- Email: [email protected]
Effects of gastric acidity on peristomal infection after percutaneous endoscopic gastrostomy placement
Journal of Hospital Infection 76 (2010) 42e45
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Effects of gastric acidity on peristomal infection after percutaneous endoscopic gastrostomy placement H. Ono a, *, S. Ito b, Y. Yamazaki c, Y. Otaki d, H. Otaki d a
Department of Internal Medicine, Seiwa Memorial Hospital, Sapporo, Japan Division of Internal Medicine, Tannan Regional Medical Center, Sabae, Japan c Division of Endoscopic Medicine, Faculty of Medical Sciences, University of Fukui, Eiheiji-cho, Fukui, Japan d Department of Internal Medicine, Otaki Hospital, Fukui, Japan b
a r t i c l e i n f o
s u m m a r y
Article history: Received 18 July 2009 Accepted 8 April 2010 Available online 1 July 2010
Peristomal infection is a common complication following percutaneous endoscopic gastrostomy (PEG) placement. This study investigated the effect of gastric acidity on peristomal infection, including type of bacteria and the relationship between bacteria cultured from the oropharynx and PEG tube site. Sixty-seven patients with dysphagia underwent PEG placement at Otaki Hospital between 1998 and 2001. Gastric acidity was evaluated by 24 h pH monitoring. Patients were observed for peristomal infection for two weeks after PEG placement, with specimens collected from the oropharynx and PEG tube site. Twenty-one (31.3%) of the patients who had undergone PEG insertion developed peristomal infections. Of 52 patients who were not colonised with meticillin-resistant Staphylococcus aureus (MRSA) in the oropharynx, 11 cases (21.2%) developed peristomal infection. The median gastric pH of infected patients (11 cases) was 5.05 2.55 (mean SD) and in patients without infection (41 cases) it was 3.06 1.83 (P ¼ 0.019). Peristomal infection developed in 66.7% (10/15) of patients carrying MRSA compared with only 21.2% (11/52) of patients who were not colonised by MRSA (P < 0.001). The incidence of peristomal infection was affected by gastric acidity and the presence of MRSA in the oropharynx. Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Gastric acidity Meticillin-resistant Staphylococcus aureus Percutaneous endoscopic gastrostomy Peristomal infection pH monitoring
Introduction In 1980, Ponsky and Garderer developed the percutaneous endoscopic gastrostomy (PEG) method.1,2 Nutrition is delivered through the PEG tube, which facilitates feeding and improves the nutritional status of patients with dysphagia. Although the PEG procedure is easy to perform, it sometimes results in complications such as peristomal infection, intragastric bleeding, and aspiration pneumonia.3,4 According to the European Society of Gastrointestinal Endoscopy Guidelines the most common complication is peristomal infection, which occurs in 30e43% of patients with a PEG tube.5
* Corresponding author. Address: Department of Internal Medicine, Seiwa Memorial Hospital, 1-5-1-1 Kotoni, Nishi-ku, Sapporo 063-0811, Japan. Tel.: þ81 11 611 1111; fax: þ81 11 631 6271. E-mail address: [email protected] (H. Ono).
The mechanism of peristomal infection is not clearly understood, but it may be associated with the patient’s general condition or with the PEG procedure. Risk factors include diabetes mellitus, malignant disease, malnutrition, or other factors that may potentiate the effect of peristomal infection.6 A gastrocutaneous fistula may be difficult to complete due to these diseases and may induce secondary bacterial infections. Therefore PEG placement must be performed safely and carefully in these patients. There are three methods of performing PEG: the pull, the push, and the introducer or direct method. During the pull or push method, the PEG tube is passed through the oral cavity to reach the stomach, increasing the risk of bacterial contamination. Deitel et al. reported that infection did not occur when they used the introducer method.7,8 In addition, oral bacteria will reach the stomach from the oropharynx via saliva. The gastric microbial flora is thought to be influenced by gastric acidity.9,10 Furthermore, H2-receptor antagonists affect gastric acidity and gastric microbial flora in ways that
0195-6701/$ e see front matter Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2010.04.011
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
significantly alter the risk of postoperative infection.10e13 Older patients who are at higher risk of cerebrovascular disease are more likely to receive PEG tube nutrition. In addition these patients often have low acid output due to chronic atrophic gastritis. In this study, we investigated the effect of intragastric acidity on peristomal infection during the course of PEG placement. We also investigated the type of bacteria and the relationship between bacteria cultured from the oropharynx and those from the PEG tube site. Methods Patients During a randomised controlled trial to assess the effect of ranitidin sulfate after PEG insertion, 67 patients (20 males and 47 females aged 54e95 years; mean: 80.7 9.3) who had undergone PEG due to dysphagia between January 1998 and May 2001 at Otaki Hospital were followed up for PEG site infection (Figure 1).14 Of the 67 patients, 54 had cerebrovascular disease (80.6%), five had Parkinson’s disease, and four had senile dementia (Table I). This study was performed according to the principles of the Declaration of Helsinki; the ethics committee of Otaki Hospital approved the protocol and written informed consent was obtained from all patients’ families. Plan for PEG placement We performed a secondary analysis for peristomal infection in these 67 patients. All patients underwent PEG placement by the pull method using a pull type kit (Bard Interventional Products, Billerica, MA, USA).14 Twenty-four-hour pH monitoring (Degitrapper 4 Mb; Synectics Medical, Stockholm, Sweden) was initiated soon after PEG placement. Ranitidine sulfate (100 mg/day, i.v.), an H2-receptor antagonist, was administered to 19 randomly selected patients to control gastric acidity during the two weeks after PEG. Piperacillin (2 g/day, i.v.) was administered to all patients for three days after the PEG placement. Specimens from the oral cavity and pharynx were collected for two days before the PEG placement, and
