Long-term Survival of Dogs After Cholecystoenterostomy: A Retrospective Study of 15 Cases (1981–2005)
Fifteen dogs with extrahepatic biliary tract disease underwent cholecystoenterostomy. Long-term survivors were significantly older at presentation (mean age 140.5 months) than dogs that survived the first 20 days after surgery but subsequently died from causes related to the surgery or hepatobiliary disease (mean age 72 months). Dogs that died during the first 20 days had significantly more complications in the hospital than dogs that survived this period. The type of underlying hepatobiliary disease (i.e., benign or malignant) was not associated with either short-term outcome or long-term survival. Eight dogs died from causes related to surgery or hepatobiliary disease. Long-term complications included hepatic abscesses, acquired portosystemic shunts, pancreatitis, and vomiting.
Introduction
Extrahepatic biliary tract conditions, such as infiltrative disease, scar tissue formation, or trauma, can result in bile duct obstruction. Early cholecystoenterostomy (CCE) is recommended to establish bile duct patency (by re-routing bile from the gallbladder to the intestine) and to provide relief of clinical signs.1–4 Biliary diversion can be achieved by performing cholecystoduodenostomy (CCD) or cholecystojejunostomy (CCJ), provided that 1) the gallbladder is intact and not directly involved in the extrahepatic biliary tract disease process, and 2) the hepatic and cystic ducts are patent.5,6 Extrahepatobiliary surgery in dogs is associated with high morbidity and mortality.3,4,7 In recent studies of biliary surgery in dogs, factors identified as possible predictors of poor outcome included advanced age, increased preanesthetic heart rate, increased serum bio-chemical values (including γ-glutamyltransferase, blood urea nitrogen, phosphorus, bilirubin, and preoperative creatinine), septic bile peritonitis, prolonged partial thromboplastin time, immediate postoperative hypotension, and biliary diversion procedures.4,7 Postoperative risk factors associated with poor survival included hypotension, dyspnea, decreased concentrations of albumin and globulin, increased concentration of bilirubin, and an increased percentage of band neutrophils on the white blood cell differential count.7 However, specific data on morbidity, mortality, and long-term survival of dogs having CCE have not been reported.
The purposes of this retrospective study are to report short- and long-term postoperative complications and to identify variables associated with short-term outcomes and long-term survival in dogs that had CCE.
Materials and Methods
Criteria for Case Selection
Medical records of dogs that had undergone CCD or CCJ from January 1981 to August 2005 at the University of Missouri–Columbia were reviewed.
Procedures
Data extracted from the medical records included the following:
-
Signalment (i.e., breed, gender, age, and weight at surgery)
-
Medical history
-
Duration of clinical signs (from onset to surgery)
-
Physical examination findings
-
Serum chemistry values (i.e., glucose, blood urea nitrogen, creatinine, albumin, total bilirubin, alanine aminotransferase, alkaline phosphatase, and amylase)
-
Prothrombin time and partial thromboplastin time
-
Measurements of vascular pressure (i.e., lowest mean direct intraoperative arterial blood pressure, lowest mean direct postoperative arterial blood pressure, lowest postoperative central venous pressure, and highest postoperative central venous pressure)
-
Surgical findings, including duration of surgery, stoma size at surgery, and placement of a jejunostomy tube
-
Histological findings from tissue biopsies, including diagnosis and level of malignancy
-
Pre- and postoperative hospitalization time
-
Complications both in hospital and after discharge
-
Short-term outcome and long-term survival
Short-term outcomes for both survivors and nonsurvivors were defined as those occurring within 20 days after surgery. This time period was chosen based on the authors’ clinical impression before the study that many dogs with biliary diversion surgery die or are euthanized (in hospital or after discharge) within the first 20 days because of complications related to surgery or underlying disease. The authors’ impression was supported by a recent study in which the survival curve for dogs undergoing biliary surgery plateaued at 20 days.7
Long-term survival was defined as the time from surgery to last contact with the owner or referring veterinarian, as long as this time was >20 days. Dogs were considered lost to follow-up if the owners or referring veterinarians could not be contacted at the time of writing of the present study. Four owners and/or referring veterinarians were contacted by telephone to determine their animals’ postoperative complications, recurrence of clinical signs, morbidity, and possible cause of death.
Statistical Analyses
Data are presented as mean ± standard deviation (SD). Fisher’s exact test was used for analysis of categorical variables. Student’s independent t-test or the nonparametric alternative Mann-Whitney rank sum test was used for comparison of continuous variables, depending on the results tests of equal variance and data distribution. A P value <0.05 was considered significant. Statistical analyses were performed using the SigmaStata software program.
