Evaluation of asymmetric dimethylarginine as an inflammatory and prognostic marker in dogs with acute pancreatitis.

Abstract Background Asymmetric dimethylarginine (ADMA) has been proposed as a severity marker in humans with acute pancreatitis (AP). Objectives To evaluate ADMA in dogs with AP compared with healthy dogs and its association with severity of disease, mortality, and indicators of the systemic inflammatory response syndrome (SIRS), including serum C‐reactive protein (CRP) concentration, WBC count, and band neutrophils. Animals Fifty‐four dogs with AP and a control group (CG) of 28 healthy dogs. Methods Cohort study including dogs with AP diagnosed using clinical and laboratory variables, abnormal canine pancreatic lipase (cPL) concentration, and compatible abdominal ultrasound examination findings performed within 48 hours of admission. Canine AP severity (CAPS) was calculated. Serum concentration of ADMA was measured using high performance liquid chromatography. Blood donor‐, staff‐, and student‐owned dogs were enrolled in the CG. Results Dogs with AP had higher median admission serum ADMA concentrations compared with the CG (62 versus 48.5 μg/dL; P = .003). Dogs with CAPS ≥11 had higher serum ADMA concentrations than did dogs with CAPS <11 (92 versus 54.6 μg/dL P = .009). Univariable analysis for mortality, CAPS score, band neutrophils, CRP, and ADMA were included in multivariable logistic regression, in which only ADMA was associated with mortality (P = .02). Survivors had a significant decrease in ADMA at first reevaluation compared to admission (P = .02). Conclusions and Clinical Importance Because serum ADMA concentrations were higher in AP dogs compared with the CG, it may have value as a biomarker in the diagnosis of AP in dogs. In addition, because ADMA was associated with mortality, it may have prognostic value.


| INTRODUCTION
Acute pancreatitis (AP) is the most common disease of the exocrine pancreas in dogs. 1 It is characterized by a variety of clinical presenting signs, including hyporexia, weakness, vomiting, diarrhea, and abdominal pain.
Not only the clinical presentation of AP but also the prognosis depends on the severity of inflammation in the pancreatic parenchyma. Parenchymal inflammation is followed by the release of inflammatory mediators, such as reactive oxygen species, reactive nitrogen species (NOS), and cytokines, which may lead to systemic inflammatory response syndrome (SIRS). 2 Specifically, during AP, after the initial activation of pancreatic enzymes, local inflammation causes neutrophil migration, and subsequent production of reactive oxygen species (ROS) and NOS, which contribute to ongoing inflammation. 2 Nitric oxide (NO) is an important molecule that can have regulatory functions in both the circulatory and immune systems. Asymmetric dimethylarginine (ADMA) has been identified as a potent endogenous inhibitor of NO synthesis. 3 Therefore, an increase in serum ADMA concentration may be an indirect and useful indicator of NOS balance and has been shown to be altered in sepsis. [4][5][6][7] To the best of our knowledge, no studies in veterinary medicine have evaluated the relationship between ADMA and AP. However, ADMA has been studied in SIRS and sepsis in humans 4,5 and in cardiovascular diseases in both humans and dogs. [8][9][10] Our hypothesis was that ADMA would be a marker of disease severity and inflammation in dogs with AP. Our aims were to evaluate serum ADMA concentrations in dogs with AP compared with a control group of healthy dogs (CG) and assess possible relationships between ADMA and severity of disease, acute inflammatory markers (C-reactive protein, WBC, and band neutrophils) and survival in dogs with AP. Abdominal ultrasound examination was performed by an experienced radiologist and was considered consistent with AP if the pancreas was hypoechoic and enlarged, with irregular shape and margins, and surrounded by hyperechoic mesentery or abdominal effusion or both. 12 In our facility, serum samples from blood donors, and staff, or student-owned healthy dogs are routinely stored at −80 C for scientific purposes and each owner signs informed consent for such usage. Dogs were considered healthy based on history, physical examination, and blood test results (CBC, serum biochemistry). Twenty-eight healthy dog serum samples were available for the CG for ADMA analysis.

