1074.Full.Pdf

Article

Peritoneal Dialysis–Related Peritonitis due to Coagulase-Negative Staphylococcus: A Review of 115 Cases in a Brazilian Center

Carlos Henrique Camargo,*† Maria de Lourdes Ribeiro de Souza da Cunha,† Jacqueline Costa Teixeira Caramori,* Alessandro Lia Mondelli,* Augusto Cezar Montelli,* and Pasqual Barretti*

Abstract *Department of Internal Medicine, Background and objectives Coagulase-negative Staphylococcus (CNS) is the most frequent cause of peritoneal Botucatu Medical dialysis (PD)–related peritonitis in many centers. This study aimed to describe clinical and microbiologic char- School, Sa˜oPaulo acteristics of 115 CNS episodes and to determine factors influencing the outcome. State University, Botucatu, Sa˜oPaulo, Brazil; and Design, setting, participants, & measurements This study reviewed the records of 115 CNS peritonitis episodes †Department of that occurred in 74 patients between 1994 and 2011 at a single university center. Peritonitis incidences were Microbiology and calculated for three consecutive 6-year periods (P1, 1994–1999; P2, 2000–2005; P3, 2006–2011) and annually. The Immunology, production of biofilms, enzymes, and toxins was evaluated. Oxacillin resistance was evaluated based on its Biosciences Institute minimum inhibitory concentration and the presence of the mecA gene. of Botucatu, Sa˜oPaulo State University, Botucatu, Sa˜oPaulo, Results The overall incidence of CNS peritonitis was 0.15 episodes per patient per year and did not vary over Brazil time (0.12, 0.14, and 0.16 for P1, P2, and P3, respectively; P=0.21). The oxacillin resistance rate was 69.6%. Toxin and enzyme production was infrequent and 36.5% of CNS strains presented the gene encoding biofilm pro- Correspondence: duction. The presence of icaAD genes associated with biofilm production was predictive of relapses or repeat Dr. Pasqual Barretti, episodes (odds ratio [OR], 2.82; 95% confidence interval [95% CI], 1.11 to 7.19; P=0.03). Overall, 70 episodes Department of fi Internal Medicine, (60.9%) resolved; oxacillin susceptibility (OR, 4.41; 95% CI, 1.48 to 13.17; P=0.01) and vancomycin use as the rst Botucatu Medical treatment (OR, 22.27; 95% CI, 6.16 to 80.53; P,0.001) were the only independent predictors of resolution. School, Sa˜oPaulo State University, Conclusions Oxacillin resistance and vancomycin use as the first treatment strongly influence the resolution Botucatu, Sa˜oPaulo, Brazil-Distrito de rate in CNS peritonitis, which reinforces the validity of the International Society for Peritoneal Dialysis guidelines fi Rubia˜o Jr, w/n, 18618- on monitoring bacterial resistance to de ne protocols for initial treatment. These results also suggest that the 970, Brazil. presence of biofilm is a potential cause of repeat peritonitis episodes. Email: pbarretti@uol. Clin J Am Soc Nephrol 9: 1074–1081, 2014. doi: 10.2215/CJN.09280913 com.br

Introduction association between biofilm production and nonreso- Peritonitis remains a major cause of technique failure lution of CNS peritonitis (10), and we showed that in peritoneal dialysis (PD) (1) and affects patients’ both b-hemolysin production and diabetes mellitus morbidity and mortality (2). In many centers, coagulase- are independent predictors of the nonresolution of negative Staphylococcus (CNS) species are its most Staphylococcus aureus peritonitis (11). Furthermore, frequent causes. In general, CNS peritonitis studies comparing the outcome of peritonitis caused presents a mild clinical course and has a high resolu- by different CNS species have not been reported. tion rate; however, CNS has shown an increasing The two largest published series of CNS peritonitis methicillin resistance rate (3,4). The CNS group com- by Szeto et al. and Fahim et al. confirmed high meth- prises .40 species (5), some of which are already well icillin resistance rates of 49.5% and 68%, respectively established as causes of PD-related peritonitis, partic- (12,13). Unexpectedly, in both studies, there were no ularly Staphylococcus epidermidis (6,7). Methicillin (or differences in the outcomes of episodes treated with oxacillin, similar and stable penicillin) resistance is vancomycin or a first-generation cephalosporin. Al- mediated by mecA (8), a gene located on the transfer- though these results may be associated with bias in able element SCCmec, which allows transmission of the initial prescription, it is also possible that some the resistance trait (9). Methicillin resistance has im- nonevaluated host or bacterial factors influenced the portant implications, including restriction of the use outcome. To provide information regarding the mi- of b-lactams (7). In addition, bacterial virulence fac- crobiologic and clinical characterization of CNS peri- tors and patient characteristics can influence the peri- tonitis, we reviewed all CNS peritonitis episodes that tonitis outcome. For instance, we demonstrated an occurred over an 18-year period. Our main objective

