Clinical manifestations of chronic pancreatitis in English cocker spaniels

Abstract Background Chronic pancreatitis (CP) is common in English cocker spaniels (ECS). It is histologically similar to IgG4‐related disease (IgG4‐RD) in humans and is characterized by duct destruction, interlobular fibrosis, and dense periductular and perivenous lymphocytic aggregates. However, the clinical manifestations of CP in ECS have not been previously described. Objectives Characterize the clinical manifestations of CP in a group of ECS, including similarities and differences to IgG4‐RD in humans. Animals One‐hundred four ECS with CP and 44 client owned control ECS without CP (both healthy and diseased controls). Methods Affected dogs were divided into 2 groups according to the methods used to diagnose CP. Case records were searched for signalment, clinical, and clinicopathological findings, and evidence of keratoconjunctivitis sicca (KCS), proteinuria, other immune‐mediated diseases, and anal sacculitis. Results Involvement of other organs was common. Affected ECS presented with a high frequency of KCS (n = 49), proteinuria (n = 47), anal gland disease (n = 36), atopy (n = 21), and other immune‐mediated diseases (n = 16). Those with parti‐color hair coats, particularly blue roan, had a strong association with CP, suggesting a link between coat color and autoimmune conditions in this breed. Conclusions and Clinical Importance English cocker spaniels with CP show clinical similarities to humans with IgG4‐RD and common involvement of other organs. Clinicians should evaluate affected Cocker Spaniels for proteinuria, keratoconjunctivitis sicca, and other potential immune‐mediated diseases.

aggregates with a predominance of IgG4-positive plasma cells on immunohistochemistry in numerous organs including pancreas, kidney, salivary glands and orbital tissue. 1,2Although not conclusively diagnostic of autoimmune disease, these distinctive features resemble those of autoimmune pancreatitis type 1 in humans, which is part of a steroid-responsive, fibro-inflammatory, multisystemic syndrome, currently known as IgG4-related disease (IgG4-RD). 3IgG4-RD in humans can affect several organs although any organ can be involved, and a predominance of IgG4 + plasma cells is found histologically, with an IgG4 + /total IgG + plasma cell ratio >40%. 4Pancreas, lymph nodes, bile ducts, lacrimal glands, salivary glands, lungs, kidneys, aorta, meninges, and thyroid gland are the most commonly affected organs, and the majority of the patients have increased serum IgG4 concentrations. 5reover, an association between atopy and IgG4-RD has been reported in human patients. 6mplete clinical findings in ECS with CP have not been reported previously.Involvement of kidney, salivary glands, and orbital tissue in affected ECS in a recent study of histological findings 2 suggests that affected dogs may have clinical signs of glomerulonephritis (GN), dry eye, and dry mouth.Immune-mediated cholangitis is also common in affected humans but was not prominent in histology of affected ECS.
A possible association with anal sacculitis and anal sac carcinoma also has been suggested, 7 and ECS with CP had an increased frequency of the same dog leukocyte antigen (DLA) type as ECS with anal sac carcinoma, 8 as well as ECS with immune mediated hemolytic anemia (IMHA). 9r aims were to describe the signalment and clinical findings in ECS with proven or suspected CP with a focus on other immunemediated diseases and anal sacculitis.Our hypothesis was that a high proportion of ECS with CP would have associated immune-mediated diseases, particularly GN and dry eye.In addition, we hypothesized that many affected dogs also would have anal sac disease but would not have associated liver disease.

| Control dogs
Age-matched control dogs were included and were also divided into 2 groups.Control group 1 consisted of healthy ECS.Surplus blood samples were obtained from ECS having pre-anesthetic blood screening for routine procedures.Control group 2 consisted of ECS affected by other conditions not related to CP and with no known immunemediated etiology.Surplus blood samples were obtained during routine clinical investigations in these dogs.
