Staphylococcal peritonitis in patients on continuous peritoneal dialysis

Initial Treatment of Dialysis Associated Peritonitis: A Controlled Trial of Vancomycin versus Cefazolin

Objective

To determine if intraperitoneal administration of vancomycin (a slowly absorbed antibiotic) im proves the management of dialysis-associated peritonitis over that obtained by using cefazolin, an equally potent, rapidly absorbed antibiotic.

Setting

A university operated teaching hospital, with patient treatment initiated at home.

Patients

One hundred thirty-one patients trained to perform peritoneal dialysis (CAPD and CCPD) and followed at the University of Iowa Hospitals and Clinics Home Dialysis Treatment Center.

Design

Patients were prospectively allocated into groups adding either vancomycin 25 mgm/L, orcefazolin 50 mgm/L to their dialysate when signs or symptoms of peritonitis developed. Treatment results were analysed using chi-square testing.

Findings

Compared to cefazolin, initial peritonitis therapy with vancomycin improved the peritonitis resolution rate [67% vs 81 %; p=0.008], reduced the incidence of hospital admissions [68% vs 48%; p=0.001], and decreased the risk of superinfection [4% vs 0%; p=0.039].

Conclusion

Vancomycin appeared to be superior to cefazolin in the treatment of peritoneal dialysis associated peritonitis.

Management of Refractory Peritonitis to Maintain the Peritoneum for Subsequent Dialysis

Peritonitis and its sequelae remain major clinical problems in treating peritoneal dialysis (PD) patients. One of these sequelae is the formation of intra-abdominal adhesions, preventing a patient from returning to peritoneal dialysis after a Tenckhoff catheter is removed for refractory peritonitis. We have recently applied a technique that appears to reduce the incidence of this severe complication. When it is determined that a catheter will be removed for refractory peritonitis, hourly peritoneal dialysis exchanges are performed for 12 hr prior to surgery. Postoperatively, the abdomen is rested for 48 hr, after which a temporary peritoneal dialysis catheter is placed at the bedside and hourly exchanges (with antibiotics) are performed for 2–3 days or until the dialysis fluid white blood cell count improves. Then the temporary catheter is removed and the abdomen is rested until the Tenckhoff catheter is replaced in 10–14 days. We treated 5 consecutive patients with refractory peritonitis (2 Pseudomonas, 1 Proteus, 1 Candida, 1 S. aureus) with this technique. All 5 patients were able to return successfully to peritoneal dialysis. At our institution over the past five years, 9 patients with refractory peritonitis due to the same organisms have had their catheters removed. Only 5 (56%) were able to return to PD. Although preliminary, our technique holds promise for those patients wishing to return to peritoneal dialysis after having a catheter removed for refractory peritonitis.

Peritonitis during Continuous Ambulatory Peritoneal Dialysis (CAPD): Risk Factors, Clinical Severity, and Pathogenetic Aspects

This study of 228 episodes of peritonitis occurring during a total observation time of 2365 treatment months over a six-year period in a uniformly selected, trained, and treated continuous ambulatory peritoneal dialysis (CAPD) population (N = 124) showed the following major findings: 1) the risk of developing peritonitis was 55% within the first year and 89% within the first three years on CAPD; 2) high age (≥60 years) and year of CAPD start were risk factors for peritonitis; 3) neither sex, diabetes, or hypoalbuminemia were found to be risk factors for peritonitis; 4) the year of start, but neither the degree of severity, nor the time of the first episode affected the risk of developing a second episode of peritonitis; 5) no specific characteristics were identified in patients with the highest mean peritonitis incidence or in the patients without peritonitis; 6) in 27% of all episodes, turbidity of the dialysate was the only clinical finding; 7) the proportion of asymptomatic episodes was lower in patients ≥ 60 years; 8) the degree of clinical severity of peritonitis was not statistically influenced by the number of previous episodes; 9) the cause of peritonitis was established in only 26% of all cases; and 10) no statistical association was found between the cause of peritonitis and patient characteristics.

A Comparison of Peritoneal Dialysis-Related Infections in Black and White Patients

Objective

To determine if black patients in our peritoneal dialysis (PD) program had higher rates of PD-related infections. The outcomes of black patients versus white patients were also reviewed.

Design

A review of prospectively collected patient demographic and PD-related infection data and out comes, from 1979 to 1991.

Patients

The 68 black patients in our PD program were matched with white control patients for age, sex, insulin dependence, time on dialysis, and mode of dialysis (CAPD or CCPD). The infection, demographic, and outcome data from the two groups were compared.

Results

Black patients had higher peritonitis rates (1.10 vs 0.82 episodeslyear, p=0.001) and exit-site infection rates (1.13 vs 0.95 episodeslyear, p=0.02) than the white control patients. Tunnel infection rates were 0.21 episodeslyear in both groups. S. epidermidis peritonitis was more common in black patients (48% of episodes vs 21% of episodes in whites, p=0.005), and S. aureus peritonitis was more common in white patients (29% vs 11% in blacks, p=0.005). The subset of black patients (n=13) on a disconnect system (Y-set) had a peritonitis rate similar to their white controls on the Y-set (0.41 vs 0.74 episodes/year, p=0.27). There were no episodes of S. epidermidis peritonitis in this subset of black patients. Black patients had fewer S. aureus exit-site infections than white patients (21% vs 41%, p=0.005). Peritonitis was the leading cause of transfer to hemodialysis in the black patients but not in the white patients.

Conclusion

The susceptibility to S. aureus and S. epidermidis infections differs in black and white patients on peritoneal dialysis for unclear reasons. Peritonitis rates in black patients can be reduced to that of white patients if a disconnect system is used.

