Traumatic Body Wall Herniation in 36 Dogs and Cats
Traumatic body wall hernias (TBWH) are serious sequelae to traumatic injury in dogs and cats. During the study period, 26 dogs and 10 cats with surgically managed TBWH were identified. Five cases (four dogs, one cat) did not have their hernias identified during the first 24 hours of hospitalization. Bite wounds were the most common cause of TBWH, accounting for 54% of canine and 40% of feline hernias. Twelve cases (nine dogs, three cats) had serious intra-abdominal injuries diagnosed in addition to their TBWH. Seventy-three percent of dogs and 80% of cats survived to hospital discharge. In addition, the authors report the occurrence of a unique cause of herniation, termed an autopenetrating hernia.
Introduction
A hernia is defined as “the protrusion of an organ or part of an organ through a defect in the wall of the anatomical cavity in which it lies.”1 True hernias are abdominal wall defects that have anatomically defined hernial rings. They usually have a complete sac of peritoneum surrounding the hernial contents2 and frequently occur at the umbilicus, inguinal canal, and femoral canal. True hernias occur due to congenital and, more rarely, traumatic etiologies.23 Herniation also can occur by traumatic disruption of the body wall at sites other than anatomical apertures.2 These hernias are classified as false hernias since they lack a hernial sac.12
Traumatic body wall hernias (TBWH) are protrusions of abdominal or thoracic contents through a trauma-induced defect in the abdominal or thoracic body wall.245 They are classified according to their anatomical location (i.e., paracostal, lateral, inguinal, femoral, prepubic, ventral).3 Blunt trauma has been reported to be the most common cause of traumatic herniation.126 Rarely, congenital agenesis or hypoplasia of the abdominal wall may cause similar abdominal hernias.26 Traumatic body wall hernias have been described relatively infrequently in dogs and cats.5–10 In one large-scale review of 600 dogs with vehicular trauma, only two cases of TBWH were reported, and a study of 132 cats with high-rise syndrome found just two cats with a TBWH.78
The pathophysiology of TBWH has been postulated to involve a considerable blunt force that is distributed over a large surface area. This area is large enough to prevent penetration of the relatively elastic skin, yet it is able to disrupt deeper, less elastic tissues such as muscle and fascia.11 Traumatic body wall hernias may also result from shearing forces that are distributed across the bony prominences of the pelvis or the caudal thoracic cage. These forces result in the shearing or tearing of muscle or tendon from its bony attachment.5
Previous reports in the veterinary literature focused on the surgical correction, concurrent injuries, and potential surgical complications of traumatic abdominal wall hernias.6910 The purpose of this retrospective study was to evaluate the clinical signs, surgical findings, concurrent injuries, postoperative complications, perioperative care, and outcome in dogs and cats with surgically corrected TBWH. In addition, the authors describe a unique cause of herniation, termed an autopenetrating hernia, which to the authors’ knowledge has not been described previously in animals.
