A Retrospective Study of Chemotherapy-Related Extravasation Events in Dogs and Cats

1 Introduction

Extravasation is defined as the inadvertent leakage of a drug from a vein into surrounding tissue, and the degree of tissue injury is determined by the drug's classification as an irritant, vesicant, or non-vesicant [1, 2]. Irritants have the potential to cause an inflammatory reaction but generally not tissue damage, whereas vesicants can cause tissue injury and necrosis [1]. Examples of irritant chemotherapies include dacarbazine and carboplatin [3]. Examples of vesicant chemotherapies include doxorubicin, vinca alkaloids, rabacfosadine, and actinomycin-D [3, 4]. Mitoxantrone is typically categorized as an irritant, however some reports classify it as a vesicant based on concentration and volume [1, 5-7]. The severity of tissue damage secondary to extravasation might be related to the drug's DNA binding properties, as this can make it more difficult to remove the drug from the tissues [1, 2]. DNA binding chemotherapy drugs include doxorubicin and mitoxantrone [1, 2]. Vinca alkaloids are classified as non-DNA binding drugs [1, 2]. There are several known risk factors for extravasation injury reported in people, including, but not limited to, the presence of small veins, recent venipuncture in the same vein, use of large-gauge catheters or butterfly catheters, and age (very young or elderly) [3, 5-7]. Additionally, drug-related extravasation risk factors can include the concentration of the drug, the volume extravasated, and the location of extravasation [5-7]. In people, common clinical signs of irritant injury include swelling, pain, discomfort, and erythema, while signs of vesicant injury can also include ulceration and necrosis [2].

There is a paucity of information in the veterinary literature on the treatment and outcome of chemotherapy extravasation injury. Publications evaluating extravasation injury or treatment response consist of single case reports or small case series [8-12]. A study published by Miller et al. described suspected carboplatin extravasation in seven dogs [8]. Of these dogs, six developed full-thickness necrosis around the vessel, and three had swelling and edema in the same region [8]. The wounds completely healed in six of the seven dogs, a median of 25.5 days after observation of extravasation injury [8]. Venable et al. published a case series describing the outcomes of four dogs with known or suspected doxorubicin extravasation receiving the neutralizing agent dexrazoxane [9]. In three dogs, complications were noted during the infusion of doxorubicin, and dexrazoxane was administered within 2 h, based on recommendations from physician-based medicine [9]. In the fourth dog, dexrazoxane was not administered until 48 h after infusion, when clinical signs of a suspected extravasation injury were noted [9]. This latter case was the only dog that developed extensive limb necrosis requiring amputation [9].

The aim of this retrospective, multicenter, descriptive study was to evaluate chemotherapy extravasation events (EEs) in dogs and cats, subsequent adverse events (AEs), and overall outcomes. To our knowledge, this is the largest descriptive study of chemotherapy extravasation injuries in veterinary medicine.

2 Materials and Methods

Records were gathered from contributing institutions through direct contact and contact through the American College of Veterinary Internal Medicine, Oncology listserv. Contributors either completed a data sheet electronically or provided a copy of the medical records for review by the primary and corresponding author. Canine and feline cases were included if they had a suspected EE during the administration of chemotherapy.

Medical records were reviewed retrospectively. Information related to signalment, tumor type, EE, treatment of the extravasation injury, and outcome was obtained from the records. Table 1 provides details of the information obtained. The severity of the lesion resulting from the EE was graded retrospectively according to the Veterinary Cooperative Oncology Group—common terminology criteria for AEs (VCOG-CTCAE v2) following investigational therapy in dogs and cats [13]. Table 2 depicts the grading scores and associated clinical changes for chemotherapy extravasation injury as described elsewhere [13]. VCOG-CTCAE v2 does not describe a Grade 0 AE. Grade 0 was used in our study to describe a case in which no clinical signs of extravasation injury were reported after known extravasation.

3 Results

4 Discussion

This is a descriptive study of chemotherapy extravasation injuries in dogs and cats. Doxorubicin was the most common extravasated drug in this study, likely a result of reporting bias given the known severity of extravasation injury with this drug. Carboplatin was the second most common extravasated drug in this study. We suspect that extravasation of other drugs that result in less severe or absent clinical signs was underreported.

