Short- and Long-Term Outcome of Dogs Following Surgical Correction of a Persistent Right Aortic Arch

Introduction

Vascular ring anomalies lead to esophageal constriction secondary to abnormal aortic arch development and are a common cause of regurgitation in young dogs. Clinical signs are typically seen shortly following weaning in most dogs, although late-onset regurgitation has been reported.^1,2 Approximately 95% of vascular ring anomalies leading to clinical signs consist of a persistent right aortic arch (PRAA) with a left ligamentum arteriosum, but other anomalies have been reported.^3–16 Purebreed dogs considered at risk include German shepherd dogs and Irish setters. A genetic predisposition has been shown in German shepherd dogs and has been investigated in greyhounds.^17–19 Currently, surgical treatment is recommended in animals with a PRAA because medical therapy alone is typically unrewarding.^20,21

Two previous studies have described the outcome of dogs discharged from the hospital following surgical treatment of a PRAA with diverging results. The study by Shires and Liu (1981) reported only 9% of dogs to be completely free of any episodes of regurgitation and 76% of dogs to have an overall acceptable outcome to the owners. The survival rate reported in that study was 80% at 3 wk following surgical treatment. Twenty-four percent of dogs that survived, and for which long-term follow-up was available, were euthanized because of continued clinical signs and complications.¹⁸ That finding is in contrast to the study by Muldoon et al. (1997) that found that 92% of dogs had an excellent outcome with no regurgitation following consumption of solid food. All of the dogs surgically treated were reported to have a long-term outcome acceptable to the owners in that study. Three out of 28 dogs with known outcome died within 2 wk of surgical treatment and no dogs died or were euthanized after that time.²²

In the authors’ experience, dogs that survive for 2 mo following surgery have a good quality of life, but the majority of dogs do not return to completely normal function. This study was performed to help clarify the conflicting results in the current literature regarding outcome following surgical correction of a PRAA. The purpose of this study was to report the survival to discharge, postdischarge survival, and long-term outcome of dogs following surgical treatment of a PRAA and to identify potential risk factors for death prior to discharge from the hospital. The authors’ hypothesis was that a young age at presentation, male gender, and belonging to the German shepherd breed would lead to increased odds for death prior to discharge from the hospital. It was also hypothesized that the majority of dogs surviving long-term would have a good or excellent outcome, but would continue to display some degree of clinical signs or require ongoing dietary modification and that owner satisfaction would be high for long-term survivors.

Materials and Methods

Medical records of all dogs examined at the Iowa State University, Colorado State University, and University of Illinois Veterinary Teaching Hospitals between August 1987 and December 2009 were reviewed. Only animals with a surgically confirmed PRAA with a left ligamentum arteriosum were included in the study. Survival to discharge and postdischarge survival evaluated mortality rate prior to and following discharge from the hospital, respectively. Long-term outcome was defined as outcome of animals surviving at least 2 mo following discharge from the hospital. Animals were included in the short-term survival analysis if medical records were available for review. Animals were included for analysis of postdischarge survival and long-term outcome if the owners could be reached via telephone or electronic correspondence to administer a survey following surgical treatment or if medical records were available documenting death or euthanasia following discharge from the hospital.

All animals were surgically treated through a fourth or fifth intercostal thoracotomy or via thoracoscopy performed by either a Diplomate of the American College of Veterinary Surgeons or by a resident under supervision of a Diplomate. The left ligamentum arteriosum was ligated with either suture or hemoclips and transected. Remaining bands restricting the esophagus were dissected. Additional intraluminal esophageal dilation and placement of a feeding tube was performed at the discretion of the surgeon. Signalment, body weight, surgical approach, and information regarding placement of a feeding tube were obtained from the medical record. Postoperative feeding recommendations were not standardized, but elevated feeding of a blenderized or liquid diet was recommended in all cases for varying duration.

To determine short-term survival, death prior to discharge from the hospital following surgical treatment was noted. Cause of death and necropsy findings were recorded if available.

