Risk factors and prevention of peritonitis in peritoneal dialysis

INTRODUCTION — Peritonitis is a major complication of peritoneal dialysis and a primary reason for patients to switch from peritoneal dialysis to hemodialysis [1]. This topic reviews the causes of peritonitis in this setting and measures to reduce the risk.

The clinical features and treatment of peritoneal dialysis-associated peritonitis are discussed elsewhere. (See "Clinical manifestations and diagnosis of peritonitis in peritoneal dialysis" and "Microbiology and therapy of peritonitis in peritoneal dialysis".)

PATHOPHYSIOLOGY — Among patients on peritoneal dialysis, peritonitis is caused by introduction of microbes to the usually sterile peritoneum in the setting of compromised host defenses. Both the introduction of microbes and the compromise of defenses are usually related to the dialysis procedure. We discuss the source of microbial contamination, the possible role of biofilms in preventing eradication of microbes by host defenses, and impairment of macrophages due to the peritoneal dialysis procedure.

Source of infection — Sources of peritonitis include intraluminal contamination (or touch contamination during or between exchanges); periluminal contamination (or extension from exit-site or tunnel infection); transvisceral migration from a bowel or, rarely, vaginal leak; and hematogenous dissemination from a remote source, as may occur during dental procedures.

Peritonitis is most often due to contamination with pathogenic skin bacteria, with Staphylococcus epidermidis and Staphylococcus aureus accounting for the majority of cases [2]. (See "Microbiology and therapy of peritonitis in peritoneal dialysis".)

Contamination with skin bacteria occurs when the patient is performing exchanges or connecting/disconnecting from the transfer set. A patient on continuous ambulatory peritoneal dialysis (CAPD) connects and disconnects from the transfer set at least four times per day and thus 1460 times per year, raising the possibility of an infection even among maximally skilled patients. The near-universal use of "Y" transfer systems has reduced the rate of peritonitis [3].

The risk correlates with the number of times the patient has to connect or disconnect. Some studies have suggested an increase in risk with CAPD compared with continuous cycling peritoneal dialysis (CCPD) because of fewer required connections and disconnections.

Even without contamination during exchanges or disconnects, bacteria can track along the peritoneal catheter and its tunnel. The catheter itself can be a nidus of retained microbes (particularly fungi) that grow directly into the material or under biofilm/slime layers, where they are protected from host defenses. (See 'Biofilms' below.)

Biofilms — A biofilm is a slimy film of bacteria or fungi that adheres to a surface. Biofilms on the catheter may contribute to relapsing or recurrent infection and to antibiotic resistance, although the clinical significance of biofilms is uncertain [4].

A large, prospective study demonstrated biofilms in peritoneal catheters removed from patients on peritoneal dialysis; cultures of scrapings from these biofilms were often positive for S. aureus or S. epidermidis [5]. There was, however, no clear relationship to clinical peritonitis [6]. Other studies also showed that biofilms can be present without any evidence of infection [7,8].

Nevertheless, biofilms may contribute to relapsing or recurrent peritonitis with the same organism [9,10]. In a study including nearly 200 patients who initiated peritoneal dialysis between 1990 and 2000 and had experienced more than one episode of culture-positive peritonitis, approximately 80 percent (157 patients) had at least one repeat infection with the same organism [10]. More than 50 percent of the peritonitis episodes among 124 patients were caused by the same organism, and 59 of 90 patients (65 percent) of those with more than four infectious episodes had at least one-half or more of their episodes caused by the same organism.

Another report found that patients with multiple episodes of peritonitis were likely to have stable biofilms, positive biofilm cultures, and a high incidence of catheter loss [9]. Troublesome organisms that notoriously form biofilms in this setting are staphylococci and Pseudomonas [11]. Intraperitoneal urokinase has been suggested as a treatment of biofilm as a potential cause of relapsing infection [12,13].

Compromised host defenses — An intact peritoneum and the defense mechanisms of the mesothelium are important barriers to the development of peritonitis in patients on peritoneal dialysis. However, these host defenses are compromised by the peritoneal dialysis procedure:

●The continuous presence of a large amount of fluid in the peritoneal cavity can impair host defenses. Mesothelial surfaces have extraordinary host defense mechanisms when the surfaces are allowed close proximity to each other [14]. The presence of dialysis solution prevents close proximity of mesothelial surfaces and also dilutes macrophage and cytokine levels.