Plasmid Profile and Slime Analysis of Coagulase-Negative Staphylococci from Capd Patients with Peritonitis

Little is known about the epidemiology of infections causing peritonitis in continuous ambulatory peritoneal dialysis (CAPD). The commonest cause, coagulase-negative staphylococci (C-NS), are normal skin flora. The main source is thought to be organisms from the patient's own skin or environment. Using plasmid profiles as an epidemiological marker, the authors identified cases in which surveillance skin cultures taken just before an episode of peritonitis were identical to those isolated from the effluent. On comparing the plasmid profiles from the effluent of patients who had multiple episodes over eight weeks, they identified two patterns. One group had different plasmid profiles between episodes of infection. The second group (the majority of the cases) had identical plasmid profiles between the initial episode and the second which occurred between 10 days and four weeks after stopping antibiotics. This suggests that, in most cases of recurrent infection studied, the second episode represented a reinfection or recurrence with the same organism (as the initial episode). Slime production did not discriminate those patients who would develop recurring peritonitis.

Analysis of early mesothelial cell responses to Staphylococcus epidermidis isolated from patients with peritoneal dialysis-associated peritonitis

The major complication of peritoneal dialysis (PD) is the development of peritonitis, an infection within the abdominal cavity, primarily caused by bacteria. PD peritonitis is associated with significant morbidity, mortality and health care costs. Staphylococcus epidermidis is the most frequently isolated cause of PD-associated peritonitis. Mesothelial cells are integral to the host response to peritonitis, and subsequent clinical outcomes, yet the effects of infection on mesothelial cells are not well characterised. We systematically investigated the early mesothelial cell response to clinical and reference isolates of S. epidermidis using primary mesothelial cells and the mesothelial cell line Met-5A. Using an unbiased whole genome microarray, followed by a targeted panel of genes known to be involved in the human antibacterial response, we identified 38 differentially regulated genes (adj. p-value < 0.05) representing 35 canonical pathways after 1 hour exposure to S. epidermidis. The top 3 canonical pathways were TNFR2 signaling, IL-17A signaling, and TNFR1 signaling (adj. p-values of 0.0012, 0.0012 and 0.0019, respectively). Subsequent qPCR validation confirmed significant differences in gene expression in a number of genes not previously described in mesothelial cell responses to infection, with heterogeneity observed between clinical isolates of S. epidermidis, and between Met-5A and primary mesothelial cells. Heterogeneity between different S. epidermidis isolates suggests that specific virulence factors may play critical roles in influencing outcomes from peritonitis. This study provides new insights into early mesothelial cell responses to infection with S. epidermidis, and confirms the importance of validating findings in primary mesothelial cells.

A Monte Carlo analysis of peritoneal antimicrobial pharmacokinetics

Peritoneal dialysis-associated peritonitis (PDAP) can be treated using very different regimens of antimicrobial administration, regimens that result in different pharmacokinetic outcomes and systemic exposure levels. Currently, there is no population-level pharmacokinetic framework germane to the treatment of PDAP. We coupled a differential-equation-based model of antimicrobial kinetics to a Monte Carlo simulation framework, and conducted "in silico" clinical trials to explore the anticipated effects of different antimicrobial dosing regimens on relevant pharmacokinetic parameters (AUC/MIC and time greater than 5 ×MIC) and the level of systemic exposure.

The nature and role of microbial biofilm in infections associated with prosthetic devices

An understanding of the role of microbial biofilm in infections associated with prosthetic devices is essential for the implementation of successful treatment regimens.

Antimicrobial activities of YycG histidine kinase inhibitors againstStaphylococcus epidermidisbiofilms

Staphylococcus epidermidis has become a significant pathogen causing infections due to biofilm formation on surfaces of indwelling medical devices. Biofilm-associated bacteria exhibit enhanced resistance to many conventional antibiotics. It is therefore, important to design novel antimicrobial reagents targeting S. epidermidis biofilms. In a static chamber system, the bactericidal effect of two leading compounds active as YycG inhibitors was assessed on biofilm cells by confocal laser scanning microscopy combined with viability staining. In young biofilms (6-h-old), the two compounds killed the majority of the embedded cells at concentrations of 100 microM and 25 microM, respectively. In mature biofilms (24-h-old), one compound was still effectively killing biofilm cells, whereas the other compound mainly killed cells located at the bottom of the biofilm. In contrast, vancomycin was found to stimulate biofilm development at the MBC (8 microg mL(-1)). Even at a high concentration (128 microg mL(-1)), vancomycin exhibited poor killing on cells embedded in biofilms. The two compounds exhibited faster and more effective killing of S. epidermidis planktonic cells than vancomycin at the early stage of exposure (6 h). The data suggest that the new inhibitors can serve as potential agents against S. epidermidis biofilms when added alone or in concert with other antimicrobial agents.

Dynamic Analysis of Peritoneal Dialysis Associated Peritonitis

Peritoneal dialysis associated peritonitis (PDAP) has a historical incidence of approximately 0.3 to 0.5 episodes per patient per year; it represents the leading cause for hospitalization in patients on peritoneal dialysis (PD) and imposes a significant burden of morbidity. PDAP is unique in that each dialysis exchange removes a relatively large fraction of the bacteria laden free intraperitoneal fluid. The attendant removal of bacteria existing in the fluid phase (planktonic bacteria) may interact with bacterial growth to modulate the rate at which the peritoneal burden of microorganisms is reduced. We investigated the potential interactions between bacterial growth dynamics, multiphase bacterial kinetics, and mechanical clearance of microorganisms using simple mathematical analyses based upon in vitro data regarding bacterial growth kinetics in peritoneal dialysate. There are strong dynamic interactions predicted between fluid phase bacterial kinetics, dialysis prescription, and the mechanical clearance of planktonic peritoneal bacteria. There are also strong interactions between fluid phase bacterial kinetics and the kinetics of biofilm/sanctuary site formation and clearance. More frequent exchanges might significantly hasten the clearance of intraperitoneal planktonic bacteria in the absence of catheter-associated bacterial biofilm. The formation of bacteria laden biofilm raises the possibility of a "commensal state," in which ongoing mechanical clearance limits the total peritoneal bacterial burden.