Materials and Methods
The small animal surgery log, maintained on all dogs and cats undergoing surgical procedures at the Foster Hospital for Small Animals, Tufts New England Veterinary Medical Center (FHSA-TNEVMC), between January 1, 1997 and December 31, 1999, was retrospectively evaluated for the surgical repair of a hernia. The medical records of patients that underwent hernia repair were retrieved, and nontraumatic body wall hernias were excluded (e.g., isolated diaphragmatic, perineal, incisional, and congenital umbilical hernias). Those cases with a TBWH were further assessed. The following information was recorded: signalment, cause of trauma, weight, and initial temperature, pulse, respiratory rate, packed cell volume (PCV), and total serum protein (TS). In addition, physical examination findings including detection of a TBWH, radiographic findings, and whether or not the TBWH was detected within 24 hours of presentation were recorded. Surgical findings including the location of the hernia, presence of other significant injuries, postoperative complications including the occurrence of postoperative infections (defined as wounds or incisions judged to be infected by the attending clinician, for which positive culture results were obtained), anesthetic and analgesic protocols, length of stay, and outcome were also recorded. Location of the hernia was classified as paracostal, lateral, inguinal, femoral, prepubic, or ventral according to a prior classification used by Waldron, et al.6 In some cases, it was not possible to distinguish between inguinal and femoral hernias due to ambiguity in the medical record. As a result, these were tabulated together, and all hernias in a femoral or inguinal location were considered to be femoral hernias. The dose of crystalloid and colloid fluids administered in the first 24 hours was calculated and reported as milliliters per kilogram. The use of blood products and antibiotics was recorded. An animal trauma triage (ATT) score was retrospectively calculated from each patient’s medical record.12
The medical record was examined to determine if the hernia could be classified as an autopenetrating hernia. An autopenetrating hernia, as described in humans, occurs when a blunt traumatic force results in a bony fracture. The jagged bone is driven through the layers of the abdominal wall, resulting in a traumatic hernia.13
Unless otherwise noted, the results are reported as mean±standard deviation (SD). An independent t-test was used to compare two-group continuous variables, and Fisher’s exact test was used to evaluate nonparametric variables. A P value of <0.05 was considered significant. Survival was defined as a patient being alive at the time of hospital discharge. All animals that were euthanized were considered nonsurvivors for the purpose of statistical analysis. Data from dogs and cats was analyzed separately, unless otherwise indicated.
Results
During the study period, 26 dogs and 10 cats underwent surgical correction of a TBWH [see Table]. One dog (case no. 6) presented 9 years after the initial trauma and was excluded from analysis except for the outcome, location, and contents of the hernia. During the study period, there were approximately 59,000 admissions and 7,500 surgical procedures performed at the FHSA-TNEVMC. Traumatic body wall hernia repair was performed in approximately 0.5% of all small animal surgical cases during this time period.
Dogs
The mean age was 7.6±4.2 years (range, 8 months to 14 years). Eighteen breeds of dogs were represented, with the most common being two (8%) each of Chihuahua, Shetland sheepdog, toy poodle, and Yorkshire terrier. Five (19%) were mixed-breed dogs. Fifteen (58%) of the dogs were female (nine spayed, six intact), and 11 (42%) were male (five neutered, six intact). Mean weight was 13.7±12.2 kg (range, 3.6 to 40.9 kg).
Fourteen (54%) of TBWH were the result of bite wounds from another dog, 10 (38%) resulted from vehicular trauma, one (4%) resulted from being kicked by a horse, and the cause of trauma was unknown in one case (4%). The mean heart rate at presentation was 146±35 beats per minute (bpm; range, 84 to 230 bpm); the mean respiratory rate was 40±17 breaths per minute (breaths pm; range, 22 to 80 breaths pm); and the mean temperature was 100.7°F±1.6° (range, 96° to 102.4°F). The mean PCV and TS were 42%±10 (range, 23% to 52%) and 5.4±1.2 gm/dL (range, 2.0 to 7.4 gm/dL), respectively.
A TBWH was identified during the initial physical examination in 17 (65%) of the cases and was suspected in an additional three (12%). The TBWH was identified radiographically in 11 (42%) of the 23 cases where the appropriate area was imaged. Radiographic findings were equivocal in an additional three (13%) of the cases.
Four (15%) cases did not have their TBWH identified until >24 hours after admission. One dog (case no. 13) had a tibial fracture and sacroiliac luxations repaired 2 days before the hernia was identified. Another case (case no. 21) had an open reduction of a hip luxation 1 day before the TBWH was identified. The remaining two cases (case nos. 8 and 22) both suffered severe polytrauma and did not have their TBWH identified until days 4 and 2 postadmission, respectively.
Surgical Findings.
As criterion for study entry, all 26 dogs underwent surgical correction of their TBWH. Fourteen (54%) of the hernias were located in the lateral paralumbar body wall; four (15%) resulted from cranial pubic ligament rupture; four (15%) were intercostal hernias; two (8%) were in a femoral or inguinal location; one (4%) was in a ventral location; and one (4%) was a paracostal hernia. According to the surgical reports in the medical records, herniated organs included intestine (n=13), omentum (n=8), bladder (n=3), lung (n=3), and liver (n=1). Significant additional surgical findings included diaphragmatic herniation (n=2), kidney avulsion (n=1), gastric rupture (n=1), partial mesenteric avulsion (n=1), and splenic laceration (n=1). Herniorrhaphy was performed in 24 cases using autogenous tissue to close the defect. In two dogs, polypropylene mesha was utilized to repair the hernia.