43
pus was collected from the PEG tube site three times after the PEG placement only if patients’ purulent discharge was present. The bacteria were identified at the Hokuriku Medical Laboratories. Definition of peristomal infection and patient groups Peristomal infection was defined as the presence of purulent discharge around the PEG tube within 2 weeks after PEG, including the inability to begin feeding through the PEG tube due to this infection, about which we referred to the study by Jain et al.15 Two endoscopists, who were certificated from Japan Gastroenterological Endoscopy Society, determined presence or absence of this infection. Patients were divided into two groups: the infected group and the non-infected group. Further groups were divided as follows: group A consisted of the infected subgroup excluding MRSA-positive patients; group B was composed of the non-infected subgroup excluding MRSA-positive patients. Statistical analysis Results were expressed as means SD and were analysed for statistical significance by the t-test or Fisher’s exact test. Age, nutritional conditions and the average median gastric pH of the cases in each group were compared using the t-test. Gender, cerebrovascular disease, nutritional methods, patients positive for MRSA, enterobacteria and pseudomonas of the oropharynx were compared using the Fisher’s exact test. Differences were significant at P < 0.05. Results Age, gender, disease, nutritional methods and conditions The average age of the infected group was 81.6 6.6 years and that of the non-infected group was 80.3 10.3 years (P ¼ 0.27). There were 7 males and 14 females in the infected group, and 13 males and 33 females in the non-infected group (P ¼ 0.39). Fifteen patients in the infected (71.4%) and 39 patients in the non-infected
PEG placement 2
3
4
7
14 (day)
24 h pH monitoring endoscopy peri-PEG culture (three times) pharynx culture (2 days) Piperacillin 2 g/day (i.v.) 3 days Ranitidine sulfate 100 mg/day (i.v)
19 cases
Ranitidine sulfate unused
48 cases
Figure 1. Protocol for percutaneous endoscopic gastrostomy (PEG) placement.
44
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
Table I Characteristics of the pre-percutaneous endoscopic gastrostomy patients No. of cases Male Female Age (years), mean SD (range) Principal diseases (no.) Cerebrovascular disease cerebral infarction cerebral haemorrhage othersa Parkinson disease Dementia Othersb Nutritional methods (no.) TPN NGF PPN þ PO PPN Nutritional conditions Albumin (g/dL) Cholinesterase (U/L) Total cholesterol (mg/dL) Total lymphocyte (/mL)
67 20 47 80.7 9.3 (54e95) 54 38 8 8 5 4 4 35 25 6 1
Discussion
3.16 0.55 205.9 73.0 161.4 39.4 1661.9 669.2
TPN, total parenteral nutrition; NGF, nasogastric tube feeding; PPN, peripheral parenteral nutrition; PO, per os. a Subarachnoid haemorrhage, five cases; cerebellar haemorrhage, two cases; brain stem infarction, one case. b Senile decay, two cases; mucopolysaccharidosis, one case; brain injury, one case.
(84.8%) had cerebrovascular disease (P ¼ 0.16) (Table II). Age, gender, the presence of disease, nutritional methods and nutritional conditions were not significantly different between two groups after the PEG placement retrospectively. Gastric acidity and peristomal infection Twenty-one (31.3%) of the 67 patients developed peristomal infections. The average median gastric pH was 4.05 2.61 in the infected group and 3.09 1.86 in the non-infected group; however, the difference was not significant (P ¼ 0.073). Fifty-two patients were negative for MRSA of the oropharynx, and 11 cases (21.2%) showed peristomal infection. The average median pH in group A (11 cases) and group B (41 cases) was 5.05 2.55 and 3.06 1.83 (P ¼ 0.019), respectively. Cultures from purulent discharge around the peri-PEG tube and oropharynx bacterial cultures In 20 of the 21 patients who developed peristomal infection, the bacteria cultured from the PEG tube site yielded the same bacteria as from the oropharynx. In 10 patients MRSA was isolated from the Table II Comparison between infected and non-infected groups after the percutaneous endoscopic gastrostomy placement Infected group 21 cases Age (years), mean (range) 81.6 6.6 Gender (male) 7 (33.3%) Disease (CVD) 15 (71.4%) Nutritional methods TPN 9 (42.9%) NGF 10 (47.6%) Nutritional conditions Albumin (g/dL) 3.10 0.55 Cholinesterase (U/L) 215.8 80.8 Total cholesterol (mg/dL) 154.3 42.2 1486.6 491.8 Total lymphocyte (/mL)
Non-infected group
PEG tube site and in six cases Enterobacteriaceae including Proteus spp., Enterobacter spp. and Escherichia coli, and three cases with Pseudomonas spp. When the oropharynx was colonised with MRSA, peristomal infection occurred in 10/15 (66.7%) cases, whereas in non-colonised patients the infection rate was 11/52 (21.2%) (P < 0.001). Seven of the 20 (35.0%) patients who had Enterobactericeae in the oropharynx developed a peristomal infection compared with only 14/47 (29.8%) who did not have enterobacteria in the oropharynx and did not develop a peristomal infection (nonsignificant difference). Similarly for Pseudomonas spp. the difference was non-significant (Table III).
