The study was performed at Murdoch University.
Diagnostic accuracy of the SNAP and Spec canine pancreatic lipase tests for pancreatitis in dogs presenting with clinical signs of acute abdominal disease
Article first published online: 16 APR 2014
© Veterinary Emergency and Critical Care Society 2014
Journal of Veterinary Emergency and Critical Care
Volume 24, Issue 2, pages 135–143, March/April 2014
How to Cite
Haworth, M. D., Hosgood, G., Swindells, K. L. and Mansfield, C. S. (2014), Diagnostic accuracy of the SNAP and Spec canine pancreatic lipase tests for pancreatitis in dogs presenting with clinical signs of acute abdominal disease. Journal of Veterinary Emergency and Critical Care, 24: 135–143. doi: 10.1111/vec.12158
The authors declare no conflict of interests.
Presented as an abstract at the Australian College of Veterinary Scientists College Science Week, Gold Coast Australia, July 2011.
- Issue published online: 16 APR 2014
- Article first published online: 16 APR 2014
- Manuscript Accepted: 26 DEC 2013
- Manuscript Received: 6 DEC 2011
- acute abdomen;
- point-of-care diagnostics;
- pancreatic disease
- Top of page
- Materials and Methods
To (i) assess the clinical diagnostic accuracy of SNAP canine pancreatic lipase (cPL) and specific canine pancreatic lipase (Spec cPL) and (ii) assess the agreement of an abnormal test result between SNAP cPL and Spec cPL in dogs presenting with acute abdominal disease.
Prospective observational cohort study.
University teaching hospital emergency center.
Thirty-eight client-owned dogs that presented with acute abdominal disease, with a known final diagnosis between March 2009 and April 2010. Dogs were retrospectively assigned into 2 groups, dogs with acute pancreatitis (AP) (Group 1) and dogs without AP (Group 2).
Paired serum samples obtained within 24 hours of presentation were analyzed using the SNAP cPL test and Spec cPL assay.
Measurements and Results
SNAP cPL clinical sensitivity and specificity was 82% (9/11 dogs of group 1) and 59% (16/27 dogs of group 2), respectively. Spec cPL clinical sensitivity and specificity was 70% (7/10 dogs of group 1) and 77% (20/26 dogs of group 2), respectively. Accuracy of the SNAP and Spec cPL for a clinical diagnosis of pancreatitis was found to be 66% and 75%, respectively. Agreement between a positive SNAP (cPL ≥ 200 μg/L) and a clinical diagnosis pancreatitis resulted in κ = 0.33. Agreement between an increased Spec (cPL ≥ 400 μg/L) and a clinical diagnosis of pancreatitis resulted in a κ = 0.43. The agreement between SNAP and Spec cPL (cPL ≥ 200 μg/L) for the entire cohort resulted in κ = 0.78.
SNAP cPL and Spec cPL results may provide a "false positive" diagnosis of pancreatitis in up to 40% of dogs presenting with acute abdominal disease. There is good overall agreement between SNAP cPL and Spec cPL; however, there were 4/38 dogs with positive SNAP cPL and "normal" Spec cPL.
canine pancreatic lipase
canine pancreatic lipase immunoreactivity
- Spec cPL
specific canine pancreatic lipase
- Top of page
- Materials and Methods
Acute pancreatitis (AP) is an important disease of dogs, with variable and nonspecific clinical signs such as abdominal pain, vomiting, and diarrhea.[1, 2] These clinical signs are also present in conditions such as septic peritonitis or intestinal obstruction, which require specific and timely interventional treatment. Traditional diagnostic methodologies, such as total serum lipase and amylase, have poor sensitivities and specificities for the diagnosis of AP in dogs.[3-5]
The canine pancreatic lipase test measures lipase of pancreatic origin, and theoretically should only be increased during times of pancreatic inflammation. The canine pancreatic lipase immunoreactivity (cPLI) assay (first a radioimmunoassay, and then subsequently an enzyme immunoassay) has been validated in dogs.[6, 7] The cPLI assay was then developed into a commercially available specific canine pancreatic lipase (Spec cPL) assay, using a recombinant peptide as the antigen and dual monoclonal antibodies for capture and detection. Spec cPL shows good correlation to and high reproducibility with cPLI. Spec cPL results < 200 μg/L are considered to be consistent with an absence of pancreatic inflammation,[8, 9] while results ≥ 400 μg/L are considered consistent with a diagnosis of pancreatitis, and a result from 200–399 μg/L is considered equivocal.[9, 10] A rapid point-of-care semiquantitative assay (SNAP cPL) has also been developed using the same dual monoclonal antibodies for capture and detection of pancreatic lipase as Spec cPL. SNAP cPL is reported to show good correlation and reproducibility compared to the laboratory-based Spec cPL. A negative SNAP cPL result corresponds to a Spec cPL concentration < 200 μg/L, and a positive result with a concentration ≥ 200 μg/L.