Three (12%) of the dogs required small intestinal resection and anastomosis, and one (4%) additional dog had a colonic resection and anastomosis. Two of the three dogs undergoing small intestinal resection and anastomosis and the one dog with a colonic resection and anastomosis had evidence of partial or full-thickness perforation of the affected bowel. The third dog that required a small intestinal resection and anastomosis had a partial mesenteric avulsion that compromised the blood flow to the affected intestine. Lung lobectomy was performed in three of four animals with intercostal hernias.
Case no. 26 suffered from an autopenetrating hernia caused by a bite from another dog. The force of the trauma had fractured the 12th rib, resulting in the fractured rib segment penetrating the diaphragm, creating a diaphragmatic defect, then penetrating through the paracostal musculature and creating a defect in the abdominal body wall through which a liver lobe herniated.
Associated Injuries.
Any injury that occurred outside of the abdominal or thoracic cavity was considered an associated injury. Fifteen (58%) of 26 dogs suffered associated injuries. Six (23%) dogs had more than one associated injury. Orthopedic injuries were identified in nine cases including dogs with pelvic fractures (n=5), long bone fractures (n=2), facial fractures (n=1), coxofemoral luxation (n=1), and a vertebral fracture (n=1). Multiple orthopedic injuries occurred in case no. 13 (tibial fracture and multiple pelvic fractures) and case no. 21 (coxofemoral luxation and facial fractures). Seven (88%) of the eight dogs with orthopedic injuries suffered from vehicular trauma. Severe bite wounds, defined as multiple bite wounds requiring surgical debridement, were noted in five dogs, and pulmonary contusion was diagnosed in three dogs. Signs of spinal cord injury were described in five (19%) of the cases, including the dog with the vertebral fracture. All five cases had posterior limb paresis with varying degrees of diminished conscious proprioception. Three cases had reduced motor function. None, however, had complete loss of motor function. Based on neurological examination, the clinical signs were felt to be consistent with a third thoracic to third lumbar (T3–L3) spinal cord lesion in all cases. The four cases with suspected spinal cord injury that survived until discharge showed normal or near-normal neurological function at the time of hospital discharge. This included the case with a vertebral fracture. Orthopedic injury was not detected in four of five cases suspected of having spinal cord injury. A definitive diagnosis was not established for the five cases with suspected neurological injury, as myelography, computed tomography (CT), or magnetic resonance imaging (MRI) were not performed.
Anesthesia and Analgesia.
Thirteen (50%) of the dogs received a preanesthetic medication. Preanesthetic drugs included butorphanol (n=6), oxymorphone and diazepam (n=5), and fentanyl (n=2). Induction agents included propofol (n=11), ketamine and diazepam (n=8), sodium thiopental (n=5), and etomidate (n=1). All dogs were maintained using isoflurane for inhalant anesthesia. Information on the anesthetic protocol was unavailable for one dog. All dogs received postoperative analgesia. The most commonly used drug was buprenorphine (n=14) followed by oxymorphone (n=6), butorphanol (n=4), and morphine (n=2). Three of the four dogs that had thoracostomy tubes also received bupivacaine, which was administered through the tube into the pleural space.
Postoperative Complications.
Postoperative complications included ventricular arrhythmias documented by electrocardiography (n=3), incisional or body wall infection (n=3), and pancreatitis (n=1). Organisms cultured from animals with postoperative infections included Staphylococcus intermedius (n=1), Pseudomonas aeruginosa (n=1), and a combined infection with Staphylococcus intermedius and Enterobacter cloacae (n=1). Two cases required surgical debridement of the incision, and the remaining case responded to antibiotic therapy alone. One dog (case no. 9) required intermittent positive-pressure ventilation after it experienced cardiopulmonary arrest during its recovery from anesthesia. This dog was mechanically ventilated for 12 hours after successful cardiopulmonary resuscitation and survived an additional 48 hours before recurrence of cardiopulmonary arrest and death.
