Introduction

Portosystemic shunts are vascular abnormalities that redirect venous return from the gastrointestinal tract away from the liver and back into systemic circulation. This abnormal circulation leads to a wide range of clinical signs and concurrent conditions including neurologic dysfunction, seizures, cystoliths, vomiting, and diarrhea. Portosystemic shunts can be acquired or congenital and may occur within the liver parenchyma (intrahepatic) or outside the liver parenchyma (extrahepatic). Congenital extrahepatic shunts occur in 0.18% of all dogs and 1.3% of pugs.¹

Although both medical and surgical management may be employed in the treatment of extrahepatic shunts, surgical attenuation of the aberrant vessel is the only way to achieve a complete cure by preventing blood flow through the abnormal vessel. Options for surgical attenuation of extrahepatic portosystemic shunts include partial or complete suture ligation, cellophane banding, and placement of an ameroid constrictor. However, surgical attenuation is not without complication, with 4.8–12% developing neurologic complications such as intractable seizures, ataxia, and disorientation postoperatively.^2–4 In approximately 50% of these cases, these complications may lead to death or euthanasia.^4,5 Unfortunately, there is currently no universally accepted theory that explains why these complications develop. Proposed etiologies include decreases in endogenous inhibitor central nervous system benzodiazepine agonist levels, imbalances in neurotransmitters, and concurrent disease processes such as idiopathic epilepsy, granulomatous meningoencephalitis, and hypoxia.^5–7

In the authors’ experience, pugs undergoing surgery for congenital extrahepatic portosystemic shunts (CEPSS) seemed to be at a higher risk than other breeds for developing severe neurologic complications postoperatively. Pugs are also specifically affected by necrotizing meningoencephalitis, which has been shown to be a common cause of intracranial signs, such as seizures, circling, and head pressing, in this breed.^8,9 It is possible that this could be a contributing factor to a perceived increase in neurologic complications following surgery for CEPSS. The objective of this study was to determine the prevalence of postoperative neurologic complications in pugs undergoing surgical correction for a single CEPSS, as well as to determine if there was any discernable connection to necrotizing meningoencephalitis in the cases that experienced postoperative mortality. The hypothesis of this study is that pugs would have a higher prevalence of postoperative neurologic complications when compared with other breeds, and that those pugs who died would have evidence of necrotizing meningoencephalitis on necropsy.

Materials and Methods

Records from the Colorado State University Veterinary Teaching Hospital from 1998 to 2010 were searched for pugs who had undergone surgical attenuation of CEPSS. Pugs were included if they met the following criteria: single CEPSS present, surgical correction was performed and follow-up of at least 1 mo postoperatively or death within the first month postoperatively. Additionally, all owners, excluding dogs who were known to have died, were contacted via letter or telephone to inquire about the current status of their dog. This group was compared with a control group of dogs of varying breeds who also underwent surgical correction of a single CEPSS at our institution during a similar time frame (1997–2005).¹⁰ Dogs in the control group were required to meet the same inclusion criteria as the pug group.

Information collected about each dog included, but was not limited to, patient age, breed, weight at the time of surgery, gender, packed cell volume, total protein, albumin, alanine aminotransferase, blood urea nitrogen, glucose, neutrophil count, serum bile acids, date of procedure, location of shunt, type of procedure (ameroid or cellophane), postoperative complications, date of last follow-up/death, and cause of death. Survival time was defined as the difference between the date of surgery and date of death/last follow-up.

Normality of the data was determined using the Shapiro-Wilk test. Normally and nonnormally distributed data were analyzed using the Kruskal-Wallis rank sum test. Student t test was used for normally distributed data only. Fisher’s exact test was used to determine significance. Statistical significance was set at P < .05.

Results

Records for 18 pugs undergoing surgical correction during the specified time frame were reviewed. Fourteen pugs were included in the study based on the inclusion criteria. There were seven females and seven males. The control cohort of dogs consisted of 19 females and 11 males of the following breeds: Yorkshire terrier (n = 9), schnauzer (n = 4), mixed breed (n = 3), Maltese (n = 2), and one each of bichon frise, Cairn terrier, cocker spaniel, Welsh corgi, dachshund, Labrador retriever, Lhasa apso, miniature pinscher, Pekingese, Shetland sheepdog, standard poodle, and West Highland white terrier. Table 1 compares age, weight, and preoperative bloodwork values between the two groups. Weight was found to be the only significant difference, with pugs weighing significantly more than dogs in the control group (P = .03). There was no significant gender difference between the two groups (P = .61). All dogs in the control group had cellophane banding performed. Within the pug group, one dog underwent placement of an ameroid constrictor, whereas the remainder of the dogs underwent cellophane banding of the aberrant vessel.

TABLE 1 Comparison of Age, Weight, and Preoperative Bloodwork Values Between the Pug Study Group and the Control Group

View Fullscreen

Four of the 14 pugs were euthanized within 1 mo after surgery because of continued or worsening neurologic disease, as compared with the control group, in which only one dog died within 1 mo postoperatively because of neurologic disease, which was a significant difference (P = .029). Overall, pugs undergoing surgical attenuation of a portosystemic shunt had an 8.6 times greater risk of being euthanized or dying secondary to postoperative neurologic complications in the perioperative period when compared with control dogs. Of the four pugs who were euthanized, two were euthanized because of intractable seizures at 2 and 3 days postoperatively, and two were euthanized because of severe worsening neurologic status at 2 days and 29 days postoperatively. In the control group, the single perioperative mortality occurred because of intractable seizures 1 day postoperatively.

