Chemotherapy extravasation: recognize and treat

A 52-year-old woman receives her second cycle of chemotherapy for breast cancer in a day hospital. Doxorubicin has been running through a peripheral catheter on the back of her hand for about twenty minutes when she says the site has started to burn. The nurse looks and sees the skin swollen and pale around the catheter. On testing, there is no blood return. The infusion pump did not alarm. The scene is fictional and was written for teaching.

This text uses the scenario as a setting for clinical reasoning in nursing. The trap is thinking the burning is "normal for chemo", slowing the rate and carrying on. With a vesicant drug, every milliliter that escapes the vein can cause necrosis that advances over weeks. The text does not replace your institution's protocol or the guidance of the pharmacy and oncology teams. If you play the case of the day on NEXO, you will recognize the challenge: deciding fast with few clues and a high cost for error.

Key points

• Extravasation is the escape of the drug from the vein into the surrounding tissue, and with vesicants it can cause necrosis.
• Burning, swelling, pallor and absence of blood return during the infusion call for an immediate stop.
• The flare reaction leaves marks along the vein, but without intense pain or swelling, and fades in 30 to 90 minutes.
• Local cold is the usual measure for anthracyclines; dry local heat accompanies hyaluronidase with vinca alkaloids.
• Dexrazoxane, for anthracyclines, should start as soon as possible and within six hours, in the opposite arm.
• Marking the borders, taking photographs and reassessing for days is part of the care, because the lesion can worsen late.

Why this is a reasoning scenario

Chemotherapy extravasation is rare, but it has disproportionate consequences. A review on anthracyclines estimates the incidence between 0.1 and 6 percent of administrations (Jordan et al., 2009), and a recent Japanese guideline cites a similar range, from 0.1 to 6.5 percent of patients receiving antineoplastics (Matsumoto et al., 2024). Because chemotherapy is widely prescribed, the total number of patients at risk is large, and it is almost always nursing staff who are at the bedside when the problem begins.

The challenge is that the first signs are subtle and look like benign reactions. The patient feels burning, but many people feel discomfort during the infusion. The swelling may be small. The pump does not alarm, because the pressure of a catheter outside the vein may not be high enough. The decision to stop depends on the eye and the suspicion of the person in the chair beside her.

What the scene teaches

The scene trains three habits. The first is knowing the drug class before starting, because the response changes depending on whether it is a vesicant, an irritant or a non-vesicant. The second is to stop first and investigate afterwards. The third is to follow up for days, since some lesions only show their severity weeks later. The post on clinical decision-making discusses how to act with incomplete information.

Vesicants, irritants and what happens in the tissue

Antineoplastics are classified by their potential for tissue injury. Vesicants can cause blisters and necrosis; irritants cause pain, inflammation and phlebitis, but rarely necrosis; and non-vesicants usually cause little reaction (Matsumoto et al., 2024). The Japanese guideline adopted the classification common to three international guidelines for 36 drugs and systematically reviewed another 31 in which there was divergence or new data (Matsumoto et al., 2024).

Among the vesicants, there is a division that is useful for reasoning. Those that bind to DNA, such as the anthracyclines (doxorubicin, epirubicin, daunorubicin, idarubicin), mitomycin and dactinomycin, enter tissue cells and are released again when they die, which keeps the lesion active. Those that do not bind to DNA, such as the vinca alkaloids and the taxanes, are metabolized more quickly in the tissue (Boschi and Rostagno, 2012). This difference explains why an anthracycline lesion can grow over weeks.

Why the back of the hand is a concern

On the back of the hand, the skin is thin and there is little tissue between it and the tendons. Extravasation there can reach structures that compromise hand function. The Japanese guideline weakly recommends considering a central venous device in patients about to start chemotherapy, citing as benefits completing the planned treatment with less pain and anxiety from punctures, and as a disadvantage the complications of the device itself (Matsumoto et al., 2024).