Length of Stay.
The median length of stay was 4.5 days (range, 1 to 14 days). The median stay for survivors was 5 days (range, 2 to 14 days) compared to 3 days (range, 1 to 6 days) for nonsurvivors. Six dogs had a length of stay >6 days, and all of these dogs ultimately survived. There was a significant difference between survivors and nonsurvivors for length of stay (P=0.04).
Supportive Therapies.
All dogs received antibiotics for bite wounds, abrasions, or perioperative prophylaxis for surgical repair of orthopedic injuries. All dogs in the study received intravenous (IV) lactated Ringer’s solution. The mean dose received in the first 24 hours was 92.8 mL/kg±40.0 (range, 49.6 to 157.5 mL/kg body weight). Five (19%) dogs received IV hetastarch. The median dose received in the first 24 hours was 14 mL/kg body weight (range, 5 to 24 mL/kg body weight). Eight (31%) of the animals in this study were transfused with blood products. Five (19%) received packed red blood cells (pRBC) and fresh-frozen plasma (FFP); one (4%) received pRBC only; one (4%) received FFP only; and one dog (4%) received fresh whole blood, pRBC, and FFP. There was no difference between survivors and nonsurvivors with respect to the administration of blood products.
Outcome.
Nineteen of 26 (73%) cases survived to hospital discharge. Of the seven nonsurvivors, three were euthanized, and the other four died as a result of their injuries. A review of the medical records showed that euthanasia was performed in all three dogs due to worsening clinical condition and not for financial reasons. Two dogs (case nos. 18 and 24) had cardiopulmonary arrests intraoperatively, and neither survived to hospital discharge. There was no significant difference between survivors and nonsurvivors with respect to age, time to presentation after injury, initial temperature, initial respiratory rate, weight, PCV, or dose of fluids received in the first 24 hours. There were significant differences between survivors and nonsurvivors for initial pulse rate (136.5±30.9 for survivors versus 178.0±31.8 for nonsurvivors; P=0.02) and initial TS (5.86±0.74 g/dL for survivors versus 4.05±1.29 g/dL for nonsurvivors; P=0.02).
Ten (71%) of the 14 cases of bite wounds survived, and seven (70%) of 10 cases of vehicular trauma survived. There was no significant difference between the cause of trauma and age, initial temperature, initial pulse rate, initial respiratory rate, PCV, TS, and dose of fluids. However, dogs with bite wounds weighed significantly less than those with vehicular trauma (6.7 kg±4.5 versus 24.3 kg±13.0; P<0.01).
Animal Trauma Triage Score (See Appendix).
An ATT score was calculated for each dog as previously described by Rockar.12 The mean ATT score for all dogs was 5.7±2.0. The mean score for all survivors was 5.1±1.7 compared to 7.3±2.4 for nonsurvivors [Figure 1]. The ATT scores differed significantly between survivors and nonsurvivors (P<0.01).
Cats
The mean age of all cats with TBWH was 5.0±4.0 years (range, 7 months to 14 years). All 10 cats were domestic shorthairs. Four (40%) of the cats were female (three spayed, one intact), and six (60%) were castrated males. Mean weight was 4.7 kg±1.3 (range, 3.0 to 7.4 kg).
Four (40%) of TBWH were the result of bite wounds; three (30%) resulted from vehicular trauma; and the cause of trauma was unknown in three (30%) of the cases. The mean heart rate at presentation was 176±28 bpm (range, 120 to 220 bpm); the mean respiratory rate was 39±10 breaths pm (range, 24 to 56 breaths pm); and the temperature was 99.0°F±2.5° (range, 96.8° to 101.5°F). The mean PCV and TS were 33%±7% (range, 22% to 42%) and 5.8±1.2 gm/dL (range, 5 to 7.8 gm/dL), respectively.
A TBWH was identified during the initial physical examination in five (50%) of the cases and was suspected in an additional three (30%). The TBWH was identified radiographically in five (63%) of the eight cases where the appropriate area was imaged. Radiographic findings were equivocal in one (13%) additional case.