Of the four pugs that were euthanized, three underwent cellophane banding and one underwent placement of an ameroid constrictor. Table 2 compares weight, age, and preoperative bloodwork values between the pugs who survived the perioperative period and those who did not. No statistical differences were found between the two groups. Necropsies were available for two of the four pugs who died in the 1-mo postoperative period. Necropsy reports were consistent with hepatic encephalopathy on both pugs and did not reveal any evidence of necrotizing meningoencephalitis.

TABLE 2 Comparison of Age, Weight, and Preoperative Bloodwork Values Between Pugs Who Survived the Perioperative Period and Those Who Did Not

View Fullscreen

Of the 10 pugs who did not die in the perioperative period, follow-up of at least 6 mo was available for 9, with a median follow-up time of 32 mo (range 6.5–72.7 mo). All nine were still alive at the time of follow-up and were reported by the owners to be doing well with no neurologic complications and no need for medical management. Serum bile acid levels were available in 8 of these 10 pugs at least 1 mo postoperatively, with the majority being taken 3–6 mo postoperatively. Seven of those eight pugs showed a decrease in bile acid levels compared with their preoperative bile acid levels, which is consistent with vessel attenuation. The mean for the preoperative bile acid measurement was 176.67 μM/L ± 84.34 standard deviation (SD) preprandial and 224.97 μM/L ± 133.06 SD postprandial. The mean for the postoperative bile acid measurement was 65.6 μM/L ± 85.1 SD preprandial and 98.4 μM/L ± 103.6 SD postprandial. Postoperative bile acid measurements were taken on average 3.1 ± 1.3 mo after surgery.

Discussion

This study found a significant increase in the perioperative mortality of pugs following surgical correction of a single CEPSS as compared with the control population. There was no evidence of necrotizing meningoencephalitis in the dogs who had a necropsy performed, and no preoperative parameters were predictive of which pugs would have an increased risk for development of neurologic complications postoperatively. This study is the first to demonstrate a significant difference in the postoperative neurologic complication rate and mortality rate between pugs and other breeds.

In this study, 28% of the pugs who underwent surgical attenuation of a portosystemic shunt were euthanized within 1 mo of undergoing surgery because of neurologic complications. This mortality rate is much higher than recently reported overall perioperative mortality rates of 5.5%, 10.1%, and 7.5% by Hunt et al., Greenhalgh et al., and Falls et al., respectively.^3,11,12

When reviewing the outcome of pugs in other portosystemic shunt studies, it was of interest that all of the pugs in two studies developed neurologic complications postoperatively; however, only three pugs were represented between the two studies, and it is difficult to determine a breed relationship with such small numbers.^2,3 None of the three pugs who developed neurologic complications died in the immediate postoperative period. In a study by Mehl et al., only 1 of 12 pugs developed neurologic complications postoperatively, which is not consistent with the outcome seen in this study.⁴ If the data from the current study are combined with the data from these three studies, 8 of 29 pugs (27.6%) developed neurologic complications, which is consistent with the numbers seen in this study; however, the majority of the pugs in the other studies survived the immediate postoperative period, unlike the pugs in the current study.^2–4

It is still uncertain why pugs appear to be at an increased risk of developing neurologic complications after surgery for portosystemic shunt attenuation. Age at diagnosis has been previously proposed as a risk factor for development of postoperative seizures in dogs with portosystemic shunts.² It is interesting to note that although not statistically significant, there was a trend toward those pugs who died having a higher average age at the time of surgery when compared with the pugs who did not die (34.5 mo versus 8.5 mo). However, pugs were not older than the control population overall, which does not support the theory that pugs are possibly diagnosed with portosystemic shunts at an older age. Necrotizing meningoencephalitis as a contributing factor to postoperative mortality is intriguing because this disease specifically affects pugs. It is possible that the neurologic signs such as seizures, circling, and head pressing seen postoperatively could actually be due to worsened encephalitis after anesthesia rather than being related to the portosystemic shunt attenuation. Unfortunately, with the limited number of necropsies available, we were unable to find any signs of this disease in the pugs who died in this series. Another theory specifically of interest with pugs is the theory of postoperative hypoxia leading to seizures.⁵ As a brachycephalic breed, pugs are predisposed to postoperative hypoxia, which, combined with hepatic encephalitis, could increase the risk of seizures. Future studies could consider measuring oxygenation via serial blood gas measurements in pugs after surgical attenuation of portosystemic shunts to see if level of hypoxia plays a role in development of postoperative neurologic complications.

Limitations of this study include the retrospective nature of the study, as well as the small population of patients evaluated. Additionally, the use of a control group of dogs included in a prior study rather than evaluating all other dogs undergoing surgical attenuation of a portosystemic shunt during the study timeframe could have potentially altered the results. Also, these cases were performed prior to the standard use of levetiracetam as an anticonvulsant prior to surgical intervention for portosystemic shunts.¹³ It is unclear if the use of this medication would have changed the outcome in any of these patients.

Conclusion

Pugs may be at an increased risk of having fatal neurologic complications after surgery to correct single CEPSS. This risk should be taken into consideration when presenting owners with treatment options and when performing this procedure. Preoperative antiseizure medications should be strongly considered for dogs undergoing portosystemic shunt attenuation. Further studies are indicated to determine the reasons behind the increased risk found in this study, as well as to determine any factors that may indicate which pugs are at a higher risk for postoperative complications.