Disorders of Visceral Healing

The small-animal surgeon routinely creates wounds in the gastrointestinal tract for biopsy, for foreign body or neoplasm removal, or for correction of strangulation obstruction. The urinary bladder wall is likewise wounded to remove calculi or to resect bladder wall tumors. The uterus is incised during cesarean section or removed in ovariohysterectomy, electively or when necessitated by infection. Although dehiscence of a skin wound is often easily remedied with appropriate local wound treatment, dehiscence of a hollow viscus is usually life threatening. Wound dehiscence of the digestive system often leads to generalized bacterial peritonitis and subsequent death. Wound dehiscence of the uterus can have the same outcome if pre-existing intrauterine sepsis is present. Leakage of cystotomy closure may also lead to uroperitoneum, uremia, hyperkalemia, and ultimate death as well. Therefore, factors that negatively affect visceral healing are potentially of great clinical significance to the surgeon.

Normal Wound Healing of the Viscera

The gastrointestinal, urinary, and reproductive tracts follow the same basic healing curve as the skin but have accelerated healing properties. The lag or inflammatory phase of healing lasts 3 to 4 days [1]. Immediately after wounding, blood vessels contract, platelets aggregate, the coagulation mechanism is activated, and fibrin clots are deposited to control hemorrhage. The fibrin clot offers some minimal wound strength on the first postoperative day, but the main wound support during the lag phase of healing comes from the sutures [2]. Enterocyte and uroepithelial regeneration begins almost immediately after wounding; however, the epithelium offers little biomechanical support [2,3]. The lag phase is the most critical period during visceral wound healing, because most dehiscences take place within 72 to 96 hours after the wound has been created [2].

The proliferative or logarithmic phase of visceral wound healing lasts from day 3 or 4 through day 14 [1]. Rapid proliferation of fibroblasts occurs logarithmically during this period. The fibroblasts produce large amounts of immature collagen, resulting in rapid gains in wound strength. The proliferative phase of wound healing is a dynamic process in which collagen synthesis takes place in the presence of collagenolysis. In the stomach, small intestine, and urinary tract, collagenase activity at the wound edge is minimal and rapid gains in tensile and bursting strength occur. At the end of 14 days, gastric and small-intestinal wound bursting strength is approximately 75% that of normal tissue. The urinary bladder heals even faster, regaining 100% of normal tissue strength at 14 to 21 days [3]. Conversely, the colon heals much more slowly owing to marked collagenase activity at the wound edge and regains only about 50% of its normal tensile strength 14 days after wounding [1]. Factors such as traumatic suturing, fecal material, bacterial contamination, and infection all increase the amount of local collagenase produced at the wound edge [2].

The maturation phase of wound healing is characterized by reorganization and cross-linking of collagen fibers. This phase extends from day 14 through day 180 in the gastrointestinal tract of the dog [1], and from day 14 through day 70 in the dog bladder [3]. As with skin wounds, the size and thickness of the scar decrease during this time without weakening the wound. The maturation phase is relatively unimportant clinically in visceral wound healing, because acceptable tensile and bursting strength have been established by the end of the proliferative phase of wound healing and leakage is virtually nonexistent at this point [2].

Factors that negatively affect Visceral Wound Healing

Nutritional Depletion

Tissue trauma, sepsis, burns, and major surgery induce major metabolic changes in small-animal patients. With each of these stresses, the animal's basic metabolic rate is accelerated and protein metabolism occurs, leading to a potential state of negative nitrogen balance. Protein-calorie malnutrition (PCM) occurs because of starvation, when a metabolic response to injury becomes prolonged, or when a hypermetabolic state occurs such as that caused by sepsis. It takes only 5 to 10 days of anorexia to compromise the immune system and deplete the body's muscular and hepatic glycogen stores [4]. When PCM is present, cell-mediated immunity is impaired, and there is a high risk of infection, anemia, and hypoproteinemia. Wound healing may also be impaired.

Caloric and protein depletion in experimental animals has been shown to inhibit dermal, muscular-fascial, and visceral healing, but only after a loss of 15% to 20% of body weight [5]. Decreases in wound breaking strength are directly proportional to the carcass weight loss. It is estimated that 75% of animals with elective surgical wounds attain functional wound union during the period of negative nitrogen balance [6]; however, extended PCM from muscle, visceral, or plasma tissue losses increases the risk for visceral wound disruption. Impaired visceral wound healing is due both to a prolonged lag phase of healing and to diminished capacity for fibroplasia within the logarithmic phase [7].