Doxorubicin EE resulted in Grade 0 (n = 2/9), Grade II (n = 4/9), and one each of Grades III–V AEs. The doxorubicin was diluted in all but one case (Case 9, Grade II AE). The cases with the most severe AEs (Case 11, Grade IV and Case 7, Grade V) were estimated to have 100% of the drug volume extravasated, while cases reported to have 25% or less drug extravasated experienced only mild–moderate AEs (Grades II–III), suggesting that the volume of drug extravasated might play an important role in the severity of doxorubicin extravasation injuries. To our knowledge, this has not been previously evaluated as a risk factor for extravasation injury in veterinary medicine. Studies evaluating a larger population are needed to explore this further. All but one doxorubicin EE case (n = 8/9) received the neutralizing agent dexrazoxane. Cases receiving dexrazoxane experienced AEs ranging from mild to severe (Grades 0–V). The one case in which dexrazoxane was not used (Case 12) developed Grade III AEs. Given the variability of the dose, timing, and frequency of dexrazoxane in this study, strong conclusions on the overall benefit of dexrazoxane and the impact of dose, timing, and frequency of this neutralizing agent in dogs and cats experiencing doxorubicin EE cannot be made. Dexrazoxane is FDA-approved for the treatment of anthracycline extravasation in humans [14, 15]. At the time of this writing, one case series and a few case reports have been published on the use of dexrazoxane for the treatment of doxorubicin extravasation in dogs or cats [9-11]. Venable et al. described three of four dogs with doxorubicin EE that received dexrazoxane between 1 and 3 h post-EE. The fourth dog did not receive dexrazoxane until 48 h after EE when clinical signs of extravasation injury were reported [9]. This latter dog had the only case in this series that required amputation [9]. In the present study, the doxorubicin EE cases with the most severe AEs requiring amputation (Cases 7 and 11) both received dexrazoxane within 2 h of the EE; therefore, it is less likely that the timing of dexrazoxane administration played a role here. Further studies to evaluate the optimal dexrazoxane dose, timing, and frequency in dogs and cats experiencing doxorubicin extravasation are needed. The majority of cases experiencing doxorubicin EE (6/9) were managed on an outpatient basis. Three of nine cases were managed as inpatients, 2/3 due to the severity of extravasation injury (Case 7 and 11), and 1/3 due to client convenience during serial administration of dexrazoxane.

This study found that carboplatin was the second most common extravasated drug and that all five cases experiencing carboplatin EE resulted in Grade III or IV AEs which required surgical debridement. In one of the five cases, it was estimated that 75% of carboplatin was extravasated. This case (Case 1) experienced Grade III AEs. In the other four cases, the estimated volume of drug extravasated was unknown (n = 3 Grade III and n = 1 Grade IV). Given the small sample size, the effect of the volume of drug extravasated with respect to carboplatin EE is unknown. Historically, carboplatin has been thought to be an irritant rather than a vesicant [6, 16]. Intracavitary carboplatin for the treatment of neoplastic effusions is not associated with tissue injury [17]. In another study, subcutaneous carboplatin was administered via an indwelling subcutaneous catheter placed at the amputation site or the interscapular region in 17 dogs as adjuvant therapy for osteosarcoma [18]. This study concluded that the surgical infection rate was higher in these dogs compared with infection rates following clean surgeries [18]. The authors of this study concluded that it is unknown if the localized infusion of carboplatin at the surgical site contributed to the increased infection rate. In another study, carboplatin caused ulcerations when administered intradermally to laboratory mice at concentrations greater than or equal to 10 mg/mL [19]. Additionally, Miller et al. evaluated carboplatin extravasation in seven dogs, and although the extravasation injuries were not graded according to VCOG-CTACE criteria, six dogs developed full-thickness necrosis and had to receive surgical debridement [8]. Our study further demonstrates the potential for carboplatin to act as a vesicant. At this time, it is unknown what factors might contribute to the AEs from carboplatin extravasation. One possibility is that carboplatin can be diluted in 5% dextrose solution, which can contribute to irritation if extravasated; however, this is unlikely the causative factor in this study given that dilution did not occur in any of the carboplatin EE reported here [20]. Based on the findings in this study as well as prior studies, it is recommended to have a heightened awareness of the risks of extravasation injury when administering carboplatin [8, 18].

In this study, vinca alkaloid EE resulted in mild–moderate (Grades II–III) AEs. In physician-based medicine, hyaluronidase is recommended for the treatment of vinca alkaloid extravasations [21, 22]. Spugnini et al. evaluated the use of hyaluronidase for the treatment of extravasation of vincristine ± mechlorethamine or doxorubicin in six dogs. In this study, the injections were repeated weekly until side effects resolved, and all dogs recovered within 6 weeks [12]. In the current study, there were five vinca alkaloid EE, and two (Cases 22 and 23) were treated once with hyaluronidase. The severity of AEs (Grades II–III) for these two cases was similar to the severity of AEs for the three cases not treated with hyaluronidase. Given the small sample size and only one-time use of this neutralizing agent compared to the Spugnini et al. study, strong conclusions on the effect of hyaluronidase on vinca alkaloid EE cannot be made in the current study.

In the current study, mitoxantrone EE (n = 2) and dacarbazine EE (n = 1) resulted in no AEs. One of the mitoxantrone EE (Case 17) was due to leakage of the chemotherapy agent into the subcutaneous space during intracavitary administration. While this does not fit the true definition of extravasation, this case was not excluded from this study, as some reports classify mitoxantrone as a vesicant that causes tissue damage based on concentration and volume [1, 5-7]. Rabacfosadine EE (n = 1) resulted in Grade III AEs and was managed in the outpatient setting. Larger studies are required to further explore the severity of AEs resulting from extravasation of these chemotherapy agents.

