Eleven-year retrospective analysis of acquired diaphragmatic hernia in 49 dogs and 48 cats

Abstract

Data on dogs and cats that underwent surgery for a traumatic diaphragmatic hernia were retrospectively collected and analyzed, with an objective to identify factors that influenced the survival rate. Forty-nine dogs and 48 cats were included. The predominant respiratory clinical sign was dyspnea, and the interval from trauma to development of clinical signs did not influence patient survival. Concurrent orthopedic and/or soft tissue injuries were identified in 48.45% of the animals. Intraoperative complications occurred in 14 dogs and 5 cats, and postoperative complications in 7 dogs and 6 cats. Intraoperative death occurred in 6.2% and postoperative death in 8.3% from 1 h to 10 d after surgery. Animals that received surgical treatment within 48 h after diagnosis had a lower risk of death. Conversely, concurrent injuries and intraoperative and postoperative complications were the main factors associated with a higher risk of death.

Introduction

Blunt or penetrating trauma of the thorax or abdomen can be associated with traumatic diaphragmatic hernias in dogs and cats (1–3). Displacement of the abdominal viscera into the thoracic cavity may compress the pulmonary lobes, with progressive atelectasis and pleural effusion, causing compromised ventilation (3,4). Affected animals may have respiratory and cardiac changes, such as tachypnea, dyspnea, cyanosis, tachycardia, and muffled heart sounds (3,5,6). In addition, clinical signs may be minimal and undetected for years when the degree of pulmonary compromise is minor (3,7), although an asymptomatic animal may eventually present with respiratory and/or gastrointestinal signs (1,8).

Diagnosis can be made based on a history of traumatic insult, clinical signs, physical examination, and imaging studies (1,4). Although radiography is commonly used to diagnose a diaphragmatic hernia, it may be difficult to identify loss of the diaphragm silhouette in the presence of associated pleural effusion (9). In addition, radiographs may lack evidence of organ herniation (1). Another imaging modality used as an adjunctive or sole diagnostic method is ultrasonography, primarily through the transhepatic abdominal and intercostal windows (10). Various surgical approaches have been reported to repair diaphragmatic defects, including ventral midline celiotomy, which is the most commonly used and can be extended to sternotomy, and lateral thoracotomy when adhesions are present (3,4,11,12). Additional approaches have been reported, such as paracostal extension (3,4). Hernia repair using a laparoscopic approach has also been suggested (13,14).

Data on dogs and cats that underwent surgery for a traumatic diaphragmatic hernia were retrospectively collected and analyzed, with an objective to identify factors that influenced the survival rate.

Materials and methods

Procedures

This study was approved by the Institutional Ethics Committee for Animal Care and Use (No. 0082/2021, CEUA). Medical records from 2010 to 2020 obtained from a veterinary teaching hospital of dogs and cats suspected of traumatic diaphragmatic rupture were reviewed. The inclusion criteria were imaging studies that confirmed the presence of diaphragmatic rupture and surgical correction of diaphragmatic injury performed via a ventral midline celiotomy. The exclusion criteria were congenital diaphragmatic hernia and conservative treatment.

The data gathered from medical records included patient signalment [breed, age, sex, body weight (BW)]; cause of injury; clinical signs; concurrent injuries; interval from trauma to development of clinical signs (≤ 2 d, 2 to 10 d, 10 to 30 d, and ≥ 30 d before admission); time of surgical intervention (≤ 24 h, from 24 to 48 h, and ≥ 48 h after the diagnosis); organ displaced into the thoracic cavity; intraoperative and postoperative complications; and intraoperative and postoperative mortality rates.

Statistical analyses

Statistical analyses were performed to determine whether the interval from trauma to development of clinical signs (acute: ≤ 2 d; chronic: > 2 d) and the timing of surgical intervention (≤ 48 h; ≥ 48 h) were associated with survival. Concurrent injuries associated with diaphragmatic hernia were also analyzed in terms of survival and the presence of complications. Fisher’s Exact test was used because the outcomes were binary (survival or not). Odds ratios (ORs) with 95% confidence intervals (CIs) were also calculated. All statistical analyses were conducted using commercial statistical software (GraphPad Prism 8.3.1; San Diego, California, USA).