The reported sensitivity for cPLI/Spec cPL ranges from 21–82%,[4, 5, 9, 11, 13] while specificity for Spec cPL is reported to range from 86–100%.[5, 9, 12, 13] All but 1 of these studies based the diagnosis of AP on histologic demonstration of pancreatic inflammation. As a result, the sensitivities and specificities may not be accurate as pancreatic inflammation was often very low or mild. The sensitivity of Spec cPL (or cPLI) has been shown to be higher in dogs with increasing histologic severity.[4, 5, 13] A recent study reported the sensitivity and specificity of the SNAP cPL to be 94 and 77%, respectively, in dogs that presented both with suspicion and without suspicion of pancreatitis.
The true diagnostic accuracy of noninvasive methodologies for determining the presence of AP is unknown due to the difficulty in obtaining a gold standard. Abdominal ultrasonography is used extensively in veterinary practice. The main finding with this diagnostic modality in AP is peri-pancreatic hyperechogenicity indicative of peri-pancreatic fat necrosis in the acute necrotizing form. Pancreatic inflammation may also develop due to duodenal reflux, ischemia, or generalized peritonitis in association with other diseases such as septic peritonitis, abdominal hemorrhage, or intestinal foreign bodies. Therefore, despite the presence of histological and ultrasonographic severe pancreatic inflammation, pancreatitis may only be secondary and not be the cause of the clinical presentation in dogs. Sensitivity of ultrasound may be influenced by animal-related factors (eg, obesity, presence of ingesta interfering with image quality), operator-related factors (eg, level of experience of operator), and technical factors (eg, quality of ultrasound equipment used). Reported sensitivities for ultrasonography for detection of AP range between 66–68%.[1, 4] Higher median histologic grading of pancreatic inflammation has been reported to correlate with ultrasonographic evidence of pancreatitis. In another recent study, a small subgroup of dogs with histologically confirmed pancreatitis of varying severity all had ultrasonographic evidence of pancreatitis. Therefore, history, clinical signs, laboratory testing, and abdominal imaging are often used together to make a clinical diagnosis of AP. Conversely, it is also possible that by relying on histologic evaluation of the pancreas, studies may be biased toward dogs with more severe disease (and a fatal outcome). Additionally, pancreatic biopsies are seldom obtained in critically ill dogs unless at postmortem, and inflammation may be unevenly distributed throughout the pancreas, or just be located in the peri-pancreatic fat.[13, 14]
The primary objective of this study was to determine the accuracy, sensitivity, and specificity of SNAP cPL and Spec cPL in dogs presenting with acute abdominal disease to an emergency center. A secondary objective of this study was to quantify the agreement between the Spec cPL and the SNAP cPL for paired serum samples taken from the same cohort of dogs.
Materials and Methods
- Top of page
- Materials and Methods
Client-owned dogs presenting to a first-opinion and referral emergency center at a university teaching hospital between March 2009 and April 2010 were recruited. Dogs were initially included if they had ≥ 2 of the following clinical signs: acute (< 2 days) onset of abdominal pain, vomiting, abdominal distension, or diarrhea. Dogs were excluded from the analysis if they did not have a definitive diagnosis made during hospitalization.
Blood was collected via jugular, cephalic, or saphenous venipuncture from all dogs within 24 hours of admission as part of diagnostic investigation. If additional samples of blood were collected specifically for the purposes of the study, signed owner consent was obtained. The study was approved by the institutional Animal Ethics Committee, fulfilling National Health, and Medical Research Council regulations. Two milliliters of blood was collected into plain serum tubes initially, centrifuged at 3,120 xg for 10 minutes, and allowed to equilibrate to room temperature prior to serum collection.