The same reasoning applies to other drugs

Extravasation is not limited to oncology. A meta-analysis of 19 studies and 6852 patients, most in intensive care, estimated the extravasation rate of vasopressors infused through a peripheral catheter at 1.43 percent (Wu et al., 2025). In a study with emergency department nurses in Turkey, 91 percent said they had received no training on extravasation after graduation, and 82 percent reported having no protocol at their workplace (Kuğu and Akyüz, 2025). Those working in units with a protocol had greater knowledge (Kuğu and Akyüz, 2025).

How the condition presents

In the patient in the scene, the condition has four clues: burning at the site, swelling, pallor around the catheter and absence of blood return during the infusion of a vesicant. No single finding on its own, but together they make extravasation the leading hypothesis. A Korean guideline describes early signs such as redness, itching and edema around the puncture, and notes that with a peripheral catheter there may be no symptom at the start (Kim et al., 2020).

Differential diagnosis

Not every burning sensation is extravasation. The table summarizes the hypotheses nursing staff need to separate at the chairside.

Hypothesis	Clues	What nursing checks
Vesicant extravasation	Burning, swelling, pallor or redness, no blood return	Blood return, affected area, drug class
Flare reaction	Marks or lines along the vein, sometimes with small blisters, without intense pain or swelling	Whether it fades within 30 to 90 minutes
Venous irritation	Pain or tightness along the vein, blood return preserved, no swelling	Vein path, irritant drug
Infiltration of a non-vesicant	Cool swelling, little pain	Type of solution and volume
Hypersensitivity	Itching, widespread hives, shortness of breath	Vital signs, skin away from the site

Table source: adapted from (Kim et al., 2020). The post on differential diagnosis with clues shows how to organize this list when time is short.

Tools that help detect it early

In a Japanese study with 257 patients receiving chemotherapy through a peripheral vein, thermography performed every 15 to 30 minutes during the infusion identified extravasation with a sensitivity of 84.6 percent and a specificity of 94.8 percent (Matsui et al., 2017). The technique is not available everywhere, but it reinforces a principle: inspect and palpate the site regularly during the infusion, not only at the start. The Japanese guideline weakly recommends checking blood return to confirm catheter position (Matsumoto et al., 2024).

Nursing management in the first minutes

At the first sign of extravasation, the Korean guideline describes the nursing sequence: stop the infusion immediately, disconnect the tubing from the catheter, aspirate the remaining drug through the catheter, administer the specific antidote and notify the physician (Kim et al., 2020). In the patient in the scene, the nurse stops the pump, leaves the catheter in place and performs aspiration before removing it.

Aspiration and catheter removal

Aspiration has limited evidence. The Japanese guideline made no recommendation on aspirating the residue, citing possible benefit and possible harm to the vessel (Matsumoto et al., 2024). In the Turkish study, the protocol item least known to the nurses was precisely aspirating the drug without exceeding 3 to 5 mL (Kuğu and Akyüz, 2025). The principle common to the sources is not to flush the line or inject anything else through the catheter, so as not to push the drug into the tissue. After that, the catheter is removed and the limb is raised.

Assess and document

• Mark the borders of the affected area with a pen, with date and time.
• Photograph the site, with a ruler, if the protocol and consent allow.
• Estimate the volume that may have leaked and the start time.
• Record pain, color, skin temperature, swelling and mobility.
• Assess perfusion and sensation in the fingers.
• Notify the medical team and the pharmacy about the drug involved.

The Korean guideline proposes monitoring pain, edema, skin temperature, mobility and the appearance of the lesion over time (Kim et al., 2020). A systematic review on the role of nursing in chemotherapy extravasation concludes that nursing plays a central role in prevention, management and treatment, and that up-to-date training improves competence and prevents more extravasations (Antúnez-Blancat et al., 2024).

Instruct the patient

The Korean guideline notes that the patient should be instructed, before the infusion, to report any change in sensation at the site, and that this instruction deserves special attention when the drug is a vesicant (Kim et al., 2020). The same guideline warns that symptoms often appear soon, but may arise days or weeks later, and that with a peripheral catheter there may be no symptom at the start (Kim et al., 2020). Discharge instructions are therefore part of the treatment. The patient needs to know what to watch for: increased redness, blisters, darkening of the skin, worsening pain, fever or difficulty moving the fingers. She also needs to know whom to call and how to keep the hand raised and protected.

