Gastrostomy Site Infections: A Review in Patients With Cancer
Jordan N. Edwards, BS
Department of Biochemistry, University of Florida, Gainesville, Florida
Fabiola Rios de Choudens, MD
Resident, Internal Medicine, University of South Florida, Moffitt Cancer Center and Research Institute, Tampa, Florida
Chandrashekar Bohra, MBBS
International Resident, Moffitt Cancer Center and Research Institute, Tampa, Florida
Ganesh Gajanan, MBBS
International Resident, Moffitt Cancer Center and Research Institute, Tampa, Florida
John N. Greene, MD
Chief of Infectious Diseases and Hospital Epidemiologist, Moffitt Cancer Center and Research Institute, Tampa, Florida
Edwards JN, Rios de Choudens F, Bohra C, Gajanan G, Greene JN. Gastrostomy site infections: a review in patients with cancer. Consultant. 2019;59(5):131-134.
ABSTRACT: This review highlights the cases of 10 patients with cancer and with culture-confirmed gastrostomy tube site infections. The main causative organisms were gram-negative bacteria, the most common of which were Pseudomonas aeruginosa, Achromobacter xylosoxidans, Enterobacter aerogenes, and Enterobacter cloacae, along with gram-positive Staphylococcus aureus. Only one patient had methicillin-resistant S aureus (MRSA) isolated from his cultures, and he had no history of MRSA colonization. Fungi (Candida albicans) were isolated in 5 of 10 patients, of whom only 2 had significant candidal growth on culture. Weight loss, a low albumin level, and prolonged neutropenia were the most common associated risk factors in the 10 patients. Most of the gram-positive infections were treated with amoxicillin-clavulanate, and most gram-negative infections were treated with levofloxacin, with a satisfactory response.
KEYWORDS: Gastrostomy tube (GT) infection, percutaneous endoscopic gastrostomy (PEG), antibiotics
Percutaneous endoscopic gastrostomy (PEG) is widely used to maintain enteral nutrition in patients who are unable to swallow. The most common complication of PEG tube placement is infection at the PEG tube site. Peristomal irritation at the site occurs commonly, but few infections require antimicrobial therapy or surgical management. The standard pull technique for PEG insertion is associated with infections in 5% to 30% of cases.1 In a hospital that is endemic for methicillin-resistant Staphylococcus aureus (MRSA), it can be the most common organism associated with these infections. In view of the paucity of literature on PEG infections in patients with cancer, this retrospective review describes infections after PEG tube placements at Moffitt Cancer Center in Tampa, Florida.
The main aims of this study were (1) to identify the most common pathogens found within PEG tube infections; (2) to identify the treatments to which these organisms responded; and (3) to identify the risk factors that might have led to these infections.
At Moffitt Cancer Center, 205 patients who received PEG tubes from January 1, 2008, to December 31, 2008, were considered for the review. Of the 205 patients, only 10 (<5%) were found to have PEG tube site infections. These patients’ cases were then further investigated for microbiology and susceptibility of pathogens and predisposing factors to infection.
The inclusion criteria for patients in this study were as follows: age greater than 19 years, culture-proven gastrostomy tube (GT) site infection, and a current or previous cancer diagnosis.
In the 1-year study period, 10 of the 205 patients (<5%) who received GTs at Moffitt Cancer Center became infected at the insertion site. Gram-negative organisms were the main causative agents, with the most common being Pseudomonas aeruginosa, Achromobacter xylosoxidans, Enterobacter aerogenes, and Enterobacter cloacae, along with gram-positive Staphylococcus aureus. Only one patient had MRSA isolated from his cultures, and he had no history of MRSA colonization. Fungi (Candida albicans) were isolated in 5 of 10 patients, of whom only 2 had significant candidal growth.
All patients with bacterial infection were initially treated empirically with cephalexin. Depending on further microbiology and culture test results, patients with gram-negative bacterial infection were switched to ciprofloxacin, levofloxacin, or cefepime, depending on sensitivities. Those with gram-positive infections required topical mupirocin at the infection site and amoxicillin-clavulanate. MRSA infection was treated with daptomycin, and no particular treatment was necessary for fungal infections. None of the patients received antimicrobial prophylaxis after PEG tube placement.