P-value
46 cases 80.3 10.3 13 (28.3%) 39 (84.8%)
NS NS NS
26 (56.5%) 15 (32.6%)
NS NS
3.19 0.54 201.3 68.5 164.7 37.6 1743.7 723.2
NS NS NS NS
CVD, cerebrovascular disease; TPN, total parenteral nutrition; NGF, nasogastric tube feeding; NS, non-significant.
The pull or push method has an inherent limitation; bacteria present in the oral cavity and pharynx may attach to the PEG tube as it passes through the oropharynx. This risk cannot be avoided during the manipulations of the pull or push method. In our series, 20/21 (95.2%) patients developed peristomal infection by the same organisms, which were cultured from the oropharyngeal specimens. This finding suggests that peristomal infection was due to contamination of the PEG tube on passage through the oropharynx. In addition, bacteria that are swallowed and collected in the stomach are attached to the gastric wall when the PEG tube is passed through the gastrocutaneous fistula. Gastric microbial flora is influenced by gastric acidity, and gastric microbial flora increases above pH 4.0.9,10,16,17 Previous reports demonstrated that H2-receptor antagonists affect intragastric acidity and gastric microbial flora, thereby significantly increasing the risk of postoperative infection.10e13 In this study we randomised about one-third of patients to receive ranitidine sulfate. As almost all patients were elderly, it was thought that they had low acid output due to chronic atrophic gastritis. Although the pH difference between patients with and without drug administration was not significant (P ¼ 0.075), administration of ranitidine sulfate for two weeks suppressed gastric acidity, promoted microbial overgrowth, and might have led to a peristomal infection. The gastric juice pH was obtained by 24 h pH monitoring on the day of PEG placement. One potential limitation of this study is that on the same day, stress factors such as pain due to the PEG placement and the endoscopic procedure may have affected gastric acid secretion. Therefore, the median pH of gastric juice may be different during the period of fistula formation. Our results suggest that in patients who are not colonised with MRSA, the peristomal infection was affected by the gastric pH on
Table III Rate of peristomal infection according to positive or negative bacterial culture from oropharynx Culture from oropharynx MRSA Positive Negative Enterobacteriaceaea Positive Negative Pseudomonas spp.b Positive Negative
Rate of peristomal infection
P-value 0.00079
66.7% (10/15) 21.2% (11/52) NS 35.0% (7/20) 29.8% (14/47) NS 29.4% (5/17) 32.0% (16/50)
MRSA, meticillin-resistant Staphylococcus aureus; NS, non-significant. a Enterobacter aerogenes, Enterobacter cloacae, Escherichia coli, Proteus mirabilis. b Pseudomonas aeruginosa, Pseudomonas fluorescens, Pseudomonas paucimobilis.
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
the day of PEG placement. Nevertheless the peristomal infection rate might not be affected by gastric acidity in MRSA carriers. Usually Staphylococcus aureus is killed rapidly at pH 2 but develops resistance to acidity by a sigB-mediated pathway.18,19 This leads to induction of sodA, which renders increased acid resistance suggesting that acid stress in turn leads to oxidative stress.20 Therefore, it is thought that the peristomal infection may not be affected by gastric acidity in patients positive for MRSA in the oropharynx, and that MRSA and pH may be independent risk factors for peristomal infection. This study suggests that MRSA in the oropharynx and the use of the pull method strongly promote peristomal infection. It was thought that MRSA might not have been affected by gastric acidity because the average median pH of 10 patients with peristomal infection caused by MRSA was 2.96 2.21. MRSA might develop resistance for gastric acidity.19 MRSA was reported as an emerging major pathogen in PEG site infections.21,22 We believe that the peristomal infection rate will decrease if MRSA can be eradicated from the oropharynx. Based on these findings, it is necessary to reduce bacterial contamination of the PEG tube. Suzuki et al. reported that peristomal infection was prevented by covering the PEG tube and Maetani et al. reported that an overtube during PEG placement reduced the risk of peristomal wound infection, whereas Horiuchi et al. reported that nasopharyngeal decolonisation reduced peristomal wound infection.23e25 In addition, bacterial overload in the gastric juice should be decreased by maintaining low gastric pH. It is thought that Enterobacteriaceae, Pseudomonas spp. and others e except MRSA e in the oropharynx are swallowed and the bacterial overload from gastric flora is induced when gastric acidity is above pH 4.0, thus increasing the risk of peristomal infection with PEG placement.16,17 Therefore, an H2-receptor antagonist or proton pump inhibitor should not be administered during the course of PEG placement. If patients have severe atrophic gastritis, which indicates low acid secretion, oropharyngeal and gastric eradication should be performed if possible before PEG placement. In conclusion, peristomal infection after PEG placement by the pull method was associated with the gastric acidity. The existence of MRSA in the oropharynx might increase the risk of peristomal infection, and MRSA in the oropharynx should be eradicated before PEG placement. Conflict of interest statement None declared. Funding source This study was supported in part by funds from Otaki Hospital, which paid for the 24 h pH monitoring device. References 1. Ponsky JL, Gauderer MWL. Percutaneous endoscopic gastrostomy: a non-operative technique for feeding gastrostomy. Gastrointest Endosc 1981;27:9e11.