Isolation of methicillin-resistant coagulase-negative staphylococci from patients undergoing continuous ambulatory peritoneal dialysis (CAPD) and comparison of different molecular techniques for discriminating isolates of Staphylococcus epidermidis

Coagulase-negative staphylococci (CNS) have emerged as an important pathogen in nosocomial infections. About 80%-90% of CNS isolates associated with hospital infections are methicillin-resistant coagulase-negative staphylococci (MRCNS). The aims of this study were to screen for MRCNS isolates in the flora of a small population of patients undergoing continuous ambulatory peritoneal dialysis (CAPD) and to evaluate the discriminatory power of different molecular methods: pulsed-field gel electrophoresis (PFGE), mecA location, ClaI/mecA polymorphism and arbitrarily primed polymerase chain reaction (AP-PCR) for characterizing isolates of methicillin-resistant Staphylococcus epidermidis (MRSE). Seventy-nine CNS isolates were recovered from the 11 CAPD patients studied. Using a methicillin screening agar and a DNA specific mecA probe we verified that 30 of the 79 (38%) CNS isolates were resistant to methicillin (MRCNS). Twenty-two of the 30 MRCNS (73%) were MRSE, 7 (23%) methicillin-resistant S. haemolyticus (MRSH(ae)) and 1 (3%) methicillin-resistant S. hominis (MRSH(om)). All patients analyzed carried MRCNS in their flora, in one or more sites. Since CAPD patients have high risk for developing peritonitis, the colonization of these patients with MRCNS might represent an additional problem, due to the therapeutic restrictions imposed by these multiresistant isolates. A wide genetic diversity was verified when the PFGE of the MRSE isolates was analyzed. The 22 MRSE isolates displayed a total of 15 PFGE different patterns (11 PFGE types and 4 subtypes). The location of mecA in the SmaI-fragmented genome DNA did not bring any additional advantage for epidemiologic characterization of the isolates. The ClaI/mecA polymorphism was able to correctly discriminate 12 from the 15 PFGE patterns. In addition, the DNA of 20 MRSE isolates were used for AP-PCR typing. These isolates belonged to 14 PFGE patterns (11 types and 3 subtypes) and displayed 15 genotypes (for the association of PFGE, mecA location and ClaI/mecA polymorphism). A total of 17 different amplification patterns was verified using the primer 1. Only for 2 genotypes, strains having identical genetic backgrounds were further discriminated by AP-PCR (2 of 15 genotypes (87%) for AP-PCR and 1 of 15 genotypes for PFGE; (93%). Concluding, our results indicated that the AP-PCR can be an alternative and useful tool for monitoring and genotyping MRSE colonization and also to molecular characterizing MRSE outbreaks in hospitals.

The abilities of a Staphylococcus epidermidis wild-type strain and its slime-negative mutant to induce endocarditis in rabbits are comparable

The abilities of a parent and mutant pair of Staphylococcus epidermidis strains, the slime-producing parent RP62A and its slime-negative mutant, to establish endocarditis in a rabbit model of aortic valve endocarditis and to accumulate and adhere to surfaces in vitro were compared. Vegetation titer and infection rate depended on the presence or absence of a catheter (P = 0.020) and on inoculum size (P < 0.001) but not on the infecting strain. The ability of the parent strain vis-à-vis its mutant to accumulate in vitro on surfaces as demonstrated in a slime test did not correlate with any enhancement in the development of endocarditis in the rabbit model. In vitro initial adherence rates were identical. Both isolates accumulated to the same reduced extent in vitro in the presence of serum, albumin, or gelatin. Adhesion was equally promoted by addition of fibronectin. These data suggest that the in vitro phenomenon of accumulation described as slime production in the absence of serum may not be an important virulence determinant in vivo.

Slime production, adherence and hydrophobicity in coagulase-negative staphylococci causing peritonitis in peritoneal dialysis

Attachment of coagulase-negative staphylococci to plastic surfaces by means of hydrophobic interaction and slime production may be important in producing catheter associated infections. In continuous ambulatory peritoneal dialysis (CAPD), the relationship between these properties and disease is unclear and the effect of dialysate fluid is not considered. For a collection of coagulase-negative staphylococci from CAPD patients, slime production and adherence were measured by colorimetric methods and hydrophobicity was determined by autoaggregation in ammonium sulphate solution. Comparison of 73 nasal isolates with 69 isolates from peritonitis showed no significant differences with respect to three properties, with the exception of a greater adherence of peritoneal isolates in dialysate because of a greater proportion of staphylococcal species other than Staphylococcus epidermidis. Fewer strains showed adherence in dialysate (12/142 8.5%) than in broth (94/142 66%) but the proportion of strains producing slime was similar. The milieu of the bacteria rather than the organisms themselves may be of greater importance in the establishment of infection.

Biochemical aberrations in a dialysis patient following parathyroidectomy

Renal osteodystrophy is a common problem in patients with end-stage renal disease. Severe secondary hyperparathyroidism unresponsive to an intensive regimen of diet, phosphate binders, and calcitriol therapy is an indication for surgical parathyroidectomy. Certain unique postoperative electrolyte aberrations are seen after parathyroidectomy, including severe hypocalcemia and hypophosphatemia (hungry bone syndrome), as well as hyperkalemia. Identifying the patient at high risk for developing these electrolyte abnormalities will facilitate perioperative and postoperative management.