Results
Signalment
Of the 15 dogs included in the study, nine were female (seven spayed) and six were male (five castrated). Dogs were mixed-breed dogs (n=4), toy poodles (n=2), and fox terrier, American Eskimo, Silky terrier, Yorkshire terrier, Boston terrier, Siberian husky, springer spaniel, German shepherd dog, and golden retriever (n=1 each). The mean age ± SD of dogs that survived long-term (140.5±19.2 months; n=4) was significantly higher than the mean age of those that died long-term (72.0±17.0 months; n=2) from causes related to surgery or hepatobiliary disease (P=0.013). The mean weight ± SD of dogs that died long-term from causes related to surgery or hepatobiliary disease (28.6±17.0 kg; n=2) was not significantly different (P=0.185; power=0.159) than that of dogs that survived long-term (13.1±8.3 kg; n=4).
Medical History and Clinical Findings
Historical complaints for the 15 dogs are presented in Table 1. Median duration of clinical signs from onset to surgery was 14 days (range 7 to 120 days). The mean ± SD rectal temperature for the 15 dogs was 38.9±0.6°C (range 38.0 to 40°C). Three dogs had temperatures >39.4°C. The mean ± SD heart rate was 105.9±28.4 beats per minute (range 40 to 140 beats per minute).
Neither short-term outcome nor long-term survival was significantly associated with any of the medical/clinical variables included in this study [Table 2].
Hospitalization and Surgery
The mean ± SD preoperative hospitalization time for the 15 dogs was 2.9±2.8 days (median 2 days, range 0 to 8 days). The mean ± SD postoperative hospitalization time was 6.9±5.3 days (median 6 days, range 1 to 20 days).
Clinical data for the 15 dogs that had CCE are presented in Table 3.
Cholecystoduodenostomy was performed in 13 dogs, and CCJ was performed in two dogs. Ten dogs underwent only CCD, while the other CCD cases included either a Bilroth I procedure or a partial pancreatectomy. The mean ± SD stoma size at surgery was 3.1±0.6 cm (range 2.5 to 4 cm; n=12). In one case, CCJ was performed for management of prior iatrogenic bile duct ligation.
Jejunostomy tubes were placed in nine dogs, and a gastrostomy tube was placed in one dog. Other procedures that were performed included open peritoneal drainage for management of septic peritonitis (in two dogs) and liver biopsy (in six dogs).
Histopathological Findings
Histopathology was available for 13 dogs [Table 3]. Underlying hepatobiliary disease was benign in 10 dogs and malignant in three.
Outcomes
By the end of the study, 11 dogs had died over a median follow- up time of 19 days (range 1 to 1259 days), with the other four dogs lost to follow-up after a median of 45 days (range 6 to 224 days). Six dogs died within the short-term period (20 days of surgery) from causes related to the surgery or underlying hepatobiliary disease; of these six dogs, five died from cardiac arrest, and one was euthanized because of persistent vomiting.
Of the 11 dogs that died, eight died from causes related to the surgery or underlying hepatobiliary disease (median follow-up 13 days, range 1 to 1259 days), and three dogs died from unknown causes. Eight dogs developed complications while hospitalized at the intensive care unit, and five developed complications after discharge. Dogs that died within 20 days of surgery had significantly more complications in the hospital than those that survived beyond the 20- day period (P=0.001).
Of those animals that died within 20 days of surgery, one dog had cardiac arrest and died during an exploratory celiotomy. This dog was admitted initially for the management of extrahepatic biliary obstruction associated with previous bile duct ligation, which occurred during a partial pancreatectomy and gastroduodenostomy performed by the referring veterinarian. The authors performed a CCJ, but the dog subsequently developed peritonitis and postoperative gastrointestinal obstruction at the previous gastroduodenostomy site. A Roux-en-Y gastrojejunostomy was performed 10 days after the first surgery, but the dog did not improve and developed sepsis. Nineteen days after the CCJ, an exploratory celiotomy revealed perforated jejunal ulcers at the Roux-en-Y anastomotic site. During this exploratory surgery, the CCJ site was found to have healed normally. However, the dog had a cardiac arrest during the procedure and died.
Of the five dogs that had complications after discharge, two died or were euthanized within 20 days of surgery, and three died or were euthanized during long-term follow-up [Table 3].
One of these three dogs received medical treatment for pancreatitis that developed 18 months after surgery. This dog originally had a CCD performed for extrahepatic biliary obstruction caused by fibrosing pancreatitis. The dog died from unknown reasons 35.9 months after surgery.
Another of the long-term survivors from CCD surgery (originally presented for the management of bile duct avulsion and bile peritonitis) showed signs of hepatobiliary disease 3.5 years after the original surgery. This dog was euthanized after an exploratory celiotomy revealed multiple acquired portosystemic shunts. The CCD site was found to be patent at the time of surgery. Histological examination of liver biopsies showed hepatic portal fibrosis, bile duct hyperplasia, bile stasis, and cholelithiasis.