Effect of Early Postoperative Enteral Feeding on Visceral Healing

Malnutrition induces intestinal mucosal atrophy, reduced motility, increased incidence of ileus, and the potential for bacterial translocation through the bowel wall, with resultant sepsis. Impaired wound healing owing to nutritional causes may be ameliorated by feeding an enteral or parenteral diet that supplies energy needs in the form of fatty acids and sugars and provides essential amino acids [4,7]. Feedings of high-protein meals after injury can optimize conditions for normal visceral wound healing. Amino acids provided through enteral nutrition are utilized for the synthesis of hexosamines, proteoglycan polymers, nucleic acids, and structural proteins such as actin, myosin, collagen, and elastin [7].

Early, if not immediate, postoperative enteral feeding has been shown to have a positive influence on the healing rate of intestinal anastomosis in dogs. Bursting pressures and collagen levels of ileal and colorectal anastomosis were compared in Beagles fed elemental diets versus those fed only electrolyte and water for four days. The dogs fed elemental diets had nearly twice the bursting strengths of the control group and nearly double the amount of both immature and mature collagen at the wound site [8]. Total parenteral nutrition (TPN) does not appear to ameliorate the mucosal atrophy or increase collagen deposition as does enteral nutrition. In human studies, the incidence of septic complications was significantly lower in people fed between 8 to 24 hours after surgery versus those maintained on TPN. Additionally, those patients fed early had a reduced incidence of postoperative ileus and reduced hospital stay [9].

Anemia

Small-animal patients with polytrauma, major burns, or malignancies are frequently anemic. Studies of experimental anemia in rats induced by phlebotomy or iron-deficient diet have shown suppression in wound healing only if the animals' blood volume remained deficient or significant malnutrition accompanied the iron-deficient diet [5]. With anemia resulting from phlebotomy, intravascular volume replacement results in normal wound healing [10]. Therefore, anemia in the absence of concurrent malnutrition or volume deficit does not appear to be a factor in suppressed visceral wound healing.

Leukopenia

Neutropenia caused by disease processes or induced by chemotherapy has been suspected of causing impaired wound healing. Contaminated wounds are more likely to become infected in the presence of neutropenia, but minimal problems are encountered with clean wounds. Neutropenic animals have been shown to have a reduced number of neutrophils at the wound site during the inflammatory phase of wound healing. However, normal progression of wound debridement occurs, owing to the presence of macrophages. When neutropenic rats are compared with controls, fibroplasia is unaffected, collagen deposition is unchanged, and wound strength is normal [10].

Lymphopenia also fails to affect wound healing in rats. A normal inflammatory response is seen during the first 24 to 48 hours of the lag phase of healing, and normal wound strength and collagen content are found at 7 and 14 days [10]. Thus, leukopenia owing to either neutropenia or lymphopenia alone fails to suppress wound healing as assessed by histologic and biochemical techniques.

The macrophage plays a vital role in wound healing. Systemic hydrocortisone-induced monocytopenia can reduce tissue macrophage levels to approximately one third that of controls. Mild inhibition of wound debridement is observed, but collagen synthesis is unaffected. However, when antimacrophage serum and hydrocortisone are combined, clearance of wound debris is much decreased and fibroplasia and collagen synthesis are reduced. Macrophages are also important in enhancing neovascularization at the wound edge [11]. Activated macrophages are associated with a higher frequency of neovascularization. Injection of activated peritoneal macrophages into dermis and subcutaneous tissues of rats immediately before wounding increases wound breaking strength at 8 days [11].

Corticosteroids

The anti-inflammatory action of glucocorticoid hormones may prolong the inflammatory phase of wound healing. Specific mechanisms include the stabilization of lysosomal membranes, mobilization of neutrophils, decreased local phagocytosis, and inhibition of deoxyribonucleic acid (DNA) synthesis [10]. Hydrocortisone and methylprednisolone have a greater anti-inflammatory affect than does dexamethasone. Corticosteroids are also associated with increased risk of wound infection. They interfere with the logarithmic phase of healing by delaying fibroblast proliferation and collagen synthesis and crosslinking.