●The composition of the peritoneal dialysis fluid may promote microbial growth and inhibit host defense. Standard dialysis fluids have a high glucose concentration, which promotes microbial growth. The high glucose concentration, the low pH, and glucose degradation products (GDPs) may also alter host response. Newer peritoneal dialysis fluids that contain glucose polymers (such as icodextrin), amino acids rather than glucose, or fluids that results in fewer GDPs may be relatively less harmful to neutrophil and macrophage function, although this has not been proven. (See "Peritoneal dialysis solutions".)

●Resident macrophages and cytokines that potentially prevent infection are constantly removed during each exchange of dialysis fluids.

As a result of factors that are specific to patients on peritoneal dialysis, a relatively small inoculum of bacteria, as often occurs with a touch contamination, can readily induce peritonitis in a patient on peritoneal dialysis; in contrast, a similar inoculation during surgical laparotomy seldom causes peritonitis [14,15].

EPIDEMIOLOGY

Incidence — The reported incidence of peritonitis ranges widely [16-21]. A retrospective study of 1677 patients on incident peritoneal dialysis revealed a first-year peritonitis rate of 42 per 100 patient-years [20]. Of 463 first peritonitis cases, 336 (72.6 percent) occurred within the first six months of peritoneal dialysis.

Risk factors — Important modifiable risk factors include recent invasive intervention (colonoscopy, sigmoidoscopy, cystoscopy, hysteroscopy) or dental procedures, nasal S. aureus carriage, and exit-site and/or tunnel infections [22-25]. (See 'Prophylactic treatment with procedures' below.)

Other factors that have been associated in various studies include constipation, smoking and chronic obstructive pulmonary disease (COPD), domestic pets, obesity, depression, hypokalemia, hypoalbuminemia, prior hemodialysis, use of bioincompatible solutions, and living far away from the peritoneal dialysis clinic [26-29].

Selected risk factors are discussed below.

●Invasive interventions – Invasive interventions (colonoscopy, sigmoidoscopy, cystoscopy, hysteroscopy) or dental procedures are risk factors for peritonitis [22-25,30-33]. In one study of 79 continuous ambulatory peritoneal dialysis (CAPD) patients undergoing colonoscopies without prophylactic antibiotics, the risk of peritonitis was 6.2 percent [23]. Similar results were observed in another study of patients on peritoneal dialysis undergoing 99 endoscopies [30]. In this study, peritonitis only occurred following invasive procedures such as biopsy, polypectomy, or intrauterine device (IUD) implantation but not following endoscopy without an additional procedure.

●S. aureus nasal carriage – The primary reservoir for S. aureus is the anterior nares [34]. Nasal carriers of S. aureus are at increased risk to develop S. aureus exit-site infections, tunnel infections, and, possibly, peritonitis [35-37]. One landmark report found that nasal carriers were four times more likely to develop exit-site infections than noncarriers; the overall peritonitis rates were not different between the groups, but all cases of S. aureus peritonitis occurred in the carriers [35]. Exit-site infections are a known risk factor for peritonitis.

●Exit-site infection – Catheter exit-site infections are associated with increased risk of peritonitis [38-40]. A secondary analysis of a randomized trial compared two antimicrobial ointments among 203 stable patients on peritoneal dialysis [39]. Over 18 months, there were 44 exit-site infections and 87 peritonitis episodes. Patients who had an exit-site infection were much more likely to develop peritonitis within 30 days, even if the exit-site infection was appropriately treated. The risk of peritonitis decreased over time, with hazard ratios of 11.1, 5.3, and 4.9 at 15, 45, and 60 days, respectively. Surprisingly, in this study, the organisms that caused peritonitis were most often unrelated to those that caused the exit-site infection, suggesting that the mechanism of increased susceptibility to peritonitis is not straightforward. However, this observation needs to be confirmed in a larger study. Regardless of the mechanism by which exit-site infections lead to peritonitis, these data support the need for aggressive prevention and treatment of exit-site infections [41]. (See "Peritoneal catheter exit-site and tunnel infections in peritoneal dialysis in adults", section on 'Treatment'.)