In physician-based medicine, cold compresses are recommended for extravasations of DNA binding agents, while warm compresses are recommended for extravasations of non-DNA binding agents [1, 21]. In the current study, given the small sample size and variability of use, the benefit of aspirating extravasated drug, compresses, and neutralizing agents is unknown. The use of supportive care and bandages was variable and dependent on individual case clinical factors and clinician preferences; thus, strong conclusions on the effect of these treatments cannot be made.

The majority of cases (20/23) had mild to moderate or no AEs associated with the EE. Three of the 23 cases had severe (Grade IV–V) AEs secondary to doxorubicin EE (n = 2, one Grade IV and one Grade V) and carboplatin (n = 1, Grade IV). Amputation was required for both latter cases of doxorubicin EE, one case of which the owners ultimately elected euthanasia instead of undergoing amputation.

The majority of EE occurred within the first four chemotherapy administrations (18/23, 78%), with extravasation occurring at the first chemotherapy administration in 7/23 (30%) of cases. Furthermore, 19 cases were fully awake and received no form of injectable or oral sedatives before chemotherapy administration. These factors highlight the need for careful consideration of the animal's compliance, especially early in the treatment protocol when personnel are naïve to how the animal will react to restraint and venipuncture. The use of chemical restraint could be considered to mitigate the potential for EE due to animal noncompliance. However, this is not a foolproof measure as one case in this study still experienced doxorubicin EE despite injectable sedation protocol.

The majority (18/23) EE occurred at a specialty hospital (academia or specialty private practice). Given that it is more common for chemotherapy to be administered in the specialty hospital setting in veterinary medicine, this finding is not surprising. In addition, selection bias is likely a factor in this population, given that cases were recruited through a specialist online listserv.

In physician-based medicine, there are several reported risk factors for extravasation injuries, including but not limited to small veins, age, and use of large gauge catheters or butterfly catheters [3, 5-7]. In the current study, there were six cases weighing less than 7.5 kg, but in these, the extravasation injuries were overall mild (Grades 0–II). The catheter gauge used for the cases that reported the size was standard 22-gauge (8/23) and 25-gauge (2/23). A butterfly catheter was used in only one case, while the majority (21/23) had chemotherapy administered via intravenous catheter. Given the small sample size, it is difficult to determine how much of a role these risk factors played in the EE in this study. Fluid pump-assisted administration was used in one case of doxorubicin EE and one case of dacarbazine EE in this study. This method of delivery is discouraged when administering drugs with known extravasation risk to veterinary species, especially in instances in which the animal is not being closely monitored or restrained, as animal movement can cause the intravenous catheter to become displaced, and the fluid pump continues to push fluid regardless of catheter patency.

The left forelimb was the most common location for EE to occur (11/23 cases, 48%). The reason for this is likely multifactorial, including that the forelimbs might be more commonly used for chemotherapy administration due to accessibility, or perhaps restraint of the animal for venipuncture in this position is more likely to be compromised. As the majority of the human population is right-handed, it is possible that placing an intravenous catheter in the left front limb could be more technically challenging, potentially leading to errors in catheter placement. Given the small sample size of this population, it is difficult to draw strong conclusions.

The majority of cases experiencing EE did not experience severe AEs and were able to go on to complete chemotherapy. Only five cases experienced long-term AEs, including scar formation or alopecia. Two cases discontinued chemotherapy due to the EE, and one case discontinued chemotherapy due to the client decision for euthanasia instead of amputation after EE.

5 Limitations

A major limitation of this study is its retrospective nature, including the assignment of AE grade. Reporting bias might also be at play given that extravasation of drugs known to cause severe injury or EEs that caused severe clinical signs were more likely reported. In addition, extravasation might not be recognized at the time it occurs and therefore not be well documented in medical records. Less severe AEs (Grades 0–I) might be underreported due to minimal clinical signs. Selection bias was introduced given that the contributors to this study are all oncology specialists, and the oncology specialty listserv was used to recruit cases. It is possible that more cases could have been obtained if the case search had been opened to general practitioners and internal medicine specialists. Furthermore, the small case numbers and variability in EEs and treatments preclude the possibility of strong conclusions.

6 Conclusion

Chemotherapy extravasation is a potentially serious complication. This study demonstrated that doxorubicin and carboplatin extravasations caused the most serious AEs. Two cases in which it was estimated that the majority of doxorubicin was extravasated resulted in more serious complications. However, due to the small sample of cases in which the volume extravasated was reported, a strong conclusion on the impact of the volume extravasated on AE severity cannot be formed. This study supports handling carboplatin as a vesicant rather than an irritant. Personnel administering chemotherapy must be aware of the risks and complications from extravasation injury and make efforts to reduce the risk. This study demonstrated that the majority of extravasations occurred within the first four treatments and that nearly all cases were fully awake and received no sedation before treatment. The use of chemical restraint can be considered by care teams to help mitigate risks, especially in cases receiving chemotherapy known to cause severe extravasation injury, although this might not be appropriate for every case depending on other clinical factors.

Disclosure

Authors declare no off-label use of antimicrobials.

Ethics Statement

Authors declare no Institutional Animal Care and Use Committee or other approval was needed. All clients consented to the treatment provided to these animals. Authors declare human ethics approval was not needed.

Conflicts of Interest

The authors declare no conflicts of interest.