One case (case no. 32) (10%) did not have its TBWH identified until >24 hours after admission. This cat originally presented with unknown trauma and was diagnosed with a proximal femoral fracture that was treated with a femoral head and neck excision. The cat represented 7 days after discharge with a primary complaint of anorexia. A TBWH in the ventral body wall was found on physical examination at that time.
Surgical Findings.
Three (27%) of the hernias were located in the ventral body wall; three (27%) were in the femoral area; two (18%) were located in the lateral paralumbar area; two (18%) resulted from cranial pubic ligament rupture; and one (9%) was a paracostal hernia. One cat (case no. 35) had both a femoral hernia and a cranial pubic ligament rupture. Herniated organs included intestine (n=5), omentum (n=5), and bladder (n=3). Three cats (case nos. 31, 33, and 35) had both small intestinal and bladder herniation. Herniorrhaphy was accomplished in all cases by apposition of autogenous tissues.
Significant concurrent surgical findings included diaphragmatic hernia (n=1) and kidney avulsion (n=1). One cat (case no. 35) had a colonic resection and anastomosis due to partial-thickness perforation of the colon. None of the cats required small intestinal resection and anastomosis.
Associated Injuries.
Six (60%) of 10 cats suffered associated injuries. Severe bite wounds (n=3) were the most common. Other associated injuries included uroabdomen due to kidney avulsion (n=1), bladder rupture identified at surgery (n=1), long bone fracture (n=1), and coxofemoral luxation (n=1). The bladder rupture and long bone fracture occurred in the same cat (case no. 29). Neurological injury was described in one (10%) of the cases. This cat (case no. 27) had unilateral hind-limb paresis. No associated vertebral injury was diagnosed.
Anesthesia and Analgesia.
Nine cats received premedication. Premedication drugs included oxymorphone and diazepam (n=6) and butorphanol (n=3). Induction agents included propofol (n=7) and ketamine and diazepam (n=2). All cats were maintained using isoflurane anesthesia. Information on anesthesia and analgesia was unavailable in one case.
Postoperative Complications.
Postoperative complications in cats included incisional or body wall infection (n=1), acute renal failure (n=1), and pancreatitis (n=1). The cat with a body wall infection (case no. 35) had a prolonged clinical course that involved recurring infections with several different multiresistant bacteria, including Escherichia coli, Staphylococcus spp., and Enterobacter cloacae. This cat developed a body wall abscess that required serial debridement with drain placement. Case no. 33 developed acute renal failure shortly after presentation. A peritoneal dialysis catheter was placed at the time of surgery. Concurrent ethylene glycol intoxication was suspected based upon the clinical and laboratory findings.
Length of Stay.
The median length of stay was 3.5 days (range, 2 to 20 days). The median length of stay for survivors was 5 days (range, 3 to 20 days), and the two nonsurvivors had a 3-day length of stay. There was no significant difference between survivors and nonsurvivors with respect to length of stay.
Supportive Therapies.
All 10 cats received antibiotics for bite wounds, abrasions, or perioperative prophylaxis for surgical repair of orthopedic injuries. Every cat with a TBWH received IV lactated Ringer’s solution. The mean dose received in the first 24 hours was 37.4 mL/kg body weight±18.5 mL/kg body weight (range, 22.3 to 49.1 mL/kg body weight). Four (40%) received IV hetastarch. The median dose received in the first 24 hours was 10.75 mL/kg body weight (range, 5.5 to 12 mL/kg body weight). Five (50%) of the cats received a transfusion with type-specific banked whole blood.
Outcome.
Eight (80%) of 10 cats survived and were successfully discharged from the hospital. Of the two nonsurvivors, one was euthanized due to a deteriorating clinical condition, and the other died as a result of its injuries. There was no significant difference between survivors and nonsurvivors with respect to age, time to presentation, initial temperature, initial respiratory rate, pulse rate, weight, PCV, TS, or dose of fluids received in the first 24 hours.
