Peritoneal Dialysis-Related Infections


peritonitis, infection, microbiology


  • Outline

  • Peritoneal Dialysis-Related Peritonitis, 509

    • Pathogenesis, 509

    • Host Defense Mechanisms of the Peritoneal Cavity, 510

    • Presentation, 511

    • Diagnosis, 511

    • Treatment of Peritonitis, 511

    • Mycobacterial Peritonitis, 515

    • Complications of Peritoneal Dialysis Peritonitis, 515

  • Catheter-Related Infections, 516

  • Prevention, 517

Peritoneal dialysis (PD)-related infection is a broad term that encompasses PD-related peritonitis and catheter-related infections (CRIs), and the latter is used as a collective term to describe exit-site infection (ESI) and tunnel infection. PD-related infections are important complications of PD and could lead to serious consequences, including catheter loss, transfer to hemodialysis, hospitalization, and death. Peritonitis is the leading cause of technique failure in PD patients. Although the mortality of a peritonitis episode is <5%, peritonitis is the major contributing cause of death in >15% of PD patients.

Every PD unit should monitor, at least on a yearly basis, the incidence of peritonitis and CRIs. The latest guideline published by the International Society for Peritoneal Dialysis (ISPD) suggests that rates of peritonitis and CRIs should be reported as the number of episodes per patient-year rather than as number of patient-month per episode, which was commonly used in the past. There is a substantial variation in the peritonitis rate reported by different countries or even different centers within the same country. However, the overall peritonitis rate should be no more than 0.5 episodes per year at risk.

Peritoneal Dialysis-Related Peritonitis


PD-related peritonitis could be caused by touch contamination, catheter-related problems, bowel pathology, gynecological disease, or systemic bacteremia. The common microbiological causes of PD-related peritonitis are summarized in Table 32.1 . Despite the advances in PD system connectology, touch contamination at the time of the PD exchange remains a common cause of peritonitis. Organisms that are commonly grown from specimens in contamination-related peritonitis are coagulase-negative staphylococcal species (CNSS) and diphtheroids ( Corynebacterium species). Nasal carriers of Staphylococcis aureus often have the same bacteria on their hands and at the exit sites, which can lead to peritonitis through either touch contamination or CRI. Organisms found in the oral cavity, such as Streptococcus species, may cause peritonitis by droplet spread or transient bacteremia (e.g., after a dental procedure).

TABLE 32.1

Microbiological Causes of Peritonitis

Bacteria, % 80

  • Gram-positive organisms


  • Gram-negative organisms


  • Polymicrobial

Mycobacterium species, % 1
Fungus, % 2.5
Culture negative, % 15

Approximately 15% to 20% of peritonitis episodes, especially those due to S. aureus or Pseudomonas aeruginosa, are caused by catheter infection. ESIs can spread to involve the catheter tunnel and then the peritoneum. Such infections often are refractory or relapsing. A warm and humid climate may favor the accumulation of sweat and dirt around the catheter exit site, and therefore the growth and colonization of bacteria, resulting in a seasonal variation in the incidence of PD-related peritonitis that peaks in the months that are hot and humid.

Many bacteria can form biofilm on the walls of catheters. Bacteria within the slime layer of biofilm are protected from host defense and antibiotics. Refractory or relapsing peritonitis episodes often are caused by the release of planktonic bacteria from the biofilm. However, biofilm also can be found in asymptomatic PD patients without peritonitis.

Gram-negative bacteria that cause PD-related peritonitis are generally considered to originate from the bowel. Most of these cases are presumably caused by the transmural movement of bacteria rather than perforation. This phenomenon is similar to the scenario of spontaneous bacterial peritonitis in patients with liver cirrhosis. Uremia per se is associated with impaired intestinal barrier function to macromolecules, bacterial fragments, and possibly bacteria. Constipation, diarrhea, and diverticular disease may further increase the risk for such happening. Gastric acid inhibitors have been reported to be associated with gram-negative peritonitis. Surgical diseases of intraabdominal organs also may result in enteric peritonitis. Traditionally, polymicrobial peritonitis is believed to be caused by the perforation of internal viscera, and surgical exploration often is recommended. Recent reports show that many patients with PD-related polymicrobial peritonitis do not have overt surgical pathology in the abdomen and response to antibiotic therapy alone often is satisfactory.

