Introduction

In dogs, many different causes of chronic diarrhoea exist. Inflammatory bowel disease (IBD) is common in dogs but differentiating it from food-responsive disease, bacterial infection or small intestinal bacterial overgrowth (SIBO) as the cause can be difficult. The diagnosis of these diseases is generally made by exclusion and intestinal biopsy. Histology is, however, non-specific for determining the cause of the inflammation, but is important for ruling out neoplastic infiltration.

General Considerations for the Doppler Examination

The Doppler examination is carried out pre- and postprandially in order to analyse the gastrointestinal haemodynamics during digestion of food. Dogs should be fasted for 12 hours and examined in the conscious state. Dogs that are nervous, panting or otherwise cannot be restrained comfortably in lateral recumbency cannot be examined with Doppler ultrasound. The use of sedatives may alter the dog's ability to eat after baseline analysis and may alter the intestinal blood flow, creating a false reading. The dogs should be placed in right lateral recumbency and the hair over the area of the left kidney should be clipped, the area cleaned with alcohol and have coupling gel applied. The coeliac (CA) and cranial mesenteric arteries (CMA) can be scanned from the left lateral abdomen. Both phased sector and curved array probes can be used for collection of the Doppler information. A lower frequency, 5 MHz transducer is ideal for larger dogs to have adequate depth penetration as the vessels of interest are located in the middle of the abdomen. Measurements and calculations of the required Doppler parameters are made with the built-in calculation software of the ultrasound unit. RI (resistive index) and PI (pulsatility index) measurements are usually included in standard calculation packages of most machines.

RI and PI measurements in both vessels are made at fasting (t₀) and at 20, 40, 60 and 90 minutes following ingestion of a meal. The test meal should not be high in fat as this can cause prolonged vasodilatation and lead to a false-positive result.

At each time-point (t₀,t₂₀,t₄₀,t₆₀ and t₉₀) and from each artery, three to five waveforms should be recorded for analysis. The RI and PI are calculated from each of three consecutive spectral tracings and averaged. The actual values and percentage decrease from baseline (time 0) are recorded and compared to established reference ranges.

Normal Doppler Findings

In healthy conscious dogs the RI and PI values calculated from spectral waveforms of the coeliac and cranial mesenteric arteries decrease during digestion until 60 minutes postprandially, after which time they increase. Postprandial vasodilatation of the mucosal vascular bed due to a combination of mechanical, hormonal and neuronal mechanisms, as well as the food composition itself, is responsible for this phenomenon. The gut hormones gastrin, cholecystokinin, vasoactive intestinal peptide, secretin and glucagon are primarily responsible for physiological vasodilatation during digestion.

During digestion, therefore, vasodilatation decreases resistance to flow, allowing an increase in diastolic flow into the vascular bed. The decreasing RI and PI values infer this response and its recovery at 90 minutes postprandially.

Two-Dimensional Grey Scale Abnormalities

Measurements of intestinal wall thickness are neither specific nor sensitive for diagnosing idiopathic inflammatory bowel disease in dogs. The range of wall thicknesses in dogs with chronic enteritis ranges from normal to slightly thickened. Wall thickness is too insensitive for detecting the presence of disease and not specific enough to ascertain the cause. The large majority of dogs with inflammatory infiltrates have normal wall layering. A small percentage of dogs with enteritis may show loss of wall layering, but these are generally associated with severe haemorrhagic, necrotising or granulomatous enteritis. Recently the meaning of altered echogenicity of the intestinal wall has been investigated. Both hyperechoic striations and speckles within the intestinal mucosa can occur.

Mild small intestinal thickening and hyperechoic striations in the mucosa are findings associated with the presence of inflammation and lymphangiectasia. The striations are thought to represent dilated lacteals. Over 80% of dogs with lymphangiectasia also have inflammatory infiltrates of the intestinal walls. The majority of these dogs have protein-losing enteropathy. Although the finding of mucosal striations is consistently associated with protein-losing enteropathy and lymphangiectasia, the cause of mucosal speckles remains unclear.

Doppler Abnormalities

Inflammatory Bowel Disease (IBD)

Compared to normal dogs, those with inflammatory bowel disease (IBD) can have an altered appearance of the spectral waveform and response to feeding. A deep diastolic notch during fasting may be present. This waveform shows a 'void' in the early diastolic period which disappears with feeding but begins to reappear 90 minutes afterwards.

