Discussion
Our results indicated that many equids with gastrointestinal disease/colic have increased serum pancreatic (DGGR) lipase concentrations. The results showed a significantly higher median lipase activity in horses with colic than those with colitis or peritonitis, suggesting that a degree of pancreatitis may be present in many colic cases, although this does not indicate causation, and, in many cases, this is likely to be a secondary response, as is noted in people and other species (Watson et al 2010; Watson 2015). An increased risk of pancreatitis, as determined by increased serum pancreatic lipase concentrations, has previously been reported in colic (Johnson et al 2019; Lanz et al 2022). However, increased pancreatic lipase activity has also been reported in animals, including horses, with a variety of other non-gastrointestinal disease. For example, in the study of Johnson et al (2019), hyperlipaseamia was identified in 16 horses affected by various conditions, including poor athletic performance, lameness, anaemia, rhabdomyolysis, laminitis, cellulitis, juvenile osteoarthritis and training. However, in the current study, DGGR lipase concentrations were significantly higher in colic cases compared to colitis and peritonitis cases, providing further evidence to suggest a possible association between colic and pancreatitis. In addition to this apparent association with colic, there were significant associations between pain severity and lipase values in this study, with higher lipase activities in horses with more severe pain. However there was no significant association with heart rate. Although heart rate is to some extent dependent on the level of pain and sympathetic response, the rate is also significantly affected by other factors including vascular volume and cardiovascular status (including hypovolaemia and systemic inflammatory syndrome), which may explain why there was an observed association with pain but not with heart rate. We also found no association between pancreatic lipase activities and breed in this study; this differs from the results of Johnson et al (2019) who found that hot-blooded horses were more frequently affected.
Johnson et al (2019), reported increased DGGR-lipase activity above the upper reference limit of 20 U/L in 40% of 109 hospitalised horses; 33 of these horses had gastrointestinal disease, of which 28 (85%) had increased pancreatic lipase activity. This compares with 39.5% of equids with gastrointestinal disease that had hyperlipaseaemia in the current study. Lanz et al (2022) measured serum DGGR-lipase activity in a larger number (192) of horses with colic; increased DGGR-lipase activity was demonstrated in 30.2% of horses with signs of colic and was above twice the upper reference limit in 15.6%. These results more closely align with the results of our study, where pancreatic lipase concentration > 2xURL was identified in 13.2% of equids with gastrointestinal disease. The marked differences in the prevalence of hyperlipaseaemia in the study of Johnson et al (2020) with that of Lanz et al (2022) and our own study could represent differences in the populations assessed; in the Johnsonet al (2019) study there were larger numbers of foals (up to a year) and Thoroughbreds in training. Compared with horses with normal pancreatic lipase concentrations, horses with colic with hyperlipaseaemia in the study of Lanz et al (2022) had higher percentages of non-survivors, surgical cases, strangulating diseases, and large bowel displacements or torsion. In the current study we were unable to demonstrate a statistically significant association between pancreatic lipase concentration and outcome or any specific type of disease, apart from identifying significant differences between medical colic and colitis cases, surgical colic and colitis cases, and surgical colic and peritonitis cases. This could also be reflective of a difference in the population of animals studied. In our study, we also found a significant difference in outcome (survival, euthanasia, died) and lipase concentrations < or > 2xURL. One other study (Bartel et al 2023) reported an elevation of pancreatic lipase concentration in one horse with colic (out of a total of 19 horses tested). It is noteworthy that the prevalence of increased DGGR lipase concentrations in horses with signs of colic in all of these studies was higher than the prevalence of primary pancreatitis reported in necropsy studies (Yamout et al 2012; Newman, 2015). This finding suggests that hyperlipaseaemia, as diagnosed by increased DGGR lipase activity, probably results from secondary pancreatic damage in many cases (Lanz et al 2022).
Assessing the relevance of increased pancreatic lipase concentrations in this and other studies is hampered by the lack of histopathological examination of the pancreas; this was not feasible in the current study. The prevalence of pancreatitis, based on the presence of hyperlipaseaemia in this and other studies, is much higher than that identified previously, before the DGGR lipase assay was available. For example, the presence of pancreatitis, based on gross post-mortem and histological examination, was diagnosed in only 4/834 (0.4%) horses necropsied over nine years in one study (Newman 2015), and was diagnosed via necropsy in only 43 cases in a 25-year period at another large teaching hospital (Yamout et al 2012). In agreement with the current and more recent studies where pancreatitis was diagnosed using serum DGGR-lipase activity (Johnson et al 2019; Lanz et al 2020), the study by Yamout et al (2012) showed that pancreatitis occurred frequently in association with gastrointestinal disease, including large bowel torsion, colonic displacements, colonic impaction, enteritis, small intestinal strangulation, etc. Pancreatitis associated with the latter diseases has been reported previously. We also diagnosed pancreatitis in a horse with entrapment of jejunum in the epiploic foramen and it seems likely that trauma to the pancreas might be the cause in such cases, since the pancreas borders the epiploic foramen; an association with this disease has been recorded previously (Johnsonet al 2019). Displacement and torsion of the large colon has also been associated with pancreatitis in several previous studies, possibly due to disruption of the blood supply to the pancreas (Yamout et al 2012) or shock-related hypoperfusion and mechanical compression (Grulke et al 2003).
