Use of octreotide for the treatment of protein-losing enteropathy in dogs: Retrospective study of 18 cases

1 INTRODUCTION

Protein-losing enteropathy (PLE) in dogs is a clinical syndrome characterized by excessive loss of protein across the enteric mucosa.¹ It occurs as a complication of a wide variety of disorders, the most common of which include chronic inflammatory enteropathy (CIE) and intestinal lymphangiectasia (IL).² Intestinal lymphangiectasia is characterized by varying degrees of lacteal dilatation, lymphatic obstruction and lymphangitis, and can be focal, segmental or diffuse throughout the small intestine (SI). Based on breed predispositions,^1-5 there appears to be a genetic susceptibility to the development of IL. However, IL can also occur secondary to conditions that alter lymph flow, such as CIE or neoplasia. Intestinal lymphangiectasia is common in dogs with PLE with histologic evidence of lacteal dilatation described in 214/469 (46%) cases.² 76% of dogs with CIE and PLE have concurrent lacteal dilatation, suggesting CIE and IL occur together frequently.⁶ As serum albumin, cholesterol, and 25-hydroxyvitamin D concentrations decrease in cases of PLE, the presence of IL increases and the prognosis worsens.^6-9 This might suggest IL is an important contributor to disease in these cases.

Despite first being described and identified as a cause of PLE over 60 years ago,¹⁰ IL in dogs remains poorly understood with no clear consensus on the optimal therapeutic management.¹¹ Furthermore, disease-associated death occurs in 48% to 54% of dogs with PLE.^{2, 5} Standard management strategies for dogs with PLE involve dietary changes, glucocorticoids (at anti-inflammatory or immunosuppressive doses), immunosuppressive agents, and supplemental and supportive therapies, often regardless of the suspected or confirmed disorder causing PLE.^{2, 11} Some dogs fail to improve clinically, biochemically, or both in response to a variety of therapies.^{1, 2, 12, 13} Octreotide, a long-acting somatostatin analogue that is purported to decrease intestinal fat absorption, inhibit gastrointestinal vasoactive peptides, and stimulate the autonomic nervous system, is successful in the treatment of some cases of primary intestinal lymphangiectasia (PIL) in children and adults.^14-22 While octreotide has been used as a treatment for dogs and cats with chylothorax,²³ and dogs with insulinoma²⁴ and gastrinoma,^{25, 26} there are no published descriptions of its use for the treatment of IL or PLE in dogs.

The objective of this study is to describe the use of octreotide in a group of dogs with PLE including reason for and details of prescription, adverse effects, and apparent response.

