduce.

Two: reduced metabolic rate. Enzyme kinetics obey a Q10 relationship, meaning reaction rates change by a factor of roughly two to three for every ten degrees Celsius. Cool tissue by four degrees and you have slowed its chemistry appreciably, including the oxygen demand of cells at the margin of an injury. The classic argument is that this limits secondary hypoxic injury. That argument is mechanistically sound and has never been convincingly demonstrated to change outcomes in human musculoskeletal injury, which is a useful example of a good mechanism failing to earn a clinical grade.

Three: neural slowing. Cooling reduces nerve conduction velocity and raises the activation threshold of nociceptors, and cold afferents themselves compete for dorsal horn processing, which is a gate-control style analgesia. This is why ice works for pain and works fast. It is also why ice before or during sport is a bad idea for anything requiring precision: cooled muscle produces lower peak force and lower rate of force development, and cooled joints have measurably worse position sense. Numbing a joint and then cutting on it is how a manageable problem becomes a season.

Four: it blunts the adaptive signal. This is the expensive one. The post-exercise anabolic response depends on the same events cold suppresses: local perfusion delivering amino acids, the inflammatory signaling that recruits satellite cells, and possibly the heat itself as a stimulus. In human biopsy work, cold water immersion after resistance exercise reduced phosphorylation of p70S6K and reduced satellite cell number in the following days compared with active recovery. If inflammation is the first half of repair, then cold is a brake pedal, and brake pedals are for when you want to slow down.

Worth flagging one contrast because it is the same logic in reverse. Heat, whether from sauna or local application, increases perfusion, raises tissue extensibility, and induces heat shock proteins, which is why it suits stiff chronic tissue and pre-training preparation rather than the acute post-injury window. The cold-versus-heat question is not a preference. It is a question about which direction you want blood flow and metabolic rate to move.