How to Ice Hands and Feet During Chemo

Cooling your hands and feet during chemotherapy infusions is a straightforward, non-drug strategy that can meaningfully reduce nerve damage and nail problems caused by certain chemo drugs. The idea is simple: cold narrows blood vessels in the fingers and toes, limiting how much toxic drug reaches those vulnerable nerve endings. A growing number of clinical trials support the approach, with one large randomized trial finding that hand cooling cut the rate of significant neuropathy roughly in half compared to the untreated side. But how you cool, for how long, and with what equipment all vary, and no single protocol has become the universal standard yet.

Why Cold Works During a Chemo Infusion

Chemotherapy drugs travel through the bloodstream and reach every part of the body, but the fingertips and toes are especially prone to damage. The nerves there are long, thin, and exposed, and certain drugs, particularly taxanes like paclitaxel and docetaxel, are known to injure them. The resulting condition, called chemotherapy-induced peripheral neuropathy (CIPN), can cause numbness, tingling, burning, and pain that sometimes lasts months or years after treatment ends.

Cooling the hands and feet during an infusion causes local vasoconstriction, meaning the small blood vessels narrow and less blood flows through. Less blood flow means less drug reaching the tissue. It also slows the metabolic activity of the nerve cells themselves, which may make them less susceptible to toxicity. The concept is the same one behind scalp cooling to prevent hair loss: reduce local drug exposure by chilling the area during the window when drug concentration in the blood is highest.

What the Evidence Actually Shows

The research base for limb cooling during chemo has grown considerably over the past decade, though results are not perfectly uniform. The largest and most cited recent trial, known as POLAR, randomized patients receiving taxane chemotherapy to wear cooling devices on one hand versus the other, so each person served as their own comparison. In the cooling arm, 29% of participants developed significant neuropathy compared to 50% on the control side.¹ That is a meaningful difference, especially considering that no approved drug reliably prevents CIPN.

A randomized trial focused specifically on foot cooling in advanced breast cancer patients found that women who cooled their feet during chemotherapy had significantly lower neuropathy scores at every assessment point, from six weeks through three months after completing treatment.² An integrative review examining multiple cooling studies reported that four out of six found a clear benefit, with evidence suggesting that colder temperatures conferred greater protection than milder cooling.³

Not every trial has produced a clear win, though. One randomized controlled trial testing cryotherapy to prevent paclitaxel neuropathy found no significant difference between the cooled and uncooled hands when comparing only its own arms. The researchers noted, however, that their control arm experienced unusually low neuropathy rates compared to similar trials. When the cryotherapy arm was compared against the pooled control groups from multiple trials, the cooled group did fare better.⁴ This kind of inconsistency is common in cryotherapy research, partly because cooling methods, temperatures, and durations differ so much across studies.

Practical Methods and Equipment

There is no single standardized protocol for limb cooling during chemotherapy, which is one of the biggest practical challenges.⁵ The approaches used in clinical trials and infusion centers generally fall into a few categories:

Frozen gloves and socks: These are commercially available or sometimes made in-house by hospitals. They are stored in a freezer and placed on the hands and feet before the infusion begins. They are inexpensive and widely accessible, but they lose their cooling capacity as the ice melts and offer limited temperature control.
Hilotherapy devices: These are medical-grade circulating-water cooling systems that deliver continuous, temperature-controlled cold through specialized mitts and boots. They maintain a consistent temperature throughout the infusion, which is a significant advantage over frozen gloves. One study comparing hilotherapy to frozen gloves found that hilotherapy resulted in significantly fewer side effects of moderate or higher severity, with about 44% of patients affected versus 61% with frozen gloves.⁶
Ice packs and instant cold packs: Some trials have used commercial instant-ice packs fixed over fingers and toes. One trial used packs applied for about 70 minutes during paclitaxel infusion, at temperatures between roughly negative 5°C and positive 5°C.⁷
In-house frozen gloves: At least one hospital in Thailand developed its own low-cost frozen gloves and showed that cold therapy using these cut the rate of neuropathy from 100% in the control group to about 49% in the treatment group.⁸

The lack of a single standard method means that what you encounter will depend heavily on your infusion center. Some centers provide hilotherapy as part of their standard offerings. Others keep frozen gloves in a freezer for patients to request. Many centers have no cooling equipment at all, in which case you may need to bring your own supplies or ask your oncology team about options.

Timing and Temperature

Getting the timing right is important. In most trials, cooling begins 15 to 30 minutes before the chemotherapy drug starts infusing and continues throughout the infusion, with some protocols extending cooling for 15 to 30 minutes after the infusion ends. The goal is to have the hands and feet already constricted and cold before the drug hits peak blood levels.

