Comparative wound healing-Are the small animal veterinarian's clinical patients an improved translational model for human wound healing research?

PERSPECTIVE ARTICLE

Comparative wound healing—Are the small animalveterinarian’s clinical patients an improved translationalmodel for human wound healing research?Susan W. Volk VMD, PhD1; Mark W. Bohling DVM, PhD2

1. Department of Clinical Studies and Animal Biology, School of Veterinary Medicine, The University of Pennsylvania, Philadelphia,Pennsylvania, and2. College of Veterinary and Comparative Medicine, Lincoln Memorial University, Harrogate, Tennessee

Reprint requests:

Dr. S. W. Volk, Department of ClinicalStudies and Animal Biology, School ofVeterinary Medicine, The University ofPennsylvania, 312 Hill Pavilion, 380 S.University Avenue, Philadelphia, PA19104-4539, USA.Tel: +1 215 898 0635;Fax: +1 215 746 2295;Email: [email protected]

Manuscript received: October 7, 2012Accepted in final form: February 28, 2013

DOI:10.1111/wrr.12049

ABSTRACT

Despite intensive research efforts into understanding the pathophysiology of bothchronic wounds and scar formation, and the development of wound care strategies totarget both healing extremes, problematic wounds in human health care remain aformidable challenge. Although valuable fundamental information regarding thepathophysiology of problematic wounds can be gained from in vitro investigationsand in vivo studies performed in laboratory animal models, the lack of concordancewith human pathophysiology has been cited as a major impediment to translationalresearch in human wound care. Therefore, the identification of superior clinicalmodels for both chronic wounds and scarring disorders should be a high priority forscientists who work in the field of human wound healing research. To be successful,translational wound healing research should function as an intellectual ecosystem inwhich information flows from basic science researchers using in vitro and in vivomodels to clinicians and back again from the clinical investigators to the basicscientists. Integral to the efficiency of this process is the incorporation of modelswhich can accurately predict clinical success. The aim of this review is to describe thepotential advantages and limitations of using clinical companion animals (primarilydogs and cats) as translational models for cutaneous wound healing research bydescribing comparative aspects of wound healing in these species, common acute andchronic cutaneous wounds in clinical canine and feline patients, and the infrastructurethat currently exists in veterinary medicine which may facilitate translational studiesand simultaneously benefit both veterinary and human wound care patients.

UNDERSTANDING COMPARATIVE WOUNDHEALING TO IMPROVE CLINICAL WOUNDCARE ACROSS SPECIESEthical and moral considerations require the use of animalmodels in wound healing research for human beings. Thesemodels have made many important discoveries possible, par-ticularly at the basic science level. However, as the questionbecomes more clinically oriented—“Will the use of treatmentX make a clinically relevant difference in the outcome whencompared to the current standard of care?”—informationobtained from existing laboratory animal models becomesriskier. For this reason, it is imperative that wound healingresearchers, particularly those whose investigations arecarried out at the organism level, become familiar with thefield of veterinary comparative wound healing.

Veterinarians have long been aware that although allanimals appear to follow the same fundamental patterns ofhealing (first intention vs. second intention), with the samebasic stages (inflammation, proliferation, remodeling), theactual healing outcomes often diverge across species, either

with regard to the timeline of healing, or the appearance of thefinal result. Some species exhibit rapid healing while othersheal much more slowly; some produce abundant granulationtissue, others much less. Oftentimes the differences in woundhealing are attributable to differences in the gross and histo-logic structure of the skin.1 For example, it is well known thatsecond intention healing in rats and mice occurs very rapidly,and that this wound closure is primarily the result of woundcontraction facilitated by the panniculus carnosus muscle.Rapid contraction is a common feature of animals with looseskin, whereas in tight-skinned species (human, porcine) con-traction is not nearly as rapid as wound closure occurs prima-rily as the result of reepithelialization. The looser skin overthe body/trunk wounds of companion animals, such as thedog, cat, and horse, has been shown to heal primarily bycontraction, while extremity wounds on these animals healin a manner more similar to truncal wounds in tight-skinnedspecies (Table 1).2

It is also important to note that significant phylogeneticdivergence is not required for species to exhibit major quali-tative or quantitative differences in wound healing. For

Wound Rep Reg (2013) 21 372–381 © 2013 by the Wound Healing Society372

example, domestic dogs and cats, both members of the orderCarnivora, exhibit very different patterns of healing of bothfirst and second intention wounds. Comparing clean acutewounds of dogs and cats reveals significant differences. Theinflammatory phase in dogs is rapid in onset and robust incharacter; swelling, redness, and exudation are all usuallyquite evident in the first 24 to 72 hours postwounding.3 Incontrast, wounds in the cat develop a relatively mild inflam-matory response with little or no swelling, redness, or exuda-tion. The immediate consequence of this difference is visiblein the early proliferative phase of second intention healing,with canine wounds producing significantly more granulationtissue than feline wounds during the same time frame. Con-sequently, canine wounds have been shown to exhibit signifi-cantly faster rates of contraction and reepithelializationcompared with feline wounds. In standardized (2 cm2 exci-sional) experimental wounds in dogs and cats, the averagereduction in wound area via contraction at 7 days was 41% fordogs, compared with only 18% in cat wounds. Similarly,

reepithelialization in feline wounds was significantly slower.At day 14 postwounding, only 13% of the initial wound areawas covered with neoepithelium in feline wounds; in contrast,an average of 44% of the initial wound area was reepithelial-ized in wounds of dogs. Removal of subcutaneous tissues hasbeen shown to significantly delay the formation of granulationtissue in both dogs and cats, but the deleterious effect ismuch more marked in cats than in dogs. Fascial abrasion andfasciotomy have been shown to reverse some of these earlydeficits in granulation tissue formation in the cat.4 Althoughthe mechanisms by which these comparative impairments towound healing occur in the cat are currently unidentified, theylikely contribute to the development of true chronic wounds inthis species such as axillary fold wounds and the phenomenonof pseudohealing (more details in the chronic wound sectionnext).

