Introduction

Foreign body–associated esophageal obstruction is a common emergency in dogs and may result over time in pressure necrosis of the esophageal wall and subsequent perforation and death.^1–4 Bones are the most commonly encountered foreign bodies in dogs, but fishhooks, rawhide chews, apples, wooden skewers, pig ears, dog biscuits, needles, safety pins, and others are occasionally reported.^1,3,5–8 Diagnosis of esophageal foreign body (EFB) obstruction is made with plain radiography or endoscopy. A total of 50–80% of the cases reported are located in the distal esophagus.^1,3,5,6,8-10 Initial attempts for distal EFB retrieval are through rigid or flexible endoscopy or fluoroscopically guided forceps.^1,3,5,6,8,9 Foreign bodies can be extracted through the mouth or advanced into the stomach where they will be digested or removed via gastrotomy.^1,3,5,6,8,9 In case of failure, surgical removal is attempted through a transthoracic esophagotomy, midline celiotomy gastrotomy, or a transdiaphragmatic gastrotomy (TG) approach.^11–13

The TG approach has not received much attention from the veterinary community as only three cases of foreign body retrieval in dogs have been published in the literature.¹¹ In the present retrospective study, the results of TG for the retrieval of distal EFBs, postoperative complications, and long-term outcome in 13 dogs are reported. We hypothesized that the TG approach for the removal of distal EFBs is an effective surgical option with no significant complications.

Materials and Methods

The records of 13 dogs who were referred to the Clinic of Companion Animals with TG for retrieval of distal EFBs between January 1997 and December 2016 were reviewed. All dogs had unsuccessful attempts for retrieval of foreign bodies or advancement into the stomach through a fiberoptic or rigid endoscopy; these attempts were performed prior to surgery under the same anesthetic period. Diagnosis of EFB was made by radiographic examination composed of two laterals and a ventrodorsal view and confirmed on endoscopy and surgery. All dogs underwent routine hematology and biochemical analysis before surgery. All foreign bodies included in the study were located in the distal esophagus immediately cranial to the diaphragm. Data retrieved from the records included signalment, history, clinical findings, duration of signs, endoscopic findings immediately after surgery, laboratory examinations, surgical findings, medical treatment, complications, duration of hospitalization, and follow-up/outcome. Endoscopic assessment of esophagitis after foreign body retrieval was based on the Savary-Miller classification.¹⁴ Long-term follow-up was performed by telephone communication with the owner or referring veterinarian. Owners were questioned on any signs of dysphagia, regurgitation/vomiting, or surgical wound complications at the time of follow-up. Outcome was considered as excellent when no clinical signs relevant to stricture formation were reported. All dogs received IV Ringer’s Lactated solution before anesthetic induction. The dogs were premedicated with acetylpromazine^a and butorphanol^b. Anesthesia was induced with IV propofol^c and maintained with isoflurane in oxygen under intermittent positive-pressure ventilation. Cefazolin^d (20 mg/kg) was administered IV for prophylaxis. A left lateral thoracotomy was performed through an eighth or ninth intercostal space based on the location of the foreign body. Following entrance to the thoracic cavity and after the left caudal lung lobe was packed off and moved cranially with a moistened laparotomy sponge, the distal esophagus was observed. The distal esophagus was dissected free, and two 1/4ʺ Penrose drain tubes were placed around the esophagus to allow manipulation of the esophagus and foreign body, rotation of the distal esophagus, and observation of the esophageal circumference. Two stay sutures were placed in the diaphragm, a linear incision was made at the central tendinous region, and the stomach was located, elevated through the diaphragmatic incision, and a 6 cm gastrotomy was made between the greater and the lesser curvatures of the stomach 3 cm away from the cardia between two stay sutures. Upon entering the gastric lumen, the suction tip was inserted and all gastric contents were removed. While the hand of an assistant surgeon manipulated the foreign body through the distal esophagus and propelled it toward the cardia, a long-curved Kocher hemostat was placed through the gastrotomy incision and into the distal esophagus to grasp and retrieve the foreign body. The gastrotomy incision was closed in two layers using a 2 or 3 metric polydioxanone^e suture. The thoracic cavity was lavaged with warm normal saline, the diaphragm was closed in a continuous 3 metric polydioxanone suture, and thoracotomy closure was achieved with simple interrupted 3 or 3.5 metric polydioxanone sutures following an intercostal bupivacaine block. A tube thoracostomy was placed in all dogs before thoracotomy closure for air and fluid evacuation as well as for intermittent intrapleural bupivacaine infusion through the tube. Aspirations were done every 3 hr after thoracostomy tube placement. Tube retrieval was performed if minimal air or fluid was aspirated following two consecutive unsuccessful aspirations. Closure of the skin and subcutis was routinely performed. Fentanyl^f or morphine^g was given for postoperative analgesia. All dogs received sucralfate slurry^h (1 g/20 kg q 12 hr), ranitidineⁱ (2 mg/kg q 12 hr), or omeprazole^j (1 mg/kg q 24 hr) orally and metoclopramide^k (0.2 mg/kg q 8 hr), subcutaneously, during hospitalization and then orally for 5–12 days postsurgery depending on the severity of esophagitis.

