Case selection
This study included canine patients referred to the SNU VMTH, between October 2018 and July 2019, and included normal dogs as controls, and dogs diagnosed with AP as cases. This study was approved by Institutional Animal Care and Use Committee of SNU (SNU-191108-1). Written or verbal informed consent for collecting blood was obtained from the owners and was recorded in the medical chart.
Inclusion criteria
Dogs enrolled in this study were screened for abnormalities during the general health examination (overall physical examination, complete blood count, serum biochemistry, survey thoracic and abdominal radiographs, and abdominal ultrasonography). AP diagnosis was based on the following criteria: 1) presence of one or more of the clinical signs of anorexia/hyporexia, diarrhea, lethargy, vomiting for <7 days; 2) concurrent presence of increased serum cPL level (>200 ng/L, VCheck cPL®, Bionote, Seoul, South Korea) or abnormal SNAP cPL test result (Idexx Laboratories, Milan, Italy); and 3) ultrasonographic evidence of pancreatitis (ProSound Alpha 7, Hitachi, Japan and UGEO H60 Samsung Medison, South Korea). Ultrasonographic findings included enlarged and hypoechoic pancreas, cavitary lesions such as abscess or pseudocyst, dilated pancreatic duct, swollen hypomotile duodenum, biliary dilatation, and peritoneal fluid [20]. Normal dogs that showed no clinical signs and did not show any abnormalities in general examination were recruited into the control group.
Exclusion criteria
Among patients with AP, dogs had undergone treatment with prednisolone for their concurrent diseases were excluded to avoid the effect of external steroids on the results. Additionally, clinical and experimental studies suggested that HMGB1 could majorly influence the pathogenesis of autoimmune diseases [1]. Patients with concurrent diseases such as immune-mediated hemolytic anemia, IMT, protein-losing enteropathy, and systemic lupus erythematosus were excluded to eliminate the possibility of synergism with the autoimmune diseases.
Development of the disease activity index on acute pancreatitis
DAI was designed to clinically quantify AP severity. The scoring system was based on the careful monitoring of patients. To capture the immediate disease activity during diagnosis in non-invasive way, clinical symptoms were grouped into 4 categories: anorexia/hyporexia, diarrhea, lethargy, and vomiting. Symptoms was scored on a scale of 0-2 as follows: 1) Appetite: 0 - no abnormalities, 1 - no enteral food intake for <3 days or decreased appetite <50%, 2 - no enteral food intake or decreased appetite >50% for >3 days; 2) Defecation: 0 - no abnormalities, 1 - Bristol type [21] 5-6, 2 - Bristol type 6-7, or presence of hematochezia or melena; 3) Vitality: 0 - Karnofsky’s performance score [22] grade 0, 1 - Karnofsky’s performance score grade 1, 2 - Karnofsky’s performance score grade 2-3; 4) Vomiting: 0 - no abnormalities, 1 - vomiting or regurgitation <3 days, 2 - vomiting or regurgitation >3 days or >5 times in a day. The scores were summed and the total score was evaluated as the DAI.
Serum collection
Blood was collected by venipuncture method and placed in heparin tubes within 48 hours of diagnosis. After the samples were centrifuged, the collected plasma was stored at -80 ℃ until analysis.
Assay for high-mobility group box chromosomal protein-1 concentrations
The serum HMGB1 concentrations were measured using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Canine High Mobility Group Box-1 Protein ELISA kit; MyBioSource, Inc., San Diego, USA) and following the manufacturer’s instructions, and each sample was tested twice for precision.
Medical records
The following data of all patients was reviewed from the medical records on the electronic charting program (E-friends, pnV, Seoul, Korea): age, breed, sex, clinical symptoms, body condition score [23], results of clinicopathologic evaluation (hematologic and serum biochemical analyses and urinalysis) and radiologic results, costs for hospital stays, length of hospital stay, and survival. The cost of hospitalization was calculated as the total cost of all expenses used during hospitalization, including diagnosis, nursing, medications, supplies, and injections.
Treatment for acute pancreatitis
Each AP patient was treated according to their specific conditions. General treatment principles involved replacing fluid loss, maintaining hydrostatic pressure, controlling nausea, and providing pain relief [24]. Furthermore, the patients were managed under hospitalization for cases that were not permitted to consume fluids orally and had severe dehydration, or high inflammation (white blood cells >15000 /ul, CRP > 35 ng/L) [25]. The patients were discharged when their appetites were restored, vitality had improved, and cPL levels were normalized. The period from the time of admission until discharge was regarded as the length of hospital stay.
Statistical analysis
Statistical analysis was performed using GraphPad Prism 6 (GraphPad Software), R software and SPSS. Shapiro-Wilk test was performed for normality. Student’s t-test was performed if the data passed equal variance test. If equal variance tests failed, Welch’s t-test was used. Differences between groups were assessed for statistical significance by using unpaired t-test. The p-value was calculated by one-tailed testing and considered as a statistically significant difference if p <0.05. After multicollinearity was screened, the correlation analysis and regression analysis were performed between level of serum HMGB1 concentration, patient’s hospitalization period, and hospitalization cost. For regression analysis, it was found that the S curve was suitable as an estimated correlations model using SPSS.