Molecular basis of intercellular adhesion in the biofilm-forming Staphylococcus epidermidis

The Staphylococcus epidermidis genes icaABC are involved in the synthesis of the polysaccharide intercellular adhesin (PIA), which is located mainly on the cell surface, as shown by immunofluorescence studies with PIA-specific antiserum. PIA was shown to be a linear beta-1,6-linked glucosaminoglycan composed of at least 130 2-deoxy-2-amino-D-glucopyranosyl residues of which 80-85% are N-acetylated, the rest being non-N-acetylated and positively charged. A transposon insertion in the icaABC gene cluster (ica, intercellular adhesion) led to the loss of several traits, such as the ability to form a biofilm on a polystyrene surface, cell aggregation, and PIA production. The mutant could be complemented by transformation with the icaABC-carrying plasmid pCN27. Transfer of pCN27 into the heterologous host Staphylococcus carnosus led to the formation of large cell aggregates, the formation of a biofilm on a glass surface, and PIA expression. The nucleotide sequence of icaABC suggests that the three genes are organized in an operon and that they are co-transcribed from the mapped icaA promoter. IcaA contains four potential transmembrane helices, indicative of a membrane location. The deduced IcaA sequence shows similarity to those of polysaccharide-polymerizing enzymes, the most pronounced being with a Rhizobium meliloti N-acetylglucosaminyltransferase involved in lipo-chitin biosynthesis (22.5% overall identity and 37.4% overall similarity). This similarity suggests that IcaA has N-acetylglucosaminyltransferase activity in the formation of the beta-1, 6-linked N-acetyl-D-glucosaminyl polymer. IcaB is secreted into the medium and contains a typical signal peptide. IcaC is hydrophobic and contains six predicted transmembrane helices distributed over its entire length, typical for an integral membrane protein. Neither IcaB nor IcaC shares similarity with known proteins, and their function is unknown. Inactivation of icaA, icaB, or icaC in pCN27 led to the complete loss of the intercellular adhesion phenotype in S. carnosus, suggesting that all three genes are involved in intercellular adhesion, PIA expression, and translocation.

Pathogenicity and virulence of coagulase negative staphylococci in relation to adherence, hydrophobicity, and toxin production in vitro

AIMS

To study the pathogenicity and virulence characteristics of Staphylococcus epidermidis, Staphylococcus haemolyticus, and Staphylococcus sapro-phyticus.

METHODS

BALB/c mice were challenged intraperitoneally with graded doses of three strains belonging to each species. LD50s were measured for each strain. Haemolysin (alpha- and delta-) and enzyme (DNAase, lipase, and esterase) production in vitro were measured qualitatively and quantitatively. Adhesion to plastic was measured and related to cell surface hydrophobicity among the strains.

RESULTS

S saprophyticus proved the most virulent (LD50 = 2.7-2.9 x 10(7) cfu/g body weight) while S epidermidis was the least virulent (LD50 = 6-8 x 10(7) cfu/g body weight). An enlarged spleen was the most common macroscopic pathological feature. Kidney, liver, and more rarely peritoneal abscesses were also seen in the infected animals. No direct correlation was found between adherence in vitro, cell surface hydrophobicity, or toxin/enzyme biosynthesis and virulence in mice.

CONCLUSION

The results show that coagulase negative staphylococci are pathogenic in BALB/c mice. It is clear that these bacteria can cause invasive disease. However, the in vitro characteristics of coagulase negative staphylococci are not related to the pathogenicity of the organisms in mice.

Managing Staphylococcus aureus catheter infection in continuous ambulatory peritoneal dialysis patients

A case of Staphylococcus aureus catheter infection in a patient on peritoneal dialysis is presented and discussed by nephrologists, a social worker, a nurse specializing in the care of peritoneal dialysis patients, and the patient involved. The focus of the multidisciplinary case discussion concerns the management of S aureus catheter infections, including catheter removal, psychosocial issues and the patient's response to the need for catheter removal, the risk factors and prevention of S aureus catheter infections in peritoneal dialysis patients, and exit site care practices.

Lack of evidence for increased adherent growth in broth or human serum of clinically significant coagulase-negative staphylococci

The relevance of adherent growth of coagulase-negative staphylococci (CNS) in vitro to their clinical significance is controversial. We have investigated the accumulation on polymer surfaces of clinically significant CNS strains, isolated from blood cultures and intravascular catheter tips, compared with controls. The adherent growth of significant CNS isolates in a microtitre tray assay was not greater than controls, and this finding was repeated when different combinations of target polymer surface, culture medium and incubation atmosphere type were investigated. The surface accumulation of CNS cultured in pooled human serum was poor compared with culture in broth. Whole cell and surface protein profiles of serum grown Staphylococcus epidermidis differed markedly from broth cultured cells. We recommend the use of human body fluids, such as serum, as useful culture media to assess whether potential virulence determinants are likely to be expressed in vivo. The microtitre tray assay of adherent growth is not helpful in determining the virulence of individual CNS isolates.

Diffusion of rifampin and vancomycin through a Staphylococcus epidermidis biofilm

Using an equilibrium dialysis chamber, we evaluated the penetration of vancomycin, rifampin, or both through a staphylococcal biofilm to simulate treatment of an infected biomedical implant. A biofilm of ATCC 35984 (slime-positive Staphylococcus epidermidis; vancomycin MIC and MBC, 1 and 2 micrograms/ml, respectively; rifampin MIC and MBC, 0.00003 and 0.00025 micrograms/ml, respectively) was established on the inner aspect of the dialysis membrane (molecular mass exclusion, 6,000 kDa). Serum containing vancomycin (40 micrograms/ml), rifampin (20 micrograms/ml), or a combination of both was introduced into the inner chamber of the dialysis unit (in direct contact with the biofilm), and serum alone was added to the outer chamber. Rifampin and vancomycin concentrations in both chambers were determined over a 72-h period. In the absence of rifampin, the concentration of vancomycin in the outer chamber exceeded the MBC for the organism after 24 h, and the MBC increased to nearly 8.0 micrograms/ml by 72 h, demonstrating that therapeutic levels of vancomycin can penetrate a staphylococcal biofilm. However, viable bacteria were recovered from the biofilm after 72 h of treatment with no apparent increase in the MIC or MBC of vancomycin. Similarly, concentrations of rifampin exceeding the MBC were detected in the outer chamber after 24 h of treatment, but viable organisms were recovered from the biofilm after 72 h of treatment. In this case, the rifampin MBCs for surviving organisms increased from 0.00025 to > 128 micrograms/ml. The combination of agents prevented the development of resistance to rifampin, improved the perfusion of vancomycin through the biofilm, and decreased the penetration of rifampin but did not sterilize the membrane. These observations provide evidence that bactericidal levels of vancomycin, rifampin, or both can be attained at the surface of an infected implant. Despite this, sterilization of the biofilm was not accomplished after 72 h of treatment.