One (33%) of three cases of vehicular trauma survived, four (100%) of four cases of bite wounds survived, and all three cases where the cause of trauma was unknown survived. There were insufficient case numbers to detect significant differences between these groups. There was no significant difference between survivors and nonsurvivors for the cause of trauma, age, weight, temperature, initial pulse rate, initial respiratory rate, PCV, TS, receipt of a transfusion, or dose of fluids in the first 24 hours.
Animal Trauma Triage Score.
The mean ATT score for all cats was 6.4±2.5. Survivors had a score of 6.0±2.3 compared to a score of 8.0±4.2 for nonsurvivors. There was no significant difference in ATT scores between surviving and nonsurviving cats.
Discussion
In this series of cases, 54% of the canine TBWH and 40% of the feline TBWH were the results of dog-bite wounds. This is in contrast to a previous report of traumatic abdominal hernias, where only 15% of the canine and 18% of the feline hernias were the results of animal attacks.6 The different populations of patients seen by the two hospitals may explain this difference.
A TBWH was diagnosed or suspected in 28 of the 36 cases (20/26 dogs, 8/10 cats) based upon physical examination findings. Radiographs were diagnostic or suggestive of a TBWH in 20 of 31 cases (14/23 dogs, 6/8 cats) where the appropriate area was imaged. It is of interest that five cases (four dogs, one cat) did not have their TBWH identified within the first 24 hours of hospitalization. Two of the dogs and the cat had orthopedic injuries corrected prior to the diagnosis of a TBWH. The delay in diagnosis of these cases may have been due to incomplete or inadequate physical examination or reduction of the hernia prior to the initial physical examination, with later reherniation of abdominal contents. Alternatively, there may have been no abdominal contents herniated at the time of trauma, but subsequent patient movement or manipulation facilitated herniation through a traumatic defect in the body wall. This delay in diagnosis is indicative of the difficulty veterinarians may encounter in identifying all injuries in patients that present with multiple traumatic injuries. In order to diagnose all TBWH within a timely manner, the clinician must maintain a high index of suspicion in patients with multiple injuries and perform detailed serial physical examinations. This is especially important in those cases that do not improve as expected, given their diagnosed injuries, or in animals with evidence of severe local swelling or contusion that cannot be explained by their previously identified injuries. In cases where the diagnosis of TBWH has not been reliably established on the basis of physical examination or routine radiography, positive contrast celiography and ultrasonography may aid in confirming the diagnosis.14
Fifty-four percent of the canine hernias in this series occurred in the lateral paralumbar body wall. This has been cited as the most common location for traumatic herniation due to decreased elasticity of the tissues in the lateral abdominal wall when compared to the costal attachments, linea alba, or the cranial pubic ligament.13 The lack of support from the rectus abdominal musculature may also predispose to lateral herniation.6 The findings of this study and reference nos. 1 and 3 are in contrast to a previous report where only 14% of the dogs had lateral hernias.6 This may again be due to differences in the etiology of trauma in the two studies. All the cranial pubic ligament ruptures in this study were the result of vehicular trauma. This is most likely due to the extreme forces required to disrupt the cranial pubic ligament only being found in cases of vehicular impact. These results are consistent with Waldron et al. and Mann et al.’s findings, where all cases of cranial pubic ligament rupture resulted from vehicular trauma.69
In this series, there were four dogs with intercostal herniation. All four of these cases resulted from bite wounds. The fact that all four cases resulted from bite wounds may reflect the unique forces applied to the body when larger dogs attack smaller dogs. The smaller dog is frequently picked up and shaken when attacked. This shaking might apply significant shearing forces to the intercostal muscles, resulting in their disruption.10 In contrast, the blunt trauma during vehicular impact is more widely distributed and may be better absorbed by the thoracic cage.