1074 Copyright © 2014 by the American Society of Nephrology www.cjasn.org Vol 9 June, 2014 Clin J Am Soc Nephrol 9: 1074–1081, June, 2014 Coagulase-Negative Staphylococcus Peritonitis in Peritoneal Dialysis, Camargo et al. 1075

was to identify microbiologic and clinical predictors of mupirocin cream at the exit site was introduced; the twin outcome. bag system was used in all patients. During period 3 (2006–2011), daily application of mupirocin was pre- scribed until December 2006, and it was replaced by gen- Materials and Methods tamicin cream in January 2007; the twin bag system was Study Population used in all patients. When patients started their treatment, they gave written Cultures were performed with the Bactec System (Becton consent for the use of their clinical and laboratorial data for Dickinson Company, Sparks, MD) and then seeded onto research purposes. Thus, the institutional research ethics blood agar. The isolates were submitted to catalase and committee approved this study and exempted it of any coagulase tests (16) and were identified according to the specific written informed consent. We reviewed all epi- recommendations of Cunha et al. (17). After identification sodes of PD-related peritonitis caused by CNS that oc- and susceptibility testing, strains were stored frozen at curred between June 1994 and December 2011. Exclusion 280°C in Trypticase Soy Broth with 15% glycerol. Pheno- criteria were the presence of concomitant exit site or tunnel typic identification was confirmed by internal transcribed infections and incomplete information. A peritonitis diagno- spacer PCR, after extraction of bacterial whole nucleic sis was based on at least two of the following: abdominal acids (18). Discrepancies were resolved using matrix-assisted pain or cloudy dialysate, dialysate white cell count .100/ml laser desorption ionization time-of-flight mass spectrometry with at least 50% neutrophilic cells, and positive culture of (19,20) using a MALDI-TOF Biotyper (Bruker Daltonik, dialysate (14). A repeat refers to an episode that occurs .4 Bremen, Germany), and/or partial sequencing of RNA weeks after completion of therapy of a prior episode with polymerase b-subunit gene rpo B (21) followed by Basic the same CNS species (14). Recurrence refers to an episode Local Alignment Search Tool analysis (22). that occurs within 4 weeks of completion of therapy of a The in vitro susceptibilities to oxacillin and vancomycin prior episode but with a different CNS species (14). A re- were determined based on minimal inhibitory concentra- lapse is an episode caused by the same CNS species, or a tions by the E test (bioMérieux, Inc., Durham, NC) and the negative culture result accompanied by signs or symptoms broth microdilution technique, respectively. The microdi- of peritonitis within 28 days of completion of antibiotic ther- lution test was performed according to Clinical and Lab- apy (14). Resolution was defined as the disappearance of oratory Standards Institute (CLSI) recommendations (23) signs and symptoms within 5 days after the beginning utilizing cation-adjusted Mueller Hinton Broth (Becton of antibiotic therapy (14). Refractory peritonitis is the failure Dickinson [BD]) and vancomycin hydrochloride (Sigma- of the effluent to clear after 5 days using appropriate antibiotics Aldrich, Shanghai, China). Standard reference strains (S. (14). Peritonitis-related death refers to the death of a patient aureus ATCC 29213; Enterococcus faecalis ATCC 29212, and with active peritonitis or to a patient who had an episode E. faecalis ATCC 51299) were utilized for quality control. within the previous 2 weeks (14). Nonresolution refers to the The susceptibility was defined based on the 2013 CLSI patients who required catheter removal by refractory perito- breakpoints (8). Strains presenting intermediate values nitis, needed a second antibiotic regimen, relapsed, or pro- were considered resistant. gressed to death. PCR amplification of bacterial DNA was performed to Patients and/or caretakers received at least 20 hours of detect the mecA gene, biofilm-related gene (icaAD), and the theoretical and practical PD training by a senior nurse. genes encoding enterotoxins A–D(sea, seb, sec,andsed), Retraining was performed if peritonitis occurred. If the TSST-1 (tst), and Panton-Valentin leukocidin (lukPV). Re- patient reported contamination, a single dose of cefazolin actions were carried out in 0.2-ml microcentrifuge tubes con- (1 g intraperitoneally) was given. taining 10 pmol of each primer, 2.0 U of Taq-DNA polymerase, 100 mM of deoxyribonucleotide triphosphates, m Data Collection 10 mM Tris-HCl (pH 8.4), 0.75 mM MgCl2,and3 l nucleic The following information was recorded for each case: acid in a total volume of 25 ml. Primer sequences and ampli- (1) episode: date, clinical findings, previous peritonitis ep- fication conditions of the PCR reactions were published else- isodes, vancomycin use, outcome (resolution, relapse, catheter where (24–27). The following S. aureus reference strains were removal, or death); (2) presence of diabetes; (3)demographics: included in the reactions as positive controls: ATCC 33591 age, sex, and race (Caucasian, non-Caucasian), and (4)dialy- (mecA positive), ATCC 13565 (sea positive), ATCC 14458 (seb sis vintage and modality. positive), ATCC 19095 (sec positive), ATCC 23235 (sed posi- Patients were initially treated within 24 hours of the tive), ATCC 51650 (tst positive), and ATCC 49775 (lukPV onset of signs or symptoms using contemporary antibiotic positive); water was used as a template for negative con- recommendations (7,14,15). The peritonitis rate (episodes trol PCR reactions. The efficiency of amplification was per patient per year) and the proportion of oxacillin-resistant monitored by electrophoresis on 1.5% agarose gel pre- isolates were calculated for three periods of 6 years. pared in 13Tris-borate-EDTA buffer and stained with During period 1 (1994–1999), antibiotic prophylaxis was ethidium bromide. The size of amplified products was not prescribed, and the connection systems were the Y set compared with a 100-bp standard and the gels were pho- (1994–1998) or the twin bag (introduced in 1999). Auto- tographed under ultraviolet transillumination. mated PD was introduced in 1998, and its indication for Production of a-andb-hemolysin was determined on prevalent patients was based only on clinical criteria; for plates containing a blood agar base supplemented with 5% incident patients, the main criterion was the patient’sop- rabbit blood and 5% sheep blood, respectively. The plates tion. During period 2 (2000–2005), antibiotic prophylaxis were incubated for 24 hours at 37°C. The formation of was not prescribed until 2003, when daily application of hemolysis zones around the isolated colonies indicated a 1076 Clinical Journal of the American Society of Nephrology

