Hemorrhagic Pleural Effusion Secondary to an Unusual Type III Hiatal Hernia in a 4-Year-Old Great Dane

Introduction

A hiatal hernia (HH) occurs when the abdominal esophagus, gastroesophageal junction, and the gastric cardia and/or fundus undergo cranial protrusion through a defective esophageal hiatus into the thoracic cavity.^1–5 Herniation can occur on an intermittent or permanent basis, with or without the involvement of other abdominal organs.^2,4–6 The clinical presentation of HH is variable and, traditionally, this disorder in small animals has been classified into four distinct types based on a human classification system. Defects of the esophageal hiatus can lead to an axial or “sliding” HH (type I), a paraesophageal or “rolling” HH (type II), or a combination of axial and paraesophageal HH (type III). A discrepancy exists with the classification of type IV HH, with the terminology of this defect being inconsistent in recent literature. Type IV HH has been defined both as a combined type III HH complicated by herniation of abdominal organs other than the stomach,^6–9 and as gastroesophageal intussusception.^1,3–5

Hiatal hernias are most commonly congenital; however, acquired HHs occurring secondary to trauma, upper airway obstruction, or tetanus have also been reported.^{3–5,10–12} In the absence of known causative factors, congenital herniation is regarded to be the result of inherent laxity of the phrenicoesophageal ligament and/or musculature of the right diaphragmatic lumbar crus, which together form the hiatal orifice.^2,5,9,10,13

Classical clinical signs of HH are usually chronic in nature and include occasional or persistent regurgitation, vomiting, abdominal pain and distension, hypersalivation, and dysphagia.^{1,2,4,7,10,12,14,15} Dyspnea secondary to aspiration pneumonia or lung compression by marked invasion of abdominal organs into the chest has also been reported.^8,9,16,17 To the authors’ knowledge, this is the first report of hemorrhagic pleural effusion causing an acute onset of dyspnea secondary to HH in a dog.

Case Report

A 4-year-old, 50.5-kg, spayed female Great Dane was presented with acute onset of dyspnea, minor abdominal distension, and a 48-hour history of intermittent postprandial retching and regurgitation of a viscous fluid. Dyspnea was first observed 2 days prior to presentation with no history of trauma or specific inciting cause. According to the owner, this was the first time the dog had demonstrated these clinical signs. Twelve hours prior to referral, thoracocentesis had been performed following the radiographic diagnosis of a pleural effusion. No significant improvement was observed. The respiratory signs worsened, and referral was recommended.

Upon presentation to a soft-tissue surgical specialist, the dog appeared subdued but alert, with marked dyspnea and minor abdominal distention. Clinical examination revealed pain on palpation of the cranial abdomen. Body temperature was slightly elevated (39.3°C), and the dog was hypotensive with a systolic blood pressure of 80 mm Hg. The mucous membranes were pink and tacky, with a capillary refill time of 1 second. Thoracic auscultation revealed decreased lung sounds (specifically over the left caudodorsal and bilateral cranioventral lung fields), and a shallow breathing pattern was noted. A sinus tachycardia of 120 beats per minute (bpm) was present, and heart sounds were muffled over both the left and right sides. The pulse quality was considered moderate, and no pulse deficits were present.

Right lateral survey radiographs of the thoracic cavity were made and revealed marked pleural effusion with loss of the cardiac silhouette and a caudodorsal, gas-filled, intrathoracic structure (consistent with the stomach) causing dorsal interruption of the diaphragmatic outline [Figure 1]. Approximately 1200 mL of nonclotting pleural fluid were removed from both sides of the chest via needle thoracocentesis, with a packed cell volume (PCV) of 15%, suggestive of frank blood.¹⁸ Immediate improvement in respiratory rate and effort was noted. Buccal mucosal bleeding and activated clotting times were within normal limits. Serum biochemical analyses and hematology results were unremarkable except for a low end of normal-range peripheral PCV (34%, reference range 35% to 70% in dogs).¹⁹

Following thoracocentesis, radiographs confirmed the presence of an oval, poorly marginated, gas-filled, soft-tissue structure protruding into the caudodorsal thoracic space [Figure 2].^15,17,20 These findings were consistent with the diagnosis of pleural effusion and an esophageal HH. No radiographic signs of aspiration pneumonia were noted.