Peritonitis may follow colonoscopy with polypectomy, hysteroscopy, endoscopy with sclerotherapy, and dental procedures. Peritonitis after dental procedures is most likely related to transient bacteremia. Vaginal leak of dialysate, the use of intrauterine devices, and endometrial biopsy are other recognized contributing factors of PD-related peritonitis. Because of the risk for peritonitis related to such procedures, antibiotic prophylaxis administered before any such procedure is necessary. Antibiotic prophylaxis should be given before colonoscopy and invasive gynecological procedures.

Host Defense Mechanisms of the Peritoneal Cavity

Uremia per se causes a wide range of defects in the immunological defense against infection, which is beyond the scope of this chapter. Both humoral and cellular factors participate in the local peritoneal defense processes against peritonitis. Under physiological circumstances, bacteria entering the peritoneal cavity are digested by resident peritoneal macrophages and neutrophils. Individual variation in the phagocyte function may partly account for interindividual variation in the risk for peritonitis.

Humoral Immunity

The concentrations of immunoglobulin G (IgG) and complement in normal peritoneal fluid are similar to serum. In PD effluent, however, these values are reduced by 100- to 1000-fold, even after several hours of dwell time. This dilutional effect severely compromises the humoral immunity within the peritoneal cavity. An inverse relationship between the frequency of PD-related peritonitis and peritoneal opsonic activity or IgG concentration has been reported.

For mechanisms that are not entirely clear, the opsonic activity of spent dialysate against gram-negative bacteria is substantially lower than that against gram-positive bacteria. This may partly account for the greater clinical severity of gram-negative peritonitis. In addition to complement, fibronectin has opsonic activity against gram-positive organisms. Low fibronectin concentrations in spent dialysate is a risk factor of PD-related peritonitis.

Cellular Immunity

The leukocyte count in PD effluent is 100- to 1000-fold less than in normal peritoneal fluid. Macrophage is the predominant cell type in noninfected spent dialysate; lymphocyte percentages vary between 2% and 84%, and polymorphonuclear leukocytes (PMNs) are usually 5% to 10%. However, baseline peritoneal leukocyte count is not associated with the risk for PD peritonitis. When there is peritonitis, neutrophils and macrophages quickly migrate from the systemic circulation and interstitial matrix into the peritoneal cavity.

Lymphocytes and cell-mediated cytotoxicity do not play a significant role in the defense against bacterial peritonitis. On the other hand, resident peritoneal macrophages constitute the first line of defense against bacterial invasion. Peritoneal macrophages probably originate from blood monocytes, and their phagocytic and bacteria-killing capacity from PD patients is normal when incubated in culture media. However, the oxidative metabolism of macrophages is reduced in dialysis solution or after repeated peritonitis episodes. PMN from PD patients exhibit decreased binding of C5a, decreased chemotaxis, and impaired opsonic activity.

Mesothelial cells along the peritoneal membrane also play important roles in the local defense against peritonitis. The interaction between mesothelial cells and peritoneal macrophages early in the course of peritonitis occurs via cell–cell interaction and secretion of various inflammatory mediators.

Effects of Peritoneal Dialysis Solutions on Peritoneal Defense

The effects of commercial PD solutions on peritoneal defense are related to dilution, high osmolality, low pH, lactate, and heat sterilization of the dialysate. In addition to the dilutional effects on intraperitoneal (IP) immunoglobulin and complement levels, decreased density of peritoneal macrophages reduces the phagocyte-bacterium encounter and thus bacterial killing. The high osmolality and low pH of dialysis solution suppress peritoneal PMN and macrophage functions. Although dialysate pH rises to physiological levels within 30 minutes of infusion, the period of low pH coincides with that of a high risk for bacterial entry. Lactate in dialysis solution may have independent adverse effects on peritoneal defense. Some studies showed that biocompatible PD solutions with normal pH and bicarbonate-buffer dialysis solution may reduce the risk for peritonitis. However, in vitro PMN function after incubation in bicarbonate-based dialysis solution also is impaired. A metaanalysis of six randomized controlled trials (RCTs) concluded that the use of neutral pH or bicarbonate-buffered PD solutions had no definite effect on the rate of peritonitis. The effect of glucose polymer–based or other novel PD solutions on leukocyte function has not been well studied.


