Pancreatic Abscess in a Cat with Diabetes Mellitus

Introduction

In the veterinary literature, pancreatic abscesses have been defined as circumscribed collections of purulent materials or purulent exudates within the pancreatic parenchyma.^1,2 Those lesions are considered to be a complication of severe necrotizing pancreatitis or chronic-active pancreatitis in dogs.^1–5 Abscess formation is generally preceded by necrosis, and the frequently observed clinical signs in animals with severe pancreatitis may include anorexia, vomiting, and lethargy.^2–6 Clinicopathological findings of fluid in the pancreas are very important for differentiating abscesses from pancreatic pseudocyst, cystic neoplasms, or pancreatic phlegmon, but differentiation using diagnostic imaging techniques alone, such as abdominal ultrasound examination, is usually not possible.^7–9 In humans, pancreatic abscesses are generally due to a secondary infection of necrotic pancreatic tissue, whereas the isolation of microbial agents from the pancreas is uncommon in dogs.^{1–3,5,10,11} In previous reports, dogs with pancreatic abscesses showed high mortality rates, which ranged from 50 to 86%, postoperatively.^1–5 In cats, however, there is a lack of information available on pancreatic abscess because of the limited number of reported cases.¹²

Diabetes mellitus is considered to be a complication of pancreatitis in cats.^13,14 To the best of the authors' knowledge, there are no previous reports of pancreatic abscesses in diabetic cats. This report describes a case of pancreatic bacterial abscess concurrent with diabetes mellitus in a Maine coon cat.

Case Report

An 11 yr old spayed female Maine coon cat weighing 6.48 kg was referred because of uncontrolled hyperglycemia. The cat had been diagnosed with diabetes mellitus by the referring veterinarian 2 mo previously. Neutral protamine Hagedorn (NPH) insulin^a had been administered by the owner q 12 hr, but the cat continued to suffer from intermittent vomiting and weight loss. At the time of referral, the cat was being administered 10 units of insulin q 12 hr, suggesting she was either insulin resistant or had a poor response to NPH insulin. On presentation, the cat had a poor body condition (body condition score was 2 out of 9) and abdominal palpation detected a firm mass in the middle of the abdominal cavity. Serum biochemical analysis detected high levels of fructosamine^b (0.593 mmol/L; reference range, 0.191–0.340 mmol/L) and feline pancreatic lipase immunoreactivity (5.4 μg/L; reference range, ≤3.5 μg/L). Other biochemical analyses, electrolytes, and complete blood cell count were within normal limits except for fasting hyperglycemia (28.47 mmol/L; reference range, 4.16–11.04 mmol/L). Urinalysis revealed a urine specific gravity of 1.031 with glycosuria. A bacterial culture of the urine was negative. The urine cortisol/creatinine ratio was normal. Endocrine analyses did not reveal abnormal levels of total thyroxine, free thyroxine, or insulin-like growth factor-1, suggesting that those were not the cause of the uncontrolled diabetes mellitus.

Abdominal radiographs revealed a bean-shaped mass in the midabdomen, and abdominal ultrasound examination showed the presence of a large cavity filled with homogeneous echogenic fluid within the parenchyma of the pancreas, which measured a maximum of 4.6 cm in diameter (Figure 1A). Percutaneous ultrasound-guided fine-needle aspiration was performed using a 21-gauge needle with an extension line to the pancreatic cavity. Approximately 30 mL of thick, yellowish, purulent fluid was drained (Figure 2A). Cytology of the fluid detected a predominance of degenerative neutrophils, which was indicative of suppurative inflammation (Figure 2B). Bacteria engulfed by phagocytes were not observed definitively in the cytological evaluation, but bacterial culture of the fluid was positive and growth of Escherichia coli was demonstrated.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

Based on those findings, the uncontrolled diabetes mellitus was considered to be related directly to the pancreatic bacterial abscess, which could have been responsible for insulin resistance or the loss of endocrine pancreatic function. However, the authors could not exclude that the uncontrolled hyperglycemia was caused by a poor response to NPH insulin or an improper diet.