The third dog was diagnosed with a liver abscess and had a left lateral hepatic lobectomy 16 months after surgery. The dog originally had a gastroduodenostomy and a CCD for the management of ulcerative pyloric gastritis. This dog was presented again 1 month after the second surgery; cholangiohepatitis was diagnosed and treated medically. The dog was admitted again 16 months after the second surgery for chronic vomiting and melena. An exploratory celiotomy revealed that the CCD site was intact and patent, and histopathology findings of biopsies of the gastrointestinal tract were unremarkable. The dog was euthanized 2 months after this final surgery because of hematemesis of unknown cause.
Discussion
In this study, the older dogs had a better prognosis than the younger dogs. Older dogs with hepatobiliary disease are usually compromised and may experience increased morbidity and mortality from long anesthesia and surgical times.8 This is consistent with a recent study in dogs having biliary surgery, in which increased age was associated with poor survival.7 This finding is also consistent with the situation in people, for whom age >60 years has been associated with increased mortality after extrahepatic biliary surgery.9,10
Weight of the animal at surgery has not been previously reported to affect long-term survival in dogs undergoing CCE. In the authors’ study, small dogs with a mean weight of 13.1 kg had a better long-term survival than those with a mean weight of 28.6 kg, although this difference was not statistically significant. The authors also noted that older or smaller dogs had fewer complications (both during hospitalization and after discharge) than younger or heavier dogs.
The procedure of choice for treatment or palliation of extrahepatic biliary obstruction in dogs is CCD or CCJ.1–7,11 Choledochoduodenostomy and choledochojejunostomy are the treatments of choice for people with benign12,13 or malignant disease14,15 that require biliary diversion. The mucosal appositional technique is currently recommended for biliary diversion in dogs.11 However, the small number of dogs undergoing CCJ in this study precluded comparisons between the two CCE techniques. Small numbers and lack of specific follow-up information similarly limit the veterinary literature on this topic.
Indications for performing choledochoenterostomy in dogs are very limited, because the small size of the normal bile duct makes anastomosis difficult.6 (The importance of large stoma size was initially addressed in people.12,16) In dogs, a large stoma (2.5 to 4 cm) has been recommended as the best way to achieve free passage of the chyme that enters the biliary tract after CCE.2,6,17 In the authors’ study, the mean stoma size in dogs was 3.1 cm, which was within recommended ranges.2,6,17
In this study, the most common indication for performing CCE was fibrosing pancreatitis (47% of cases), followed by malignant disease (20% of cases). These findings are similar to those reported by others.1,3 However, in another retrospective study of 60 dogs having extrahepatic biliary surgery, only 17 of the dogs had extrahepatic biliary obstruction caused by pancreatitis (n=12) or neoplasia (n=5).4 Chronic pancreatic disease may cause bile duct obstruction, because the distal end of the bile duct is in such close proximity to the pancreas. Scar tissue around or inside the bile duct can directly obstruct flow, or the bile duct can be compressed by adjacent fibrotic tissue or pancreatic swelling, edema, and inflammation.1,2,5,11,18
An early high mortality rate has been reported in dogs after CCE.1,3,4,7 In a study of 60 dogs undergoing extrahepatic biliary surgery, 17 died within 2 weeks after the procedure.4 In a more recent study of 34 dogs undergoing biliary surgery, nine died from biliary disease within 20 days. These findings are consistent with those of the authors’ study, in which 40% of the dogs died within 20 days of surgery. An operative mortality of 7% was reported in human patients after having undergone biliary enteric bypass for benign and malignant disease.15 Malignant disease has been reported as a risk factor for mortality in people undergoing biliary surgery.9,10 In the authors’ study, the type of disease did not seem to affect either short-term outcome or long-term survival.
Complications during hospitalization appeared to significantly affect short-term outcome in this study. Based on the authors’ findings, dogs that experience complications while in the hospital are more likely to die within 20 days of surgery than those that have no complications. Cardiac arrest was a common fatal postoperative complication reported in both this study and in the literature.4
Postoperative hypotension was seen in this study, similar to reports of others.4,7 Possible causes include prolonged anesthesia, inadequate fluid administration, or sepsis.4 Immediate postoperative hypotension has been significantly associated with outcomes of dogs undergoing extrahepatic biliary surgery.4,7 However, such was not the case in this study, possibly because of the low statistical power associated with the small number of cases for which pressure measurements were available.