Patients with peritonitis typically present with cloudy dialysis effluent and abdominal pain. Other common symptoms at presentation include diarrhea, vomiting, chills, and reduction in PD outflow. Although theoretically possible, abdominal pain with clear PD effluent seldom indicates peritonitis. A small but non-negligible group of patients present with cloudy PD effluent but no abdominal pain. The severity of illness varies widely and partly depends on the etiological microorganism. For example, S. epidermidis or diphtheroids often cause minimal abdominal pain. On the other hand, virulent organisms such as S. aureus, P. aeruginosa, and fungi often cause severe abdominal pain and, not uncommonly, diarrhea. Fever and hypotension indicate systemic sepsis, bacteremia, and severe peritonitis.


Most practicing nephrologists can diagnose PD-related peritonitis on clinical grounds. Nonetheless, a standard set of diagnostic criteria has been designed to facilitate dialysis unit audit and assist in making the diagnosis by nonspecialists. The latest ISPD guideline states that peritonitis should be diagnosed when at least two of the following are present: (1) clinical features consistent with peritonitis (i.e., abdominal pain and/or cloudy dialysis effluent); (2) dialysis effluent white cell count >100/μL or >0.1 × 10 9 /L (after a dwell time of ≥2 hours), with >50% PMN; and (3) positive dialysis effluent culture. For patients receiving machine-assisted PD with rapid cycles, the PD effluent white blood cell (WBC) count may be <100/μL during peritonitis. In this circumstance, one should rely on the differential WBC count, and a PMN >50% is indicative of peritonitis.

The differential diagnosis for infectious peritonitis includes chemical peritonitis, peritoneal eosinophilia, hemoperitoneum, pancreatitis, chylous effluent, and malignancy. Peritoneal eosinophilia usually occurs early in the course of PD, resolves spontaneously after 2 to 6 weeks, and is usually not associated with infection. The mechanism is believed to be allergic reaction to the plasticizers on the dialysis tubing. IP administration of amphotericin can cause chemical peritonitis. Sterile chemical peritonitis also has been reported after icodextrin PD solution. The episodes are characterized by mild abdominal discomfort, dialysate leukocytosis with a predominance of macrophages, and the absence of systemic symptoms. PD patients who develop pancreatitis may present with abdominal pain and cloudy dialysis effluent. Typically, culture of the effluent yields no bacterial growth, and the effluent amylase level is >100 U/L. Chylous ascites is a rare cause of sterile cloudy effluent, and the effluent WBC count is normal. Patients with intraabdominal malignancy may have cloudy effluent, and the diagnosis can be established by cytological evaluation.

Treatment of Peritonitis

Initial Evaluation

PD patients presenting with cloudy effluent should be presumed to have peritonitis and should be treated as such until the diagnosis can be confirmed or excluded. In a patient presenting with possible peritonitis, evaluation should include questioning about possible touch contamination, adherence to sterile exchange technique, recent procedures that may lead to peritonitis, and change in bowel habits. A formal root-cause analysis is valuable in preventing further episodes of peritonitis, but it can be performed after the patient improves with treatment. The physician should review any history of recent peritonitis episode so as to determine the possibility of relapsing episode or antibiotic resistance. In addition to the usual physical examination, one must carefully search for coexisting ESI or tunnel infection. PD effluent should be tested for WBC count, differential, and Gram stain and should be cultured whenever peritonitis is suspected. Blood-culture bottle is the preferred technique for bacterial culture of PD effluent. At present, there is insufficient evidence to support the use of novel laboratory and molecular techniques for the diagnosis of peritonitis.

