| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Part 5: Peritonitis |
Department of Nephrology, Dialysis and Internal Diseases, The Medical University of Warsaw, Warsaw, Poland
Correspondence to: J. Matuszkiewicz-Rowinska, Department of Nephrology, Dialysis and Internal Diseases, The Medical University of Warsaw, ul. Banacha 1a, Warsaw 02-097 Poland. jotmatrow{at}o2.pl
 |
ABSTRACT |
|---|
 TOP ABSTRACT RISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
Fungal peritonitis (FP) is a rare but potentially fatal complication of
chronic peritoneal dialysis (PD), associated with high morbidity and mortality
ranging between 20% and 30%. If not leading to death, the inflammatory process
usually causes irreversible damage to the peritoneal membrane with subsequent
dropout from PD therapy. Fungal peritonitis accounts for 3% – 6% of all
peritonitis episodes; however, in some areas, the numbers can be much higher.
The most common cause of the disease is Candida, predominately C.
albicans, C. parapsilosis, and—more recently— C.
glabrata; other yeasts and filamentous fungi such as Aspergillus,
Paecilomyces, Penicillium, and Zygomycetes are found, but much
less frequently. The main factors associated with the development of FP
include previous antibiotic therapy, particularly for bacterial peritonitis,
when two important operative mechanisms coincide: fungal overgrowth in the
gastrointestinal tract and declining peritoneal defense because of
peritonitis.
The management of FP poses a difficult challenge. Prompt initiation of
therapy is critical, but no typical clinical picture has emerged, and the
infecting organism can be difficult to isolate. The approach to the disease
has changed considerably in recent years, and the 2005 guidelines from the
International Society for Peritoneal Dialysis list FP as a strong indication
for immediate catheter removal with temporary hemodialysis. The conventional
antifungal regimens include fluconazole, amphotericin B, and flucytosine alone
or in combination, optimally based on fungal sensitivities. The newer agents
such as caspofungin and voriconazole have the potential to alter treatment
strategies for FP, but further studies are required to clarify the precise
role of these agents in this group of patients.
KEY WORDS: Fungal peritonitis; peritonitis prophylaxis; peritoneal catheter removal.
Fungal peritonitis (FP) is a relatively rare but very serious disease. In most studies, it accounts for 3% – 6% of all peritonitis episodes complicating peritoneal dialysis (PD) in adults; however, in some centers, the numbers can be higher. Mortality rates are very high, varying between centers in the range of 15% to 50%. Fungal peritonitis is associated with a considerable risk of technique failure, mainly through loss of peritoneal membrane function, with up to 40% of patients being unable to resume PD and needing definite conversion to hemodialysis because of peritoneal adhesions, sclerosis, and irreversible membrane damage. In cases involving C. parapsilosis and some filamentous fungi, the mortality and technique failure percentages are higher.
Most FP cases are caused by yeasts, with Candida species accounting for 70% – 90% in adults and 80% – 100% in the pediatric population. Filamentous fungi (moulds) such as Aspergillus, Penicillium, and the other yeasts are much less common, and together represent about 10% of cases. In most centers, C. albicans is still predominant as a single pathogen; however, four other major Candida species—C. glabrata, C. parapsilosis, C. tropicalis, and C. krusei—are becoming increasingly recognized causes of FP. Some investigators have already found cases of C. parapsilosis to outnumber cases of C. albicans (1,2).
|
RISK FACTORS |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
Antibiotics, by killing the bowel flora, promote fungal colonization of the intestinal or genitourinary tract, with subsequent invasion across the mucosal barrier to the peritoneal cavity. Ongoing inflammation makes the peritoneum more susceptible to that process. Other suggested risk factors are immunosuppression and malnutrition, bowel perforation, diverticulitis (a common cause of polymicrobial peritonitis), and possibly, certain comorbidities such as diabetes or neoplastic diseases. In women, vaginal candidiasis should be considered as a possible source of fungi. Among reported interesting and rare causes of FP was an outbreak of peritonitis resulting from fecal carriage of C. parapsilosis (3).
If bacterial peritonitis is a risk factor, a question then arises: Does the risk vary with different bacterial pathogens? The answer is not clear, but the suggestion has been made that gram-negative infections may be associated with a greater risk of subsequent FP. Chou et al. retrospectively analyzed 216 cases of catheter-related peritonitis in 123 dialyzed patients (4). In 19% of those cases, a peritonitis episode was followed by FP within the next 6 months. When analyzed by cause, risk of FP was significantly higher after polymicrobial or gram-negative infection (22% and 42.1% respectively vs 4.7% after gram-positive and 5.8% after culture-negative peritonitis; p < 0.0001). The 26% mortality rate and 70% technique failure rate were significantly worse than the outcomes seen in patients with bacterial peritonitis only (7.2% and 14.5% respectively; p < 0.001).