45
2. Gauderer MWL, Ponsky JL, Izant Jr RJ. Gastrostomy without laparotomy: a percutaneous endoscopic technique. J Pediatr Surg 1980;15:872e875. 3. Mamel JJ. Percutaneous endoscopic gastrostomy. Am J Gastroenterol 1989;84: 703e710. 4. Taylor CA, Larson DE, Ballard DJ, et al. Predictors of outcome after percutaneous endoscopic gastrostomy: a community-based study. Mayo Clin Proc 1992;67: 1042e1049. 5. Rey JF, Budzynska A, Axon A, Kruse A, Nowak A. Guidelines of the European Society of Gastrointestinal Endoscopy (E.S.G.E.) antibiotic prophylaxis for gastrointestinal endoscopy. European Society of Gastrointestinal Endoscopy. Endoscopy 1998;30:318e324. 6. Lee JH, Kim JJ, Kim YH, et al. Increased risk of peristomal wound infection after percutaneous endoscopic gastrostomy in patients with diabetes mellitus. Dig Liver Dis 2002;34:857e861. 7. Deitel M, Bendago M, Spratt EH, Burul CJ, To TB. Percutaneous endoscopic gastrostomy by the “pull” and “introducer” methods. Can J Surg 1988;31:102e104. 8. Maetani I, Tada T, Ukita T, Inoue H, Sakai Y, Yoshikawa M. PEG with introducer or pull method: a prospective randomized comparison. Gastrointest Endosc 2003;57:837e841. 9. Heyland DK, Cook DJ, Schoenfeld PS, Frietag A, Varon J, Wood G. The effect of acidified enteral feeds on gastric colonization in critically ill patients: results of a multicenter randomized trial. Canadian Critical Care Trials Group. Crit Care Med 1999;27:2399e2406. 10. Cothran DS, Borowitz SM, Sutphen JL, Dudley SM, Donowitz LG. Alteration of normal gastric flora in neonates receiving ranitidine. J Perinatol 1997;17: 383e388. 11. Lachman L, Howden CW. Twenty-four-hour intragastric pH: tolerance within 5 days of continuous ranitidine administration. Am J Gastroenterol 2000;95:57e61. 12. Muscroft TJ, Youngs D, Burdon DW, Keighley MR. Cimetidine and the potential risk of postoperative sepsis. Br J Surg 1981;68:557e559. 13. Muscroft TJ, Deane SA, Youngs D, Burdon DW, Keighley MR. The microflora of the postoperative stomach. Br J Surg 1981;68:560e564. 14. Ono H, Azuma T, Miyaji H, et al. Effects of percutaneous endoscopic gastrostomy tube placement on gastric antral motility and gastric emptying. J Gastroenterol 2003;38:930e936. 15. Jain NK, Larson DE, Schroeder KW, et al. Antibiotic prophylaxis for percutaneous endoscopic gastrostomy. A prospective, randomized, double-blind clinical trial. Ann Intern Med 1987;107:824e828. 16. Giannella RA, Broitman SA, Zamcheck N. Gastric acid barrier to ingested microorganisms in man: studies in vivo and in vitro. Gut 1972;13:251e256. 17. Hill M. Normal and pathological microbial flora of the upper gastrointestinal tract. Scand J Gastroenterol Suppl 1985;111:1e6. 18. Cotter PD, Hill C. Surviving the acid test: responses of gram-positive bacteria to low pH. Microbiol Mol Biol Rev 2003;67:429e453. 19. Chan PF, Foster SJ, Ingham E, Clements MO. The Staphylococcus aureus alternative sigma factor sigmaB controls the environmental stress response but not starvation survival or pathogenicity in a mouse abscess model. J Bacteriol 1998;180:6082e6089. 20. Clements MO, Foster SJ. Stress resistance in Staphylococcus aureus. Trends Microbiol 1999;7:458e462. 21. Chaudhary KA, Smith OJ, Cuddy PG, Clarkston WK. PEG site infections: the emergence of methicillin resistant Staphylococcus aureus as a major pathogen. Am J Gastroenterol 2004;97:1713e1716. 22. Rao GG, Osman M, Johnson L, Ramsey D, Jones S, Fidler H. Prevention of percutaneous endoscopic gastrostomy site infections caused by methicillinresistant Staphylococcus aureus. J Hosp Infection 2004;58:81e83. 23. Suzuki Y, Urashima M, Ishibashi Y, et al. Covering the percutaneous endoscopic gastrostomy (PEG) tube prevents peristomal infection. World J Surg 2006;30:1450e1458. 24. Maetani I, Yasuda M, Seike M, et al. Efficacy of an overtube for reducing the risk of peristomal infection after PEG placement: a prospective, randomized comparison study. Gastrointest Endosc 2005;61:522e527. 25. Horiuchi A, Nakayama Y, Kajiyama M, Fujii H, Tanaka N. Nasopharyngeal decolonization of methicillin-resistant Staphylococcus aureus can reduce PEG peristomal wound infection. Am J Gastroenterol 2006;101:274e277.