Empirical Therapy

As just emphasized, PD patients presenting with cloudy effluent should be presumed to have peritonitis, and empirical antibiotic therapy should be initiated as soon as possible after appropriate microbiological specimens have been obtained. Many patients with PD-related peritonitis could be managed on an outpatient basis. The decision to hospitalize a patient depends on clinical factors (e.g., severity of signs and symptoms) as well as practical ones (e.g., feasibility of administering IP antibiotics).

The recommended treatment regimen has been described in detail by the latest ISPD guideline. In essence, empirical antibiotic regimens should be center-specific and cover both gram-positive and gram-negative organisms. Gram-positive organisms could be covered by vancomycin or a first-generation cephalosporin, whereas gram-negative organisms could be covered by a third-generation cephalosporin or an aminoglycoside.

For gram-positive coverage, a metaanalysis showed that vancomycin-based regimens result in a higher complete cure rate than first-generation cephalosporin, but there is no difference in the rate of primary treatment failure, relapse, or catheter removal. Theoretical concern of inducing vancomycin resistance would favor the use of first-generation cephalosporin, except for PD units that have a high rate of methicillin-resistant organisms. For the coverage of gram-negative organisms, previous studies showed that aminoglycosides and third- or fourth-generation cephalosporins are equally effective Aminoglycosides are inexpensive, and there is no evidence that short courses of aminoglycosides accelerate the loss of residual renal function. However, repeated or prolonged aminoglycoside treatment (>3 weeks) is associated with a high incidence of vestibular and oto-toxicity and should be avoided. Table 32.2 summarizes the algorithm for the initial assessment and empirical antibiotic therapy of PD-related peritonitis. Suitable agents and dosages are listed in Box 32.1 . In general, IP antibiotics are the preferred route of administration unless the patient has features of systemic sepsis.

TABLE 32.2

Algorithm of the Initial Assessment and Therapy for Peritoneal Dialysis-Related Peritonitis

Goal Steps
Evaluation Clinical evaluation; examine exit site and catheter tunnel.
Collect dialysis effluent for cell count, differential count, Gram stain, and bacterial culture.
Treatment Start intraperitoneal antibiotics as soon as possible
Allow to dwell for at least 6 h
Empirical gram-positive and gram-negative coverage, based on patient history and center sensitivity patterns
Gram-positive coverage:
first-generation cephalosporin or vancomycin
Gram-negative coverage:
third-generation cephalosporin or aminoglycoside
Adjunctive measures Consider adjuvant treatment: pain control; IP heparin; antifungal prophylaxis.
Education and assess IP injection technique.
Ensure follow-up arrangements.

IP , Intraperitoneal.

Adapted from Li PK, Szeto CC, Piraino B, et al. ISPD peritonitis recommendations: 2016 update on prevention and treatment. Perit Dial Int . 2016;36:481-508.

BOX 32.1

  • Gram-positive coverage

    • Cefazolin

      • intermittent: 15–20 mg/kg IP daily

      • continuous: 500 mg/L IP loading, 125 mg/L IP maintenance

      • APD: 20 mg/kg IP daily in long day dwell

    • Vancomycin

      • intermittent: 15–30 mg/kg IP every 5–7 d

      • APD: as above, in long day dwell

  • Gram-negative coverage

    • Ceftazidime

      • continuous: 500 mg/L IP loading, 125 mg/L IP maintenance

      • APD: 1000–1500 mg IP daily in long day dwell

    • Aminoglycoside (gentamicin/netilmicin/tobramycin)

      • intermittent: 0.6 mg/kg IP daily

      • APD: as above, in long day dwell

APD , Automated peritoneal dialysis; IP , intraperitoneal.

Empirical Initial Therapy for Peritoneal Dialysis-Related Peritonitis

Adapted from Li PK, Szeto CC, Piraino B, et al. ISPD peritonitis recommendations: 2016 update on prevention and treatment. Perit Dial Int . 2016;36:481-508.