Figure 1. | Annual rate (episodes per patient per year) of CNS peritonitis from June 1994 to December 2011 and introduction of changes in peritoneal dialysis practice. Dashed line represents a trend curve of peritonitis rate. APD, automated peritoneal dialysis; CNS, coagulase- negative Staphylococcus. positive result. Lipolytic activity was determined on plates variables was tested and, if statistically significant inter- containing blood agar enriched with 0.01% CaCl2/2H2O actions were presented, one of them was excluded. Uni- and 1% Tween 80. A positive result was defined as the variate logistic regression was performed to select formation of opacity around the colony after incubation variables for the final model, using P.0.20 as the elimi- at 37°C for 18 hours, followed by incubation at room tem- nation criterion, and to evaluate the association between perature for 24 hours (28). The production of lecithinase microbiologic factors and repeat or relapse episodes. The was evaluated using Baird-Parker medium. The formation Hosmer and Lemeshow test was used to assess the fitof of an opaque halo around the colony indicated positivity logistic regression models. A P value ,0.05 was consid- (29). Nuclease (DNAse) and thermonuclease (TNAse) pro- ered significant. duction were detected using the metachromatic Toluidine blue O agar diffusion-DNA technique as described by Lachica et al. (30). Supernatants obtained by the sac culture Results method described by Donnelly et al. (31) were transferred Between 1994 and 2011, 878 peritonitis episodes were to the wells of plates containing metachromatic Toluidine diagnosed; CNS caused 135 of them. After the exclusion blue O agar. The culture supernatant was first heated in a criteria were applied, 115 episodes were enrolled. The water bath for 20 minutes for the detection of TNAse. Nu- clease (DNAse and TNAse) activity was evaluated by measuring the diameter of pink halos (in millimeters) formed on the medium. Positive results were interpreted Table 1. Clinical findings at initial presentation in 115 by comparing the halos with those obtained for a standard episodes of peritoneal dialysis–related peritonitis caused by coagulase-negative Staphylococci DNAse and TNAse-positive S. aureus strain (ATCC 25923). Sign or Symptom Frequency, n (%) Statistical Analyses fl Chi-squared or Fisher’s exact tests were used to compare Cloudy dialysis ef uent 106 (92.2) frequencies. Incidences were compared using the Poisson Abdominal pain 89 (77.4) Nausea or vomiting 47 (40.9) regression model. Multivariate analysis by logistic regres- Fever 17 (14.8) sion was used to determine independent predictors of out- Hypotension 12 (10.4) comes (resolution or nonresolution). Collinearity among Clin J Am Soc Nephrol 9: 1074–1081, June, 2014 Coagulase-Negative Staphylococcus Peritonitis in Peritoneal Dialysis, Camargo et al. 1077