Due to the risks posed by performing surgery for a pancreatic abscess in a cat with uncontrolled diabetes, medical management was chosen as the first line of therapy, and treatment was restricted to antibiotic administration and percutaneous ultrasound-guided drainage. Before the positive bacterial culture test, amoxicillin trihydrate/clavulanate potassium^c [30 mg/kg per os (PO) q 12 hr] was prescribed with ranitidine^d (2 mg/kg PO q 12 hr), while subcutaneous injection of glargine insulin^e (1.5 U q 12 hr) was initiated to control hyperglycemia instead of NPH insulin. At the time of discharge, the owner was informed of the importance of maintaining good glycemic control by feeding the cat with a low carbohydrate, high protein diet. Two weeks later (day 14), the cat returned for re-evaluation of the pancreatic abscess and diabetes mellitus. Intermittent vomiting was still observed by the owner. Despite being drained 2 wk before, the abscess had increased in size to 3.27 cm diameter, which was considerably larger than the size of the abscess (1 cm) after the first drainage. A second ultrasound-guided drainage of the pancreas was performed, and 20 mL of yellowish fluid was aspirated. No other evidence related to the deterioration or progression of the pancreatic abscess was observed. The antibiotic was changed to cefaclor^f (20 mg/kg PO q 8 hr), which was considered to be more effective than treatment with amoxicillin trihydrate/clavulanate potassium based on the results of an antimicrobial susceptibility test. Depending on the outcome of the 12 hour blood glucose curves made by owner at home, insulin administration was continued by administrating 1.5 U of glargine q 12 hr. The nadir glucose concentration ranged from 6.55 to 12.77 mmol/L for 1 wk, and the mean blood glucose concentration ranged from 9.0 to 14.41 mmol/L.

On day 60, the size of the abscess was clearly reduced within the pancreatic parenchyma and additional drainage was not required. Based on the blood glucose curves recorded by the owner, the insulin dose was gradually decreased to 0.5 U. Five months after diagnosis, the cat had gained over 15% body weight (body condition score was 4 out of 9), and the diabetes mellitus had entered remission. The large cavity had disappeared within the pancreas and there was no recurrence, although the ultrasound findings still showed heterogeneous pancreatic parenchyma with several small cystic lesions (Figure 1B). At that time, the levels of fructosamine and feline pancreatic lipase immunoreactivity had declined to within the normal range (0.294 mmol/L and 3.3 μg/L, respectively). Vomiting was not observed any more. Antibiotic therapy was prescribed for 4 mo. Six months after the cessation of treatment with insulin and antibiotics, the cat remained healthy without any sign of exocrine pancreatic insufficiency or cobalamin deficiency.

Discussion

In this case, a diabetic cat presented with a pancreatic abscess, and investigations were made to determine the cause of its uncontrolled hyperglycemia. At the time of referral, the cat was being administered 10 U of NPH insulin q 12 hr. That suggested insulin resistance, although it was impossible to rule out uncontrolled hyperglycemia caused by a poor response to NPH insulin.

Pancreatic abscesses generally form if an episode of pancreatitis is sufficiently severe to cause parenchymal necrosis.¹⁵ Pancreatic abscesses are considered to be a late complication of pancreatitis in humans and dogs.^1–5,10,11 However, pancreatic abscesses have rarely been reported in cats in the veterinary literature, particularly those related to the incidence of feline pancreatitis.¹⁴ In dogs, a total of 73 cases of pancreatic abscesses in five separate references have been reported, whereas only 2 cases were reported among 8 cats with severe acute pancreatitis in a single reference.^1–5,12 In the current case, purulent fluid obtained from the pancreatic abscess was positive for bacteria. In contrast, canine pancreatic abscesses are usually described as sterile, while microbial agents are isolated from most human pancreatic abscesses.^{1–3,5,11,16}

The anatomic characteristics of the feline pancreas in the gastrointestinal tract differ from those of dogs. The origin of bacterial infection in the pancreas could not be explained in the patient described in this report, but the distinguishing size or morphology of the pancreatic ducts in cats might affect bacterial invasion or the regurgitation of bile from the gastrointestinal tract.^17–20 Feline acute necrotizing pancreatitis may be sterile initially, but necrosis and inflammation of the pancreas promote colonic bacterial translocation and colonization of the pancreas with a related ascending infection.^20–22 Therefore, the development of pancreatic abscesses with accompanying bacterial infections are possible in cats. However, further study will be necessary to investigate the incidence and significance of feline pancreatic abscesses with infections.

