All 15 dogs of this study revealed some abnormal findings in US and ERC/ERCP examinations. The most agreement between US and ERC occurred for assessing the normality or abnormality of the CBD. In case of disagreement of both modalities, the ERC findings of the ductal structures were in line with the available laparotomy, laparoscopy, or autopsy findings more often than the US findings. In contrary, for the gallbladder the US findings were closer to the laparotomy, laparoscopy, or autopsy findings. In a human study, US was in agreement with abnormal ERCP findings in 101 of 120 patients (sensitivity 84%) and of normal ERCP findings in 70 of 73 patients (specificity 95%) [16]. US has been shown to be more sensitive in diagnosis of dilated or strictured bile ducts than finding choledocholithiasis [16]. ERCP is considered complimentary to US specifically when the problem is located within the pancreatobiliary ductal anatomy.
The comparison of US and ERC results revealed different degrees in agreement and disagreement dependent on the reported anatomical structure. The major papilla was visible in all 15 dogs scanned by endoscopy for ERC examination, but was reported only in six dogs by using US. This is in agreement with a previous study, which found that the major papilla was visualized ultrasonographically in 42% of dogs without evidence of abdominal disorders [17]. The visibility was negatively influenced by the presence of air or food in the gastrointestinal tract and increased body weight [17]. US and endoscopy for ERC/ERCP disagreed concerning the major papilla in three dogs of our study, which may be due to contractile activity of the sphincter of Oddi [18] or different types of the major papilla [19]. Since no histopathology reports were available for the major papilla in these three dogs, however, it is impossible to show which modality had the correct diagnosis.
US and ERC agreed in a normal or dilated CBD in 10/15 dogs. A recent study showed that US is a useful modality for detection of the canine CBD diameter and that US results were not significantly different from those of CT [20]. In human medicine, bile duct dilatation has been detected by US with the sensitivity of 85–95% for experienced examiners [21]. In all three dogs (Dogs 1, 5, 10) that showed normal CBD in US but abnormal in ERC, pathology revealed abnormal findings of the hepatobiliary system which highlights the fact that a normal US examination cannot rule out the need for ERC. Two dogs (Dogs 2 and 7) had a dilated CBD in US but the duct did not fill with contrast material in ERC. In Dog 2, this was found to be due to infiltration of carcinoma within the intestinal wall, causing pressure to the duct, and in the other dog, precipitated bile was detected within the gallbladder at autopsy. MRCP could be helpful in cases where space-occupying masses prevent contrast material visibility within the biliary ducts [4].
The gallbladder was visible and could be assessed more often by US than by ERC, and the US findings were more in agreement with the laparotomy, laparoscopy, or autopsy findings where available. It can be speculated that the diagnostic value of US is preferentially for the gallbladder, and ERC for ductal structures. In human medicine, it has been shown that US has high sensitivity for detecting cholecystitis [22]. Another study showed that CT was significantly more sensitive for diagnosing cholecystitis than US [23], but because of the speed and portability, US is still used as the initial imaging technique for evaluating patients with suspected acute cholecystitis.
Extrahepatic ducts were not mentioned in any of the US reports. Extrahepatic ducts are usually poorly visualized owing to overlying bowel gas and the normal intrahepatic ducts are not visible in US [24]. Dilated intrahepatic ducts were reported in one dog in US, but ERCP revealed an actual dilation of extrahepatic ducts. Since ERCP has the ability to show the entire biliary system in one view, it can be more reliable in localizing the abnormality.
The technical success rate of ERC and ERP was 88.2% and 66.7%, respectively. This is comparable with previously reported success rates of 67–75% for ERC in dogs and cats and of 70% for ERP in dogs c8, 9, 10]. Failures have been reported to be related to difficulties in the duodenal papilla cannulation in a small-sized dog, intraduodenal contents, changes of the duodenal mucosa, and decrease in gastric tone [9].
In the six dogs that underwent ERP, the duodenal and gastric branches of the main pancreatic ducts were clearly visible with no signs of duct abnormalities in five dogs. In one dog (Dog 15), with serum cPL concentration in the reference range, the duodenal branch of the pancreatic duct did not go caudally but cranially to the diaphragm. In humans, abnormal contrast pancreatograms are used to grade the stage of chronic pancreatitis by the Cambridge classification system from mild to severe [25]. It remained open whether the ERP finding of the dog in our study was a variation of normality, an artifact, or a possible pathologic finding, since the dog underwent no direct examination of the pancreas by invasive diagnostic methods.