Results

Of the 108 dogs and cats diagnosed with a diaphragmatic hernia, 49 dogs and 48 cats met the inclusion criteria. Most dogs were mixed breeds (n = 31; 63.3%). Purebred dogs were represented by the following breeds: 3 each of pinscher and poodle; 2 each of Labrador retriever, Lhasa Apso, dachshund, and Maltese; and 1 each of Yorkshire terrier, Australian cattle dog, beagle, and German spitz. Except for 1 Siamese cat, most cats were of mixed breeds. Dogs had a mean age of 46.5 ± 39.3 mo and a mean BW of 7.9 ± 5.3 kg. Twenty-four dogs were male and 25 were female. Cats had a mean age of 33.6 ± 34.8 mo and a mean BW of 3.0 ± 1.3 kg. Twenty-eight cats were male and 20 were female. The data are presented in Table 1.

Table 1.

Signalment of 49 dogs and 48 cats presented with traumatic diaphragmatic rupture.

Signalment	Dogs	Cats	All patients
Species	49 (51.5%)	48 (49.5%)	97 (100%)
Female	24 (49%)	28 (58.3%)	52 (53.6%)
Male	25 (51%)	20 (41.7%)	45 (46.4%)
Mixed breeds	31 (63.3%)	47 (97.9%)	78 (80.4%)
Purebreds	18 (36.7%)	1 (2.1%)	19 (19.6%)
Mean age (mo), (range)	46.5 (2 to 144)	33.6 (2 to 156)	40.16 (2 to 156)
Mean body weight (kg), (range)	7.9 (1.2 to 33)	3.02 (0.5 to 7.4)	5.48 (0.5 to 33)

The cause of the diaphragmatic rupture was motor vehicle accidents in 38 dogs (77.6%) and 35 cats (72.9%). Other causes in cats were falling from a height (n = 1, 2%) and kicking (n = 1, 2%). In addition, 1 dog (2%) was attacked by another. The cause was unknown in 11 cats (22.9%) and 10 dogs (20.4%); however, these were free-roaming animals.

The most common clinical sign at presentation was dyspnea in 99% of the animals (N = 96). Coughing was the major clinical sign in 1 cat (2%). Other clinical signs were present in 21.9% of the animals (n = 10 dogs, n = 11 cats), including lameness (n = 4 dogs, n = 6 cats; 10.4%), shock (n = 3 dogs, n = 1 cat; 4.2%), hematuria (n = 3 cats; 3.1%), inability to bear weight on the hind limbs (n = 1 dog, n = 1 cat; 2.1%), stupor (n = 1 cat; 1%), and syncope (n = 1 dog; 1%).

Concurrent injuries were detected in 47 (48.4%) animals. The most prevalent orthopedic injury was pelvic fracture (n = 6 dogs, n = 3 cats; 9.27%), followed by femoral fracture (n = 1 dog, n = 5 cats; 6.18%), tibial fracture (n = 3 dogs, n = 2 cats; 5.1%), hip luxation (n = 2 cats, n = 2 dogs; 4.12%), rib fracture (n = 2 dogs; 2.1%), fracture of both the humerus and radius/ulna (n = 1 cat; 1%), and sacrococcygeal luxation (n = 1 cat; 1%). The most prevalent soft tissue injuries were hepatic lacerations (n = 2 dogs, n = 1 cat; 3.1%) and traumatic abdominal hernia (n = 2 dogs, n = 1 cat; 3.1%). Rupture of the vena cava and intestinal perforation were detected in 1 dog each (1%), and urinary bladder rupture and skin wounds were present in 1 cat each (1%). Other lesions such as myiasis, pneumonia, scleral hemorrhage, and chemosis were detected in 1 dog each (1%), and abscess, sepsis, megaesophagus, and dead fetuses were detected in 1 cat each (1%). In addition, 1 dog had a traumatic brain injury (1%).

The interval for development of clinical signs following trauma was as follows: 53.6% (n = 25 dogs, 27 cats) up to 2 d, 21.6% (n = 8 dogs, 13 cats) from 2 to 10 d, 7.2% (4 dogs and 3 cats) from 10 to 30 d, and 5.2% (n = 5 dogs) more than 30 d before admission. The interval from trauma to development of clinical signs was unknown in 7 dogs and 5 cats (12.37%). The interval from diagnosis to surgical intervention was ≤ 24 h in 72.2% of the animals (n = 26 dogs, n = 44 cats), from 24 to 48 h in 11.3% (n = 1 dog, n = 10 cats), and ≥ 48 h in 16.5% (n = 11 dogs, 5 cats).