In dogs, increased pancreatic lipase activity may be also observed in a variety of other non- pancreatic diseases, including portal hypertension, gastrointestinal, renal, and endocrine disorders (Ralliset al 1996; Prümmer et al 2020; Serrano et al 2021); in the present study, increased DGGR-lipase concentration was identified secondary to several other conditions, including non-specific colic, enteritis, grain overload and post-foaling colic. Acute pancreatitis has previously been reported in a donkey with grain overload (Kawaguchi et al 2004).
Spontaneous gastric reflux in association with severely elevated DGGR lipase activity was identified in 3 horses is this study, and gastric reflux on passage of a nasogastric tube was present in a further 8 horses with severe hyperlipaseaemia; in total 11 of 15 (73%) horses with severely raised pancreatic lipase concentration had gastric reflux. This association has been recorded previously (Buote 2003; Kawaguchi et 2004; Waitt et al 2006; Bakos et al 2008; Yamout et al 2012; Newman 2015; Lohmann and Allen 2015). The presence of gastric reflux in a horse showing signs of acute colic should, therefore, raise suspicion of pancreatitis if no other reason for the gastric reflux (such as small intestinal obstruction) can be identified. Likewise, the identification of distended loops of small intestine or thickened small intestinal walls of FLASH scanning could indicate possible pancreatitis if no other cause of these findings can be recognised. Distended loops of small intestine have been identified either on rectal palpation or ultrasonographically in cases of acute pancreatitis previously (Buote 2003; Kawaguchi et al 2004; Bakos et al 2008; Schmidtet al 2010; Yamout et al 2012; Newman 2015; Edery et al 2015; Gomez et al 2015), however this finding is commonly seen in horses with primary diseases causing small intestinal obstruction (Beccati et al 2011).
We were unable to visualise the pancreas on abdominal ultrasonography. Ultrasonographic identification of pancreatic damage, including a mass effect between the liver and right dorsal colon has been previously described in a 24-year-old Warmblood gelding with recurrent colic (Lacket al 2021). However, detailed ultrasonographic examination of this area is not a specific aim of the FLASH procedure which was undertaken in animals in this study.
In most animals which had elevated serum pancreatic lipase concentrations at admission, and where sequential assays were undertaken, the DDGR lipase concentrations reduced as the clinical condition improved. This has been reported previously in two horses with right dorsal displacement of the colon (Johnson et al 2019). However, in the latter study, one of the horses showed a recurrence of pain and a subsequent elevation of plasma DGGR lipase concentration 8 days after the initial episode of colic, suggesting that some horses may be prone to recurrent episodes of pancreatitis, although this might reflect recurrence of the gastrointestinal disease rather than recurrence of primary pancreatitis.
There are several limitations of this study. We were unable to confirm the presence of pancreatitis in any of the horses with elevated serum DGGR lipase concentrations by post-mortem examination or histopathology. This histopathological confirmation has not been reported in any other studies, and requires further investigation. Traditionally, histopathology has been considered to be the gold standard for the diagnosis of pancreatitis and for distinguishing acute pancreatitis from chronic pancreatitis (Cridge et al 2021). However, the histopathological appearance does not always reflect the clinical presentation of affected patients, and reliance on a histopathologic diagnosis for acute pancreatitis is considered to be questionable because lesions can be highly localised and the immediate clinical relevance of lesions is unclear (Mansfield et al 2012). The work of Johnson et al (2019) indicates that DGGR lipase is highly specific for the equine pancreas, and it is widely used as a marker of pancreatitis in other species (O’Brien et al 2014; Cridgeet al 2021). In addition, similar to previous reports of the use of DGGR lipase in equids (Johnson et al 2019; Lanz et al 2020), the sample size for this study was small. Future studies incorporating larger numbers and correlation with specific diagnostic imaging and post-mortem/histopathological findings are warranted. The results of this study should be appraised with reference to the population of animals examined. In this study, all of the animals were privately owned, and most commonly used for pleasure, so the results cannot necessarily be extrapolated to other patient groups such as competition horses and racehorses. A key disadvantage of convenience sampling that was used in this study is that the sample lacks clear generalizability. We were reliant on the availability of surplus blood samples, and there was a likely bias towards sampling surgical colics, colitis cases and peritonitis cases because these cases would be more likely to have surplus blood samples available.
In conclusion, the results of this study indicate that pancreatitis, as evidenced by increased serum DGGR lipase concentrations, is frequently present in horses presenting with signs of colic. In many cases this seems likely to represent a secondary pancreatitis. This observation appears to be more common in certain diseases including colon torsion, colonic displacements, enteritis, entrapment of the small intestine in the epiploic foramen, etc. In addition, horses with colic presenting with gastric reflux (including spontaneous reflux) and horses with ultrasonographic evidence of distended/thickened loops of small intestine may be more likely to have elevated DGGR lipase concentrations. In most cases, the DGGR lipase concentration reduces once the underlying disease resolves. However, in some cases, it is possible that primary acute pancreatitis may cause colic signs, although distinguishing these from secondary pancreatitis associated with gastrointestinal disease may be challenging.