Those results were recently confirmed in a Dutch pediatric PD population (5). The authors of the Dutch study compared previous episodes of various bacterial peritonitis episodes in two groups: all studied patients and those who developed FP. They found twice as many gram-negative organisms in the FP group (46% vs 23%). The authors speculated that the antibiotics used for gram-negative infections created more severe disturbances in the normal intestinal flora. This issue needs to be clarified.
Several studies have analyzed the risk factors for FP, but few of them have addressed the predictors of poor outcome of this complication. In those studies, a higher mortality rate and a greater risk of technique failure were associated with abdominal pain, bowel obstruction, and the catheter remaining in situ (6–8).
|
DIAGNOSIS |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
Gram stain of peritoneal fluid can help to establish an early diagnosis in up to 30% of cases. However, the only fungi that are usually gram-positive on a smear are Candida species. Growth rates for fungal cultures are generally slow, varying from several days to several weeks, and therefore the diagnosis can be considerably delayed. The results of serologic tests for systemic fungal pathogens are often difficult to interpret and may be unreliable. Recent diagnostic advances include detection of fungal cell-wall components such as galactomannan, β-D-glucan, or genomic DNA amplified by polymerase chain reaction. Those tests may improve diagnostic ability in the future, but they are still under investigation.
Currently, therefore, the initial diagnosis of FP is often based on a high index of clinical suspicion (culture-negative results, failure of antibiotic therapy), especially in a patient with predisposing factors.
|
PROPHYLAXIS |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
|
The only randomized study to date was reported by the Lo group (12). They prospectively studied 397 continuous ambulatory PD (CAPD) patients, randomized into two groups (with and without nystatin prophylaxis) for 2 years. Nystatin tablets 500 000 U were given 4 times daily concurrently with any antibiotic therapy, whatever the cause of the infection. This strategy resulted in a significant reduction in the incidence of all FP episodes, and the probability of FP-free 2-year survival was significantly higher in the nystatin group (0.974 vs 0.915; p < 0.05). Surprisingly, however, no important difference in the rate of antibiotic-related FP was observed between the studied groups.
Recently Wong et al. (15) found a significant reduction in the incidence of antibiotic-related FP in nystatin-treated patients compared with a control group (1 FP episode in 3431 patient–months vs in 888 patient–months, p < 0.05). The overall cumulative probability of antibiotic-related FP-free survival in 4 years was 0.947 in the treated group and 0.991 in the control group; the effect persisted even after adjustment for the peritonitis rate. In that study, the FP rate in the nystatin group was only slightly and insignificantly lower than that in the control group. The patient cohort was large, and importantly, the baseline incidence of FP was relatively low. However, the study was again retrospective and historically controlled, and additionally, most of the patients in the control group were using a disconnect system (34% vs 92% in the nystatin group). The improved connectology in the treated group could obviously be a potent confounder.
Until the results of larger randomized trials become available, the guidelines from the International Society for Peritoneal Dialysis (ISPD) recommend antifungal prophylaxis only in centers with high FP rates. For programs with lower rates, there is no clear message, and decisions have to be made on an individual basis.
|
MANAGEMENT |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
Another issue is the management of the patient who refuses the catheter removal or whose bad general condition makes removal infeasible. Recently, Boer et al. presented an interesting approach to that problem: they successfully treated 8 episodes of Candida peritonitis with the catheter remaining in situ (18). They used standard antifungal therapy (oral flucytosine 500 mg twice daily and intraperitoneal fluconazole 150 mg every other day) with additional intra-catheter instillation of amphotericin B lock solution. Patients being treated with automated PD were switched to CAPD. After every exchange, the catheter and connecting system were filled with 10 mL 0.1 mg/mL amphotericin B solution, which was then flushed out with the next drain. In this way, the catheter lumen was exposed to a high concentration of the antibiotic for almost 24 hours daily. Treatment was continued for 4 weeks. No treatment failures or relapses occurred during follow-up (1 – 7 years). If this strategy proves to be successful in larger groups of patients, it can become an interesting option—at least in selected cases.
Antifungal Therapy: No uniform recommendations have been established regarding drug selection or doses and combinations. Choice of therapy is based on case series, anecdotal reports, and opinions. To be effective, treatment should be provided early and often empirically. Until the diagnosis of FP is validated, antimycotics can be given by all three routes: intraperitoneal, intravenous, or oral. After the catheter is removed, the switch to systemic therapy is necessary.
