Case study in canine intestinal lymphangiectasia

Abstract

A 9.52 kg, 9-year-old, spayed female beagle was presented with the chief complaint of abdominal distention of 1 week’s duration. A presumptive diagnosis of canine intestinal lymphangectasia was arrived at by exclusion of other causes for the patient’s ascites. The patient was successfully treated with dietary modification and immunosuppressive therapy.

A 9.52 kg, 9-year-old, spayed female beagle was presented to the Fox Valley Veterinary Clinic (North Aurora, Illinois, USA) with the chief complaint of abdominal distention of 1 wk duration (Figure 1). The physical examination was relatively unremarkable, except for ascites with a pronounced abdominal fluid wave and a body condition score (BCS) of 2.5/5, based on a scale where a score of 1/5 is severely emaciated and a score of 5/5 is morbidly obese. From the history obtained from the client at the time of presentation, the patient’s attitude and appetite were normal, but for about 3 wk prior to presentation, her stools were semisoft in the early morning, became progressively looser through the day, and eventually became unformed. There was no change in the frequency of defecation, though defecation seemed to be taking more time. Sporadic episodes of vomiting had been noted, and the dog had had a dry, intermittent cough since being adopted 2 y previously. She was currently being treated with a hyposensitization program for inhalant allergies to Tyrophagus putrescentiae, a type of grain mite, and had been for a period of approximately 9 mo at the time of presentation. The patient was not on any other medications.

Figure 1.

Patient at initial presentation

Hyperadrenocorticism was considered as a possible cause for the dog’s “potbelly” appearance, because older, spayed, female beagles have been reported to be at an increased risk for developing this disease (1), but this was thought to be less likely in light of the history and physical examination. So 2 broad categories of conditions causing ascites were focused on 1) Forms of right-sided heart failure that could cause congestion in the systemic vasculature, such as tricuspid valve regurgitation, pulmonic stenosis, pericardial effusion, pericarditis, and heartworm disease; and 2) forms of hypoalbuminemia caused either by decreased albumin production or increased albumin loss, resulting in low intravascular oncotic pressure. The decreased albumin production would be due to a hepatic insufficiency (HI), while the increased albumin loss could be due to a protein-losing nephropathy (PLN), protein-losing enteropathy (PLE), acute or chronic blood loss, or starvation. With the client’s approval, the patient was admitted for preliminary diagnostic tests.

Blood was submitted for a complete blood (cell) count (CBC), biochemical analysis, values for postprandial bile acid, and an antigen heartworm test. Thoracic radiographs were taken to evaluate the heart and look for evidence of pleural effusion. A lead II electrocardiograph (ECG) was also taken.

Results from the antigen heartworm test (Heska Solo Step HC; Heska Corporation, Fort Collins, Colorado, USA) were negative, and no evidence of heart disease was apparent on either the ECG or the thoracic radiographs. Evaluation of the rest of thorax and visible extrathoracic structures showed no evidence of pleural effusion, but the liver did appear small. In light of these findings and the normal cardiac auscultation during the physical examination, right-sided heart disease was ruled out and the ascites was attributed to hypoalbuminemia, most likely due to HI, PLN, or PLE.

An abdomenocentesis was performed and ultrasonographs of the abdomen were taken. The abdomenocentesis yielded a clear transudate, with no cellularity on cytologic examination. The fluid had a specific gravity of 1.006 and total solids < 2.0 g/100 mL. The ultrasonographs showed no abnormalities, except for the marked ascites and what appeared to be a small liver. No nodularity was seen within the liver, which appeared homogenous and hyperechoic. A tentative diagnosis of liver fibrosis resulting in HI was made, and the client was given a guarded prognosis.

With client consent, an ultrasound-guided liver biopsy was obtained. An activated clotting time (ACT) was performed to quickly screen for abnormalities of the intrinsic clotting factors that may have been affected by the HI. The results of the ACT were within normal limits, so the patient was premedicated with atropine sulfate (generic; Phoenix Scientific, St. Joseph, Missouri, USA), 0.04 mg/kg bodyweight (BW), SC, and sedated with meditomidine (Domitor; Pfizer, Exton, Pennsylvania, USA), 8.8 μg/kg BW, IM, and butorphanol tartrate (Torbugesic; Fort Dodge Animal Health, Fort Dodge, Iowa, USA), 0.22 mg/kg BW, IM. In addition to providing sedation, these agents also provided analgesia for the procedure. Four liver samples were collected percutaneously. The patient was released later that night with no treatment prescribed, pending the results of the biopsy and other tests.

