Retrospective Evaluation of Complications Associated With Surgically Placed Gastrostomy Tubes in Dogs (2010–2020): 133 Cases

1 Introduction

Early enteral nutrition is essential in hospitalized and critically ill patients and is associated with reduced mortality rates and improved outcomes [1-4]. Enteral nutrition provides nutrients through a functional gastrointestinal tract and results in improved gastrointestinal blood flow, motility, and stimulation of growth factors that attenuate systemic inflammation [4-8]. These effects lower the risk of bacterial translocation due to reduced intestinal permeability, attenuated mucosal atrophy, and improved intestinal barrier function [4-7] Enteral nutrition may be provided through temporary feeding tubes, such as nasogastric or nasoesophageal tubes, or through longer-term feeding options, including esophagostomy tubes, gastrostomy tubes (G-tubes), and jejunostomy tubes. G-tubes are generally well-tolerated and can be placed endoscopically or surgically [10]. They are beneficial for long-term nutritional support and allow for the feeding of blended commercial diets [10-12]. Surgically placed G-tubes have the advantage of direct visualization during placement and improved adhesion between the stomach and body wall in comparison to endoscopically placed G-tubes [11].

G-tubes facilitate the provision of adequate nutrition and a positive owner experience; however, complications such as stoma site infection or secondary septic peritonitis have been reported [1, 13]. A previous study evaluating G-tubes in dogs with septic peritonitis found a 33.3% complication rate, with 75% of complications being minor and not requiring tube removal [1]. Complications have previously been associated with systemic findings, including albumin concentrations [1]. Hypoalbuminemia delays healing time, with some studies showing an increased risk of gastrointestinal dehiscence and increased morbidity and mortality [14].

The purpose of the current study was to evaluate the incidence of surgically placed G-tube complications and outcomes in a population of dogs with diverse disease processes. Secondary objectives were to evaluate the association between the primary disease process, serum albumin concentration, total plasma protein concentration, and complications. We hypothesized that G-tubes would be associated with a low frequency of complications.

2 Materials and Methods

3 Results

A total of 138 dogs with surgically placed G-tubes were identified through the computerized search. No jejunostomy tubes were placed during this study period. Five cases were excluded based on death or euthanasia within the first 24 h of hospitalization, leaving 133 cases in the study. Sixty-one dogs were spayed females, 55 were neutered males, 10 were intact males, and seven were intact females. The most common breeds included mixed breed (n = 20), Golden Retriever (n = 9), Labrador Retriever (n = 9), Cocker Spaniel (n = 8), Yorkshire Terrier (n = 7), Beagle (n = 5), Chihuahua (n = 5), Schnauzer (n = 5), and Cairn Terrier (n = 5), as well as 33 other various dog breeds with fewer than five individuals each. The median weight was 10.1 kg (IQR: 6.3–19.7 kg). The decision to place a G-tube was made by the clinician based on a dog's nutritional status before surgery, concern for nutritional status after surgery, and the intention to provide enteral nutrition after surgery. Primary disease processes included hepatobiliary disease (n = 55), prolonged anorexia (n = 25), septic peritonitis (n = 22), esophageal disease (n = 15), genitourinary tract disease (n = 9), and other diseases (n = 7; Table 1).

All G-tubes were polyurethane tubesb; no low-profile tubes were used. Tube size was recorded in 60 dogs, with a median of 21-French (range: 10–28 French), which was the most commonly used. During laparotomy, all tubes were placed in the body of the stomach and sutured to the left body wall by a board-certified surgeon or surgical resident. Internal fixation was performed in all dogs with a purse-string and interlocking box pattern using 2-0 or 3-0 polydioxanone or poliglecaprone sutures. All dogs had a finger trap performed with nylon sutures as a method of external fixation. Tubes were in place a minimum of 24 h before use, and all dogs received a commercial, blended diet through the G-tube. Individual feeding plans were at the clinician's discretion. G-tubes were in place for an average of 15 days (IQR: 13–25 days).