Results

Breeds represented were West Highland white terrier (n = 5) and one of each of boxer, English setter, German shorthair pointer, Labrador retriever, mixed-breed dog, Pekingese, Pomeranian, and poodle. Six dogs were females, 3 were spayed females, 3 were males, and 1 was a castrated male. Mean age at presentation was 38.6 mo (range 3–156 months) and mean weight was 10.5 kg (range 2.5–25 kg). Mean duration of clinical signs from onset to presentation was 5.8 days (range 1–10 days). Physical examination abnormalities included depression (n = 8), inappetence (n = 4), regurgitation (n = 13), vomiting (n = 5), diarrhea (n = 2), esophagodynia (n = 1), fever (n = 2), and hypothermia (n = 1). Case 1 showed anemia (reference range 37.1–55.00) and case 6 had neutrophilic leukocytosis (reference range 6000–17,000 K/μL). EFBs were visualized in plain radiographs in all 13 dogs, and increased soft tissue density around foreign body was seen in 8 dogs. Rigid and fiberoptic endoscopic attempts to remove the foreign bodies failed in all 13 dogs. Esophageal obstruction was caused by bones in 10 dogs, and fishhook entrapment was seen in 3 dogs. Intercostal thoracotomy space was based on the location of the EFB as was seen in thoracic radiographs. TG was performed in 10 dogs through the eighth intercostal space thoracotomy and in 3 dogs through the ninth intercostal space. Bones, including pork chops, lamb chops, and calf vertebrae, were removed from 10 dogs and fishhooks were removed from 3 dogs. In case 5, the gastrotomy failed to retrieve a bone foreign body and a distal esophagotomy was performed for its removal. Esophagotomy was closed in one layer with simple interrupted 2 metric polydioxanone sutures. Postsurgical esophagitis as assessed through endoscopy in all dogs detected grade 1 esophagitis in 5 dogs, grade 2 in 2, grade 3 in 3, and grade 4 in 3. Twelve dogs began eating in a mean time of 35 hr after surgery. Four dogs were introduced to soft food 24 hr postsurgery, 6 dogs in 48 hr, 1 dog in 72 hr, and 1 dog (case 6) on day 7 postsurgery. This case was anorectic on day 3 after surgery. Gastrostomy tube placement was declined by the owner. Thoracic radiographs showed no signs of pleural effusion, and no signs of perforation were detected. The dog received parenteral nutrition on day 4 postsurgery for 3 days and started to eat on day 7 postsurgery. Thoracostomy tubes were removed after a mean duration of 15 hr (range 6–24 hr). Case 5 developed pyothorax and died 1 day postsurgery. This dog also had an esophagotomy for the removal of a foreign body. Esophagoscopy performed following completion of esophageal closure failed to detect any perforation. The dog developed pyothorax within 24 hr after surgery, which was diagnosed through cytology of pleural effusion that was obtained through the thoracostomy tube. Necropsy revealed esophageal perforation in the right lateral esophageal wall. Case 10 showed regurgitation 6 days after surgery despite the conservative management he received. A new esophagoscopy performed 2 wk after surgery showed esophageal stricture, which was managed with two balloon dilatations spaced 2 wk apart and prednisolone^l, metoclopramide, and sucralfate for 10 days. The dog recovered completely from stricture and was free of clinical signs of esophageal disease 12 mo after surgery. Mean hospitalization time for the 12 dogs was 3.4 days (range 2–10 days). Twelve dogs were available for follow-up. All 12 dogs were free of clinical signs after a mean follow-up time of 17 mo (range 12–48 mo). Clinical data of 13 dogs are presented in Table 1.

