Spontaneous Resolution of Postoperative Chylothorax Following Surgery for Persistent Right Aortic Arch in Two Dogs
Two young dogs underwent surgical management of a persistent right aortic arch (PRAA) and developed chylothorax postoperatively. In both cases, the surgical procedure and anesthetic recovery were uncomplicated and routine. Following surgery, both patients appeared bright, alert, responsive, and previous signs of regurgitation had resolved. Dyspnea and tachypnea developed 1–2 days postoperatively in each patient, and chylous effusion was detected on thoracocentesis. For each case, a diagnosis of chylothorax was based on cytology and triglyceride concentrations of the aspirated pleural fluid. Similar protocols for monitoring were used in the treatment of each patient’s chylothorax. The duration and volume of chylous effusion production were closely monitored via routine thoracostomy tube aspiration. Both dogs rapidly progressed to recovery with no additional complications. With diligent monitoring, chylothorax secondary to surgical trauma can resolve in a rapid, uncomplicated manner.
Introduction
Chylothorax is the abnormal accumulation of chyle in the pleural cavity.1 The normal flow of chyle originates in the proximal small intestines and enters into lymphatic vessels via the intestinal lacteals. Chyle is then directed toward the principal lymphatic channel, the thoracic duct. The thoracic duct arises as a cranial continuation of the cisterna chyli, the dilated, sac-like portion of the lymph channel that lies retroperitoneally. The cisterna chyli is considered to transition into the thoracic duct as it attains its minimum width near the level of the diaphragmatic crura. Termination of the thoracic duct occurs at the lymphaticovenous junction on the cranial vena cava, where chyle enters into venous circulation.2
It is well established that the anatomy of the thoracic duct is inconsistent in the dog, as more than 20 variations of origin, course, and termination have been demonstrated.2 Most anatomic variation is found at the caudal portion of the thoracic duct as it arises from the cisterna chyli in a plexiform manner.2 The unpredictable nature of this structure in the dog likely increases the risk of thoracic duct trauma during intrathoracic surgical procedures.
During embryologic development, vascular ring anomalies can develop in up to 20% of dogs and cats.3,4 Of the clinical vascular ring anomalies, persistent right aortic arch (PRAA) accounts for up to 95% of cases in dogs.5 Overall success following surgery has improved throughout the last 2 decades. Muldoon et al. (1997) reported an excellent outcome in 92% of cases and a good outcome in 8% of cases following surgery.6 The most common complication postoperatively is persistent regurgitation.4
The purpose of this case report is to describe two separate cases of chylothorax following surgical management of a PRAA in two young dogs. Intraoperative surgical trauma to anomalous branches of the thoracic duct likely contributed to the postoperative chylothorax in both cases.
Case Report
Case 1
An 8 wk old male bichon frise was referred to Kansas State University Veterinary Medical Teaching Hospital for evaluation of a vascular ring anomaly. The young dog had a 1 wk history of intermittent regurgitation of increasing frequency and two documented episodes of hypoglycemia. Prior to presentation, pertinent diagnostics by the referring veterinarian included a barium contrast esophagogram, which revealed megaesophagus cranial to the base of the heart. In addition, a complete blood count (CBC) and a serum biochemical profile, which revealed a mild neutrophilia (12.71 × 109/L; reference range, 3–12 × 109/L) and a mild monocytosis (2.54 × 109/L; reference range, 0.3–2 × 109/L).
On physical examination, the rectal temperature, heart rate, and respiratory rate were within normal limits. The only significant abnormalities were a low body condition score (2/5) and mildly harsh right lung sounds. CBC and serum biochemical analysis revealed hypoglycemia (3.22 mmol/L; reference range, 4.05–6.27 mmol/L), as well as a mild anemia (30%; reference range, 37–55% ) and hypoproteinemia (45 g/L; reference range, 60–80 g/L). Pre- and postprandial bile acids were normal.