2 MATERIALS AND METHODS

3 RESULTS

Twenty-four dogs were screened for inclusion into the study. Four dogs were ineligible due to an incomplete diagnostic investigations (lack of urinalysis excluding proteinuria in 2 dogs, abdominal ultrasound not performed in 2 dogs). An additional 4 dogs were excluded because the required follow-up information was not available or able to be obtained. Thus, 18 dogs were included in the study. Eleven dogs were neutered males, and 7 dogs were spayed females. The median age of the dogs was 7 (range, 2-13) years. Breeds included Maltese (2), mixed breed canine (2), Yorkshire terrier (2), and 1 each of Australian cattle dog, Boston terrier, Catahoula leopard, English bulldog, French bulldog, German shepherd dog, goldendoodle, Labrador retriever, miniature dachshund, miniature pincher, pug, and Shetland sheepdog. Reported clinical signs at the time of diagnosis included diarrhea (17/18; 94%), decreased appetite (7/18; 39%), weight loss (6/18; 33%), vomiting (5/18; 28%), and clinical signs (eg, abdominal distention, difficulty breathing) associated with cavity effusions (5/18; 28%). Median duration of clinical signs before initial presentation was 12 weeks (range, 3-208 weeks). The median body weight at time of initial diagnosis was 9.8 kg (range, 2.9-29 kg). BCS was recorded at time of diagnosis for 15/18 dogs with median of 3 (range, 1-6). MCS at the time of diagnosis was available for 9/18 dogs with median of 2 (range, 1-3). The median serum albumin concentration at the time of diagnosis was 1.6 g/dL (range, 0.8-2.4 g/dL) (Table 1). Additional relevant biochemical data from the time of initial evaluation is also reported in Table 1. All dogs had proteinuria excluded by negative dipstick or UPC <0.5. Nine dogs had normal serum bile acids activity, including 3/4 (75%) dogs with normal serum globulin concentrations. The other 9 dogs had liver insufficiency as the cause of hypoalbuminemia excluded or deemed highly unlikely based on comprehensive results of serum chemistry panel, additional diagnostic investigation including abdominal ultrasound, and case follow-up. Fourteen dogs had fecal flotation performed with no parasites detected. Of those 14 dogs, 3 dogs were also tested for giardia and found to be negative for giardia antigen in the feces. The remaining 4 dogs were treated with broad spectrum anthelmintics at the time of initial diagnosis. All dogs had TLI concentrations >5.0 ng/mL excluding EPI as the cause of their clinical signs and hypoalbuminemia. Twelve dogs were receiving corticosteroids prescribed by a previous veterinarian at the time of their initial presentation, thus hypoadrenocorticism was unable to be excluded in those cases. In the other 6 dogs, hypoadrenocorticism was excluded via a baseline cortisol concentration >2 μg/dL. Abdominal ultrasound was performed in all cases. The most frequently reported abnormalities included peritoneal effusion (12/18; 67%), SI hyperechoic mucosal striations (12/18; 67%), SI mucosal thickening (8/18; 44%), hyperechogenicity to the SI mucosal layer (4/18; 22%), and 3/18 (17%) dogs each had SI speckling and altered SI wall layering. One dog had mild jejunal and hepatic lymphadenopathy. One dog, with an ultimate diagnosis of lipogranulomatous lymphangitis had severe segmental thickening and altered layering localized to the distal jejunum and proximal ileum, otherwise changes were described as diffuse throughout the SI. Endoscopy was performed in 9/18 (50%) of cases with apparent lacteal dilatation noted in 4/9 (44%) cases. An additional 4 dogs had laparotomy performed with biopsies of the SI obtained; in one of those cases serosal and mucosal lacteal dilatation was noted on gross examination. Video capsule endoscopy was performed in 1 dog which revealed diffuse apparent small intestinal lacteal dilatation and multifocal GI erosions and ulcerations. Histopathology of the small intestine was available for 13/18 (72%) of cases. For cases that had endoscopy performed (n = 9), duodenal biopsy was performed in all cases, with ileal biopsy concurrently performed in 4/9 (44%) of cases. For dogs that had biopsies performed via laparotomy, the following tissue was available for evaluation: duodenum, jejunum, and ileum in 1 dog, jejunum and ileum in 1 dog, and ileum only in 2 dogs. Of the 13 dogs that had histopathology of the SI performed, 12 (92%) had histologic evidence of lymphangiectasia. Lacteal dilatation was noted as mild in 4/12 (33%), moderate in 6/12 (50%), and severe in 2/12 (17%). Inflammatory infiltrates of the SI were described in all 13 dogs who had histopathology performed. The infiltrate was described as lymphoplasmacytic in 6/13 (46%) dogs, lymphoplasmacytic with eosinophils in 5/13 (38%) dogs, lymphoplasmacytic with neutrophils in 1 (8%) dog, and lymphoplasmacytic with histocytes in 1 (8%) dog.

After the initial evaluation, 16 dogs were prescribed a veterinary therapeutic diet (n = 12 low fat and n = 4 hydrolysate diet) and 2 dogs were prescribed home-cooked diets formulated by a board-certified veterinary nutritionist. Median fat content for the veterinary therapeutic low fat diets prescribed was 18.7 g/1000 kcal (range, 18.1-23.2 g/1000 kcal). The median fat content for the 4 dogs prescribed a hydrolysate diet was 29.75 g/1000 kcal (range, 26.5-47.9 g/1000 kcal). The fat content for the 2 dogs initially prescribed home cooked diets were 11 g/1000 kcal and 15 g/1000 kcal. Additionally, 17/18 (94%) of dogs were either newly prescribed (n = 13) or maintained (n = 4) on prednisone (n = 16) or prednisolone (n = 1). Information regarding how long the dogs had previously been treated with steroid for was only available for 2 dogs. One dog had been treated with steroids for 6 months previously and another for 3 months. Median dose of steroid prescribed was 2 mg/kg, PO, daily (range, 0.5-2.5 mg per kg, PO, q daily). Additional medications prescribed at T0 or in the interim between T0 and T1 included: azathioprine (n = 1), calcitriol (n = 2), calcium gluconate (n = 4), chlorambucil (n = 2), clopidogrel (n = 4), cobalamin (n = 10), cyclosporine (n = 4), enrofloxacin (n = 1), maropitant (n = 3), metronidazole (n = 10), omeprazole (n = 3), ondansetron (n = 1), pancreatic enzymes (n = 1), tylosin (n = 1), and sucralfate (n = 1).