The storage of the SNAP cPL kits,1 sample handling and testing procedure was according to manufacturer's instructions.2 Testing of SNAP cPL was either performed at the time of collection and the remaining serum frozen at minus 20°C, or the serum sample was kept refrigerated for < 1 week and then allowed to equilibrate to room temperature before being tested and then frozen. One of 2 qualified veterinary nurses and one of the authors (MH) performed all of the in-house point-of-care SNAP cPL tests and were unaware of the final diagnoses at the time. Results of the SNAP and Spec cPL tests were not paired with individual dogs at the time a diagnosis was assigned, and the test result was not revealed to veterinarians in charge of the clinical case. If the person performing the in-house test was unsure of the result due to the test and reference spot intensity similarities, the test was repeated. If similar results were obtained a second time, the result was recorded as an abnormal result. A later batch analysis of Spec cPL3 concentration was performed on the frozen serum that had been stored up to 18 months. This was shipped overnight, refrigerated, to a regional laboratoryc for analysis.
The SNAP cPL was recorded as either visually normal or abnormal, where abnormal corresponds to a cPL ≥ 200 μg/L. The agreement between a clinical diagnosis of pancreatitis and a visually abnormal SNAP cPL was quantified by the kappa (κ) coefficient. Further, the agreement between a clinical diagnosis of pancreatitis and a Spec cPL ≥ 400 μg/L was also quantified by the kappa (κ) coefficient. A Spec cPL ≥ 400 is considered consistent with pancreatitis.[5, 9, 10]
A cut-off concentration of cPL ≥ 200 μg/L, as measured by Spec cPL, was used for the agreement between the SNAP cPL and Spec cPL for all dogs. Spec cPL above or below this concentration, with visually abnormal or normal SNAP cPL, respectively, was considered necessary for agreement between the 2 tests.
All agreements were made using McNemar's test, and quantified by the kappa (κ) coefficient.4 Results for the Spec cPL were between the values of 30 μg/L and 1000 μg/L, which represents the limits of the range reported by the laboratory performing the assays.c Reported results of ≤ 30 μg/L or ≥ 1000 μg/L were calculated as 30 μg/L or 1000 μg/L, respectively, for all analyses.
The diagnosis of AP in dogs was achieved by taking into consideration the history, physical exam findings, and the results of complete blood count, biochemical analysis, and abdominal ultrasonography performed by a veterinary radiologist. Diagnostics for dogs without AP were performed as indicated for each individual dog to enable reaching a diagnosis. A diagnosis for these dogs was also reached based on the history, physical exam findings, complete blood count, serum biochemical analysis, abdominal ultrasound (by either a veterinary radiologist or emergency clinician), blood gases and electrolytes, thoracic and abdominal radiography and computed tomography, echocardiography, coagulation assessment, body fluid analysis and bacterial culture and sensitivity, cytology, biopsy, histopathology, immunohistochemistry, and surgery.
The results of the diagnostic workup were reviewed by 3 of the authors (2 board certified in emergency and critical care and 1 board certified in internal medicine) upon completion of the study to determine the definitive diagnosis. The authors were blinded to the results of the SNAP and Spec cPL tests at that time. Dogs were diagnosed with AP if they had ultrasonographic or histologic support for pancreatic inflammation or necrosis with no other identifiable disease. Supportive ultrasonographic evidence of AP was defined and reported by the veterinary radiologist to include the presence of an enlarged, hypoechoic pancreatic tissue surrounded by hyperechoic peripancreatic mesentery, with or without peritoneal effusion, biliary duct dilatation and corrugation or thickening of the duodenal wall.[1, 15] Additionally, in order to be given a final diagnosis of clinical AP, a minimum of 6 months follow-up was required to ensure exocrine pancreatic neoplasia was unlikely.
For analysis, dogs were allocated to 1 of 2 groups, based on the above criteria:
- Group 1: Dogs with AP as their primary disease.
- Group 2: Dogs with confirmed disease other than AP. These dogs may have had pancreatic inflammation but was considered inconsequential and not the primary cause of their clinical presentation.
- Top of page
- Materials and Methods
Samples were collected from 64 client-owned dogs, with 26 dogs excluded as no definitive diagnosis could be determined, leaving 38 dogs for analysis (Figure 1). Vomiting was present in 28 (74%), diarrhea in 8 (21%), abdominal pain in 33 (87%), and abdominal distension in 7 dogs (18%) (Table 1). No sample on testing had an indeterminate SNAP cPL result. Serum was available for Spec cPL measurement in 36 of 38 dogs. Twenty-nine (76%) of these samples were either hemolyzed (n = 25/38 [66%]), icteric (n = 3/38 [8%]), or lipemic (n = 6/38 [16%]), or had a combination of these characteristics (n = 6/38 [16%]).