Even though most of the noninfected patients (195 of 205) did receive chemoradiation and corticosteroids for cancer during this period, the most significant risk factor for PEG tube infections was weight loss—7 of the 10 infected patients had significant weight loss prior to infection, and 2 had low albumin levels. One of the 10 had a history of longstanding neutropenia. Weight loss and low albumin levels might have resulted in not only the immunological impairment that might have led to the infections, but also the loosening of the external bumper (separation from the skin), leading to entry of pathogens. All of our patients with PEG site infection had loosening of the flap or bumper from the GT site.
The general complications associated with PEG tube insertion are aspiration,1 peritonitis,2 hemorrhage,3 gastrocutaneous fistulas,4 and infections.5 The GT site is associated with infections in 5% to 30% of cases.1,6 The presence of wound infection is associated with an increase in morbidity and mortality.7,8 In their study of GT site infections in noncancer patients, Pien and colleagues reported an infection rate of 4.8%, and the most common infecting organisms were staphylococci, gram-negative bacteria, and yeast.9 Cancer patients have the additional burden of chemotherapy and in few cases concomitant corticosteroid therapy, which predisposes them to various infections.10 Rolston and colleagues described the microbiology of PEG tube insertion sites in cancer patients, with the most common organisms isolated being Candida species, S aureus, and P aeruginosa.11 In that study, all infections were local infections, meaning cellulitis, complicated skin, and skin structure infections including abdominal wall abscesses. These findings were similar to those of our study, in which Pseudomonas and Enterobacter species were involved in most cases, along with S aureus and Candida species. There was also the presence of A xylosoxidans, E aerogenes, and E cloacae in 4 of 10 patients, which was unique to our study.
With regard to MRSA infection, Mainie and colleagues concluded that patients with prior MRSA colonization had a significantly higher risk of developing symptomatic MRSA infection of the PEG site.8 However, there was still a significant risk (15%) of developing MRSA infection at the PEG site for patients with no known prior MRSA colonization.8 At least 2 studies have found that nasopharyngeal decontamination of patients with MRSA (in addition to standard prophylactic antibiotics) significantly reduced the incidence of wound infections.12,13 Another study found that administration of a third-generation cephalosporin intravenously and a povidone-iodine spray to the abdominal wall prior to the procedure reduced the rate of wound infections compared with intravenous cephalosporin or povidone-iodine spray used separately.14 Studies suggest that antibiotic prophylaxis significantly reduces the risk of peristomal wound infection associated with GT insertion.15 Antibiotic prophylaxis, therefore, may be considered as a general measure in preventing infection following PEG tube placement.16 None of our patients were on any prophylaxis following the GT insertion per usual guidelines.
A major complication of GT site infections is necrotizing fasciitis.17,18 Most of our patients were placed empirically on cephalexin after infection was suspected and prior to culture results. For gram-positive organisms, amoxicillin-clavulanate and mupirocin were used. Antibiotic selection was based on previous studies on gram-positive infections around gastrostomy sites in which β-lactam antibiotics were used; in the case of MRSA infections, vancomycin, daptomycin, or linezolid were used.19,20 For gram-negative organisms such as Pseudomonas and Enterobacteriaceae around the surgical site, Goswami and colleagues found that P aeruginosa was sensitive to ciprofloxacin (83.78%) and meropenem (51.35%), and that Escherichia coli was sensitive to levofloxacin (72.41%) and ciprofloxacin (62.07%).21 Ciprofloxacin has been used successfully for the treatment of P aeruginosa GT site infections.22 Mahmood reported that 80% of Pseudomonas and Enterobacteriaceae causing PEG tube site infections were resistant to aminoglycosides, especially gentamicin.23
Distinguishing colonization of normal skin and gastrointestinal tract flora from infection with a true pathogen is difficult and relies heavily on clinical judgement. Deep-tissue biopsy with quantitative cultures is not recommended unless surgery is required. It is unclear whether all culture-proven pathogens must be included in the coverage of the antibiotic selection, but an antibiotic with activity against the culture isolate should be favored. However, because susceptibility results can take days to become available, if the infection is improving, the current antibiotic should not be changed to include activity against the cultured isolate.