Animals do not need to come from different genera toexhibit significant differences in wound healing; in fact, evenwithin the same species, striking differences in wound healing

Table 1. Comparative cutaneous wound healing among species

SpeciesCommon

name Skin type 2o closure

Inflammationcharacter andtiming of peak

Granulationamount and

character

2 cm2 squareexcisional wound

area reduction

Rodentia Rats,mice

Loose Contraction Robust; peaksbetween days4 and 7postwounding54,55

Good Mice: 50% or greater contractionin 3 to 7 days56,57

Rats: 50% contraction in 12.5days,58 up to 80–90% of totalclosure by contraction59,60

Homosapiens

Humans Tight Reepithelialization Robust; 48–72 hours Good Reepithelialized in 10–12 days61;overall 20–40% of woundclosure by contraction62,63

Porcine Pigs Tight Reepithelialization Robust at 7 dayspostwounding64

Good; note hypertrophicscar formation in RedDuroc pigs

20% reduction in wound area by11 days postwounding65

Carnivora Dogs Trunk: loose Contraction Robust; up to 72hours3

Abundant, wellvascularized3

Trunk wounds:Day 7, 41% contraction;Day 14, 44% reepithelialized3

Extremities:tight

Reepithelialization

Carnivora Cats Trunk: loose Contraction Mild3 Sparse, fibrotic3 Trunk wounds:Day 7, 18% contraction;Day 14, 13% reepithelialized38

Extremities:tight

Reepithelialization

Equus feruscaballus

Horses Trunk: loose Contraction Weak, butpersistent66,67

Exuberant granulationtissue (GT) at allwound margins66

� 12 weeks to heal 2.5 ¥ 3.5 cmwounds; contraction: 59% at 9weeks

Extremities:tight

70%Reepithelialization

Exuberant, fibrotic,irregular surface66,68

Contraction: 12% of 2.0 ¥ 3.5 cmwounds at 9 weeks68

Equus feruscaballus

Ponies Trunk: loose Contraction Rapid and robust66 Pink, well vascularized,regular66,68

7–9 weeks to heal 2.0 ¥ 3.5 cmwounds; contraction: 76% of2.0 ¥ 3.5 cm wounds at 9weeks68

Extremities:tight

Reepithelialization Smooth, wellvascularized66 overallbut with someexuberant GT at distalborder66,68

7–9 weeks to heal 2.0 ¥ 3.5 cmwounds; Contraction: 49% of2.0 ¥ 3.5 cm wounds at 9weeks68

Volk and Bohling Companion animals as preclinical models and clinical wound care patients

Wound Rep Reg (2013) 21 372–381 © 2013 by the Wound Healing Society 373

have been observed at the gross, cellular, and molecularlevels. The best-described example is the comparison betweenhorses and ponies, both members of the same species andsubspecies, Equus ferus caballus. In spite of their manygenetic and phenotypic similarities, horses and ponies exhibitsignificant differences in first and second intention woundhealing.2 For example, in an experimental comparison ofsecond intention healing on horses and ponies, the latter wereshown to have a more rapid onset and resolution of woundinflammation, leading to more rapid wound contraction andepithelialization.

Genetic variations affecting wound healing also existwithin the human race between different ethnic populations.For example, an increased propensity for fibroproliferativescarring to occur in patients of African American, Mediterra-nean, and Hispanic decent compared with Caucasian is wellknown.5 Genetic variation differences in healing have alsobeen described in nonhuman species and this knowledgeexploited in the development of models that uniquely displayphenotypes similar to that seen in human wound pathology.One such example is the female Red Duroc pig, which hasbeen proposed as a model for fibroproliferative scarring inpeople. The female Red Duroc pig develops raised, thick scarsfollowing deep dermal injury, as compared with normalhealing of the same wounds in theYorkshire pig (same speciescontrol).6 The Mexican Hairless dog also exhibits similarfibroproliferative scarring after injury; these lesions are notseen in other breeds of dogs.7 Knowledge of such interspeciesand intraspecies differences can help the human woundhealing researcher choose the best animal model for the ques-tion at hand, and also helps in the interpretation of results andextrapolation of data into other species and other clinicalsituations.

NATURALLY OCCURRING ANIMALMODELS OF ACUTE TRAUMATICCUTANEOUS INJURIES

Thermal injuries

The basic pathophysiology of burn wounds—the local andsystemic burn wound response—appears to be fairly uniformacross species lines. In fact, much of what is known todayabout the pathophysiology and treatment of burns in humanshas been gleaned from experimental animal models. Althoughthese models (chiefly rodents, pigs, and sheep, but also dogsand cats to a lesser degree) have yielded much valuable infor-mation regarding thermal injury pathophysiology and therapy,there are significant social and ethical drawbacks of suchstudies that are difficult to ignore. Given such ethical consid-erations of performing burn research in mammals in vivo8,9

and the limitations of in vitro assays for such purposes, onealternative to the use of lab-created burn wounds in animalswould be the use of naturally occurring burns in small animalveterinary patients.

Burn wounds are relatively uncommon in domesticanimals; however, when they do occur, they often requireadvanced wound care techniques. As such, these naturallyoccurring animal burns represent a valuable resource forhuman wound healing research. Most commonly, animal burnwounds are caused by accidental or malicious environmentalexposure (i.e., scalding water, hot surfaces such as mufflers,

hot stoves or radiators, or open fire) or may be iatrogenic(secondary to electric heating pads/lamps or improperlygrounded electrocautery units). Most burn wounds seen bysmall animal veterinarians do not exceed 20% of total surfacebody area and therefore do not have significant secondarysystemic effects.10 Animals that have sustained more exten-sive burns, particularly those involving the extremities, areoften euthanized based on the guarded prognosis, and finan-cial considerations of the owner. Larger wounds may lead tocontractures and/or wounds that fail to fully heal withoutsurgical reconstruction (Figure 1). Even less extensive inju-ries often require referral to a veterinary specialist foradvanced wound care, with costs often exceeding severalthousand dollars of out-of-pocket costs to the owner (withfurther escalation of costs in critical patients) and time-intensive therapies associated with in-office care. For suchpractical reasons as these, pet owners are motivated to seekclinical trials which may provide financial compensation or asuperior treatment option that may be otherwise unavailableto a beloved pet. Therefore, although severe thermal burns arenot as frequent in animals as in human beings, there can stillbe significant opportunity for physicians to use companionanimal models in translational clinical research projects.