Practical Aspects of Antibiotic Therapy

For many antibiotics, IP regimen can be given as continuous (i.e., in each exchange) or intermittent dosing (i.e., once daily to once every few days). Because of its concentration-dependent bactericidal effect, IP aminoglycoside should preferably be administered as daily intermittent dosing. There is no evidence that monitoring aminoglycoside levels mitigates toxicity risk or enhances efficacy. IP vancomycin could be administered intermittently, usually at a dosing interval of every 4 to 5 days. Redosing is probably appropriate to keep serum vancomycin levels >15 μg/mL.

For cephalosporins, basic pharmacological principles would favor continuous dosing because this group of agent kills bacteria in a time-dependent manner. However, there are few data to support that continuous dosing is more efficacious than intermittent dosing. Moreover, intermittent dosing regimens are valuable in some situations. For example, some elderly or debilitated patients would require helpers or healthcare visitors for the administration of IP antibiotics, and intermittent dosage may be the only practical solution when the injection could only be performed once or twice daily and drug stability in PD solution is a concern.

Intermittent dosage is also the only possible means of administering antibiotics for patients treated with machine-assisted PD. In this case, the intermittent IP dosing could be given in the day dwell. Because extrapolation of pharmacokinetic data from continuous ambulatory peritoneal dialysis (CAPD) to machine-assisted PD may result in significant underdosing when antibiotics are given IP intermittently, a higher daily dose often is required. Alternatively, patients on machine-assisted PD who develop peritonitis may temporarily switch to CAPD. However, it is not always practical to switch because patients may not be familiar with the exchange technique, and the supplies for CAPD may not be immediately available.

Adjuvant Therapy

Extensive rapid-cycle peritoneal lavage during the first 24 hours of peritonitis does not affect the clinical outcome and is not advisable for routine clinical use. Patients with cloudy PD effluent may benefit from the addition of IP heparin (e.g., 500 units/L) to prevent occlusion of the catheter by fibrin. Antifungal prophylaxis, preferably by oral nystatin, should be given during antibiotic therapy to prevent secondary fungal peritonitis. The benefit of IP urokinase for the treatment of relapsing or refractory peritonitis has not been confirmed.

Because of the systemic inflammatory response and increased peritoneal protein loss, protein-energy wasting develops quickly during PD peritonitis. During peritonitis, ultrafiltration (UF) by PD is usually reduced. Protein-energy wasting should be screened for and appropriate supplement should be considered in patients with prolonged peritoneal inflammation. In patients with diabetes, glycemic control may worsen during peritonitis because of the rapid glucose absorption. Blood glucose monitoring with appropriate adjustments of insulin dosage may be needed. Fluid overload is a frequent complication, and the usual PD regimen may need to be adjusted according to the clinical condition. A recent study suggested that the use of icodextrin solution during acute peritonitis resulted in better UF and fluid control in CAPD patients.

Therapy for Specific Organisms

Gram-Positive Microorganisms

The treatment for gram-positive peritonitis is outlined in Table 32.3 . Peritonitis episodes due to CNSS, S. aureus, Streptococcus, and Enterococcus species are distinctly different in presentation, pathogenesis, and outcome. Treatment must therefore be individualized.

TABLE 32.3

Treatment Regimen for Gram-Positive Peritonitis

Enterococcus Staphylococcus aureus Coagulase-Negative Staphylococcus
At 24–48 h

  • Stop cephalosporins

  • Start IP vancomycin

  • Consider adding IP aminoglycoside

  • Stop ceftazidime or aminoglycoside, continue IP cefazolin

  • Consider adding oral rifampin 450–600 mg/d for 5–7 d

  • If methicillin resistant or clinically not responding, change to IP vancomycin

  • Stop ceftazidime or aminoglycoside, continue IP cefazolin

  • If methicillin resistant or clinically not responding, change to IP vancomycin

Duration of Therapy

  • 21 d

  • 21 d

  • 14 d

IP , Intraperitoneal.

Adapted from Li PK, Szeto CC, Piraino B, et al. ISPD peritonitis recommendations: 2016 update on prevention and treatment. Perit Dial Int . 2016;36:481-508.