The negative effect on wound healing is critically dependent on the dose and timing of steroid administration. If glucocorticoids are started 3 days or more after creation of the wound, progression of cellular infiltration and resulting fibroplasia are histologically unchanged and produce no negative effect on wound tensile strength at 7 days [10]. Impairment of wound healing by corticosteroids is most obvious at 7 to 14 days, owing to their negative effect on fibroplasia during the logarithmic phase. However, by 3 to 4 weeks after injury the wound tensile strength in animals receiving corticosteroid treatment approaches that of control animals.

Dose is also an important factor. In rats, small doses (5 mg/kg) of hydrocortisone have no adverse effect on wound tensile strength, whereas larger doses significantly decrease it. The nutritional state of the animal is another important variable. Hydrocortisone at a dose of 5 mg/kg had no effect on the tensile strength of incised wounds in rabbits fed a regular diet but produced a marked inhibitory effect in rabbits that were 25% underweight. Along with the healing effects on the gastrointestinal tract, corticosteroid-induced gastric and colonic perforations are well documented in dogs being treated for concurrent spinal cord disease [10].

It is possible that anabolic steroids may counteract the negative effects of glucocorticoids on wound healing. In one study, rats given 10 mg hydrocortisone for eight days had significantly weaker dermal wounds. However, rats supplemented with 5 mg testosterone propionate or nandrolone phenpropionate had similar wound strength to that in control animals [12]. In other studies, lower doses of testosterone did not reverse the detrimental effects of cortisone in rats five days after wounding [12].

Nonsteroidal Anti-Inflammatory Drugs

Prostaglandins (PG) have been implicated as a major mediator in wound healing, particularly during the inflammatory phase. The PG-induced inflammatory response proceeds via lipid mediators metabolized from arachidonic acid through the cyclooxygenase pathway. Corticosteroids are thought to inhibit phospholipase A2 activity in cell membranes, resulting in reduced release of prostaglandin precursors (arachidonic acid). Nonsteroidal anti-inflammatory drugs (NSAID), on the other hand, inhibit the action of cyclooxygenase in the prostaglandin-formation cascade. Excessive doses of older NSAIDs such as aspirin, indomethacin, and phenylbutazone have been shown to significantly delay increases in skin wound strength in animals during the inflammatory phase. Flunixin meglumine was evaluated for its effects on dermal, fascial, gastric, and colonic wound healing in rats. Flunixin, 1.1 mg/kg b.i.d., significantly decreased the tensile strength of wounds in the skin and linea alba, but it did not affect visceral bursting strength at five days after surgery. At 14 days after surgery, no significant difference in wound strength was seen between the group given flunixin and control groups [13]. Other than liver toxicity, the newer Cox-2-inhibiting NSAIDs such as carprofen or deracoxib do not appear to negatively affect alimentary healing and may, in fact, reduce postoperative ileus by their analgesic and anti-inflammatory properties. The newer NSAIDs may work by reducing the release of cytokines (TNF, IL-1, IL-6, and IL-8) during the inflammatory stage of wound repair [14]. They probably do not affect the proliferative phase of visceral healing when fibroplasia is the factor influencing gain in visceral wound strength [13].

Radiation

Radiation has a negative effect on wound healing for several reasons. Injury to the chromatin of dividing fibroblasts may decrease their accumulation in the wound. Collagen production is reduced in most irradiated wounds. Progressive fibrosis of blood vessels results in decreased blood flow and lower oxygen tension after radiation therapy. This delayed or impaired perfusion delays wound healing and also increases the risk of infection [15].

Confusion exists about when to initiate radiation therapy in the perioperative period. Initiation of radiation therapy prior to or at the time of surgery may increase chances of dermal or visceral wound dehiscence. In one study of rats, a single dose of 2000 rad was given at seven days before, on the day of skin wounding, and at seven days thereafter. The wounds that were irradiated one week before surgery and at the time of surgery were significantly weaker than those irradiated one week after surgery. Because radiation interferes principally with the logarithmic proliferative phase of wound healing, it would be ideal to initiate radiation a minimum of 7 to 14 days after visceral wounding has occurred [10]. When suturing previously irradiated viscera, a non-absorbable suture material or slowly absorbable synthetic monofilament absorbable suture should be chosen. Irradiated bladder wounds sutured with polyglycolic acid were stronger than those sutured with chromic catgut, but no difference was noted in non-irradiated bladders [15].