Eleven dogs were diagnosed with AP (Group 1). Breeds in this group included Australian Cattle Dog (n = 2), Fox Terrier (n = 1), Border Collie (n = 1), Jack Russell Terrier (n = 1), Labrador (n = 1), Maltese cross (n = 1), Akita (n = 1), Siberian Husky (n = 1), Miniature Schnauzer (n = 1), and Cocker Spaniel cross (n = 1). Ages ranged from 1.5 to 13 years (median 9 years, mean 8 years). There were 7 females (6 neutered), and 4 males (3 neutered). Abdominal ultrasonography was consistent with AP in all 11 dogs, and no dog underwent surgery. Three dogs were euthanized (postmortem confirmed diagnosis in 1; no postmortem examination was permitted in 2), with no clinical recurrence in the surviving 8 dogs at 6 months follow-up. No dog with AP had azotemia. Abdominal effusions were noted in 5 (45%) of these dogs, but no abdominal fluid was collected for analysis.
Primary disease other than AP was diagnosed in 27 dogs (Group 2). Breeds included Labrador (n = 3), Siberian Husky (n = 2), Border Collie (n = 2), Rottweiler (n = 2), German Shepherd (n = 2), and the remaining 16 dogs were represented by single or mixed breeds. Ages ranged from 17 weeks to 15 years (median 10 years, mean 9 years). There were 9 neutered females, 3 entire females, 12 neutered males, and 3 entire males. Fifteen of these dogs were euthanized, and 1 dog died.
Five of the dogs from Group 2 had full post mortem performed, which confirmed the absence of pancreatic or peri-pancreatic inflammation or necrosis. Diagnoses included anaplastic large T-cell lymphoma of the liver (n = 1); small intestinal infarction with bilateral adrenomegaly (n = 1); pancreatic islet cell carcinoma with erosive enterocolitis (n = 1); pancreatic islet cell carcinoma with hepatic metastasis (n = 1); pancreatic carcinoma with hepatic, duodenal, lymph node, and lung metastasis (n = 1). Of the 3 dogs with pancreatic islet cell carcinoma, histology did not identify inflammation or necrosis associated with pancreatic tissue, although in 1 there was virtually no recognizable pancreatic tissue present. None of the dogs with pancreatic carcinoma tested positive with either SNAP cPL or Spec cPL. Both remaining dogs had abnormal SNAP cPL tests but only the dog with small intestinal infarction had an increased Spec cPL result consistent with pancreatitis.
|Group 1 (pancreatitis)||Group 2 (not pancreatitis)|
|SNAP cPL abnormal||SNAP cPL normal||SNAP cPL abnormal||SNAP cPL normal|
|Parameter||(n = 9)||(n = 2)||(n = 11)||(n = 16)|
|Increased urea concentration||1||0||2||2|
|Septic or inflammatory||0||0||6||3|
The 22 dogs that did not have postmortem examination in Group 2 were diagnosed with small intestinal foreign bodies (n = 6), hemoperitoneum due to splenic or concurrent hepatic masses (n = 3), pyometra (n = 2), hepatic abscessation (n = 2), emphysematous cholecystitis (n = 2), abdominal mass and concurrent septic peritonitis (n = 1), large solitary hepatic mass invading the caudal vena cava (n = 1), septic peritonitis due to a ruptured jejunal mass (n = 1), septic peritonitis due to intestinal foreign body (n = 1), prostatic abscessation (n = 1), pericardial effusion (n = 1), and hepatic lymphoma (n = 1).
Abdominal ultrasound by a veterinary radiologist was performed in 14 dogs in Group 2. Abdominal surgery was performed in 14 dogs in Group 2, of which 8 dogs did not undergo abdominal ultrasonography prior. One dog had abdominal and thoracic computed tomographic examination. No surgery or advanced abdominal imaging was performed in 4 dogs of group 2. A final diagnosis of hemoperitoneum was confirmed in 3 of the 4 dogs by abdominocentesis. All 3 had large abdominal masses on ultrasonography (performed by emergency resident), and were reported as splenic (n = 2) or hepatic (n = 1). The dog identified as having a hepatic mass also displayed a septic component as evidenced by intracellular bacteria by cytology. The remaining dog had a postmortem only. The final diagnosis was pancreatic carcinoma with hepatic metastasis.
