| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Part 8: PD Peritonitis |
Center for Health and Biological Sciences,1 Pontifícia Universidade Católica do Paraná, Curitiba, and Department of Internal Medicine,2 University Hospital, Botucatu School of Medicine, UNESP, Sao Paulo, Brazil
Correspondence to: R. Pecoits–Filho, Centro de Ciências Biológicas e da Saúde, Rua Imaculada Conceição, 1155, Curitiba PR 80215 Brazil. r.pecoits{at}pucpr.br
 |
ABSTRACT |
|---|
 TOP ABSTRACT GEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
Despite improvements in connectology, peritoneal dialysis
(PD)–associated peritonitis contributes significantly to morbidity and
modality failure in patients maintained on PD therapy. A broad spectrum of
organisms—gram-positive, gram-negative, fungal, anaerobic—are
involved in this complication. In addition, a significant percentage of
episodes involve polymicrobial and culture-negative infection. Technological
advances are being developed to minimize the incidence of access-related
complications such as peritonitis. Many traditional factors such as exit-site
infection and poor technique have been already identified. In the present
review, we discuss the geographic, patient selection, and clinical issues that
can affect peritonitis rates in different areas of the world and in different
centers in the same area.
KEY WORDS: Peritonitis; infection; risk factors; complications.
The growth and clinical results of peritoneal dialysis (PD) have historically been challenged by peritonitis, which, despite significant improvement over the years, continues to be a major cause of technique failure and patient morbidity and mortality (1–3). The organisms that typically cause peritonitis and the peritonitis rate vary around the world and even in different centers in the same geographic region, but the reasons for these variations are relatively unexplored.
The causes of PD-related peritonitis include mainly gram-positive, gram-negative, fungal, and anaerobic organisms (3), but culture-negative peritonitis is also relatively commonly seen. The characteristics of the local flora and the rate of bacterial recovery in culture both potentially affect the results of peritonitis treatment. Understanding the factors that predispose to peritonitis in a particular region or center can lead to the development of specific prevention strategies, which can lead in turn to a significant improvements in clinical results.
In the present article, we review the geographic, patient selection, and clinical practice issues potentially responsible for regional differences in peritonitis rates and patterns. We also propose an approach to improving clinical results based on an understanding of those regional differences.
|
GEOGRAPHIC ISSUES |
|---|
|
TOPABSTRACTGEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
A more consistently described risk factor for peritonitis is the local weather. A study performed in Hong Kong found substantial seasonal variation in the incidence of PD-related peritonitis, which peaked during the warm and humid months. This seasonal variation was present in peritonitis episodes caused by both gram-positive and gram-negative organisms, and the monthly peritonitis rate was significantly correlated with average temperature and, more strikingly, with humidity (5). Accordingly, a study analyzing infection over a wide range of temperature and humidity (6) showed a clear correlation between peritonitis incidence and climatic factors. A Brazilian study (7) also reported that catheter-related infections were more frequent in warmer months. On the other hand, a North American study (8) found no monthly or seasonal differences in gram-positive and gram-negative peritonitis, but a greater frequency of culture-negative episodes during the hot and humid months.
A more humid environment appears not only to favor the accumulation of sweat and dirt around the catheter exit site, but also to enhance the growth and persistence of bacteria on dialysis tubing and other environmental reservoirs. Strategies aiming to reduce humidity and heat may lead to a reduction in the peritonitis rate.
|
PATIENT SELECTION ISSUES |
|---|
|
TOPABSTRACTGEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
First, peritonitis rates can be influenced by various patient comorbidities such as advanced age, presence of diabetes, and disorders of nutrition (9). The influence of age on peritonitis incidence has been reported by several authors. Although most studies showed a higher peritonitis rate in older patients (10), a recent study (11) reported that the patient's age at the start of dialysis therapy is not predictive of peritonitis incidence. Another study (12) evaluated the impact of diabetes mellitus on the peritonitis risk and found no differences between patients with and without diabetes. In contrast, diabetic patients were observed to have a higher risk for a first peritonitis episode, with diabetes being an independent factor associated with that risk (11). Hypoalbuminemia has been similarly associated with higher rates of peritonitis (11), suggesting that immunologic impairment related to malnutrition may lead to a higher peritonitis risk (13). Previous studies showed an association between peritonitis incidence and nutrition status measured using anthropometric data (14) and creatinine appearance rate (15). However, another study (16) observed that patients with a higher body mass are exposed to a higher risk of peritonitis, suggesting that obesity is also involved in peritonitis risk.
Taken together, the foregoing findings suggest that systemic (and perhaps intraperitoneal) inflammation can be a major factor influencing peritonitis occurrence. Indeed, we recently observed that patients with higher levels of plasma and dialysate interleukin-6 at the initiation of PD presented a higher risk of developing peritonitis during the first year on dialysis (17).
Socioeconomic condition has been also consistently described as a factor determining peritonitis risk, although isolated analysis of the impact of poverty has not been extensively studied. In most studies, racial factors can confound the results. For example, with demographic and economic factors and comorbidities controlled for, the time to development of a first episode of peritonitis was significantly longer in African-American patients than in Caucasian patients (18). Nevertheless, some authors disagree with those data, having found no differences between races (19). An Australian study (20) observed that the time to development of a first episode of peritonitis was shorter in Aboriginal than in non-Aboriginal patients. In addition, a recent study (11) compared patients with and without social security and observed a significant difference between the two groups. Compared with patients having social security, patients lacking social security had 3 times the chance of developing peritonitis. In Brazil, education level and family income are important factors, with the probability of developing peritonitis being lower in patients with a higher educational level or a higher family income (21).