A practical approach is to give the prescribed local measure in writing, with the times of application and the date of the next reassessment. When the service uses photographs, comparing them helps show whether the area has grown. If the patient has another cycle scheduled, the team needs to decide on the access before the day of the infusion, considering the affected limb and the possibility of a central device. Recording the event in the medical record and in the incident reporting system also helps the service review its processes.

Antidotes and measures by drug class

The specific approach depends on the drug. A pediatric review summarizes the antidotes and local measures by agent (Boschi and Rostagno, 2012), and the table below shows the main ones.

Extravasated drug	Local measure	Antidote or treatment cited
Anthracyclines	Local cold	Dexrazoxane or topical DMSO
Mitomycin, mitoxantrone	Local cold	Topical DMSO
Vinca alkaloids	Local dry heat	Hyaluronidase
Docetaxel	Local cold	Hyaluronidase
Paclitaxel	Local cold	No specific antidote
Oxaliplatin	Local heat	Sodium thiosulfate

Table source: (Boschi and Rostagno, 2012). Choices may vary between guidelines and institutions, and the decision on the antidote is made by the medical team together with the pharmacy.

Cold and heat

The Japanese guideline gives a weak recommendation for cold compresses as a local measure to reduce inflammation, pain and burning, and does not recommend heat as a stand-alone measure (Matsumoto et al., 2024). The Korean guideline describes cold or heat applications for 15 to 20 minutes, every 4 hours, for 24 to 48 hours, noting that the benefit has not been clearly demonstrated, although it is usual supportive care (Kim et al., 2020). For vinca alkaloids, the table by Boschi and Rostagno pairs local heat with hyaluronidase (Boschi and Rostagno, 2012). The Japanese guideline observes that heat alone appears harmful in extravasation of vinca alkaloids, while the combination of hyaluronidase with heat did not cause ulceration in the studies cited (Matsumoto et al., 2024).

Dexrazoxane for anthracyclines

Two prospective multicenter studies used intravenous dexrazoxane for three days: 1000 mg/m² on the first and second days and 500 mg/m² on the third, in the arm opposite the extravasation, starting as soon as possible and within at most six hours (Jordan et al., 2009). In the 54 patients analyzed, the treatment prevented necrosis requiring surgery in 53 (Jordan et al., 2009). The review by Jordan et al. cites maximum doses of 2000 mg on the first and second days and 1000 mg on the third, a halved dose when creatinine clearance is below 40 mL/min, a 1 to 2 hour infusion in a large vein of another limb, and starting the second and third days at the same time as the first, with a tolerance of 3 hours (Jordan et al., 2009).

For the patient in the scene, with a body surface area of 1.7 m², this would correspond to 1700 mg on the first and second days and 850 mg on the third, within the maximum limits. The calculation and the prescription are the responsibility of the medical team and the pharmacy. Two details fall to nursing. The first is that the cold compress must be removed at least 15 minutes before the dexrazoxane infusion, to allow blood flow in the area (Jordan et al., 2009). The second is that, in the studies, topical DMSO was not allowed together with dexrazoxane (Jordan et al., 2009). The Japanese guideline gives a weak recommendation for dexrazoxane, citing fewer surgeries and shorter recovery (Matsumoto et al., 2024).

Hyaluronidase for vinca alkaloids

Hyaluronidase breaks down hyaluronic acid, loosens bonds in the subcutaneous tissue and favors the diffusion and absorption of the extravasated drug (Boschi and Rostagno, 2012). The review by Boschi and Rostagno advises applying 1 mL of the hyaluronidase solution in five injections of 0.2 mL at the extravasation site, with a 25-gauge or finer needle, changing the needle with each injection (Boschi and Rostagno, 2012). The Korean guideline lists hyaluronidase at a concentration of 1500 U/mL, kept refrigerated, in the extravasation kit (Kim et al., 2020); with this presentation, 1 mL corresponds to 1500 U. The Japanese guideline notes that hyaluronidase is not approved for this use in Japan (Matsumoto et al., 2024), which shows that availability varies between countries.