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Effects of gastric acidity on peristomal infection after percutaneous endoscopic gastrostomy placement H. Ono a, *, S. Ito b, Y. Yamazaki c, Y. Otaki d, H. Otaki d a
Department of Internal Medicine, Seiwa Memorial Hospital, Sapporo, Japan Division of Internal Medicine, Tannan Regional Medical Center, Sabae, Japan c Division of Endoscopic Medicine, Faculty of Medical Sciences, University of Fukui, Eiheiji-cho, Fukui, Japan d Department of Internal Medicine, Otaki Hospital, Fukui, Japan b
a r t i c l e i n f o
s u m m a r y
Article history: Received 18 July 2009 Accepted 8 April 2010 Available online 1 July 2010
Peristomal infection is a common complication following percutaneous endoscopic gastrostomy (PEG) placement. This study investigated the effect of gastric acidity on peristomal infection, including type of bacteria and the relationship between bacteria cultured from the oropharynx and PEG tube site. Sixty-seven patients with dysphagia underwent PEG placement at Otaki Hospital between 1998 and 2001. Gastric acidity was evaluated by 24 h pH monitoring. Patients were observed for peristomal infection for two weeks after PEG placement, with specimens collected from the oropharynx and PEG tube site. Twenty-one (31.3%) of the patients who had undergone PEG insertion developed peristomal infections. Of 52 patients who were not colonised with meticillin-resistant Staphylococcus aureus (MRSA) in the oropharynx, 11 cases (21.2%) developed peristomal infection. The median gastric pH of infected patients (11 cases) was 5.05 2.55 (mean SD) and in patients without infection (41 cases) it was 3.06 1.83 (P ¼ 0.019). Peristomal infection developed in 66.7% (10/15) of patients carrying MRSA compared with only 21.2% (11/52) of patients who were not colonised by MRSA (P < 0.001). The incidence of peristomal infection was affected by gastric acidity and the presence of MRSA in the oropharynx. Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Gastric acidity Meticillin-resistant Staphylococcus aureus Percutaneous endoscopic gastrostomy Peristomal infection pH monitoring
Introduction In 1980, Ponsky and Garderer developed the percutaneous endoscopic gastrostomy (PEG) method.1,2 Nutrition is delivered through the PEG tube, which facilitates feeding and improves the nutritional status of patients with dysphagia. Although the PEG procedure is easy to perform, it sometimes results in complications such as peristomal infection, intragastric bleeding, and aspiration pneumonia.3,4 According to the European Society of Gastrointestinal Endoscopy Guidelines the most common complication is peristomal infection, which occurs in 30e43% of patients with a PEG tube.5
* Corresponding author. Address: Department of Internal Medicine, Seiwa Memorial Hospital, 1-5-1-1 Kotoni, Nishi-ku, Sapporo 063-0811, Japan. Tel.: þ81 11 611 1111; fax: þ81 11 631 6271. E-mail address: [email protected] (H. Ono).
The mechanism of peristomal infection is not clearly understood, but it may be associated with the patient’s general condition or with the PEG procedure. Risk factors include diabetes mellitus, malignant disease, malnutrition, or other factors that may potentiate the effect of peristomal infection.6 A gastrocutaneous fistula may be difficult to complete due to these diseases and may induce secondary bacterial infections. Therefore PEG placement must be performed safely and carefully in these patients. There are three methods of performing PEG: the pull, the push, and the introducer or direct method. During the pull or push method, the PEG tube is passed through the oral cavity to reach the stomach, increasing the risk of bacterial contamination. Deitel et al. reported that infection did not occur when they used the introducer method.7,8 In addition, oral bacteria will reach the stomach from the oropharynx via saliva. The gastric microbial flora is thought to be influenced by gastric acidity.9,10 Furthermore, H2-receptor antagonists affect gastric acidity and gastric microbial flora in ways that
0195-6701/$ e see front matter Ó 2010 The Hospital Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2010.04.011
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
significantly alter the risk of postoperative infection.10e13 Older patients who are at higher risk of cerebrovascular disease are more likely to receive PEG tube nutrition. In addition these patients often have low acid output due to chronic atrophic gastritis. In this study, we investigated the effect of intragastric acidity on peristomal infection during the course of PEG placement. We also investigated the type of bacteria and the relationship between bacteria cultured from the oropharynx and those from the PEG tube site. Methods Patients During a randomised controlled trial to assess the effect of ranitidin sulfate after PEG insertion, 67 patients (20 males and 47 females aged 54e95 years; mean: 80.7 9.3) who had undergone PEG due to dysphagia between January 1998 and May 2001 at Otaki Hospital were followed up for PEG site infection (Figure 1).14 Of the 67 patients, 54 had cerebrovascular disease (80.6%), five had Parkinson’s disease, and four had senile dementia (Table I). This study was performed according to the principles of the Declaration of Helsinki; the ethics committee of Otaki Hospital approved the protocol and written informed consent was obtained from all patients’ families. Plan for PEG placement We performed a secondary analysis for peristomal infection in these 67 patients. All patients underwent PEG placement by the pull method using a pull type kit (Bard Interventional Products, Billerica, MA, USA).14 Twenty-four-hour pH monitoring (Degitrapper 4 Mb; Synectics Medical, Stockholm, Sweden) was initiated soon after PEG placement. Ranitidine sulfate (100 mg/day, i.v.), an H2-receptor antagonist, was administered to 19 randomly selected patients to control gastric acidity during the two weeks after PEG. Piperacillin (2 g/day, i.v.) was administered to all patients for three days after the PEG placement. Specimens from the oral cavity and pharynx were collected for two days before the PEG placement, and