Vomiting after CCE was a common in-hospital complication in this study. Vomiting could be caused by gastric inflammation/atony (as was the situation for one of the cases) or by pancreatic ischemia related to shock and sepsis. The literature suggests several other surgical complications that might be associated with vomiting, including pancreatitis11,19 and localized ileus, cholangitis and cholangiohepatitis secondary to enterobiliary reflux,11 chronic active fibrosing pancreatitis,18 and postoperative pancreatitis after partial pancreatectomy.20
One dog in this study developed perforated jejunal ulcers at a Roux-en-Y anastomosis site. Experimental studies in dogs have shown that Roux-en-Y choledochojejunostomy is associated with increased gastric acid secretion21 and impaired pancreatic function.22 It has been suggested that bile diversion away from the duodenum (which was performed in the dog of this study) might lead to impairment of duodenal mechanisms for neutralization of gastric hyperacidity, which in turn might result in formation of peptic ulcers.23
Little has been published on long-term survival and long-term complications of dogs undergoing CCE. In this study, five dogs developed complications after discharge; two died or were euthanized within 20 days of surgery, and the other three died or were euthanized during the long-term follow-up period. One of the latter three dogs developed pancreatitis 18 months after surgery, which resolved with medical management. Pancreatitis has been reported to occur after pancreatic surgery as a result of secondary pancreatic ischemia.24 Such pancreatic ischemia can be caused by hypotension during general anesthesia or by occlusion of venous outflow to the cranial abdomen during surgery.19,24 In the current case, secondary pancreatic ischemia seems unlikely, given the long interval of 18 months between surgery and the pancreatitis. However, it is not unusual for animals with pancreatitis to experience recurrent disease.25
One dog in this study developed multiple portosystemic shunts, hepatic periportal fibrosis, bile duct hyperplasia, bile stasis, and cholelithiasis 3.5 years after undergoing CCD for bile duct avulsion. The 4-cm stoma reported in this dog was within the recommended range, and the CCD site was found to be patent during the second surgery. However, this dog may have developed hepatobiliary lesions because of cholangitis associated with duodenal reflux, despite a widely patent anastomosis.
The development of cholangitis after biliary enterostomy is believed to be caused by obstruction of the stoma,12,26 which results in ascending infection from the prolonged presence of intestinal contents within the gallbladder.11 In experimental studies, duodenal reflux in dogs undergoing biliary enterostomy may cause minimal clinical signs, provided that the size of the stoma is large enough to allow duodenal contents to exit the biliary system.17,26 In contrast, cholangitis may develop despite patent anastomosis, possibly because of severe bacterial contamination and abnormalities in the biliary tree that promote bile stasis.27 Dogs undergoing CCE that have a stoma equal in size to that of the intestinal lumen may develop portal fibrosis and ductal proliferation up to 6 months after surgery.28
Another dog in this study had a liver abscess 16 months after undergoing CCD. Hepatic abscesses are an uncommon manifestation of liver disease in dogs,29 but they have been reported as complications of biliary-enteric anastomosis in people.30–32 Cholangiohepatitis (with31 or without anastomotic stricture30) has been suggested as a risk factor for hepatic abscessation.32 Recurrent cholangiohepatitis and abscess formation after enterobiliary reflux of abnormal gastrointestinal flora have been described as complications of biliary surgery in people.27,31 In one study, surgery for hepatic abscesses was associated with 50% mortality.32
Cholangiohepatitis from duodenal reflux into the anastomosis could have caused the hepatic abscess in one case of this study. This dog developed cholangiohepatitis 1 month after hepatic lobectomy. The same dog had episodes of vomiting and melena and was euthanized 18 months after the second surgery for recurrent hematemesis (cause unknown). Possible causes of gastrointestinal bleeding include gastrointestinal ulceration associated with chronic hepatobiliary disease33 and/or endotoxemia. Endotoxemia has been reported to increase gastric secretion, and it has been associated with hemorrhage from acute gastric ulcers in human patients with end-stage liver disease.34 Gastrointestinal bleeding after biliary surgery has been associated with a significant risk of postoperative mortality in people.9,10
The limitations of this study suggest cautious interpretation of the authors’ findings. The relatively small number of cases precluded application of multivariable analysis and allowed only tentative detection of a few prognostic factors for short-term outcome or long-term survival. Also, the authors did not search for interactions and cannot definitely state that none were among the factors analyzed. In addition, because medical and surgical treatments of the cases varied, it was difficult to differentiate treatment-related effects from differences in medical/surgical technique.
However, the authors’ report is the only one to date that has addressed long-term outcomes and complications of dogs undergoing CCE. Results suggest that CCE in dogs is associated with high morbidity and mortality. Long-term complications seen in dogs after CCE included hepatic abscessation, acquired portosystemic shunts, pancreatitis, and vomiting. Dogs that died within 20 days had significantly more complications in the hospital than dogs that survived longer. Older dogs had a better long-term survival than younger dogs. Further studies are needed to focus on earlier diagnosis and improved methods of treatment.
SigmaStat 2.035 for Windows; SPSS Inc., Chicago, IL 60606