overall peritonitis rate was 0.89 and the CNS peritonitis rate respectively (P=0.18).The mecA gene was detected in 55 was 0.15 episodes per patient per year. The incidences of isolates (47.8%), correlating in 73.3% with oxacillin resis- CNS peritonitis were 0.12, 0.14, and 0.16 in periods 1, 2, and tance detected by the E test. All strains were susceptible to 3, respectively (P=0.21). There was a nonstatistically signif- vancomycin. Species-specific patterns of susceptibility are icant decline after introduction of the Y set system (1995) shown in Table 3. Analysis of the five most frequent spe- (P=0.10) and a slight increase of incidence over time cies revealed similar oxacillin resistance rates except S. (P=0.19) (Figure 1). warneri strains, which had a lower rate than S. haemolyticus The 115 peritonitis episodes occurred in 74 patients; 47 (P=0.02).Enzyme-,toxin-,andbiofilm-encoding genes patients had one, 16 had two, and 11 had three or more were detected mainly in S. epidermidis isolates (Table 4). episodes. Demographic and clinical data at the first episode At least one enterotoxin gene was found in 20 strains showed that the patient’s mean age was 53.6615.9 years, (17.4%), the simultaneous presence of sec and sed was ver- 43 were women (58.1%), 60 were Caucasians (81.1%), 35 ified in one strain of S. epidermidis and one of S. hominis, were aged.60 years (47.3%), 30 presented with diabetes and another S. epidermidis strain carried sea, sec,andsed (40.5%), and 59 were treated by continuous ambulatory genes. Toxic shock syndrome toxin-1 and Panton-Valentine PD (79.7%). Eighty-eight cases (76.5%) were new episodes, leukocidin encoding genes (tst and lukPV, respectively) were 19 (16.5%) were repeat or recurring, and eight (6.6%) were not detected. relapses. Clinical findings at the initial presentation are A multivariate logistic regression model showed that shown in Table 1. oxacillin susceptibility (odds ratio [OR], 4.41; 95% confi- Vancomycin was used in 81 episodes (70.4%; as the dence interval [95% CI], 1.48 to 13.17; P=0.01) and vanco- initial treatment in 40 episodes and as a second regimen in mycin use as the first treatment (OR, 22.27; 95% CI, 6.16 to 41 episodes), whereas cefazolin was initially prescribed in 80.53; P,0.001) were the only independent predictors of 75 cases. Overall, 70 episodes (60.9%) resolved. Among resolution. Clinical, demographic, and dialysis-related fac- nonresolved infections, 18 (15.7%) required catheter re- tors (e.g., sex, race, age, diabetes, dialysis vintage, number moval because of refractory peritonitis, 14 (12.2%) relapsed, of previous peritonitis episodes, and PD modality) did not 10 (8.7%) were cured with a second antibiotic, and 3 (2.6%) influence the outcome. In addition, factors such as CNS progressed to death. Figure 2 expresses the outcomes ac- species, toxin and enzyme production, or the presence of cording to oxacillin susceptibility and initial therapy. the mecAandicaAD genes were not associated with the S. epidermidis was the most frequent cause (62.6%), fol- outcome. On the other hand, the presence of the mecA(OR, lowed by Staphylococcus haemolyticus (11.3%), Staphylococcus 3.11; 95% CI, 1.17 to 8.27; P=0.02) and icaAD (OR, 2.82; warneri (7%), Staphylococcus cohnii (5.2%), and Staphylococcus 95% CI, 1.11 to 7.19; P=0.03) genes were associated with hominis (4.3%). Seven other species were identified, account- relapse and repeat episodes. ing for 9.6% of episodes (Table 2). The resolution rate was 53.4% in a second analysis Oxacillin resistance was detected in 80 strains (69.6%), enrolling only new episodes (n=88). Oxacillin susceptibility with rates of 62.7%, 84%, and 70.9% in periods 1, 2, and 3, (OR, 4.64; 95% CI, 1.14 to 18.3; P=0.03) and vancomycin as

Figure 2. | Clinical outcomes of coagulase-negative peritonitis, according to oxacillin susceptibility and initial antibiotic therapy. 1078 Clinical Journal of the American Society of Nephrology

the first treatment (OR, 25.13; 95% CI, 4.89 to 129.04; country (33) could explain, at least in part, these contrast- P,0.001) remained the only independent predictors of res- ing findings. olution. The P values from Hosmer and Lemeshow tests The CNS peritonitis rate between 1994 and 1999 was for the first and the second multivariate models were 0.72 lower than that observed in contemporary series present- and 0.43, respectively. ing a temporal decrease in the incidence of these infections (4,34). In addition, we observed a further increase in the incidence of CNS peritonitis, concomitant with a decline in Discussion the frequency of culture-negative episodes from 15.1% to Although our center has experienced a reduction in 9.9% between 1990 and 2010 (35). Perhaps the chance of peritonitis rates over the last 2 decades (32), the incidence recovering from a CNS is higher in settings with a low of CNS peritonitis did not vary. After an initial reduction culture-negative rate. Holley et al. (36) observed improve- associated with Y set use, a trend to increase occurred over ment in dialysate culture positivity proportional to high- time. By contrast, we recently reported a strong decline in er S. epidermidis recovery, which is in agreement with the S. aureus peritonitis rate (11) similar to that observed in our findings. Thus, in this study, the maintenance of in- other centers (4). It is possible that the introduction of safer cidence despite introduction of new technologies and dialysis connections and antibiotic prophylaxis had a antibiotic prophylaxis may be explained by the low ini- greater effect on S. aureus than CNS peritonitis prevention. tial rate or may have been blunted by an increase in CNS A high prevalence of S. aureus nasal carriage in our identification. Our overall CNS peritonitis rate was similar to that reported by Fahim et al. (13) and was greater than that observed by Szeto et al. (12). The first study showed a Table 2. Causative agents of 115 episodes of peritoneal greater frequency of diabetes among patients with CNS dialysis–associated peritonitis caused by coagulase-negative peritonitis. In our series, the frequency of patients with Staphylococci diabetes was similar to the results of Fahim et al. (13). Moreover, these authors had results of only 12% of culture- Staphylococcus Species Frequency, n (%) negative episodes (13). We speculate that the frequency of patients with diabetes and a low culture-negative rate may S. epidermidis 72 (62.6) fi S. haemolyticus 13 (11.3) explain the similarities between our ndings and those of S. warneri 8(6.9) Fahim et al. (13). S. cohnii 6(5.2) The predominance of S. epidermidis was noticeable, ac- S. hominis 5(4.3) counting for .60% of the episodes. The remaining epi- S. saprophyticus 4(3.5) sodes were caused by 11 other species. In this context, S. capitis 2(1.7) our study presents novel information on CNS peritonitis; S. caprae 1(0.9) this information is still scarce despite the International So- S. lugdunensis 1(0.9) ciety for Peritoneal Dialysis (ISPD) guideline recommen- S. pasteuri 1(0.9) dations, which emphasize the importance of distinctive S. sciuri 1(0.9) CNS species causative of peritonitis (7). S. xylosus 1(0.9) Total 115 (100.0) A high oxacillin resistance rate was observed, which likely explains the number of episodes treated by vancomycin.

