Pneumatosis coli in a domestic ferret (Mustela putorius furo)

Abstract

A 4-year-old spayed female ferret was presented with acute diarrhea and partial anorexia. Pneumatosis coli and segmental enteropathy were identified by ultrasonography and radiography. Fecal culture did not identify any pathogenic bacteria. Medical management of concurrent diseases and antibiotic therapy resulted in resolution of clinical signs and pneumatosis coli.

Ferrets are frequently affected by gastrointestinal diseases of various origins (1). Pneumatosis coli (PC) is an uncommon medical condition characterized by the presence of gas within the intestinal wall, usually in the submucosal or subserosal layer. The condition is most commonly known as pneumatosis cystoides intestinalis or pneumatosis intestinalis when the intestines are affected. It is also reported as intestinal emphysema, cystic lymphopneumatosis, and abdominal or intestinal gas cysts (2). The term gastric pneumatosis is used when the stomach is affected. These conditions are considered physical or radiographic findings and reflect the presence of an underlying condition, including trauma as well as gastrointestinal, systemic, pulmonary, autoimmune, and iatrogenic diseases (3,4).

Case description

A 4-year-old, spayed female domestic ferret was presented to the Zoological Medicine Service of the Centre Hospitalier Universitaire Vétérinaire, Université de Montréal, with a 3-day history of diarrhea and partial anorexia. The animal had been adopted 5 mo before presentation. Previous medical history was unknown. The owner reported recent loss of fur on the tail and generalized pruritus.

On physical examination, the ferret was thin and mildly lethargic. Clinical findings included a 3-mm rectal prolapse, mucoid diarrhea, generalized pruritus, and hypotrichosis of the tail. The rectal mucosa was erythematous and moderately swollen. On abdominal palpation, multiple small firm mobile masses were identified in the mid-abdomen. The right adrenal gland was suspected to be enlarged. Blood glucose measured with a portable glucometer (AlphaTrak, Abbott Laboratories, Illinois, USA; using a canine setting) after a 3-hour fasting period was 3.4 mmol/L [reference range (RR): 4.4 to 6.5 mmol/L] (5).

Abdominal ultrasonography showed moderate to severe mesenteric lymphadenomegaly (up to 15 mm in thickness) with hypoechoic parenchyma and hyperechoic fat surrounding the lymph nodes. Segmental asymmetrical wall thickening of the jejunum was also observed. There was gas dissecting the colonic wall with parallel disposition compared to the wall layering, consistent with pneumatosis coli (Figure 1). The right adrenal gland was moderately enlarged (thickness: 6.5 mm; RR: 1.3 to 3.7 mm) (6) with heterogeneous parenchyma and hyperechoic speckles casting acoustic shadowing within it. Abdominal radiographs revealed linear radiolucency within the colonic wall, confirming the presence of pneumatosis coli (Figure 2). This section of the colon was severely dilated and mainly filled with gas and a small amount of fecal material.

Figure 1.

Abdominal ultrasonography of the ferret.

Longitudinal (A) and transverse (B) ultrasonographic evaluations of the colon revealed hyperechoic lines within the colonic wall (arrows) consistent with pneumatosis coli.

Figure 2.

Abdominal radiography of the ferret. These images were acquired in right lateral recumbency on the day of presentation (A) and on first re-evaluation 5 wk later (B). Arrows show linear radiolucency within the colonic wall consistent with pneumatosis coli (A). Similar findings were absent on re-evaluation (B).

Differential diagnoses for segmental enteropathy included lymphoma, bacterial enteritis, inflammatory bowel disease, eosinophilic enteritis, proliferative bowel disease, coccidiosis, and epizootic catarrhal enteritis with concomitant pneumatosis coli. Presumptive diagnosis of adrenal gland disease was also established based on hypotrichosis and ultrasonographic findings. Despite the absence of pancreatic nodules visualized on ultrasonographic examination, an insulinoma was suspected based on the cut-off value of 3.3 mmol/L and the use of the AlphaTrack with a canine setting, which can overestimate blood glucose in hypoglycemic ferrets (7,8). A blood glucose concentration value remained to be measured using a validated laboratory analyzer to confirm hypoglycemia (8).

A therapeutic plan was established based on the presumptive diagnoses and was to be modified if needed in light of forthcoming bacterial culture results. The plan included leuprolide acetate (Lupron Depot; Abbvie, Saint-Laurent, Quebec), 100 μg/kg body weight (BW), IM, q1mo, prednisolone (Pediapred; Sanofi-Aventis, Laval, Quebec), 1 mg/kg BW, PO, q12h until re-evaluation, cephalexin (Novo-Lexin; Teva Canada, Toronto, Ontario), 25 mg/kg BW, PO, q12h for 14 d, and sucralfate (Sulcrate; Aptalis Pharma Canada, Mont-St-Hilaire, Quebec), 100 mg/kg BW, PO, q12h for 30 d.