The median time from diagnosis of PLE until T1 (initiation of octreotide) was 5 months (range, 3-48 months). The median time from T0 (time of first evaluation by clinician prescribing octreotide) to T1 was 41 days (range, 0-1080 days). Ongoing clinical signs were noted in all dogs at T1 and included diarrhea, vomiting, progressive weight loss, and ongoing effusions. At T1 all dogs were receiving some form of steroid therapy (prednisone, prednisolone, or budesonide PO or dexamethasone SQ) and all dogs were eating veterinary therapeutic diets or home-cooked diets formulated by a board-certified veterinary nutritionist. At T1, 5 dogs were being fed home cooked diets with a median fat content of 13 g/1000 kcal (range, 11-15 g/1000 kcal). Nine dogs were being fed veterinary therapeutic low fat diets with a median fat content of 18.25 g/1000 kcal (range, 18.1-23.2 g/1000 kcal). The 4 remaining dogs were being fed a veterinary therapeutic hydrolysate diet with a median fat content of 26.5 g/1000 kcal (range, 26.5-47.9 g/1000 kcal). Median albumin at T1 was 1.7 g/dL (range, 1.0-3.1 g/dL). Additional relevant biochemical data from all dogs at T1 is provided in Table 1.

Generic octreotide acetate was prescribed in all cases. The median dose was 20 μg/kg, SQ, daily with a range of 4-39 μg/kg, SQ, daily. The dose was divided q12 in 12 dogs, and q8 in 6 dogs. At the time of octreotide initiation (T1), 12/18 dogs had no other changes made to their therapeutic plan. The other 6 dogs had changes in diet, increases in dose or frequency of immunosuppressive therapies, introduction of new immunosuppressives, tapering of steroid or immunosuppressive therapies at T1, or a combination of these changes. Two dogs were changed from a veterinary prescription low fat diet to a home cooked diet. An additional 2 dogs had their total daily dose of cyclosporine increased, 1 dog had cyclosporine administration initiated, 1 dog was switched from corticosteroid administered orally to corticosteroid administered subcutaneously, and 1 dog had chlorambucil administration discontinued. Owners reported adverse effects attributed to the octreotide in 3/18 (17%, 95% CI [4%, 41%]) dogs. In 1 dog, apparent pain associated with the injection of octreotide was reported, which was mitigated with the addition of sodium bicarbonate to the needle hub. The additional 2 dogs had urgent and watery diarrhea reported to occur within 30 minutes of administration of octreotide, which led to discontinuation of the medication 8 days into therapy for 1 dog. In the other dog, the dose was reduced from 10 μg/kg, SQ, q8h to 5 μg/kg, SQ, q 8h and this apparent adverse effect resolved. The median time from T1 to T2 for all dogs was 20 days (range, 8-38 days).