Postdischarge survival was determined based on owner survey responses and medical record evaluation. Death or euthanasia at any time following surgical treatment was recorded if made apparent by the medical record or owner survey and survival times and follow-up time were determined. Cause of death and necropsy findings were recorded if available.

Long-term evaluation of outcome in dogs surviving > 2 mo following surgical treatment was based on owner surveys. The owner survey included questions regarding clinical signs prior to and following surgery, including frequency of regurgitation and the need for dietary modifications at different time points. Dogs were considered to require dietary modifications if they had a continued need for elevated feeding and were unable to eat dry dog food without further modifications, such as moistening, mixing with canned food, or soaking in water prior to feeding. Dogs only tolerating canned food were considered to require continued diet alteration. Based on owner responses, the earliest and latest time of improvement of regurgitation frequency was determined. In addition, owners were asked to rate their own satisfaction with the long-term outcome on a scale of 1 (very unsatisfied) to 6 (very satisfied) and were asked if they would have the surgery performed again based on their knowledge and experience at the time of the survey.

Long-term outcome was categorized as excellent, good, or poor, based on the owners’ responses to survey questions. An excellent outcome was assigned to dogs that no longer displayed any regurgitation, did not require any dietary modification, and did not have a relapse of clinical signs at any time. Dogs requiring continued modification or restriction of food type and/or showing continued regurgitation with a frequency of < 1/wk were classified as having a good outcome. Dogs were assigned to the poor outcome group if they had continued or recurring regurgitation daily to weekly regardless of diet and/or death or euthanasia was a direct result of clinical signs associated with a PRAA or as a result of surgical complications. Complications considered attributable to a PRAA included regurgitation (either persistent or recurring) and aspiration pneumonia. Subjective owner satisfaction with long-term results was not considered during the above-described classification process.

Results

Fifty-two dogs were included in the analysis of short-term outcome. Purebreeds represented more than once included the German shepherd dog (n = 12), Labrador retriever (n = 8), greyhound (n = 4), cocker spaniel (n = 2), Welsh springer spaniel (n = 2), Jack Russell terrier (n = 2), and Australian cattle dog (n = 2). Eight dogs were mixed-breeds, and the remaining 12 dogs represented 1 of each of the following breeds: beagle, Boston terrier, boxer, Dalmatian, field spaniel, fox terrier, golden retriever, Great Dane, Portuguese water dog, Weimaraner, Welsh terrier, and wheaten terrier. Thirty-five dogs were female, 16 dogs were male, and the gender of 1 dog was not recorded. Median age at the time of presentation was 2.5 mo (mean, 4.2 mo; range, 1–72 mo) and median weight was 3.4 kg (mean, 4.5 kg; range, 0.7–24.3 kg). All dogs had clinical signs of regurgitation > 1/day, and 1 dog had aspiration pneumonia at the time of presentation and was treated until it resolved prior to surgery. Forty-four dogs were treated through a thoracotomy and 8 dogs underwent thoracoscopy. Four dogs required conversion from a thoracoscopic procedure to an open thoracotomy. A gastrostomy feeding tube was placed in 5 dogs at the time of surgical correction of the PRAA, and 2 additional dogs received feeding tubes within 7 days of the initial surgery and prior to initial discharge from the hospital.

Forty-eight dogs (92%) survived to the time of discharge. Two dogs died of cardiac arrest shortly following surgery, 1 dog died following development of aspiration pneumonia, and 1 dog was euthanized after developing aspiration pneumonia A necropsy was performed in the dog euthanized following development of aspiration pneumonia showing severe purulent hemorrhagic bronchopneumonia as well as severe locally extensive necropurulent ulcerative esophagitis. Necropsies were not performed in the other 3 dogs. Based on logistic regression analysis, neither age at presentation, gender, nor belonging to the German shepherd breed increased the odds of death prior to discharge from the hospital.