There are a number of risk factors for the development of GT site infections. Among the common causes of surgical site infections are existing infection, low serum albumin concentration, older age, obesity, smoking, diabetes mellitus, and ischemia secondary to vascular disease or irradiation.24 The most common predisposing factors in our study were weight loss, low albumin levels, and prolonged neutropenia. Another important factor we found to be present in all of our patients was loosening of the flap or bumper from the GT site. This has also been cited as an important complication of GT site insertion due to leaking of contents onto the skin or internal structures.25 Skin maceration and irritation from moisture or gastric secretions can produce chemical-induced dermatitis and secondary site infection.
Table 1 compares the organisms isolated from the GT site in our study with those reported in the literature. Table 2 compares treatment administered for the involved organisms.
Table 1. Comparison of Common Organisms Found at the GT Site
Common Organisms According to the Literature
Common Organisms in Our Study
Pseudomonas aeruginosa, Proteus mirabilis, Escherichia coli, Klebsiella pneumoniae
Pseudomonas aeruginosa, Enterobacter cloacae, Achromobacter xylosoxidans, Enterobacter aerogenes
Enterococcus species, Staphylococcus aureus
60% of persons with MRSA colonization of the skin before PEG tube insertion develop MRSA infection of the PEG site
Only 1 of our patients had MRSA colonization
Candida species are the most commonly described organism
Table 2. Treatment Comparison for GT Site Infections
Traditional Treatment According to the Literature
Treatment in Our Study
Ciprofloxacin and other fluoroquinolones
Ciprofloxacin, levofloxacin, and cefepime
Nafcillin and clindamycin
Mupirocin and amoxicillin-clavulanate
Vancomycin and teicoplanin
Topical antifungal treatment for 2 weeks (oral fluconazole for severe Candida infections)
No particular treatment given
Infection at the GT site is an important complication in patients with cancer and more so in such patients with associated risk factors. Antibiotic prophylaxis and appropriate wound care may reduce GT insertion site infection. If infection develops, prompt recognition is key, and culture of the area should be obtained. The choice of treatment, whether topical or systemic antibiotics, should be individualized and guided by the severity of the wound and the culture results. Patients with risk factors such as weight loss, low albumin, and immunosuppression should be followed closely for evidence of local wound infection. Finally, loosening of the flap or bumper from the GT site may be the cause of the irritation and infection and may need adjustment to prevent gastric secretion leakage.
- Foutch PG. Complications of percutaneous endoscopic gastrostomy and jejunostomy: recognition, prevention and treatment. Gastrointest Endosc Clin N Am. 1992;2(2):231-248.
- Foutch PG, Woods CA, Talbert GA, Sanowski RA. A critical analysis of the Sacks-Vine gastrostomy tube: a review of 120 consecutive procedures. Am J Gastroenterol. 1988;83(8):812-815.
- Larson DE, Burton DD, Schroeder KW, DiMagno EP. Percutaneous endoscopic gastrostomy: indications, success, complications, and mortality in 314 consecutive cases patients. Gastroenterology. 1987;93(1):48-52.
- Patwardhan N, McHugh K, Drake D, Spitz L. Gastroenteric fistula complicating percutaneous endoscopic gastrostomy. J Pediatr Surg. 2004;39(4):561-56
- Gençosmanoglu R, Koç D, Tözün N. Percutaneous endoscopic gastrostomy: results of 115 cases. Hepatogastroenterology. 2003;50(51):886-888.
- Safadi BY, Marks JM, Ponsky JL. Percutaneous endoscopic gastrostomy. Gastrointest Endosc Clin N Am. 1998;8(3):551-568.