In the 11 dogs of group 1, 9 (82%) tested positive with SNAP cPL and 2 (18%) tested negative (Table 2). Pancreatic lipase was measured in 8 of 9 positive SNAP dogs, and 100% (8/8) had Spec cPL concentrations ≥ 200 μg/L (range 320–1,000 μg/L; median 800 μg/L, mean 748 μg/L). There was insufficient serum for Spec cPL testing in the remaining dog. The 2 dogs with negative SNAP cPL results both had Spec cPL concentrations of 30 μg/L.
|Spec-cPL result (μg/L) n = 11|
|Dog||SNAP cPL positive||Dog||SNAP cPL negative|
In the 27 dogs in Group 2, 11 (41%) tested positive with SNAP cPL and 16 (59%) tested negative (Table 3). Pancreatic lipase was measured in 10 of 11 positive SNAP dogs, and 6 had Spec cPL concentrations ≥ 400 μg/L, and 4 had Spec cPL < 200 μg/L. There was insufficient serum for Spec cPL testing in one SNAP positive dog in Group 2. The remaining 16 dogs in Group 2 with a negative SNAP cPL all had Spec cPL concentration < 200 μg/L (median 30 μg/L, mean 51 μg/L, range 30–121 μg/L). Eight of the 11 dogs (73%) with a positive SNAP cPL in Group 2 had abdominal effusions (septic/inflammatory in 6). In the 16 dogs with a negative SNAP cPL in Group 2, 8 (50%) had abdominal effusions (with 3 being septic/inflammatory). Therefore, 9/27 (33%) of dogs in Group 2 had septic or inflammatory abdominal effusions, with 6 of these 9 having a positive SNAP cPL. Of the 4 dogs with positive SNAP cPL and normal Spec cPL results, 2 (50%) had septic or inflammatory abdominal effusions.
|SNAP cPL positive||SNAP cPL negative|
|Dog||Diagnosis||Spec cPL (μg/L)||Dog||Diagnosis||Spec cPL (μg/L)|
|1||Small intestinal foreign body||Insufficient sample||12||Small intestinal foreign body||30|
|2||Small intestinal foreign body||30||13||Small intestinal foreign body||30|
|3||Small intestinal foreign body and septic peritonitis||105||14||Small intestinal foreign body||30|
|4||Small intestinal infarction with bilateral adrenomegalya||568a||15||Small intestinal foreign body||78|
|5||Hepatic T-cell lymphomaa||68a||16||Jejunal mass/septic peritonitis||121|
|6||Hepatic mass with invasion of the caudal vena cava||404||17||Hepatic lymphoma||30|
|7||Hepatic masses/septic peritonitis||672||18||Hemoperitoneum/splenic mass||30|
|8||Hepatic/splenic masses with hemoperitoneum||720||19||Hemoperitoneum/splenic mass||30|
|9||Hepatic abscess||1000||20||Emphysematous cholecystitis||30|
|10||Hemoperitoneum/septic peritonitis||550||21||Emphysematous cholecystitis||30|
|11||Pyometra and septic peritonitis||30||22||Pyometra||30|
|26||Pancreatic carcinoma with ulcerative enterocolitisa||83a|
|27||Pancreatic and hepatic carcinoma with hemoperitoneuma||121a|
Azotemia was present in 3 dogs in Group 2, with creatinine ranging from 220 to 323 μmol/L (ref: 44–159 μmol/L) and urea ranging from 10 to 28.2 mmol/L (ref: 2.5–9.6 mmol/L). These 3 dogs all had positive SNAP cPL results, with 2/3 also having Spec cPL concentration ≥ 200 μg/L. Urea alone was increased in 4 dogs in this group ranging from 10.1 to 26.9 mmol/L (ref: 2.5–9.6 mmol/L). Only 2 of these dogs (50%) had a positive SNAP cPL test, and none had Spec cPL ≥ 200 μg/L.
The clinical sensitivity and specificity for SNAP cPL was 82% (9/11 dogs of group 1) and 59% (16/27 dogs of group 2), respectively. The clinical sensitivity and specificity for Spec cPL was 70% (7/10 dogs of Group 1) and 77% (20/26 dogs of Group 2), respectively. Accuracy of the SNAP and Spec cPL with a clinical diagnosis of AP was 66 and 75%, respectively.