●Break in technique – Obvious breaks in technique (such as when and where the open end or unsecured end of the catheter touches a nonsterile object [eg, ungloved hands, table outside sterile field, patient's skin, etc]) are associated with increased risk of infection [42,43]. Breaks in technique may be more likely shortly after patients switch from one peritoneal dialysis modality to another (eg, switching from CAPD to continuous cycling peritoneal dialysis [CCPD]) or change dialysis product vendors, when patients may perform connection procedures that are inappropriate for their new equipment.

●Underlying gastrointestinal pathology – Peritonitis can be induced by underlying pathology of the gastrointestinal tract, resulting in significant morbidity and mortality [44,45]. These conditions include cholecystitis, appendicitis, ruptured diverticulum, and perforation during endoscopy [46].

●Constipation – Chronic constipation and the acute treatment of constipation increase the risk of peritonitis [46-49].

●Factors of uncertain significance

•Continuous ambulatory peritoneal dialysis – Some [50-53], but not all [16,54,55], studies have suggested that the rate of peritonitis is higher with CAPD compared with CCPD. This is presumed to be due to fewer required connections and disconnections. However, with the use of contemporary systems, the selected modality likely has little clinical impact on an individual patient's peritonitis risk [56].

•Use of nonbiocompatible dialysate – As noted above, repeated instillation of peritoneal dialysis fluids may alter host defense (see 'Compromised host defenses' above). More biocompatible fluids theoretically could reduce the risk of peritonitis. However, the data are inconclusive and do not support the use of biocompatible dialysate solely to prevent peritonitis. (See 'Modality and choice of dialysate' below.)

Randomized clinical trials as well as observation studies have come to different conclusions regarding the efficacy of biocompatible fluids [57-62]. While one trial (BALANZ) of 183 patients found that biocompatible fluids were associated with a reduced rate of peritonitis (0.3 versus 0.49 episodes per patient-year) [57], two previous trials (267 and 50 patients, respectively) did not find a difference in peritonitis rates in patients receiving biocompatible fluid [61,62].

•Hypokalemia – Hypokalemia may be a risk factor for peritonitis [63-65], possibly by reducing gut motility. In an open-label, 52-week, randomized trial of 167 patients on peritoneal dialysis who had a history of hypokalemia, potassium chloride administration based on a protocol to maintain serum potassium levels between 4 and 5 mEq/L resulted in fewer patients developing peritonitis than potassium chloride administration based on physician judgement when serum potassium levels were less than 3.5 mEq/L (13 versus 24 patients) [66].

•Gastric acid suppression – The use of gastric acid suppressants (ie, histamine 2 receptor antagonists [H2RAs] and proton pump inhibitors [PPIs]) may be a risk factor for peritonitis [67-69]. In a meta-analysis of six observational studies that included 829 patients on peritoneal dialysis, H2RA use was associated with a higher risk of enteric peritonitis (odds ratio 1.40; 95% CI 1.01–1.93) [70]. Some observational data suggest that the risk of peritonitis may be greater with H2RAs than with PPIs [67,69].

PREVENTION — It is important to prevent peritonitis. Peritonitis often results in morbidity, transfer to hemodialysis, and, occasionally, death [71]. (See "Microbiology and therapy of peritonitis in peritoneal dialysis", section on 'Prognosis'.)

We agree with the International Society of Peritoneal Dialysis (ISPD) guidelines that all peritoneal dialysis programs should monitor peritonitis-related events (in episodes/patient year at risk) as part of a Continuous Quality Improvement (CQI) program [22]. The ISPD guidelines have cited a target overall rate of ≤0.5 episodes/year at risk [22]. A rate higher than this, or detection of trends where overall rates are increasing, should provoke a review of training practices and clinical practices [22].

Preventive measures are discussed below. Our approach is largely consistent with the ISPD guidelines [22].

Considerations prior to placement of the peritoneal dialysis catheter (including catheter design with respect to body habitus) are discussed elsewhere. (See "Placement of the peritoneal dialysis catheter".)