Corticosteroid and surgery

The Japanese guideline does not recommend local corticosteroid injection and gives a weak recommendation for topical corticosteroid (Matsumoto et al., 2024). In a Japanese study on docetaxel extravasation in more than 46 thousand patients, only 34.3 percent of those who had the event received injectable or topical corticosteroid, and 11.4 percent were assessed by dermatology or plastic surgery (Yanagisawa et al., 2025). Surgical assessment is warranted when there is necrosis, persistent pain or a lesion that does not improve.

Step-by-step reasoning in the case

First, the suspicion: burning, swelling and pallor during the infusion of doxorubicin, a vesicant that binds to DNA.

Second, confirmation at the bedside: there is no blood return. The picture does not resemble flare, because there is swelling and pain.

Third, the stop: the nurse interrupts the infusion, disconnects the tubing and aspirates through the catheter without flushing the line.

Fourth, the documentation: removes the catheter, raises the hand, marks the borders, takes photographs and records the time and the estimated volume.

Fifth, the communication: notifies oncology and the pharmacy, reporting the drug, the time and the affected area.

Sixth, the local measures: applies a cold compress according to the protocol and removes it at least 15 minutes before dexrazoxane, if it is prescribed.

Seventh, the follow-up: instructs the patient to report any worsening and schedules reassessments on the following days.

Common mistakes

The first mistake is slowing the rate and continuing. With a vesicant, the suspicion calls for an immediate stop.

The second mistake is flushing the line to "test" the catheter. This pushes more drug into the tissue.

The third mistake is using heat with an anthracycline. The usual local measure is cold.

The fourth mistake is leaving the cold compress on during the dexrazoxane infusion. It must be removed at least 15 minutes before.

The fifth mistake is discharging the patient without a follow-up plan. A case report describes chemotherapy extravasation that progressed with soft tissue necrosis, with an appearance that resembled infection, and required debridement (Nguyen et al., 2024).

Prevention and training

In an Egyptian study, an in-service training program significantly improved nurses' knowledge and practice for reducing chemotherapy extravasation (Mohammed et al., 2023). An appraisal of guidelines with the AGREE II instrument found adequate quality in five guidelines on the topic, of which four had a high level (Bahrami et al., 2019). The practical step is for each service to have an accessible protocol and a kit ready for care. The post on diagnostic pause shows how to stop and review a picture before moving on.

How NEXO helps train this reasoning

To use the case of the day as training, try three habits. Before the first guess, ask which complication would be most serious if you were wrong. With each clue, ask whether it changes the urgency or only the cause. After the result, note which discreet clue you let slip by.

Three exercises with cases

Classify the drug. For each case with infusion, say whether the medication is a vesicant, an irritant or a non-vesicant.

Separate flare from extravasation. List the clues that tell one from the other.

Build the record. Write down what you would document in the first ten minutes.

The post about simulation of care shows how to train patient assessment, the post about interprofessional clinical reasoning discusses how nursing, pharmacy and medicine decide together, and the post about signs, symptoms and history shows how to read the patient's complaints.

The post about case-based learning in clinical practice shows how cases help fix these habits, and the post about diagnostic uncertainty discusses how to decide when the data do not yet add up.

The limits on using the game are described on the NEXO economy page. If you research the teaching of clinical reasoning, the page for researchers explains how to get in touch. Professionals who write or review cases find information on curation and on creators, and institutions can learn about the proposal on the partnerships page.

Frequently asked questions

How do you tell flare from extravasation?

Flare causes blotches or lines along the vein, without intense pain or swelling, and clears in 30 to 90 minutes (Kim et al., 2020). Extravasation usually involves pain, swelling and loss of venous return.

Should I aspirate through the catheter before removing it?