Regulation of slime production in Staphylococcus epidermidis by iron limitation

Slime production by most strains of Staphylococcus epidermidis was enhanced by conditions of iron limitation produced by the addition of ethylenediamine-di-o-hydroxyphenol acetic acid to the growth medium. The density of the biofilm which formed on the base of microtiter plates was dependent on the degree of iron limitation, the stage of the growth cycle, and the nutritional state of the initial inoculum. One repeatedly slime-negative S. epidermidis strain, passaged in tryptic soya broth containing ethylenediamine-di-o-hydroxyphenol acetic acid, expressed high levels of slime after two passages. These observations suggest that iron limitation is one factor that regulates slime production by S. epidermidis. These findings could explain inconsistencies between the in vivo observation that biofilms invariably form on implanted catheters and the in vitro finding that some isolates from catheter-associated infection fail to produce slime.

Identification of Staphylococcus epidermidis and Staphylococcus hominis from blood cultures by testing susceptibility to desferrioxamine

Testing susceptibility to desferrioxamine has recently been described as a method for the identification of Staphylococcus epidermidis. This method was compared to a commercial test and the tube coagulase test for the identification of staphylococci from blood cultures and other fluid specimens. A total of 216 isolates was tested over a 13-month period. Sensitivity of the desferrioxamine test in identifying isolates of Staphylococcus epidermidis and Staphylococcus hominis was 97.3%, while specificity was 91.8%. When isolates displaying discrepant desferrioxamine results were characterized using recently described interpretive criteria, sensitivity and specificity of the desferrioxamine test improved to 100%. The desferrioxamine test was reliable, inexpensive and simple to perform, and should prove useful in the diagnostic laboratory.

The effects of magnesium, calcium, EDTA, and pH on the in vitro adhesion of Staphylococcus epidermidis to plastic

The effects of increasing concentrations of magnesium (Mg2+), calcium (Ca2+) or EDTA, and pH on the adhesion of five slime-positive strains of Staphylococcus epidermidis (Se+) to plastic were examined using an in vitro microwell assay. The addition of Mg2+ (as either MgSO4 or MgCl2) to the bacterial suspension in concentrations as low as 16 microM significantly enhanced the adhesion of all test strains to plastic (P < 0.001). Similarly, the addition of Ca2+ (as CaCl2) in concentrations exceeding 128 microM produced a significant increase in the adhesion of all test strains, but not to the extent observed with Mg2+. In contrast, the adhesion of all test strains to plastic was significantly reduced in the presence of EDTA at concentrations greater than 8 mM. However, EDTA in concentrations as low as 0.25 mM caused a significant decrease in the adhesion of two strains of Se+. The effect of pH was variable, but at a pH of 5.0 and 6.0, the adhesion of all test strains was significantly reduced compared to control values at a pH of 7.0. Two strains showed a significant increase in adhesion at a pH of 8.0. We also compared the effects of these variables on the adherence of a slime-negative phase variant derived from a slime-positive parent strain. With the exception of pH, the adhesion of both strains in response to increasing divalent cations or EDTA was similar. These data indicate that, in addition to hydrophobic interactions, ligand-specific binding, and slime production, pH and divalent cations, especially Mg2+, are important determinants of the adhesion of S. epidermidis to plastic surfaces in vitro.(ABSTRACT TRUNCATED AT 250 WORDS)

Scanning electron microscopy of bacterial biofilms on indwelling bladder catheters

Fifty Foley bladder catheters that had been indwelling for periods ranging from 3 to 83 days (mean 35 days) were examined for the presence of bacterial biofilm. Scanning electron microscopy on freeze-dried cross-sections and fixed, critical point-dried longitudinal sections revealed biofilm formation on the luminal surfaces of 44 of the catheters. Culture of urine samples and sonicates from catheters revealed that the prevalence of bacteriuria was less than that of catheter colonization. A wide range of nosocomial species were found colonizing the catheters, Escherichia coli being most often isolated. The bacterial composition of the biofilms ranged from single species to mixed communities containing up to four species. There was no relationship between the length of time that the catheter had been in situ and the extent of biofilm formation. The biofilms varied in thickness from 3 to 490 microns and were visible as layers of bacterial cells up to about 400 cells deep, embedded in a matrix.

Microbiological aspects of peritonitis associated with continuous ambulatory peritoneal dialysis

The process of continuous ambulatory peritoneal dialysis has provided a useful, relatively inexpensive, and safe alternative for patients with end-stage renal disease. Infectious peritonitis, however, has limited a more widespread acceptance of this technique. The definition of peritonitis in this patient population is not universally accepted and does not always include the laboratory support of a positive culture (or Gram stain). In part, the omission of clinical microbiological findings stems from the lack of sensitivity of earlier microbiological efforts. Peritonitis results from decreased host phagocytic efficiency with depressed phagocytosis and bactericidal capacity of peritoneal macrophages. During episodes of peritonitis, fluid movement is reversed, away from the lymphatics and peritoneal membrane and toward the cavity. As a result, bloodstream infections are rare. Most peritonitis episodes are caused by bacteria. Coagulase-negative staphylococci are the most frequently isolated organisms, usually originating from the skin flora, but a wide array of microbial species have been documented as agents of peritonitis. Clinical microbiology laboratories need to be cognizant of the diverse agents so that appropriate primary media can be used. The quantity of dialysate fluid that is prepared for culture is critical and should constitute at least 10 ml. The sensitivity of the cultural approach depends on the volume of dialysate, its pretreatment (lysis or centrifugation), the media used, and the mode of incubation. The low concentration of microorganisms in dialysate fluids accounts for negative Gram stain results. Prevention of infection in continuous ambulatory peritoneal dialysis patients is associated with the socioeconomic status of the patient, advances in equipment (catheter) technology, and, probably least important, the application of prophylactic antimicrobial agents.