The agreement of SNAP cPL with a clinical diagnosis of primary AP in all dogs (Table 4) resulted in a κ of 0.33 (95% CI: 0.06–0.61). Agreement was also calculated to assume the 2 dogs in Group 1 that tested normal on SNAP cPL were falsely diagnosed with pancreatitis. If these dogs were moved to Group 2 for analysis, agreement of SNAP cPL and a clinical diagnosis of AP would result in a κ of 0.44 for all dogs (95% CI: 0.20–0.67). The agreement of Spec cPL with a clinical diagnosis of primary AP (Table 5) in all dogs resulted in a κ of 0.43 (95% CI: 0.12–0.74).
|Group 1||Group 2|
|Group 1||Group 2|
|Spec cPL ≥ 400 μg/L||7||6|
|Spec cPL < 400 μg/L||3||20|
The agreement between SNAP cPL and Spec cPL for all dogs (Table 6) resulted in a κ of 0.78 (95% CI: 0.59–0.98). The agreement between SNAP cPL and Spec cPL concentration for dogs with disease of nonpancreatic origin resulted in κ 0.65 (95% CI: 0.35–0.94). The agreement between SNAP cPL with Spec cPL concentrations in dogs with primary AP resulted in κ = 1.0.
|Spec cPL ≥ 200 μg/L||Spec cPL < 200 μg/L|
- Top of page
- Materials and Methods
This study suggests that SNAP and Spec cPL tests have poor agreement with a clinical diagnosis of primary AP in dogs presenting with compatible historical and clinical signs. This was predominantly due to the tests yielding a large number of clinically relevant false positives (SNAP cPL: 11/27 dogs or 41%, Spec cPL: 6/26 dogs or 23%).
Sixty percent of SNAP positive dogs without AP had Spec cPL concentrations greater than 400 μg/L, with no dogs having a value between 200–400 μg/L. In this study, having a second reference spot in the SNAP cPL test to indicate concentrations above 400 μg/L would not have improved specificity. The specificity of Spec cPL was greater than SNAP cPL. This was because 4 dogs without AP had visually abnormal SNAP cPLs, but Spec cPL concentrations well below 200 μg/L. The specificity of SNAP cPL and Spec cPL in this study is lower than previously reported.[5, 9, 12, 13] The authors feel this is probably due to a population of exclusively sick dogs with similar clinical presentations being tested, without reliance on histologic diagnosis and a known final diagnosis.
The sensitivity of the Spec cPL is consistent with that reported previously,[4, 5, 11, 13] although that of the SNAP cPL was lower. There were only a small number of false negatives (2/11 dogs or 18%) with the SNAP cPL. The sensitivity of Spec cPL was lower than SNAP cPL due to 1 dog with a cPL concentration of 320 μg/L being below the diagnostic cut-off of 400 μg/L.
There was good agreement between SNAP cPL and Spec cPL results. This was greater in dogs with a clinical diagnosis of primary pancreatitis than in those without. Further, all dogs testing normal on SNAP cPL had Spec cPL concentrations below 200 μg/L (18/18 dogs or 100%).
There are several possible explanations for the 11/27 dogs without AP that tested positive with SNAP cPL. One possible explanation is that pancreatic inflammation may develop due to diffuse abdominal inflammation, as found in dogs with septic peritonitis. Additionally, any condition that causes hypoperfusion of the pancreas, or ischemia and reperfusion of the splanchnic circulation may cause pancreatic inflammation, as the pancreas is exquisitely sensitive to disturbances of microcirculation. Increased total serum lipase activity has been reported in dogs with duodenal foreign bodies and acute gastroenteritis. This may potentially be due to production of lipase by organs other than the pancreas, or due to duodenal reflux causing subclinical pancreatitis. In studies that have assessed specificity of cPL, diagnosis was based on postmortem analysis from referral centers[5, 12] and dogs with intestinal foreign bodies were not included in the sample populations. Therefore it is remains unclear whether cPL concentrations are increased in dogs with duodenal foreign bodies. However, none of the dogs with intestinal foreign bodies in this study had increased Spec cPL, and only 2 were reported in the duodenum. It is possible that other isoforms of lipase were being measured. This is considered unlikely as pancreatic lipase has been localized to the pancreas in immunohistochemical studies, and is too low to be quantified in dogs with exocrine pancreatic insufficiency.