As stated previously, many of the current recommendationsin human burn treatment, such as early excision, grafting, aswell as topical treatments, were first developed in animalmodels. Therefore, it should come as no surprise that theseand other state-of-the-art burn wound care techniques arestandard care in veterinary referral centers. Like the earlieranimal models, these patients could provide a valuableresource for developing and testing novel therapeutics prior toinitiation of clinical trials in humans. In fact, recent worksuggests that murine models are particularly poor at success-fully predicting the efficacy of drug candidates and medicalinterventions which target diseases which possess an inflam-matory component, such as burn wounds.11 Although it

Figure 1. Scarring with contracture in a young mixed breeddog following severe burn injury. This patient presented to areferring veterinary surgeon with a nonhealing wound andcontracture which prevented a normal stance and ambulation(the left hindlimb is unable to be extended caudally nor the leftforelimb cranially from its position in the photo). The dog wasrescued following malicious injury (doused with gasoline andset on fire) and healing was allowed to proceed via secondintention prior to referral. Photo courtesy of Dr. Lillian Aronson.

Companion animals as preclinical models and clinical wound care patients Volk and Bohling


remains unclear whether the predictive index for studies incanine/feline models will be high, the combination of thecurrent poor predictive index of murine models for translationof novel therapies, the natural occurrence of such wounds inclinical veterinary patients, and the similarity of additionalsupportive treatments required for survival of extensivewounds in an intensive care setting available at referral vet-erinary institutions which is unavailable for murine modelssuggest that such effort in exploring these models is war-ranted. Developing infrastructure to perform multicenter trialswould be essential to enroll the large numbers of casesrequired for clinical trials in small animals; fortunately, suchclinical trials are becoming increasingly commonplace in vet-erinary medicine.

In addition to thermal injury, radiation burns (or “acutelocal radiation toxicity” to radiation oncologists) occur fre-quently when radiation teletherapy is applied to animals fortumor control. It is interesting to note that species differencesalso exist in the response to this type of wound. In the clinicalexperience of the authors, dogs tend to have a significantlymore severe inflammatory response to radiation treatmentthan cats do, given the same dosing intensity. Mild to severeacute radiation dermatitis is seen in >90% of canine patientsreceiving fractionated radiation therapy.12 Similar to humans,veterinary patients suffer from morbidity associated withthese lesions and may experience treatment delays associatedwith their existence, which may limit the success of theironcologic therapy. Successful completion of a placebo-controlled, randomized, double-blind prospective clinical trialin canine patients undergoing radiation therapy for soft tissuetumors establishes feasibility to assess efficacy of therapieswhich limit acute radiation dermatitis or improve subsequenttherapies for canines and humans.13

Bite wounds

Bite wounds are rather common in dogs and cats and couldeasily serve as models for infected puncture wounds in man.In particular, dog-on-dog bite wounds would seem to offer aparticularly attractive model for human medical research,because of the frequency of dog bites sustained by humansand the seriousness of these wounds. A few statistics bear thisout: A 2003 Center for Disease Control survey reported thatapproximately 4.5 million Americans are bitten by dogs eachyear.14 Over 800,000 of these cases require professionalmedical attention15 and 9,500 required hospitalization, at anaverage cost of $18,200 per patient.16 The combination ofextreme tearing of muscle tissue and the infected wound envi-ronment that is seen in dog bite wounds is nearly impossibleto replicate in experimental models (Supporting InformationFigure S1). Ready access to large numbers of cases shouldmake dog bite wounds a natural opportunity for translationalresearch.

Dog attacks account for 85% of all bite wounds in people.17

Although cat-inflicted bite wounds are significantly lesscommon, approximately 400,000 people are bitten by catsannually, with the actual number probably much higher due tounderreporting. Even more significant is the high infectionrate, up to 80% infection, in humans who are bitten by cats asopposed to a 15–25% infection rate with dog bites.17,18 Con-tributing to differences in infection rates may be the type oftrauma and the inoculum delivered by canine vs. felineaggressors. Larger dogs can cause more significant crush

injury due to higher pressures exerted by their larger jaws. Notsurprisingly, morbidity and mortality escalates when smallerpatients are victims, such as that seen with “big dog, littledog” altercations or as seen with human pediatric victims.19,20

In contrast to the crushing injury and damage to soft tissuestructures leading to punctures, lacerations, and avulsions bydogs, cat-inflicted bite wounds tend to result in deep punc-tures which provide an ideal anaerobic environment to facili-tate growth of the most common bacterial isolate from cat bitewounds, Pasteurella multocida.21 Similarly, differences inbacterial flora as well as injury type likely contribute to thehigher incidence of abscess formation, as opposed to purulentdrainage without abscessation, seen in cat-inflicted bitewounds compared with those by dogs.22 True infection rates incanine and feline patients are unknown. These data are diffi-cult to ascertain as there is significant case selection bias dueto the fact that many patients with uncomplicated (nonin-fected or non–life-threatening) wounds are not presented to aveterinarian, unless the site becomes infected, as well as thefact that virtually all dogs and cats presented for bite woundsare placed on antibiotic prophylaxis whether cultures havebeen obtained or not.19 The potential seriousness of infectedbite wounds is magnified due to the recent emergence ofmethicillin-resistant Staphylococcus spp. in pets,23 a factorthat may complicate treatment in human patients and onewhich argues for more research in the treatment of thesewounds. As with dogs, pet cats are frequently bitten by con-specifics, and these naturally occurring cases are a potentialsource of case material for research.

Ballistic wounds

Cats and dogs in certain demographics may present to emer-gency clinics for gunshot wounds, with handgun injury mostprevalent in urban areas and rifle/shotgun injuries occurringwith greater frequency in rural areas. Presenting signs aredictated by location of impact and whether high- or low-energy injury causes extensive soft tissue and orthopedictrauma or more limited soft tissue trauma. In a retrospectivestudy examining 84 cases of gunshot injury in dogs and catspresenting to an urban veterinary referral hospital, the major-ity of wounds were sustained on the limbs and thorax.24 Mostpatients were young (<3 years of age), sexually intact, malelarge breed dogs with a history of being allowed to roamunsupervised. Of 52 animals with limb injuries, 23 were asso-ciated with fractures. Dogs with penetrating abdominalwounds or vertebral injury were found to have a worse prog-nosis compared with those with thoracic or limb injuries,although the majority of animals for which owners pursuedadequate treatment were found to survive (66/77). These casesmay provide a translational model to examine interventionsthat improve clinical outcome in human victims of domesticviolence as well as military personnel sustaining combat inju-ries that involve significant composite tissue damage.