Results of the biochemical, postprandial bile acids values, and the CBC were received the following day (Table 1, day 0). Mildly elevated aspartate aminotransferase (AST; SGOT) and creatinephosphokinase (CK) levels were noted and attributed to the muscle loss noted in the physical examination (low BCS). All other liver function tests were within the normal ranges, including the postprandial bile acids. The patient was moderately panhypoproteinemic. Most of the remaining biochemical results were within the normal limits and those that were not were easily attributed to individual variation. Results from the CBC were within normal limits, except for moderately low absolute lymphocyte and eosinophil counts. The interpretation of these results excluded hyperadrenocorticism, HI, and acute or uncompensated chronic blood loss. Two days later, the exclusion of HI was further supported by liver biopsy results that showed only mild hydropic degeneration with attendant cholestasis.

Table 1.

Selected biochemical and complete blood cell count results

	Normal range	Day 0	Day 15	Day 22	Day 48	Day 62	Day 78	Day 97	Day 181
Total protein	51–78 g/L	25	27	31	30	36	34	44	44
Albumin	25–36 g/L	11	13	13	14	20	18	24	22
Globulin	28–45 g/L	14	14	18	16	16	16	20	22
Cholesterol	2.89–8.48 mmol/L	3.03	2.59	4.34	3.08	2.28	2.89	4.16	3.89
Calcium	2.05–3.09 mmol/L	2.07	2.07	2.24	2.02	2.20	2.27	2.57	2.05
AST (SGOT)	5–55 μ→/L	60	52	49	60	28	34	26	43
CK	10–200 μ→/L	335	271	168	427	151	132	128	166
Absolute lymphpocyte count	1000–4800/μ→L	476	—	344	585	246	565	534	860

AST — aspartate aminotransferase; CK — creatinephosphokinase

Having ruled out HI and acute blood loss as a cause for the hypoalbuminemia, PLN, PLE, and semicompensated for chronic blood loss became the next 3 most likely causes. Chronic blood loss and PLN were the easiest to rule out diagnostically. The client confirmed that there had been no melana or frank blood in the stools, and 3 fecal flotations and a direct fecal smear ruled out gastrointestinal parasites as a possible cause for chronic blood loss. So these results, in conjunction with results from the CBC and the history, ruled out chronic blood loss. A urine sample, collected via cystocentesis, showed only trace protein. The urinalysis results, in conjunction with normal renal values obtained from the earlier biochemical analysis, were used to rule out a PLN. A urine protein:creatinine ratio could have been used to further evaluate for renal protein loss (2), but this was not considered necessary in this case.

By exclusion, PLE was left as the final cause for the hypoalbuminemia (3). The differential diagnoses for PLE include generalized disorders, as well as primary gastrointestinal diseases (2,3). The generalized disorders include congestive heart failure, nephrotic syndrome, and metastatic neoplasia. No evidence was found in the diagnostic tests that had been done that would indicate any of these general disorders as a cause for the hypoproteinemia, leaving primary gastrointestinal causes for investigation.