Dogs with decreased renal function have been shown to have increased serum total lipase activities.[3, 21] One study has shown that Spec cPL is not increased in dogs with experimentally induced chronic kidney failure, but this has not been verified in dogs where there may be a naturally occurring acute decline in glomerular filtration. In the study cohort reported here there were 3 dogs with azotemia, although none had anuric kidney failure. These 3 dogs were not diagnosed with AP, but all had positive SNAP cPL results, and 2 also had Spec cPL concentrations ≥ 200 μg/L. The clinical diagnoses in these 3 azotemic dogs were splenic and hepatic masses with hemoperitoneum, septic peritonitis secondary to a perforating intestinal foreign body, and small intestinal thrombosis. The dog with intestinal thrombosis underwent postmortem examination and no histological evidence of pancreatitis was noted. The remaining 2 dogs underwent abdominal surgery with no gross evidence of pancreatitis recorded in the surgical reports. However, histological examination of the pancreas was not performed and therefore concurrent microscopic pancreatitis could not be ruled out. Additional studies are required to further elucidate the role of azotemia on cPL concentrations, particularly in acute disease.
All 3 dogs with pancreatic carcinoma had negative SNAP cPL and Spec cPL results. This may be due to very little associated inflammation, as documented in 2 dogs, or a lack of functional pancreatic tissue, as observed in the remaining dog.
Four dogs in this study had positive SNAP cPL results but Spec cPL concentrations < 200 μg/L. Operator error in interpreting the SNAP cPL results is a possible explanation for this discrepancy, but is considered unlikely. All 3 individuals interpreting the test were trained personnel, and made the observations after performing the test strictly according to the manufacturer guidelines. The manufacturer reports 96–100% agreement between SNAP cPL and Spec cPL for normal samples (cPL < 200 μg/L) and 88–92% agreement for abnormal samples (cPL ≥200 μg/L). It is also reported that visual discrepancy occurs mostly at Spec cPL concentrations around 200 μg/L. For the 4 discrepant results, SNAP cPL was abnormal, but all had Spec cPL concentrations < 105 μg/L. This makes visual discrepancy unlikely. Prolonged storage of serum samples prior to measurement of Spec cPL may also be a contributing factor, as all 4 of these samples were frozen for greater than 6 months. The stability of Spec cPL has been reported to be unchanged after 21 days at room temperature, refrigerated, at –20°C, and at –80°C. Additionally, a study evaluating lipase activity and Spec cPL in dogs with experimentally induced chronic renal failure utilized stored samples that were more than 20 years old, and demonstrated significantly elevated pancreatic lipase concentrations in one dog. Instability of canine pancreatic lipase in serum frozen at –20°C is therefore thought to be an unlikely cause of the discrepancies in this study.
Hemolysis, icterus, and lipemia were frequently present in the samples in this study, but these factors have been shown not to interfere with the visual interpretation of SNAP cPL tests or with measurement of Spec cPL concentrations. It is theoretically possible that an unknown protein was present in the serum of these 4 dogs that caused interference. However, given that both SNAP cPL and Spec cPL utilize the same dual monoclonal antibodies, abnormal results would be expected for both tests if cross-reactivity was present. However, the stability of a potential cross-reacting inflammatory protein may not be as long lived during storage as cPL, and not be detectable at the time of Spec cPL measurement.