Inclusion criteria for control dogs consisted of no clinical evidence of CP, GN, KCS, anal sac disease or any other immune-mediated diseases.Referring veterinarians were asked to perform STT on control dogs to rule out KCS and submit a blood and voided urine samples from each dog.Only dogs with normal DGGR lipase or Spec cPL or both and no previous clinical signs suggestive of CP, normal tear production, and UPC <0.4 were included as controls.Dogs that presented with normal or slightly increased DGGR lipase (<2x the upper limit of the reference interval) were used as controls only if they did not have any signs or history compatible with CP and the other diseases mentioned above.Exclusion criteria for control dogs consisted of corticosteroid or any other immunosuppressive treatment within the 6 months before obtaining the blood and urine samples.

| Clinical pathology
The DGGR lipase test, UPC ratio, packed cell volume (PCV), and platelet count were performed at Central Diagnostic Services, University of Cambridge, and were offered without charge for cases and controls.Many cases also had cPLI, serum biochemistry profiles, and CBCs performed during clinical investigations.
Serum lipase activity was measured using a commercial DGGR lipase assay for dogs (Lipase assay DGGR; Randox) validated in house.The reference interval was 0 to 44 IU/L. 10 However, a subsequent study from the same laboratory suggested that, using this reference interval, some control dogs may have mildly increased DGGR lipase but <3x the upper limit of the reference interval. 11Pancreatic lipase immunoreactivity was measured using the Spec cPL assay in most cases (Idexx laboratories, Wetherby, UK).Results >400 μg/L were considered positive for pancreatitis.Some cases were diagnosed with SNAP cPL (Idexx laboratories, Wetherby, UK).
Most cases had a complete urinalysis and negative sediment examination findings in addition to measurement of UPC.Most controls only had measurements of urine specific gravity and urine UPC.

| Other conditions and coat color
Atopy and other immune-mediated diseases were diagnosed based on the clinical records of the attending veterinarian.Hypothyroidism was diagnosed based on concurrent low circulating thyroxine (T4) and high circulating thyroid-stimulating hormone concentrations.Where available, coat color of both cases and controls was recorded.Coat colors were categorized according to the Kennel Club breed standards for the ECS: (https://www.thekennelclub.org.uk/breed-standards/gundog/spanielcocker/).

| Statistical analysis
The ages of cases and controls were compared using the Mann-Whitney U test.Odds ratios (OR) and 95% confidence intervals (CI) were calculated for disease association with coat color.
Fisher's exact test was performed to determine if the OR was significant or not.The level of significance was set at P < .05.
Statistical analysis was performed using SPSS software (IBM, version 22.0).

| Clinical features in cases
One-hundred four ECS with CP were recruited.There were 51 females and 53 males.Median age was 9.1 years (range, 2.7-16 years; Table 1).Group 1 included 56/104 (53.8%)CP cases (definite and confirmed on histology [n = 9] or highly probable [n = 47]) and group 2 (likely CP) included 48/104 (46.1%) cases.Figure 1 2 summarizes the frequency of these conditions in affected ECS from groups 1 and 2. The proportion of cases presented with these conditions in the case study population is shown in Figure 2.

| Clinical features in control dogs
No significant difference was found in age between CP cases and controls (median age, 9.1 years, range 2.7-16.0years vs. 9.7 years; range 7.0-13.3years: Mann-Whitney U test, P = .30).
Seven dogs from control group 1 had slightly increased DGGR lipase (<2x above the upper limit of the laboratory reference interval) but did not have other signs of pancreatitis.Eight dogs from group 2 had slightly increased DGGR lipase (<2x the upper limit of the reference interval) with no other signs of pancreatitis.Five of them had intervertebral disc disease.Parti-color coats had a significantly higher frequency in cases than in control dogs (80.9% vs 55.9%, P = .01).Blue roan parti-color coat was overrepresented in cases compared with controls (52.8% vs 11.8%), indicating association of blue roan coat color with CP in ECS (odds ratio [OR], 8.39; 95% confidence interval [CI], 2.73-25.80;P < .001; Figure 4).