Thoracic radiographs were obtained for all animals; however, an ultrasound of the thorax and abdomen was necessary to provide additional diagnostic information in 8.2% of the animals (n = 6 dogs, n = 2 cats). The most common radiographic findings were loss of diaphragmatic silhouette, presence of abdominal gas shadows in the thoracic cavity related to small intestine loops or stomach, loss of the cardiac silhouette, and displacement of the lung fields. The absence of diaphragm between the liver and heart made the ultrasonographic diagnosis easier.

Information about the organ displaced into the thoracic cavity was available for 30 dogs (30.9%) and 25 cats (25.8%). Intestines were the most frequently herniated organ in the thoracic cavity (n = 18 dogs, n = 17 cats; 36.1%), followed by the liver (n = 18 dogs, n = 12 cats; 30.9%), stomach (complete or partial) (n = 17 dogs, n = 12 cats; 29.9%), spleen (n = 6 dogs, 7 cats; 13.4%), omentum (n = 1 dog, n = 3 cats; 4.1%), and kidney (n = 1 dog; 1%). Seventeen animals (17.5%) had only 1 organ displaced into the thoracic cavity, 19 (19.6%) had 2 organs, 10 had 3 organs (10.3%), 8 (8.2%) had 4 organs, and 1 (5.2%) had 5 organs herniated into the thoracic cavity. All hernias were repaired primarily with simple interrupted or continuous sutures. Monofilament nylon was the most used non-absorbable suture.

Intraoperative and postoperative complications occurred in 32.98% of patients (n = 21 dogs, n = 11 cats). Intraoperative complications occurred in 19 patients (n = 14 dogs, n = 5 cats; 19.6%), including hepatic laceration (n = 4 dogs, n = 1 cat; 5.1%), adhesions between thoracic and abdominal organs (n = 4 dogs, n = 1 cat; 5.1%), pneumothorax (n = 3 dogs, n = 1 cat; 4.1%), hypotension (n = 1 dog, n = 1 cat; 2.1%), pregnancy loss due to trauma (n = 1 dog; 1%), intestinal rupture (n = 1 dog; 1%), and cardiorespiratory arrest responsive to resuscitation maneuvers (n = 1 cat; 1%). Iatrogenic liver lacerations were associated with surgical manipulation, or breakdown of adhesions between the liver and other organs or diaphragm. Hemorrhage associated with hepatic lacerations was controlled using direct compression or topical hemostatic agents. All animals with adhesions developed clinical signs 30 d after trauma. Intestinal resection and anastomosis were used to treat intestinal rupture. In addition, ovariohysterectomy was completed to treat the cat with dead fetuses and a dog with pregnancy loss.

Thirteen animals (13.4%) experienced postoperative complications, including pneumothorax (n = 5 dogs, n = 4 cats; 9.3%), refractory hypotension (n = 1 dog; 1%), pleural effusion (n = 1 cat; 1%), and recurrence of diaphragmatic rupture (n = 1 dog; 1%). In addition, 1 cat died of an unknown cause 10 d after surgery. Of the 13 patients with postoperative pneumothorax, clinical signs developed 30 d after trauma.

The total mortality rate was 14.4% (n = 14; 7 dogs and 7 cats), 6.2% intraoperatively (n = 5 dogs, n = 1 cat), and 8.2% postoperatively (n = 2 dogs, n = 6 cats) from 1 h to 10 d after surgery. Intraoperative death was associated with hepatic laceration resulting from the original trauma (n = 3 dogs, n = 1 cat; 4.1%), rupture of the vena cava (n = 1, 1%), and refractory hypotension (n = 1, 1%). All animals that died intraoperatively and postoperatively developed clinical signs ≤ 24 h after trauma. Postoperative death was associated with pneumothorax (n = 2 dogs, n = 2 cats; 4.1%); hypotension (n = 1 dog; 1%); traumatic hepatic laceration (n = 1 dog; 1%); and pregnancy loss due to trauma, that required an ovariohysterectomy during hernia correction and contributed to the deterioration of the dog’s general condition (n = 1 dog; 1%). Two cats died of unknown causes; however, 1 of them had a previous diagnosis of megaesophagus that may have contributed to death.

The interval from trauma to development of clinical signs, assessed as acute or chronic (OR: 1.1; 95% CI: 0.4 to 3.7; P = 1.0), did not influence patient survival. However, the interval from diagnosis to surgical treatment identified a lower risk of death in animals that received surgical intervention within 48 h after diagnosis (OR: 0.14; 95% CI: 0.04 to 0.4; P = 0.0018). The presence of concurrent injuries associated with diaphragmatic hernia was associated with a higher risk of death (OR: 5.84; 95% CI: 1.5 to 20.4; P = 0.008), as well as the presence of complications (OR: 20.21; 95% CI: 4.1 to 93.4; P < 0.0001).