Aerobic and anaerobic bacterial cultures of the stools were submitted during the initial examination. Sensitivity testing on aerobic bacteria was performed. Enterococcus spp., Enterobacteriaceae, and Clostridium perfringens were identified. Fecal flotation was planned but could not be performed following a laboratory error.

Upon re-evaluation 5 wk later, diarrhea had resolved. The owner reported that the animal was more active, fur was growing back, and pruritus had resolved following the first visit. However, a decrease in general activity and a relapse of the generalized pruritus had been noted 4 d prior to the recheck examination. Treatment recommendations had been followed except for prednisolone that was discontinued by the owner after 3 wk of administration and for leuprolide acetate injection that should have been administered a week earlier. Abdominal ultrasound showed a less pronounced abdominal lymphadenomegaly and resolution of the segmental jejunal enteropathy and pneumatosis coli (Figure 2). The size of the right adrenal gland was decreased compared to the first evaluation, but the gland was still enlarged. Abdominal radiograph confirmed resolution of the pneumatosis coli. On repeated aerobic and anaerobic bacterial cultures, Enterococcus spp., Enterobacteriaceae, C. perfringens, and Bacillus spp. were identified. No parasites were observed on fecal flotation.

Considering the resolution of diarrhea, rectal prolapse, and segmental enteropathy, no antibiotic treatments were initiated despite the bacterial culture results. Further treatment with prednisolone was declined by the owner. Leuprolide acetate injections were continued as prescribed on initial presentation.

The ferret was presented for re-evaluation 10 wk after initial presentation. Hair was growing back, pruritus had resolved, and general activity was back to normal. No gastrointestinal clinical signs were reported. The results of an ultrasonographic examination were similar to those of the first re-evaluation. Aerobic and anaerobic bacterial cultures revealed Lactobacillus spp., C. perfringens, and an unidentified coliform bacterium.

Discussion

Pneumatosis cystoides intestinalis is uncommon in veterinary medicine. It is occasionally seen in swine (9). It has been previously described in a gnotobiotic pigs inoculated with an enteropathogenic Escherichia coli (9), in a single rabbit doe (Oryctolagus cuniculus) (2), and in horse neonates (10). There have been several reports of pneumatosis coli in dogs (11–14). Gastric pneumatosis has been described in a red ruffed lemur (Varecia variegata rubra) with non-specific gastrointestinal signs (15), in 4 black and white ruffed lemur (Varecia variegata variegata) associated with eosinophilic gastritis (16), in a Bengal slow loris (Nycticebus coucang bengalensis) with a hepatocellular carcinoma and an esophageal leiomyosarcoma (17), in dogs with gastric dilatation-volvulus syndrome (18), and in a domestic cat following gastrointestinal surgery (19). To the authors’ knowledge, this is the first report of pneumatosis coli in a domestic ferret.

Three mechanisms have been described concerning the origin of pneumatosis cystoides intestinalis in humans: gas translocation from the lumen to the intestinal wall, intrusion of gas-producing bacteria within the intestinal wall, and intrusion of gas from lung alveolar rupture, tracking through the mediastinum to the retroperitoneum and mesentery (3,4). Alteration of the intestinal mucosal or immune barrier integrity eases the transfer of gas or bacteria into the mural compartment, potentially leading to pneumatosis cystoides intestinalis. Increased intraluminal pressure can also lead to the passage of gas through the mucosal barrier, even in the absence of mucosal or immune compromise, though a combination of both situations is common. A theory called “counterperfusion supersaturation” has been proposed, hypothesizing that the increased intraluminal hydrogen tension in patients affected by some bacteria exceeds the blood nitrogen tension, leading to the diffusion of hydrogen toward the submucosal vessels (20). It has also been proposed that fluctuations in intra-abdominal pressure in patients affected by pulmonary diseases could lead to the formation of pneumatosis cystoides intestinalis instead of alveolar rupture and gas movement through the mediastinum (21).

A wide variety of underlying conditions have been reported in humans affected by pneumatosis cystoides intestinalis. Although many mechanical or traumatic causes have been described, multiple inflammatory, autoimmune, infectious, pulmonary, iatrogenic, and oncologic conditions have been associated with pneumatosis cystoides intestinalis (3,4). Approximately 15% of human cases are idiopathic, while 85% are considered secondary (3). Humans with this condition are most commonly asymptomatic, but clinical signs can include abdominal pain, diarrhea, nausea/vomiting, hematochezia, melena, constipation, tenesmus, and excessive mucus or gas production (16,22). Complications directly associated with pneumatosis cystoides intestinalis occur in 16.3% of cases according to a study in human medicine (22). They are usually the result of gastrointestinal obstruction by intestinal wall dilatation (22).