Degloving injuries

Degloving injuries represent another relatively commonwound presentation in small animal veterinary practice. Mostfrequently this injury occurs when an animal is hit by a carand the foot scrapes along the pavement, resulting in a loss notonly of skin but substantial loss of bone as well. This wound



is in fact more correctly termed a shearing wound, althoughcommonly referred to as a degloving wound given theremoval of skin surrounding the distal extremity (Figure 2).These wounds are somewhat challenging to treat as theyrequire reconstruction/treatment of associated orthopedicinjury as well as effective debridement, infection control, andprotection of remaining underlying structures until the woundcan either heal by second intention or be resurfaced, usuallywith a skin graft.

Ischemic wounds

Bandages are commonly applied to the lower limbs of smallanimal patients to provide support and immobilization follow-ing orthopedic injury either pre- or postoperatively, as well asto protect wound sites. Bandage-related injuries may occurwhen a bandage is improperly applied or when the bandage isimproperly managed at home by the owner (Figure 3). Ifcaught early enough, the ischemic wound can be reperfusedonce the bandage is removed. Unfortunately, improperbandage management often results in an ischemic wound thatcannot be reperfused. As a result, full thickness necrosis andskin loss secondary to bandage complications in veterinarypatients require intensive wound management just as they doin human patients, and similar strategies of wound manage-ment are routinely applied. Many treatments commonly usedin human patients with ischemic wounds are also used inveterinary patients, including pressure-relieving bandaging,special bedding, negative pressure wound therapy, pulsatilelavage, and a broad variety of topical anti-infectives andwound-healing stimulants. If these measures fail, the situationoften necessitates skin grafting, or amputation of digits oreven entire limbs. As is the case in human medicine, theprognoses for such cases may be guarded: In a small caseseries of nine dogs and two cats with ischemic bandage inju-ries, only four animals fully returned to function.25 The avail-ability of these cases in veterinary medicine, their amenabilityto clinical intervention, the clinical need for improved out-comes and good access to clinical follow-up make them anideal opportunity for collaboration with researchers interestedin human wound healing.

IMPAIRED WOUND HEALING IN SMALLANIMAL PATIENTSA combination of advances in veterinary care and the demandfor such care by the pet-owning public has led to an increase

in the life span of companion animals over the last fourdecades. Pets are living longer with a wide variety of degen-erative, infectious, immune-mediated, and neoplastic diseasesdue to advances in treating these diseases and associatedsymptoms; however, certain disease states and/or their thera-pies have been implicated in delayed wound repair.

Endocrinopathies and immunosuppression

Corticosteroids are used to treat many inflammatory, immune-mediated, and neoplastic conditions in dogs and cats. Theiruse therefore puts such patients at risk for impaired woundhealing through its effects on multiple reparative cells duringthe process of healing. Immunosuppression associated withsuch infectious diseases such as feline leukemia virus andfeline immunodeficiency virus may also predispose cats toimpaired wound repair. Although common endocrinopathiesof cats and dogs, such as hypothyroidism, hyperadrenocorti-cism, and diabetes mellitus, are often implicated in impairedwound healing in small animal patients, there have been nosuch studies to date documenting their effects on cutaneouswound repair in these patients. Cats and dogs with endocrino-pathies have been shown to be 8.2 times as likely to developpostoperative wound infections as patients without an identi-fied endocrinopathy, although none of the patients in thisstudy were identified diabetics.26

Oncologic treatments

The small animal oncologic patient is also at risk for impairedwound healing. The cytotoxic, antiproliferative, and/or

Figure 2. Shear injury in a canine patient involved in a roadtraffic accident (hit by car). A large skin defect over the distalextremity exposes damage to underlying skeletal and softtissues (an existing open articular fracture is not evident in thisphoto).

Figure 3. Ischemic injury occurring secondary to an improp-erly managed bandage. A splint was initially placed for conser-vative management of a distal radius and ulna fracture. Thepatient was referred for ischemic injury to the distal ante-brachium secondary to improper splint application and main-tenance. Initially, open wounds were noted on the cranialaspect of the elbow with necrosis of skin distally on theforelimb (A). Pulses were poor in the distal limb and aPseudomonas sp. was cultured from the wounds. One weeklater, healthy granulation tissue can be seen at the proximalaspect of the wound over the proximal radius and ulna but thedistal extremity revealed widespread necrosis of the distallimb (B), necessitating amputation.



antimetabolic side effects of chemotherapeutic agents on cellscritical to the wound healing process are well documented.Additionally, as stated previously, radiation therapy may bethe appropriate treatment modality for neoplasia in dogs andcats, with the untoward side effects radiation-induced derma-titis and chronic ulcer formation in up to 14% of patients27

(Supporting Information Figure S2). Furthermore, healing ofsurgical incisions is inhibited if radiation is performed in theearly postoperative period or if surgery is performed in apreviously irradiated field.

Surgical site infection

Surgical site infection (SSI) in small animal patients has beenwidely reported in the veterinary literature for both opensurgical procedures as well as minimally invasive procedures.The reported incidence of SSI in small animal surgicalpatients is 1.7 to 18.1%.28–33 While the reported incidences ofSSI in humans vary widely, several reasons may account forthe apparent higher rate (percentages) of SSI in humans com-pared with those in animals, such as duration of surgicalprocedures and postoperative hospitalization, inherentlybetter immune surveillance in animal patients, or the lack ofstandardized reporting of SSI in veterinary medicine. The useof certain anesthetics, an increased surgical or anestheticduration, surgical site preparation (hair clipping) prior toanesthetic induction, concurrent endocrinopathy, increasedpatient weight, number of persons in the operating room, andmale sex have been identified as risk factors for SSI in dogsand cats.26,28,31–33 As with human patients, SSI can have aprofound impact on patient care by prolonging wound man-agement with associated increased client cost, requiring revi-sion surgeries and increasing overall morbidity and mortalityof patients. Similar to human hospitals, multidrug resistantSSIs are of increasing concern for small animal patients.34,35