Larger, controlled studies should be developed to more clearly identify the racial or socioeconomic differences (if any) that are risk factors for peritonitis. More importantly, studies need to determine if alternative education and training programs adapted to the local reality will be effective in reducing the peritonitis risk in PD patients with various socioeconomic and ethnic backgrounds.
|
CLINICAL PRACTICE ISSUES |
|---|
|
TOPABSTRACTGEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
In general, centers with many years of experience in the care of PD patients are able to adopt more effective approaches for minimizing the incidence of peritonitis. Indeed these dialysis units typically have effective patient education and training programs that play a key role in peritonitis prevention. Center experience is a non-measurable factor that can influence the peritonitis incidence and partly explain the differences between countries and units (22).
By far, the most important advance in PD technology (for its effect on peritonitis rates) is better connectology. A dramatic reduction in peritonitis incidence followed the switch from glass bottles to plastic bags and the implementation of the closed-system Y-set in daily practice. Further reductions accompanied the introduction of completely dischargeable systems such as twin bags (22). A recent systematic review of 485 patients (23) observed that the Y-set and twin-bag systems were superior to conventional spike systems in the prevention of peritonitis in PD. Differences between regions and centers in the incorporation of safer connection systems may possibly contribute to variations in the peritonitis rate.
Also of importance are the historical changes observed in peritonitis rates and profiles of the relevant flora. During the 1990s, the incidence of peritonitis declined significantly. That change was attributed to improvements in connectology, development of new dialysis solutions, improvements in living environments, and education of patients (24). Several factors—such as exit-site infection and catheter colonization by coagulase-negative staphylococci—are thought to contribute to the development of infection (2). Recently, improvements and newer techniques in exit-site care have led to a decline in the incidence of peritonitis caused by both S. aureus and S. epidermidis (3). A study performed in a single center in China (24) observed, over a decade, a decline in the incidence of peritonitis attributable to a single gram-positive organism. The authors proposed that their findings reflected mainly a decrease in coagulase-negative staphylococcus peritonitis associated with use of the double bag.
Polymicrobial peritonitis is an infrequent complication, usually occurring late in the course of PD in patients with recurrent infections and a high peritonitis rate, most of them involving gram-negative or fungal pathogens (or both). Most of these episodes appear unrelated to an intra-abdominal process (6). Fungal peritonitis is an uncommon but potentially life-threatening complication in which Candida species are the most common pathogens, accounting for 67% of episodes (25). Culture-negative peritonitis is also still frequently reported; however, improvements in culture technique have been associated with a decline in its prevalence (3).
The International Society for Peritoneal Dialysis has developed guidelines on the treatment of PD-related peritonitis, which have been important in suggesting appropriate treatment and, more recently, in stimulating prevention. The cornerstone of peritonitis management is broad empiric antibiotic coverage for both gram-positive and gram-negative bacteria (3). The large variation in recommended antibiotics seen in the most recent guidelines confirms that no single treatment protocol is applicable in every PD center. Whether a particular protocol is suitable for a given center depends on the epidemiology and susceptibility patterns in the causative micro-organisms and on the occurrence or emergence of multiresistant strains. Thus, initial treatment should be adapted to the needs of each center (26). Potentially, centers using antibiotic regimens compatible with the local causative organisms and their drug susceptibilities will present lower incidences of repeat or relapsing peritonitis, and therefore lower peritonitis rates overall. Locally based schedules should be developed as a strategy to improve care for PD-related peritonitis.
A major factor associated with the reduction in peritonitis rates has been the introduction of local antibiotic application to the exit site as a preventive measure. Numerous studies evaluating the efficacy of intranasal or exit-site mupirocin application in the dialysis population have been performed. Most of these studies demonstrated reduction in the risk of S. aureus peritonitis and exit-site infection in mupirocin-treated patients (27). However re-colonization is common, resulting in the need for extended treatment, and hence mupirocin-resistant strains have been described (28). Because mupirocin is not effective against gram-negative infections, gentamicin cream applied to the catheter exit site was recently reported to be effective in reducing peritonitis rates, particularly peritonitis from gram-negative organisms (29). Thus, centers routinely using prophylactic protocols tend to have lower rates of exit-site infections and peritonitis.
Comparisons of peritonitis rates by PD modality have produced conflicting results. Studies dating from the 1990s reported a lower incidence in automated PD than in continuous ambulatory PD (30), but a more recent study (31) found that PD performed with a cycler was associated with a slight but significantly higher risk of peritonitis. Future studies will need to clarify whether differences in the availability of these methods can in part explain disparities in the peritonitis rate between geographic regions.
Finally, the availability of alternative solutions may potentially influence peritonitis rates. Conventional solutions have an acidic pH, high concentrations of glucose (with associated hyperosmolarity) and lactate, and the presence of glucose degradation products (GDPs). This non-physiologic composition potentially impairs the peritoneal host defense and may therefore contribute to the development of PD-related peritonitis. Thus, the availability of alternative solutions in different geographic regions may influence the incidence of infections and explain varying peritonitis rates. However comparisons between conventional solutions and those with low GDPs and neutral pH have shown no differences with regard to peritonitis incidence (32,33).
|
CONCLUSIONS |
|---|
|
TOPABSTRACTGEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
|
|
REFERENCES |
|---|
|
TOPABSTRACTGEOGRAPHIC ISSUESPATIENT SELECTION ISSUESCLINICAL PRACTICE ISSUESCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