Signalment and clinical details of case and control dogs can be found in Tables S1-S3.

| DISCUSSION
We describe for the first time a large number of ECSs with CP compared with an age-matched control group of the same breed.Dogs with other immune-mediated diseases were excluded from the control group because surplus blood samples from some of these dogs were used for other studies of immunologic disease.In addition, exclusion criteria included immunosuppressive treatment in dogs in the control group because such treatment could mask relevant clinical signs.
Affected dogs were significantly more likely to have parti-colored coats compared with unaffected dogs and showed some similarities with IgG4-related disease humans in that many dogs had concurrent potentially immune-mediated diseases.However, unlike humans, very few affected dogs (2.9%) had concurrent liver disease.A high frequency of KCS was found in the ECS with CP in our study (47.1%), whereas other studies have shown a prevalence of KCS of 4% in the general dog population. 12Comparisons could not be made with the control population in our study because absence of KCS was an inclusion criterion for the control groups.In humans, orbital tissue frequently is involved in the course of IgG4-RD.These patients often present with dry eye, which may be associated with lacrimal gland or orbital nerve involvement or both. 13The high frequency of KCS observed in our study suggests a similar link between CP and KCS in ECS.Therefore, it is relevant to evaluate tear production by the use of STT and determine the presence of dry eye in affected ECS.However, the prevalence of KCS in the ECS breed is unknown.Both CP and KCS could be common in ECS without necessarily having a shared pathogenesis.
In humans with IgG4-RD, xerostomia (dry mouth resulting from decreased or absent saliva production) is present in 30% of patients.
Histologic involvement of salivary glands (IgG4-related sialadenitis) is observed in 27% to 53% of human patients with IgG4-RD and submandibular glands are more frequently affected, but parotid, sublingual, and labial salivary glands also can be involved. 14We described infiltration of IgG4 positive plasma cells in the salivary glands of ECS in a previous histologic study. 2 Difficulties in clinical evaluation and inconsistent reporting in cases and controls precluded evaluation of xerostomia in our study.Xerostomia is rarely evaluated in dogs but does occur, 15 and it is a known cause of persistent dental disease in humans. 16Repeated dental treatments might be a surrogate marker for suspected xerostomia in dogs, but it would be difficult to determine the actual cause of persistent dental disease in dogs in a retrospective study.In humans, salivary flow is measured by stimulating saliva production by having the patient chew a piece of wax for 5 minutes and then expectorating.Such a procedure would not be easily applied to dogs.Further investigation would require a reliable method to measure salivary production in dogs.Assessing the size and firmness of salivary glands on clinical examination may be valuable in future studies of ECSs with CP.
In addition, proteinuria with suspected GN was common in ECS with CP (45.2%).Suspicion of GN in most dogs was based on a UPC >1 (32.7%) or by a history of proteinuria when a urine sample was not available (9.6%) and was confirmed on histology in only a few dogs (2.9%).In a manner similar to CP, GN be diagnosed on histopathology but renal biopsies are uncommonly performed in either humans or dogs because biopsy is a relatively invasive procedure.Proteinuria therefore was used as a surrogate marker of potential GN.Persistent proteinuria is a clinicopathological sign of GN, and a UPC >1 was used as a marker of possible GN in our study.A relatively low cut-off for abnormal UPC was used so as not to miss any potential cases of GN.A UPC >1 is more than the borderline proteinuria as defined by the International Renal Interest Society (http://www.iris-kidney.control samples were collected by voiding whereas many cases had cystocentesis performed.It is possible that some control dogs had unidentified active sediment which would have increased the UPC.Nonetheless, the UPC was significantly higher in the cases than controls.