For initial therapy, the ISPD guidelines recommend administration of amphotericin B combined with flucytosine, which can be subsequently replaced by newer agents, according to species identification and minimum inhibitory concentration (17). However, amphotericin B given intravenously becomes more than 90% protein-bound and diffuses poorly from blood into peritoneal fluid, and intraperitoneal administration should be avoided because of peritoneal irritation that may produce severe abdominal pain (19). In this situation, some centers prefer to start therapy with fluconazole, either alone or in combination with flucytosine, leaving amphotericin for selected patients (for example, those who are immunosuppressed or who have had significant prior exposure to azoles). The doses of intravenous amphotericin B vary from 0.5 mg/kg to 1 mg/kg daily depending on the pathogen (yeasts vs moulds).
Fluconazole may be given intraperitoneally (200 mg in 1 exchange every 24 – 48 hours) or intravenously or orally (100 – 200 mg daily), because it is well absorbed from the gastrointestinal tract and has good intraperitoneal penetration. However, because the drug is very popular, resistance may eventually limit its use, and futhermore, fluconazole may be not a good option for treatment of non-Candida species. In these cases, a switch to amphotericin or a newer azole may be required.
Flucytosine has very good peritoneal penetration after oral dosage, but the risk of development of resistance is so high that the drug has to be used in combination with other agents. Because of bone marrow toxicity and hepatotoxicity, flucytosine should not be used in patients with bone marrow depression or hematologic disease, or in those taking drugs that suppress bone marrow. During treatment, regular monitoring of serum drug concentration and hematologic and hepatic status is necessary. The usual oral dose is 500 mg twice daily.
The role of the newer antifungal agents such as the extended-spectrum azoles or caspofungin in PD-related FP has not yet been defined, but the improved safety profiles and broad-spectrum activity of these drugs may potentially change treatment strategies. In particular, these agents may represent an important therapeutic alternative to amphotericin B in patients with peritonitis resulting from infection with moulds and some non-Candida yeasts.
Experiences with the newer triazole compounds such as voriconazole and posaconazole are very limited. The best known of these agents, voriconazole, is available for both oral and intravenous administration and offers activity against Candida and Aspergillus species and many other filamentous fungi except Zygomycetes. Fortunately, in the latter case, oral posaconazole has proven effective. Both agents are rather well tolerated, although it must be remembered that they carry a risk of hepatotoxicity. Careful monitoring of liver function is therefore recommended in treated patients. Additionally, voriconazole can cause adverse visual events and rash.
Three cases of successful voriconazole therapy have been described (20–22). In one case, the drug was given (6 mg/kg loading dose every 12 hours the first day, followed by 4 mg/kg as a maintenance dose intravenously every 12 hours) to a patient transplanted and immunosuppressed during an active infection caused by a Fusarium species (21). Recently, Sedlacek et al. reported a case of successful 6-month posaconazole therapy (800 mg daily in 2 doses) in a patient with liposomal amphotericin B–resistant PD-related Mucor peritonitis (23).
Caspofungin, micafungin, and anidulafungin represent a new class of antifungals, namely echinocandins. With their high clinical efficacy, broad-spectrum activity against Candida species and several moulds, and low rate of treatment-related adverse events, these drugs may become an attractive treatment option in patients with FP attributable to Aspergillus and non-responding non-albicans Candida, or in patients intolerant to other antifungal therapies. But again, only anecdotal reports are available concerning the use of these agents in the PD population. One report described a patient intolerant to amphotericin B in whom caspofungin was used as monotherapy (70 mg intravenous loading dose, then 50 mg daily), and a second depicted an infection refractory to amphotericin B, for which caspofungin (50 mg daily) was given as an additional drug (24,25). Both patients successfully completed a 2-week course of treatment without any adverse events. In two other cases reported by Molina et al., the results of caspofungin therapy (combined with fluconazole and flucytosine) were disappointing: although peritonitis resolved, both patients could not resume PD (26). There is also one other report about successful treatment of severe C. parapsilosis peritonitis refractory to conventional therapy treated with micafungin (27).
The ISPD has given no clear message regarding the length of therapy for dialysis-related FP and the optimal timing of peritoneal catheter replacement. Treatment for at least 10 days after catheter removal is recommended; however, in many cases, treatment is continued for 4 – 6 weeks.
|
SUMMARY |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTRISK FACTORSDIAGNOSISPROPHYLAXISMANAGEMENTSUMMARYREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