An 18-gauge catheter was placed into each cephalic vein, and the dog was treated with oxygen support and intravenous fluid resuscitation (4 mL/kg bolus of hydroxyethyl starch^a colloid fluid, followed by 40 mL/kg per hour of compound sodium lactate crystalloid solution^b) for signs of compensated cardiovascular shock. After 30 minutes, the heart rate had decreased to 95 bpm, systolic blood pressure had increased to 110 mm Hg, and pulse quality was stronger. Once the dog was hemodynamically stable, crystalloid fluids^b were continued intravenously (IV) at 4 mL/kg per hour. The dog was administered acepromazine^c (0.01 mg/kg) and methadone hydrochloride^d (0.3 mg/kg) together intramuscularly as preanesthetic medications 20 minutes prior to induction with propofol^e (4 mg/kg IV) to allow tracheal intubation. General anesthesia was maintained with isoflurane^f in oxygen. Manual, intermittent, positive-pressure ventilation was initiated shortly after intubation and continued until the dog was mechanically ventilated at a rate of 10 to 12 breaths per minute with a tidal volume of 12 to 15 mL/kg (adjusted to maintain the arterial partial pressure of carbon dioxide [PaCO₂] at 35 to 45 mm Hg). The dog was positioned in dorsal recumbency, and compound sodium lactate solution^b was administered at a rate of 10 mL/kg IV per hour.

The dog was clipped and aseptically prepared for a ventral midline celiotomy and median sternotomy. To expose the diaphragm and cranial aspect of the abdominal cavity, a celiotomy was performed that extended from the xiphoid to 5 cm caudal to the umbilicus. On gross inspection, the abdominal contents appeared to be viable and in accurate anatomical position, with the exception of the stomach. The orad stomach, including the gastroesophageal junction and the gastric cardia, was displaced cranially through the esophageal hiatus of the diaphragm and was firmly positioned within the caudodorsal thoracic cavity. The esophageal hiatus was enlarged, and the phrenicoesophageal ligament was considered to be redundant [Figure 3; black arrow]. Further inspection revealed a second defect with regular tissue margins and no evidence of fibrosis; the defect measured 7 cm in diameter and was in the adjacent tendinous portion of the diaphragm [Figure 3; white arrows]. The defect originated directly from the left ventrolateral margin of the esophageal hiatus and allowed herniation of the gastric fundus and a portion of gastric body into the thorax. No membranous sac or fibrous adhesions were found around the herniated gastric structures, but a distinct communication between the abdominal and thoracic cavities was observed.

Reduction of the herniated structures was accomplished by gentle traction on the stomach. This revealed a grossly abnormal surface of gastric serosa with rupture of the short gastric vessels. Evidence of clotting of these ruptured vessels was seen, and no ongoing hemorrhage was observed. The gastric serosa was hyperemic but was assessed visually to be viable, and a partial gastrectomy was not performed. Examination of the caudal thoracic cavity (through the diaphragmatic defect) revealed a large volume of hemorrhagic fluid in the pleural space. Fluid (800 mL) was removed from the thorax via suction. No other inciting cause of the effusion was identified at the time of surgical exploration.

The esophageal hiatus was reduced to a diameter of 2 cm by plication of the diaphragmatic crural muscles ventral to the defect with horizontal mattress sutures of 0 polypropylene.^g Routine herniorrhaphy of the second adjacent defect was accomplished with the same suture material and pattern. An esophagopexy using 2–0 polydioxanone^h in a simple interrupted pattern and a fundic tube gastropexy using a DePezzar-type feeding tube placed through the left abdominal wall were performed. Tube gastropexy was chosen over other gastropexy techniques to allow postoperative gastric decompression and feeding as required.

Prior to closing the celiotomy, thoracocentesis was performed by placing a 14-gauge catheter connected to a three-way stopcock and 50-mL closed syringe system through the muscular portion of the diaphragm. Air was evacuated from the pleural space until negative pressure was established. The volume of fluid removed at this time was negligible. A temporary thoracostomy tube was not placed.

The dog recovered well from surgery with no dyspnea. Initial postoperative treatment consisted of methadone hydrochloride^d (0.2 mg/kg IV q 4 hours for 4 days), cephalexinⁱ (20 mg/kg per os q 12 hours for 7 days), and continuation of IV crystalloid fluids (4 mL/kg per hour of compound sodium lactate solution^b for 4 days). The dog was offered a low-residue diet^j from an upright position and began to eat voluntarily within 2 hours. The peripheral PCV dropped immediately postoperatively to 24% but steadily increased to 29% by 3 days after surgery. Right lateral and dorsoventral thoracic and abdominal survey radiographs were obtained 4 days postoperatively and confirmed an intact diaphragm [Figure 4; a], normal anatomical position of the stomach [Figure 4; b], and minimal pleural fluid [Figure 4; c]. The dog was asymptomatic when discharged from the hospital 4 days postoperatively. The gastrostomy tube was removed after 30 days when radiographs confirmed all structures were still in position.

At 9 months postoperatively, the owners reported the dog to be clinically normal with no further respiratory or gastrointestinal signs.