Nine dogs (6.7%) experienced a complication associated with the surgically placed G-tube, while 124 dogs (93%) had no complications noted. The most common complication was stoma site infection or inflammation, which was observed in eight dogs (6%). All dogs with a stoma site infection or inflammation were treated with topical cleaning of the stoma site with chlorhexidine or iodine for a median of 3 days (range: 2–7). No G-tube cultures were performed, and no systemic antimicrobials were administered. The median time to stoma site infection or inflammation was 6 days (range: 4–13 days). No dogs required additional procedures to manage stoma site infections or inflammation, and no G-tube was removed prematurely for these reasons. One dog (0.7%) developed septic peritonitis secondary to gastrointestinal leakage diagnosed 5 days after G-tube placement that required surgical revision through a midline exploratory laparotomy. The dog had primary liver disease, and the diagnosis of abdominal tube leakage was made based on clinical status and the presence of abdominal effusion with intracellular bacteria. Leakage between the abdominal wall and stomach at the site of the G-tube was found during surgical exploration. The previously placed G-tube was removed, and the stomach was closed in a routine fashion, with an active drain placed in the abdomen. A new G-tube was placed that exited the left body wall at a different location, with three purse-string sutures and an interlocking box suture placed out of concern for leakage due to the fragility of the stomach wall. The abdomen was flushed and closed in a routine fashion, and the dog was treated with systemic antimicrobials, with no cultures performed. The dog survived to discharge.

Regarding the overall survival of this patient population, 116 dogs (87%) survived to discharge, 10 (7.5%) dogs died, and seven (5.3%) were euthanized. There was no association between the type of underlying disease process and the development of a G-tube complication (P = 0.915). Serum albumin and total plasma protein concentrations noted within 6 h of surgical exploration and G-tube placement were compared between dogs with and without complications, and no association was found. Median serum albumin concentrations in dogs with complications were 2.95 g/dL (IQR: 1.85–3.45 g/dL), compared to 2.6 g/dL (IQR: 2.1–3.2 g/dL) in dogs without complications (P = 0.615). Median total plasma protein concentration in dogs with complications was 6.45 g/dL (IQR: 4.6–6.65 g/dL) compared to 5.2 g/dL (IQR: 4.4–6.1 g/dL) in dogs without complications (P = 0.615).

4 Discussion

G-tubes can be useful for early enteral nutrition and medication administration. The current study found a low in-hospital complication frequency of surgically placed G-tubes in dogs with a variety of primary disease processes. Most complications observed in this study were mild, requiring topical therapy only, and did not necessitate premature tube removal. However, one dog did develop septic peritonitis secondary to G-tube placement that required surgical intervention. No dog died or was euthanized due to surgically placed G-tube complications.

Enteral nutrition is important to maintain gastrointestinal barrier function, reduce inflammation, and promote a less negative energy balance and albumin production [2, 4]. Therefore, both institutions in the current study are proactive in providing assisted enteral nutrition early in the course of hospitalization if patients are not willing or able to voluntarily consume adequate calories. During biliary surgery, septic peritonitis, and some other surgical conditions, the surgery team is already accessing the abdominal cavity, and relatively little surgical time is added by placing a G-tube in these cases. An esophagostomy tube would also be a reasonable choice in many of these patients; however, esophagostomy tube placement would require repositioning the patient, preparing a second surgical site, and ensuring correct placement with postprocedure radiography. Additional benefits of G-tubes include the ability to evacuate the stomach in situations where poor gastrointestinal motility and subsequent regurgitation are a concern, such as after biliary surgery or septic peritonitis [1].

