Materials and Methods
Serum samples for pancreatic lipase (DGGR substrate) analysis were
obtained prospectively by convenience sampling of horses and donkeys
presenting to Bell Equine Veterinary Clinic for investigation of
gastrointestinal disease, where surplus blood taken for routine clinical
evaluation of the cases was available. For the purposes of this study,
gastrointestinal disease was defined as equids presenting with clinical
signs of acute colic and/or acute diarrhoea. Serum was harvested from
clotted blood samples collected from eligible equids on admission to the
hospital; in some cases, serum was also collected on consecutive days
when blood was being drawn for ongoing clinical management. Consent was
obtained from the owners for the use of left-over biological material
for research purposes, both by the owners signing the hospital consent
form and verbally. Serum was acquired by centrifugation of clotted blood
samples and submitted to the laboratory within 24 hours of collection.
Ethical approval was provided by the CVS Ethics Committee
(CVS-2022-005).
Where available, results of peripheral haematological and plasma
biochemical analyses (including white blood cell and neutrophil counts,
packed cell volume, total protein and albumin concentrations, and
lactate concentrations) and peritoneal total nucleated cell counts and
lactate concentrations for samples obtained on admission to the hospital
were recorded. The duration of illness prior to admission and severity
of abdominal pain on admission were recorded, as were the heart rate,
results of FLASH abdominal ultrasonography (including presence or
absence of distended loops of small intestine, thickened intestinal
walls, excessive peritoneal fluid, gastric distension, abdominal masses,
intestinal intussusception, evidence of colonic displacement and hepatic
abnormalities), and volume of any gastric reflux (recorded as
spontaneous reflux or reflux obtained on passage of a nasogastric tube).
The severity of colic pain was categorised as mild, moderate, or severe,
based on a behaviour-based simple descriptive scale (Mair and Smith
2005). Any other significant clinical history or physical examination
findings were also recorded. Cases were categorised as medical colic
(i.e. colic that resolved with non-surgical treatment), surgical colic /
euthanased without surgery (surgical cases euthanased without surgery
for financial or welfare reasons), colitis, peritonitis and others. The
diagnosis of colitis was based on the typical clinical, ultrasonographic
and clinical pathological findings including diarrhoea, tachycardia,
fever, leukopenia, hypoproteinemia, etc (Shaw and Stämpfli 2018). The
diagnosis of peritonitis was based on the presence of an elevated total
nucleated cell count and elevated total protein concentration in the
peritoneal fluid, with cut-off values for nucleated cell count
of > 10 × 109/L and total protein
concentration > 25 g/L (Brownlow et al 1981). The
final diagnosis (including non-specific colic) and outcomes (survived,
euthanased or died) were recorded.
Serum pancreatic lipase concentrations were recorded and categorised as
normal (0-20 U/L), mildly elevated (21–49 U/L), moderately elevated
(50– 199 U/L) or markedly elevated (> 200 U/L), as
previously described (Johnson et al 2019). In addition,
pancreatic lipase concentrations that were less or more than twice the
upper reference limit (< or >2 x URL) were also
recorded (as described by Lanz et al 2022). Continuous data were
assessed for normality graphically and using Shapiro-Wilk tests. Median,
interquartile range (IQR) and range were reported for non-normally
distributed data. Categorical data are presented showing the count and
percentage. Non-parametric Wilcoxon rank-sum tests and Kruskal-Wallis
tests were used for continuous variables. Post-hoc analysis following a
statistically significant Kruskal-Wallis was undertaken using a Dunn’s
test. Chi-squared tests were used for categorical variables.
Correlations between two continuous variables were assessed using
Spearman’s rank correlation. Statistical significance was set at p
< 0.05. Data were analysed using Stata 17.0.