The primary gastrointestinal causes for PLE can be divided into 2 broad categories: 1) disorders that increase mucosal permeability and 2) lymphatic disorders (3). Increased mucosal permeability can be further divided into disorders of parasitism; mechanical obstruction, such as intussusception and chronic foreign body; and inflammation, such as food allergies and inflammatory bowel disease (IBD). Lymphatic disorders can be caused by intestinal lymphangiectasia (IL) and lymphosarcoma. Results from the previous diagnostic tests ruled out parasitic and mechanical causes for increased mucosal permeability. Food allergies and IBD caused by food-related allergies were not considered likely differential diagnoses, because the patient’s diet consisted of only a specially formulated low allergen food (Hill’s prescription diet canine z/d low allergen; Hill’s Pet Nutrition, Topeka, Kansas, USA). Results from the physical examination and imaging studies discounted generalized or localized lymphosarcoma, so the top differential diagnoses became IL and IBD due to non-food-related allergies, with IL being the more likely, because it is the most common disorder associated with PLE (2,4). The only way to definitively diagnose either of these conditions is histologically through collection of intestinal biopsies via endoscopy or laparotomy (2,3,5,6). Endoscopy has the advantages of being less invasive and allowing visualization of lesions to biopsy, provided that the condition is not focal and outside the effective reach of the endoscope. Laparotomy has the advantage of allowing visual evaluation of all of the abdominal organs, including the entire gastrointestinal tract and the gastrointestinal lymphatics. However, it is associated with a high occurrence of postsurgical complications, including death or dehiscence due to delayed wound healing associated with hypoalbuminemia (2–7). A concern in collecting a biopsy sample by either method is that mild forms of the disease may not yield appreciable lesions and a diagnostic sample may not be obtained (6). In this patient, an endoscopic biopsy would have been the preferred choice, but the client was not interested in having a biopsy taken via either method, due to the cost of endoscopy and possible complications of laparotomy. In the client’s opinion, the patient’s quality of life was still favorable, based on the patient’s good attitude and appetite, so it was decided to repeat the biochemical analyses to reevaluate the panhypoproteinemia and perform a fecal alpha-1 proteinase inhibitor (fecal ∞₁-PI) test to screen for PLE. Fecal ∞₁-PI is of a similar molecular weight to albumin, but, unlike albumin, it is resistant to proteolytic degradation within the gastrointestinal tract. Therefore fecal ∞₁-PI can be used as a marker for the intestinal loss of albumin (8). Although the fecal ∞₁-PI test has a high sensitivity for hypoalbuminemia in cases of PLE, its specificity is low. However, the specificity can be increased by ruling out other causes of hypoproteinemia (8), such as was done in this case. The client was also willing to institute a conservative treatment for IL, which consisted of a dietary change to reduce long-chain triglycerides (LCT) in the diet. Long-chain triglycerides in the diet stimulate the loss of lymph to the gastrointestinal lumen through chylomicron formation (9). There is some evidence that LCT may be proinflammatory, which could also stimulate the loss of lymph from inflamed lacteals (6). It has been reported that reducing the presence of LCT in the diet can decrease lymph flow 10-fold in the mesentary (6). It was hoped that a fat restrictive diet (Hill’s prescription diet r/d; Hill’s Pet Nutrition) would decrease lymph loss to the gastrointestinal lumen, enough to reduce the patient’s hypoalbuminemia and ascites. In addition, the hyposensitization treatments for inhalant allergies were stopped to reduce inflammatory stimuli in general, though there was no evidence to suggest that those antigens were causing a direct problem.

The patient was admitted to hospital to ensure proper sample handling during the collection of the stool samples for the fecal ∞₁-PI test and thus minimize the chances of receiving a false negative. Although fecal ∞₁-PI is resistant to degradation in the gastrointestinal tract, it degrades if left at room temperature. Freezing will slow this process, but repeated freezing and thawing must be avoided, because it accelerates the degradation. Three stool samples from 3 separate defecations were collected for the fecal ∞₁-PI test and the patient was released. The samples for fecal ∞₁-PI testing were submitted to the Gastrointestinal Laboratory, Texas A&M University, College Station, Texas, USA. Expected turn-around time for the test was about 3 to 4 wk. A recheck examination was scheduled to coincide with the return of the fecal ∞₁-PI test results. Blood was also collected at this time for biochemical analyses. The results of these analyses (Table 1, day 15) were similar to those of the original analyses. The patient was still moderately panhypoproteinnemic. Although hypoalbuminemia is the most consistent clinicopathologic finding in IL, globulin levels are often affected as well, and panhypoproteinnemia is supportive of a diagnosis of IL (6,7). Special attention was also paid to cholesterol and calcium levels, because these values, when low, are also supportive of a diagnosis of IL. Hypocholesterolemia is attributed to gastrointestinal loss and lipid malabsorption. Hypocalcemia is mainly an artifact of low serum albumin concentration, but it may be exacerbated by decreased intestinal calcium absorption (6). Cholesterol levels in the patient were mildly decreased and calcium levels were at the low end of the normal range. In hindsight, the CBC should have been repeated to monitor the absolute lymphocyte count, because lymphopenia is a relatively consistent finding in patients with IL and, when present, lends additional support to a diagnosis of IL (6,7).