Temperature ranges vary across studies, typically falling between negative 5°C and about 10°C at the skin surface. Systematic reviews have noted that it is difficult to recommend a single ideal temperature or duration because the studies differ so much in methods and measurement.⁹ What does seem consistent is that colder is generally better for protection, but only up to the point that the patient can tolerate the cold. Pushing past someone’s comfort threshold leads to dropout, which defeats the purpose entirely.

If you are using frozen gloves that lose temperature over time, you will likely need to swap them out partway through the infusion. Some centers keep multiple pairs in the freezer so a fresh set is ready when the first warms up. With hilotherapy, the circulating water system maintains a steady temperature automatically, so this issue does not arise.

The Tolerability Problem

This is the elephant in the room for limb cooling: a meaningful number of people simply cannot tolerate it. Having your hands and feet encased in ice or near-freezing wraps for an hour or more is uncomfortable, and for some people it is genuinely painful. The dropout rates across trials reflect this.

In one paclitaxel cryotherapy trial, about a third of participants could not tolerate the cold, and only 38% completed all four planned cryotherapy sessions.¹⁰ A larger multicenter trial of frozen gloves reported that 34% of patients discontinued the intervention, mostly because of discomfort.¹¹ The Thai trial with in-house gloves reported four patients dropping out due to pain, though the overall discontinuation rate was lower.⁸

You can improve your chances of sticking with it by starting with the cooling before you feel anxious about it, bringing warm blankets for the rest of your body, and communicating honestly with your nurse about your pain level. Some patients tolerate hilotherapy better than frozen gloves because the temperature is more controlled and less aggressively cold. If you find the cold unbearable, pulling one hand out for a few minutes and then returning it is better than giving up entirely. Any cooling is likely better than none, even if you cannot maintain it the full time.

Nail Protection

Beyond neuropathy, taxane chemotherapy is notorious for causing nail changes: darkening, ridging, loosening, and sometimes complete nail loss. Cooling the fingertips and toes during infusion protects nails through the same vasoconstriction mechanism that protects nerves. Reducing blood flow to the nail bed limits how much drug reaches it.

A study using intrapatient comparison, where one hand was cooled and the other served as control, found a significant benefit of cooling on nail toxicity from paclitaxel across visits, though the effect was less clear with docetaxel.¹² A broader review of regional limb cooling concluded that it holds promise as a low-risk, non-drug intervention for preventing multiple chemotherapy-induced toxicities, including nail damage, though the evidence base still shows notable variability.¹³

For many patients, nail protection is actually the more immediately motivating reason to try cooling. Neuropathy develops gradually and can be hard to detect early, but nail damage is visible and distressing. If your oncology team offers cooling primarily for neuropathy prevention, it is worth knowing that you are likely getting nail protection as a bonus.

Oxaliplatin and Cold Sensitivity

If you are receiving oxaliplatin, the most common chemo drug used for colorectal and other digestive system cancers, the question of cold application gets complicated. Oxaliplatin is famous for causing acute cold sensitivity, where touching anything cold triggers intense tingling, pain, or muscle cramps in the hands and throat. Patients are routinely warned not to drink cold beverages or touch cold surfaces for days after infusion. So applying ice to your hands and feet might sound like the last thing you would want to do.

Yet a randomized trial specifically testing hilotherapy during oxaliplatin infusions found surprisingly strong results. The rate of patients free from significant neuropathy at 12 weeks was 100% in the hilotherapy group versus about 81% in the control group. At 24 weeks, the benefit persisted at 66% versus 49%. Patients in the cooling group also experienced dramatically less acute numbness, tingling, and cold sensitivity in their fingers and toes.¹⁴ This counterintuitive finding suggests that cooling during the infusion itself, when the drug is entering the blood, is a different situation than touching cold objects hours or days later once the drug has already distributed to nerve tissue. The mechanism of acute cold sensitivity with oxaliplatin involves a different pathway than the chronic neuropathy that cooling aims to prevent.

That said, this does not mean you should independently decide to ice your hands during an oxaliplatin infusion. This needs to be a conversation with your oncology team, because the tolerability dynamics are different and the evidence is still newer than for taxanes.

Compression as an Alternative

The POLAR trial tested both cooling and compression gloves head to head. Tight surgical gloves, which work by compressing the fingertip blood vessels rather than chilling them, also reduced neuropathy: 24% in the compression arm versus 38% in the control arm.¹ This is good news for people who cannot tolerate cold, because tight gloves achieve a somewhat similar effect through a different mechanism.