Proper training and home visits — All patients must be trained in aseptic techniques and instructed in the definition of and proper response to contamination [72,73]. Training should be conducted by nursing staff with appropriate qualification and training [74].

After the initial training, patients should be monitored and various steps of the peritoneal dialysis therapy reviewed from time to time to be sure that patients are following the protocol correctly. Brief retraining should be considered after any episode of peritonitis.

The benefits of training as well as retraining were reported in a study of new patients who were randomly assigned to either standard or enhanced training (adult learning) [75]. At two years, enhanced training was associated with significantly fewer exit-site infections, as well as a trend toward a decreased peritonitis rate.

Ongoing education is also essential for nurses providing patient training. In a single-center, observational study, there was a negative association between trainers' length of time in practice and incidence of gram-positive peritonitis among incident patients [76].

Home visits by the peritoneal dialysis nursing staff may be helpful to identify and avoid potential problems. The location within the home where exchanges take place, the presence of pets, and the overall cleanliness of the home environment are a few of the factors to review in infection prevention [77].

Special attention is needed in cases when patients transition from one device to another (such as a newer model of the current manufacturer or a device from a different manufacturer). In these situations, the sequence of steps or actual steps for the disconnect or connect procedure may be different. In addition, patients may revert to previously learned techniques or habits (therapeutic drift), thereby increasing the risk of contamination or malfunction of the new device. Thus, during training, the staff will need to anticipate any old habits that could accidentally be used with the new device. New policies and procedures should be developed to preemptively address these issues for consistent training of patients who are in transition to new devices.

We ask patients to remove faucet aerators from their homes in order to minimize the risk of Pseudomonas peritonitis. Pseudomonas bacteria commonly colonize aerators used on faucets [78]. However, studies have not demonstrated that this practice reduces the risk of peritonitis.

Monitoring for depression should be part of routine assessment. Depression has been associated with increased risk of peritonitis. (See 'Risk factors' above.)

Transfer set — For all patients, we recommend the use of a "disconnect" or "Y" system rather than a spike system. Multiple reports have demonstrated a reduction in peritonitis risk with single-use disconnect systems [3,79-85]. This was best shown in a systematic review of seven trials consisting of 485 patients in which the use of the disconnect set, compared with the standard spike system, was associated with a lower risk of peritonitis (relative risk [RR] 0.64, 95% CI 0.53-0.77) [3]; a similar benefit with the disconnect system in terms of the peritonitis rate (RR 0.49, 95% CI 0.40-0.61) was reported upon analysis of eight trials and 7417 patient-months.

Disconnect systems include a standard Y system and a double-bag Y system in which both dialysis solutions and drain bags are preattached.

If available, we suggest the use of disconnect systems with double-bag systems [86]. Some [54,73,85,87-92], though not all [87], studies suggest that, compared with standard Y systems, double-bag systems may be associated with lower peritonitis rates since they more effectively flush out possible contamination during attachment (figure 1) [54,73,85,87-92]. In a study of 147 patients who were randomly assigned to a double-bag system, a disconnect Y set without attached double bag, or a conventional straight transfer system, the average peritonitis-free interval was highest in the twin-bag system group (24.8 months versus 12 and 6.1 months for the standard Y-set and straight transfer system groups, respectively) [91].

However, a systematic review of randomized, prospective studies found no statistically significant difference with double-bag and Y systems concerning peritonitis rates (RR 0.90, 95% CI 0.49-1.66) and exit-site/tunnel infection rates (RR 1.04, 95% CI 0.52-2.06) [3]. Despite this, there was a trend in favor of double-bag systems for experiencing peritonitis episodes (RR 0.59, 95% CI 0.35-1.01), which was based upon data from three trials consisting of 292 patients [87,88,91]. There is limited evidence that different double-bag systems may not be associated with the same peritonitis rates [93].

Modality and choice of dialysate — We do not choose the peritoneal dialysis modality based on risk of peritonitis. With the use of contemporary connection systems, the modality likely has little clinical impact on an individual patient's peritonitis risk, and many individual patient variables inform the choice of modality [56]. (See "Evaluating patients for chronic peritoneal dialysis and selection of modality".)