43
pus was collected from the PEG tube site three times after the PEG placement only if patients’ purulent discharge was present. The bacteria were identified at the Hokuriku Medical Laboratories. Definition of peristomal infection and patient groups Peristomal infection was defined as the presence of purulent discharge around the PEG tube within 2 weeks after PEG, including the inability to begin feeding through the PEG tube due to this infection, about which we referred to the study by Jain et al.15 Two endoscopists, who were certificated from Japan Gastroenterological Endoscopy Society, determined presence or absence of this infection. Patients were divided into two groups: the infected group and the non-infected group. Further groups were divided as follows: group A consisted of the infected subgroup excluding MRSA-positive patients; group B was composed of the non-infected subgroup excluding MRSA-positive patients. Statistical analysis Results were expressed as means SD and were analysed for statistical significance by the t-test or Fisher’s exact test. Age, nutritional conditions and the average median gastric pH of the cases in each group were compared using the t-test. Gender, cerebrovascular disease, nutritional methods, patients positive for MRSA, enterobacteria and pseudomonas of the oropharynx were compared using the Fisher’s exact test. Differences were significant at P < 0.05. Results Age, gender, disease, nutritional methods and conditions The average age of the infected group was 81.6 6.6 years and that of the non-infected group was 80.3 10.3 years (P ¼ 0.27). There were 7 males and 14 females in the infected group, and 13 males and 33 females in the non-infected group (P ¼ 0.39). Fifteen patients in the infected (71.4%) and 39 patients in the non-infected
PEG placement 2
3
4
7
14 (day)
24 h pH monitoring endoscopy peri-PEG culture (three times) pharynx culture (2 days) Piperacillin 2 g/day (i.v.) 3 days Ranitidine sulfate 100 mg/day (i.v)
19 cases
Ranitidine sulfate unused
48 cases
Figure 1. Protocol for percutaneous endoscopic gastrostomy (PEG) placement.
44
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
Table I Characteristics of the pre-percutaneous endoscopic gastrostomy patients No. of cases Male Female Age (years), mean SD (range) Principal diseases (no.) Cerebrovascular disease cerebral infarction cerebral haemorrhage othersa Parkinson disease Dementia Othersb Nutritional methods (no.) TPN NGF PPN þ PO PPN Nutritional conditions Albumin (g/dL) Cholinesterase (U/L) Total cholesterol (mg/dL) Total lymphocyte (/mL)
67 20 47 80.7 9.3 (54e95) 54 38 8 8 5 4 4 35 25 6 1
Discussion
3.16 0.55 205.9 73.0 161.4 39.4 1661.9 669.2
TPN, total parenteral nutrition; NGF, nasogastric tube feeding; PPN, peripheral parenteral nutrition; PO, per os. a Subarachnoid haemorrhage, five cases; cerebellar haemorrhage, two cases; brain stem infarction, one case. b Senile decay, two cases; mucopolysaccharidosis, one case; brain injury, one case.
(84.8%) had cerebrovascular disease (P ¼ 0.16) (Table II). Age, gender, the presence of disease, nutritional methods and nutritional conditions were not significantly different between two groups after the PEG placement retrospectively. Gastric acidity and peristomal infection Twenty-one (31.3%) of the 67 patients developed peristomal infections. The average median gastric pH was 4.05 2.61 in the infected group and 3.09 1.86 in the non-infected group; however, the difference was not significant (P ¼ 0.073). Fifty-two patients were negative for MRSA of the oropharynx, and 11 cases (21.2%) showed peristomal infection. The average median pH in group A (11 cases) and group B (41 cases) was 5.05 2.55 and 3.06 1.83 (P ¼ 0.019), respectively. Cultures from purulent discharge around the peri-PEG tube and oropharynx bacterial cultures In 20 of the 21 patients who developed peristomal infection, the bacteria cultured from the PEG tube site yielded the same bacteria as from the oropharynx. In 10 patients MRSA was isolated from the Table II Comparison between infected and non-infected groups after the percutaneous endoscopic gastrostomy placement Infected group 21 cases Age (years), mean (range) 81.6 6.6 Gender (male) 7 (33.3%) Disease (CVD) 15 (71.4%) Nutritional methods TPN 9 (42.9%) NGF 10 (47.6%) Nutritional conditions Albumin (g/dL) 3.10 0.55 Cholinesterase (U/L) 215.8 80.8 Total cholesterol (mg/dL) 154.3 42.2 1486.6 491.8 Total lymphocyte (/mL)
Non-infected group
PEG tube site and in six cases Enterobacteriaceae including Proteus spp., Enterobacter spp. and Escherichia coli, and three cases with Pseudomonas spp. When the oropharynx was colonised with MRSA, peristomal infection occurred in 10/15 (66.7%) cases, whereas in non-colonised patients the infection rate was 11/52 (21.2%) (P < 0.001). Seven of the 20 (35.0%) patients who had Enterobactericeae in the oropharynx developed a peristomal infection compared with only 14/47 (29.8%) who did not have enterobacteria in the oropharynx and did not develop a peristomal infection (nonsignificant difference). Similarly for Pseudomonas spp. the difference was non-significant (Table III).