Either medical records or owner survey results were available for 28 of the surviving dogs. The remaining 20 dogs were lost to follow-up. Five of the 28 dogs (18%) were euthanized within 2 mo following surgery. Three of those 5 dogs had no improvement of regurgitation and the remaining 2 dogs showed initial improvement but had a relapse of regurgitation at 2 wk and 7 wk following surgery, respectively. Three dogs developed aspiration pneumonia, and 2 dogs developed distant infection. Necropsies were performed in four of the 5 dogs, confirming the presence of bronchopneumonia in 2 dogs and documenting distant infection in 2 dogs. One dog with distant infection developed osteomyelitis of the seventh lumbar vertebra and the sacrum and 2 dogs developed sepsis with α-hemolytic Streptococcus spp. isolated from the liver and spleen during necropsy.

Owner surveys regarding long-term outcome among survivors were available for 23 dogs. Seventeen dogs were alive at the time of the study and 6 dogs were dead. Five dogs had been euthanized or died for unrelated reasons and 1 dog had been euthanized because of regurgitation and recurring aspiration pneumonia. Median follow-up time was 60 mo (mean, 69 mo; range, 7–164 mo).

Nineteen owners reported that their dogs never regurgitated with their regular diet at the time of follow-up. Four of those owners indicated that their dog would regurgitate at infrequent intervals if their dog ate too fast (n = 1) or ate food other than their regular diet (n = 3). The survey results pertaining to timing of improvement of regurgitation, degree of improvement, dietary tolerance, and residual frequency of regurgitation are shown in Table 1. Clinical outcome was scored as excellent in 7 dogs (30%), good in 13 dogs (57%), and poor in 3 dogs (13%). Of the dogs with poor clinical long-term outcome, 1 dog improved transiently with a decreased frequency of regurgitation during the first mo postoperatively but commenced regurgitating several times daily thereafter. That dog required continued medical management with diet modification and elevated feeding 21 mo following surgical treatment at the time of the study. One dog showed persistent regurgitation a few times/wk but < 1/day while being fed a small kibble dry diet. The third dog with poor clinical outcome showed resolution of clinical signs (never regurgitated, tolerated dry food) with an excellent outcome initially, but suffered a relapse of multiple daily regurgitations 66 mo following surgical treatment and was euthanized because of recurrent episodes of aspiration pneumonia. A necropsy was not performed in that dog. Owner satisfaction scores for long-term outcome were available for 22 dogs. Eighteen owners indicated a satisfaction score of 6 out of 6, three owners indicated a score of 5 out of 6, and one owner rated satisfaction as 1 out of 6.

TABLE 1 Summary Data for Outcome in Dogs Following PRAA Surgery

*Indicates death at time of follow-up

N/A, not applicable.

View Fullscreen

Discussion

This study found that there is a considerable perioperative risk for death following surgical correction of a PRAA with a left ligamentum arteriosum, with 8% of dogs not surviving to discharge. Further, 18% of dogs surviving to the time of discharge either died or were euthanized within 2 mo of surgical treatment due to lack of improvement or relapse of clinical signs and development of complications.

This study showed that surgical correction of a PRAA with a left ligamentum arteriosum leads to a good or excellent clinical outcome in dogs in 87% of cases surviving long-term, but complete resolution of clinical signs is seen in only 30% of all long-term survivors. Out of 23 cases with long-term follow-up, 57% exhibited either persistent mild clinical signs or required continuous medical management but had good quality of life according to owners. Those findings confirmed the authors’ hypothesis that the majority of dogs continue to exhibit clinical signs or require continued dietary modification, which is consistent with the report by Shires and Liu (1981).¹⁸ Based on the study findings, surgical treatment is highly rewarding, but does not necessarily lead to complete resolution of clinical signs. Owners need to be informed that continued medical management may be needed throughout the life of the dog. As was expected based on the study authors’ clinical experience, owner satisfaction is high despite the presence of some degree of clinical signs or the requirement for continued dietary modification.