We do not select peritoneal dialysis solution on the basis of peritonitis risk. As noted above, biocompatible fluids could reduce the risk of peritonitis. (See 'Risk factors' above.)

However, randomized clinical trials as well as observation studies have come to different conclusions regarding the efficacy of biocompatible fluids [57-62]. While one trial (BALANZ) of 183 patients found that biocompatible fluids were associated with a reduced rate of peritonitis (0.3 versus 0.49 episodes per patient-year) [57], two previous trials (267 and 50 patients respectively) did not find a difference in peritonitis rates in patients receiving biocompatible fluid [61,62].

There are insufficient data to inform the choice of peritoneal dialysis solution based on a peritonitis risk reduction strategy.

Approach after breaks in technique — We change the transfer set after all recognized break in technique. We give antibiotics in selected cases depending on the risk of infection:

●If the contamination occurred distal to a closed transfer set clamp but the tubing and dialysate bag are not leaking (referred to as a "dry contamination"), we just change the transfer set. We do not give antibiotics, because, in our experience, the risk of resulting infection is relatively low.

●If there was a leak in the dialysate bags, or a break in tubing distal to an open clamp, but fluid was not being infused, we change the transfer set, and we give three days of oral cephalexin (500 mg twice daily). If the patient is allergic to cephalosporins, we give ciprofloxacin 500 mg daily for three days. Some clinicians use one dose of intraperitoneal antibiotics instead.

●If there was a leak in tubing proximal to a closed clamp or if contamination occurred while peritoneal fluid was being infused, we change the transfer set and we give one dose of intraperitoneal antibiotics. This type of contamination is associated with the highest risk of infection. We generally use intraperitoneal cefazolin for prophylaxis in this setting, but for patients colonized with methicillin-resistant staphylococcus aureus (MRSA) or at high risk of MRSA, intraperitoneal vancomycin plus an additional intraperitoneal antibiotic such as cefepime, ceftazidime, or an aminoglycoside is a reasonable alternative (table 1 and table 2). (See "Antimicrobial prophylaxis for prevention of surgical site infection in adults", section on 'Role of vancomycin'.)

Prophylactic treatment with procedures — Our standard practice prior to procedures is described below. However, there has been no systematic evaluation of efficacy, and approaches vary [94].

All procedures — For patients on peritoneal dialysis undergoing surgery or invasive procedures involving the abdomen or pelvis, including minimally invasive gastrointestinal, gynecologic, and urologic procedures, we suggest first draining the peritoneal fluid. We also drain the dialysate fluid before dental procedures. The presence of peritoneal dialysis fluid may promote bacterial growth and dilute concentrations of native host defense mechanisms. (See 'Source of infection' above and "Issues in patients on peritoneal dialysis undergoing surgery", section on 'Draining the abdomen before surgery'.)

We give antibiotics prior to selected procedures. Specific regimens are described below. (See 'Colonoscopy and upper endoscopy' below and 'Gynecologic procedures' below and 'Dental procedures' below.)

Colonoscopy and upper endoscopy — Dialysate fluid is drained the day of the procedure, and the abdomen is kept dry (see 'All procedures' above). We keep the abdomen dry until the day following the procedure (ie, the first day after the procedure) unless a biopsy or polypectomy is performed. For patients who undergo biopsy or polypectomy, we aim to keep the abdomen dry for longer periods; optimally, we restart peritoneal dialysis on the third day after the procedure unless dialysis is needed sooner.

We give prophylactic antibiotics prior to colonoscopy and upper endoscopy. Other experts, however, only give prophylactic antibiotics before colonoscopy (but not upper endoscopy), since evidence supporting such treatment before upper endoscopy is lacking. We administer a single intraperitoneal dose of gentamicin 160 mg in the overnight dwell or in cycler bags one night prior to the procedure, in addition to prescribing oral metronidazole 500 mg to be taken four times daily for three days (starting the day prior to the procedure) [94]. However, some clinicians only use oral antibiotics [95].

Prophylactic antibiotics decrease the risk of peritonitis following colonoscopy [23,30,95]. In one study of 77 continuous ambulatory peritoneal dialysis (CAPD) patients who underwent 97 colonoscopies, no cases of peritonitis were diagnosed in 18 patients who received prophylactic antibiotics, but five cases were diagnosed among 79 patients who did not [23].