Coagulase-Negative Staphylococcal Species

Peritonitis episodes caused by CNSS should be treated with IP first-generation cephalosporins or vancomycin, according to the antimicrobial susceptibility, for 2 weeks. These episodes often are secondary to touch contamination. Exchange technique of the patient and adherence to aseptic technique should be checked. Nowadays, some centers have very high prevalence of methicillin resistance, and vancomycin may need to be considered as empirical therapy. On the other hand, vancomycin may be slightly less effective than first-generation cephalosporins when the bacteria are methicillin-sensitive. Relapsing coagulase-negative staphylococcus peritonitis suggests colonization of the PD catheter with biofilm, and catheter removal should be considered. The benefit of IP urokinase and oral rifampicin for the prevention of relapsing episode remains controversial.

Staphylococcus aureus

Peritonitis episodes caused by S. aureus should be treated with effective antibiotics for 3 weeks. These episodes often are secondary to ESIs or tunnel infections, which should be carefully searched for. Ultrasound study of the catheter tunnel may help to diagnose occult tunnel infections and identify a subgroup of patients who are likely to require catheter removal. Screening for nasal S. aureus carrier also should be considered, especially in patients with relapsing or repeated peritonitis episodes caused by S. aureus.

Two retrospective studies found that the initial empiric antibiotic choice between vancomycin and cefazolin had similar clinical outcomes. One study showed that the use of adjuvant rifampicin for 5 to 7 days may reduce the risk for relapsing or repeat S. aureus peritonitis. However, rifampicin is a potent liver enzyme inducer and interaction with other concomitant medications may be problematic. For patients with concomitant S. aureus ESI or catheter tunnel infection, catheter removal should be considered.

Streptococcal Species

Streptococci frequently originate from the mouth, although Streptococcus bovis typically comes from the colon. Peritonitis episodes caused by streptococci usually respond well to IP cefazolin or vancomycin treatment. Patients often are advised to wear a face mask during PD exchange to prevent further episodes of streptococcal peritonitis, but there is no published evidence to support this practice.

Enterococcus Species

Peritonitis episodes caused by Enterococcus species are usually severe. In about half of the cases of enterococcal peritonitis, other coexisting organisms could be isolated, and intraabdominal pathology must be considered. Since enterococcal species have intrinsic resistance to cephalosporins, and ampicillin may be unstable in PD solution, enterococcal peritonitis should be treated with IP vancomycin for at least 3 weeks. IP aminoglycoside could be added as adjunct for severe cases. Treatment of peritonitis episodes caused by vancomycin-resistant enterococcus should be individualized. Linezolid, quinupristin/dalfopristin, daptomycin, or teicoplanin are all valid options.

Gram-Negative Organisms

Peritonitis due to gram-negative organisms often is associated with fever, nausea, vomiting, and abdominal pain. The treatment of gram-negative peritonitis is summarized in Table 32.4 . Satisfactory therapeutic responses have been reported with either IP aminoglycoside or ceftazidime. Other third- or fourth-generation cephalosporins are probably equally effective. Alternatively, fluoroquinolones, such as levofloxacin, ciprofloxacin, and moxifloxacin, have the advantage of good oral bioavailability, and can be used with acceptable result.

TABLE 32.4

Treatment Regimen for Gram-Negative Peritonitis

Single Gram-Negative Organism Pseudomonas Species Multiple Gram-Negative or Anaerobes
At 24–48 h

  • Stop cefazolin or vancomycin

  • Continue IP ceftazidime or aminoglycoside

  • Adjust antibiotics according to sensitivity

  • Continue ceftazidime or aminoglycoside

  • Treat with both IP ceftazidime and IP aminoglycoside

  • Adjust antibiotics according tosensitivity

  • Stop cefazolin

  • Continue IP ceftazidime or aminoglycoside

  • Start IP vancomycin and oral or intravenous metronidazole

  • Consider surgical evaluation

Duration of Therapy

  • 14–21 d

  • 21 d

  • 21 d

IP , Intraperitoneal.