Discussion

Data on dogs and cats that underwent surgery for traumatic diaphragmatic hernias were reviewed in this study. A retrospective study of 60 cases revealed the percentage of traumatic diaphragmatic hernias was approximately the same in dogs (51.5%) and cats (49.5%) (15), as observed herein. Other studies have reported a relatively high prevalence in dogs (6,16–18). Distribution according to was similar in dogs; however, females were over-represented in cats. In contrast, some studies reported a higher prevalence of traumatic diaphragmatic rupture in male cats (6,18,19). Although the mean age (46.5 mo) at diagnosis for dogs was comparable to that of other studies, the mean age of 2.9 y for cats was lower than the means of 3.7 and 4.7 y reported in other studies (6,18). Motor vehicle accidents have been considered the most frequent cause of trauma associated with diaphragmatic hernias in both cats and dogs (5,6,11,15,19), as verified in the present study. Although the cause was unknown in 10.3% of cats and 7.2% of dogs, these animals face several street dangers due to their free-roaming behavior. Other studies have identified a high prevalence of unknown causes (5,6,18).

Most animals (99%) in the current study had dyspnea at presentation. However, respiratory distress is a nonspecific sign of traumatic disruption of the diaphragm, which is influenced by herniated organs, the size of the diaphragmatic defect, and the presence of other injuries (2,3). Diaphragmatic tears with herniated organs do not always alter the ability to ventilate and the animal may compensate for reduced pulmonary capacity (2). In a study of 34 dogs and 16 cats with diaphragmatic hernia of ≥ 2 wk duration, dyspnea and vomiting were the most prevalent clinical signs (8). Other respiratory signs may develop such as fatigue after exercise and intermittent chronic cough (1). Anorexia, lethargy, and weight loss may also be observed in chronic diaphragmatic hernias as non-specific signs (8). Despite some identified differences in clinical signs for acute versus chronic diaphragmatic hernias, the interval from trauma to development of clinical signs did not influence survival in this cohort of animals.

In the present study, the intestine (36.1%), liver (30.9%), stomach (29.9%), and spleen (15.5%) were herniated into the thorax more often than other organs. Liver, small intestine, and stomach have been reported as the most frequent organs involved in traumatic diaphragmatic hernia; however, any abdominal organ can be herniated (1,15). Size and location of the diaphragmatic tear may influence the organs that can herniate into the thorax (3).

Concurrent orthopedic and soft tissue injuries were identified in 48.45% of the animals. Similar findings have been reported in other studies and are associated with the magnitude and extent of trauma (6,18). Fractures were the most common orthopedic injury in the present study, suggesting severe trauma. Approximately 2% of dogs with fractures owing to motor vehicle accidents also have traumatic diaphragmatic hernias (4). Hepatic laceration and abdominal hernia were the most frequent soft tissue injuries, and both lesions were related to abdominal trauma. Furthermore, a high risk of death was related to hepatic laceration in the present study.

Pneumothorax was the major postoperative complication in the present study (9.3%). In a retrospective study with a chronic diaphragmatic hernia in 34 dogs and 6 cats, pneumothorax was the most common transient postoperative complication (8). Conversely, in a study of 56 cases, canine deaths were related to pneumothorax with a degree of hemothorax, whereas feline deaths were due to pulmonary edema (16).

The total mortality rate was 14.4% in the present study, with 6.2% of deaths occurring intraoperatively and 8.2% of deaths occurring postoperatively. The relationship between the timing of surgical intervention and the mortality rate has been debated (4,20). The mortality rate for acute traumatic diaphragmatic hernia has been reported to range from 6.3 to 20% (20). For chronic traumatic diaphragmatic hernia defined as > 14 d, the mortality rate was reported to be 14% in 1 study (8) and 0 to 19.4% in another study (18). The interval from diagnosis to surgical intervention identified a lower risk of death in animals that underwent surgery within 48 h after diagnosis in the present study. Conversely, survival was not related to the interval from trauma to surgery, or from admission to surgery in a study with 17 cats and 79 dogs (18).

In summary, a review of traumatic diaphragmatic rupture over 11 y revealed that most dogs and cats were of mixed breeds that suffered motor vehicle accidents. The predominant clinical respiratory sign was dyspnea; however, the interval from trauma to development of clinical signs did not influence patient survival. Concurrent orthopedic and/or soft tissue injuries and intraoperative and postoperative complications were the main factors associated with a higher risk of death in both species.