Porcupine Quill Migration in the Thoracic Cavity of a German Shorthaired Pointer

Introduction

Dogs commonly encounter porcupines in certain regions of the United States. Quills are loosely attached to the porcupine's body and are readily released upon contact with an assailant. At the tip of those quills are backward-facing barbs that facilitate forward migration through body tissues if they are not promptly removed.^1,2 Quills frequently break off at the skin surface, which makes locating and removing the quills more difficult, and resulting in progressive quill migration into body tissues.^1,3,4 Case reports in dogs indicate quills can migrate to many different locations, including the thoracic cavity, the orbit, joints, the temporal fossa, the brain, and the vertebral canal.^4–8 Identification and removal of the remaining quills is imperative for a successful outcome; however, the most appropriate imaging modality to identify and locate porcupine quills remains questionable. Depending on the circumstance, one or many imaging modalities may be necessary to optimize visualization. Quills are not generally seen with standard radiographic techniques; however, some reports have described an effective use of ultrasound.^4,5 Despite the availability of multiple options, there is currently no imaging modality that consistently and accurately recognizes quills. This case report describes porcupine quill migration into multiple lung lobes, the mediastinum, and heart of a 7 yr old German shorthaired pointer and the use of computed tomography for identification of quills.

Case Report

A 7 yr old, spayed female, German shorthaired pointer weighing 23 kg was referred to the Emergency Service of the Ryan Veterinary Hospital at the University of Pennsylvania with 2 days of lethargy and acutely increased respiratory effort. Two weeks prior to presentation, the dog had had encountered a porcupine while hunting in North Dakota. Within 24 hr of the encounter, the dog was examined by a veterinarian and the quills were removed from the left forepaw, sternal area, and mouth. Ten days later, several more quills were removed from the dog's thoracic region and the left axilla by a second veterinarian. The dog was subsequently treated with enrofloxacin^a, and deracoxib^b. Following the procedure performed by the second veterinarian, the patient experienced an acute episode of retching, wheezing, and diarrhea. That event was self-resolving, but she remained lethargic and was taken back to the hospital 2 days later when she began exhibiting mild respiratory distress, taking short, shallow breaths.

At the time of presentation to the Emergency Service, the dog's vital signs were within normal limits (temperature, 39.0° C; heart rate, 150 beats/min; respiratory rate, panting), appropriate mucous membrane color (pink and moist), and rapid capillary refill time (<1 sec). It was also noted that she had bounding pulses; harsh lung sounds in the left dorsal, right dorsal, and right ventral lung fields; and friction rubs in the right dorsal lung field. There was an absence of lung sounds in the left ventral lung field. Additional remarkable findings included a sinus arrhythmia and an enlarged left prescapular lymph node. Pulse oximetry revealed the patient to be oxygenating at 98%. Thoracocentesis was performed and yielded approximately 2 mL of serosanguinous fluid. The thorax was assessed with ultrasound and revealed areas that appeared to be consistent with pericardial effusion. A complete blood count, serum biochemical analysis, prothrombin time, and partial thromboplastin time were performed. Serum biochemical analysis revealed elevations in alanine aminotransferase (ALT, 166 U/L; reference range, 16 to 91 U/L), aspartate aminotransferase (AST, 75 U/L; reference range, 23 to 65 U/L), and alkaline phosphatase (ALP, 368 U/L; reference range, 20 to 155 U/L). The complete blood count revealed a mild anemia characterized by a decreased red blood cell count (4.56 × 10¹²/L; reference range, 5.83 to 8.87 × 10¹²/L), decreased hemoglobin (113 g/L; reference range, 133 to 205 g/L), and a decreased hematocrit (0.34; reference range, 0.40 to 0.60).

Three-view thoracic radiographs were obtained, and the lung lobes exhibited a mild, diffuse interstitial pattern. The lung lobe margins were retracted from the body wall, exhibited pleural fissure lines, and were surrounded by a uniform soft-tissue opacity consistent with pleural effusion. Gas bubbles were present in the cranioventral pleural cavity, and there was a mediastinal shift to the left side. The ventrodorsal projection revealed that the heart was rounded from the 1–3 o'clock position, suggestive of pericardial fat, loculated effusion, or left auricular enlargement; however, lack of cardiac pathology noted on the lateral projection was more consistent with loculated pericardial effusion. The right lateral projection revealed a broad-based, soft-tissue opacity dorsal to the second sternebra. Those findings were consistent with an asymmetric, bilateral pneumohydrothorax that was worse on the left side than on the right. The changes to the cardiac silhouette were not consistent with a specific disease process; however, positional artifact, pericardial fat, or loculated pericardial effusion were considered as possible causes. There was no evidence of radiopaque thoracic foreign bodies, and the interstitial pattern and mediastinal shift were considered to be the result of atelectasis. It was concluded that further imaging was needed for localization and identification of possible remaining quills.