TABLE 1 Clinical Data of 13 Dogs Who Had Transdiaphragmatic Gastrotomy for the Retrieval of Distal Esophageal Foreign Bodies

View Fullscreen

Discussion

In the study reported here, TG provided successful extraction of the majority of EFBs lodged in the distal esophagus. In a recent report concerning endoscopic removal of gastric foreign bodies and EFBs in dogs, those that weigh <10 kg showed a significantly increased risk of complications.³ In the study presented here, the mean age of our dogs was 3.2 yr, which was similar to the previously reported age^15–17 but lower than the age of >4 yr reported by others.^{3, 5, 6, 7, 8, 10, 12, 13, 18}

Clinical signs reported in the dogs in our study are generally consistent with those reported by other authors.^1,3,5–7,10 Mean duration of clinical signs of 5.6 days of our study was greater than the mean 3.9 days of those that reported surgical treatment of EFB^10,12,13 but lower than those of a previous study that reported a mean duration of 6.4 days for 7 dogs who had esophagotomy for foreign body removal.¹

The presence of an EFB provokes esophageal inflammation that may result in esophagitis, stricture formation, or partial- or full-thickness perforation and subsequent pneumothorax, pyothorax, sepsis, or death.^2,3,6,19 Duration of clinical signs from onset to presentation of >3 days correlated with an increased risk of complications or death in two recent studies of gastric foreign bodies or EFBs.^3,8 However, this figure was disputed by another study of 222 dogs with EFB in whom the duration of clinical signs was not associated with risk of death.⁷

Esophagoscopy is a noninvasive procedure that is initially performed for the removal or advancement of EFB in the stomach that is lodged in the distal esophagus.^{1,3,5–8,10,20,21} Esophagoscopy was initially performed in all dogs of the present study but failed to retrieve or push the foreign body into the stomach. Possible reasons for endoscopic retrieval failure in this study included firm entrapment of the foreign body in the esophagus to allow uncomplicated retrieval with the esophageal forceps, suspicion of esophageal wall perforation, or high risk of esophageal or major vessel laceration.^7,10