The median time from T1 to T2 (time of first evaluation after the initiation of octreotide) for the 12 dogs in which no other changes to therapy were made was 21 days (range, 8-38 days). Median serum albumin for these dogs at T1 was 1.8 g/dL (range, 1-3.1 g/dL) and 2.1 g/dL (range, 1.1-3.8 g/dL) at T2. In these 12 dogs with no other changes made to their therapies, apparent improvement in clinical signs after initiation of octreotide was reported in 6/12 (50%). The improvements in clinical signs noted were improved consistency of stool (n = 4), reduction in peritoneal effusion (n = 1), improvement in appetite (n = 2), and better energy level (n = 2). Changes in serum albumin between T1 and T2 for the 6 dogs with an apparent improvement in clinical signs vs. the 6 dogs with no apparent improvement in clinical signs are shown in Figure 1. Of the 6 dogs with an apparent improvement in clinical signs in response to octreotide, 5 had an improvement in serum albumin concentration at T2 compared to T1, and 1 dog's serum albumin concentrations was considered unchanged. Serum albumin concentration was improved in 1 dog and unchanged in 5 dogs at T2 compared to T1 in the dogs with no apparent improvement in clinical signs in response to octreotide. Between T1 and T2, dogs with an apparent improvement in clinical signs (n = 6) had varied changes in BCS with 1 dog having a reduction in BCS, 4 dogs having no change in BCS (n = 4), and 1 dog having an improvement in BCS. Of the dogs evaluated for response with no apparent improvement in clinical signs, 2 had a reduction in BCS and 4 had no change in BCS. Serum cholesterol concentrations increased in all 6 dogs with an apparent clinical response to octreotide, and in 3/6 dogs with no apparent response to octreotide (Figure 2). Biochemical data from the 12 dogs who were evaluated for a response to octreotide therapy between T1 and T2 is provided in Table 2.

The dose of octreotide for the 6 dogs with apparent clinical improvement in response to octreotide was 10-20 μg/kg, SQ, q 8-12h (median dose 30 μg/kg, SQ, daily, range 30-60 μg/kg, SQ, daily). The dose of octreotide for dogs that did not show clinical improvement in response to octreotide was 2-10 μg/kg, SQ q 8-12h (median dose 20 μg/kg, SQ, daily, range 4-30 μg/kg, SQ, daily). All but 1 dog (5/6) showing an apparent clinical response to octreotide was eating a veterinary prescription low-fat diet (n = 2) or a veterinary nutritionist formulated homecooked diet (n = 3) that was formulated to be low in fat at the time octreotide was initiated (T1). The median fat content at T1 for the dogs with an apparent improvement in clinical signs in response to octreotide was 15 g/1000 kcal (range, 14-26.5 g/1000 kcal). Of the 6 dogs that did not improve, 3 dogs were eating veterinary prescription low-fat diets, the other 3 were eating veterinary prescription hydrolyzed diets. Of the 3 dogs eating veterinary prescription hydrolyzed diets, 2 had previously been fed veterinary therapeutic low fat diets. The median fat content at T1 for the dogs with no apparent improvement in clinical signs in response to octreotide was 20.7 g/1000 kcal (range, 18.1-47.9 g/1000 kcal). Eight of the 12 dogs that had no other therapies started or changed at the time of octreotide initiation had histopathology available. Four of those dogs were considered responders, all of which had IL documented. In 2 of these dogs, the IL was noted to be mild, in the other 2 the IL was noted to be moderate. The other 4 were considered non-responders, all of which also had IL documented histologically, and similarly, 2 were noted to have mild IL, and the other 2 moderate IL.

At T2, octreotide was discontinued in 2 dogs; 1 due to urgent and watery diarrhea occurring 10 minutes after administration of octreotide and the other due to apparent lack of efficacy. The dose or frequency was decreased in 2 dogs, and the dose or frequency was increased in 3 dogs. Octreotide was continued at the same dose and frequency for the remainder of the dogs. The median time from T2 to T3 (second evaluation after initiation of octreotide) for all dogs receiving octreotide (n = 16) was 16.5 days (range, 7-60 days). Median serum albumin for these 16 dogs at T3 was 2.25 g/dL (range, 1.4-3.8 g/dL). Long-term follow-up data was available for 17 dogs (median of 11 months post-diagnosis, range, 2-72 months). Twelve of the 17 dogs are alive at the time of manuscript preparation, 6 of which are still receiving octreotide at a median dose of 20 μg/kg, SQ, daily (range, 12-30 μg/kg, SQ, daily). The remaining 5 dogs were euthanized due to ongoing PLE and clinical consequences. Last known serum albumin concentration was available for 10/12 dogs alive at the time of manuscript preparation. The median serum albumin concentration for these dogs was 2.5 g/dL (range, 2.2-3.6 g/dL) and was measured a median of 12 months post-diagnosis with a range of 2-36 months. Five of the dogs considered responders to octreotide were alive at the time of manuscript preparation; 1 dog was euthanized due to ongoing PLE and clinical consequences.