NATURALLY OCCURRING CHRONICWOUNDS IN THE SMALL ANIMAL PATIENT

Pressure ulcers

Pressure ulcers in small animals, particularly dogs, are causedby the compression of soft tissues over bony prominencesagainst underlying surfaces on which the patient rests(Figure 4). The resulting pressure ulcers are very comparablein appearance with those seen in human beings, and veteri-nary patients may be presented for treatment with ulcersranging from National Pressure Ulcer Advisory Panel(NPUAP) Stage I to Stage IV.36

Veterinary small animal patients with pressure ulcers aretypically neurologic patients who are para- or tetraplegic ororthopedic patients which have difficulty rising and ambulat-ing for a variety of reasons, including the presence of multiplefractures. Perhaps the largest at-risk group and source of themajority of pressure ulcers in small animals, however, aregeriatric dogs that have advanced osteoarthritis, particularlyof the pelvic limbs. Besides being a very large group ofpatients, these dogs should be of particular interest to humanresearchers who focus on pressure ulcers, because of thesimilarities in the patients. For example, not infrequently theypossess the same risk factors as their human counterparts—

advanced age, obesity, and in some cases, type 2 diabetes.Veterinarians incorporate many of the same preventative mea-sures utilized in human medicine to prevent pressure ulcers inthe medical management of at-risk patients, including chang-ing the position of patients frequently, providing sufficientlypadded bedding, and maintaining adequate nutrition andcleanliness. Similarly, the treatment measures for pressureulcers in veterinary patients are much the same as for humans.Debridement, bandaging to prevent continued trauma, anddiligent nursing care may be sufficient to treat mild ulcerswhile more severe lesions require extensive surgical and con-tinued postoperative management.

One dog breed is worth special mention in regard to pressureulcers. The greyhound breed is known to be particularly sus-ceptible to the development of such lesions due to their thinskin, short hair, and angular conformation.37 With its naturalpredilection for dermal pressure ulcers, the greyhound hasbeen proposed to serve as a model for human pressure ulcers.36

Feline pseudohealing and axillary

(indolent pocket) wounds

The term “pseudohealing” has been used to describe a form ofaberrant cutaneous wound healing in small animal species,although it occurs predominantly in the cat. The clinical con-sequence of such healing is revealed by complete dehiscenceof a surgical wound, with little or no associated bleeding,following normal physiologic stress (for example, when a catjumps from the floor up onto a chair). The wound bed and edgesare typically covered with a poorly perfused, fibroproliferativetissue. It has been hypothesized that disruption of the subcu-taneous tissues during the creation of the original wound orsurgical defect may be responsible for this phenomenon. It hasbeen experimentally shown that removal of subcutaneoustissues creates a negative impact on cutaneous wound healing,and this effect is much more pronounced in cats than in dogs.38

Indolent pocket wounds are a special case of chronicwounds and appear to be seen predominantly (though notexclusively) in the axillary region of cats. In the typical pre-sentation, a foreleg becomes entrapped within its collar whicheventually cuts into the axilla and causes a wound in the

Figure 4. Pressure ulcer over the point of the elbow in ageriatric German shepherd dog. Difficulties in rising and ambu-lating due to severe coxofemoral osteoarthritis secondary tohip dysplasia were predisposing factors in the development ofthis pressure ulcer.



axillary region.39 The presence of foreign material (collar) inthe wound may prevent healing and allows chronicity tobecome established, but even after the collar is removed, thesewounds are predisposed to failure to heal. Invariably, a deeppocket wound forms, lined with chronic granulation tissue,and seems incapable of either contracting or of reepithelial-ization (Supporting Information Figure S3). Surgical treat-ment is indicated; some have successfully treated thesepatients by providing vascular supply via a pedicle graft ofgreater omentum brought into the wound via subcutaneoustunnel or other vascularized graft.39–42

While there are no known direct correlates to either of theaforementioned conditions in human wound healing, thesenatural models of aberrant healing may still be of practicalinterest to researchers in search of translational models forincisional dehiscence (pseudohealing) or chronic woundswhich are characterized by fibrotic granulation tissue that haslost the ability to facilitate wound contraction and epithelial-ization (indolent pocket wounds).

VETERINARY WOUND CAREThe need for improved clinical outcomes in veterinary woundcare has led to a significant increase in commercially avail-able products in the last decade for use in companion animals,including topical agents, dressings, biologics, closure devices,and negative pressure wound therapy. Dogs, and to a lesserextent cats, have been used to investigate efficacy of variouswound care interventions in experimental models as transla-tional models for both veterinary and human applications.43–49

In addition, veterinary wound care has advanced through itsadaptation of products and techniques commonly used inhuman wound care, such as an increased use of negativepressure wound therapy in recent years.49–52

In conclusion, efficient translation of research findings inwound healing research to wound care interventions capable ofsignificant improvements in outcome measures relies upon acoordinated approach of discovery and development, investi-gations of efficacy in a preclinical model, and finally in thetarget clinical population where sustainability of such inter-ventions must be determined feasible. It is clear that laboratoryanimal models may not be the most accurate models to predictclinical efficacy in human wound healing research due tolimitations in accurately modeling clinical conditions.11 Natu-rally occurring wounds in canine and feline companionanimals may provide superior models to examine interventionfor the treatment of true chronic wounds or for injuries whichare difficult or impossible to recreate in the laboratory setting(due to both technical or ethical reasons). In addition, veteri-nary patients are outbred with a relatively longer life, andcommonly develop the same comorbidities (such as diabetesand obesity) that affect their human counterparts. Their com-panion status also exposes them to environmental factors (sec-ondhand smoke and environmental toxins) that may impacthealing in their owners. Whether it is for their young orgeriatric pets, owners are demanding increasingly advancedclinical care that is similar to what they themselves mightreceive. In fact, in some cases advanced wound therapies (suchas stem cell therapy) are available for pets prior to theiravailability in human medicine. To meet these demands, clini-cal trials (including multi-institutional trials) are being rou-tinely offered for a variety of veterinary ailments.53 Pet-owningclients have been willing to enroll their pets into studies in

which both minimally invasive (biopsies, bloodwork, imagingunder anesthesia, etc.) and noninvasive (repeated physicalexaminations, gait analysis, etc.) procedures are performed todetermine potential treatment benefits that may directly helptheir pet or to help others with similar conditions.