Surgical extraction of distal EFB can be accomplished via transthoracic esophagotomy or gastrotomy.^12,13 Gastrotomy can be performed transdiaphragmatically or through a midline celiotomy approach.^11,13 Esophagotomy may be related to a high incidence of postoperative complications including incisional dehiscence.^{1,2,4,15,16,18} Lack of esophageal serosa, esophageal segmental blood supply, paucity of omental coverage and esophageal motility, and tension at the incision are considered risk factors for the development of dehiscence.⁴ However, recent studies report a successful clinical outcome following esophagotomy for the retrieval of distal EFB.^10,12 In a study by Deroy et al., the removal of EFB by esophagotomy or esophagoscopy had a similar outcome.¹⁰ Gastrotomy through a cranial midline celiotomy was recently reported for the extraction of distal EFB.¹³ In the study by Artsens et al., a gastrotomy incision between the lesser and the greater curvatures was made. The incision was of such length to allow easier identification of the cardia and introduction of one hand into the gastric lumen to manipulate long forceps for grasping the EFB. With this technique, a successful extraction of EFB was performed in all 12 cases.¹³ In the study reported here, TG was successful in EFB extraction in 12 out of 13 cases. TG was elected in the present study for removal of distal EFBs for the following reasons: The gastrotomy performed following intercostal thoracotomy and diaphragmatic incision was closer to the cardia to allow easier inspection of the foreign body and of a shorter length than that of Aertsens et al.¹³ With this approach, inspection of the lateral and medial wall of the distal esophagus to exclude perforations and manipulation of the distal esophagus was accomplished to allow mobilization of the foreign body toward the cardia and easier extraction of the foreign body through gastrotomy. Gastrotomy closure is associated with a low incidence of dehiscence compared with esophagotomy.^4,13 Left intercostal thoracotomy through the eighth or ninth space allowed for decision making of EFB extraction through esophagotomy with the same intercostal incision if gastrotomy attempts failed. However, our approach required an intercostal thoracotomy, which is more invasive and painful than gastrotomy through a midline celiotomy.^22,23 In our approach, the possible risk of gastric contents spillage into the thoracic cavity was greater than midline celiotomy gastrotomy.

In the study presented here, mean time for thoracostomy tube removal was less than thoracostomy tube removal in the transthoracic esophagotomy study.¹¹ This difference may be because a few dogs in that study had pleural effusions preoperatively that required monitoring. This was in contrast to our dogs, for whom no pleural effusions were evident preoperatively; only one dog developed postoperative pleural effusion, but this dog died within 24 hr after surgery.

Most of our dogs began eating within 48 hr after surgery; this figure was longer than that of the gastrotomy study, which food was offered 12 hr after surgery, and shorter than that of transthoracic esophagotomy study, in which dogs began eating within 3 days after surgery.^12,13 It seems that gastrotomy performed transdiaphragmatically or through ventral midline for EFB extraction allows for a quicker introduction to feeding than esophagotomy.

In the present study, the survival rate following extraction of EFB through TG was 92%, a figure that compared favorably with the transthoracic esophagotomy and midline celiotomy gastrotomy, for which survival rates ranged from 93 to 83%.^12,13

Postoperative reported complications following surgical extraction of EFB include esophagitis, esophageal stricture, esophageal perforation, mediastinitis, pyothorax, dehiscence or leakage of esophagotomy wound, and death.^{1,2,8,12,13,15,19}

Esophagitis is a common complication following endoscopical or surgical EFB extraction.^10,19 In a study of 60 dogs with esophagitis following EFB removal, 2 patients with mild esophagitis developed aspiration pneumonia, 5 out of 6 with moderate to severe esophagitis showed stricture, and 1 out of 6 developed a diverticulum.¹⁹ In our study, despite treatment to prevent esophagitis, 1 dog with severe esophagitis developed perforation that was a result of the EFB extraction or the inflammatory process by itself and died, and another dog developed esophageal stricture that was managed successfully with balloon dilatation. Medical management of esophagitis is recommended in all cases following EFB extraction to prevent serious complications.¹⁹

Perforation of the esophagus is a serious complication of EFB following surgical extraction.^2,12,15 One dog of the present study died of pyothorax associated with a perforation of the contralateral side of the distal esophagus after the EFB was extracted through an esophagotomy. The dog died within 24 hr following surgery. Inspection of the distal esophagus and postoperative esophagoscopy failed to reveal any perforation. Repetitive endoscopic attempts for EFB retrieval is considered a risk factor for esophageal perforation.⁷

This study has some limitations including its retrospective nature, the small population size, and no availability of endoscopic or imaging follow-up to evaluate for the presence of esophageal stricture.

Conclusion

TG is an effective technique for distal EFB extraction. TG allowed for the visualization of the cardia and distal EFB extraction, inspection of the entire distal esophagus, manipulation and mobilization of the foreign body through the esophageal wall toward the stomach, and decision making for foreign body extraction, with the same thoracotomy, through esophagotomy, in case of gastrotomy failure. In most cases, EFB extraction was successful and clinical signs resolved.