Table 3. Vancomycin and oxacillin susceptibility patterns of coagulase-negative Staphylococci isolated from peritoneal dialysis– associated peritonitis m MIC50, g/ml Staphylococcus Species n Oxacillin Resistance Rate, n (%) mecA Gene, n (%) Vancomycin Oxacillin

S. epidermidis 72 2.0 1.0 50 (71.4) 42 (58.3) S. haemolyticus 13 1.0 4.0 11 (84.6) 8 (61.5) S. warneri 8 1.0 0.25 3 (37.5) 1 (12.5) S. cohnii 6 1.5 0.5 5 (83.3) 0 S. hominis 5 0.5 0.38 3 (60.0) 4 (80.0) S. saprophyticus 4 1.0 0.38 3 (75.0) 0 S. capitis 2 1.0 0.38 2 (100) 0 S. xylosus 1 1.0 0.38 1 (100) 0 S. sciuri 1 1.0 0.5 1 (100) 0 S. pasteuri 1 1.0 0.75 1 (100) 0 S. lugdunensis 1 0.5 0.75 0 0 S. caprae 10.250.125 0 0 Total 115 1.0 0.75 80 (69.6) 55 (47.8)

MIC50, minimal inhibitory concentration that inhibited the growth of 50% of isolates. Clin J Am Soc Nephrol 9: 1074–1081, June, 2014 Coagulase-Negative Staphylococcus Peritonitis in Peritoneal Dialysis, Camargo et al. 1079

Oxacillin resistance, detected phenotypically, had a 73% con- cordance with mecA detection. This was influenced mainly

AD by the low breakpoints from the 2013 CLSI (8), once 35% of resistant strains were mecA negative, whereas only 8.6% of susceptible were mecA positive. Similar resistance profiles were observed among CNS species, although S. warneri and S. haemolyticus trended toward low and high oxacillin resis- tance rates, respectively. Beyond peritonitis, an increasing CNS oxacillin resis- tance rate has been observed in cases of outpatient and nosocomial infections at our hospital, particularly among outpatient infections, with a current rate of .50% (unpub- lished data). The SENTRY Antimicrobial Surveillance Pro- gram has documented a high and slightly increasing D encoding genes, respectively; TNAse,

– methicillin resistance rate among CNS around the world (37). Oxacillin susceptibility was an independent predictor of resolution. Methicillin resistance has been associated with an increased risk of relapse in CNS peritonitis (13) and with sea seb sec sed ica worse outcomes of S. aureus infections (38,39). The largest

, enterotoxin A report of CNS peritonitis showed that detection of methi- isolated from peritoneal dialysis–associated peritonitis cases

sed cillin resistance and prompt change of the treatment to

and vancomycin improved the cure rate (12). Vancomycin had high in vitro activity against all of the sec, , strains, with minimum inhibitory concentration values seb