To report the mortality rates for dogs surgically treated for a PRAA, all cases with long-term follow-up data and dogs known to have died or been euthanized related to a PRAA at any time following discharge were included. The survival rate for dogs surviving to the time of discharge was 82%; however, if all dogs known to not have survived following surgical treatment are included, the overall survival rate in this study was 21 of 32 dogs (72%), which is lower than the 80% 3 wk survival reported by Shires and Liu (1981).¹⁸ Muldoon et al. (1997) did not directly report mortality, but they did report that 3 of 47 dogs (6.4%) were euthanized within 2 wk of surgery.²² A large number of dogs were lost to follow-up after discharge from the hospital and were, therefore, not included in the analysis of postdischarge survival. It is possible that this study overestimates the mortality rate following surgical treatment of a PRAA with left ligamentum arteriosum. A dog either developing complications or not improving within 2 mo of surgery may be more likely to present for re-evaluation and be less likely to be lost to follow-up and may, therefore, create a bias toward negative outcome. All known short-term deaths occurred within 2 mo of surgical treatment. Survival beyond that time appears to carry a favorable long-term prognosis because only three long-term survivors had an overall poor outcome, although no statistical analysis regarding a cut-off for survival was performed. Of those dogs, only 1 failed to show any significant improvement. One dog was placed in the poor outcome group because of the frequency of regurgitation; however, the dog was fed a dry kibble diet and the owner did not consider it necessary to change the diet because of the perceived good quality of life and high outcome satisfaction despite the frequency of regurgitation. The study authors consider it likely that a change in diet would have led to reassignment to the good outcome category. One dog showed complete resolution of clinical signs for 5 yr following surgical treatment, followed by recurrence of regurgitation and the development of repeated aspiration pneumonia. It is possible that the clinical signs were unrelated to the initial vascular ring anomaly, and no follow-up radiographs or necropsy were available. This dog was included in the poor long-term outcome category because it fit the criteria for complications associated with a PRAA. Removal of that dog from the poor outcome category and either reassigning it into the excellent outcome category or censoring it from the long term analysis would, however, have a minor effect on the overall results regarding outcome. This is especially true if excellent and good outcome categories are treated as a single group with acceptable outcome, as owners appear to do based on the satisfaction survey.

Risk factors for survival to discharge could not be established. Shires and Liu (1981) previously reported a poorer survival rate in dogs presenting between 2 wk and 2 mo of age compared with the rest of the population, and surmised that German shepherd dogs may have a poorer prognosis than other breeds, although statistical significance was not reached.¹⁸ Those findings were not confirmed in this study, although rejection of the authors’ null hypothesis may represent a type II error.

This study has limitations similar to previous studies because it relies on owner responses rather than objective clinical parameters for outcome evaluation. In addition, owner surveys pertained to dogs that had received surgical treatment up to 22 yr prior to the time of the study, which was necessary to accumulate enough cases for meaningful statistical analysis. Long-term outcome results may, therefore, be skewed by the owners’ ability to recall details, especially regarding the timing of improvement. In the authors’ opinion, the criteria for evaluating clinical outcome (i.e., frequency of regurgitation, type of food fed, relapse of clinical signs) were both specific and broad enough to allow accurate recollection by owners.

Risk factors for death following surgical correction of a PRAA remain elusive. Diagnostic modalities and treatment were not standardized because of the retrospective nature of the study. An inclusion of an objective evaluation of the severity of clinical signs or the degree of esophageal constriction, segmental dilation, or motility for risk factor analysis would be ideal. However, fluoroscopic studies or esophagography were inconsistently available and esophageal dimension and function parameters could, therefore, not be analyzed as predictors for outcome. Similarly, surgical technique and placement of an enteric feeding tube were not analyzed as predictors of outcome. Further studies are needed to identify factors that increase the odds of mortality or predict long-term outcome in dogs undergoing surgical correction of a PRAA.

Conclusion

This study shows a mortality rate of 8% prior to discharge and an additional 18% mortality among survivors to discharge. For dogs that survive, overall owner satisfaction is high. Long-term outcome is good or excellent for 87% of dogs, but only 30% of long-term survivors are either completely free of clinical signs or do not require further dietary modifications. Risk factors for death prior to discharge were not identified.