In our study 29.8% of the ECS with CP had a history of anal sacculitis, whereas other studies have shown a prevalence of 12.5% for anal sacculitis in the general dog population. 19A previous unpublished histological study of anal sac tissue from ECS with anal sacculitis showed histopathological findings morphologically similar to IgG4-RD in humans, with lymphocytic and plasmacytic infiltration, fibrosis, and increased numbers of IgG4 + plasma cells. 20This finding suggests that anal sacculitis in ECS could be an IgG4-RD sequela.More studies are needed to define the pathophysiology of anal sacculitis in ECS compared with other breeds, including defining lymphocyte subsets and numbers on histology and response to immunosuppressive drugs.
Involvement of other organs (eg, kidney, lacrimal glands, anal sacs) in addition to the pancreas was seen in 72% of ECS with CP, suggesting a multisystemic chronic inflammatory disease, similar to IgG4-RD.However, the possibility of co-occurrence of independent disease entities should be considered rather than multiorgan disease.Also, the extent of involvement of other organs may have been underestimated because of a lack of relevant imaging studies or biopsy specimens.
Atopy and atopic dermatitis were present in 20.2% of our study population.Other studies have reported atopy and atopic dermatitis in approximately 10% of the general dog population. 21A potential link has been described between atopy and IgG4-RD in humans, although it remains controversial.A study in the United Kingdom has reported a frequent clinical history of allergy (63%) and atopy (40%) in human patients with IgG4-RD 6 and similar findings have been reported in Japanese AIP patients 22 but not in other studies. 23Concurrent immunemediated conditions such as hypothyroidism, IMTP, IMHA, IMPA, IBD, and hypoadrenocorticism were seen in 16 (15.4%)ECS with CP, supporting a predisposition to immune-mediated disease, which is already recognized in the veterinary literature.Dogs affected by atopic dermatitis, IMTP or IMHA are frequently young to middle-aged. 24,25Dogs with CP frequently are older.In our study, the median age of affected ECS was 9.1 years whereas the mean age of dogs with IMHA in another study was 6.6 ± 2.9 years 25 and most dogs with atopy present between 6 months and 3 years of age. 24These findings could suggest an effect of immunological aging on the type of disease.A previous genetic study showed an increased frequency of a single DLA haplotype in ECS with CP. 7 This haplotype also is more frequent in ECS with IMHA, 9 suggesting that CP may be part of a multiorgan immune-mediated disease in ECS in which this DLA haplotype may play an important role.
Chronic pancreatitis causes permanent destruction of pancreatic tissue and progressive loss of exocrine and endocrine function, which eventually can progress to EPI or DM or both. 26These diseases reportedly develop only if 80% to 90% of functional pancreatic mass is lost, because the pancreas has extensive functional reserve capacity.Thus, many dogs will not reach end stage disease during their lives.A complex cause and effect relationship exists between pancreatitis and DM in dogs.A recent study in primary-care clinics in the UK indicated that pancreatitis is associated with the risk of dogs developing DM.The study reported an apparent annual prevalence of DM of 0.26% and an annual incidence risk of 0.09% in dogs ≥3 years of age.The same study showed an increased hazard of death after diagnosis of DM in cocker spaniels compared with crossbreeds. 27No data currently is available on the prevalence of EPI in dogs with CP.In our study, 6.7% and 10.6% of ECS with CP had concurrent DM and EPI, respectively.These high frequencies emphasis the importance of actively evaluating their presence in ECS with CP.
The inflammatory lesions in IgG4-related pancreatic disease in humans frequently cause diffuse enlargement of the organ or, in some cases, a benign pancreatic mass, making differentiation from pancreatic cancer difficult. 28In our study, only 5 affected dogs had mass-like pancreatic lesions identified on ultrasonography.This pancreatic feature appears to be caused by fibrosis and inflammation.