Orthogonal radiographs of the thorax and abdomen were performed. No abnormal findings were noted on abdominal radiographs. Thoracic radiographs revealed a mild interstitial pattern in the right middle lung lobe, megaesophagus cranial to the base of the heart, and the trachea was deviated dorsally and to the left. Based on signalment, history, clinical signs, physical examination findings, and radiograph findings, the diagnosis was consistent with aspiration pneumonia and a suspected vascular ring anomaly.
The patient was admitted to the intensive care unit and given an IV dextrosea bolus (0.5 mg/kg diluted 1:1 with 0.9% salineb) and administered a corn syrupc orally. Additional treatments that were initiated included administration of a crystalloidd with 2.5% dextrose (70 mL/kg/day IV), famotidinee (0.5 mg/kg IV q 12 hr), sucralfate slurryf (500 mg per os [PO] q 8 hr), and ampicillin Na/sulbactam Nag (22 mg/kg IV q 8 hr). Serial blood glucose measurements indicated persistent mild hypoglycemia, and the dextrose concentration in the crystalloid was increased to 5%. Additional blood glucose samples were normal. The patient consistently maintained a strong interest in food; however, he regurgitated a large amount of his ingested contents following each meal.
Three days after admission, a left lateral thoracotomy was performed through the left fourth intercostal space by a third year surgery resident under the supervision of a board-certified veterinary surgeon. The PRAA was confirmed. The ligamentum arteriosum was identified, and two circumferential ligatures using 4-0 polypropyleneh were placed. The ligamentum arteriosum was transected between the circumferential ligatures. An oroesophageal tube was then placed to better visualize the remaining constricting fibroelastic bands superimposed over the narrowed portion of the esophagus. As the fibroelastic bands were broken down, a serous fluid of unknown origin began to accumulate in the thoracic cavity. The amount of fluid was small, and the origin could not be determined. Due to the small amount of serous fluid, a thoracostomy tube was placed using an 8-French red rubber catheteri and secured with 4-0 nylonj.
The patient had an uneventful recovery in the intensive care unit. Analgesic therapy was provided using fentanylk (5 μg/kg/hr IV), tramadoll (5 mg/kg PO q 8 hr), and meloxicamm (0.1 mg/kg subcutaneously q 24 hr). Over the next 15 hr, the puppy ingested 10 mL portions of soft food at 3 hr intervals without any episodes of regurgitation. The preoperative protocol of famotidine, sucralfate, and ampicillin Na/sulbactam Na was continued postoperatively at the same dose, frequency, and route. Ten hr postoperatively, the patient was clinically normal, and the thorax ausculted similar to before surgery. No fluid or gas had been aspirated from the chest tube and it was removed.
Sixteen hr after anesthetic recovery, the patient had increased respiratory effort. At that time the patient maintained a bright, alert, and responsive attitude, and cardiac sounds were normal on auscultation. Over the next several hr, the patient’s respiratory effort and rate progressively increased and the patient developed muffled heart sounds, hypoxemia (O2 saturation was 84%), inappetence, and lethargy.
Two thoracocenteses were performed, and a total of 30 mL of milky white fluid were aspirated from the pleural space. Thoracic radiographs revealed evidence of pleural effusion. Analysis of the aspirated fluid included a total protein of 38 g/L, free lipid droplets and mature lymphocytes on cytologic evaluation, and a significantly elevated triglyceride concentration (32.62 mmol/L; reference range in serum, 0.31–2.43 mmol/L). A diagnosis of chylous effusion was made based on those findings. A right thoracostomy tube was placed and aspirated q 4–6 hr.
The volume of aspirated chylous fluid was closely monitored for any notable changes in the amount of fluid production. With each successive day of monitoring, there was a marked decline in the volume of fluid production. Approximately 1 day after surgery, a total of 81.38 mL (104.33 mL/kg) of chyle were removed from the pleural space; 2 days after surgery, 52.4 mL (66.33 mL/kg) of chyle were removed; 3 days after surgery, 10.3 mL (12.88 mL/kg) were removed; 4 days after surgery, 1.7 mL (1.79 mL/kg) were removed; and 5 days following surgery, no chylous effusion was aspirated from the thoracostomy tube.