Simultaneous catheter replacement and removal in refractory peritoneal dialysis infections

The present report describes more than nine years of experience with simultaneous removal and replacement of the chronic peritoneal dialysis (PD) catheter in treating refractory mechanical and infectious complications. Simultaneous catheter replacement and removal not only succeeded in 22 of 23 cases with non-infectious mechanical complications, allowing uninterrupted PD in all but three, but also succeeded in 30 (83%) of 36 additional cases with persisting or recurring infection. Simultaneous procedures failed in six (17%) of 36 infectious cases, due to persistent infection in two cases and to procedure complications in four cases. The clinical characteristics of the 30 successful infectious cases were compared to the characteristics of both the six failures, as well as 29 additional infectious cases in which the peritoneal catheter was removed but was not replaced because of a variety of serious complications (such as pancreatitis, abscess, sepsis, or fluid overload). At the time of simultaneous catheter replacement in successful cases, temporary control of active peritoneal inflammation was achieved more frequently (80% vs. 31%, P less than 0.001) with a shorter interval of antibiotic coverage (6.4 +/- 0.9 vs. 14.7 +/- 1.3 days, P less than 0.001). In addition, the successful cases were characterized by significantly more Staphylococcal infections (70% vs. 26%, P less than 0.001) and significantly fewer Pseudomonal or fungal infections (6% vs. 59%, P less than 0.001), although successful cases included some non-Pseudomonal, non-enteric gram negative infections (23%).(ABSTRACT TRUNCATED AT 250 WORDS)

Risk factors for tunnel infections in continuous peritoneal dialysis

Little data are available about risk factors for peritoneal catheter subcutaneous tunnel infection. Therefore, we analyzed tunnel infections occurring in our program over a 10.5-year period. One hundred twenty-nine tunnel infections occurred in 92 of 411 patients (22%) on peritoneal dialysis for a mean of 19 +/- 19 months. Tunnel infection rate was 0.19 per year. By 1 year, 15% of patients had a tunnel infection, and by 2 years, 23%. Tunnel infection rates decreased with increasing time on peritoneal dialysis: 2.4 per year for patients on peritoneal dialysis less than 1 year, 0.8 per year for patients on dialysis 1 to 2 years, and 0.4 per year for patients on dialysis greater than 2 years (all different at P less than 0.01). Organisms were cultured in 109 tunnel infections: gram-positive cocci in 77 episodes (71%) [Staphylococcus aureus 57, 52%], and gram-negative bacilli in 24 episodes (22%). Tunnel infection rates were higher in diabetics than in nondiabetics (0.27 per year v 0.16 per year, respectively; P less than 0.001 by life-table analysis of time to first infection) and also higher in women than in men (0.23 per year v 0.17 per year, P less than 0.001). Tunnel infection rates were 0.35 per year for diabetic women, 0.20 per year for diabetic men, 0.18 per year for nondiabetic women, and 0.15 per year for nondiabetic men (groups different, P less than 0.001). Race and age were similar in patients with and without tunnel infections. Catheter loss was 80% when tunnel infection was associated with peritonitis and 40% when tunnel infection alone was present (P less than 0.001). We conclude that the risk of tunnel infection is highest early in the course of peritoneal dialysis and that diabetic women, for unclear reasons, are at the highest risk.

In vitro measurement of the adherence of Staphylococcus epidermidis to plastic by using cellular urease as a marker

A rapid and sensitive in vitro assay was developed to quantitatively assess the adherence of Staphylococcus epidermidis to a hydrophobic plastic surface. The assay is based upon the detection of cell-associated urease activity as a marker of bacteria remaining adherent to the polystyrene microwells of flat-bottomed, 96-well tissue culture plates. Using ATCC 35984, a slime-producing strain of S. epidermidis, the assay could detect as few as 3 x 10(3) bacteria and was linear to 3.5 x 10(7) bacteria. The adherence of both slime-positive and slime-negative coagulase-negative staphylococci could be evaluated by using this method. This assay could be used to examine factors which influence the adherence of individual S. epidermidis strains to hydrophobic surfaces and to develop agents or coating materials which suppress the adherence of coagulase-negative staphylococci to biomedical implants.

Antibiotic susceptibility of coagulase-negative staphylococcal blood isolates with special reference to adherent, slime-producing Staphylococcus epidermidis strains

The epidemiologic relationship between slime-producing coagulase-negative staphylococci and antibiotic resistance was evaluated by analyzing the in vitro susceptibility of 278 blood isolates to 10 different antimicrobial agents. The strain collection consisted of 60 coagulase-negative staphylococcal septicemia isolates, recovered from at least 2 successive blood cultures from 59 septicemia patients and 218 single blood isolates defined as insignificant. When the slime-producing Staphylococcus epidermidis isolates were compared with all non-slime-producing S. epidermidis isolates, the slime-producing isolates exhibited higher MIC50 and/or MIC90 values than the non-slime-producing isolates, and were also more commonly resistant against many of the antimicrobials tested. The association between slime production and resistance against tobramycin, gentamicin and rifampicin was statistically significant (p less than 0.05). A similar finding was observed among the S. epidermidis septicemia isolates when they were analyzed separately from the single (suspected blood contaminant) S. epidermidis isolates. Further characterization of the septicemia isolates revealed that the positive correlation between slime production and antibiotic resistance reflected, to some extent, multiresistance of 1 epidemic slime-producing S. epidermidis strain. No difference in resistance was observed between the slime-producing and non-slime-producing single S. epidermidis blood isolates.

Association of coagulase-negative staphylococcal slime production and adherence with the development and outcome of adult septicemias

The relationship of coagulase-negative staphylococcal slime production and adherence with the development and outcome of bloodstream infections in two Finnish hospitals was evaluated. Analysis of 64 strains from 62 adult septicemias disclosed 34 (53%) adherent slime producers. In comparison, only 142 (29%) of 489 single blood culture isolates were adherent slime producers. Although tube adherence test-positive strains were significantly (P less than 0.001) more common among the septicemia strains than among clinically insignificant isolates, almost half of the septicemia cases were caused by tube test-negative strains. Thus, regarding any single patient isolate, a cautious posture to the clinical impact of positivity in the tube adherence test seems warranted. Moreover, adherence and slime production, as such, apparently played no role in the clinical outcome of these infections. The epidemiologic findings revealed that slime-producing coagulase-negative staphylococci were common in the hospital environment and suggested that epidemic spread of such strains was influenced by antimicrobial therapy. Collectively, these results indicate that, at least in these two hospitals, positivity in the tube was of minor importance in guiding clinical decisions in treating adult septicemias.