The exact cause of pneumatosis coli was not determined in this case, but multiple etiologies are possible. Ulcerative colitis, as well as Clostridium spp. infections, have been reported as possible causes for pneumatosis cystoides intestinalis in humans (3,4). In this patient, Enterococcus spp., C. perfringens, and Enterobacteriaceae were identified on initial presentation. These bacteria were still present on the first re-evaluation despite resolution of gastrointestinal signs and pneumatosis coli. These findings were expected considering the weak efficacy of cephalexin against most of these bacteria and the time lapse between the end of the treatment and the re-evaluation. C. perfringens, as well as several Enterobacteriaceae, have been previously described as part of the fecal microbiota of healthy ferrets (23). Their role in this case is probably minimal considering their continued presence after the resolution of clinical signs. However, no distinction was made among the different bacteria of the family Enterobacteriaceae and some of them could have been susceptible to cephalexin. Determination of toxin status via enzyme-linked immunosorbent assay (ELISA) would have been required to link the presence of C. perfringens with the diarrhea (24). Infection by another bacterium, not identified on routine aerobic and anaerobic bacteriology, remains possible. Bacterial enteritis was unlikely to be the primary cause of pneumatosis coli in this patient. However, the use of antibiotics may have contributed to reduction of bacterial overgrowth, leading to resolution of pneumatosis coli (25). Cephalexin, though uncommonly prescribed in cases of diarrhea in ferrets, was used while we awaited the results of bacterial culture. The ferret’s poor general condition and several client-associated constraints led to the choice of this antibiotic. The treatments were not modified when bacterial culture results were available due to the improvement in the animal’s condition and refusal of the owner to try another medication. As inflammatory bowel disease and eosinophilic enteritis were part of the differential diagnoses, prednisolone could have played a role in the resolution of pneumatosis coli as well.

Insulinoma was highly suspected in this ferret because of the low blood glucose (3.4 mmol/L) following a 3-hour fast. The AlphaTrak with a canine setting has been shown to be the portable glucometer with the best agreement to a validated laboratory analyzer for the evaluation of ferrets’ blood glucose, although results can be overestimated (9,26). Hypoglycemia is usually defined as a blood glucose value of < 3.3 mmol/L, although different reference ranges have been described in ferrets (5). Thus, hypoglycemia was likely but not confirmed and evaluation by a validated laboratory analyzer was recommended, but declined by the owner.

In this case, pneumatosis coli was diagnosed using both ultrasonography and radiography. The condition is characterized by linear, curvilinear or cystic-shaped radiolucency within the colonic wall on radiography (27). Barium enema can occasionally be helpful for the diagnosis of pneumatosis coli, though the condition can be confused with intestinal polyposis (4). The diagnosis can also be made using ultrasonography by identifying bright echoes casting acoustic shadowing within the colonic wall (28). Computed tomography is now considered the most sensitive diagnostic technique in the identification of pneumatosis coli, allowing the distinction of gas in the bowel wall from intraluminal gas (3). Histopathological examination of gas cysts usually reveals granulomatous inflammation characterized by eosinophils, lymphocytes, multinucleated giant cells, macrophages, and plasma cells (4,16). Endoscopy has been used in human medicine to confirm the presence of pneumatosis cystoides intestinalis, but the distinction between gas cysts and other structures is often impossible without aspiration (4).

Treatment of pneumatosis coli relies mainly on resolution of the primary condition responsible for the accumulation of submucosal gas. It has been suggested that oxygen therapy by facemask, head tent, hyperbaric means, or mechanical ventilation to maintain an arterial oxygen partial pressure higher than 300 mmHg was effective in treating pneumatosis cystoides intestinalis (3). It is uncertain whether this therapy acts by direct effect on anaerobic bacteria or by diffusion of oxygen into the bowel wall, increasing the gas partial pressure and thus forcing the accumulated gas into the intestinal lumen. Radiographic criteria supportive of a surgical intervention in human include the presence of pneumoperitoneum or portal venous gas (29). Other criteria include blood pH < 7.3, bicarbonate level < 20 mEq/L, lactate level > 2 mmol/L, amylase level > 200 U/L, and disseminated intravascular coagulation (3). Complete blood cell count and biochemistry were not performed in this ferret because of budget limitations and the absence of criteria supportive of a surgical intervention.

In conclusion, radiography and ultrasonography were useful in establishing a diagnosis when pneumatosis cystoides intestinalis was suspected in a ferret. The exact cause of pneumatosis coli was not identified in this ferret, but a multifactorial enteropathy is deemed likely.