The patient continued to receive small meals q 3 hr. Following surgical management of the PRAA, only one episode of regurgitation occurred. Blood glucose levels were checked periodically for the initial 3 days postoperatively and were consistently within normal limits. After 5 days of monitoring, the thoracostomy tube was removed, and the patient continued to make substantial progress in overall condition.
At the time of discharge, all previously administered medications had been discontinued, and the patient was sent home on amoxicillin trihydrate/clavulanate potassiumn (13.75 mg/kg PO q 12 hr for 4 wk) for continued treatment of aspiration pneumonia. Eight days after being discharged, the patient returned for a follow-up evaluation. At that time, thoracic radiographs were unremarkable, and the young dog was found to be healthy on physical examination.
Case 2
A 12 wk old female Maltese was presented to Purdue University Veterinary Teaching Hospital for evaluation of an acute onset of postprandial regurgitation that started when the patient was transitioned to solid food at 9 wk of age. On physical examination, the rectal temperature, heart rate, and respiratory rate were within normal limits. The only significant physical exam abnormality was a low body condition score (2/5). CBC and serum biochemical analysis revealed mild anemia (35.7%; reference range, 37–55%) and hypoproteinemia (55 g/L; reference range, 60–80 g/L). Thoracic radiographs revealed that the trachea was deviated dorsally and slightly to the left, with mild intrathoracic cranial esophageal dilation. Based on signalment, history, clinical signs and radiograph findings, the diagnosis was consistent with a vascular ring anomaly.
The patient underwent surgical management by a board-certified surgeon via a left thoracotomy made at the fourth intercostal space in the manner described above. A PRAA was confirmed. The ligamentum arteriosum was double ligated with 3-0 silko prior to transection. Intraoperatively, no abnormal fluid accumulations were noted. A thoracostomy tube was placed prior to routine closure of the thoracotomy site. Anesthetic recovery was uneventful, and postoperative medications included hydromorphonep (.05 mg/kg subcutaneously q 5 hr), cefazolinq (22 mg/kg IV q 8 h), and famotidine (.5 mg/kg IV q 12 h). The thoracostomy tube was removed the night of surgery. The following afternoon, the patient developed dyspnea. Over a 12 hr period, three thoracocenteses were performed, rendering a total of 40 mL of fluid from the pleural space. Cytology results indicated that the majority of the cell population was lymphocytes, with a total protein of 43 g/L. Chylous effusion was diagnosed based on the characteristic milky white appearance of the fluid, cytologic evaluation, and significantly elevated triglyceride concentration of the pleural fluid (5.6 mmol/L). A thoracostomy tube was placed, and the volume of aspirated fluid was closely monitored. On day 1 after chest tube placement, a total of 2.52 mL (3.15 mL/kg) of chyle was removed from the pleural space; on day 2, 2.57 mL (0.71 mL/kg) of chyle were removed; on day 3, .3 mL (.38 mL/kg) were removed; and on day 4, 0.1 mL (0.13 mL/kg) were removed. Five days after surgery, no chylous effusion was aspirated from the thoracostomy tube and it was subsequently removed.
As in case 2, this patient also resumed eating and had no documented episodes of regurgitation. Postoperative blood glucose levels were within normal limits. The patient was discharged 1 wk after initial presentation. At that time, all previously administered medications had been discontinued, and the patient was discharged with amoxicillin trihydrate/clavulanate potassium (13.75 mg/kg PO q 12 h). Approximately 10 mo later, the dog presented to Purdue University Teaching Hospital for an elective ovariohysterectomy. At that time, the dog was clinically healthy, with no long-term complications from either the PRAA repair or chylothorax.