Adherence measured by microtiter assay as a virulence marker for Staphylococcus epidermidis infections

Staphylococcus epidermidis strains isolated from clinical sources showed a wide range of abilities to adhere to glass and plastic materials. The degree of adherence depended on a number of factors, most notably, the composition of the growth medium. Adherence was enhanced by the addition of glucose or oleic acid to the growth medium and inhibited by serum. We have demonstrated a statistically significant association between the quantitative assessment of adherence to polystyrene tissue culture plates and clinical relevance. No such association was found when adherence was assessed by the qualitative adherence assay. Possible new approaches for assessing the clinical relevance of coagulase-negative staphylococcal isolates are discussed.

Staphylococcus aureus colonization and infection in patients on continuous ambulatory peritoneal dialysis

Staphylococcus aureus is the most common cause of peritonitis in patients undergoing peritoneal dialysis in Brazil. Using restriction endonuclease analysis of plasmid DNA, we investigated the importance of chronic carriage of S. aureus in the development of peritonitis in patients on continuous ambulatory peritoneal dialysis at the Division of Nephrology, Escola Paulista de Medicina, Sao Paulo, Brazil. A total of 117 isolates (30 patients) of S. aureus were available for typing, including 51 isolates (22 patients) from the nares, 58 isolates (27 patients) from pericatheter skin, and 8 isolates (6 patients) from peritoneal fluid, from patients with peritonitis. Restriction endonuclease subtyping showed that although most patients harbored more than one subtype of S. aureus, in the majority of patients nasal and/or pericatheter skin isolates with identical restriction endonuclease digest patterns were recovered on more than one occasion. Furthermore, 95% of patients with both nasal and pericatheter colonization were colonized with the same subtypes at both sites. All of the patients with peritonitis were infected with a subtype which colonized the nares, pericatheter skin, or both. These results demonstrate the importance of an endogenous source of S. aureus in the development of continuous ambulatory peritoneal dialysis-associated peritonitis.

Methicillin-resistant staphylococcal infections in an outpatient peritoneal dialysis program

In view of the increasing concern about hospital-acquired methicillin resistance, we examined the sensitivities and outcome of staphylococcal infections related to outpatient peritoneal dialysis over a 5-year period. Data on all episodes of peritonitis (n = 360) and catheter infections (n = 507) were gathered prospectively from January 1984 to December 1988. The numbers of patients on peritoneal dialysis each year ranged from 136 in 1984 to 109 in 1987. Fifteen methicillin-resistant staphylococcal infections (MRSI) related to outpatient peritoneal dialysis occurred. Three were due to methicillin-resistant Staphylococcus aureus found in infected exit sites (2.3% of all S aureus catheter infections). Two of these infections occurred in a continuous ambulatory peritoneal dialysis (CAPD) patient who carried methicillin-resistant S aureus in his nares. The other 12 methicillin-resistant organisms were coagulase-negative staphylococci that caused peritonitis. There was a significant increase in the percentage of episodes of coagulase-negative staphylococci peritonitis caused by methicillin-resistant organisms; from 5% (3/57) in 1984 through 1986 to 28% (9/32) in 1987 through 1988 (P less than 0.005). In view of the high percentage of coagulase-negative staphylococci peritonitis that is methicillin-resistant, vancomycin rather than cephalosporins should be used for initial treatment.

A review of Staphylococcus aureus exit-site and tunnel infections in peritoneal dialysis patients

Staphylococcus aureus peritoneal exit-site and tunnel infections are a source of considerable morbidity for peritoneal dialysis patients. These infections are difficult to resolve, can lead to peritonitis, and often require removal of the peritoneal catheter. Staphylococcal nasal carriage is the major risk factor for S aureus exit-site infections and peritonitis episodes. In the future, the identification of patients who are S aureus nasal carriers and then treatment of the carriage state with rifampin may prove to be a means of decreasing infection rates. The best treatment for S aureus exit-site and tunnel infections has not been established. Treatment regimens in general use include oral antibiotics or intraperitoneal vancomycin. The optimal length of therapy is also unclear. Since the development of the disconnect peritoneal dialysis system, S aureus, rather than the Staphylococcus epidermidis, is the leading cause of peritonitis. To further decrease peritonitis rates, attention must now be directed at catheter-related peritonitis episodes, with S aureus the most common cause of such episodes. Controlled, prospective studies designed to investigate methods of preventing and treating S aureus exit-site infections in peritoneal dialysis patients are needed.

CAPD peritonitis. Incidence, pathogens, diagnosis, and management

Peritonitis is a frequent complication of CAPD. Sixty percent of all patients on CAPD will have at least one episode of peritonitis during the first year of this mode of dialysis. Most of the episodes of peritonitis are caused by touch contamination of the dialysis tubing or by extension of the catheter exit site or tunnel infection. Coagulase-negative and coagulase-positive Staphylococcus are the two most common organisms, accounting for 50% or more of all CAPD peritonitis. Other gram-positive and gram-negative bacteria and fungi account for the rest. Intraperitoneal antibiotic treatments are usually effective in eradicating the infection. The choice of antibiotics depends on organisms isolated from cultured dialysate. Fungal peritonitis and, occasionally, Pseudomonas peritonitis require removal of the catheter to eradicate the infection. Prompt identification and treatment of peritonitis are essential to ensure success of a CAPD program. Although with newer techniques, like Y-connector or ultraviolet light system, the rate of peritonitis has declined; however, it has still remained the major complication of the CAPD program.