Effects of Cancer on Wound Healing

Cancer has been incriminated in delayed wound healing, but impaired nutrition is probably the underlying cause. Except as a result of cancer cachexia, adverse wound healing owing to cancer is difficult to document. The presence of macroscopic tumor remote from the wound or microscopic tumor within the wound does not appear to impair healing [10]. Humans with esophageal cancer who have residual microscopic tumor at the surgical margins did not have a significantly higher rate of leakage at the anastomosis than did patients with no tumor present [16]. No evidence exists that the presence of tumor directly impedes healing; in selected cases, it may actually accelerate healing. In excisional biopsy for invasive cutaneous melanoma, those portions of the incision that contained residual tumor were stronger than those that did not. Further evidence indicates that presence of a tumor may have a positive effect on the wound strength of distant surgical wounds. Wounds were stronger in rats with hepatomas than in control rats. Increased wound strength in the presence of tumor may be due to the presence of growth factors that act as fibroblast mitogens. Other tumors such as mast cell tumors, which release vasoactive substances such as histamine, may retard wound healing [17].

Effects of Chemotherapy

Chemotherapeutic agents affect wound healing in several ways. Neutropenia, a common side effect of chemotherapy, increases the risk of infection but does not interfere directly with wound healing. Interference with the inflammatory stage of wound healing is often not seen unless therapy is begun preoperatively [10]. Many chemotherapeutic agents exert their antineoplastic effect by interfering with DNA replication, RNA production, protein synthesis, or cell division [12]. Inhibition of fibroblast proliferation, reduced formation of collagen, and impaired neovascularization have been documented subsequent to chemotherapy in animals. Therefore, the greatest potential for impairment of wound healing would be during the proliferative or logarithmic phase, when fibroblasts are dividing and are metabolically active. In addition, anorexia and cachexia are common side effects of chemotherapy and may have an additive adverse effect with preexisting cancer cachexia. This can exacerbate negative nitrogen balance and PCM. Similar to corticosteroid therapy, the effects of chemotherapy are largely affected by time of administration and dose of the drug. The relative effects of various antineoplastic agents on wound healing are listed in Table 37-1.

Vincristine

Vincristine is commonly used to treat lymphosarcoma and other carcinomas in animals. The drug exerts its effect by binding with intracellular microtubular systems, causing mitotic arrest [10,12]. Impairment of dermal wound healing by this agent appears to be minimal. In mice treated with vincristine, mildly reduced dermal wound strength was seen 3 days after surgery, but not at 7 or 21 days thereafter.18 Clinical trials in humans using preoperative vincristine do not indicate increased incidence of wound dehiscence or morbidity.

Vinblastine

Vinblastine is another plant alkyloid used to treat mast cell tumors and in some lymphoma protocols. This compound acts by binding to microtubules in the mitotic spindle, thereby preventing cell division. There is no direct association with disruption of wound healing, but gastroenterocolitis and myelosuppression are common [19].

Doxorubicin

Doxorubicin (Adriamycin), an antitumor antibiotic, acts by inhibiting DNA and RNA synthesis. Doxorubicin inhibits cell division during the proliferative phase of wound healing. Doxorubicin given to rats reduced wound breaking strength at 5, 10, 15, 20, and 30 days after wounding when compared with controls [20]. In addition, doxorubicin impaired wound healing when it was given up to 5 weeks prior to wounding or as much as 4 weeks after wounding. The effects of doxorubicin on wound healing are dose dependent. Animals given high levels of the drug show significantly impaired wound healing, whereas those given normal therapeutic doses show little if any effect on wound healing [17].

Cyclophosphamide

Cyclophosphamide is an alkylating agent that acts by cross-linking DNA strands and, thereby, preventing cell division. Cyclophosphamide interferes with fibroblast production and collagen production and also decreases neovascularization [10]. Experimental studies in animals indicate that cyclophosphamide significantly impedes healing at therapeutic doses. Rats given cyclophosphamide at the time of wounding had decreased strength of skin wounds at 1, 3, and 5 weeks after wounding. Larger doses of cyclophosphamide have also been shown to delay scar maturation in rats. However, in clinical studies of human patients, no adverse effect on visceral wound healing was observed, regardless of the time of administration of cyclophosphamide [18].

Methotrexate

Methotrexate is an antimetabolite folic acid antagonist that prevents DNA maturation. High doses of methotrexate are associated with increased rates of wound infection in mice [21]. The effect is dose-dependent and is reversible with concurrent folic acid administration. Wound breaking strength is decreased and rate of wound healing was also retarded for as long as 21 days after wounding. However, these effects are not observed if methotrexate is given in therapeutic doses at the time of surgery. Methotrexate given to humans at the time of surgery increases the rate of wound dehiscence. This complication is minimized by delaying chemotherapy 10 to 14 days or by concurrently administering folic acid. In mice, intraperitoneal methotrexate given in therapeutic doses causes decreased wound strength on day three but not on day 7 or 21 [10].