P-value
46 cases 80.3 10.3 13 (28.3%) 39 (84.8%)
NS NS NS
26 (56.5%) 15 (32.6%)
NS NS
3.19 0.54 201.3 68.5 164.7 37.6 1743.7 723.2
NS NS NS NS
CVD, cerebrovascular disease; TPN, total parenteral nutrition; NGF, nasogastric tube feeding; NS, non-significant.
The pull or push method has an inherent limitation; bacteria present in the oral cavity and pharynx may attach to the PEG tube as it passes through the oropharynx. This risk cannot be avoided during the manipulations of the pull or push method. In our series, 20/21 (95.2%) patients developed peristomal infection by the same organisms, which were cultured from the oropharyngeal specimens. This finding suggests that peristomal infection was due to contamination of the PEG tube on passage through the oropharynx. In addition, bacteria that are swallowed and collected in the stomach are attached to the gastric wall when the PEG tube is passed through the gastrocutaneous fistula. Gastric microbial flora is influenced by gastric acidity, and gastric microbial flora increases above pH 4.0.9,10,16,17 Previous reports demonstrated that H2-receptor antagonists affect intragastric acidity and gastric microbial flora, thereby significantly increasing the risk of postoperative infection.10e13 In this study we randomised about one-third of patients to receive ranitidine sulfate. As almost all patients were elderly, it was thought that they had low acid output due to chronic atrophic gastritis. Although the pH difference between patients with and without drug administration was not significant (P ¼ 0.075), administration of ranitidine sulfate for two weeks suppressed gastric acidity, promoted microbial overgrowth, and might have led to a peristomal infection. The gastric juice pH was obtained by 24 h pH monitoring on the day of PEG placement. One potential limitation of this study is that on the same day, stress factors such as pain due to the PEG placement and the endoscopic procedure may have affected gastric acid secretion. Therefore, the median pH of gastric juice may be different during the period of fistula formation. Our results suggest that in patients who are not colonised with MRSA, the peristomal infection was affected by the gastric pH on
Table III Rate of peristomal infection according to positive or negative bacterial culture from oropharynx Culture from oropharynx MRSA Positive Negative Enterobacteriaceaea Positive Negative Pseudomonas spp.b Positive Negative
Rate of peristomal infection
P-value 0.00079
66.7% (10/15) 21.2% (11/52) NS 35.0% (7/20) 29.8% (14/47) NS 29.4% (5/17) 32.0% (16/50)
MRSA, meticillin-resistant Staphylococcus aureus; NS, non-significant. a Enterobacter aerogenes, Enterobacter cloacae, Escherichia coli, Proteus mirabilis. b Pseudomonas aeruginosa, Pseudomonas fluorescens, Pseudomonas paucimobilis.
H. Ono et al. / Journal of Hospital Infection 76 (2010) 42e45
the day of PEG placement. Nevertheless the peristomal infection rate might not be affected by gastric acidity in MRSA carriers. Usually Staphylococcus aureus is killed rapidly at pH 2 but develops resistance to acidity by a sigB-mediated pathway.18,19 This leads to induction of sodA, which renders increased acid resistance suggesting that acid stress in turn leads to oxidative stress.20 Therefore, it is thought that the peristomal infection may not be affected by gastric acidity in patients positive for MRSA in the oropharynx, and that MRSA and pH may be independent risk factors for peristomal infection. This study suggests that MRSA in the oropharynx and the use of the pull method strongly promote peristomal infection. It was thought that MRSA might not have been affected by gastric acidity because the average median pH of 10 patients with peristomal infection caused by MRSA was 2.96 2.21. MRSA might develop resistance for gastric acidity.19 MRSA was reported as an emerging major pathogen in PEG site infections.21,22 We believe that the peristomal infection rate will decrease if MRSA can be eradicated from the oropharynx. Based on these findings, it is necessary to reduce bacterial contamination of the PEG tube. Suzuki et al. reported that peristomal infection was prevented by covering the PEG tube and Maetani et al. reported that an overtube during PEG placement reduced the risk of peristomal wound infection, whereas Horiuchi et al. reported that nasopharyngeal decolonisation reduced peristomal wound infection.23e25 In addition, bacterial overload in the gastric juice should be decreased by maintaining low gastric pH. It is thought that Enterobacteriaceae, Pseudomonas spp. and others e except MRSA e in the oropharynx are swallowed and the bacterial overload from gastric flora is induced when gastric acidity is above pH 4.0, thus increasing the risk of peristomal infection with PEG placement.16,17 Therefore, an H2-receptor antagonist or proton pump inhibitor should not be administered during the course of PEG placement. If patients have severe atrophic gastritis, which indicates low acid secretion, oropharyngeal and gastric eradication should be performed if possible before PEG placement. In conclusion, peristomal infection after PEG placement by the pull method was associated with the gastric acidity. The existence of MRSA in the oropharynx might increase the risk of peristomal infection, and MRSA in the oropharynx should be eradicated before PEG placement. Conflict of interest statement None declared. Funding source This study was supported in part by funds from Otaki Hospital, which paid for the 24 h pH monitoring device. References 1. Ponsky JL, Gauderer MWL. Percutaneous endoscopic gastrostomy: a non-operative technique for feeding gastrostomy. Gastrointest Endosc 1981;27:9e11.