Many protocols advise aspirating the residue before removing the catheter (Kim et al., 2020), but the evidence is limited and the Japanese guideline made no recommendation (Matsumoto et al., 2024). Follow the local protocol and never flush the line.

Cold or heat?

Cold for most vesicants, including anthracyclines (Matsumoto et al., 2024). Local heat with hyaluronidase for vinca alkaloids (Boschi and Rostagno, 2012).

What is the dexrazoxane regimen?

1000 mg/m² on the first and second days and 500 mg/m² on the third, starting within six hours, in the opposite arm, with maximum doses of 2000, 2000 and 1000 mg (Jordan et al., 2009).

For how long should the lesion be monitored?

For days to weeks, because a vesicant lesion that binds to DNA can progress late (Boschi and Rostagno, 2012).

Final checklist

• I checked the drug class before starting the infusion.
• I confirmed blood return and inspected the site during the infusion.
• I stopped the infusion at the first suspicious sign.
• I did not flush the line and I aspirated according to the protocol.
• I marked the borders, took a photo and recorded time and volume.
• I notified oncology and pharmacy.
• I applied the right local measure for the drug.
• I removed the cold at least 15 minutes before dexrazoxane, if prescribed.
• I arranged follow-up and gave the patient guidance.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. Antúnez-Blancat A, Gago-Valiente FJ, García-Iglesias JJ, et al. The Role of Nursing in the Management of Chemotherapy Extravasation: A Systematic Review Regarding Public Health. Healthcare (Basel). 2024. PMID 39057599.
2. Bahrami M, Karimi T, Yadegarfar G, Norouzi A. Assessing the Quality of Existing Clinical Practice Guidelines for Chemotherapy Drug Extravasation by Appraisal of Guidelines for Research and Evaluation II. Iran J Nurs Midwifery Res. 2019. PMID 31772914.
3. Boschi R, Rostagno E. Extravasation of antineoplastic agents: prevention and treatments. Pediatr Rep. 2012. PMID 25396033.
4. Jordan K, Behlendorf T, Mueller F, Schmoll HJ. Anthracycline extravasation injuries: management with dexrazoxane. Ther Clin Risk Manag. 2009. PMID 19536310.
5. Kim JT, Park JY, Lee HJ, Cheon YJ. Guidelines for the management of extravasation. J Educ Eval Health Prof. 2020. PMID 32668826.
6. Kuğu E, Akyüz N. Emergency department nurses' knowledge and practices related to extravasation injuries of non-cytotoxic medications. Ulus Travma Acil Cerrahi Derg. 2025. PMID 40910712.
7. Matsui Y, Murayama R, Tanabe H, et al. Evaluation of the Predictive Validity of Thermography in Identifying Extravasation With Intravenous Chemotherapy Infusions. J Infus Nurs. 2017. PMID 29112585.
8. Matsumoto K, Ryushima Y, Sato J, et al. Extravasation associated with cancer drug therapy: multidisciplinary guideline of the Japanese Society of Cancer Nursing, Japanese Society of Medical Oncology, and Japanese Society of Pharmaceutical Oncology. ESMO Open. 2024. PMID 39389005.
9. Mohammed HS, Mohammad ZA, Azer SZ, Khallaf SM. Impact of In- Service Training Program on Nurses' Performance for Minimizing Chemotherapy Extravasation. Asian Pac J Cancer Prev. 2023. PMID 37898860.
10. Nguyen M, Borders L, Wesolow JT, Greene J. Chemotherapy Extravasation Causing Soft-Tissue Necrosis Mimicking Infection: A Longitudinal Case Study. Cureus. 2024. PMID 38434604.
11. Wu W, Yang X, Kou L. Extravasation, thrombosis, and infection with vasopressor infusion through peripheral intravenous catheters: a systematic review and meta-analysis. Cardiovasc Diagn Ther. 2025. PMID 40948710.
12. Yanagisawa R, Inose R, Muraki Y. Frequency, patient characteristics, and clinical management for extravasation with docetaxel: a descriptive study using a large Japanese medical claims database. J Pharm Health Care Sci. 2025. PMID 41291948.