It should be noted that many of the above traits that maymake canine and feline patients with naturally occurringwounds excellent preclinical models also detract from theiruse in exploratory models. The variability introduced by non-uniform wounding in outbred animals that are not standard-ized for age, body-condition score, exposed to complex anddivergent environmental backgrounds is less desirable in theexploratory research critical to developing new targets. Fur-thermore, although the molecular and cellular tools for use inthe dog and cat have increased exponentially over the lastdecade, they surely are inferior to those available for rodentmodels. These reagents as well as the availability of criticalgenetically altered strains of mice allow for efficiency indefining mechanisms at play in either normal or pathologicresponses and in response to specific interventions.

Finally, further development of specific wound pathologiesin canine and feline patients as accurate translational models isrequired, including defining incidence and characterizing thecomparative aspects of wound healing parameters in similarconditions between humans and cats or dogs.Although the vastmajority of wounds that present to veterinarians are of an acutenature and heal without complication, the types of woundsdescribed herein are not uncommon in veterinary practice andas naturally occurring wounds could provide an opportunity tomore accurately assess treatment strategies prior to initiatingclinical trials in human patients.Although the true incidence ofsuch conditions remains to be defined, it is likely that suchstudies will necessitate multi-institutional collaborations.These may include both academic and private practitionersexperienced and dedicated to the advancement of woundcare in animals such as board-certified veterinary surgeons(http://www.acvs.org) and members of organizations suchas the Veterinary Wound Management Society (http://www.vwms.org). No practice-based research networks whichfocus on wounds currently exist, although a general veterinarymedical database (http://www.vmdb.org) which compilespatient encounter data from nearly all North American veteri-nary colleges has been in existence for nearly 50 years.

Therefore, although the use of cats and dogs as transla-tional models is certainly not appropriate or practical forwidespread use in wound healing research, they may providea superior model for specific investigations. The most efficienttranslational approach to advancing novel wound care thera-pies will work within an intellectual ecosystem in whichcollaborations of investigators apply the most appropriatemodel for the hypothesis being examined with a clear under-standing of the strengths and limitations of models utilized ateach stage. The strengths of naturally occurring small animalmodels, their availability through partnerships with clinicaltrial centers, and the motivation of pet owners to enroll theirpets in such trials may provide a unique opportunity forassessing efficacy and safety of vulnerary agents, while duallyadvancing veterinary and human wound care.

ACKNOWLEDGMENTSNo external funding was received for this perspective review.SWV would like to acknowledge support from NIAMS



(K08AR053945). The authors wish to thank Laura K. S.Parnell for the original idea of this mini-review as well ascritical review in its preparation. This work was presented atthe 2012 annual meeting of the Wound Healing Society inAtlanta GA.

Conflict of Interest: The authors report no conflict ofinterest.

REFERENCES1. Sullivan TP, Eaglestein WH, Davis SC, Mertz P. The pig as a

model of human wound healing. Wound Repair Regen 2001; 9:66–76.

2. Wilmink JM, Nederbragt H, van Weeren PR, Stolk PW, Barn-eveld A. Differences in wound contraction between horses andponies: the in vitro contraction capacity of fibroblasts. EquineVet J 2001; 33: 499–505.

3. Bohling MW, Henderson RA, Swaim SF, Kincaid SA, WrightJC. Cutaneous wound healing in the cat: a macroscopic descrip-tion and comparison with cutaneous wound healing in the dog.Vet Surg 2004; 33: 579–87.

4. Mitsui A, Mathews KG, Linder KE, Kruse MA, Roe SC. Effectsof fascial abrasion, fasciotomy, and fascial excision on cutaneouswound healing in cats. Am J Vet Res 2009; 70: 532–8.

5. Shih B, Bayat A. Genetics of keloid scarring. Arch Dermatol Res2010; 302: 319–39.

6. Zhu KQ, Carrougher GJ, Gibran NS, Isik FF, Engrav LH. Reviewof the female Duroc/Yorkshire pig model of human fibroprolif-erative scarring. Wound Repair Regen 2007; 15: S32–S39.

7. Kimura T. Hairless descendants of Mexican hairless dogs: anexperimental model for studying hypertrophic scars. J CutanMed Surg 2011; 15: 329–39.

8. Emanuelsson P, Kratz G. Characterization of a new in vitro burnwound model. Burns 1997; 23: 32–6.

9. Xu J, Fei Z, Yu Y-M, Xu W, Rhodes A, Tompkins RG, SchulzJT. In vivo rabbit hindquarter model for assessment of regionalburn hypermetabolism. J Appl Physiol 2003; 94: 135–40.

10. Pavletic MM, Trout NJ. Bullet, bite, and burn wounds in dogsand cats. Vet Clin Small Anim 2006; 36: 873–93.

11. Seok J, Warren HS, Cuenca AG, Mindrinos MN, Baker HV, XuW, et al. Inflammation and Host Response to Injury, Large ScaleCollaborative Research Program. Genomic responses in mousemodels poorly mimic human inflammatory diseases. Proc NatlAcad Sci U S A 2013; 110: 3507–12.

12. Flynn AK, Lurie DM. Canine acute radiation dermatitis, asurvey of current management practices in North America. VetComp Oncol 2007; 5: 197–207.

13. Flynn AK, Lurie DM, Ward J, Lewis DT, Marsella R. Theclinical and histopathological effects of prednisone on acuteradiation-induced dermatitis in dogs: a placebo-controlled, ran-domized, double-blind, prospective clinical trial. Vet Dermatol2007; 18: 217–26.

14. Sacks JJ, Kresnow M. Dog bites: still a problem? Injury Prev2008; 14: 296–301.

15. Weiss HB, Friedman D, Coben JH. Incidence of dog bit injuriestreated in emergency departments. JAMA 1998; 279: 51–3.

16. Holmquist L, Elixhauser A. Emergency department visits andinpatient stays involving dog bites, 2008. 1–14. Rockville, MD:Agency for Healthcare Research and Quality, 2010.