| MATERIALS AND METHODS
Each AP dog had a CBC, serum biochemical profile, coagulation profile, urinalysis, and venous blood gas analysis performed as part of routine care for AP. Data from CBC as well as serum creatinine and ionized calcium concentrations at presentation were recorded.
Canine acute pancreatitis severity (CAPS), a recently validated clinical scoring system for short-term mortality in dogs with AP, 13 was calculated for each dog at presentation (T0). The previously described cutoff of 11 was used to divide the dogs into 2 groups, because it has been shown to be the most sensitive (89%) and specific (90%) cutoff for survival. 13 The CAPS score was calculated as follows: CAPS score = 8 × Ireland Inc., Wicklow, Ireland). The presence of SIRS was assessed using proposed criteria. 14 A dog was assigned to the SIRS group if at least 2 of the following 4 criteria were present: hyperthermia or hypothermia (>39.7 or <37.8 C), tachycardia (>160 beats/min), tachypnea (> 40 breaths/min), WBC < 4000/μL or >12 000/μL or band neutrophils >10%. 13,14 Mortality was evaluated at hospital discharge and dogs were divided into survivors (discharged from the hospital) and nonsurvivors (died or euthanized for the worsened clinical condition despite treatment).
Approximately 2 weeks (10-20 days) after discharge, survivors were clinically reevaluated. A CBC and serum biochemical profile were performed, and surplus serum was stored at −80 C until analysis.
Serum ADMA concentrations were measured both at T0 (hospital admission) and at the first reevaluation (T1). Serum concentrations of ADMA were determined using high performance liquid chromatography with fluorescence detection, as previously described. 11 Briefly, blood was centrifuged immediately after collection, and serum was frozen at −80 C and stored until analysis. For serum ADMA concentration, the intra-and interassay coefficients of variation were <9 and <11%, respectively. The lower limit of quantification, at a signal-tonoise ratio of 10, was 5 μg/dL for ADMA using a 0.2 mL sample volume.

| Statistical analysis
All continuous variables were tested using the Kolmogorov-Smirnov normality test. Normally distributed variables were expressed using the mean ± SD, and nonnormally distributed variables were expressed using the median and range.
The T0 serum ADMA concentration was compared between AP dogs and CG using a Mann-Whitney U test.
A Mann-Whitney U test was performed to determine if T0 ADMA was different between dogs with and without SIRS and between CAPS score groups (< or ≥11). Spearman's or Pearson's correlation test was performed between ADMA and serum C-reactive protein (CRP), WBC, and band neutrophil count.
A univariable analysis was performed to identify variables associated with mortality using t-tests or the Mann-Whitney U test based on data distribution (age, weight, WBC, band neutrophil count, CRP, and T0 ADMA) or chi-square test (CAPS score groups and SIRS). All variables with a P-value <.2 were subjected to multivariable analysis. Afterward, multivariable backward stepwise binary logistic regression was performed to assess the association of variables with mortality, using the variables previously identified in the univariable model as significantly associated with mortality. If a variable remained significantly associated with mortality (P < .05), it was considered to be associated with the death of the dog.

