• Open Access

Successful Treatment of Invasive Pulmonary Aspergillosis in a Neonatal Foal


Corresponding author: Dr Hugo Hilton, William R. Pritchard Veterinary Medical Teaching Hospital, School of Veterinary Medicine, University of California, One Shields Avenue, Davis, CA 95616; e-mail: hghilton@ucdavis.edu.

A 55 kg, 14-day-old, Thoroughbred colt foal was examined because of acute respiratory distress of 4 hours duration. The colt had been born at 320 days gestation and parturition was normal. Serum IgG 48 hours after birth was >800 mg/dL.a The colt was hospitalized for treatment of diarrhea 7 days after birth; acidemia (pH 7.247; reference range 7.32–7.44), hyponatremia (108 mEq/L; reference range 132–146 mEq/L), hypochloremia (89 mEq/L; reference range 99–109 mEq/L), and leukocytosis (22,500 cells/μL; reference range 5,300–14,000 cells/μL) due to neutrophilia (18,913 neutrophils/μL; reference range 3,400–11,900 neutrophils/μL) were documented. Thoracic radiography performed during hospitalization identified a thin-walled gas-filled structure within the right caudo-dorsal lung field and increased soft-tissue opacity consistent with regional pulmonary consolidation (Fig 1). The radiographic diagnosis was congenital pulmonary bulla. The foal responded to IV administration of fluids and antibiotic therapy but a CBC identified continued leukocytosis on days 2 and 4 of hospitalization (19,690 and 23,650 cells/μL, respectively). The foal was discharged from the hospital with a recommendation for administration of ceftiofur sodium (10 mg/kg SQ q12h for 14 days) and amikacin (21 mg/kg IM q24h for 7 days).

Figure 1.

 Lateral thoracic radiograph demonstrating the cavitary lesion identified in the right caudo-dorsal lung field (arrows).

The foal presented for examination 24 hours after discharge with signs of respiratory distress including marked abdominal effort and nostril flare. Heart rate was 120 bpm, respiratory rate 60 bpm, and rectal temperature 100.1°F. Mucous membranes were pale and cyanotic. Thoracic auscultation identified reduced bronchovesicular sounds bilaterally. Thoracic ultrasonography and radiography identified free gas and pulmonary collapse within the right and left pleural cavities consistent with bilateral pneumothorax. The previously identified pulmonary bulla in the right caudo-dorsal lung field appeared attached to the dorsal parietal pleura (Fig 2). Bilateral thoracocentesis and aspiration of pleural gas relieved respiratory distress. Thoracic radiographs performed after thoracocentesis revealed bilateral inflation of the lungs with minimal residual free air within the thoracic cavity.

Figure 2.

 Lateral thoracic radiograph demonstrating bilateral pneumothorax and pulmonary collapse. The previously identified cavitary lesion appears attached to the dorsal parietal pleura in the right caudo-dorsal thorax (arrows).

A CBC identified anemia (28.3%; reference range 30–46%), leukocytosis (34,640 cells/μL) because of neutrophilia (32,562 cells/μL) and hyperfibrinogenemia (600 mg/dL; reference range 100–400 mg/dL). Serum biochemistry identified hyperglycemia (304 mg/dL; reference range 122–205 mg/dL), venous blood gas analysis indicated acidemia (pH 7.219) with both respiratory and metabolic components (PvCO2 70.5 mmHg; reference range 40–50 mmHg, lactate 2.1 mmol/L; reference range 0–2 mmol/L). A diagnosis of bilateral pneumothorax secondary to rupture of the pulmonary bulla was suspected based on the clinical presentation, diagnostic imaging findings, and previous identification of the pulmonary bulla. Within 1 hour of the initial stabilization, the foal was again noted to be in respiratory distress. A 2nd thoracocentesis was performed and an indwelling thoracic catheter was placed. Conservative management was considered unlikely to be successful given the rapid reaccumulation of air. Surgical management was recommended to assess the viability of the remaining lung and to remove or repair the pulmonary defect. The foal received ampicillin (21 mg/kg IV q8h), amikacin (25 mg/kg IV q24h), and flunixin meglumine (1.1 mg/kg IV q 12h).

The colt was administered diazepam (0.05 mg/kg IV) and anesthesia was induced with ketamine hydrochloride (2 mg/kg IV). After oro-tracheal intubation, an esophageal dilation balloon, with a maximal inflated diameter of 25 mm, was passed into the right mainstem bronchus under endoscopic guidance. The balloon was inflated to occlude flow of anesthetic gases to the right lung.