The dog was anesthetized with hydromorphone^c (0.18 mg/kg IV), diazepam^d (0.45 mg/kg IV), and propofol^e (4.05 mg/kg IV); maintained on isoflurane^f (1–1.4%) and O₂ (15–20 mL/kg/min); and placed in a wedge trough in sternal recumbency. Computed tomography (CT) of the thorax was performed with a 16-slice helical CT scanner^g. Pre- and post-contrast CT scans encompassed the length of the thoracic cavity and consisted of 5 mm by 5 mm transverse images reconstructed into 2.5 mm by 2.5 mm thick slices. Contrast agent^h (2.2 mL/kg) and saline was administered IV prior to the second scan. Pre-contrast and pre-thoracocentesis, pre-contrast and post-thoracocentesis, post-thoracocentesis and immediate post-contrast, and delayed post-contrast scans were performed. The precontrast and prethoracocentesis images showed a large amount of gas in the left pleural space that displaced the heart to the right. Thoracocentesis extracted roughly 3 L of air from the left hemithorax but did not produce a significant change on imaging. Smaller amounts of free air were present in the right side of the pleural cavity and in the caudal mediastinum. Those findings were consistent with a severe left-sided tension pneumothorax, a mild right-sided pneumothorax, and a mild pneumomediastinum. Pleural effusion was present in the ventral thorax, with the left being worse than the right. Evidence of pericardial effusion was noted predominantly on the left side. Multiple, hypoattenuating, curvilinear foreign bodies, up to 1.3 mm thickness and of varying lengths, were observed in the left cranial lung lobe, the left caudal lung lobe, in the body wall, and perforating the descending aorta, the left ventricle, and the pericardium. Additional suspected foreign bodies appeared to be entering the bronchus of the left cranial lung lobe, in the right ventral thorax body wall, in the ventral mediastinum, perforating the main pulmonary artery, along the esophagus at the level of the tracheal carina, and along the left ventrolateral wall of the esophagus. The ventral subcutaneous tissues cranial to the sternum exhibited non-enhancing soft-tissue attenuation through subcutaneous fat. The soft tissues of the left pectoral and right triceps muscles exhibited multifocal, oblong, hypoattenuating regions with mild regional enhancement. Those findings were considered to be consistent with the previous history and suspected porcupine quill migration. Foreign body migration was considered to be a likely cause of the pneumothorax, pleural effusion, and pericardial effusion.

Immediately following CT, due to concerns for the tension pneumothorax, an exploratory thoracotomy was performed via a median sternotomy. Enrofloxacin (15 mg/kg IV) and ampicillinⁱ (22 mg/kg IV) were administered perioperatively. Quills found embedded within the subcutaneous tissues of the thoracic body wall and the mediastinum were removed. The lung lobes were visually assessed and palpated. Quills were identified in the left cranial, left caudal, and right cranial lung lobes. Each of those lung lobes had erythematous and irregular areas on their surfaces. A partial lobectomy of the cranial aspect of the left cranial lung lobe, as well as complete lobectomies of the left caudal and right cranial lung lobes were performed with a thoraco-abdominal stapler using a 30 mm vascular cartridge^j. Following the ligation, transaction, and removal of all lobes, the pleural cavity was filled with warm sterile saline to ensure there had been adequate occlusion of each main stem bronchus. Samples of the excised lung were submitted for histopathology.

The pericardium was opened, and 2-0 silk^k stay sutures were applied to either side of the incision. Those sutures were used to suspend the heart within the pericardial tissues, forming a pericardial “basket.” The myocardium and great vessels were inspected. Quills were seen in the main pulmonary artery and left ventricle. A horizontal mattress suture using 5-0 polypropylene^l was placed through two pledgets that were placed on either side of each quill. The quills were removed as the sutures were tightened. One more quill was discovered in the fat between the aortic root and main pulmonary artery and was removed without damage to either vessel. A 14-French chest tube^m was capped with a three-way stopcockⁿ and then sutured in place with 2-0 nylon^o purse string at the site of the stoma followed by a finger trap suture pattern. The thoracic cavity was closed with 18-gauge stainless steel wires^p placed around the sternebrae in a simple interrupted pattern. The pectoral muscles were closed with 0 polydioxanone^q suture in a simple continuous pattern. A 20 mL syringe was then attached to the chest tube, and the pleural space was aspirated until negative pressure was achieved. The subcutaneous tissues were closed with a simple continuous pattern of 3-0 polydioxanone, and the skin was closed with simple interrupted sutures of 3-0 nylon.