Coagulase-negative staphylococci: incidence, pathogenicity, and treatment in the 1990s

Coagulase-negative staphylococci (CNS), which historically have been viewed as contaminants when recovered in culture media, are now recognized as opportunistic pathogens of increasing importance in hospital-acquired infections. They are frequently found colonizing prosthetic devices and intravenous lines. CNS are capable of producing a variety of infections including deep-seated infections such as endocarditis and meningitis. Staphylococcus epidermidis is the most commonly isolated CNS and it appears to be the most resistant to antibiotics, making antimicrobial therapy challenging. Treatment of the infection will very often require removal of a prosthetic device, if present. An adequate infection control program is imperative in prophylaxis against this infection.

Effects of subinhibitory concentrations of vancomycin or cefamandole on biofilm production by coagulase-negative staphylococci

The density of the biofilm layer produced on a plastic surface by 23 clinical isolates and 1 reference strain of slime-positive, coagulase-negative staphylococci was measured following growth in subinhibitory concentrations (sub-MICs) of cefamandole or vancomycin ranging from 2 to 0.008 micrograms/ml. All strains were susceptible to less than or equal to 2 micrograms of each agent per ml. The mean biofilm density produced by each strain was calculated from a total of eight determinations at each sub-MIC and was compared with the mean biofilm density of a drug-free control after correcting for differences in growth. The results showed that the density of the biofilm layer produced by 10 (42%) of 24 strains and 13 (54%) of 24 strains was significantly increased (P less than 0.006) at one or more sub-MICs of cefamandole or vancomycin, respectively. In contrast, the density of the biofilm produced by 9 (38%) of 24 and 2 (8%) of 24 strains was significantly reduced at one or more sub-MICs of cefamandole and vancomycin, respectively, and the biofilm density of 7 of these strains was decreased only when the sub-MIC was one-half the MIC. The biofilm density of six strains (five versus cefamandole and one versus vancomycin) was both enhanced and reduced by different sub-MICs of the same agent. None of the strains produced a detectable biofilm at or above the MIC for the strain. These data indicate that antimicrobial agents such as cefamandole or vancomycin could potentially enhance the biofilm matrix produced by certain slime-positive, coagulase-negative staphylococci on the surface of a biomedical implant if concentrations of these agents fall below the MIC for the infecting strain.

Prevalence of peritonitis-associated coagulase-negative staphylococci on the skin of continuous ambulatory peritoneal dialysis patients

The predominance of coagulase-negative staphylococci as normal skin flora is thought to be a factor in their association with episodes of peritonitis in patients undergoing continuous ambulatory peritoneal dialysis. We investigated the prevalence of peritonitis-associated strains on the skin of 28 patients undergoing peritoneal dialysis. Coagulase-negative staphylococci were the most frequently isolated organisms, comprising 47% of peritoneal dialysis fluid isolates and 59% of body site isolates. A total of 142 coagulase-negative staphylococci were speciated, tested for their antimicrobial sensitivity and slime production, and identified by phage typing and plasmid-profile analysis. Staphylococcus epidermidis was the most commonly identified species from both peritoneal dialysis fluid (73%) and body sites (53%). Multiple antibiotic resistance was common, and the greater proportion of isolates were resistant to methicillin; 63.6% of peritoneal dialysis fluid isolates and 61.7% of body-site isolates. S. haemolyticus isolates were significantly more resistant to methicillin than other species. By phage typing and plasmid-profile analysis it was shown that peritonitis was rarely caused by skin-colonizing strains. In only 3 of 14 patients were peritonitis-associated strains isolated as skin colonizers, and no patients developed peritonitis due to organisms previously isolated as skin colonizers.

Species identification, antibiotic sensitivity and slime production of coagulase-negative staphylococci isolated from clinical specimens

Two hundred and seventy-five consecutive clinical isolates of coagulase-negative staphylococci, including strains associated with disease, contaminants and skin colonizers, were speciated, tested for slime production and for their sensitivity to a range of antibiotics. Staphylococcus epidermidis was the most commonly identified species, comprising 63% of all isolates. Slime production was detected in half the strains of Staph. epidermidis, Staph. haemolyticus and Staph. saprophyticus but was rare in other species. Most Staph. haemolyticus strains and approximately half of the Staph. epidermidis strains were resistant to five or more antibiotics. A significant association was found between slime production and multiple antibiotic resistance. For catheter-associated strains, clinical relevance was predictable by species i.e. Staph. epidermidis. Multi-resistant slime-positive Staph. haemolyticus strains, although infrequently associated with disease, were common skin colonizers, presumably acquired from the hospital environment. We describe a practical and inexpensive scheme for the speciation of human coagulase-negative staphylococcal isolates.

Lack of correlation of slime production with pathogenicity in continuous ambulatory peritoneal dialysis peritonitis caused by coagulase negative staphylococci

In order to evaluate the association of slime production by coagulase-negative staphylococci (CNS) and pathogenicity in the setting of continuous ambulatory peritoneal dialysis (CAPD), we prospectively studied 23 CAPD patients with positive dialysate cultures for CNS during a 12-mo period. Patients were categorized into groups with peritonitis or contaminants. In addition, 24 skin flora isolates of CNS from nine uninfected CAPD patients were compared to the other two groups. We found 14 patients with peritonitis and nine with contaminants. Fifty percent of the peritonitis isolates produced slime, compared to 67% of the contaminants (p = 0.72) and 54% of the skin flora isolates (p = 0.93). Slime production was not associated with specific antibiotic susceptibility patterns or speciation. In this prospective study, slime production was not associated with pathogenicity in CAPD peritonitis caused by coagulase-negative staphylococci.

Quantitative epifluorescence assay of adherence of coagulase-negative staphylococci

A quantitative epifluorescence assay was developed to assess the adherent qualities of clinical isolates of coagulase-negative staphylococci. This method compared favorably with the standardized tube adherence assay, yet demonstrated a wide range of surface binding affinities among the slime-producing and non-slime-producing strains of coagulase-negative staphylococci examined. This technique can also be adapted to evaluate the adherence of bacteria to other plastics and biomedical materials used to manufacture implantable medical devices.