The most common primary disease process in the current study was hepatobiliary disease and biliary surgery. Biliary surgery is associated with high mortality in dogs, up to 73% in some studies, and pancreatitis is one of the most common complications seen in the postoperative period [15-17]. The second most common disease process was septic peritonitis, which is also associated with high mortality rates [1, 4]. Based on the concern associated with high mortality rates and the importance of enteral nutrition in the postoperative period, it is common practice at our institutions to place G-tubes in ill dogs undergoing surgery for biliary disease or septic peritonitis.

The current study evaluated in-hospital complications associated with surgically placed G-tubes in a population of dogs with diverse underlying conditions. G-tubes were most commonly placed at the time of abdominal exploratory surgery. No association was found between G-tube complications and primary etiology, although the overall incidence of complications was relatively low (6.7%), and most complications were minor.

Other studies have investigated G-tube complications in single disease processes or in relatively small numbers of patients [1, 11, 18, 19]. In a previous study of 12 dogs with surgically placed G-tubes, the most common disease processes included esophageal disease (33.3%), anorexia (16.7%), cholangiohepatitis (12.7%), and septic peritonitis, skull trauma, and gastric dilatation and volvulus (8.3% each) [11]. Of these dogs, one mild complication and one complication of moderate severity resulted in an overall complication frequency of 16.7% for surgically placed G-tubes. No severe complications were noted [11]. Two studies have evaluated complications of surgically placed G-tubes in dogs with septic peritonitis. One found a minor complication rate of 26% within 21 days of placement, while another found a minor complication rate of 25% during hospitalization; in both studies, discharge around the G-tube site was the most common problem observed [1, 19]. Major complications requiring surgical intervention were noted in one of the studies, with an incidence of 8.3% [19]. Two dogs had migration of the G-tube out of the stomach, resulting in leakage of food into the subcutaneous space [1]. The incidence of major complications, specifically septic peritonitis, in the current study was 0.75%, lower than previously reported. This difference in complication frequencies may be due to the diverse patient population or number of patients evaluated.

Intestinal dehiscence has been associated with preexisting peritonitis; however, other studies noted no association between peritonitis and the risk of dehiscence [20-22]. The current study did not find an association between primary disease, including septic peritonitis, and the development of in-hospital G-tube complications. Previous studies have evaluated preoperative albumin concentrations and outcomes in septic peritonitis; however, a possible association between serum albumin concentration and G-tube complications has not been previously evaluated [1]. While dogs in the current study had serum albumin and plasma total protein concentrations recorded within 6 h of G-tube placement, no association was found between presenting serum albumin or total plasma protein concentrations and subsequent G-tube complications.

The main limitation of the study is its retrospective nature, which led to a lack of standardization regarding patient selection and G-tube placement. Additionally, details such as G-tube size and postoperative monitoring were not available in all records. While the dogs’ primary underlying conditions were evaluated, the degree of peritonitis at the time of G-tube placement could not be determined due to the retrospective nature of the study. Although serum albumin concentration was evaluated and was not associated with increased complications, the median serum albumin concentration was 2.6 g/dL, and the overall frequency of complications was low. This study could not determine whether dogs with more severe hypoalbuminemia are at increased risk of developing more complications with G-tube placement, partially due to the low number of complications. Additionally, this study only looked at in-hospital complications, and longer-term follow-up may be associated with more complications.

In conclusion, this study noted a low in-hospital complication frequency associated with surgically placed G-tubes in a diverse population of dogs. Nearly all complications were minor, requiring topical treatment alone. Surgically placed G-tubes provide a means for early enteral nutrition and easier enteral medication administration in postoperative patients; therefore, they should be considered for critically ill patients undergoing laparotomy and those deemed poor candidates for endoscopically placed tubes.

Author Contributions

Miranda Buseman: conceptualization, formal analysis, writing—review and editing. Bianca Reyes: Formal analysis, writing—review and editing. Valery Scharf: Formal analysis, writing—review and editing. Lingnan Yuan: Formal analysis, writing—review and editing.

Conflicts of Interest

The authors declare no conflicts of interest.