Discussion
Although idiopathic chylothorax is the predominant cause of chylothorax in veterinary medicine, several specific etiologies have been reported in the literature. They include the development of chylothorax secondary to neoplasia, congenital abnormalities of the thoracic duct, fungal infections, lung lobe torsion, heartworm disease, thrombosis of the cranial vena cava, right-sided heart failure, diaphragmatic hernia, trauma, constrictive pericarditis, peritoneopericardial diaphragmatic hernia, thrombosis secondary to jugular catheter, vascular access port, endocardial pacing wire placement, tricuspid dysplasia and atrial septal defect, and double-chambered right ventricle and tricuspid dysplasia.7–23
After a thorough search of the veterinary literature, only one other veterinary case of idiopathic chylothorax occurring apparently secondary to surgical intervention has been documented. In that previously reported case, a cat developed chylothorax secondary to ligation and transection of the left brachiocephalic vein during a thoracotomy for intrathoracic trachea surgery.24 The cat’s chylothorax was likely a sequella of impaired lymphaticovenous drainage, whereas direct trauma to an anomalous branch of the thoracic duct was the probable etiology in both of the cases reported herein.
Human medicine subclassifies chylothorax etiologies into congenital, traumatic (blunt, penetrating, surgical, diagnostic procedures), neoplastic, infectious, and miscellaneous according to the DeMeester classification scheme.25,26 Surgery constitutes the majority of cases from traumatic etiologies. In human medicine, it has been documented that postoperative iatrogenic chylothorax occurs in 0.25–0.5% of cardiovascular surgeries and in 3–4% of esophageal surgeries.25,27–30
Confirmation of chylothorax is made based on fluid analysis showing high lymphocyte counts and high fat content.31,32 More specifically, it is an elevation in chylomicrons that supports the diagnosis of chylothorax.31,33 Additionally, one study in human medicine found that a pleural effusion triglyceride concentration > 1.24 mmol/L is highly suggestive of a chylous effusion, whereas a value < 0.56 mmol/Lwas suggestive of a nonchylous effusion.34 There are numerous additional methods to confirm chylothorax that largely depend on clinician preference. The presence of lipid droplets can be found on cytology with the use of Sudan III staining. Ether extraction can also be used to identify chylous fluid by dissolving the chylomicrons, which transitions the milky white chyle into a clear fluid. Finally, a pleural cholesterol:triglyceride ratio < 1:1 and/or a triglyceride concentration that is higher in the pleural fluid compared with the serum are indicative of chylous effusion.17,35 Although lymphocytes are the predominant cell type of chylous fluid, chronicity can result in the presence of more neutrophils.36 In both of those cases, the diagnosis of chylothorax was confirmed via cytology and evaluation of significantly elevated pleural fluid triglyceride concentrations.
In 1993, Hodges et al. published an experimental study of induced chylothorax on six healthy dogs.37 A 2.5 cm longitudinal incision was made at the caudal thoracic duct in three dogs, and a complete transection of the caudal thoracic duct in the remaining three dogs. In the patients with the longitudinal incision, chylothorax resolved within 5 days. In the complete transection patients, the chylothorax resolved by 10 days postoperatively. Lymphangiography studies showed that lacerated ducts were patent compared with none of the transected thoracic ducts. During histopathologic examination, healed thoracic ducts could not be differentiated from normal thoracic ducts. Similar to Hodges et al.’s experimental study, the leakage of chylous effusion subsided by day 5 following surgery.
As in human surgery, damage to the thoracic duct is rarely recognized intraoperatively because the duct is relatively collapsed in the fasting state and the content of the chyle blends with normal serous fluid in the thorax. Humans with insidious leaks during surgery often present several days postoperatively for evaluation of chest discomfort and dyspnea.38 Consistent with human medicine, thoracic duct trauma was not recognized during surgery in both cases, and the diagnosis of chylothorax was delayed in the patients described in this report by 1–2 days after surgery. Fluid production < 2 mL/kg/day and a general trend downwards are traditional guidelines for removing a chest tube.39 Because thoracostomy tubes placed after PRAA surgery is primary for evacuation of air, both tubes were directed dorsally during placement. On retrospective evaluation, the original thoracostomy tube could have remained in place until at least 2 days postoperatively, allowing time for detection of chylous effusion and/or the thoracosotomy tube could have been directed more ventral in the cranial thorax. Chylothorax is a rare complication, but it doesn’t seem warranted to prolong placement of the chest tube. It seems reasonable that clinicians should consider keeping patients a minimum of 36–48 hr postoperatively before discharge for monitoring.