Adapted from Li PK, Szeto CC, Piraino B, et al. ISPD peritonitis recommendations: 2016 update on prevention and treatment. Perit Dial Int . 2016;36:481-508.

Pseudomonas Species

Recent antibiotic therapy is the major risk factor of Pseudomonas peritonitis. Patients with immunosuppression are also at higher risk for Pseudomonas peritonitis. Pseudomonas peritonitis should be treated with two antibiotics (e.g., IP gentamicin or oral ciprofloxacin with IP ceftazidime or cefepime) for 3 weeks. When there is concomitant ESI or tunnel infection, catheter removal is usually necessary. In this circumstance, simultaneous removal and reinsertion of a new PD catheter (with a new tunnel and exit site) may avoid the need of temporary hemodialysis and often is successful in preventing further relapsing episodes.

Other Gram-Negative Bacteria

Single-organism gram-negative peritonitis may be secondary to touch contamination, ESI, or transmural migration from constipation or colitis. In general, non- Pseudomonas gram-negative peritonitis should be treated with effective antibiotics for at least 3 weeks. A retrospective study suggests that treatment with two antibiotics may reduce the risk for relapse and recurrence compared with treatment by a single agent. In recent years, there has been an increase in incidence and probably recognition of peritonitis episodes caused by gram-negative bacilli with extended-spectrum beta-lactamases and carbapenem-resistant Enterobacteriaceae . Their treatment should be individualized. A number of new second-line antibiotics, such as ceftaroline, ceftolozane plus tazobactam, and tigecycline, are potentially effective.

Polymicrobial Peritonitis

When multiple enteric organisms (multiple gram-negative or mixed gram-negative and gram-positive organisms) are grown from the PD effluent, it often indicates the presence of intraabdominal pathology. Treatment should include oral or intravenous metronidazole plus IP vancomycin and either IP aminoglycoside or IP ceftazidime for a minimum period of 3 weeks. Similarly, the presence of anaerobic bacteria or fungus as part of the mixed microbial growth from the PD effluent usually suggests underlying surgical pathology. On the other hand, if multiple gram-positive organisms are grown from PD effluent, intraabdominal pathology is less likely and the prognosis is usually favorable. These patients should be treated with effective antibiotics for 3 weeks. Routine surgical intervention for this group of patients is often not necessary.

Culture-Negative Peritonitis

In approximately 15% of episodes that meet the criteria for peritonitis on the basis of cell count and clinical features, culture of the PD effluent would not yield any pathogenic organisms. Most of the culture-negative peritonitis could be explained by recent antibiotic therapy or technical problems during the collection of PD effluent for culture. Negative PD effluent cultures on day 3 warrant a repeat dialysis effluent WBC count with differential. If the culture-negative peritonitis is resolving at that time, the empirical antibiotic that covers gram-positive organisms (i.e., first-generation cephalosporin or vancomycin) should be continued for 2 weeks, although it remains controversial whether the empirical antibiotic that covers gram-negative organisms should be stopped. In general, IP aminoglycoside should be discontinued to avoid extended or repeated exposure, whereas third-generation cephalosporins may be continued, especially in centers with a high background incidence of gram-negative peritonitis. If the culture-negative peritonitis is not resolving on day 3, special culture techniques should be considered for isolation of unusual organisms. Fungus and mycobacterial species may not be recovered by routine bacterial culture of PD effluent and should also be considered as the causative organisms.


Fungal peritonitis occurs in patients undergoing PD at the rate of 0.01 to 0.19 episodes per dialysis-year, accounting for 3% to 6% of episodes. More than 70% of the episodes of fungal peritonitis are caused by Candida species. Recent antibiotic therapy, frequent episodes of bacterial peritonitis, and immunosuppression are the major risk factors of fungal peritonitis. Patients often are severely ill with marked abdominal tenderness.