Staphylococci # m , 2.0 g/ml. Resistance to vancomycin is seldom observed

sea in staphylococci, even in CNS in which glycopeptide re- sistance is more frequent (40). The initial vancomycin prescription was a predictor of resolution, possibly because of the high oxacillin resistance rate. However, its use as a second treatment was associated with a low resolution rate. We speculate that a delay in the lm-related genes;

fi vancomycin prescription or its use in patients with severe infections could explain these findings. CNS strains had low expressive enzyme and toxin AD, bio production, although the icaAD gene-encoding biofilm ica was found in approximately 35% of them. However, these factors were not associated with the outcome. This finding differs from the observed in S. aureus episodes, in which b-hemolysin production was a predictor of a poor out- come (11). Even we have not used techniques for direct observation of biofilm, the presence of icaAD was associ- ated with relapse and repeat peritonitis, reinforcing that -Hemolysin Lipase Lecithinase DNAse TNAse fi b bio lm increases the risk for relapse (41). However, its associationwithrepeatedinfections was not previously reported. Interestingly, Nessim et al. (42) reported that prior CNS peritonitis was associated with an increased risk of CNS peritonitis within the subsequent year. This result may be associated with the presence of biofilm in the -Hemolysin

a catheter of patients suffering a repeat CNS episode. Our results suggest that catheter removal should be considered 865 4 (50.0)4 3 (50.0)20 4 (80.0)1 1 (25.0)10 0 2 (25.0)10 2 0 (33.3)1 0 1 0 0 (20.0)1 0 0 0 6 (75.0) 0 00000 1 0 0 (100.0) 0 3 (50.0) 0 0 0 00000 4 (50.0) 0 0 0 0 00000 1 (16.7) 1 (100.0) 1 (100.0) 2 1 (25.0) (25.0) 1 (20.0) 2 (25.0) 0 1 0 (25.0) 0 0 0 1 (16.7) 0 0 1 (100.0) 0 0 0 0 0 0 0 1 (12.5) 0 0 0 0 0 0 0 0 0 0 1 (100.0) 0 0 0 1 (100.0) 0 1 (20.0) 0 1 (25.0) 1 0 (20.0) 0 0 0 0 0 0 0 2 (50.0) 0 0 0 0 0 0 0 0 7213 21 (29.2) 10 (76.9) 14 (19.4) 7 (53.8) 24 (33.3) 1 12 (7.7) (16.7) 5 (6.9) 1 (7.7) 2 (2.8) 0 2 (2.8) 3 (4.2) 10 (13.9) 0 3 (4.2) 40 (55.6) 0 0 2 (15.4) 0 0 n for patients presenting with high peritonitis rates and re- peat infections. This study had limitations such as the small number of

(%) unless otherwise indicated. DNAse, deoxyribonuclease; cases and its single-center characteristic, which limits wide

n extrapolation of its results. However, this work presents Species novel information regarding CNS peritonitis, such as a detailed characterization of the species and the identifica- tion of the mecA gene as well as the association of biofilm with repeat peritonitis. Furthermore, this is the first Latin American study analyzing a long-term series of CNS PD- thermonuclease. S. cohnii S. hominis S. saprophyticus S. capitis S. xylosus S. sciuri S. pasteuri S. lugdunensis S. caprae Total 115 44 (38.3) 28 (24.3) 35 (30.4) 21 (18.3) 8 (7.0) 6 (5.2) 2 (1.7) 3 (2.6) 15 (13.0) 4 (3.5) 42 (36.5) S. haemolyticus S. warneri S. epidermidis Table 4. Rates of enzyme productionStaphylococcus and detection of biofilm- and toxin-encoding genes in coagulase-negative Data are given as related peritonitis. 1080 Clinical Journal of the American Society of Nephrology