The foal was placed in left lateral recumbency and the right thorax was aseptically prepared and draped for exploratory thoracoscopy. After skin desensitization with 3 mL, 2% lidocaine, a 12 mm skin incision was made at the 10th intercostal space, approximately 3–5 cm ventral to the dorsal costal arch. A Veress needle was introduced through the intercostal muscles and into the pleural cavity to induce a pneumothorax. Once collapse of the right lung was confirmed by free manipulation of the Veress needle, a grid approach was continued through the intercostal muscle layers extending through the pleura into the pleural cavity. A 30°, 50-cm laparoscope was introduced through the pleura and the right hemithorax was explored. A laparoscopic cannula was not used due to limitations of space between the 10th and 11th ribs. An open cavity was visualized at the caudo-dorsal and lateral aspect of the lung lobe, which was adhered to the dorso-lateral pleura from the 12th to the 14th ribs. A separate laparoscopic portal was localized with a 3.5-in., 18-G spinal needle and a 2nd 12 mm incision was made after infiltration of 2 mL, 2% lidocaine at the level of the 14th intercostal space, 2.5 cm ventral to the border of the epaxial muscles. The laparoscope was then introduced in the 2nd portal to allow a more complete visualization of the cavity. The cavity was deemed too large and too distant from the caudal and lateral lung border to attempt laparoscopic Endo-GIAb resection.

The esophageal balloon in the right mainstem bronchus was deflated and removed. A 10-cm thoracotomy was performed at the level of the 14th intercostal space by extending the previously made incision ventrally. After application of a Finochietto retractor, the adhesion was manually broken and the caudal aspect of the right lung lobe was exteriorized. To stabilize the lesion for resection, stay sutures (# 2 Vicryl) were placed on healthy parenchyma cranially, dorsal and ventral to the lesion. The lesion was resected in a V-shaped pattern with an Endo-GIA 3.5-mm staplerb (1 cartridge fired dorsally and 1 ventrally) and a TA-90,c 3.5-mm stapler (2 cartridges fired over each other at the cranial aspect of the “V”). Gaping of the tissue and associated bubbling of air was seen at the cranial aspect of the wedge-shaped resection (where the TA-90 stapler was used). This area was oversewn with # 0 Polydioxanone (PDS II) suture material in a continuous Lembert pattern.

The right lung lobe was reinflated and no air leakage was identified. A 10-Fr, fenestrated silastic Blake draind was placed by inserting the attached sharp trocar from the thoracotomy incision to exit the ventral aspect of the 10th intercostal space. Two milliliters of 0.5% bupivicainee was infused around the intercostal nerves immediately cranial and caudal to the thoracotomy incision to provide additional analgesia before anesthetic recovery. After routine closure, a 60-mL syringe inserted in a 3-way stopcock attached to the chest drain was used to reestablish negative intrathoracic pressure.

In the immediate postoperative period the foal was administered fluids (Plasmalyte 148, 4 mL/kg/h and dextrose, 4 mg/kg/min IV), morphine sulfate (0.1 mg/kg IV q6h), and intranasal oxygen (10 L/min). One hour after extubation the foal was ambulatory and making attempts to nurse.

The resected tissue measured 66 × 68 × 34 mm with a central excavation measuring 21 × 21 × 34 mm (Fig 3). Histologically up to 80% of the pulmonary parenchyma was effaced by an area of central necrosis surrounded by granulation tissue and neutrophilic and histiocytic inflammation. In several microscopic fields, abundant mats of dichotomously branching, parallel walled, and septate fungal hyphae were identified. These were comingled with conidia and brown-gold-pigmented spores at the interface between viable and necrotic tissue (Fig 4). These findings suggested that the foal had a severe fungal pneumonia with hyphal morphology consistent with Aspergillus spp. The foal was administered voriconazole (10 mg/kg PO q24h) pending fungal culture and sensitivity results.

Figure 3.

 Formalin-fixed gross section of the resected tissue showing cavitation and necrosis of the lung surrounded by granulation tissue and inflammation. Bar = 1 cm.

Figure 4.

 Photomicrograph showing mats of dichotomously branching, parallel walled, and septate fungal hyphae characteristic of Aspergillus spp. *comingled with conidia (arrows) and brown-gold-pigmented spores (diamonds) in necrotic tissue. Stain: Hematoxylin & Eosin.

Indwelling thoracic drains in the right and left thoracic cavities were removed 2 and 3 days after surgery, respectively. Postoperative radiographs and ultrasonographic examination of the chest did not identify accumulation of either fluid or free air within the thorax. CBC performed 2 days after surgery identified anemia (26%) and hyperfibrinogenemia (600 mg/dL); all other hematologic values were within normal limits. The foal was discharged 4 days after surgery. He continued to receive antimicrobials (trimethoprim sulfamethoxazole 30 mg/kg PO q12h for 7 days) and antifungal therapy (voriconazole 10 mg/kg PO q24h for 24 days).

Fungal culture confirmed the presence of Aspergillus fumigatus that was sensitive to voriconazole at a minimum inhibitory concentration (MIC) of 0.25 mcg/mL at 24 hours and 0.5 mcg/mL at 48 hours. The Aspergillus spp. identified was also susceptible to itraconazole (MIC 0.5 mcg/mL at 24 hours and 48 hours) and posaconazole (MIC 0.06 mcg/mL at 24 hours and 0.25 mcg/mL at 48 hours) but resistant to fluconazole (MIC > 64 mcg/mL at 24 hours).