Gynecologic procedures — We drain the dialysate fluid prior to gynecologic procedures, including hysteroscopy, placement of intrauterine device (IUD), cervical biopsy, and routine gynecologic exams (ie, pelvic exam and Papanicolaou [Pap] test). Our approach to restarting peritoneal dialysis after gynecologic procedures varies according to whether a biopsy or polypectomy is performed and is the same as for patients undergoing colonoscopy and endoscopy. (See 'Colonoscopy and upper endoscopy' above.)

For hysteroscopy, placement of IUD, and cervical biopsy, we also give prophylactic antibiotics. The antibiotics are the same as those administered prior to colonoscopy and endoscopy (see 'Colonoscopy and upper endoscopy' above). We generally do not give prophylactic antibiotics for routine gynecologic exams (ie, pelvic exam and Pap test without cervical biopsy or IUD placement). However, some clinicians give oral antibiotics for such exams.

Prophylactic antibiotics decrease the risk of peritonitis following gynecologic procedures [23,30,96]. In a study of 45 patients on peritoneal dialysis who underwent 125 endoscopies including hysteroscopy, peritonitis occurred in patients who were not treated with antibiotics but not in those who were [30]. All episodes of peritonitis occurred following endoscopy with invasive procedures (including biopsy, polypectomy, or IUD implantation).

Dental procedures — We drain the dialysate fluid prior to dental procedures and restart peritoneal dialysis the day after the procedure.

We give oral antibiotics prior to invasive or potentially invasive dental procedures, including routine dental cleanings. Dental procedures are associated with a high risk of bacteremia and have been associated with peritonitis among patients on peritoneal dialysis [97]. We give 2 grams of amoxicillin orally two hours prior to the procedure.

Prophylactic antibiotics for catheter exit-site leaks — For patients with peritoneal catheter exit-site leaks, we suggest prophylactic antibiotic therapy. Because patients with exit-site leaks are at risk of developing peritonitis [98], we typically administer antibiotics for the duration of the leak. We generally use intraperitoneal cefazolin for prophylaxis in this setting, but for patients colonized with MRSA or at high risk of MRSA, intraperitoneal vancomycin plus an additional intraperitoneal antibiotic such as cefepime, ceftazidime, or an aminoglycoside is a reasonable alternative (table 1 and table 2). (See "Antimicrobial prophylaxis for prevention of surgical site infection in adults", section on 'Role of vancomycin'.)

Our overall approach to the management of pericatheter leaks is detailed elsewhere. (See "Noninfectious complications of peritoneal dialysis catheters", section on 'Pericatheter leakage'.)

Prevention and treatment of catheter infections — We use measures to prevent and treat exit-site and tunnel infections. (See "Peritoneal catheter exit-site and tunnel infections in peritoneal dialysis in adults", section on 'Prevention' and "Peritoneal catheter exit-site and tunnel infections in peritoneal dialysis in adults", section on 'Treatment'.)

Exit-site and tunnel infections are important risk factor for peritonitis, particularly for infections caused by S. aureus and Pseudomonas aeruginosa.

Prevention of exit-site infections and prompt treatment of any exit-site infection are logical steps in prevention of peritonitis [99].

Antifungal prophylaxis during antibiotic therapy — Treatment with systemic antibiotics is a major risk factor for the development of fungal peritonitis among patients on peritoneal dialysis. Antifungal prophylaxis during the course of antibiotic therapy may reduce the risk of subsequent fungal peritonitis. Our approach to antifungal prophylaxis in this setting is discussed elsewhere. (See "Fungal peritonitis in peritoneal dialysis" and "Microbiology and therapy of peritonitis in peritoneal dialysis", section on 'Role of antifungal prophylaxis'.)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topic (see "Patient education: Peritoneal dialysis (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Pathophysiology – Among patients on peritoneal dialysis, peritonitis is most often due to contamination with pathogenic skin bacteria, with Staphylococcus epidermidis and Staphylococcus aureus accounting for the majority of cases. Contamination with skin bacteria occurs when the patient is performing exchanges or connecting/disconnecting from the transfer set. (See 'Pathophysiology' above.)