Discussion

Esophageal HHs are uncommon in small animals. Chinese shar pei and English bulldog breeds are reported to be predisposed to congenital HHs.^{5,12,14,16,20,21} To the authors’ knowledge, this is the first reported case of HH in a Great Dane. Diagnosis of HH is based primarily on history, physical examination findings, and radiographic assessment—although confirmation of a tentative diagnosis can be achieved with a barium contrast esophagram or fluoroscopic studies of the upper gastrointestinal tract.^{2,10,12,15,20–22} Acquired HHs have been reported in animals with a history of trauma, upper airway obstruction, or tetanus; however, the most common cause is a congenital anomaly of the structures forming the esophageal hiatus.^{3–5,10–12,14} Weakened musculature of the right diaphragmatic lumbar crus and/or laxity of the surrounding phrenicoesophageal ligament may predispose to HH.^{9,10,13,16,23} With no history of trauma or known inciting cause in this giant-breed dog, the etiology was suspected to be a distinct variation of a congenital defect, despite the older age at the onset of clinical signs. The second diaphragmatic defect was also suspected to be caused by a concurrent congenital weakening of the adjacent tendinous portion of the diaphragm that, with no history of trauma, likely acutely ruptured secondary to the occurrence of the initial HH. The authors emphasize that this is just supposition, as the underlying cause cannot definitively be identified.

Reported clinical signs of HHs are generally gastrointestinal and tend to be chronic in nature.^{1,2,4,10,12,14,15} In this dog, the primary clinical sign was an acute onset of dyspnea. Gastrointestinal signs were far less prominent features of this dog’s history and may have been unrelated to the HH. In contrast, dyspnea has been previously reported as a less prominent presenting sign attributed to respiratory compromise secondary to aspiration pneumonia and/or lung compression by herniated viscera.^8,9,16,17 In this case, no radiographic evidence of aspiration pneumonia was seen, and the marked dyspnea was assessed as unlikely to be caused by atelectasis alone. The immediate clinical improvement following thoracocentesis also suggests the pleural effusion was a major cause of impaired ventilation. The surgical findings, normal coagulation profile, and absence of hemorrhage postoperatively suggest the hemorrhage arose from the torn, short gastric vessels. To the authors’ knowledge, this is the first reported case of dyspnea caused by hemorrhagic pleural effusion occurring secondary to a HH in a companion animal species.

The majority of human and veterinary reports on HHs reference the type I to IV classification system, although controversy exists not only with respect to the definition of the type IV HH, but also to the definition of the type II, paraesophageal HH. Typically, herniation of the fundus in a type II HH is described as occurring directly adjacent to the esophagus, through the actual esophageal hiatus.^2,3,5–7 In contrast, other reports refer to the “true paraesophageal hiatal hernia,” which has been specifically described by Auger and Riley.⁸ They describe a “true type II” HH as being different from the conventionally defined type II HH, with herniation of the fundus alongside the esophagus, but with protrusion through a distinct diaphragmatic defect, separate from the esophageal hiatus. This interpretation has been supported by Miles et al who described a type II HH as the “displacement of a portion of the stomach through a diaphragmatic defect adjacent to the esophageal hiatus.”¹ Park also described the stomach as “herniat[ing] through, or alongside of, the esophageal hiatus,” and Kirkby et al suggested that “portions of the stomach and perhaps other abdominal organs are displaced into the caudal mediastinum through a defect adjacent to the esophageal hiatus.”^15,17 Important to note is that although these variations of paraesophageal HHs are not commonly distinguished, it appears that a “true type II” HH may represent a distinct form of the disorder. The authors, therefore, suggest official reclassification of the type II HH, wherein the type IIa is described as occurring through the actual esophageal hiatus, and the type IIb is described as a separate entity from the conventionally defined type IIa because of the herniation through an adjacent, yet distinct, diaphragmatic rent.

According to these descriptions and newly proposed classification, the case presented here is an unusual type III HH—that is, a combination of a conventional type I through the esophageal hiatus and a type IIb through a distinct diaphragmatic defect adjacent to the hiatus.

The prognosis for dogs with HHs is variable and dependent on the chronicity of clinical signs, the degree and type of herniation, and the method of medical and/or surgical treatment.^2,12 Based on the outcome of this case, the prognosis for resolution of hemorrhagic pleural effusion and other clinical signs occurring secondary to a type III HH may be good with prompt and effective surgical intervention.

Conclusion

An unusual case of combined axial and paraesophageal (type III) HH was diagnosed in a mature Great Dane. The dog was presented with acute onset of dyspnea. Hemorrhagic pleural effusion caused by ruptured short gastric vessels of the herniated stomach was identified to be the primary cause of the acute respiratory compromise. The dog was treated surgically with full resolution of signs and no recurrence after 9 months. The paraesophageal component of the type III HH in this case was herniation of the stomach alongside the esophagus, through an adjacent and distinct defect in the diaphragm, which the authors suggest should be reclassified as a type IIb HH.

Acknowledgments

The authors are grateful to Mr. Cameron J. Broome for his encouragement and advice in the preparation of this manuscript. The authors also thank Ms. Marieke Helleman, the veterinary surgeon, who referred this case for further investigation and treatment.