In human medicine, open abdominal surgical interventions following medical stabilization have commonly been performed to treat pancreatic abscesses and to confirm the diagnosis.¹⁰ However, there has been a reduced emphasis on early surgical approaches because improved noninvasive imaging, therapeutic techniques, and intensive care have become more readily available.¹¹ Recently, it was suggested that initial treatments for pancreatic abscesses should consist of percutaneous, endoscopic, and surgical drainage together with aggressive antimicrobial therapy.¹¹ Cats and dogs with pancreatic abscesses may also respond favorably to surgical drainage, but the risks and expense of surgery may outweigh the benefits.¹ One study (the largest in this field) of 36 dogs that underwent a surgical procedure reported that the prognosis was fair to grave: 14 out of 36 dogs were euthanized and 8 out of 36 dogs died postoperatively.² Recently, it was reported that one of two cats with pancreatic abscesses caused by severe acute pancreatitis died after surgical intervention, while the other cat exhibited insulin-dependent diabetes mellitus postoperatively.¹² In one case report, a cat with multiple, recurrent pancreatic cysts exhibited a rapid clinical course resulting in diabetes mellitus 1 mo after surgery, and it was euthanized because of nonketotic hyperosmolar diabetes.²³ Those results are likely to be related to the severity of the pancreatic lesions, although there are no reports of correlations among the severity score, surgical intervention, and prognosis.

In the current case, the cat was already suffering from uncontrolled diabetes mellitus and there were no other severe clinical signs; therefore, the benefit of a surgical intervention to remove the abscess was questionable. As a result, prompt percutaneous drainage was selected as the best procedure to manage the abscess. To prevent the leakage of pus from the cavitary lesion and hemorrhage caused by laceration of the vessels, a small-gauge needle and maximum drainage were employed under ultrasound-guidance with a color-power Doppler. No complications were observed, although not all of the pus could be drained using the 21-gauge needle attached to an extension line. Percutaneous drainage of an abdominal abscess with a fine needle might be simpler, less invasive, and less expensive; however, if the pus is very viscous, a larger needle or an indwelling catheter might be more appropriate.^24,25 It is also helpful to confirm abscessation by performing a cytological examination or an aerobic/anaerobic culture. Until recently, that approach has been shown to be a safe and effective alternative to operative drainage of abscesses or cysts in several organs, such as liver, pancreas, and prostate in veterinary medicine.^24–27 However, if proper skills are not applied, potential complications can arise, including contamination of the abdominal cavity with fluid from the abscess, hemorrhage, and secondary infection. Therefore, it is important to have a clear perception of risks involved in performing percutaneous drainage of the abdominal abscess with a fine needle. Fortunately, a combination of antibiotics and percutaneous ultrasound-guided drainage of the pancreatic bacterial abscess was successful for the management of the patient described herein.

It was unclear what role treating the pancreatic abscess had on the cat's recovery because at the time of presentation the cat did not exhibit signs of a sepsis. Additionally, it was possible that diabetes mellitus remission was the result of switching from NPH insulin to glargine insulin. Cytological analysis of the fluid aspirated from the pancreatic abscess revealed the presence of extracellular coccoid bacteria that were initially considered to be contaminants introduced during the examination. However, infection with Escherichia coli was finally confirmed, and appropriate antibiotics were administered based on an antibiotic sensitivity test. In dogs with pancreatitis, the administration of antibiotics is still controversial. However, antibiotics could be of some benefit for the treatment of cats with pancreatitis because inflammation of the pancreas can result in colonic bacterial translocation or ascending infection from the duodenum as mentioned above.^20,22 As in this case, cats that are suspected of having a bacterial infection are likely to need prompt initiation of antibiotic administration.

Deterioration or progression of pancreatitis can lead to a loss of pancreatic function or insulin resistance in dogs and cats.^13,14 Therefore, pancreatitis is an important cause of feline diabetes mellitus, although pancreatitis has also been described as a common concurrent disease in diabetic cats.^28,29 In a study of 104 cats with diabetes mellitus, acute pancreatitis was identified in 2 out of 37 cats that underwent necropsy and chronic pancreatitis was identified in 17 out of 37 cats. There was a trend for inferior glycemic control in cats with pancreatitis compared to those without.²⁸ In the current case, the cat experienced diabetes mellitus and a pancreatic abscess concurrently, but it was not clear whether the pancreatic abscess was a consequence of acute pancreatitis or a complication of chronic pancreatitis because no biopsy of the pancreas had been performed.

Conclusion

To the best of the authors' knowledge, this is the first report of a pancreatic bacterial abscess concurrent with diabetes mellitus in a cat. The subject's response to treatments with antibiotics and ultrasound-guided percutaneous drainage of the pancreatic bacterial abscess were successful, and the prognosis was excellent with complete remission of the diabetes mellitus.