The success rate of EST was 81.8% with no major procedure-related complications during or immediately after the procedure. In humans, biliary EST is associated with several complications, both in the short- and the long-term [13, 14]. Due to the invasive nature of the procedure, these complications are inevitable and observed in some patients, depending on patient- and procedure-related factors [11]. The short-term complications of EST are bleeding, perforation, pancreatitis and cholangitis [13, 14]. Early identification and appropriate management of complications is essential to reduce mortality and morbidity. Of the known short-term complications of EST, only mild bleeding occurred in one dog of our study. In human medicine, acute pancreatitis is the most common post-ERCP complication, with an incidence of 2–10% [26]. One dog (Dog 14) showed acute severe necrotizing pancreatitis six months after EST, but there are no laboratory or clinical records of the dogs until six months after EST. It can be suggested that an acute pancreatitis developing six months after EST is unlikely to have been caused by the procedure.
Only four of nine dogs showed clinical and laboratory improvement after EST. This finding is similar to the findings of previous studies reporting clinical improvement after EST in a dog with papillary stenosis [9] and in one dog with signs of extrahepatic duct obstruction treated by EST and biliary stenting [10]. One dog (Dog 6), regardless of clinical and laboratory improvement after EST, developed recurrent fever four months after EST and bacterial culture revealed ESBL E. coli in the bile and feces. A study in humans showed that the prevalence of post-ERCP biliary tract infections was 4% of the patients [27]. It remained open whether the ESBL E.coli infection in Dog 6 was a complication of ERC or an ascending infection from intestine after EST.
The finding of dilated CBD in ERC without evidence of obstruction in two icteric dogs, was of interest. The cause of this dilation could not be identified. However, after EST there was marked bile flow into the duodenum. A report on six cats with sphincter of Oddi pathology speculated that patients without obvious pathology at the papilla might have sphincter dysfunction with extrahepatic biliary obstructions due to a chronic inflammatory enteropathy [28]. Histopathology of the duodenal biopsy samples in Dog 8 revealed mural enteritis at the papilla.
Four dogs underwent EST without improvement and were euthanized within days after the procedure. Clear causes of posthepatic and intrahepatic cholestasis were established by autopsy in three dogs. Dog 2 had an exocrine pancreatic carcinoma infiltrating the duodenal wall, Dog 7 a moderate chronic cholangiohepatitis and cystic mucinous hyperplasia of the gallbladder mucosa and Dog 8 moderate chronic cholangiohepatitis and severe chronic mural enteritis. No clear cause was found for Dog 5 with a CBD stricture and increasing hyperbilirubinemia after EST, since hepatic lipidosis was the sole histopathological finding in the liver. Prolonged cholestasis after ERC may be one possible reason. It is a very rare but recognized complication following ERCP in humans [29]. The exact mechanism for prolonged jaundice post-ERCP has remained unclear but it has been suggested that it may be directly related to the radiocontrast medium, perhaps due to an idiosyncratic adverse reaction resulting in disruption of the hepatocyte canalicular membrane or interruption of transport pumps with subsequent intrahepatic cholestasis and jaundice [29, 30].
Findings of patients without short-term improvement after EST, and the occurrence of an infection with ESBL, acute necrotizing pancreatitis and biliary peritonitis within 3 to 10 months after EST revealed that EST can be of no help or has the possibility to harm the patient in the short or long run. The indication for EST should be the result of a thorough diagnostic work up of the individual patient suspect for a biliary tract disorder to rule out hepatic or pancreatic diseases requiring different therapeutic approaches. The study revealed that ERC has some advantages over US in visualizing pathologies of the biliary ductal system, but it is technically demanding and with some risk of complications. MRCP is a less invasive alternative to diagnostic ERC since it was shown to have good accuracy for the diagnosis of hepatobiliary and pancreatic duct disorders in dogs [4]. MRCP could be a beneficial imaging tool for selecting patients that require ERC and subsequent EST. MRCP does not only help to visualize the biliary and pancreatic duct systems but also the surrounding soft tissues which is impossible with ERCP. As established in human medicine, endoscopic treatment should include biliary stenting when diagnosing distal or long CBD strictures by MRCP or ERC, especially with malignant lesions that are not eligible for open surgery [31]. Endoscopic CBD stenting has been proven possible in dogs [10]. It is very likely that the establishment of MRCP in the diagnostic work up of dogs with suspected disorders of the biliary duct system could give a better indication and a more successful outcome of endoscopic interventions [4].
There are some limitations in the present study. The retrospective nature of the study caused lack of systematic reports for US and ERC/ERCP and EST. The US reports were written for clinical purposes and by different veterinary radiologists. Lack of systematic collection of laboratory, ultrasonographic and clinical follow-up data after the procedures made comparison difficult in some cases. There was also a lack of histopathologic confirmation of the lesions in some patients.