4 DISCUSSION

Octreotide acetate was most commonly prescribed in dogs with PLE that had ongoing clinical signs after weeks to months of administration of corticosteroids, immunosuppressives, or both and a variety of therapeutic diet trials. Intestinal lymphangiectasia was suspected or histologically confirmed in all dogs prescribed octreotide. Of the 12 dogs that had octreotide prescribed with no concurrent changes to their therapies, 6 appeared to show clinical improvement after the introduction of octreotide. Octreotide acetate was well tolerated in dogs with PLE.

Despite treatment with standard therapies such as therapeutic diets, glucocorticoids, immunosuppressives, and variety of supportive care measures, approximately 50% of dogs with PLE have disease-associated death.^{2, 5} Thus investigation into novel approaches both with adjustments of standard therapies and introduction of new strategies is necessary. Intestinal lymphangiectasia is a frequent cause or contributor to PLE in dogs.^{2, 6} In humans with PIL non-responsive to standard dietary therapy and supportive care, octreotide acetate has been used with varying success.^{14-16, 18, 19, 21, 29} One review article described 9 case reports of children and adults with PIL treated with octreotide, 7 of which reported improvement in response to octreotide therapy.¹⁸ Octreotide is a somatostatin analogue. Information is limited on the exact mechanism of action of somatostatin in decreasing chyle production and intraluminal pressure within the lymphatic system, however it is thought to exert these effects through inhibition of gastrointestinal vasoactive peptides and stimulation of the autonomic nervous system.²¹ In rodent and dog models of disease, somatostatin reduces intestinal absorption of fat and concentrations of triglycerides within the thoracic duct.^{30, 31} Somatostatin and its analogues reduce lymphatic fluid outflow in children and adults who have sustained iatrogenic thoracic duct injuries.^{32, 33} Evidence of its benefit for the treatment of PIL in children and adults is limited to case reports and case series.^{14-16, 18, 19, 21, 29} Thus, standardized dosing recommendations are not available, however typically an induction dose of 5-20 μg/kg, SQ, every 12 hours is followed by a maintenance dose in which the patient is switched to slow-release octreotide and given a similar dose every 3-4 weeks.^{18, 21, 29} In some instances, patients with PIL treated with octreotide achieved complete clinical and biochemical remission,¹⁶ however more commonly there is reduced the requirement for albumin infusions and smaller volume of stool.²¹

The lack of control group and the wide variations in dosing and concurrent therapies make it similarly challenging to draw any conclusions about the benefit of octreotide in our group of dogs with PLE. Anecdotally, 50% of dogs for which the addition of octreotide was their sole change in therapy appeared to benefit from it. Though no direct conclusions can be drawn, we can report that the median dose of octreotide in the 6 dogs that might have benefitted from octreotide was 30 μg/kg, SQ, daily with a range of 30-60 μg/kg, SQ, daily and the majority of the dogs that possibly benefitted from octreotide were eating veterinary prescription or veterinary nutritionist formulated low-fat diets.

Adverse effects of octreotide were reported in 3/18 dogs and included apparent pain associated with injection in 1 dog, and urgent watery diarrhea in 2 dogs. Common adverse effects of octreotide that have been described in humans include nausea, diarrhea, flatulence and abdominal cramping.^{17, 34} Other less common but reported adverse effects include headache, hyperglycemia, constipation, pruritus, rash, alopecia, and dizziness.^{17, 34} Adverse effects in humans are noted to spontaneously resolve in many patients within 7-14 days.³⁴ Information regarding the frequency of these adverse effects in human patients is lacking.

Limitations of this study are present due to its retrospective nature and the lack of a control group. Additionally, some of the cases included were missing diagnostic data to fully characterize their disease process, including specialized infectious disease testing. Even in cases where endoscopy was performed, sections of the intestine biopsied varied, which could affect the diagnosis.^{35, 36} Because of this and the uncontrolled nature of the study, we recognize that conclusions cannot be drawn regarding the benefit of octreotide for the treatment of PLE or its clinical consequences, but rather we can report its usage and the incidence of adverse effects. A prospective and controlled clinical study would be necessary to determine if and when octreotide should be utilized as a therapy for dogs with IL and PLE.

CONFLICT OF INTEREST DECLARATION

The authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION

The authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

The authors declare no IACUC or other approval was needed.

HUMAN ETHICS APPROVAL DECLARATION

The authors declare human ethics approval was not needed for this study.