The dog recovered in the intensive care unit and was maintained on IV fluids^r (4 mL/kg/hr) and a colloid^s (2 mL/kg/hr). Constant rate infusions of fentanyl^t (3–4 μg/kg/min) and lidocaine^u (30 μg/kg/min) were administered to alleviate post-operative pain. The dog's direct blood pressure, central venous pressure, respiratory rate and effort, temperature, packed cell volume, total solids, and serum electrolyte levels were monitored frequently. The chest tube was aspirated frequently to monitor and remove the fluid and air that accumulated within the pleural cavity.

The dog had a consistently low diastolic arterial blood pressure (≤37 mm Hg; reference range, 60–100 mm Hg) in conjunction with normal systolic arterial blood pressures (104 mm Hg; reference range, 100–160 mm Hg), resulting in bounding pulses. A persistent grade II/VI diastolic murmur was ausculted. An electrocardiogram revealed an intermittent accelerated idioventricular rhythm (heart rate, 150 beats/min) and multiple fusion beats. Two-dimensional Doppler echocardiography showed that the left ventricle had normal motion, size, and wall thickness. Both atria were mildly enlarged, and the right ventricle showed evidence of moderate dilation and paradoxical septal motion during early diastole. The latter phenomenon was characterized by a paradoxical “bounce” of the interventricular septum as it initially deviated towards, then away from, the free wall of the left ventricle during diastole. Multiple sets of hyperechoic parallel lines measuring 0.5–2 cm in length appeared to be in the right atrium, right ventricle, left atrium, and at the aortic root/aortic valve annulus.

Based on the echocardiographic images, it appeared that several quills spanned the aortic root below the level of the valve (Figure 1). Additionally, a small, round, and mildly mobile hyperechoic structure consistent with a thrombus was adhered to the posterior leaflet of the tricuspid valve. Color Doppler echocardiography revealed severe tricuspid regurgitation with an eccentrically directed regurgitant jet, severe pulmonic insufficiency, and moderate to severe aortic insufficiency. A spectral Doppler study also revealed increased tricuspid regurgitant velocity suggestive of increased right ventricular pressure secondary to either outflow obstruction or pulmonary hypertension. It was concluded that there were likely multiple intracardiac foreign bodies in all chambers of the heart causing a tricuspid valve thrombus, right heart enlargement, severe pulmonic regurgitation, moderate tricuspid regurgitation, severe aortic regurgitation, mild mitral regurgitation, and an accelerated idioventricular rhythm. The prognosis was guarded due to the severe aortic insufficiency and further complicated by the unknown extent of damage to the aortic valve/root. Surgery to remove those quills would require cardiopulmonary bypass, and given the progressive deterioration of the patient in conjunction with the poor prognosis of the procedure, the owners elected to euthanize their dog.

View Figure

• Download as Powerpoint
• Download Figure

A complete necropsy was performed. Porcupine quills were discovered within the pectoral muscle near the cranial aspect of the sternum, the thoracic inlet, the stomach, the cecocolic junction, the right caudal lung lobe, the left atrium, and the aortic root. The quill embedded within the aortic root resulted in tearing of the left coronary cusp of the aortic valve (Figure 2). The quill in the right caudal lung lobe was found in the main pulmonary artery of the lobe. Throughout the thoracic cavity there was an acute diffuse fibrinous pleuritis with moderate serosanguineous to hemorrhagic effusion.

View Figure

• Download as Powerpoint
• Download Figure

Histopathological analysis revealed a focal acute rupture of the left coronary cusp of the aortic valve with surface fibrin and subacute valvulitis. Severe diffuse fibrinous epicarditis was discovered in the aortic valve, in the right ventricular free wall, on the tricuspid valve, and on the left ventricle. The right ventricular free wall revealed multifocal myocardial loss with extensive fibrosis. A subacute to chronic thrombus was present on the tricuspid valve and there was a subacute to chronic thrombus coupled with subacute valvulitis within the aortic valve. Both the tricuspid valve and the interventricular septum revealed a severe myxomatous degeneration.