17. Davies HD. When your best friend bites: a note on dog and catbites. Can J Infect Disease 2000; 11: 227–9.

18. Ball V, Youngren BN. Emergency management of difficultwounds: part 1. Emerg Med Clin North Am 2007; 25: 101–21.

19. Shamir MH, Leisner S, Klement E, Gonen E, Johnston DE. Dogbite wounds in dogs and cats: a retrospective study of 196 cases.J Vet Med A Physiol Pathol Clin Med 2002; 49: 107–12.

20. Boat BW, Dixon CA, Pearl E, Thieken L, Bucher SE. Pediatricdog bite victims: a need for a continuum of care. Clin Pediatr2012; 51: 473–7.

21. Freshwater A. Why your housecat’s trite little bite could causeyou quite a fright: a study of domestic felines on the occurrenceand antibiotic susceptibility of Pasteurella multocida. ZoonosesPublic Health 2013; 55: 507–13.

22. Abrahamian FM, Goldstein EJC. Microbiology of animal bitewound infections. Clin Microbiol Rev 2011; 24: 231–46.

23. Bramble M, Morris D, Tolomeo P, Lautenbach E. Potential roleof pet animals in household transmission of Methicillin-resistantStaphylococcus aureus: a narrative review. Vector-borne Zoonot2011; 11: 617–20.

24. Fullington RJ, Otto CM. Characteristics and management ofgunshot wound in dogs and cats: 84 cases (1986–1995). JAVMA1997; 210: 658–62.

25. Anderson DM, White RAS. Ischemic bandage injuries: acase series and review of the literature. Vet Surg 2000; 29: 488–98.

26. Nicholson M, Beal M, Shofer F, Brown DC. Epidemiologicevaluation of postoperative wound infection in clean-contaminated wounds: a retrospective study of 239 dogs andcats. Vet Surg 2002; 31: 577–81.

27. Arthur JJ, Kleiter MM, Thrall DE, Pruitt AF. Characterization ofnormal tissue complications in 51 dogs undergoing definitivepelvic region irradiation. Vet Radiol Ultrasound 2008; 49: 85–9.

28. Eugster S, Schawalder P, Gaschen F, Boerlin P. A prospectivestudy of postoperative surgical site infections in dogs and cats.Vet Surg 2004; 33: 542–50.

29. Vasseur PB, Levy J, Dowd E, Eliot J. Surgical wound infectionrates in dogs and cats. Data from a teaching hospital. Vet Surg1988; 17: 60–4.

30. Mayhew PD, Freeman L, Kwan T, Brown DC. Comparison ofsurgical site infection rates in clean and clean-contaminatedwounds in dogs and cats after minimally invasive vs opensurgery: 179 cases (2007–2008). J Am Vet Med Assoc 2012; 240:193–8.

31. Brown DC, Conzemius MG, Shofer F, Swann H. Epidemiologicevaluation of postoperative wound infections in dogs and cats. JAm Vet Med Assoc 1997; 210: 1302–6.

32. Heldmann E, Brown DC, Shofer F. The association of propofolusage with postoperative wound infection rate in clean wounds:a retrospective study. Vet Surg 1999; 28: 256–9.

33. Beal MW, Brown DC, Shofer F. The effects of perioperativehypothermia and the duration of anesthesia on postoperativewound infection rate in clean wounds: a retrospective study. VetSurg 2000; 29: 123–7.

34. Bergström A, Gustafsson C, Leander M, Fredriksson M, Grön-lund U, Trowald-Wigh G. Occurence of methicillin-resistantStaphylococci in surgically treated dogs and the environment ina Swedish animal hospital. J Small Anim Pract 2012; 53: 404–10.

35. Weese JS. A review of post-operative infections in veterinaryorthopaedic surgery. Vet Comp Orthop Traumatol 2008; 21:99–105.

36. Swaim SF, Bradley DM, Vaughn DM, Powers RD, Hoffman CE.The Greyhound Dog as a model for studying pressure ulcers.Decubitus 1993; 6: 38–40.

37. Swaim SF, Angarano DW. Chronic problem wounds of doglimbs. Clin Dermatol 1990; 8: 175–86.



38. Bohling MW, Henderson RA, Swaim SF, Kincaid SA, WrightJC. Comparison of the role of the subcutaneous tissues in cuta-neous wound healing in the dog and cat. Vet Surg 2006; 35: 3–14.

39. Brinkley CH. Successful closure of feline axillary wounds byreconstruction of the elbow skin fold. J Small Anim Pract 2007;48: 111–5.

40. Brockman DJ, Pardo AD, Conzemius MG, Cabell LM, Trout NJ.Omentum-enhanced reconstruction of chronic nonhealingwounds in cats: techniques and clinical use. Vet Surg 1996; 25:99–104.

41. Lascelles BDX, Davison L, Dunning M, Bray JP, White RAS.Use of omental pedicle grafts in the management of non-healingaxillary wounds in 10 cats. J Small Anim Pract 1998; 39: 480.

42. Gray MJ. Chronic axillary wound repair in a cat with omentali-sation and omocervical skin flap. J Small Anim Pract 2005; 46:499–503.

43. Swaim SF, Gillette RL, Sartin EA, Hinkle SH, Coolman SL.Effects of a hydrolyzed collagen dressing on the healing of openwounds in dogs. Am J Vet Res 2000; 61: 1574–8.

44. Swaim SF, Bradley DM, Spano JS, McGuire JA, Hoffman CE,Trachy RE. Evaluation of multipeptide-copper complex medica-tions on open wound healing in dogs. J Am Anim Hosp Assoc1993; 29: 519–25.

45. Gillette RL, Swaim SF, Sartin EA, Bradley DM, Coolman SL.Effects of bioactive glass on healing of closed skin wounds indogs. Am J Vet Res 2001; 62: 1149–53.

46. Winkler JT, Swaim SF, Sartin EA, Henderson RA, Welch JA.The effect of a porcine-derived small intestinal submucosaproduct on wounds with exposed bone in dogs. Vet Surg 2002;31: 541–51.

47. Scardino MS, Swaim SF, Sartin EA, Steiss JE, Spano JS,Hoffman CE, Coolman SL, Peppin BL. Evaluation of treatmentwith a pulsed electromagnetic field on wound healing, clinico-pathologic variables and central nervous system activity of dogs.Am J Vet Res 1998; 59: 1181.