Twelve of the cases (nine dogs, three cats) had associated body cavity injuries identified at surgery. This is similar to the frequency of intra-abdominal injuries found in a study of human traumatic abdominal hernias, where 30% of cases had intracavitary injuries.13 Many of the injuries seen in this series were consistent with those reported to occur following blunt abdominal trauma (e.g., diaphragmatic hernia, kidney avulsion, bladder rupture, and mesenteric avulsion).115
Four of the five animals requiring intestinal resection and anastomosis showed evidence of intestinal serosal penetration from bite wounds in addition to their TBWH. The region of resected intestine was partially or fully penetrated in all four of these cases. The remaining case of intestinal resection and anastomosis had a mesenteric avulsion. The low frequency of intestinal injuries in cases without direct serosal injury might reflect the ability of tubular organs to withstand considerable blunt trauma without significant injury.15 Due to the high frequency of serious concurrent injuries in this series, the authors believe surgical exploration of the underlying abdominal or thoracic viscus is indicated during all TBWH repairs.
Orthopedic injuries were present in nine dogs and two cats. Ten of the 11 cases with orthopedic injury had sustained vehicular trauma, and the cause of injury was unknown in the remaining case. This frequency probably resulted from the significantly greater forces generated during vehicular impact. Clinically significant neurological injury was described in six of the cases (five dogs, one cat), and a spinal fracture was diagnosed in one of the dogs. The remaining cases all had posterior paresis thought to result from spinal cord trauma that was not a result of vertebral fracture or luxation (e.g., hemorrhage in or around the spinal cord or traumatic intervertebral disk herniation). None of these cases had myelography, CT, or MRI performed; however, five of the animals with signs of spinal cord injury had normal or near-normal neurological function at the time of discharge.
Four of the cases (three dogs, one cat) developed postoperative infections at the site of surgery. Three of the four cases that developed infectious complications had bite wounds from a dog attack. This might have been the result of inoculation of bacteria at the time of injury, which later developed into a local infection. The only case of vehicular trauma that had an infectious complication was a dog in which a polypropylene mesh was used to close the defect in its body wall. The use of implantable mesh is a risk factor for the development of wound infection, and this probably played a role.16 The choice of appropriate therapeutic antibiotics should ideally be based upon culture and sensitivity results. Pending culture and sensitivity testing, empirical antibiotic choice should be guided by knowledge of the likely bacterial pathogens.17 Antibiotic sensitivities of bacteria cultured from bite wounds vary widely and cannot be accurately predicted.1819 In one study, 30% of the staphylococci were resistant to penicillin.20 Due to the uncertain resistance patterns, some authors recommend a combination of antibiotics (e.g., a penicillin with an aminoglycoside or a second-generation cephalosporin combined with a fluoroquinolone).18 Other authors feel there is no evidence to support the use of empirical antibiotic therapy in patients with bite wounds.21
This study found a difference between surviving and nonsurviving dogs with respect to length of hospitalization. None of the nonsurvivors had a length of stay >6 days, while eight (30%) of the survivors had a length of stay >6 days. This difference likely reflects the prolonged hospitalization required in patients with multiple injuries that ultimately survive and suggests that nonsurvivors may die earlier in the course of their hospitalization.
Patients in this study received both lactated Ringer’s solution and hetastarch. The exact reason for the fluid choice in each case was not available in the patient records. While crystalloid fluids such as lactated Ringer’s solution are rapidly redistributed to the interstitial space, colloids such as hetastarch have a greater ability to remain in the intravascular space and thus potentially provide greater volume support.22 Thirteen of the 36 patients (8/26 dogs, 5/10 cats) in this study received transfusions. This transfusion rate may emphasize the severe degree of trauma found in patients with a TBWH. Transfusions are usually administered to restore the hematocrit to a level where clinical signs attributable to anemia are no longer evident, to treat a coagulopathy, to provide oncotic support, or to restore more normal levels of hemoglobin in order to improve oxygen delivery during anesthesia. Unfortunately, the exact reason for transfusion in this study could not be readily ascertained from the medical records, although FFP alone was given to one animal while all the others received blood products that included pRBCs and stored whole blood.
The overall mortality rate of 25% (seven dogs, two cats) in this group of patients appears to be higher than in previous reports. In Waldron, et al.’s review, only two (9%) of the 22 cases died, while Mann, et al. found that 20% of the cases with cranial pubic ligament rupture did not survive.69 This disparity may be explained at least in part by the different distribution of injuries in the two previous studies compared to this study. The actual mortality rate for patients with TBWH may be much higher, since this study excludes any cases that died or were euthanized before surgical intervention. An accurate gauge of overall mortality in the authors’ hospital population is not possible due to an inability to identify those patients with a TBWH that were euthanized or died before surgical intervention. For most animals in this study, a specific cause of death could not be accurately determined from the retrospective review of the medical record.