Discussion

Porcupine quill injuries are often treated promptly and carry a good prognosis with complete removal.¹ As the time from the traumatic event to the time of quill removal is prolonged, there is an increased incidence of complications associated with quill migration.¹ The delay makes it more likely for additional structures to be damaged as quill migration continues.

Modern imaging modalities have been utilized to identify the presence of quills in deeper tissues following migration.^5,8 CT remains the gold standard in veterinary medicine for imaging the thoracic cavity because rapid scan times result in decreased motion artifact and cross-sectional imaging eliminates superimposition, providing accurate imaging down to 1 mm.^9,10 Those improvements facilitate visualization of foreign objects within the thoracic cavity; however, due to the small size of porcupine quills together with motion artifact associated with the heart beat and respiration, it may be difficult to definitively diagnose the presence and exact location of quills within the pleural cavity. That task is further complicated because the quills exhibit a soft-tissue density similar to that of the inflammation associated with migratory tracts and with lung consolidation secondary to pneumothorax.

Echocardiography is currently the preferred imaging modality of the heart because it is widely available in clinical practice and because it is capable of producing accurate results.¹¹ The transthoracic echocardiogram appeared to identify multiple quills associated with the heart; however, the number and location of those quills was inconsistent with the necropsy results. It is quite possible that some quills may remain unidentified, misidentified, or that they may emerge as a result of continued migration. Although limitations associated with user skill and experience persist, echocardiography has improved the overall assessment of cardiac function and of intracardiac anatomy.¹² As it becomes more widely available, transesophageal echocardiography may facilitate perioperative evaluations of the heart for the presence of additional quills; however, that imaging modality was not available at the time of surgery.

The postmortem finding of a quill within the main pulmonary artery of the right caudal lung lobe raises two interesting questions. First, why was the quill not detected in that location during surgery? Given the small size of the structures, it is possible that although multiple quills were identified and removed, some quills could have been missed during visual evaluation and, more importantly, during palpation of the lobe. It is also possible that quill was located in either the right ventricle or main pulmonary artery at the time of surgery and only migrated into the right caudal lung lobe postoperatively.

Second, the intra-arterial location of that quill and the location of the quill removed from the main pulmonary artery at surgery raises the question of how the quills entered the vasculature. Any discussion of this question is somewhat speculative; however, at least two quills were located in the pulmonary vasculature and the description of the dog's original presentation included the removal of multiple quills removed from the sternal area. One possible explanation for those quills could be that, after quilling the cranioventral thorax, the quills may have moved through superficial tissues until they penetrated one or both of the jugular veins and subsequently migrated through the right atrium and ventricle. It is also possible that the quill remaining in the aortic root had migrated in a similar manner initially but was then forced across the interatrial septum and out through the left atrial wall across the aortic root.

This is the first report of presumed intravascular quill migration and it illustrates the potential for severe injury if prompt and complete quill removal is not performed after a porcupine encounter. Due to the small size of the quills as well as the various challenges in imaging the thoracic cavity, the sensitivity and specificity of computed tomography and echocardiography these modalities would seem to be adequate at best. Given the likelihood that quill migration will have an adverse effect on the lung lobes first, CT scanning would be the logical choice for imaging; however, even with increased precision it can be extremely difficult to discern the location and extent of quill penetration. Echocardiography was similarly misleading. While it did appropriately identify a quill within the heart, the echocardiographic report suggested the presence of additional quills. In spite of the benefits of multiple imaging modalities in this case, not all quills could be identified and removed during surgery. Consequently, it is not possible to accurately recommend an overarching diagnostic protocol for affected patients. It must be acknowledged that both modalities are useful and can provide much needed information; however, the utmost importance must be placed on careful patient evaluation and prompt removal of the quills prior to extensive quill migration.

Conclusion

This is the first report of presumed intravascular porcupine quill migration in a dog. Porcupine quill injuries constitute a serious threat that may result in significant morbidity and mortality days or even weeks after the initial trauma. Identification and immediate removal of all quills is critical in preventing progressive quill migration; however, small or fragmented quills may become embedded in deeper tissues prior to extraction. Intravascular quill migration should be considered as a possibility that may exacerbate individual pathologies and hinder localization of all quills. Numerous imaging modalities are available to search for deeper quills; however, it is not possible to recommend an overarching diagnostic protocol because these diagnostics often fail to identify all of the migrating quills.