- Ho C-S, Yeung EY. Percutaneous gastrostomy and transgastric jejunostomy. Am J Roentgenol. 1992;158(2):251-25
- Mainie I, Loughrey A, Watson J, Tham TC. Percutaneous endoscopic gastrostomy sites infected by methicillin-resistant Staphylococcus aureus: impact on outcome. J Clin Gastroenterol. 2006;40(4):297-300.
- Pien ECT, Hume KE, Pien FD, Gastrostomy tube infections in a community hospital. Am J Infect Control. 1996;24(5):353-358.
- Gea-Banacloche JC, Segal BH. Infections in the cancer patient. In: DeVita VT Jr, Lawrence TS, Rosenberg SA, eds. Cancer: Principles and Practice of Oncology. 9th ed. Philadelphia, Pa: Wolters Kluwer Lippincott Williams & Wilkins; 2011:2262-2299.
- Rolston KVI, Mihu C, Tarrand JJ. Current microbiology of percutaneous endoscopic gastrostomy tube (PEG tube) insertion site infections in patients with cancer. Support Care Cancer. 2011;19(8):1267-1271.
- Thomas S, Cantrill S, Waghorn DJ, McIntyre A. The role of screening and antibiotic prophylaxis in the prevention of percutaneous gastrostomy site infection caused by methicillin-resistant Staphylococcus aureus. Aliment Pharmacol Ther. 2007;25(5):593-597.
- Radhakrishnan NV, Shenoy AH, Cartmill I, et al. Addition of local antiseptic spray to parenteral antibiotic regimen reduces the incidence of stomal infection following percutaneous endoscopic gastrostomy: a randomized controlled trial. Eur J Gastroenterol Hepatol. 2006;18(12):1279-1284.
- Naik RP, Joshipura VP, Patel NR, Haribhakti SP. Complications of PEG—prevention and management. Trop Gastroenterol. 2009;30(4):186-194.
- Gossner L, Keymling J, Hahn EG, Ell C. Antibiotic prophylaxis in percutaneous endoscopic gastrostomy (PEG): a prospective randomized clinical trial. Endoscopy. 1999;31(2):119-124.
- Lipp A, Lusardi G. A systematic review of prophylactic antimicrobials in PEG placement. J Clin Nurs. 2009;18(7):938-948.
- Greif JM, Ragland JJ, Ochsner MG, Riding R. Fatal necrotizing fasciitis complicating percutaneous endoscopic gastrostomy. Gastrointest Endosc. 1986;32(4):292-294.
- Cave DR, Robinson WR, Brotschi EA. Necrotizing fasciitis complicating percutaneous endoscopic gastrostomy. Gastrointest Endosc. 1986;32(4):294-296.
- Stevens DL. Treatments for skin and soft-tissue and surgical site infections due to MDR Gram-positive bacteria. J Infect. 2009;59 suppl 1:S32-S39.
- Nichols RL, Florman S. Clinical presentations of soft-tissue infections and surgical site infections. Clin Infect Dis. 2001;33 suppl 2:S84-S93.
- Goswami NN, Trivedi HR, Goswami APP, Patel TK, Tripathi CB. Antibiotic sensitivity profile of bacterial pathogens in postoperative wound infections in a tertiary hospital in Gujarat, India. J Pharmacol Pharmacother. 2011;2(3):158-164.
- Obritsch MD, Fish DN, McLaren R, Jung R. National surveillance of antimicrobial resistance in Pseudomonas aeruginosa isolates obtained from intensive care unit patients from 1993 to 2002. Antimicrob Agents Chemother. 2004;48(12):4606-4610.
- Mahmood A. Bacteriology of surgical site infections and antibiotic susceptibility pattern of the isolates at a tertiary care hospital in Karachi. J Pak Med Assoc. 2000;50(8):256-259.
- Cheadle WG. Risk factors for surgical site infections. Surg Infect (Larchmt). 2006;7 suppl 1: S7-S11.
- Arrowsmith H. Nursing management of patients receiving gastrostomy feeding. Br J Nurs. 1996;5(5):268-273.