| Serum ADMA concentrations between AP dogs and CG and association between ADMA and CAPS score and SIRS
The AP dogs had significantly higher median serum ADMA concentration compared with the CG (62 μg/dL; interquartile range [IQR], 50.51 versus 48.5 μg/dL; IQR, 39.8; P = .003; Figure 1). Those AP dogs with CAPS ≥11 had higher serum concentrations of ADMA than did dogs with CAPS <11 F I G U R E 1 Box and whisker plot comparing median serum ADMA concentrations between dogs with acute pancreatitis (AP) and healthy dogs (CG). The line represents the median, box depicts the 25th and 75th percentiles, and the whiskers express the range of data. There is considerable overlap in the data between the 2 groups although the median concentration of ADMA of the AP group is significantly higher than CG (Mann-Whitney U test, P = .003) F I G U R E 2 Box and whisker plot comparing serum ADMA concentrations between dogs with acute pancreatitis divided into groups using the previously validated canine acute pancreatitis severity (CAPS) cutoff 12 (CAPS ≥11 and <11). Dogs with CAPS ≥11 showed a significantly higher median ADMA concentration than dogs with CAPS<11 (Mann-Whitney U test, P = .009) (92 μg/dL; IQR, 64.7 versus 54.6 μg/dL; IQR, 36.2; P = .009; Figure 2).
The SIRS was present in 28 dogs (51.8%). No difference in median serum ADMA concentration was found between dogs with and without SIRS (56.3 versus 68.4 μg/dL; P = .24). In addition, no significant correlation was found between serum ADMA concentration and CRP, WBC, or band neutrophil count (P = .35, P = .83, and P = .38, respectively).

| Mortality analysis
The univariable analysis results are shown in Table 1. In the binary logistic regression analysis including band neutrophils, CRP, serum ADMA concentrations, and CAPS score, only serum ADMA concentration was significantly associated with survival (P = .02; Table 2).

| DISCUSSION
In AP in dogs, serum ADMA concentration is associated with the severity of the disease and mortality. The serum ADMA concentrations in dogs with AP and healthy control dogs, respectively, had broad overlap but no healthy animal had serum ADMA concentration that exceeded 100 μg/dL.
Many studies in humans have taken serum ADMA concentration into account, which plays an important regulatory role in the synthesis of NO from L-arginine. 11 Unlike symmetric dimethylarginine (SDMA), which is almost exclusively excreted by the kidneys, ADMA is mainly metabolized by the enzyme dimethylarginine dimethylaminohydrolase in the cytosol of liver and kidney tissue. 4 In humans and experimental animals, ADMA concentration is a wellknown marker of endothelial dysfunction, and increased concentrations are associated with hypercholesterolemia, hypertension, diabetes and other conditions, including SIRS and sepsis. [4][5][6][7]11 Serum ADMA concentration also is a recognized marker of cardiac diseases in humans, and it is significantly increased in humans with valvular disease or idiopathic cardiomyopathy. 11 However, compared to human medicine, in small animals the clinical and prognostic relevance of serum ADMA concentration still is unclear. In 2 studies, serum ADMA concentration was significantly Univariable analysis of selected parameters between survivors (n = 37 dogs) and nonsurvivors (n = 17) Note: Normally distributed variables were expressed using mean ± SD, while the nonnormally distributed variables were expressed using median and range. The variables in bold are those that were entered into the multivariable logistic regression procedure after elimination of variables with a univariable mortality association of P > .20. Abbreviations: ADMA, asymmetric dimethylarginine; CAPS, canine acute pancreatitis severity; CRP, c-reactive protein; SIRS, systemic inflammatory response syndrome; WBC, white blood cells.

T A B L E 2
Results of the multivariable backward stepwise binary logistic regression model for mortality occurring AP in dogs is very often accompanied by or caused by other conditions. However, such comorbidities could not necessarily be ruled out, and they were not taken into account in evaluating mortality.
In conclusion, our study suggests an association between high serum ADMA concentration and death in dogs with AP. Finally, as reported in humans, in our study serum ADMA concentration was significantly decreased over time in dogs that survived.

CONFLICT OF INTEREST DECLARATION
Authors declare no conflict of interest.

OFF-LABEL ANTIMICROBIAL DECLARATION
Authors declare no off-label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION
The study was approved by the Scientific Ethical Committee for Animal Testing of the University of Pisa Approval No. 16749/2017.

HUMAN ETHICS APPROVAL DECLARATION
Authors declare human ethics approval was not needed for this study.