●Risk factors – Important modifiable risk factors include recent invasive intervention (colonoscopy, sigmoidoscopy, cystoscopy, hysteroscopy) or dental procedures, nasal S. aureus carriage, and exit-site and/or tunnel infections. (See 'Risk factors' above.)

●Prevention – Our approach to the prevention of peritonitis is consistent with the International Society of Peritoneal Dialysis (ISPD) guidelines. In addition to careful training of patients, we use the following measures in all patients on peritoneal dialysis:

•Preferred transfer set – We recommend a single-use "disconnect" or "Y" system rather than the older spike system (Grade 1B). Multiple reports have demonstrated a reduction in peritonitis risk with single-use disconnect systems.

We also suggest the use of disconnect systems with double-bag systems rather than standard Y system (Grade 2C). Some studies suggest that, compared with standard Y systems, double-bag systems may be associated with lower peritonitis rates. (See 'Transfer set' above.)

•Management of breaks in technique – Following all breaks in technique, we change the transfer set. For patients with a leak in the dialysate bags or break in tubing in the setting of an open transfer set clamp, or proximal to a closed transfer set clamp, we suggest giving antibiotics (Grade 2C). This type of contamination is associated with a high risk of infection. Our approach to antibiotic administration is as follows (see 'Approach after breaks in technique' above):

-For patients with a leak in the dialysate bags, or a break in tubing distal to an open clamp, providing fluid is not being infused, we give oral cephalexin (500 mg twice daily).

-For patients who have a leak in tubing proximal to a closed clamp or if contamination occurred while peritoneal fluid was being infused, we give one dose of intraperitoneal antibiotics. We generally use intraperitoneal cefazolin for prophylaxis in this setting, but for patients colonized with methicillin-resistant staphylococcus aureus (MRSA) or at high risk of MRSA, intraperitoneal vancomycin plus an additional intraperitoneal antibiotic such as cefepime, ceftazidime, or an aminoglycoside is a reasonable alternative (table 1 and table 2).

•Draining the abdomen – For patients on peritoneal dialysis undergoing surgery or invasive procedures involving the abdomen or pelvis, including minimally invasive gastrointestinal, gynecologic, and urologic procedures, we suggest first draining the peritoneal fluid (Grade 2C). We also drain the dialysate fluid before dental procedures. The presence of peritoneal dialysis fluid may promote bacterial growth and dilute concentrations of native host defense mechanisms (see 'All procedures' above). The preferred duration of a dry abdomen after colonoscopy, upper endoscopy, and gynecologic procedures varies according to whether a biopsy or polypectomy is performed. (See 'Colonoscopy and upper endoscopy' above and 'Gynecologic procedures' above.)

•Use of prophylactic antibiotics

-Colonoscopy and invasive gynecologic procedures – For patients on peritoneal dialysis undergoing colonoscopy or invasive gynecologic procedures, we suggest prophylactic antibiotics (Grade 2C). Although we also give prophylactic antibiotics prior to upper endoscopy, other experts do not. We administer a single intraperitoneal dose of gentamicin 160 mg in the overnight dwell or in cycler bags one night prior to the procedure, in addition to prescribing oral metronidazole 500 mg to be taken four times daily for three days (starting the day prior to the procedure). (See 'Colonoscopy and upper endoscopy' above and 'Gynecologic procedures' above.)

-Dental procedures – For patients on peritoneal dialysis undergoing invasive dental procedures, we suggest prophylactic antibiotics (Grade 2C). Dental procedures are associated with a high risk of bacteremia and have been associated with peritonitis among patients on peritoneal dialysis. A preferred regimen is a single oral dose of amoxicillin (2 grams) given two hours prior to dental procedures. (See 'Dental procedures' above.)

-Catheter exit-site leaks – For patients with peritoneal catheter exit-site leaks, we suggest prophylactic antibiotic therapy (Grade 2C). Because patients with exit-site leaks are at risk of developing peritonitis, we typically administer antibiotics for the duration of the leak. (See 'Prophylactic antibiotics for catheter exit-site leaks' above.)