However, compression does not work for every drug. A trial of compression gloves for vincristine-induced neuropathy in lymphoma patients found absolutely no benefit. The rate of significant neuropathy was identical, at about 14%, in both the compressed and uncompressed hands.¹⁵ This likely reflects differences in how vincristine damages nerves compared to taxanes. The upshot: compression gloves are a viable alternative for taxane regimens specifically, but you should not assume they transfer to other drugs.

Cold Therapy for Children

Most of the cooling research has been done in adults with breast or colorectal cancer, but pediatric patients face neuropathy too, especially from vincristine, which is one of the most widely used drugs in childhood cancer protocols. A pilot trial of nurse-led cold therapy in children receiving vincristine found high acceptability among both children and parents, with full adherence to the intervention and no adverse events. The cooled group showed significant improvements in pain, tingling, motor function, and quality of life, though sensory neuropathy specifically did not reach statistical significance.¹⁶

The tolerability angle is especially relevant for children. Adults often describe the cold as the worst part of the experience, but this pilot found that children tolerated it well when nurses managed the process and parents were involved. It is a small, early study, so stronger evidence is still needed, but it opens the door for families to have the conversation with their pediatric oncology team.

What You Can Do on Your Own

If your infusion center does not offer any formal cooling equipment, you can bring your own supplies. The simplest option is commercial instant cold packs or gel packs stored in a cooler with ice. Wrap them in a thin cloth to avoid direct contact with skin, and apply them to your hands and feet about 15 minutes before the drug starts infusing. Keep replacing them as they warm. Some patients use ziplock bags filled with ice and water, which conform well to the shape of the hand.

The main limitation of do-it-yourself approaches is inconsistent temperature control. Researchers have flagged this as a real problem: ice packs are difficult to control, can become painful at their coldest, and lose cooling capacity as they melt.¹⁷ A frozen glove that starts at negative 15°C and warms to room temperature over 45 minutes is delivering very different cooling than a hilotherapy device held at a steady 10°C. Both “count” as cryotherapy, but the patient experience and the protective effect likely differ.

If going the DIY route, keep a few things in mind. First, tell your nurse what you are doing. They need to know in case of frostbite risk or if the cold interferes with IV access in the hands. Second, bring extra packs. You will probably need two or three rotations over the course of a typical infusion. Third, do not put bare ice directly on skin. A thin cloth or paper towel barrier prevents cold injury while still allowing enough cooling to constrict blood vessels. Fourth, cover the rest of your body with blankets. Being systemically cold and miserable makes the local hand and foot cold much harder to endure.

The Bigger Evidence Picture on Supportive Cryotherapy

Cooling hands and feet fits into a broader use of regional cold during chemotherapy. Scalp cooling to prevent hair loss is now well established and commercially available. Oral cryotherapy, where patients suck on ice chips during certain drug infusions, has shown benefit against mouth sores. A review of cryotherapy across all body sites found encouraging evidence that regional cooling reduces the burden of several chemotherapy-related side effects, including oral mucositis, hair loss, eye irritation, and nail loss, though most studies lacked blinded controls and used different clinical endpoints.³ The underlying principle is the same everywhere: cool the tissue you want to protect, and less drug reaches it.

One finding from that integrative review is worth highlighting for anyone considering limb cooling: a relatively greater degree of cooling compared to milder cooling appears to confer more benefit.³ In practical terms, this means that if you can tolerate colder temperatures without bailing out, you are likely getting more protection. But this has to be balanced against the dropout problem. The most protective cooling protocol in the world does nothing for you if you stop using it after the first session. Finding your personal sweet spot, cold enough to work, tolerable enough to sustain, is the real challenge, and it is worth experimenting over your first couple of infusion cycles to figure out what you can handle.

Conditions That May Complicate Cooling

Not everyone is a good candidate for hand and foot cooling. People with Raynaud’s phenomenon, a condition in which the fingers and toes already overreact to cold by turning white and painful, may experience severe vasospasm from cryotherapy. If you have been diagnosed with Raynaud’s or have a history of cold-triggered color changes in your fingers, discuss this with your oncologist before attempting cooling. Peripheral artery disease, cryoglobulinemia, and cold urticaria are other conditions where applying prolonged cold to the extremities could cause harm.

Diabetes with existing peripheral neuropathy is a gray area. You already have impaired nerve function in your feet, which means you may not reliably feel whether the cold is causing tissue damage. If you have diabetic neuropathy, your oncology team may want to monitor you more closely during cooling or adjust the temperature upward to reduce risk.

People receiving certain combination regimens should also check with their team. If the IV chemotherapy is delivered through a hand vein and you are cooling that same hand, the cold-induced vasoconstriction could theoretically slow the infusion or affect drug delivery. Most protocols call for cooling the opposite hand from the IV site, or switching to a port-accessed infusion, but this is worth confirming with your nurse before you start wrapping your hands in frozen gloves.