Dogs with IBD show a lower RI and PI during digestion compared to normal dogs. The postprandial increase in mean diastolic blood flow velocity in dogs with IBD is also below that of normal dogs. Furthermore, those with inflammatory infiltrates with protein-losing enteropathy (PLE) and lymphangiectasia histologically, have no significant increase in mean diastolic flow post-prandially. The lower RI and PI values as well as the lack of increase in both mean total and mean diastolic blood flow during digestion would suggest a lack of the dynamic nature of the vascular bed to feeding. Histological alterations of the mucosa with reduction of the villar pattern and infiltration will also involve the capillary bed. Such disruption, if severe enough, could also result in lack of adequate blood flow to the mucosa during digestion which may result in malabsorption of nutrients. Villar infiltration and architectural destruction, lymphangiectasia and inflammatory cell infiltrates are likely reasons why an adequate haemodynamic response does not occur in affected dogs.

Food Hypersensitivity

Food hypersensitivity in dogs and humans is defined as an adverse immunological response to certain food components. A large variety of dietary antigens is known. One of the major contributors of food hypersensitivity is gluten. Wheat for example, which is a common food component, contains gluten. Gluten therefore can act as a prototype in research on food hyper-sensitivities. Hypersensitivity to gluten has been described in the Irish Setter and Soft-Coated Wheaten Terrier (SCWT). Recent findings on food hypersensitivity in SCWTs have shown that they have a familial syndrome of PLE which is associated with IBD of varying morphology and may include lymphangiectasia. Their disease is multifactorial, including hypersensitivity to various allergens.

A potential field of application for early detection of pathohaemodynamics with Doppler ultra-sound is in the diagnosis of chronic enteropathies due to canine food allergies. The current gold standard for the diagnosis is the restriction diet followed by remission and then relapse of disease when the dog is challenged with the suspected antigen. Initial investigations in a colony of allergic SCWTs showed that the haemodynamic response of the gastrointestinal tract to allergens is quantifiable with Doppler ultrasound. Significant differences were shown between Doppler parameters obtained while feeding a maintenance diet vs. an allergen-containing one. The RI and PI values recorded from the coeliac and cranial mesenteric artery pre-and postprandially after feeding a known allergen for four days were significantly different on days 2 and 4 than when the dog was fed its regular diet. A local acute inflammatory response with increased and prolonged vasodilatation most likely explains the lowered RI and PI values. The release of local vasodilatory mediators such as histamines, interleukins and others have been demonstrated in different hypersensitivities, such as atopic disease, and may contribute to the intestinal response.

Important Points

 Two-dimensional grey scale ultrasonography has limited sensitivity and specificity for diagnosing inflammatory bowel disease and determining its cause.

 Dogs with inflammatory bowel disease may have a normal to mildly thickened small intestinal wall thickness.

 The mucosal echogenicity may be increased in dogs with chronic inflammatory infiltrates.

 Hyperechoic mucosal striations may be seen in dogs with inflammatory infiltrates and lymphangiectasia, and is associated with protein-losing enteropathy.

 A sonographically normal intestinal wall does not rule out the presence of inflammation.

 Doppler ultrasonography is a feasible and reproducible method for assessing the gastrointestinal blood pre- and postprandially in dogs.

 Dogs with inflammatory bowel disease demonstrate inadequate haemodynamic responses during digestion of food.

 Increases in mean diastolic blood flow are below normal in dogs with chronic enteropathies.

 Dogs with food allergies have prolonged vasodilation and lower RI and PI values during digestion after eating food components to which they are allergic.

References

1.  Delaney F, O'Brien RT, Waller K. Ultrasound evaluation of small bowel thickness compared to weight in normal dogs. Veterinary Radiology and Ultrasound 2003; 44(5): 577-580.

2.  Hall EJ, Simpson KW. Diseases of the small intestine. In: Ettinger, SJ; Feldmann, EC. eds. Textbook of Veterinary Internal Medicine (5th edition). Philadelphia: WB Saunders, 2000; 1182-1238.

3.  Kircher P, Spaulding K, et al. Dopppler investigations of gastrointestinal blood flow in a canine model of food allergy. Journal of Veterinary Internal Medicine 2004; 18: 605-611.

4.  Penninck D, Smyers B, et al. Diagnostic value of ultrasonography in differentiating enteritis from intestinal neoplasia in dogs. Veterinary Radiology and Ultrasound 2003; 44(5): 570-575.

5.  Sutherland-Smith J, Penninck DG, et al. Ultrasonographic intestinal hyperechoic mucosal striations in dogs are associated with lacteal dilation. Veterinary Radiology and Ultrasound 2007; 48(1): 51-57.