In conclusion, the odds of resolution were strongly in- 14. Piraino B, Bailie GR, Bernardini J, Boeschoten E, Gupta A, fluenced by oxacillin susceptibility and initial vancomycin Holmes C, Kuijper EJ, Li PK, Lye WC, Mujais S, Paterson DL, Fontan MP, Ramos A, Schaefer F, Uttley L; ISPD Ad Hoc Advisory prescription, emphasizing the pertinence of ISPD guidelines Committee: Peritoneal dialysis-related infections recom- (7,14) regarding antimicrobial resistance monitoring in every mendations: 2005 update. Perit Dial Int 25: 107–131, 2005 dialysis center toward a rational definition of peritonitis 15. Keane WF, Bailie GR, Boeschoten E, Gokal R, Golper TA, Holmes treatment protocols. CJ, Kawaguchi Y, Piraino B, Riella M, Vas S; International Society for Peritoneal Dialysis: Adult peritoneal dialysis-related perito- nitis treatment recommendations: 2000 update. Perit Dial Int 20: Acknowledgments 396–411, 2000 The authors thank Dr. Rogerio Carvalho Oliveira and Dr. Thyago 16. Koneman EW, Winn WC: Koneman’s Color Atlas and Textbook Proença de Moraes for their technical support. of Diagnostic Microbiology, 6th Ed., Philadelphia, Lippincott This work was supported by a grant from the São Paulo Research Williams & Wilkins, 2006 Foundation. This agency played no role in the study design, data 17. Cunha ML, Sinzato YK, Silveira LV: Comparison of methods for the identification of coagulase-negative staphylococci. Mem Inst collection and analysis, decision to publish, or manuscript prepa- Oswaldo Cruz 99: 855–860, 2004 ration. 18. Couto I, Pereira S, Miragaia M, Sanches IS, de Lencastre H: Identification of clinical staphylococcal isolates from humans by Disclosures internal transcribed spacer PCR. J Clin Microbiol 39: 3099–3103, P.B. is an active member of the Baxter Latin American Renal 2001 Advisory Board. 19. Spanu T, De Carolis E, Fiori B, Sanguinetti M, D’Inzeo T, Fadda G, Posteraro B: Evaluation of matrix-assisted laser desorption ioniza- tion-time-of-flight mass spectrometry in comparison to rpoBgene References sequencing for species identification of bloodstream infection staphylococcal isolates. Clin Microbiol Infect 17: 44–49, 2011 1. Davenport A: Peritonitis remains the major clinical complication 20. Carpaij N, Willems RJ, Bonten MJ, Fluit AC: Comparison of the of peritoneal dialysis: The London, UK, peritonitis audit 2002- 2003. Perit Dial Int 29: 297–302, 2009 identification of coagulase-negative staphylococci by matrix- 2. Pe´rez Fontan M, Rodrı´guez-Carmona A, Garcı´a-Naveiro R, assisted laser desorption ionization time-of-flight mass spec- Rosales M, Villaverde P, Valde´s F: Peritonitis-related mortality in trometry and tuf sequencing. Eur J Clin Microbiol Infect Dis 30: patients undergoing chronic peritoneal dialysis. Perit Dial Int 25: 1169–1172, 2011 274–284, 2005 21. Mellmann A, Becker K, von Eiff C, Keckevoet U, Schumann P, 3. Zelenitsky S, Barns L, Findlay I, Alfa M, Ariano R, Fine A, Harding Harmsen D: Sequencing and staphylococci identification. Emerg G: Analysis of microbiological trends in peritoneal dialysis- Infect Dis 12: 333–336, 2006 related peritonitis from 1991 to 1998. Am J Kidney Dis 36: 1009– 22. National Center for Biotechnology Information: BLAST: Basi- 1013, 2000 cLocal Alignment Search Tool. Available at: http://blast.ncbi. 4. Kim DK, Yoo TH, Ryu DR, Xu ZG, Kim HJ, Choi KH, Lee HY, Han nlm.nih.gov/Blast.cgi. Accessed August 21, 2013 DS, Kang SW: Changes in causative organisms and their anti- 23. Clinical and Laboratory Standards Institute: Methods for Dilution microbial susceptibilities in CAPD peritonitis: A single center’s Antimicrobial Susceptibility Tests for Bacteria That Grow Aero- experience over one decade. Perit Dial Int 24: 424–432, 2004 bically M07-A9, Wayne, PA, Clinical and Laboratory Standards 5. Euze´by JP: List of prokaryotic names with standing in nomen- Institute, 2012 clature - genus Staphylococcus. Available at: http://www. 24. Murakami K, Minamide W, Wada K, Nakamura E, Teraoka H, bacterio.net/s/staphylococcus.html. Accessed September 1, Watanabe S: Identification of methicillin-resistant strains of 2013 staphylococci by polymerase chain reaction. J Clin Microbiol 29: 6. Schnitzler N, Meilicke R, Conrads G, Frank D, Haase G: 2240–2244, 1991 Staphylococcus lugdunensis: Report of a case of peritonitis and 25. Johnson WM, Tyler SD, Ewan EP, Ashton FE, Pollard DR, Rozee an easy-to-perform screening strategy. J Clin Microbiol 36: 812– KR: Detection of genes for enterotoxins, exfoliative toxins, and 813, 1998 toxic shock syndrome toxin 1 in Staphylococcus aureus by the 7. Li PK, Szeto CC, Piraino B, Bernardini J, Figueiredo AE, Gupta A, polymerase chain reaction. J Clin Microbiol 29: 426–430, 1991 Johnson DW, Kuijper EJ, Lye WC, Salzer W, Schaefer F, Struijk 26. da Cunha ML, Calsolari RA, Ju´nior JP: Detection of enterotoxin DG; International Society for Peritoneal Dialysis: Peritoneal di- and toxic shock syndrome toxin 1 genes in Staphylococcus, with alysis-related infections recommendations: 2010 update. Perit emphasis on coagulase-negative staphylococci. Microbiol Im- Dial Int 30: 393–423, 2010 munol 51: 381–390, 2007 8. Clinical and Laboratory Standards Institute: Performance Stan- 27. Arciola CR, Baldassarri L, Montanaro L: Presence of icaAand dards for Antimicrobial Susceptibility Testing M100-S23, Wayne, icaD genes and slime production in a collection of staphylo- PA, Clinical and Laboratory Standards Institute, 2013 coccal strains from catheter-associated infections. J Clin Micro- 9. Martins A, Cunha ML: Methicillin resistance in Staphylococcus biol 39: 2151–2156, 2001 aureus and coagulase-negative staphylococci: Epidemiological and 28. Jessen O, Faber V, Rosendal K, Eriksen KR: Some properties of molecular aspects. Microbiol Immunol 51: 787–795, 2007 Staphylococcus aureus, possibly related to pathogenicity. Part 1. 10. da Cunha ML, Caramori JC, Fioravante AM, Batalha JE, Montelli A study of 446 strains from different types of human infection. AC, Barretti P: Significance of slime as virulence factor in co- Acta Pathol Microbiol Scand 47: 316–326, 1959 agulase-negative staphylococcus peritonitis in CAPD. Perit Dial 29. Matos JE, Harmon RJ, Langlois BE: Lecithinase reaction of Int 24: 191–193, 2004 Staphylococcus aureus strains of different origin on Baird-Parker 11. Barretti P, Moraes TM, Camargo CH, Caramori JC, Mondelli AL, medium. Lett Appl Microbiol 21: 334–335, 1995 Montelli AC, da Cunha ML: Peritoneal dialysis-related peritonitis 30. Lachica RV, Genigeorgis C, Hoeprich PD: Metachromatic agar- due to Staphylococcus aureus: A single-center experience over diffusion methods for detecting staphylococcal nuclease activity. 15 years. PLoS ONE 7: e31780, 2012 Appl Microbiol 21: 585–587, 1971 12. Szeto CC, Kwan BC, Chow KM, Lau MF, Law MC, Chung KY, 31. Donnelly CB, Leslie JE, Black LA, Lewis KH: Serological identi- Leung CB, Li PK: Coagulase negative staphylococcal peritonitis fication of enterotoxigenic staphylococci from cheese. Appl in peritoneal dialysis patients: Review of 232 consecutive cases. Microbiol 15: 1382–1387, 1967 Clin J Am Soc Nephrol 3: 91–97, 2008 32. Oliveira LG, Luengo J, Caramori JC, Montelli AC, Cunha ML, 13. Fahim M, Hawley CM, McDonald SP, Brown FG, Rosman JB, Barretti P: Peritonitis in recent years: Clinical findings and pre- Wiggins KJ, Bannister KM, Johnson DW: Coagulase-negative dictors of treatment response of 170 episodes at a single Brazilian staphylococcal peritonitis in Australian peritoneal dialysis pa- center. Int Urol Nephrol 44: 1529–1537, 2012 tients: Predictors, treatment and outcomes in 936 cases. Nephrol 33. Giarola LB, Dos Santos RR, Bedendo J, da Silva Ju´nior WV,Borelli Dial Transplant 25: 3386–3392, 2010 SD: HLA molecules and nasal carriage of Staphylococcus aureus Clin J Am Soc Nephrol 9: 1074–1081, June, 2014 Coagulase-Negative Staphylococcus Peritonitis in Peritoneal Dialysis, Camargo et al. 1081