The initial pulse rate and total solids were significantly different between surviving and nonsurviving dogs. Pulse rate may have been increased in these animals due to pain, anxiety, hypotension, arrhythmia, or other causes of hemodynamic instability. Total solids may be decreased because of blood loss, tissue damage, endothelial dysfunction, or preexisting disease. The cause of these differences between survivors and nonsurvivors could not be accurately determined from a retrospective review of the medical records. Measurement of pulse rate, respiratory rate, mucous membrane color and capillary refill time, blood pressure, urine output, PCV, and total solids can be useful to evaluate blood volume and hemodynamic status.23 A recent human study confirmed the importance of measurement and correction of resuscitation endpoints such as blood pressure, central venous pressure, urine output, and central venous oxygen saturation.24 It identified improved survival when certain treatment endpoints were achieved using a goal-directed approach.
This study found that those patients with a TBWH as a result of bite wounds were significantly smaller than those patients suffering their injuries as a result of vehicular trauma. This is consistent with previously published information.126 This size disparity is thought to result from the increased likelihood of a smaller animal receiving trauma during an attack that is of sufficient force to cause a TBWH.
The mean ATT score was significantly different for surviving and nonsurviving dogs. The ATT score was designed to predict the likelihood that a patient would survive 7 days beyond presentation for emergency care following traumatic injury.12 The survivors in this study had a higher ATT score (5.4±2.3) than the survivors in Rockar’s study (2.4±0.2), although the overall survival rates in the two studies were similar (75.0% versus 77.2%).12 The higher ATT score in the current study may be due to the fact that the deep muscular injury and penetration into the abdominal or thoracic cavity seen with a TBWH resulted in a minimum of three points being assigned to the ATT score in all patients. Applying the ATT score in a retrospective manner can pose difficulty, since it requires the interpretation of the clinical record in order to assign a point value. Some of the values (e.g., temperature and heart rate) are easily recorded retrospectively, while others (e.g., pulse quality and respiratory effort) were not specified in some medical records and therefore are more difficult to interpret and assign scores. Despite this difficulty, the ATT score may provide authors and readers a useful tool to compare patient populations in studies of traumatized animals.
Case no. 26 suffered from an autopenetrating hernia. To the authors’ knowledge, this type of injury has not been previously reported in the veterinary literature. The blunt abdominal force applied to the caudal thoracic area resulted in a fracture of the 12th rib, with the fracture fragment penetrating through the diaphragm, peritoneum, and the muscular layers of the abdominal wall, resulting in subsequent herniation of the liver through the abdominal defect. The skin remained intact due to its superficial position and elasticity. This injury fulfills the criteria for a traumatic hernia.2 This hernia differs from the paracostal hernias described in this series, as paracostal hernias typically occur when shearing forces cause separation of the abdominal musculature from their attachments on the last rib. In contrast, autopenetrating hernias result from sharp, dissecting trauma caused by a fracture.13
Conclusion
Although not common, a TBWH represents a serious physical insult with significant morbidity and mortality. Based on this study, the initial heart rate and TS may help the clinician to identify patients at a higher risk for death; however, further studies are indicated to better assess this finding. Both orthopedic and intra-abdominal injuries are commonly associated with TBWH. Traumatic body wall hernias may not be identified on initial examination; careful attention to secondary and tertiary physical examinations and use of imaging techniques may assist in diagnosis of TBWH in a timely manner. Surgical exploration of underlying tissues and body cavities is indicated in all patients with a TBWH due to the high frequency of concurrent injuries to underlying viscera.
Marlex mesh; Bard Pharmaceuticals, Philadelphia, PA
Citation: Journal of the American Animal Hospital Association 39, 1; 10.5326/0390035
A comparison of animal trauma triage (ATT) score for surviving and nonsurviving dogs with traumatic body wall hernias.