Catheter removal is indicated once fungi are identified in PD effluent, and an appropriate antifungal agent should be continued for at least 2 weeks after catheter removal. Traditionally, the therapy for fungal peritonitis is a combination of amphotericin B and flucytosine. However, this regimen has a high incidence of adverse effects, and the response often is less than satisfactory because amphotericin B has poor penetration through the peritoneal membrane. Fluconazole is commonly used for peritonitis episodes caused by many Candida species and Cryptococcus, but this agent is fungistatic. Echinocandins (especially caspofungin) is now often advocated for the treatment of fungal peritonitis caused by Aspergillus species and nonalbicans Candida species, whereas posaconazole and voriconazole are reasonable choices for episodes caused by other filamentous fungi. However, it should be noted that intravenous voriconazole is contraindicated in dialysis patients because it contains cyclodextrin in the solvent and may cause neurotoxicity in patients with severe renal impairment.

Mycobacterial Peritonitis

Tuberculous Peritonitis

Tuberculous peritonitis is uncommon in the Western world, but is more common in Asian countries. Contrary to the common belief, the WBCs in the effluent, at least during the early phase of peritonitis, are predominantly PMN cells, and an acid-fast bacilli stain of an effluent specimen is usually negative. Abnormal chest radiograph and ascitic fluid lymphocytosis only identify one-third of the cases. Conventional microbiological diagnostic methods are slow and have a low sensitivity. New fluid medium (e.g., Septi-Chek, BACTEC; Becton Dickinson) could reduce the time to develop and positive culture. Overall diagnostic yield could further be improved by centrifuging 50 to 100 mL effluent, followed by culturing the sediment in both solid and fluid media. Alternatively, mycobacterial DNA polymerase chain reaction can be performed on dialysis effluent, although false-positives are common. Peritoneal biopsy often is recommended for expedite diagnosis of tuberculous peritonitis in clinically suspicious cases, but this strategy is not practical for PD patients.

Regarding the treatment, standard antituberculous chemotherapy is highly effective, but it takes at least 12 months to complete the course. Drug interactions between rifampicin and other concomitant medications are common. Removal of PD catheter is usually not necessary, but UF problems may occur as a delayed complication.

Nontuberculous Mycobacterial Peritonitis

In subtropical countries, there is an increase in incidence of peritonitis caused by nontuberculous mycobacteria in the past decade. More than half of the cases are caused by rapidly growing species, such as Mycobacterium fortuitum and M. chelonae. It has been postulated that extensive use of topical antimicrobial ointment (especially gentamicin) for exit-site care may predispose patients to nontuberculous mycobacterial infection of the exit site. The treatment regimen for nontuberculous mycobacterial peritonitis is not well established, but catheter removal is usually necessary.

Reassessment After Therapy

Most patients with PD-related peritonitis show considerable clinical improvement after 2 to 3 days of antibiotic therapy. Antibiotic therapy should be adjusted to narrow spectrum agents, as appropriate, once culture results and sensitivities are known. If the patient has not shown definitive clinical improvement after 3 days of empirical therapy, effluent cell counts, Gram stain, and cultures should be repeated. PD effluent WBC count ≥1090/mm 3 on day 3 predicts treatment failure. Theoretically, catheter removal should be considered if the response to antibiotic therapy is poor after 4 days of antibiotic therapy (see next section), but this recommendation is difficult to follow in practice as it is limited by the availability of surgeon, operating theatre, as well as patients’ own preference.

Catheter Removal

The definitions of refractory, relapsing, recurrent, and repeat peritonitis are summarized in Table 32.5 . PD catheter should be removed promptly in refractory peritonitis. Timely catheter removal also should be considered for relapsing peritonitis, refractory ESI and tunnel infection, and fungal peritonitis. In most patients, temporary hemodialysis is needed after catheter removal. However, simultaneous removal and reinsertion of catheters is a safe and effective method for the treatment of refractory ESI as well as for the prevention of relapsing peritonitis after dialysis effluent clears up.

Feb 24, 2019 | Posted by in NEPHROLOGY | Comments Off on Peritoneal Dialysis-Related Infections

Full access? Get Clinical Tree

Get Clinical Tree app for offline access