Surprisingly, a strong association between CP and dogs with parti-color coats was observed in our study (80.9% of the affected ECSs were parti-color).In particular, the blue roan parti-color coat was overrepresented in cases compared with controls (52.8% vs Fifteen dogs in the control population (7 from control group 1 and 8 from control group 2) had slightly increased serum DGGR activity.This increase was above the upper limit of the laboratory reference interval, but <3x the upper limit, which is the value often used clinically to signify a clinically relevant increase. 30This slight increase in the control dogs may be explained by hydrolysis of other lipases such as lipoprotein lipase or gastric lipase detected by the DGGR assay or could reflect a laboratory reference interval that is too narrow (which has been previously suggested 11 ), normal biological variability, or the presence of pancreatic lesions that were not identified histologically because our control group did not have histology performed. 11r study had limitations inherent in its initial design and control selection.Control dogs were included if they did not have evidence of CP but also if they did not have evidence of GN, KCS, anal sac disease or any other immune-mediated diseases.Therefore, it was not possible to compare the prevalence of these diseases between groups.
Future work including a control group without CP but also without exclusion of the conditions mentioned above would be needed to determine if these conditions are part of a multiorgan disease or if they are coexisting independent disease entities.Another limitation was the use of surrogate markers and not histopathology for the diagnosis of CP and GN.There was potential for bias in selection of both cases and controls such that they may not be representative of all ECS.However, involvement of the breed society in recruitment resulted in inclusion of dogs from multiple veterinary practices, which should minimize selection bias.
In conclusion, we characterized the clinical manifestations of CP in ECS and suggest potential clinical similarities to IgG4-RD in humans, including high frequency of KCS, proteinuria, atopy and other concurrent immune-mediated diseases that represent part of a multiorgan inflammatory syndrome.Anal sacculitis is also common in affected dogs.Especially interesting is the predisposition of parti-color ECS, in particular blue roans, to CP. Future work is indicated, involving follow-up of affected dogs to obtain current clinical records, development of a serum IgG4 test specific for dogs, and studies of the effects of immunosuppression in affected dogs to improve understanding of this disease and its manifestations.
English cocker spaniels affected by CP and client owned control ECS were prospectively recruited for this study between 2008 and 2019.Blood and serum samples were obtained from affected and control dogs and stored at À80 C until further analysis.Case material was obtained from dogs presented to the Queen's Veterinary School Hospital, University of Cambridge, and from private veterinary practices in the United Kingdom.Specifically, affected and control dogs were recruited with support from The Cocker Spaniel Club and The Kennel Club with the assistance of their breed health coordinators.Clinical records and surplus blood samples from clinical investigations were used with informed owner consent.Data regarding signalment and clinical status was collected from the clinical records.The study was approved by the Department of Veterinary Medicine, University of Cambridge Ethics Committee (CR153 and CR291).

2. 1 . 1 |
Cocker spaniels affected by CP Chronic pancreatitis was diagnosed either by pancreatic histology or by increased lipase concentrations measured using a canine pancreatic lipase immunoreactivity (cPLI) test or a 1,2-o-dilauryl-racglycero-3-glutaric acid-(6 0 -methylresorufin) ester (DGGR) lipase assay or both, abnormal findings on pancreatic ultrasonography and presence of compatible clinical signs.All affected dogs had clinical signs consistent with CP (recurrent presentation with ≥2 of: vomiting, diarrhea, lethargy, anorexia, abdominal pain, and weight loss).Cases were divided into 2 groups because methods used to diagnose CP varied among dogs.Case group 1: definite CP was diagnosed based on pancreatic histology and highly probable CP was diagnosed based on both an increased cPLI or DGGR lipase or both and pancreatic ultrasonographic findings consistent with pancreatitis.Case group 2: likely CP was diagnosed based on increased cPLI in the absence of ultrasonographic findings, increased DGGR lipase, increased amylase or conventional lipase, and clinical signs consistent with CP.Signalment details were recorded and case records were examined for the presence of any other disease conditions and with particular emphasis on diseases with reported histological associations with CP in ECS.The conditions were recorded as follows: keratoconjunctivitis sicca (KCS), diagnosed based on the Schirmer tear test (STT <15 mm/min) or a clinical diagnosis in the clinical records, GN (based on renal histology or suspected based on a urine protein: creatinine ratio [UPC] >1 on either a voided or cystocentesis urine sample) or a clinical diagnosis in the clinical records, anal sacculitis (history of recurrent expression of anal glands or infection), anal sac adenocarcinoma (histologically confirmed), atopic dermatitis, other immune-mediated diseases, and diseases caused by end-stage CP (eg, exocrine pancreatic insufficiency [EPI] and diabetes mellitus [DM]).Other concomitant diseases also were recorded.