45
2. Gauderer MWL, Ponsky JL, Izant Jr RJ. Gastrostomy without laparotomy: a percutaneous endoscopic technique. J Pediatr Surg 1980;15:872e875. 3. Mamel JJ. Percutaneous endoscopic gastrostomy. Am J Gastroenterol 1989;84: 703e710. 4. Taylor CA, Larson DE, Ballard DJ, et al. Predictors of outcome after percutaneous endoscopic gastrostomy: a community-based study. Mayo Clin Proc 1992;67: 1042e1049. 5. Rey JF, Budzynska A, Axon A, Kruse A, Nowak A. Guidelines of the European Society of Gastrointestinal Endoscopy (E.S.G.E.) antibiotic prophylaxis for gastrointestinal endoscopy. European Society of Gastrointestinal Endoscopy. Endoscopy 1998;30:318e324. 6. Lee JH, Kim JJ, Kim YH, et al. Increased risk of peristomal wound infection after percutaneous endoscopic gastrostomy in patients with diabetes mellitus. Dig Liver Dis 2002;34:857e861. 7. Deitel M, Bendago M, Spratt EH, Burul CJ, To TB. Percutaneous endoscopic gastrostomy by the “pull” and “introducer” methods. Can J Surg 1988;31:102e104. 8. Maetani I, Tada T, Ukita T, Inoue H, Sakai Y, Yoshikawa M. PEG with introducer or pull method: a prospective randomized comparison. Gastrointest Endosc 2003;57:837e841. 9. Heyland DK, Cook DJ, Schoenfeld PS, Frietag A, Varon J, Wood G. The effect of acidified enteral feeds on gastric colonization in critically ill patients: results of a multicenter randomized trial. Canadian Critical Care Trials Group. Crit Care Med 1999;27:2399e2406. 10. Cothran DS, Borowitz SM, Sutphen JL, Dudley SM, Donowitz LG. Alteration of normal gastric flora in neonates receiving ranitidine. J Perinatol 1997;17: 383e388. 11. Lachman L, Howden CW. Twenty-four-hour intragastric pH: tolerance within 5 days of continuous ranitidine administration. Am J Gastroenterol 2000;95:57e61. 12. Muscroft TJ, Youngs D, Burdon DW, Keighley MR. Cimetidine and the potential risk of postoperative sepsis. Br J Surg 1981;68:557e559. 13. Muscroft TJ, Deane SA, Youngs D, Burdon DW, Keighley MR. The microflora of the postoperative stomach. Br J Surg 1981;68:560e564. 14. Ono H, Azuma T, Miyaji H, et al. Effects of percutaneous endoscopic gastrostomy tube placement on gastric antral motility and gastric emptying. J Gastroenterol 2003;38:930e936. 15. Jain NK, Larson DE, Schroeder KW, et al. Antibiotic prophylaxis for percutaneous endoscopic gastrostomy. A prospective, randomized, double-blind clinical trial. Ann Intern Med 1987;107:824e828. 16. Giannella RA, Broitman SA, Zamcheck N. Gastric acid barrier to ingested microorganisms in man: studies in vivo and in vitro. Gut 1972;13:251e256. 17. Hill M. Normal and pathological microbial flora of the upper gastrointestinal tract. Scand J Gastroenterol Suppl 1985;111:1e6. 18. Cotter PD, Hill C. Surviving the acid test: responses of gram-positive bacteria to low pH. Microbiol Mol Biol Rev 2003;67:429e453. 19. Chan PF, Foster SJ, Ingham E, Clements MO. The Staphylococcus aureus alternative sigma factor sigmaB controls the environmental stress response but not starvation survival or pathogenicity in a mouse abscess model. J Bacteriol 1998;180:6082e6089. 20. Clements MO, Foster SJ. Stress resistance in Staphylococcus aureus. Trends Microbiol 1999;7:458e462. 21. Chaudhary KA, Smith OJ, Cuddy PG, Clarkston WK. PEG site infections: the emergence of methicillin resistant Staphylococcus aureus as a major pathogen. Am J Gastroenterol 2004;97:1713e1716. 22. Rao GG, Osman M, Johnson L, Ramsey D, Jones S, Fidler H. Prevention of percutaneous endoscopic gastrostomy site infections caused by methicillinresistant Staphylococcus aureus. J Hosp Infection 2004;58:81e83. 23. Suzuki Y, Urashima M, Ishibashi Y, et al. Covering the percutaneous endoscopic gastrostomy (PEG) tube prevents peristomal infection. World J Surg 2006;30:1450e1458. 24. Maetani I, Yasuda M, Seike M, et al. Efficacy of an overtube for reducing the risk of peristomal infection after PEG placement: a prospective, randomized comparison study. Gastrointest Endosc 2005;61:522e527. 25. Horiuchi A, Nakayama Y, Kajiyama M, Fujii H, Tanaka N. Nasopharyngeal decolonization of methicillin-resistant Staphylococcus aureus can reduce PEG peristomal wound infection. Am J Gastroenterol 2006;101:274e277.