48. Karayannopoulou M, Tsioli V, Loukopoulos P, Anagnostou TL,Giannakas N, Savvas I, Papazoglou LG, Kaldrymidou E. Evalu-ation of the effectiveness of an ointment based on Alkannins/Shikonins on second intention wound healing in the dog. Can JVet Res 2011; 75: 42–8.

49. Demaria M, Stanley BJ, Hauptman JG, Steficek BA, Fritz MC,Ryan JM, et al. Effects of negative pressure wound therapy onhealing of open wounds in dogs. Vet Surg 2011; 40: 658–69.

50. Guille AE, Tseng LW, Orsher RJ. Use of vacuum-assistedclosure for management of a large skin wound in a cat. J Am VetMed Assoc 2007; 230: 1669–73.

51. Ben-Amotz R, Lanz OI, Miller JM, Filipowicz DE, King MD.The use of vacuum-assisted closure therapy for the treatment ofdistal extremity wounds in 15 dogs. Vet Surg 2007; 36: 684–90.

52. Owen L, Hotson-Moore A, Holt P. Vaccuum-assisted woundclosure following urine-induced skin and thigh muscle necrosisin a cat. Vet Comp Orthop Traumatol 2009; 22: 417–21.

53. Brown DC. Control of selection bias in parallel-group controlledclinical trials in dogs and cats: 97 trials (2000–2005). J Am VetMed Assoc 2006; 229: 990–3.

54. Lundberg C. Inflammatory reaction and collagen accumulationin an experimental model of open wounds in the rat. A compari-son between gauze and Debrisan treatment. Scand J PlastReconstr Surg 1985; 19: 11–6.

55. Sharma C, Dalferes ER, Radhakrishnamurthy B, Rosen EL, Ber-enson GS. Glycoprotein biosynthesis during inflammation innormal and streptozotocin-induced diabetic rats. Inflammation1985; 9: 273–83.

56. Bullard KM, Lund L, Mudgett JS, Mellin TN, Hunt TK, MurphyB, et al. Impaired wound contraction in Stromelysin-1-deficientmice. Ann Surg 1999; 230: 260–5.

57. Hayati F, Maleki M, Pourmohammad M, Sardari K, Mohri M,Afkhami A. Influence of short-term, repeated fasting on the skinwound healing of female mice. Wounds 2011; 23: 38–43.

58. Karr BP, Bubak PJ, Sprugel KH, Pavlin EG, Engray LH. Plateletderived growth factor and wound contraction in the rat. J SurgRes 1995; 59: 739–42.

59. McGrath MH, Simon RH. Wound geometry and the kinetics ofwound contraction. Plast Reconstr Surg 1983; 72: 66–72.

60. Rudolph R. Location of the force of wound contraction. SurgGynecol Obstet 1979; 148: 547–51.

61. Holt DR, Kirk SJ, Regan MC, Hurson M, Lindblad WJ, BarbulA. Effect of age on wound healing in healthy human beings.Surgery 1992; 112: 293–7.

62. Tranquillo RT, Murray JD. Mechanistic model of wound con-traction. J Surg Res 1993; 55: 233–47.

63. Olsen L, Sherrat JA, Maini PK. Mechanochemical model foradult dermal wound contraction and the permanence of the con-tracted tissue displacement profile. J Theor Biol 1995; 177:113–28.

64. Rothwell S, Sawyer E, Dorsey J, Floumoy WS, Settle T,Simpson D, et al. Wound healing and the immune response inswine treated with a hemostatic bandage composed of salmonthrobin and fibronogen. J Mater Sci Mater Med 2009; 20: 2155–66.

65. Yao F, Visovatii S, Johnson CS, Chen M, Slama J, Wenger A,Eriksson E. Age and growth factors in porcine full-thicknesswound healing. Wound Repair Regen 2001; 9: 371–7.

66. Wilmink JM, Stolk PW, van Weeren PR, Barneveld A. Differ-ences in second-intention wound healing between horses andponies: histological aspects. Equine Vet J 1999; 31: 61–7.

67. Lepault E, Céleste C, Doré M, Martineau D, Theoret C. Com-parative study on microvascular occlusion and apoptosis in bodyand limb wounds in the horse. Wound Repair Regen 2005; 13:520–9.

68. Wilmink JM, van Weeren PR, Barneveld A. Differences insecond-intention wound healing between horses and ponies:macroscopic aspects. Equine Vet J 1999; 31: 53–60.

69. Wilmink JM, Veenman JN, van den Boom R, Rutten VP,Niewold TA, Broekhuisen-Davies JM, et al. Differences in poly-morphonucleocyte function and local inflammatory responsebetween horses and ponies. Equine Vet J 2003; 35: 561–9.

Supporting InformationAdditional Supporting Information may be found in theonline version of this article at the publisher’s Website:

Figure S1. Severe bite wounds to the hindlimbs andperineum of a Labrador retriever. Crushing, avulsion, anddevitalization of underlying tissue occurs with dog-on-dogbite wounds.

Figure S2. Radiation induced chronic ulcer in a canineoncologist patient. This photograph shows a large nonhealingulcerated lesion over the shoulder (head of the patient istoward the left). This patient had previously received adjunc-tive radiation therapy for treatment of a soft tissue sarcomafollowing excision. This ulcerated lesion was successfullytreated by prestretching of the skin followed by excision ofthe lesion and primary closure. Photo courtesy of Dr. LillianAronson.



Figure S3. Feline chronic axillary wounds. (A) A largenonhealing wound is seen in the right axilla of a cat. Thispatient had been found as a stray more than 1 month prior andtherefore the inciting trauma is unknown. Over the course ofthe month, the general appearance remained unchanged. Onpresentation, the wound displays poorly vascularized granu-

lation tissue covering its surface and showed little to no evi-dence of contraction and epithelialization. (B) A chronicaxillary wound of 1.5 years duration in a cat adopted from ananimal shelter. At the time of rescue by the shelter, the cat wasfound with a flea collar embedded in the left axillary region.Photos courtesy of Dr. David Holt.



Documents

Comparative wound healing-Are the small animal veterinarian's clinical patients an improved translational model for human wound healing research?