About 1 wk after initiating the new diet, the patient was presented for vomiting. Two days prior to presentation, the patient had vomited about half a meal of food, shortly after feeding. On the day of presentation, the patient vomited 3 times in a 10-minute period, shortly after feeding. The vomitus contained food all 3 times, but frank blood was also noted with the 3rd vomition. A physical examination yielded the same findings as at the time of the original presentation for ascites. The patient was given aminopentamide hydrogen sulfate (Centrine; Fort Dodge Animal Health), 0.2 mg, IM, as a fast onset antiemetic. The following medications were prescribed for use at home: metoclopramide HCl (generic; PLIVA, East Hanover, New Jersey, USA), 5 mg, PO, q8h for 5 d and sucralfate (generic; Merckle GmbH, Blaubeuren, Germany), 0.5 g, PO, q8h for 5 d, as an antiemetic and gastro-protectant, respectively. It was thought that the vomiting might be due to the increased volume of the fat restrictive diet that was being fed compared with the volume of the previous diet. The increased volume was necessary to maintain an equivalent caloric intake. The client was advised to feed smaller meals more frequently, to prevent over-filling the stomach. Another biochemical analysis and CBC were performed to make sure that the vomiting did not indicate a more complicated problem. The results of the biochemical analyses and CBC (Table 1, day 22) showed some improvement in the levels of total protein, globulin, cholesterol, and calcium. Albumin and absolute lymphocyte values remained relatively unchanged.

The results of all 3 fecal ∞₁-PI test samples were consistent with PLE (Table 2). On physical examination, the ascites was less pronounced than on original presentation, but it was still prominent. No other significant changes were noted on physical examination. The client reported that the consistency of the stools had improved slightly, but that they were still soft with a gradual progression to being loose as the day progressed. There had been no further vomiting since the last physical examination. The client also felt that the patient’s attitude had been a little quieter lately, but her appetite was normal. Based on the clinical signs, fecal ∞₁-PI test results, and biochemical and CBC profiles collected thus far, it was felt that IL was the underlying condition for this PLE. Obtaining a biopsy was discussed again as a means to get a definitive diagnosis, but the client did not consider that this was necessary when weighed against the costs involved and possible complications. The use of corticosteroids as an immunosuppressive agent to reduce inflammation of the assumptively inflamed lacteals of the intestine, and thus decrease loss of lymph to the gastrointestinal lumen (2,4,5), was considered, but the client was reluctant to start a medication that would be needed indefinitely. It was decided to continue the dietary therapy for another month and then, if there was no improvement, to add corticosteroids to the treatment regimen. Blood was submitted for a biochemical analysis and CBC, with an additional follow-up biochemical analysis and CBC scheduled in 2 wk, and another, at the time of the recheck physical examination, in 4 wk.

Table 2.

Any individual sample > 15.0 μg/g is consistent with protein-losing enteropathy

	Sample A	Sample B	Sample C
Fecal ∞-1 PI (μg/g)	29	51.7	53.2

The results of the biochemical analysis and CBC (Table 1, day 48) did not show any improvement over the previous results; however, the results from the 2-week follow-up (Table 1, day 62) did show an improvement of the total protein and albumin values, although both were still below the reference range. The cholesterol and absolute lymphocyte values were at their lowest since original presentation. No change in treatment was made at this time.