One limitation of this study is the use of ultrasonography alone to diagnose AP. The diagnosis of AP by ultrasound has been reported to have a sensitivity ranging from 66 to 68%.[1, 4] In these studies, there may have been primary disease other than AP as no final diagnosis was discussed, and mild or chronic forms of pancreatitis could also have been present, reducing the sensitivity of ultrasound. The other study evaluating ultrasonographic detection of pancreatitis was performed between 1986 and 1995, commencing well over a 20 years ago. The authors believe the diagnostic sensitivity of ultrasound is likely to be much higher now than in the earlier studies due to improved equipment and operator expertise. Additionally, the specificity of ultrasound for the diagnosis of AP due to the presence of hyperechogencity associated with peri-pancreatic fat necrosis is well accepted. Therefore, the authors feel that the number of false positives in dogs diagnosed with primary AP were negligible. A further limitation was the absence of ultrasonographic evaluation of the pancreas in nearly half of the dogs of group 2. This may have enabled comparison between the positive tests of group 2 and ultrasonographic findings. All dogs in the AP group were treated for AP, and had no recurrence of clinical signs within 6 months of discharge for all survivors, making concurrent pancreatic neoplasia or other abdominal disease such as septic peritonitis unlikely. Of the 3 dogs in this group that were euthanized, consent for postmortem examination was only obtained for 1 dog and declined in the other 2. The postmortem examination confirmed the presence of severe pancreatic inflammation and necrosis. To determine if a false positive diagnosis of AP may have influenced this agreement, dogs in group 1 that had a negative SNAP cPL/Spec cPL test were moved to group 2 for analysis. Analysis of agreement for a clinical diagnosis of pancreatitis then gave a κ of 0.44, which still represents poor agreement.
A further limitation of this study was the small number of dogs analyzed. Many dogs were excluded due to an absence of a definitive diagnosis. Therefore, bias toward severe disease may have arisen as dogs with mild pancreatitis may not have had a diagnosis made after full diagnostic work-up. It is unknown what effect inclusion of those dogs would have on the analysis of diagnostic accuracy, sensitivity, or specificity.
The manufacturer currently recommends performing a quantitative Spec cPL assay following SNAP cPL at initial presentation. Although there was a good concordance between SNAP cPL and Spec cPL concentration overall in this study, there was some discordance in dogs without primary AP. However, the use of additional testing such as abdominal imaging, along with stringent assessment of clinical and historical findings to make a diagnosis of acute pancreatitis may preclude the necessity of follow-up testing. Currently, there are no published data that show changes in serum concentrations of pancreatic enzymes to correspond to clinical improvement, or to guide treatment regimes.
- Top of page
- Materials and Methods
This study indicated a poorer specificity of cPL for diagnosing AP than previously reported, although sensitivity was similar. There was reasonable agreement between SNAP cPL and Spec cPL results. Measurement of Spec cPL had a better agreement than SNAP cPL for a clinical diagnosis of AP, but overall both produced poor agreement. A positive SNAP cPL or Spec cPL may be indicative of pancreatic inflammation, however this cannot readily determine the primary reason for clinical presentation. Conversely, a negative SNAP cPL or Spec cPL < 200 μg/L appears to be moderately specific, with a small number of dogs (2/11; 18%) diagnosed with AP having false negative results.
- Top of page
- Materials and Methods
The authors would like to thank all clinical staff at Murdoch University Veterinary Hospital who assisted in management and recruitment of cases, and the staff of the Anatomical Pathology Department at Murdoch University in the collection, preparation, and handling of samples, particularly Dr Mandy O'Hara, BSc Hons, BVMS, MACVSc, DACVP, and Louise FitzGerald, BSc(Vet)(Hons), BVSc(Hons), MANZCVSC.
SNAP cPL Test Kit, Idexx Laboratories Inc., Westbrook, ME.
Packet Insert, Idexx Laboratories Inc.
Spec cPL ELISA, Idexx Laboratories, Brisbane, Queensland, Australia.
PROC FREQ, SAS v9.1, SAS Institute, Cary, NC.
- Top of page
- Materials and Methods
- 4Sensitivity of serum markers for pancreatitis in dogs with macroscopic evidence of pancreatitis. Vet Ther 2008; 9:263–273., , , et al.
- 5Sensitivity and specificity of canine pancreas-specific lipase (cPL) and other markers for pancreatitis in 70 dogs with and without histopathological evidence of pancreatitis. J Vet Intern Med 2011; 25:1241–1247., , , et al.
- 10Performance validation and method comparison of an in-clinic enzyme-linked immunosorbent assay for the detection of canine pancreatic lipase. J Vet Diagn Invest 2011; 23:115–119., , , et al.
- 11Serum canine pancreatic lipase immunoreactivity (cPLI) concentrations in dogs with spontaneous pancreatitis [abstract]. J Vet Intern Med 2001; 15:274., , , et al.
- 17Duodenal foreign body mimicking acute pancreatitis. J Am Anim Hosp Assoc 1993; 29:351–354., , .
- 22Serum lipase activity and canine pancreatic lipase immunoreactivity (cPLI) concentration in dogs with experimentally induced chronic renal failure. Vet Res 2010; 3:58–63., , .