5-Fluorouracil

5-Fluorouracil (5-FU) has been used extensively in small animal patients with carcinomas and in humans with intestinal carcinomas. Because it is commonly given systemically or intraperitoneally after resection of carcinomas, its effect on intestinal wound healing has been studied directly but results have been conflicting. No impairment of healing of colon anastomoses occurred in rats when chemotherapy was given at operation and at 4, 7, and 11 days postoperatively [22]. However, rats given 5-FU alone or 5-FU in combination with levamisole were noted to have bursting strengths of healing colon anastomoses only 65% of that seen in the control group. In vitro studies showed that both of these agents significantly inhibit fibroblast proliferation. These effects were more pronounced in animals with weight loss or nutritional deficiencies [23]; however, the effect was dose related, and mice given therapeutic doses had minimal adverse effects. Although 5-FU is related to methotrexate in mode of action, the apparent risk for impaired visceral wound healing is much less with 5-FU than it is with methotrexate when given in therapeutic doses [10].

Cisplatin and Carboplatin

Cisplatin, a heavy metal compound, has been used for various carcinomas including transitional cell carcinomas in dogs. Its effects on visceral wound healing have been evaluated in rats. Large- and small-bowel anastomoses were performed and compared at 4, 7, 14, and 28 days postoperatively [24]. Cisplatin was given in doses of 5 mg/kg 1 and 5 days preoperatively. At four days after surgery, wound breaking strength of both the large- and small-bowel anastomosis were too low to measure [10,27]. At all times during the study, wound breaking strength of treated animals was below that of controls. From these results and from studies in humans it appears that cisplatin is one of the most dangerous chemotherapeutic agents with regard to the potential for interfering with visceral wound healing. Carboplatin is a less toxic cousin to cisplatin that can be used in cats. Gastrointestinal irritation with this drug is reduced when compared with cisplatin, but the mechanism of action is the same and so caution should be exercised when using this drug concurrently with visceral surgery [25].

Newer Chemotherapeutic Agents Requiring Further Study

Lomustine

This alkylating agent is used primarily in the treatment of canine mast cell tumor. No mention has been made of adverse wound healing, but vomiting, diarrhea, and stomatitis are common side effects [25].

Mechlorethamine HCL

This alkylating agent is used as a rescue drug for lymphoma as well as for intracavitary administration of some neoplasms. It has significant gastrointestinal effects, which can lead to severe vomiting and diarrhea and may halt therapy [25].

Mitoxantrone

This synthetic antitumor antibiotic has been used with some success in the treatment of lymphoma and transitional cell carcinoma, and as a radiation sensitizer against squamous cell carcinoma. Its side effects include myelosuppression, vomiting, and diarrhea [25].

Conclusion

Healing of visceral wounds is negatively affected by a variety of factors. Chronic weight loss of 15% to 20% owing to cancer cachexia or other reasons has a negative effect on visceral wound healing. Correction of cachexia as well as early postoperative enteral feeding appears to increase collagen deposition and bursting wound strength. Glucocorticoids have a negative effect on wound healing when given in large doses prior to the third day after wounding. NSAIDs appear to affect the early inflammatory phase of wound healing, but do not appear to interfere with the proliferative phase of wound healing or have a significant negative effect on visceral healing strength. Radiation therapy interferes with fibroblast mobilization, replication, and collagen synthesis. It also causes sclerosis of the microvasculature, thereby reducing oxygenation at the wound site. Whenever possible, radiation therapy should be initiated after visceral wound healing is complete. The negative effects of cancer on wound healing appear to be secondary to nutritional deficiencies rather than direct tumor impairment of wound healing. Visceral wound healing may actually be mildly augmented owing to release of growth factors by the neoplasm. Effects of chemotherapeutic agents on visceral wound healing are variable. Drugs such as vincristine, vinblastine, and azathioprine seem to be safe when used in therapeutic doses. Drugs such as cyclophosphamide, methotrexate, 5-FU, and doxorubicin have been shown to delay wound healing in both experimental and clinical studies. Cisplatin appears to significantly impair intestinal wound healing in rats and should be used with caution after intestinal surgery. Newer antineoplastic drugs require further evaluation.