At the 4-week recheck and physical examination, the patient seemed a bit depressed. There was no appreciable change in the ascites since the last examination. Although the patient’s weight had remained at 9.5 kg, the BCS was reevaluated as only a 2/5 and the hair coat was dull in appearance. No other abnormalities were found. The client reported that although the appetite remained good, the patient was lethargic and uninterested in the daily routine, except for mealtimes. Although the most recent biochemical analyses had shown some small improvement in the hypoalbuminemia, the general well-being of the patient appeared to have declined. Prednisolone (generic; Vetamix, Shenandoah, Iowa, USA) was dispensed at an immunosuppressive dosage of 2 mg/kg BW, PO, q12 h for 5 d, then to be decreased to 1 mg/kg BW, PO, q12h for at least 6 wk or until a change in clinical signs was seen (4). If remission of the disease was obtained, the dose would be lowered to the lowest effective dose to maintain remission of clinical signs (4). On day 78, there were no significant improvements (Table 1, day 78), but the cholesterol and absolute lymphocyte values had returned to approximately their levels at day 48. A recheck examination in 2 wk was scheduled.

Four days after beginning the prednisolone, the client reported that the ascites was gone and that the patient’s attitude was brighter and she was more active (Figure 2). In addition, the client felt that the consistency of the stools had improved, although they remained soft. The patient was reweighed and the 1 mg/kg dose of prednisolone adjusted to reflect the current weight of 7.25 kg. It was felt that medium-chain triglycerides (MCTs) could be helpful in regaining the patient’s body condition. Although there is debate about the mechanism of absorption for MCTs, it is felt that MCTs can provide some of the favorable aspects, mainly increased calories, of dietary lipids, without stimulating lymph loss (5). Medium-chain triglycerides’ oil was not readily available from the practice’s regular suppliers or local pharmacies, but an alternative source was found at a local health food store. Pure coconut oil, which is an MCT, was added to the diet at 1 tablespoon, q12h. It has been reported that MCT oil is unpalatable to most patients, but the client reported that the patient found the pure coconut oil very palatable. The plan for recheck was modified and a follow-up examination, biochemical analyses, and CBC were scheduled for in 3 wk.

Figure 2.

Patient after 4 d of corticosteroid treatment.

At the follow-up examination, the patient was bright, alert, and active. The client reported that the patient was active and playful at home and that the consistency of the stools had continued to improve but were occasionally soft. The patient’s weight had increased to 7.62 kg. No evidence of ascites was appreciated (Figure 3). The only abnormalities noted were a BCS of 2.5 and a dull hair coat. The biochemcal analyses and CBC were repeated (Table 1, day 97). The albumin and globulin levels had increased to a point where the patient was now only mildly panhypoproteinemic. The cholesterol and calcium values were within the reference ranges, though the absolute lymphocyte count remained relatively unchanged. It was decided to continue the prednisolone, as planned, and to continue the pure coconut oil in the diet until a BCS of 3/5 was attained.

Figure 3.

Patient after approximately 1 mo of corticosteroid treatment.

Although regular monitoring of protein levels would have been appropriate (3,4), the client felt the patient was doing well and it was unnecessary. Fecal consistency was used as an indicator of corticosteroid effectiveness in determining the lowest effective every-other-day (eod) dose. The patient had been maintained free of clinical signs at a level of 2.5 mg of prednisolone eod. A BCS of 3/5 was attained at a body weight of 9.07 kg, at which time the pure coconut oil was removed from the diet. The 3/5 BCS has been maintained on the fat restricted diet. Biochemical analyses and a CBC were obtained 181 d after the original presentation (Table 1, day 181). The albumin, globulin, cholesterol, and calcium values were relatively unchanged from the last biochemical analysis at 97 d. The CBC showed an increase in the absolute lymphocyte count, although it was still below the reference range.

The 3 cornerstones for treatment of IL include addressing the underlying cause, the use of anti-inflammatory medications, and dietary modification. But as is demonstrated by this case, a favorable outcome can be achieved, even if the underlying cause cannot be determined. Unfortunately, resolution is rare and in most cases remission of clinical signs is the best that can be achieved for patients with IL. Even in patients where remission can be attained, the longest reported survival time after diagnosis of IL is 2 y (6). As of this writing, 507 d since the initial presentation for ascites, the patient remains free of clinical signs on a maintenance dose of 2.5 mg (0.28 mg/kg) of predinisolone eod. and continues to maintain a BCS of 3/5 on the restricted fat diet.