isolated from dialysis and kidney transplant patients at a hospital dialysis: Review of 245 consecutive cases. Clin J Am Soc Nephrol 2: in Southern Brazil. BMC Res Notes 5: 90, 2012 245–251, 2007 34. Moraes TP, Pecoits-Filho R, Ribeiro SC, Rigo M, Silva MM, 39. Govindarajulu S, Hawley CM, McDonald SP, Brown FG, Rosman Teixeira PS, Pasqual DD, Fuerbringer R, Riella MC: Peritoneal JB, Wiggins KJ, Bannister KM, Johnson DW: Staphylococcus dialysis in Brazil: Twenty-five years of experience in a single aureus peritonitis in Australian peritoneal dialysis patients: Pre- center. Perit Dial Int 29: 492–498, 2009 dictors, treatment, and outcomes in 503 cases. Perit Dial Int 30: 35. Caramori JCT: Vinte anos de dia´lise peritoneal no Hospital das 311–319, 2010 Clı´nicas da Faculdade de Medicina de Botucatu [thesis]. Botucatu, 40. Srinivasan A, Dick JD, Perl TM: Vancomycin resistance in Sao Paulo, Brazil, Sa˜o Paulo State University; 2011 staphylococci. Clin Microbiol Rev 15: 430–438, 2002 36. Holley JL, Moss AH: A prospective evaluation of blood culture 41. Dasgupta MK: Biofilms and infection in dialysis patients. Semin versus standard plate techniques for diagnosing peritonitis in Dial 15: 338–346, 2002 continuous ambulatory peritoneal dialysis. Am J Kidney Dis 13: 42. Nessim SJ, Nisenbaum R, Bargman JM, Jassal SV: Microbiology of 184–188, 1989 peritonitis in peritoneal dialysis patients with multiple episodes. 37. SENTRYAntimicrobial Surveillance Program: http://gpcsusceptibilitydata. Perit Dial Int 32: 316–321, 2012 com/content/methicillin-resistant.asp?page=content&nav=2& subnav=1&subnav2=1&state=0. Accessed December 7, Received: September 5, 2013 Accepted: February 12, 2014 2013 38. Szeto CC, Chow KM, Kwan BC, Law MC, Chung KY, Yu S, Leung Published online ahead of print. Publication date available at www. CB,LiPK:Staphylococcus aureus peritonitis complicates peritoneal cjasn.org.