It has been demonstrated in humans that the volume of chyle drainage on the fifth day after surgery predicts the chances of success. If the drainage is < 10 mL/kg/day, the chances of resolution with monitoring are high.40 Only one of the patients described in this report exceeded that volume of chyle production; however, by the fifth day after surgery, the amount of aspirated fluid from each patient was far below that value.
Compared with other etiologies, surgical traumatic causes of chylothorax have a shorter interval from occurrence to diagnosis and a shorter duration and smaller volume of chyle leakage.41 As was evident in the cases reported herein, conservative management of surgical traumatic chylothorax is often successful in human patients provided that nutritional and fluid requirements are met.41 Leakage of chyle from the lymphatic system leads to a concurrent loss of proteins, immunoglobulins, fat, vitamins, electrolytes, and water.25 Therefore, monitoring is necessary to identify and address fluid and nutritional deficits. The patients described in this report were frequently offered oral feedings in addition to receiving IV fluid therapy with 2.5–5% dextrose. Case 1 was documented to regurgitate once postoperatively, and no episodes of regurgitation were documented in case 2. Both patients ate ravenously and were fed q 3–4 hr. Postoperative blood glucose levels on both patents were normal. A decision was made to not pursue additional blood work and supplemental forms of nutrition because of each patients’ good appetite, one episode in case 1 and no episodes of regurgitation in case 2, normoglycemia measurements, small patient body weight, and financial considerations.
Although these two cases of chylothorax were likely secondary to surgical trauma, one review of chylothorax cases in dogs found that 70% were of unknown etiology.42 Medical management of idiopathic chylothorax is typically unrewarding, and surgical intervention is often required.4 Thoracic duct ligation is the most widely accepted technique; however, various adjunctive surgical procedures have been proposed.43 Those procedures include subtotal pericardectomy, thoracic omentalization, cisterna chyli ablation, active and passive pleuroperitoneal shunting, pleurovenous shunting, and thoracic duct embolization.43–51
The inability to entirely visualize the thoracic duct and its branches intraoperatively increases the risk of iatrogenic trauma. In the veterinary literature, both lymphangiography and methylene blue coloration of the lymphatic vessels have been described as means to better distinguish those structures during surgery.37,51–59 Currently, the above-described techniques are recognized for their use in thoracic duct ligation procedures; however, further investigation may be warranted to explore their potential role as a prophylactic approach to protect lymphatic vessels during surgical correction of a PRAA.
Principles of medical management of chylothorax include providing reduction of chyle flow, drainage of the pleural cavity, nutritional support, and prevention of septic complications. All of those components of medical therapy were provided to the two patients described in this report and likely contributed to the rapid, uneventful resolution of their condition. Although definitive confirmation of thoracic duct trauma could not be obtained in those patients, the manifestation of chylothorax was most consistent with an etiology of surgical trauma based on the time of onset (1–2 days postthoracotomy) and the resolution within 5 days. The first surgery was performed by a third year surgery resident under the direct supervision of a board-certified veterinary surgeon. The second surgery was performed by a separate board-certified veterinary surgeon. Both board-certified surgeons had performed the procedure many times before without complications and aggressive surgical technique was not suspected.
Conclusion
Chylothorax has not been previously described as a complication associated with surgical management of a vascular ring anomaly. Both of the cases presented in this report involved young, small-breed canines that had a similar history, clinical presentation, and course of chylothorax onset, progression, and resolution. By emphasizing the parallels between the two cases, future cases of similar nature can be approached with increased attentiveness and surveillance for complications, such as chylothorax secondary to surgical intervention. Both of these cases demonstrated that surgical intervention is not necessary for postoperative chylothorax following PRAA surgery.
Contributor Notes
L. Barbur’s present affiliation is the Veterinary Teaching Hospital, University of Georgia, Athens, GA.