presents the groups and their clinical and diagnostic findings.Lethargy, weight loss, inappetence, cranial abdominal pain, diarrhea, and intermittent vomiting were the most common clinical signs.Ultrasonographic changes in the pancreas compatible with CP were described in 47 cases.A heterogeneous mass-like lesion in the pancreas was reported in 5/47 (10.6%)ECS that underwent ultrasonography.Forty-nine dogs were diagnosed with KCS (STT results available [n = 32]; history of KCS confirmed by referring veterinarian [n = 17]).Thirty-eight of these dogs had bilateral KCS and the remaining 11 had unilateral KCS.Forty-seven dogs presented with suspected or confirmed GN (suspected with UPC >1 [n = 34], with history of proteinuria [n = 10]; and confirmed on histology [n = 3]).The UPC median of the dogs with UPC performed was 4.1 (range, 1-13.7).Thirty-two dogs had UPC ≥2.Anal sac disease was identified in 36 dogs (anal sacculitis [n = 31]; anal sac adenocarcinoma [n = 5]).Twenty-one dogs had a history of atopy.There was a history of other immune-mediated conditions such as hypothyroidism (n = 6), immune-mediated thrombocytopenia (IMTP; n = 4), IMHA (n = 4), immune-mediated polyarthritis (IMPA; n = 2), inflammatory bowel disease (IBD; n = 4), and hypoadrenocorticism (n = 2) in 16 cases although no dogs were receiving immunosuppressive treatment for these conditions at the time of enrollment in the study.Table
Forty-four ECS that fulfilled the strict inclusion criteria were recruited as controls.Fifteen dogs belonged to control group 1 (healthy ECS), and 29 belonged to control group 2 (diseased control group).Group 1 consisted of 7 females and 8 males and group 2 of 10 females and 19 males.Median age was 10.0 years for group 1 (range, 7.0-11.3years) and 9.2 years for group 2 (range, 7.0-13.3years; Table
com/pdf/IRIS-DOG-Treatment_Recommendations_2023.pdf).Ideally, repeat urine samples would have been taken to confirm abnormal UPC, as recommended by the American College of Veterinary Internal Medicine consensus statement on the diagnosis of glomerular disease in dogs, 17 but doing so was not possible in our study because only 1 urine sample was submitted from each dog.Increased UPC is not specific for GN and could occur as a result of amyloidosis, hypertension, urinary tract infections, inflammation associated with ongoing CP or other disorders.Therefore suspicion of GN is only presumptive without a biopsy.Also several dogs did not have a UPC performed and their medical records simply mentioned the existence of proteinuria.Proteinuria alone is difficult to interpret because it varies with urine concentration and among different samples.17,18Sample collection and analysis methods differed between cases and controls.Most cases had complete urinalysis and sediment examination to rule out urinary tract inflammation whereas most controls did not, and most F I G U R E 3 Number and percentage of ECSs affected by chronic pancreatitis with concurrent multiorgan pathology.F I G U R E 4 Comparison of proportions of different coat colors between cases and controls.(A) Parti-color and (B) blue roan coat proportions in cases (dark gray) and controls (light gray).P values indicate statistical significance of differences.
T A B L E 1 Signalment and clinical details of cases and controls included in this study.
Number of ECS dogs that presented with conditions associated with CP.
T A B L E 2 F I G U R E 2 Proportion of ECS dogs affected by CP presenting with other concurrent conditions.[Correction added after first online publication on 31 May 2024. Figure 2 has been revised.].