The foal remained healthy during treatment with voriconazole. At examination 28 days after discharge from the hospital, the foal was in good body condition, bright, and responsive. Rectal temperature, heart rate, and respiratory rates were within normal limits. No abnormalities were detected on CBC or serum biochemistry. Thoracic radiographs identified a mild focal interstitial opacity in the caudodorsal lung associated with the previous surgical site. There was no evidence of pneumothorax or diffuse pneumonia. The foal is reportedly well 8 months after surgery.

Pulmonary aspergillosis is an infrequent cause of pneumonia in the horse and is often associated with acute enterocolitis.1–3 Pulmonary disease is believed to develop as a result of profound neutropenia and translocation of fungal organisms across a disrupted intestinal mucosa. In animals without loss of gastrointestinal integrity, immunosuppression, prolonged administration of antibiotics or corticosteroids and overwhelming exposure to fungal spores may also result in disease.3–6 The case fatility rate of horses affected with pulmonary aspergillosis is high and likely reflects the difficulty in antemortem diagnosis and the poor response to therapy in debilitated or immunocompromised animals.3

Reports of pulmonary aspergillosis are rare in neonatal foals. Those reported have been in weanling foals and were found in association with enteric salmonellosis.7 Pulmonary aspergillosis is a major cause of illness and death in neonatal human patients with hematologic malignancies, transplant recipients and those receiving immunosuppressive therapy.8 Although respiratory disease resulting from opportunistic pathogens such as adenovirus and Pneumocystis carinii is a common finding in foals with immunodeficiency, pulmonary aspergillosis has not been reported.9 The source of infection in this case is unclear. There was no indication of immunosuppression in this foal and although the foal did have signs consistent with acute enteritis, the cavitary lung lesion was found on presentation to our hospital only 3 days after the onset of the foal's diarrhea, presumably too soon for the lesion to have developed as a result of enterocolitis. Other possible routes of infection included in utero inoculation or postpartum inhalation.

Cavitary lung disease resulting from pulmonary aspergillosis is a rare finding and has not been reported previously in the foal. In recent reports from the human literature 20% of adult patients with invasive aspergillosis presented with radiographic evidence of cavitary lung lesions whereas in a review of pediatric patients <5 years of age, a lower incidence of cavitiation was found.10,11 The development of cavitary lesions has been associated with a rapid rise in neutrophil count following a neutropenic episode.12,13 This pathogenesis is consistent with the clinical findings in this case. Although neutropenia was not documented it could be postulated that the foal was neutropenic at the onset of acute enteritis with subsequent neutrophilia confirmed. Differential diagnoses for pulmonary cavitary lesions include bullous disease and coccidiomycosis.11 This case demonstrates that an infectious cause of cavitation should be suspected in animals with regional consolidation of the surrounding pulmonary parenchyma and compatible clinical and laboratory findings.

Early lung resection in combination with aggressive antifungal medication has been shown to be safe, effective, and diagnostic in humans suspected of having pulmonary aspergillosis.14–17 The success of the procedure in this case is likely because of the localized nature of the lesion and the lack of systemic dissemination. In addition, the fungal culture obtained from the resected tissue served to ensure accurate antifungal therapy.

Effective antifungal treatment for affected horses is restricted to a few drugs because of unfavorable pharmokinetics, limited absorption, or potential toxicity.3 Amphotericin B deoxycholate has been used widely in both human and equine patients for the treatment of pulmonary aspergillosis, however, a recent study in humans concluded that initial therapy with voriconazole led to better clinical responses, improved survival, and resulted in fewer adverse effects.18 Systemic absorption of voriconazole after IV and PO administration to horses is good.19,20 The dosage used in this foal (10 mg/kg PO q24h) was based on available safety and pharmokinetic data in adult horses and human patients.19,20 Although a lower dosage (3–5 mg/kg PO q12h or q24h) might have achieved appropriate plasma concentrations, it was felt that, despite the considerable cost of the drug (approximately US $2,500 for 24 days), given the severity of the infection, the unknown MIC values at the initiation of treatment and the potential higher volume of distribution in the neonate that this regimen was appropriate. Fluconazole has also been used as an antifungal treatment in horses although this drug is not recommended for the treatment of filamentous fungi such as Aspergillus or Fusarium spp. and as this and other reports suggest, resistance to this drug may be widepsread.21,22 Drug-associated adverse events reported in humans treated with voriconazole include visual abnormalities, skin reactions, and increases in serum activity of liver derived enzymes.18 No adverse effects were observed in this foal and there were no abnormalities identified on physical or hematologic examination 28 days after discharge.


aSNAP Foal IgG Test Kit. IDEXX Laboratories Inc, Westbrook, ME

bEndo GIA 3.5-mm stapler, United States Surgical Corporation, Norwalk, CT

cTA-90 3.5-mm stapler, United States Surgical Corporation

d10-Fr Blake Silastic Drain, Ethicon Inc, Somerville, NJ

eHospira Inc, Lake Forest, IL


The authors thank Drs Peter Pascoe, Robert Pottie, and Tracy Peterson for their help with this case.